1 NO. 90-CI-6033 JEFFERSON CIRCUIT COURT DIVISION ONE (1) 2 3 JOYCE FENTRESS, ET AL. PLAINTIFFS 4 5 VS. DEPOSITION FOR PLAINTIFFS 6 7 SHEA COMMUNICATIONS, ET AL. DEFENDANTS 8 * * * * * * * * * * 9 10 DEPONENT: DR. ROBERT ZERBE 11 DATE: AUGUST 9, 1994 12 13 * * * * * * * * * * 14 15 16 REPORTER: KATHY NOLD 17 18 KENTUCKIANA REPORTERS SUITE 260 19 730 WEST MAIN STREET LOUISVILLE, KENTUCKY 40202 20 (502) 589-2273 Page 1 1 * * * * * * * * * * 2 3 UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF INDIANA 4 INDIANAPOLIS DIVISION 5 IN RE ELI LILLY AND COMPANY ) Prozac Products Liability ) MDL Docket No. 907 6 Litigation ) 7 * * * * * * * * * * 8 NO. 91-02496-A 9 JACKIE LYNN BIFFLE, ET AL ) IN THE DISTRICT ) COURT OF 10 V. ) DALLAS COUNTY, TEXAS ) 11 ELI LILLY & COMPANY AND ) 14TH JUDICIAL DISTA PRODUCTS COMPANY ) DISTRICT 12 * * * * * * * * * * 13 NO. 92-14775-E 14 RICHARD HAROLD CROSSETT, JR., ) IN THE 15 CHAD H. CROSSETT, AMY MICHELLE ) DISTRICT CROSSETT AND KRISTEN ANN CROSSETT, ) COURT OF 16 INDIVIDUALLY AND AS SURVIVORS OF ) AND ON BEHALF OF THE ESTATE OF ) 17 JOCQUETTA ANN CROSSETT, DECEASED ) ) 18 V. ) DALLAS COUNTY, ) TEXAS 19 ELI LILLY & COMPANY, DISTA ) PRODUCTS COMPANY, TEXAS ) 20 PSYCHIATRIC COMPANY, INC. ) D/B/A/ HCA WILLOW PARK ) 101ST JUDICIAL 21 HOSPITAL, JAMES K. WITSCHY, M.D., ) DISTRICT AND DOUG BELLAMY, ED.D. ) Page 2 1 * * * * * * * * * * 2 NO. A-921,405-C 3 MARIA GUADALUPE REVES ) IN THE 4 INDIVIDUALLY AND AS NEXT ) DISTRICT COURT FRIEND OF GRANT JULIAN REVES ) OF 5 A MINOR CHILD, AND ON BEHALF ) OF THE ESTATE OF CHRISTIAN ) 6 MARIE REVES, DECEASED ) ) ORANGE COUNTY, 7 V. ) TEXAS ) 8 ELI LILLY & COMPANY, DISTA ) PRODUCTS COMPANY, RAVIKUMAR ) 9 KANNEGANTI, M.D., HOSPITAL ) CORPORATION OF AMERICA, A ) 10 TENNESSEE CORPORATION, HEALTH ) SERVICES ACQUISITION CORP., ) 11 A DELAWARE CORPORATION, ) HCA PSYCHIATRIC COMPANY, A ) 12 DELAWARE CORPORATION, TEXAS ) PSYCHIATRIC CO., INC.. A/K/A ) 13 AND/OR D/B/A HCA BEAUMONT ) NEUROLOGICAL HOSPITAL, AND HCA ) 14 HEALTH SERVICES OF TEXAS, INC. ) 128TH JUDICIAL A/K/A AND/OR BEAUMONT ) DISTRICT 15 NEUROLOGICAL HOSPITAL ) Page 3 1 * * * * * * * * * * 2 IN THE CIRCUIT COURT OF COOK COUNTY, ILLINOIS 3 COUNTY DEPARTMENT - LAW DIVISION 4 RENATO DI SILVESTRO, Individually ) and as Special Administrator of ) 5 the Estate of JOHN DI SILVESTRO, ) Deceased, ) 6 ) Plaintiff, ) 7 ) v. ) No. 91 L 7881 8 ) ROBERT L. NELSON, et al., ) 9 ) Defendants, ) 10 ) GEORGE MELNICK, M.D. and PETER ) 11 FINK, M.D. ) ) 12 Respondents in Discovery.) 13 * * * * * * * * * * Page 4 1 IN THE CIRCUIT COURT OF THE SIXTH JUDICIAL CIRCUIT CHAMPAIGN COUNTY, ILLINOIS 2 LINDA GARDNER, Individually and ) 3 as Special Administrator of ) the Estate of SHANE GARDNER, ) 4 deceased, ) ) 5 Plaintiff, ) ) 6 v. ) No. 91 L 1066 ) 7 ELI LILLY AND COMPANY, a foreign ) corporation, ) 8 ) Defendant. ) 9 10 * * * * * * * * * * Page 5 1 SUPERIOR COURT OF THE STATE OF CALIFORNIA 2 FOR THE COUNTY OF LOS ANGELES 3 DR. MARIUS SAINES, etc., et al., ) Case No: 4 ) SC 008331 Plaintiffs, ) 5 ) vs. ) 6 ) ELI LILLY & COMPANY, a corporation; ) 7 DISTA PRODUCTS COMPANY, a division ) of Eli Lilly & Company; and DOBS 1- ) 8 100, inclusive, ) ) 9 Defendants. ) ____________________________________) 10 11 * * * * * * * * * * 12 NO. 93-8792-D 13 DAVID KUNG, DALE KUNG COHEN ) IN THE DISTRICT ROBERT KUNG, AND TIMOTHY KUNG, ) COURT OF 14 INDIVIDUALLY AND AS SURVIVORS ) AND STATUTORY BENEFICIARIES ) 15 OF MAY YUN KUNG, DECEASED ) ) 16 VS. ) DALLAS COUNTY ) T E X A S 17 ELI LILLY AND COMPANY, DISTA ) PRODUCTS COMPANY, AND MONIQUE ) 18 KUNKLE, PH.D. ) Page 6 1 * * * * * * * * * * 2 IN THE DISTRICT COURT OF JOHNSON COUNTY, KANSAS 3 CIVIL COURT DEPARTMENT 4 EUGENE HUSLIG, AS ADMINISTRATOR ) 5 AND EXECUTOR AND ON BEHALF OF ) THE ESTATE OF DEBORAH G. WEATHERS ) 6 HUSLIG, DESCEASED, AND AS SURVIVING ) HUSBAND AND HEIR AT LAW OF DEBORAH ) 7 G. WEATHERS HUSLIG, DECEASED, ) AND IN HIS INDIVIDUAL CAPACITY AS ) 8 HUSBAND OF DEBORAH G. WEATHERS ) HUSLIG, DECEASED, AND RONALD C. ) 9 WEATHERS, SON OF DEBORAH G. ) WEATHERS HUSLIG, DECEASED, ) CASE NO.: 10 ) 94 C 192 PLAINTIFFS, ) 11 ) VS. ) 12 ) COURT NO. 7 MARY L. BILLINGSLEY, EXECUTOR OF ) CHAPTER 60 13 THE ESTATE OF THAD BILLINGSLEY, ) M.D., DECEASED D/B/A THE BENESSERE ) 14 CENTER, SUSAN C. JOHNSON, PH.D., ) BILLINGSLEY ENTERPRISES, INC., ) 15 F/K/A THAD H. BILLINGSLEY, M.D. ) CHARTERED, D/B/A THE BENESSERE ) 16 CENTER, ELI LILLY AND COMPANY, ) AND DISTA PRODUCTS COMPANY, ) 17 ) DEFENDANTS. ) Page 7 1 * * * * * * * * * * 2 3 CAUSE NO. 93-04911-A 4 LINDA JILL WELCH, CARLINDA 5 WELCH REX, CONNAN ROSS WELCH AND CHAD MICHAEL WELCH, 6 INDIVIDUALLY AND AS SURVIVORS AND STATUTORY BENEFICIARIES 7 OF CARL EUGENE WELCH, DECEASED PLAINTIFFS 8 V. 9 ELI LILLY AND COMPANY, DISTA PRODUCTS COMPANY, NOE NEAVES, 10 M.D., AND MINITH-MEIER CLINIC, P.A. DEFENDANTS Page 8 1 THE DEPOSITION OF DR. ROBERT ZERBE TAKEN AT 2 THE UNIVERSITY HOTEL, ANN ARBOR, MICHIGAN, ON 3 AUGUST 9, 1994; SAID DEPOSITION TAKEN PURSUANT TO 4 NOTICE IN ACCORDANCE WITH THE RULES OF CIVIL 5 PROCEDURE. 6 * * * * * * * * * * 7 A P P E A R A N C E S 8 9 PAUL SMITH COUNSEL FOR PLAINTIFFS 10 745 CAMPBELL CENTER 2 8115 NORTH CENTRAL EXPRESSWAY 11 DALLAS, TEXAS 75206 12 STEPHEN M. LORE COUNSEL FOR ELI LILLY AND COMPANY AND DEPONENT 13 FREEMAN & HAWKINS 4000 ONE PEACHTREE CENTER 14 303 PEACHTREE STREET, N.E. ATLANTA, GEORGIA 30308-3243 15 CURTIS G. OLTMANS 16 ELI LILLY AND COMPANY LILLY CORPORATE CENTER 17 INDIANAPOLIS, INDIANA 46285 18 BARTON BROWN COUNSEL FOR DR. BILLINGSLEY 19 WALLACE, SAUNDERS, AUSTIN, BROWN & ENOCHS 10111 W. 87TH ST. 20 P.O. BOX 12290 OVERLAND PARK, KANSAS 66282 21 Page 9 1 I N D E X 2 3 DEPOSITION OF DR. ROBERT ZERBE 4 5 CONTINUING EXAMINATION BY MR. SMITH 11 6 7 8 CERTIFICATE 275 9 ERRATA 276 10 EXHIBITS 11 PLAINTIFFS' EXHIBIT NO. 11................35 PLAINTIFFS' EXHIBIT NO. 12................47 12 PLAINTIFFS' EXHIBIT NO. 13................52 PLAINTIFFS' EXHIBIT NO. 14................64 13 PLAINTIFFS' EXHIBIT NO. 15................74 PLAINTIFFS' EXHIBIT NO. 16................88 14 PLAINTIFFS' EXHIBIT NO. 17...............120 PLAINTIFFS' EXHIBIT NO. 18...............148 15 PLAINTIFFS' EXHIBIT NO. 19...............155 PLAINTIFFS' EXHIBIT NO. 20...............162 16 PLAINTIFFS' EXHIBIT NO. 21...............167 PLAINTIFFS' EXHIBIT NO. 22...............179 17 PLAINTIFFS' EXHIBIT NO. 23...............185 PLAINTIFFS' EXHIBIT NO. 24...............211 18 PLAINTIFFS' EXHIBIT NO. 25...............218 PLAINTIFFS' EXHIBIT NO. 26 ..............221 19 PLAINTIFFS' EXHIBIT NO. 27...............225 PLAINTIFFS' EXHIBIT NO. 28...............228 20 PLAINTIFFS' EXHIBIT NO. 29...............240 PLAINTIFFS' EXHIBIT NO. 30...............249 21 PLAINTIFFS' EXHIBIT NO. 31...............259 PLAINTIFFS' EXHIBIT NO. 32...............271 Page 10 1 * * * * * * * * * * 2 3 CONTINUING EXAMINATION 4 BY MR. SMITH: 5 Q. Doctor Zerbe, we've met. In 6 fact, I was asking you questions when we 7 adjourned your deposition last time, correct? 8 A. Yes. 9 Q. You're still under oath, even 10 though we've reconvened your deposition and not 11 administered another oath; you're still under 12 oath from the previous deposition and your 13 answers can and will be used in evidence in the 14 trials of these cases. Do you understand that? 15 A. Yes. 16 Q. If I ask you any question that 17 you don't understand or would like for me to 18 repeat, please free to let me know. Would you do 19 that for me? 20 A. Yes. 21 Q. In going over an abstract of 22 your previous deposition, I was a little unsure 23 concerning what your duties were at specific 24 times. You began with Lilly in 1982, is that Page 11 1 right? 2 A. Yes. 3 Q. October of '82? 4 A. No, June. 5 Q. June, '82. And I believe you 6 started off as a clinical pharmacologist or in 7 the clinical pharmacology department there? 8 A. Yes. 9 Q. And you did that until when? 10 A. Approximately October of 1983. 11 Q. Then what was your next duty 12 after October, 1983? 13 A. I was director of the 14 neuroendocrine division. 15 Q. How long did you have that 16 title? 17 A. Until approximately June of 18 1985. 19 Q. Then in June, 1985 is that 20 when you moved to England? 21 A. Yes. 22 Q. And what was your job duty at 23 that time? 24 A. I was director for clinical Page 12 1 investigation in Europe. 2 Q. How long were you employed in 3 that capacity? 4 A. Until about March of 1987. 5 Q. Then you came -- did you come 6 back? 7 A. No. 8 Q. You remained in England? 9 A. Yes. 10 Q. All right. What was your job 11 title? 12 A. I was the managing director at 13 Lilly Research Center. 14 Q. In Erl Wood? 15 A. Yes. 16 Q. Was that a promotion, moving 17 from director of European clinical investigations 18 to becoming manager of European -- or the Lilly 19 Research Center -- 20 A. Yes. 21 Q. -- in Erl Wood? 22 A. Yes. 23 Q. How long did you have that 24 job? Page 13 1 A. Until January or February of 2 1989. 3 Q. How did your job duties differ 4 when you changed from director of European 5 clinical investigations in March, 1987 to become 6 the managing director of the Lilly Research 7 Center in Erl Wood? 8 A. As a director for clinical 9 investigation in Europe, I was responsible for 10 coordinating the clinical research activities of 11 the various affiliates. As the managing director 12 at Erl Wood, I was really responsible for the 13 management of the site, which included other 14 parts of research in addition to clinical -- I 15 was not actually responsible directly for the 16 clinical area at all, that reported back to 17 Indianapolis, and I had a responsibility for the 18 other activities at the Erl Wood site. So it was 19 really very different. 20 Q. So once you became the 21 managing director at the Lilly Research Center in 22 Erl Wood, you didn't have direct hands-on 23 management of European clinical trials? 24 A. That's correct. Page 14 1 Q. Then you moved back to the 2 United States in January or February of '89? 3 A. Yes. 4 Q. And became vice-president of 5 Lilly Research Labs? 6 A. No, I came back as an 7 executive director, Lilly Research Labs. 8 Q. Had Prozac, fluoxetine 9 hydrochloride been approved in any European 10 country by January, 1989 when you left Erl Wood? 11 A. Yes. 12 Q. Do you recall which countries? 13 A. I know that the first approval 14 in Europe was in Belgium, and it was well before 15 that. I can't say exactly which countries it was 16 approved at that time, but it had been approved 17 in Belgium. 18 Q. Was Prozac approved in Belgium 19 prior to it being approved in the United States 20 in late 1987? 21 A. I believe it was. 22 Q. Did the Belgium government 23 ever raise any questions concerning activation, 24 suicide potential or potential for violent Page 15 1 aggressive behavior in connection with Prozac? 2 A. I don't recall, I don't recall 3 that being an issue at all. 4 Q. Do you recall what the next 5 country was that approved Prozac after Belgium? 6 A. I really don't know what the 7 next country was. 8 Q. Do you recall any other 9 countries that approved Prozac prior to approval 10 by the United States Food and Drug Administration 11 in December, 1987? 12 A. I believe it was approved in 13 South Africa before it was approved in the U.S. 14 Q. Did Belgium and South Africa 15 require submission of package inserts by Lilly or 16 proposed package inserts by Lilly? 17 A. I don't recall the specific 18 content of those applications. 19 Q. I understand that, but 20 generally speaking, for those governments, would 21 they have made it a requirement that some 22 information be transmitted with the product or 23 some information about the product transmitted to 24 physicians in connection with the prescribing of Page 16 1 fluoxetine hydrochloride? 2 A. I think I understand your 3 question, but just to be certain could you 4 restate the question for me? 5 Q. Well, in most countries -- in 6 the United States, part of the approval process 7 requires that a sponsor of a medication present 8 to the regulatory body what the sponsor deems to 9 be the appropriate prescribing information 10 describing the product as far as the chemical 11 properties, the clinical pharmacology, the 12 actions of the drug, contraindications, side 13 effects, things of that nature. Do you recall 14 whether or not that was required in Belgium or 15 South Africa? 16 A. Again, I can't be certain, I 17 don't know -- I can't, you know, recite the 18 regulations, but it would be my assumption that 19 that would have been required in the application. 20 Q. Do you know of any country 21 that does not require that the sponsor provide 22 information concerning adverse events that occur 23 during the clinical trial to be described in the 24 prescribing information concerning Prozac? Page 17 1 A. I would think that any country 2 would require some description of adverse events. 3 Not all require or expect exactly the same 4 description of adverse events, if you understand 5 what I'm saying. 6 Q. Well, when the United States 7 government approved Prozac for marketing in the 8 United States, there was a specific part of the 9 labeling or prescribing information that listed 10 the various adverse events that had occurred 11 during the clinical trial phase of the product. 12 Is that your understanding? 13 A. Yes. 14 Q. And I'm wondering if other 15 countries in Europe, for instance, require the 16 same thing generally? 17 A. Again, speaking in general 18 terms, they would require some description of 19 adverse events. However, they may choose to 20 include some things that are different or choose 21 to exclude some things that are different than -- 22 in one country to another. 23 Q. Were you involved in the 24 registration process of Prozac in Belgium? Page 18 1 A. Not directly. 2 Q. Were you indirectly, since 3 during that period of time, from June '85 to 4 March '87 you were the director of European 5 clinical investigations? 6 A. I don't recall any direct 7 involvement in that process. I believe the 8 filing was probably before I went to Europe. 9 Q. In the European countries 10 while you were the director of European clinical 11 investigations, were the results of your European 12 investigations provided to the various European 13 regulatory bodies that were -- that had Prozac 14 under consideration? 15 A. Would you repeat the question? 16 Q. I assume that there were a 17 number of clinical trials ongoing in Europe 18 during the period of time that you were director 19 of European clinical investigations, is that 20 correct? 21 A. Yes. 22 Q. And that was directly your 23 job, was to oversee that clinical trial process, 24 correct? Page 19 1 A. Oversee may be a bit of an 2 overstatement, it was really more coordinating. 3 The general affiliates were directly responsible 4 for those trials, all we insured was that the 5 objectives of the trials were, you know, 6 complimentary, that it really blended into a 7 European program, that we really understood the 8 drug. But we didn't directly oversee very many 9 of those trials. 10 Q. The data generated by those 11 trials were being transmitted to you in some 12 fashion, either from Indianapolis to you or 13 directly to you from the European affiliates? 14 A. The data -- yes, eventually we 15 saw reports from the trials, but many times we 16 did not manage the data directly, those were 17 managed by the affiliates. 18 Q. All right. And then would 19 those affiliates transmit that data to 20 Indianapolis? 21 A. No, not always. Generally 22 speaking, the trials of the type that you're 23 describing done in that period of time would have 24 more likely been managed at the affiliate, and Page 20 1 Indianapolis would receive not raw data -- I'm 2 speaking generally, would generally not receive 3 raw data, but would receive a report or a summary 4 of the information. 5 Q. All right. Would that data, 6 the report, be sent to you also in some form? 7 A. I would say generally, yes, 8 but I don't think there was any standard 9 operating procedure which would have required it 10 as part of the process. 11 Q. My question is, I guess, what 12 was Lilly doing to ensure that the European 13 clinical trial data was being transmitted to the 14 various regulatory bodies in Europe? 15 A. Well, of course the most 16 important aspect from the review process 17 standpoint was to ensure that affiliates 18 understood adverse events -- or that the 19 regulatory agencies that those affiliates 20 represented were aware of new adverse events that 21 might emerge, and that was handled primarily 22 through the Drug Experience Network system, the 23 DEN system. 24 Q. All right. When was the DEN Page 21 1 system operational? 2 A. Well, in one form or another 3 the process for the Drug Experience Network -- 4 and again, it had different names at different 5 times, I believe, but it was in existence 6 virtually the whole time I was around, from at 7 least 1983 on, the DEN system was operative. I'm 8 not sure it was exactly in that form, but it was 9 the process for collection and dissemination of 10 adverse events was around about that time. 11 Q. The DEN system was collecting 12 adverse events, is that correct? 13 A. That's correct. 14 Q. Is that all that the DEN 15 system does is collect adverse events that occur 16 either in the clinical trial process or in 17 post-marketing if there is -- if the product is 18 indeed being marketed? 19 A. That's correct, it's only 20 adverse events. 21 Q. We've heard the term global 22 tracking system or global tracking network. Do 23 you understand what I'm talking about? 24 A. I believe so, yes. Page 22 1 Q. It's my understanding that 2 that is another computer data base, another 3 method that Lilly uses to monitor clinical trials 4 throughout the world, is that right? 5 A. I'm not sure exactly what you 6 mean by monitor, it's a system that is designed 7 to understand what clinical trials are going on. 8 Q. All right. What is that 9 called, is it global tracking network or system? 10 A. Global project tracking. 11 Q. When did that system become 12 operational? 13 A. My recollection was that the 14 system, at least the framework of the system was 15 designed, I don't know if operational is the 16 right word, but it was designed probably in about 17 1980 -- I would say even '84, perhaps. But I 18 make the distinction that it was designed because 19 it really wasn't very operational until really 20 much later. 21 Q. How much later? 22 A. I would say at least the late 23 '80s. And again, because I wasn't in 24 Indianapolis over that period of time, it's hard Page 23 1 for me to assess, but it was very difficult to 2 get the affiliates to comply all the time with 3 it. It's not a regulatory requirement, it's a 4 managerial requirement, and it was hard to have 5 consistent data in GPT. 6 Q. All right. Would you say that 7 by the time you left Erl Wood in January or 8 February, 1989 that the global project tracking 9 system was functioning pretty well? 10 A. I guess I -- maybe even a bit 11 later than that, 19 -- maybe 1990, '91. 12 Q. Now to be sure I understand, 13 the global project tracking system was designed 14 exclusively to monitor Lilly clinical trials, is 15 that correct? 16 A. I'm not sure what you mean by 17 exclusively. And it gets very complicated when 18 you talk about trials done by other companies 19 with our compounds, et cetera, and -- so, exactly 20 what do you mean? 21 Q. Let's talk about just Lilly 22 clinical trials, all right. Now, I understand 23 the DEN system, the DEN system is designed to 24 collect and collate adverse events, correct? Page 24 1 A. Correct. 2 Q. But the global project 3 tracking system is designed to monitor the status 4 of various Lilly clinical trials, correct? 5 A. Correct. 6 Q. In other words, I would assume 7 that by the time it's fully operational, in 1991, 8 for instance, you can punch into that and 9 determine whether or not there's a clinical trial 10 going on in, say, the United Kingdom, is that 11 right, a Lilly clinical trial? 12 A. You can get a listing of 13 clinical trials in England. 14 Q. All right. Now will that 15 system give you the status of those trials in 16 England? 17 A. I think the validity of the 18 data in GPT for that sort of thing was suspect. 19 Q. Why? 20 A. Pardon? 21 Q. Why was it suspect? 22 A. I think because it was only as 23 good as the discipline that the affiliates had in 24 updating information, and that didn't always Page 25 1 occur. 2 Q. Well, were there a set of 3 guidelines and instructions given to the 4 affiliates on how and when to use the global 5 project tracking system? 6 A. I believe there were written 7 guidelines. 8 Q. Was it expected that the 9 affiliates would follow those guidelines? 10 A. It was expected. However, it 11 was not -- as I mentioned, it was not a 12 regulatory requirement, so the diligence may not 13 have been the same on the part of the affiliates 14 as it was on other aspects. 15 Q. Well, obviously Lilly has gone 16 to a great deal of time and expense in setting up 17 this system, correct? 18 A. Correct. 19 Q. It was something that was 20 deemed by management at Lilly that would be 21 beneficial and necessary in the conduct of their 22 business worldwide? 23 A. It was deemed to be valuable, 24 of value, yes. Page 26 1 Q. And management at least 2 intended that since they were expending the time 3 and money to set this system up, that it would 4 become operational and that the affiliates and 5 Lilly employees throughout the world would enter 6 information required into that system, correct? 7 A. I think the goal was to get 8 compliance with GPT, yes. 9 Q. Do you know of any instance 10 where anybody in management at Lilly relaxed any 11 requirements in connection with reporting to the 12 global project tracking system? 13 A. I can't describe any specific 14 instances. 15 Q. Were there complaints 16 generally that the affiliates were not utilizing 17 the global project tracking system as had been 18 intended? 19 A. I'm sorry, I didn't understand 20 the question. 21 Q. Were there complaints 22 generally by management at Lilly that you heard 23 concerning the fact that any particular affiliate 24 or worldwide employee of Lilly wasn't doing what Page 27 1 they were supposed to do in connection with 2 inputting into this global project tracking 3 system so that somebody at Lilly in Indianapolis 4 would know what was going on with the clinical 5 trials worldwide? 6 A. I think there were periodic 7 police efforts to reemphasize the compliance with 8 the GPT expectations, but it was not as friendly 9 a process, it was not as centrally coordinated a 10 process until the late '80s or '90s to really 11 allow it to be done in any really reliable way. 12 It was a good concept, it was not necessarily 13 required for regulatory purposes worldwide, it 14 was a management tool, and it didn't always get 15 the priority that other things like DEN got, for 16 example. 17 Q. By the time you left Lilly in 18 June, 1990 -- a little over a year ago -- 19 A. Uh-huh. 20 Q. -- were you confident in any 21 data that you received from the global project 22 tracking system in connection with its accuracy 23 or completeness? 24 A. I was personally confident in Page 28 1 the listing of the studies, particularly the U.S. 2 studies. I was much less confident in the 3 numbers related to the status of the studies. 4 Q. All right. For instance, in 5 May of 1993, before you left Lilly, if you wanted 6 to know how many patients were enrolled in a 7 study in France, for instance, and you punched it 8 up into the global project tracking system and 9 got a number, are you telling me that you would 10 wonder whether or not that really reflected the 11 number of individuals that had perhaps been 12 enrolled in a particular clinical trial? 13 A. That's what -- exactly what 14 I'm saying, that is that I wasn't confident in 15 the individual numbers related to enrollment? 16 Q. But as far as the fact that a 17 trial was being conducted in a particular 18 location at a particular time, you were 19 relatively confident with respect to that 20 information? 21 A. Particularly in the U.S. 22 Somewhat less confident that everything was in 23 there internationally, although much more so in 24 the 1990s than in the mid-1980s. Page 29 1 Q. Who was responsible for the 2 global project tracking system, say, in 1993 when 3 you left? 4 A. I believe -- I would be 5 speculating on the name, I'd probably best not 6 speculate on the name. 7 Q. Who is the last individual 8 that you are reasonably sure was responsible for 9 the system? 10 A. I believe -- I believe that 11 Linda Cummings, but I'm not sure about that. 12 Q. In what division or section 13 would she have been in? 14 A. It was an unusual 15 responsibility, and I'm not sure exactly how she 16 reported, whether she reported to regulatory and 17 Max Talbott or reported within the CRA division, 18 one of the CRA divisions, the groups of CRAs. 19 Q. CRAs would have been under 20 medical, would they not? 21 A. Yes. 22 Q. And regulatory would not 23 necessarily have been under medical? 24 A. Not necessarily. Page 30 1 Q. Was it the intent, as far as 2 you know, of Lilly that they, in submitting 3 applications for registration of Prozac in a 4 particular country, that they would submit all 5 data to that particular country that had been 6 submitted to the United States Food and Drug 7 Administration? 8 A. That's a very general 9 question. Generally the applications were 10 tailored to the requirements of those individual 11 countries. So it is possible there would be 12 differences in the types of data submitted, but 13 that was dictated by the regulators. 14 Q. All right. In the United 15 States, in order to secure approval of Prozac in 16 the United States, Lilly had to submit to the 17 Food and Drug Administration data supporting 18 efficacy and safety of the product, correct? 19 A. Yes. 20 Q. And to do that, Lilly did 21 clinical trials? 22 A. Yes. 23 Q. And there were certain of 24 those clinical trials that were deemed either by Page 31 1 Lilly or by the Food and Drug Administration as 2 pivotal trials? 3 A. Yes. 4 Q. And those pivotal trials were 5 trials which Lilly and/or the Food and Drug 6 Administration considered as having scientific 7 validity or following the appropriate scientific 8 method in order to establish efficacy and safety 9 of a product, correct? 10 A. Yes. 11 Q. So you knew that, as far as 12 you were concerned -- and when I say you, I mean 13 Lilly -- when that data was submitted to the Food 14 and Drug Administration, I believe in September 15 1983 was when the IND/NDA was submitted, that 16 Lilly had what they felt was sufficient data to 17 support sufficient -- to support efficacy and 18 safety, correct? 19 A. Yes. 20 Q. Now, my question is, would 21 Lilly have in all instances, as far as you know, 22 submitted the pivotal trial data that was 23 submitted to the FDA to the other countries in 24 which it was requesting regulatory approval? Page 32 1 A. As far as I know, the same 2 data were used for those applications. 3 Q. Various regulatory agencies 4 might have required some additional studies or 5 may not have looked as closely at some of the 6 U.S. pivotal trials, but that data was submitted 7 regardless? 8 A. That would be my 9 understanding, yes. 10 Q. For instance, we know in 11 Germany, the German regulatory agency, the BGA 12 wanted and requested some additional European 13 inpatient clinical trial data, correct -- 14 A. Yes. 15 Q. -- do you recall that? So 16 that was a particular requirement in addition to 17 United States regulatory requirements that was 18 requested by a foreign regulatory body, correct? 19 A. Yes. 20 Q. But the BGA had the pivotal 21 trial data that was submitted to the FDA as far 22 as you know? 23 A. Yes. 24 Q. And that was generally the Page 33 1 practice in submitting data to any foreign 2 country in connection with the registration 3 process, to submit at least the United States 4 pivotal trial, clinical trial data? 5 A. Are you talking about for all 6 drugs? 7 Q. For Prozac. 8 A. For Prozac. Because of the 9 timing, the availability of the U.S. data, I 10 would assume that information was submitted. 11 Q. You've never seen any 12 information that the U.S. pivotal clinical trial 13 data was not submitted to any foreign regulatory 14 body? 15 A. I never saw any information, 16 any information that would suggest it wasn't 17 submitted. 18 Q. All right. And you generally 19 are of the opinion that as far as you know all 20 U.S. pivotal trial data was generally submitted 21 in all countries? 22 A. As far as I know, although it 23 may have been presented in a somewhat different 24 way in terms of format, emphasis, et cetera. Page 34 1 Q. But the data would be 2 presented? 3 A. In one form or another, yes. 4 Q. Well, there wasn't any 5 clinical trial data done in the U.S. pivotal 6 trials that wasn't submitted by virtue of putting 7 it in a different form to a foreign regulatory 8 body? 9 A. No. The only point that I was 10 going to make was that some aspects of it, 11 because it's tailored to the regulatory agency, 12 may be deemphasized at their request, that they 13 may be interested in a certain type of study. 14 Q. Give me an example of 15 something you recall where a particular study or 16 body of data was deemphasized to some foreign 17 regulatory body. 18 A. I don't know of any specific 19 instances, I was just trying to make the point 20 that regulators may look at the same data 21 presented a slightly different way and reach 22 somewhat different conclusions. 23 (PLAINTIFFS' EXHIBIT NO. 11 WAS 24 MARKED FOR IDENTIFICATION AND Page 35 1 RECEIVED IN EVIDENCE.) 2 Q. Doctor Zerbe, you're free to 3 look at all of Exhibit 11, however I'm just going 4 to direct my questions to the first page which is 5 the typewritten portion. 6 (Witness reviews document) 7 Q. Doctor Zerbe, Exhibit 11 is a 8 document that was prepared at Lilly by either 9 Doctor Heiligenstein or Beasley. Do you recall 10 this document? 11 A. I don't. 12 Q. Do you recall an examination 13 at Lilly of the question concerning whether or 14 not fluoxetine, Prozac, was an activating or 15 sedating drug? 16 A. There were a whole series of 17 continuing discussions about the profile of the 18 drug, including this issue. 19 Q. Of whether or not Prozac was 20 activating versus sedating? 21 A. Well, the whole pharmacologic 22 profile, yes, I mean specifically issues such as 23 this, the characteristics of activation or 24 sedation, yes. Page 36 1 Q. And this appears to be an 2 analysis of clinical trial data designed to look 3 at the issue, does it not? 4 A. Yes. 5 Q. And for your education, this 6 was a document that was done at Lilly in 7 connection with the Lilly clinical trials on 8 Prozac, and I just don't recall whether it was 9 Heiligenstein or Beasley that did this. Will you 10 accept my word for that? 11 A. Yes. 12 Q. Exhibit 11 in connection with 13 this analysis says fluoxetine may produce both 14 activation, nervousness, anxiety, agitation, 15 insomnia, and sedation, somnolence, asthenia, 16 correct? 17 A. In general, that's the tone. 18 I don't know whether you're referring to some 19 specific statement. 20 Q. I'm reading the first 21 sentence. 22 A. The first sentence, okay. 23 Q. Did I read that correctly? 24 A. Yes. Page 37 1 Q. The first paragraph goes on to 2 say approximately nineteen percent of patients 3 might be expected to report activation during 4 acute therapy with fluoxetine which was not 5 present prior to therapy, and which could be 6 attributed to fluoxetine, correct? 7 A. There's an additional, I 8 think, in trials or something like that at the 9 end, yes. 10 Q. I'm going to read that in just 11 a second. 12 A. Okay, I couldn't tell if that 13 was part of the next sentence or not. 14 Q. It's a little hard to read, 15 but it looks like it says next, in trials, 16 thirty-eight percent of fluoxetine treated 17 patients reported new activation, but nineteen 18 percent of placebo treated patients also reported 19 new activation, yielding a difference of nineteen 20 percent attributable to fluoxetine, correct? 21 A. Yes, sir. 22 Q. And in your review of the 23 underlying figures supporting this, does that 24 appear correct? Page 38 1 A. I can't say exactly without 2 studying it exactly where those specific figures 3 came from, but in general, that's the tone. 4 Q. You don't have any reason to 5 question the accuracy of the statements I just 6 read you? 7 A. Again, I can't find the 8 specific numbers, but I have no reason to 9 question that. 10 Q. And that pretty well reflects 11 what you knew about Prozac at the time you left 12 Lilly in July -- in June of 1993, that Prozac was 13 more activating than those patients who were 14 taking placebo? 15 A. Was that a question? 16 Q. Yes. 17 A. Could you restate it? 18 Q. Does that reflect. 19 A. Does that reflect. I think 20 the way that I read this document, I think it's 21 probably one of many considerations of this issue 22 and probably doesn't include the whole data base 23 with the numbers that are included here, I don't 24 know what the source of the information was, and Page 39 1 is probably one analysis. I think you can 2 interpret this specific report in that way, yes. 3 Q. Have you ever seen any 4 analysis of the fluoxetine clinical trial data 5 that concludes that in fact Prozac is sedating 6 and not activating? 7 A. Well, I think this suggests 8 that in different patients it can be activating 9 and in other patients it can be sedating. 10 Q. But generally speaking, if 11 you're going to see activation side effects of 12 nervousness, anxiety, agitation and insomnia, 13 you're more likely to see those type of side 14 effects in Prozac treated patients than you are 15 going to see with respect to sedating side 16 effects, which is somnolence and asthenia, 17 correct? 18 A. Well, as I read the numbers -- 19 and forgive me, one would have to study this much 20 more carefully, but as I read the numbers, it's 21 something like thirty-eight percent with 22 activation, twenty-eight percent with sedation. 23 Those are not markedly different numbers if 24 that's the implication of your question. I know Page 40 1 there's a statistical way to look at that, I 2 assume there is, and it may actually be in here, 3 I don't know, but those aren't markedly different 4 numbers. 5 Q. If you don't agree, you're 6 going to be the first individual that's ever been 7 associated with Eli Lilly and Company that hasn't 8 said that in general Prozac is more activating 9 than sedating. Is that your testimony here? 10 A. Well -- 11 MR. LORE: No, that's not what he 12 said, Paul. Go ahead and answer. 13 Q. Then say it's not my 14 testimony. 15 MR. LORE: Go head and answer if you 16 can, Doctor. 17 A. I was really responding to 18 these specific numbers, and the way I interpreted 19 your question, you said based on the data I 20 thought anyway, based on the data are you 21 concluding, and I'm just saying I can't conclude 22 based on the thirty-eight/twenty-eight percent 23 that those are markedly different numbers. What 24 I said was it has both activating properties and Page 41 1 sedating properties in different individuals and 2 I think that the data supports that point. 3 Q. But in taking the individuals 4 as a whole, it's more likely that it's going to 5 be activating than sedating, correct? 6 A. Numerically, yes. 7 Q. Well, statistically. 8 A. Well, I just haven't seen the 9 statistics, I don't know. I mean maybe it's in 10 here, I just haven't had a chance, Paul, to look 11 at those statistics. 12 MR. LORE: Paul, are you directing 13 your questions to this document that you placed 14 before him? 15 Q. Either this document or what 16 you know about the issue in general from your -- 17 A. I'm just trying to think 18 statistically the approach that one would take to 19 reach the conclusion that statistically it's more 20 frequent -- there's more frequent activation than 21 sedation. Maybe that's been done. Numerically, 22 I think your statement is correct. 23 Q. All right. And that's what 24 you recall about the clinical trial experience in Page 42 1 connection with Prozac, is it not? 2 A. Yes. 3 Q. That there were more instances 4 of activation than there were of sedation? 5 A. That's a general recollection. 6 Q. All right. Read with me the 7 first two sentences of the third paragraph of 8 this document. It says fluoxetine is activating 9 relative to tricyclics and tricyclics are 10 sedating relative to fluoxetine. The difference 11 in activation actually attributed to the drugs, 12 nineteen percent versus four percent, is greater 13 than the difference in sedation, thirteen percent 14 versus twenty-three percent, correct? 15 A. That's what it says, yes. 16 Q. Do you agree with this 17 statement that fluoxetine is activating relative 18 to tricyclics? 19 A. Well, again, just looking at 20 the individual data to support the point, I was 21 looking for P values difference. I really did 22 not find that in the data, but maybe I missed it. 23 Q. Well, if you look with me at 24 the second sentence in the third paragraph, it Page 43 1 says the difference in activation actually 2 attributable to the drugs, nineteen percent 3 versus four percent, is greater than the 4 difference in sedation, thirteen percent versus 5 twenty-three percent. Obviously, there's a 6 statistical difference between nineteen percent 7 and four percent, is there not? 8 A. Well, not always. I mean, you 9 know, it depends on what the numbers are 10 supporting it, and that was the number I was 11 looking for. And I'm trying to find it, maybe 12 you can point me in the direction that 13 demonstrates that statistical difference, I don't 14 know, but I just could not find it in the tables 15 quickly. Do you know where it is in the tables? 16 Q. Well, the tables examine 17 agitation and nervousness, correct, and sedation, 18 during the clinical trials, also examines the 19 number of patients who discontinued the drug 20 because of these adverse events, does it not? 21 A. That's correct. I mean -- 22 Q. I don't see anything in the 23 tables that would indicate that the conclusion 24 given by one of the psychiatrists at Lilly, Page 44 1 Doctor Heiligenstein or Doctor Beasley, would be 2 inaccurate, where it says fluoxetine is 3 activating relative to tricyclics and tricyclics 4 are sedating relative to fluoxetine. Are you 5 saying you disagree with that statement? 6 A. Well, I would turn your 7 statement around. I think the burden of proof is 8 to support the statement, not the burden of proof 9 to dispute the statement, and I just don't see 10 those -- I don't see the statistical support for 11 that. If you can point that out to me, that's 12 fine, but I just don't see the statistical 13 support for that conclusion. Numerically, yes, 14 but statistically, no. 15 Q. It's a Lilly document done by 16 a Lilly scientist. 17 A. Sure. 18 Q. That's what I know coming up 19 with a conclusion. And if you see anything that 20 makes the conclusion drawn by the Lilly scientist 21 suspect, then I need to know. 22 A. Well, I'm searching for the 23 data in this document that supports that 24 statement statistically. Page 45 1 Q. All right. 2 A. And I just don't see that. 3 Q. You just don't see it? 4 A. I don't see anything 5 specifically that disputes it, but it may be more 6 a numerical than a statistical statement, if you 7 know what I mean. 8 Q. All right. Well, maybe at the 9 lunch break you can look at it a little closer 10 and give us some insight as to whether or not you 11 see any statistical basis for the statement made 12 by the Lilly scientist. So I'm clear, are you 13 saying that generally speaking Prozac is not an 14 activating drug versus a sedating drug? 15 A. That question could be 16 interpreted a number of ways. Would you maybe 17 restate the question? 18 Q. Well, it's my understanding 19 that the clinical trial data itself established 20 that there were more patients who reported 21 activating side effects than sedating side 22 effects, is that not true? 23 A. Numerically I believe that's 24 correct, yes. Page 46 1 Q. And that in comparing Prozac 2 with other tricyclic antidepressants, that there 3 were more Prozac patients that reported 4 activating side effects than those patients 5 taking tricyclics, is that correct? 6 A. Again, I think numerically 7 that's correct. 8 Q. And in connection with 9 tricyclic antidepressants versus Prozac, there 10 were more individuals reporting sedating side 11 effects who were taking tricyclics than those 12 taking Prozac, is that correct? 13 A. Again, numerically I think 14 that's what the data says, yes. 15 Q. And generally that's true with 16 respect to the properties of this drug, isn't it? 17 A. I think in general, yes. 18 (PLAINTIFFS' EXHIBIT NO. 12 WAS 19 MARKED FOR IDENTIFICATION AND 20 RECEIVED IN EVIDENCE.). 21 Q. Exhibit 12 has general 22 comments and specific comments. You're free to 23 read the specific comments, but my questions are 24 only going to be directed to the general comments Page 47 1 section of Doctor Cohen's analysis. 2 A. Okay. 3 Q. Feel free though to look 4 through the specific comments if you like, I want 5 you to be comfortable with what you're seeing. 6 A. It's hard to associate the 7 document because I don't know what they're 8 referring to, so it probably doesn't do a lot of 9 good to look through in detail the specific 10 comments. 11 Q. Right. Exhibit 12 is a 12 document authored by Doctor Marlene L. Cohen, 13 correct? 14 A. Yes. 15 Q. And she's a Lilly scientist, 16 is she not? 17 A. Yes. 18 Q. Do you personally know Doctor 19 Cohen? 20 A. Yes. 21 Q. Do you respect her judgment as 22 a scientist? 23 A. Yes. 24 Q. She is an individual with Page 48 1 specific expertise in brain 2 neuropsychopharmacology, is she not? 3 A. Actually, I would not have 4 considered Marlene an expert as much in CNS 5 pharmacology as cardiovascular pharmacology, 6 that's my recollection. Marlene was much more 7 involved in cardiovascular, as I recall. 8 Q. It was Doctor Leigh Thompson's 9 and, I believe, Doctor Mel Perelman's comments 10 that Doctor Cohen had specific knowledge 11 concerning CNS neurotransmission. 12 A. Well, again, I wouldn't 13 dispute their judgment, per se, but that's not 14 what -- I recall more her interest in serotonin 15 and cardiovascular. And I don't believe she was 16 in the CNS division, but I could be wrong, CNS 17 pre-clinical area, but maybe they're right and 18 I'm wrong, I don't know. 19 Q. But at least she is an expert 20 in serotonin? 21 A. Yes, she's worked a lot in 22 serotonin. 23 Q. And it appears in Exhibit 12 24 she has made a review of the fluoxetine safety Page 49 1 update, correct? 2 A. Yes. 3 Q. And the safety update is what? 4 A. Well, periodically new safety 5 information is provided. I presume this is a 6 formal safety update, I mean that could be any 7 number of different documents, but the one you're 8 referring to would be a regulatory document 9 updating the safety -- new information generated 10 by the safety of the compound. 11 Q. Well, that's the -- the safety 12 update is a sort of term of art meaning that it's 13 something that's going to be submitted to a 14 regulatory agency generally, isn't that correct? 15 A. That's the way I would 16 interpret it. But safety update is such a 17 general term, it's tough to say exactly what she 18 was referring to, but that would be my 19 assumption. 20 Q. That it was something that was 21 going to be submitted to the Food and Drug 22 Administration. 23 A. That would be my assumption, 24 yes. Page 50 1 Q. Look with me at point 8 of 2 Doctor Cohen's review of the safety data. She 3 says there, in point 8, quote, CNS stimulation, 4 agitation and insomnia: From the data taken in 5 concert, it might appear that there is a somewhat 6 greater incidence of CNS stimulation with 7 fluoxetine as compared with placebo. For 8 example, a greater incidence of insomnia and 9 anxiety were reported, paren, 9-95, 9-113, 9-370, 10 agitation seems pronounced, 9-160, and a decrease 11 in REM sleep was reported, and there may be a 12 suggestion of manic psychosis, 9-107, 9-111. As 13 with other antidepressant agents used in the 14 depressed population, mania and/or CNS 15 stimulation may be a minor component of the 16 actions of fluoxetine, end quote. Correct? 17 A. Well, it didn't say and/or, 18 but it said or mania or CNS stimulation. 19 Q. Okay. I couldn't read it 20 because of that stamp, that's the reason I said 21 and/or. Can you see it, is that word or? 22 A. To me it clearly says or. 23 Q. Mania or CNS stimulation may 24 be a minor component of the actions of Page 51 1 fluoxetine, correct? 2 A. Yes. 3 Q. Did you see this document -- 4 A. I don't recall seeing it. 5 Q. -- in your employment with 6 Lilly? 7 A. I don't recall seeing it. 8 Q. This at least in part supports 9 the assertions set forth in Exhibit 11 in 10 connection with the agitation, activation versus 11 sedation aspects, does it not, or fits in with 12 it? 13 A. It's compatible with that 14 assessment, yes. 15 Q. Compatible is a good word. 16 (PLAINTIFFS' EXHIBIT NO. 13 WAS 17 MARKED FOR IDENTIFICATION AND 18 RECEIVED IN EVIDENCE.) 19 Q. Exhibit Number 13 appears to 20 be a document authored by you dated February 7, 21 1990, is that correct, Doctor? 22 A. Yes. 23 Q. And you were writing to 24 apparently a physician whose name has been Page 52 1 redacted, correct? 2 A. Yes. 3 Q. Do you recall authoring this 4 letter? 5 A. In general, yes. 6 Q. Is that your signature -- 7 A. Yes, it is. 8 Q. -- on page two of the exhibit? 9 A. Yes, it is. 10 Q. Can you detail for me the 11 circumstances concerning your writing this 12 letter? 13 A. Again, it's been some time 14 ago, and I really can only describe in general, 15 but I believe it's related to the Wesbecker case. 16 Q. Wesbecker? 17 A. Yes, Wesbecker case, and a 18 query from a physician involved in the case about 19 our information on violent behavior and 20 fluoxetine. 21 Q. You say a query by a physician 22 involved in the case. Which physician are you 23 speaking of? 24 A. I don't recall the name of the Page 53 1 physician offhand, and I couldn't be certain -- 2 well, I couldn't be certain exactly who it was. 3 Q. Well, Doctor Lee Coleman was 4 Mr. Wesbecker's psychiatrist. Does that name 5 ring a bell to you as being potentially -- 6 A. I don't believe it was Doctor 7 Coleman. The name is vaguely familiar, but I 8 don't believe that's who the letter was to. 9 Q. But it was -- your 10 recollection is this Exhibit 13 letter dated 11 February 7, 1990 is a letter by you to a 12 physician involved in the Wesbecker case, is that 13 correct? 14 A. That's my recollection. 15 MR. SMITH: Do you know who it is, 16 Steve? 17 MR. LORE: I do not. 18 MR. SMITH: Can you get that 19 information for me, maybe call somebody over the 20 lunch hour? 21 MR. LORE: Okay. 22 MR. SMITH: Because I'm obviously 23 entitled to know. 24 MR. LORE: If it is in fact in that Page 54 1 case, if it is a treating physician, physician 2 involved in that case. 3 MR. SMITH: He said it was. 4 MR. LORE: That's who he believes it 5 is. 6 A. I meant not necessarily a 7 treating physician, but somebody involved in the 8 legalities associated with the case, the legal 9 case. 10 Q. Well, you think this might 11 have been an expert? 12 A. I don't know if it was an 13 expert or a medical examiner or somebody like 14 that requesting information. 15 MR. LORE: I'll check, Paul. 16 MR. SMITH: All right. 17 Q. In any event, one of the 18 physicians -- this physician that was involved in 19 some way in the Wesbecker case had written to 20 Lilly or you had talked with him on February -- 21 no, there would have been a telephone 22 conversation on February 5th, is that right? 23 A. That's what is indicated here, 24 yes. Page 55 1 Q. And you're sending him some 2 additional thoughts and comments and data in 3 connection with violence and Prozac, correct? 4 A. That's correct. 5 Q. Do you recall speaking with 6 this physician along with Doctors Masica and 7 Heiligenstein? 8 A. Again, really only very 9 general, I don't recall the details of exact -- 10 you know, the exact details, but yes. 11 Q. Tell me what details you do 12 recall. 13 A. My recollection of this is 14 that this was a call to the medical examiner, and 15 this is the letter following that conversation 16 where we offered -- and to be honest with you, I 17 don't remember whether he initiated the query or 18 whether we offered to provide any background 19 information regarding the event. 20 Q. All right. There's been some 21 testimony that there were some conversations with 22 the coroner, and I assume that the coroner and 23 the medical examiner are going to be one and the 24 same? Page 56 1 A. That's what I was referring 2 to, yes. 3 Q. And there was speaker 4 conversations between the coroner in Louisville 5 and Doctor Leigh Thompson, yourself, and Masica 6 and Heiligenstein. Do you recall that? 7 A. I honestly don't remember 8 Doctor Thompson being involved in those 9 conversations, but I could be wrong. 10 Q. In any event, he was raising 11 the question concerning whether or not there had 12 been reports to Lilly of violence in connection 13 with Prozac previously, is that right? 14 A. That's right. As I 15 reconstruct it, having sort of read through this, 16 I think he received a package of information from 17 the Church of Scientology and he was checking 18 whether what they said was correct and so forth, 19 and I think we're responding to that as providing 20 the information that was the background to that 21 specific -- those specific cases. 22 Q. And apparently that CCHR 23 document had twenty-nine reports of violent 24 aggressive behavior that were reported by Lilly Page 57 1 to the FDA prior to February 7, 1990, is that 2 right? 3 A. That's what is indicated here. 4 Q. All right. Then you go on to 5 say in the fifth paragraph, as you might imagine, 6 violent behavior can be characterized in a 7 variety of ways. Such characterizations include: 8 Aggressive behaviors; self destructive acts, 9 including suicide; and physical violence directed 10 toward others. Correct? 11 A. That's what is said, yes. 12 Q. And when you wrote this, did 13 you write this letter in coordination or 14 assistance with Doctors Masica and Heiligenstein? 15 A. Yes, I did. 16 Q. Because they are 17 psychiatrists? 18 A. That's correct. 19 Q. And you wanted their input in 20 how to characterize violent behavior? 21 A. What I think this is referring 22 to is the apparently high number of events that 23 were reported to the FDA, and the point made was 24 that it can be characterized in a wide variety of Page 58 1 ways. It doesn't mean that this is the ideal and 2 perfect way, but simply that that's the kind of 3 inclusive definition. 4 Q. Well, there you were including 5 aggressive behaviors and self destructive acts, 6 including suicide, right? 7 A. That's -- I'm not sure what 8 you mean by I was including. What I think this 9 statement says or what it was intended to say was 10 that violent acts can be, in different people's 11 mind, a whole variety of things, and these are 12 some of the examples. But I'm not suggesting 13 this is sort of the ideal definition, this is the 14 kind of all-inclusive term -- set of 15 all-inclusive terms for violence. 16 Q. So it's terms used by various 17 people at various times to categorize violent 18 aggressive behavior? 19 A. Yes. 20 Q. And you included 21 self-destructive acts, including suicide, as one 22 of the terms that are used by some to categorize 23 violent behavior. 24 A. That's right. Page 59 1 MR. SMITH: Let's take a quick break. 2 (A SHORT RECESS WAS TAKEN.) 3 Q. Did you ever go to Louisville 4 and speak with anybody in Louisville in 5 connection with the Wesbecker case? 6 A. No. 7 Q. Do you know if anybody from 8 Lilly did? 9 A. I don't recall anybody going 10 to Louisville to talk about the Wesbecker case. 11 Q. Do you recall if anybody spoke 12 by phone with Doctor Lee Coleman in connection 13 with the Wesbecker case? 14 A. Doctor Heiligenstein or Masica 15 may have, I just don't know. I don't know. 16 Q. Do you recall anything else 17 you did in connection with the Wesbecker case 18 other than author this letter and participate in 19 that -- 20 A. Phone conversation. 21 Q. -- phone conversation? 22 A. Again, it seemed like there 23 was more -- there was another phone conversation 24 besides this one, but I couldn't recall whether Page 60 1 it was before or after this or I couldn't be 2 certain there was another one. 3 Q. Do you recall, if there was 4 another one, who would have been present at that 5 phone conversation and who that conversation 6 would have been with? 7 A. It probably would have been 8 the same people, and the conversation would have 9 been with the coroner. 10 Q. Do you remember what specific 11 or generally what questions the coroner was 12 raising other than apparently as reflected by 13 your letter, his questions concerning whether or 14 not there had been reports of Prozac in 15 connection with violent aggressive behavior? 16 A. Again, I don't even remember 17 the specifics of a second conversation, and 18 that's purely speculation. But the only -- this 19 is really the extent of the conversation, as I 20 recall it, related to the Lilly experience. 21 Q. Did you ever do anything to 22 make any inquiry in connection with the Prozac 23 data, either the clinical trial data or the data 24 generated by the post-marketing spontaneous Page 61 1 reporting system, in connection with whether or 2 not Prozac might cause violent aggressive 3 behavior? 4 A. Whenever an issue like this 5 came up, our standard approach was to try to 6 understand everything we could about the specific 7 case and to understand everything in our data 8 base related to the type of event. And part of 9 my responsibility was to see that that process 10 was orchestrated, and Doctor Heiligenstein was, 11 as I recall, central in the actual data 12 evaluation of the question of violence. 13 Q. All right. Did you ever see 14 that data? 15 A. Only what Doctor Heiligenstein 16 reported in summaries that he had, things that he 17 had summarized. I didn't look at individual 18 cases. 19 Q. Did you look at his summary? 20 A. I think there was more than 21 one summary, but that would be -- routinely, I 22 would have looked at the summaries, yes. 23 Q. What is your recollection of 24 what those summaries included? Page 62 1 A. That there was no suggestion 2 that violence occurred more frequently with 3 fluoxetine than one would have expected in the 4 general population or even in comparative trials, 5 that the rates were the same or actually perhaps 6 lower. 7 Q. Well, did you look at -- do 8 you recall whether or not that summary looked at 9 aggressive behaviors? 10 A. My recollection was that 11 Doctor Heiligenstein's evaluation looked at the 12 whole spectrum of aggressive behaviors. 13 Q. Did that review include 14 self-destructive acts as you mentioned here in 15 your letter of February 17th? 16 A. That, I don't recall. I don't 17 recall whether that was included in his analysis 18 or not included. 19 Q. In fact, it wasn't, was it, 20 suicide wasn't included in his analysis? 21 A. I don't know. I think suicide 22 was a separate issue, that suicidal ideation and 23 suicide was a separate issue. That was being 24 evaluated, and I'm sure Doctor Heiligenstein had Page 63 1 very sound scientific reasons for not including 2 that if it wasn't. 3 Q. Do you know if he included 4 physical violence directed towards others? 5 A. Again, my recollection was 6 that he was looking at events that could be 7 characterized -- that were similar to those that 8 Mr. Wesbecker, Wesbecker demonstrated. 9 Q. But it didn't include suicide, 10 self-directed violence or self-destructive 11 behavior, did it? 12 A. Again, as best I can recall, 13 no, it didn't. 14 MR. LORE: Paul, before we leave this 15 document, this was to the medical examiner, 16 Exhibit 13 was from Doctor Zerbe to the medical 17 examiner, Doctor Greathouse. 18 (PLAINTIFFS' EXHIBIT NO. 14 WAS 19 MARKED FOR IDENTIFICATION AND 20 RECEIVED IN EVIDENCE.) 21 Q. Doctor Zerbe, Exhibit 14 22 appears to be minutes of a meeting that occurred 23 on the 18th of December, 1985, is that correct? 24 A. Yes. Page 64 1 Q. It says the meeting was at 2 DHSS. Can you tell us what DHSS stands for? 3 A. I think it's for the 4 Department of Health and Social Services. 5 Q. Is that a regulatory body with 6 some particular country? 7 A. UK. 8 Q. England? 9 A. Yes. 10 Q. And it says the subject of 11 that meeting is the fluoxetine PL application, is 12 that right? 13 A. Yes. 14 Q. What is PL application? 15 A. Product license. 16 Q. You were there from Lilly? 17 A. Yes. 18 Q. And there was a Doctor Wood 19 and Fawcett there from the Department of Health 20 and Social Services in England, is that right? 21 A. Yes. 22 Q. This apparently is a meeting 23 in connection with an application by Lilly to 24 register Prozac in England, correct? Page 65 1 A. Yes. 2 Q. And that had not been granted 3 at the time of this meeting, had it? 4 A. No. 5 Q. Did you author this document? 6 A. No, I didn't. 7 Q. Do you know who did? 8 A. I can't be certain, I don't 9 know exactly who authored it. 10 Q. It was somebody at Lilly, 11 though? 12 A. Yes, I believe so. 13 Q. And were you the senior 14 individual there from Lilly at this meeting? 15 A. No. 16 Q. Who was? 17 A. Doctor Gennery was my boss. 18 Q. Who is Doctor Gennery, what 19 was his title? 20 A. He was group medical director. 21 Q. For European affairs? 22 A. Yes. 23 Q. And you were the director of 24 European clinical investigations at the time? Page 66 1 A. That's right. 2 Q. When was Prozac finally 3 approved for use in England? 4 A. I don't recall the specific 5 date. 6 Q. Was the application for Prozac 7 to be marketed in England ever formally rejected? 8 A. I don't believe so. 9 Q. That application was withdrawn 10 for a period of time, was it not? 11 A. I believe that was the case. 12 Q. Because of concerns that the 13 Department of Health and Human Services in 14 England had raised? 15 A. That's right. 16 Q. Go with me to page two of this 17 document towards the middle of the page, it will 18 be the fifth full paragraph where it starts with 19 regard. Do you see that? 20 A. Yes. 21 Q. Apparently the DHS is saying, 22 quote, with regard to the animal data, fluoxetine 23 is perceived by the authorities to cause adverse 24 effects at much lower dosages than other drugs, Page 67 1 such as imipramine. In the long-term studies, 2 there were adverse effects which caused concern, 3 in particular, liver degenerative changes and CNS 4 toxicity such as aggression and convulsion, 5 correct? 6 A. Yes. I mean that's what that 7 statement says, yes. 8 Q. And in connection with the CNS 9 toxicity of aggression and convulsions, was there 10 anything done by Lilly with respect to the 11 concerns raised by the DHSS in England to 12 alleviate their concerns with respect to CNS 13 toxicity such as aggression and convulsions? 14 A. I don't remember the exact 15 timing, but the issue of aggression was fairly 16 extensively studied in animals, and I think maybe 17 even prior to this time there were data that 18 suggests just the opposite effects, that in fact 19 it was effective in decreasing aggression in 20 certain animal models. 21 Q. Are you aware of Doctor Stark, 22 Paul Stark -- I mean Doctor Irwin Slater's cat 23 study? 24 A. No, I'm not familiar with it. Page 68 1 Q. That found that his cats 2 became more aggressive while on fluoxetine. 3 A. I'm not aware of that study. 4 Q. And that the cats that were 5 formerly his friends, I think as he states it, 6 became so aggressive that they were growling and 7 hissing. 8 MR. LORE: And I'm not sure that he 9 mentioned that they were his friends, that 10 comment. 11 Q. They were at least the 12 handler's friends. 13 A. Well, I was not aware of those 14 studies. I don't know at this time whether the 15 data were available, but in some rodent models, 16 there were -- there was evidence that it actually 17 decreased aggression. So I think that -- I'm not 18 sure what that statement was based on, but that 19 was not the kind of general tone with regard to 20 aggression. 21 Q. Well, we see that Doctor Cohen 22 has found in her review of the safety update CNS 23 stimulation in instances of anxiety, agitation, 24 and even mania, don't we? Page 69 1 A. Well, I mean, again, I think 2 you may have over-simplified and generalized what 3 Doctor Cohen said, I think she states it with a 4 fair number of reservations in there. 5 Q. Why don't you read what Doctor 6 Cohen said in connection with that. 7 A. Okay. CNS stimulation, 8 agitation, and insomnia, colon, from the data 9 taken in concert, a qualifier, it might appear, a 10 qualifier, that there's a somewhat, a qualifier, 11 greater incidence of CNS stimulation with 12 fluoxetine as compared with placebo. That's what 13 I was referring to. I think Doctor Cohen is 14 almost describing a most conservative worst case 15 kind of analysis there. 16 Q. Does anything say there that 17 she's describing a most conservative worst case 18 analysis? 19 A. As I read it, and I don't 20 recall having read it before, but as I read it 21 for the first time, there are a lot of qualifiers 22 in the statement. 23 Q. There's a lot of citations to 24 specific instances where she found this Page 70 1 occurring, isn't there? 2 A. There are studies cited, but 3 no data cited. 4 Q. Well, she points to specific 5 volume and page, doesn't she? 6 A. I don't know what these 7 numbers are, I thought these were -- I thought 8 these were study numbers, but it may be volumes 9 and pages. 10 Q. They're specific volumes and 11 pages of the safety update itself, Doctor Zerbe. 12 A. Again, they may be citations, 13 but they aren't data cited here. 14 Q. It looks like she's being 15 pretty specific in pointing to those instances of 16 CNS stimulation, she's showing you where she sees 17 that, isn't she? 18 A. Again, that may be your 19 interpretation of what's here, I think she's just 20 taking a very conservative, critical, worst case 21 analysis as a good scientist would. 22 Q. How do you know, did you ever 23 talk to her about that? 24 A. No, I -- Page 71 1 Q. Did she ever use the term, 2 this is a conservative, critical, worst case 3 scenario? 4 A. I'm basing in my judgment 5 reading this paragraph. 6 Q. I'm talking to you. The words 7 say what they say, don't they? 8 A. Right. 9 Q. I'm talking to you. Did you 10 ever speak with her about this safety update? 11 A. I don't recall. 12 Q. Did you ever speak with her 13 concerning her views of the data reflected in the 14 safety update? 15 A. I don't recall such a 16 conversation. 17 Q. Paragraph eight there, and 18 nowhere in her analysis does she say this is the 19 most conservative, worst case scenario, does it? 20 A. Does she ever say that? 21 Q. Right. 22 A. No, she doesn't. 23 Q. And the qualifiers that she 24 puts in there are stated in and of themselves, Page 72 1 aren't they, she uses terms might and somewhat, 2 doesn't she? 3 A. She uses a number of 4 qualifiers in the statement. 5 Q. Might and somewhat are used 6 there, aren't they? 7 A. Yes, might, somewhat are used. 8 Q. All right. What she's saying 9 there is compatible with what the -- what's being 10 said concerning CNS toxicity by the English 11 government in December of 1985, isn't it? 12 A. Well, compatible, perhaps, but 13 I'm not sure -- this is even second-hand 14 information, and I don't recall exactly what was 15 said at that time. 16 Q. You were at the meeting, 17 weren't you? 18 A. I was at meeting, but I don't 19 recall specifically what was said. 20 Q. There's notes there for our 21 benefit that reflect what was said, isn't there? 22 A. But it's really a pretty small 23 part of that whole letter, perhaps taken out of 24 context a bit. But it is compatible if that's Page 73 1 your question. 2 Q. Whether it's taken out of 3 context or not, I read to you exactly what was 4 said by the notes, didn't I? 5 A. Yes. 6 Q. And I read to you, and you 7 have before you exactly what Doctor Cohen said 8 several years later when she made a review of the 9 safety update, correct? 10 A. Yes. 11 Q. And they are indeed 12 compatible, aren't they? 13 A. Yes. 14 (PLAINTIFFS' EXHIBIT NO. 15 WAS 15 MARKED FOR IDENTIFICATION AND 16 RECEIVED IN EVIDENCE.) 17 Q. Doctor Zerbe, Exhibit 15 18 appears to be a document dated June 3, 1986 and 19 signed by Doctor John R. Hall, correct? 20 A. Yes. 21 Q. You are shown as the 22 individual that received a copy of the document, 23 correct? 24 A. Yes. Page 74 1 Q. Apparently this document 2 concerns a meeting that occurred on May 27, 1986 3 at Erl Wood where an independent consultant was 4 called in to examine questions raised by the 5 British regulatory authorities, correct? 6 A. Yes. 7 Q. Specifically the issue was 8 whether or not Prozac was the cause or was in any 9 way related to phospholipidosis, is that right? 10 A. Phospholipidosis, yes. 11 Q. And phospholipidosis was a 12 matter that had been raised by several regulatory 13 bodies, had it not? 14 A. Yes. 15 Q. And the document here says in 16 the first paragraph that this professor was 17 invited to Erl Wood because he had experience 18 with the CSN, which is the British regulatory 19 body, is it not? 20 A. Yes. 21 Q. And had experience with all 22 European regulatory bodies and with the FDA, 23 correct? 24 A. Yes. Page 75 1 Q. Do you recall who this 2 professor was? 3 A. I'm sorry, Doctor Zbinden is 4 the professor, is the toxicologist, Professor 5 Zbinden. 6 Q. All right. 7 A. As I recall. He was the 8 expert, I don't know what name is marked out. 9 Q. Well, it's not marked out in 10 the present portion, but it is in the body of the 11 document, correct? 12 A. That's what it appears to be. 13 Q. There's another individual 14 marked out, is there not, under the present? 15 A. Yes. 16 Q. Do you know who that is? 17 A. I really don't. 18 Q. But you think when they're 19 speaking of professor, and the professor that was 20 invited to Erl Wood, they're speaking of 21 Professor Zbinden? 22 A. Yes. 23 Q. Where is he located now? 24 A. He's in Europe, I believe Page 76 1 Switzerland, but I'm not sure. 2 Q. You said he's the one in 3 regulatory that's had experience with all 4 European regulatory bodies and with the Food and 5 Drug Administration, is that right? 6 A. Well, again, I didn't write 7 the memo and I'm not sure what the background is 8 and what they're referring to, but Doctor Zbinden 9 is a known toxicologist. 10 Q. Must be a world renowned 11 toxicologist? 12 A. I think that's fair to say, 13 yes. 14 Q. There on the third page, the 15 entire -- the first two pages of the document 16 deals exclusively with phospholipidosis, is that 17 right? 18 A. I'm sorry, could you repeat 19 the question? 20 Q. The first two pages deal with 21 his comments concerning phospholipidosis, is that 22 right? 23 A. Yes. 24 Q. And then the overall Page 77 1 conclusions is recommendations concerning what 2 needs to be done, in part? 3 A. In part, yes. 4 Q. And under overall conclusions, 5 the second paragraph says we must emphasize that 6 the side effects of fluoxetine are related to its 7 pharmacology, i.e. CNS stimulation, et cetera, 8 but of course the structure of the drug is 9 amphiphilic, therefore we see PLP, which is 10 phospholipidosis, correct? 11 A. Yes. 12 Q. What does the word amphiphilic 13 mean? 14 A. Amphiphilic means -- I'm not 15 sure of the chemical nature, but basically it has 16 both soluability in water and fats. So it's a 17 chemical -- it's a chemical term related to the 18 nature of the molecule. 19 Q. And phospholipidosis is what? 20 A. Phospholipidosis is just the 21 conclusion of phospholipids in intracellular 22 globules that you sometimes see in toxicology. 23 Q. It's a physical phenomena, is 24 that right? Page 78 1 A. Yes, a physical phenomena. 2 Q. And it can cause lung 3 problems? 4 A. Theoretically I suppose it 5 could, although I'm not sure there have been any 6 specific problems associated with it. 7 Q. The concern was that it might 8 cause lung problems, is that right? 9 A. That's correct. 10 Q. And some of this has been seen 11 in animal studies during the clinical trials, is 12 that right? 13 A. During the animal trials. 14 Q. Right. 15 A. Not in the clinical trials. 16 It had been seen at high doses. 17 Q. And that's what everybody -- 18 these regulatory bodies were concerned about? 19 A. Yes. 20 Q. And that's why this eminent 21 toxicologist had been brought in? 22 A. Yes. 23 Q. To help you in explaining this 24 to the regulatory bodies, is that right? Page 79 1 A. It's not only explaining to 2 the regulatory bodies, but to be sure that we 3 thoroughly understood the issue. 4 Q. All right. And the overall 5 conclusion was that the pharmacology of 6 fluoxetine has particular side effects, correct? 7 A. Well, again, I think you're 8 focusing on a very specific point. I don't even 9 know who to ascribe this statement to. I would 10 read this as a strategy that John Hall had 11 suggested, that the pharmacology be the focus of 12 the discussion, not the toxicology. 13 Q. All right. Now who is Doctor 14 John Hall? 15 A. He was a medical advisor at 16 the UK affiliate. 17 Q. Is Doctor John Hall an 18 employee of Eli Lilly and Company? 19 A. No. 20 Q. Was he -- would he be 21 characterized as an outside consultant? 22 A. No. He's not currently an 23 employee, he was an employee at that time. 24 Q. At the time -- Page 80 1 A. Yes. 2 Q. -- he was a full-time employee 3 of Eli Lilly and Company? 4 A. Yes. 5 Q. And his position was -- let's 6 emphasize the -- well, let's quote him directly, 7 what he says. He says here, quote, we must 8 emphasize that the side effects of fluoxetine are 9 related to its pharmacology, i.e. CNS 10 stimulation, et cetera, right? 11 A. That's what is said here. 12 Again, I can't tell by the text who -- whether 13 that's his view or whose view it is exactly, but 14 that's what's stated here, yes. 15 Q. Well, he authors the document. 16 A. Yes, he authors the document. 17 Q. He uses the term we. 18 A. Yes. 19 Q. When you say -- do you have 20 any idea who he's referring to when he means we? 21 A. He's referring to the company, 22 I presume, but what I meant was it's tough to say 23 the way it's written exactly who is giving that 24 advice and whether Doctor Hall feels it's the Page 81 1 appropriate advice. It's just vague, that's the 2 only point I was making. 3 Q. We can narrow it down probably 4 to Doctor Hall or Doctor Zbinden, can't we? 5 A. I think that's probably true, 6 yes. 7 Q. And there, when the issue of 8 phospholipidosis comes up, the strategy was to 9 emphasize the side effects of fluoxetine related 10 to its pharmacology, which were CNS stimulation, 11 correct? 12 A. Well, I'm not sure it's fair 13 to say the strategy was, I think he's suggesting -- 14 somebody is suggesting that that be the strategy. 15 Q. All right. I assume that you 16 knew Doctor Hall? 17 A. Yes. 18 Q. Did you work with Doctor Hall? 19 A. Some, yes. 20 Q. What was his specialty? 21 A. I don't recall exactly. I 22 know he had worked extensively with fluoxetine at 23 the time. 24 Q. And he was knowledgable with Page 82 1 respect to the properties of fluoxetine? 2 A. I think that's fair to say, 3 yes. 4 Q. And, therefore, when he says 5 we must emphasize that the side effects of 6 fluoxetine are related to its pharmacology, i.e. 7 CNS stimulation, he says that based on some 8 experience with fluoxetine, correct? 9 A. Well, again -- we may be 10 quibbling on a point, but I can't say exactly 11 that was his view or Zbinden's view, who had much 12 less of a knowledge, I just don't know. 13 Q. Well, in the next paragraph, 14 he says he feels we should not talk too much 15 about -- I can't read that. 16 A. Amiodarone. 17 Q. Amiodarone lung changes. So 18 apparently the next paragraph is Doctor Zbinden's 19 view, and the paragraph on side effects of 20 fluoxetine being CNS stimulation is Doctor Hall's 21 view? 22 A. I think it's tough to say as 23 these things are written, but that's exactly, you 24 know -- it could just be the structure that he's Page 83 1 used. 2 Q. That's a reasonable 3 interpretation, though, is it not, that when he 4 uses the word we -- 5 A. That's one reasonable 6 interpretation. 7 Q. Again, Doctor Zerbe, this is 8 compatible with what we see in Doctor Cohen's 9 comments concerning CNS stimulation, is it not? 10 A. It's compatible, but, you 11 know, honestly, it's a very small part of this 12 whole document, and I don't think -- I think it 13 would be unfair to conclude that either Doctor 14 Hall's or Doctor Zbinden's view is that that's a 15 dominant feature of this drug from this 16 statement, that's all -- 17 Q. If you take every one of these 18 exhibits I've shown you so far this morning, we 19 have CNS stimulation and activation present, 20 don't we, Doctor Zerbe? 21 A. The issue of CNS stimulation 22 was one that was discussed at length. Whether 23 the conclusion was firmly ingrained in anybody's 24 mind with regard to that, I think is a whole Page 84 1 other matter. And the things that you've cited 2 are really pretty small items within the context 3 of the larger picture, I guess that's the only 4 point that I would like to make. 5 Q. Well, I don't know whether 6 they're -- that's my point, Doctor Zerbe, I don't 7 know whether they're small items or not. Here we 8 have, in Exhibit 15, a medical advisor in Erl 9 Wood, employed by Lilly, talking in connection 10 with a world renowned toxicologist, don't we? 11 A. Yes. 12 Q. And that medical advisor or 13 the world renowned toxicologist is saying we must 14 emphasize that the side effects of fluoxetine are 15 related to its pharmacology, i.e. CNS 16 stimulation, correct, I mean it's there in black 17 and white, isn't it? 18 A. Yes, but I think that that's 19 taken out of context with regard to the whole 20 document. The document itself -- the expert was 21 not called in to evaluate the issue of 22 stimulation or the CNS pharmacology, the 23 toxicologist was called in to look at 24 phospholipidosis. And virtually the whole Page 85 1 document, except for that almost parenthetic 2 statement, is related to phospholipidosis, it's 3 not related to that point. 4 Q. Is it parenthetic when it says 5 here we must emphasize that the side effect 6 profile -- side effects of fluoxetine are related 7 to its pharmacology? That doesn't look 8 parenthetic to me, Doctor, when he's talking 9 about how you emphasize something with respect to 10 a regulatory body and with respect to a property 11 of a medication that's now being taken by ten 12 million people -- 13 MR. LORE: That's not a question, 14 though, that's your statement. 15 Q. -- is it? 16 A. Technically it's not in 17 parentheses, but it does say, i.e. -- I want to 18 read it correctly -- i.e. CNS stimulation, et 19 cetera. You know, it was almost a throwaway 20 comment, as I read it. 21 Q. That's your interpretation. 22 A. That's my interpretation, 23 that's right. 24 Q. Whether or not somebody would Page 86 1 consider that a throwaway comment would have to 2 take these other documents into consideration 3 also, wouldn't they? 4 MR. LORE: Not necessarily, Paul. 5 He's already answered your question, I think, 6 several times. 7 Q. Well, Doctor Zerbe, do you 8 have an opinion in connection with whether or not 9 Prozac is CNS stimulating? 10 A. My view is that it is not 11 sedating and -- or not predominantly sedating, I 12 guess is the way I would summarize it, as opposed 13 to truly stimulating. But that's my opinion. 14 Q. Do you have an opinion 15 concerning whether or not Prozac is CNS 16 activating? 17 A. I'm not sure how you define 18 CNS activating. 19 Q. Let's define it as is defined 20 by Doctor Beasley or Heiligenstein in their 21 Exhibit 11, their activation, sedation paper, 22 where he compares it with placebo and other 23 tricyclics, and finds that there are more 24 instances of CNS activation. Page 87 1 A. Well, I can't dispute those 2 figures, so, I mean, activation as defined as 3 they have done it, numerically is higher. Again, 4 in this document -- I had trouble linking the 5 statistics to the figures, but -- 6 Q. All right. 7 (PLAINTIFFS' EXHIBIT NO. 16 WAS 8 MARKED FOR IDENTIFICATION AND 9 RECEIVED IN EVIDENCE.) 10 Q. Doctor Zerbe, Exhibit 16 is 11 three pages of Eli Lilly E-mail dated September 12 13th and 14th, 1990, is it not? 13 A. Yes. 14 Q. And you're copied under the 15 September 14th, 07:31 listing, are you not? 16 A. Yes. 17 Q. It appears that the top part 18 of Exhibit 16 is an E-mail authored by Doctor 19 Leigh Thompson, is that right? 20 A. Yes. 21 Q. Directed to yourself and 22 others, is that right? 23 A. Yes. 24 Q. Now, let's kind of identify Page 88 1 the individuals listed on the top part. Doctor 2 Dan Masica, correct? 3 A. Yes. 4 Q. Doctor Masica was who, what 5 was Doctor Masica's job in September, 1990? 6 A. Doctor Masica was the director 7 for the CNS division. 8 Q. And what was your -- 9 A. I was Doctor Masica's boss. 10 Q. All right. And Doctor 11 Heiligenstein, Beasley and Wheadon were 12 individuals who were directly responsible to 13 Doctor Masica, correct? 14 A. Yes. 15 Q. And Doctor Masica was directly 16 responsible to you, is that right? 17 A. Yes. 18 Q. Now Gordon Gilad or Gilad 19 Gordon, who was he? 20 A. Gilad was another physician 21 like Heiligenstein, Beasley and Wheadon, but had 22 more of an interest in epidemiology and was 23 involved in evaluation of these events. 24 Q. Heiligenstein, Beasley and Page 89 1 Wheadon are psychiatrists? 2 A. Yes. 3 Q. Masica is a psychiatrist? 4 A. No, I don't believe so. 5 Q. So you're the boss of all of 6 these doctors here, are you not, either directly 7 or indirectly? 8 A. Yes. 9 Q. Except for Doctor Leigh 10 Thompson? 11 A. Well, I thought you were still 12 referring to that portion. 13 Q. Yes, I am. 14 A. Yes. 15 Q. And what was the difference in 16 positions in you and Doctor Leigh Thompson in 17 October, 1990? 18 A. I reported to Doctor Thompson, 19 but he had broader responsibilities. I'm not 20 sure at that time whether he had responsibilities 21 for some preclinical research or not, but he had 22 broader responsibilities, obviously, and I 23 reported to him. 24 Q. But he was, at that time, Page 90 1 director of medical affairs or vice-president? 2 A. He was a vice-president, and 3 it wouldn't have been called vice-president of 4 medical affairs, it would have been vice 5 president of Lilly Research Labs or something 6 like that. 7 Q. And you weren't a 8 vice-president in September of 1990, were you? 9 A. I don't believe so. 10 Q. Was Doctor Leigh Thompson the 11 individual to whom you were regularly reporting 12 on a daily basis in September, 1990? 13 A. He was my boss, and I guess -- 14 I don't know if on a daily basis is quite the 15 right characterization, but he was my direct 16 supervisor, the person I would have interacted 17 with. 18 Q. If you had a problem in 19 connection with management, you would first turn 20 to Doctor Thompson? 21 A. Generally, yes. 22 Q. And Doctor Thompson would turn 23 to you to look to see that individuals such as 24 Doctor Masica or those under him were fulfilling Page 91 1 their duties? 2 A. Yes. I'm not sure it's quite 3 that strictly hierarchical, he might turn 4 directly to Doctor Masica or he might turn 5 directly to Doctor Wheadon and so forth, it 6 wasn't always strictly up and down the line. 7 Q. So if there are lines to be 8 drawn, we've got those lines drawn now, don't we? 9 A. Yes, that's the way -- that's 10 right, the hierarchy is described. Whether it 11 works that way is another matter. 12 Q. Well, certainly you wouldn't 13 have had any objection to Doctor Thompson 14 directing a question directly to Doctor Masica or 15 any of his staff? 16 A. That's correct. 17 Q. And in the same instance, if 18 in September of 1990 you happened to see Doctor 19 Mel Perelman and had an issue that you wanted to 20 raise to him, even though technically you were to 21 be reporting directly to Doctor Leigh Thompson, 22 you wouldn't have any problem in speaking of an 23 issue with Doctor Mel Perelman? 24 A. That's correct. Page 92 1 Q. And Doctor Leigh Thompson 2 wouldn't be particularly offended by that? 3 A. I don't believe so, no. 4 Q. Unless you said something 5 offensive about Doctor Thompson. 6 A. That's right. And you don't 7 do those kinds of things. 8 Q. Apparently Doctor Leigh 9 Thompson was going to make a presentation to the 10 board of directors of Eli Lilly and Company in 11 September, 1990, correct? 12 A. That's what it appears to be, 13 yes. 14 Q. For the record, I believe that 15 board presentation was made on September 17, 16 1990; if you'll trust my representation. And he 17 had given Doctor Masica and you some drafts of 18 verbatims on Monday, is that right? 19 A. I don't know what day of the 20 week it was, but I guess it says here Monday 21 morning. That's the way he's characterized it, 22 yes. 23 Q. Do you recall specifically 24 Doctor Thompson giving you these drafts and Page 93 1 verbatims, and this proposed presentation to the 2 board of directors? 3 A. Really only in general terms, 4 I don't remember the specific contents of it. 5 Q. Do you recall this E-mail now 6 that you've had an opportunity to look at it? 7 A. Really only, again, generally. 8 I recall the sort of general background. 9 Q. Doctor Thompson was somewhat 10 miffed that the information he was requesting 11 came later, correct? 12 A. Yes, I think that's a fair way 13 to characterize it. 14 Q. And apparently you had -- he 15 says he had given you drafts of this material, 16 but apparently you had delegated specific 17 responsibilities to the psychiatrists involved, 18 is that right? 19 A. That's, I think, correct, yes. 20 Q. And apparently Doctor Thompson 21 is not upset with you because this is something 22 that he expected them to specifically reply to, 23 is that right, by virtue of the fact that they're 24 psychiatrists? Page 94 1 A. That's the way I would read 2 it, yes. And for the technical issues, that 3 would have been the way we would have operated. 4 Q. Do you recall Doctor Thompson 5 coming and directing criticism to you about it? 6 A. About this specific event? 7 Q. Yes. 8 A. Well, again, I can't remember 9 whether it was related to this one, but there 10 were times that he was concerned about the rate 11 at which the data could be generated to address 12 the point. 13 Q. Okay. On suicidality? 14 A. Yes, suicidality and -- yes, 15 primarily suicidality, I think that's right. 16 Q. But he wouldn't come to you 17 and fuss about something you had or hadn't done, 18 his complaints concerned activities of those that 19 you had responsibility for, is that right? 20 A. Well -- 21 Q. Or was he fussing at you 22 because you should have been on individuals who 23 you were supervising to get data to them? 24 A. I took my share of pressure Page 95 1 from Doctor Thompson, that's the way I would 2 characterize it. 3 Q. But this was -- this 4 particular project wasn't something that you 5 worked on yourself, was it, or was it? 6 A. In terms of dealing with the 7 individual data, no, in terms of coordinating the 8 activities, yes. And that's kind of the role 9 that I played in it. 10 Q. Okay. So you knew in advance 11 Doctor Leigh Thompson was going to make a 12 presentation to the board of directors concerning 13 suicidality? 14 A. Again, I don't know how much 15 in advance. We may have only known on Monday or 16 shortly before the time that he was going to do 17 it that this was requested. 18 Q. All right. And you delegated 19 the physical looking at the data and looking at 20 the presentation he was going to be make to the 21 psychiatrists who worked under you? 22 A. Yes. 23 Q. Then apparently Doctor 24 Heiligenstein is reporting in the balance of this Page 96 1 exhibit what he, Doctor Beasley and Doctor 2 Wheadon had concluded in reviewing the data -- or 3 reviewing the data that was to be presented to 4 the board, is that right? 5 A. That's the way I read it, yes. 6 Q. Doctor Heiligenstein says that -- 7 under verbatim four, at the bottom of page one, 8 quote, we feel that caution should be exercised 9 in the statement that suicidality and hostile 10 acts in patients taking Prozac reflects the 11 patient's disorder and not a causal relationship 12 to PZ, end quote, correct? 13 A. That's right. 14 Q. Doctor Heiligenstein goes on 15 to say, post-marketing reports are increasingly 16 fuzzy, and we have assigned yes, reasonably 17 related on several reports, correct? 18 A. Yes. 19 Q. Now you're not copied on this 20 particular portion of Exhibit 16, are you? 21 A. No. 22 Q. Apparently the psychiatrists 23 are reporting directly back to Doctor Thompson? 24 A. Yes, that would appear to be Page 97 1 the case, yes. 2 Q. Have you ever seen Exhibit 16 3 before? 4 A. I don't recall seeing it, but 5 I obviously would have as, you know, part of this 6 having been copied. 7 Q. Well, the point is, you 8 weren't copied. 9 A. Well, when it's printed out, 10 the forwarded message would be copied. So when 11 Leigh's message came to me, I would have had the 12 whole thing. 13 Q. Okay, all right. So you are 14 copied in the latter transmission, so he would 15 have pulled up the transmission from Doctor 16 Heiligenstein directly to Doctor Thompson, is 17 that right? 18 A. Yes, that's right. 19 Q. Did you talk with any of the 20 psychiatrists under your supervision once you got 21 this memo? 22 A. We were, you know, continually 23 talking, so it's hard to say what -- how this 24 specific memo came into our conversation. But we Page 98 1 would have talked to them, you know, literally 2 daily regarding this issue at this particular 3 point. 4 Q. Well, did you agree with the 5 psychiatrists at Lilly where they say we feel 6 that caution should be exercised in a statement 7 that, quote, suicidality and hostile acts in 8 patients taking Prozac reflect the patient's 9 disorder and not a causal relationship to PZ, 10 meaning Prozac, end quote? 11 A. Well, I think worded in that 12 way, I would agree with it. I think our position 13 always was we wanted to be cautious, we wanted to 14 be sure that the data we had and were presenting 15 was accurate, and so I think this was a period 16 when the psychiatrists were very cautious, very 17 probing, trying to be sure that we really 18 understood the event and what data we had. So I 19 supported them in that kind of a position, that 20 is exercising caution. We still hadn't reached a 21 conclusion about what the wording should be, but 22 certainly it was advisable to exercise caution. 23 Q. Well, what the psychiatrists 24 are saying, Doctor Zerbe, is that Doctor Thompson Page 99 1 exercise caution when you tell the board of 2 directors of Eli Lilly and Company that 3 suicidality and hostile acts are related to the 4 patient's condition as opposed to Prozac itself, 5 isn't it? 6 A. Well, that's what they're 7 saying. Based on the knowledge at that 8 particular point in time -- 9 Q. September 13th. 10 A. -- our advice, as the 11 psychiatric group, should be that you should 12 exercise caution in, you know, a verbatim that 13 totally discounts this issue. And not all the 14 information was available at that particular 15 point and so caution was appropriate. 16 Q. What information was available 17 was, was -- they go on to explain, don't they, 18 they say post-marketing reports were increasingly 19 fuzzy and we have assigned yes, reasonably 20 related on several reports, correct? 21 A. That's what they say, yes. 22 Q. What they're speaking of is 23 post-marketing reports of suicidality and 24 hostility, aren't they? Page 100 1 A. I assume. I don't know that 2 this explicitly states that, but -- it doesn't 3 explicitly state that. 4 Q. That's the subject, isn't it? 5 A. That's the subject, yes. 6 Q. It's not talking about nausea 7 or skin rash, is it? 8 A. No. 9 Q. The presentation to the board 10 of directors is going to have to do with 11 suicidality and hostile acts, isn't it? 12 A. I'm not sure that the 13 presentation was confined to that, I just don't 14 know exactly, but that's certainly on point, 15 verbatim number four is the topic. 16 Q. That is the topic, being 17 suicidality and hostile acts, correct? 18 A. Yes. 19 Q. And what Doctor Heiligenstein 20 goes ahead and says there is that we've been 21 getting reports of suicidality and hostility and 22 that we've assigned yes, reasonably related on 23 several of those reports, correct? 24 A. That's what he said, yes. Page 101 1 Q. Did you know that at that 2 time? 3 A. Well, I mean, I would have 4 known it as a result of ongoing discussions that 5 some of those events may be classified as yes, 6 reasonably related, that was always up to the 7 physician, and we never -- you know, we never 8 questioned or argued about that point. But that 9 does not mean that they're conclusively related, 10 and that's an important point, it just means that 11 other alternative explanations, at the point of 12 which the box was checked, were not evident. 13 Q. When you say we left this up 14 to the physicians, what are you talking about? 15 A. About the psychiatrists 16 working with, you know, the physicians that 17 reported. You know, the management didn't 18 second-guess those things and go back and say, 19 you know, why did you put this as possibly 20 related. 21 Q. The point is is that the 22 people who were checking yes, reasonably related, 23 were Doctor Heiligenstein, Doctor Beasley and 24 Doctor Wheadon, weren't they, because they were Page 102 1 the psychiatrists reviewing the reports? 2 A. Yes, I assume it was them. It 3 may have been somebody else working within that 4 division, but basically they would have been 5 consulted on the -- 6 Q. But it was the Lilly 7 employees, the Lilly psychiatrists, the Lilly 8 experts, who were checking yes, reasonably 9 related, correct? 10 A. That's right. 11 Q. And it was their job, the 12 Lilly experts, the Lilly psychiatrists, to 13 investigate reports of suicidality and hostility, 14 wasn't it? 15 A. Yes, that was one of their 16 responsibilities, yes. 17 Q. They were charged with 18 monitoring adverse events in connection with 19 Prozac, weren't they? 20 A. Yes. 21 Q. And it was their 22 responsibility to investigate adverse events in 23 connection with Prozac, wasn't it? 24 A. Yes. Page 103 1 Q. And they were to contact 2 physicians or reporters of suicidality and 3 hostility, weren't they? 4 A. When possible, yes. 5 Q. And make an investigation 6 concerning whether or not suicidality and 7 hostility were related to Prozac, weren't they? 8 A. That, in collaboration with 9 other outside investigators, yes. 10 Q. And with respect to particular 11 instances, on occasion they would make a 12 determination concerning whether or not the 13 instances of suicidality or hostility were 14 reasonably related to the use of the drug, 15 correct? 16 A. They would make those 17 assessments based on their knowledge at that 18 particular point. 19 Q. Yes, of that particular case? 20 A. Of that particular case. 21 Q. In other words, they weren't 22 making a statistical analysis in connection with 23 particular cases, were they? 24 A. Well, not for this -- the Page 104 1 purpose of this, no. 2 Q. They were making a factual 3 analysis of the facts that occurred in a 4 particular case, weren't they? 5 A. I'm not sure -- I mean you can -- 6 you call it factual analysis, but the reality is 7 that they were operating frequently on very 8 limited information. 9 Q. And sometimes very specific 10 detailed information, weren't they? 11 A. Sometimes. 12 Q. But they were making -- 13 A. In most of the cases in which -- 14 well, in general, the sort of operative rule was, 15 if there was a question, then it was included in 16 that category. 17 Q. Well -- 18 A. And, therefore, when there was 19 lack of information, it tended to be listed as, 20 you know, reasonably, possibly related. We were 21 very concerned about any perception that we were 22 hiding information. 23 Q. Let's talk about that, let's 24 talk about that. Drs. Beasley, Heiligenstein Page 105 1 and Wheadon were charged with the responsibility, 2 in the psychiatric area, because they were 3 psychiatrists, of reviewing instances of adverse 4 psychiatric reports in connection with the use of 5 Prozac, correct? 6 A. Yes. 7 Q. And they were charged with 8 investigating reports of particular instances 9 where the issue was this adverse reaction of 10 suicide, suicidal ideation or hostility related 11 to the use of Prozac, weren't they? 12 A. Well, possibly related to the 13 use of Prozac. 14 Q. Reasonably, possibly related. 15 A. Reasonably, possibly related. 16 Q. And what they were supposed to 17 do was get all information available to them 18 concerning a particular case, weren't they? 19 A. Within their capabilities. 20 Q. Sometimes it was -- they were 21 able to, sometimes they weren't able to get all 22 information. 23 A. That's correct. 24 Q. But they were expected to use Page 106 1 whatever means was necessary to get as much 2 information as possibly existed at the time. 3 A. To understand the event as 4 best they possibly could, I think that's a fair 5 summary. 6 Q. In other words, they were 7 expected to, and did on occasion, talk with the 8 treating physician? 9 A. Yes. 10 Q. They were expected to, and did 11 on occasion, obtain medical records? 12 A. Yes. 13 Q. They were expected to, and did 14 on occasion, talk with nurses and family 15 personnel? 16 A. Right. 17 Q. They were expected to, and did 18 on occasion -- 19 A. I'm not sure how often they 20 talked to families. 21 Q. If that was reasonable to do, 22 and if it was possible, they would -- they were 23 to do that, weren't they? 24 A. I think if your point is that Page 107 1 they would attempt to thoroughly understand those 2 events, they -- that was part of their 3 responsibility, to try to thoroughly understand 4 the events. 5 Q. Right. 6 A. But it was not always 7 possible. 8 Q. I understand that in some 9 instances you may have a lot of information and 10 in some instances you wouldn't have as much 11 information. 12 A. Right. 13 Q. But whether you had 14 information or not, they were charged to secure 15 as much information as possible? 16 A. That's right. 17 Q. In other words, we'll get to 18 some other documents, but the issue of Prozac and 19 suicidality was one that was of great concern to 20 management and Eli Lilly, correct? 21 A. Yes. 22 Q. And there wasn't supposed to 23 be any stone left unturned in examining this 24 issue, was there, in September of 1990? Page 108 1 A. That's right. 2 Q. And these psychiatrists, these 3 employees of Lilly, were expected to get as much 4 information as is possible about a particular 5 instance, correct? 6 A. I mean, I think I've answered 7 that question, they are trying to get all they 8 can about it, but they don't always have that. 9 Q. I understand that, but they 10 always are expected to get as much as is 11 possible? 12 A. I think that's fair to say, 13 yes. 14 Q. These are also psychiatrists 15 that are employed by Eli Lilly and Company 16 because they are supposedly reasonably 17 intelligent individuals? 18 A. Sure. 19 Q. And you as their indirect 20 supervisor did indeed have respect for their 21 medical judgment, did you not? 22 A. Yes. 23 Q. You don't know of any instance 24 where any of these individuals, Doctor Beasley, Page 109 1 Doctor Wheadon or Doctor Heiligenstein, exercised 2 poor medical judgment, do you? 3 A. Well, I can't think of any at 4 the moment, no. 5 Q. If you had seen that, you 6 would have issued some type of criticism of them, 7 would you not? 8 A. Well, yes. 9 Q. Did you ever complain to 10 Doctor Masica, their direct supervisor, 11 concerning any instances of poor medical judgment 12 that you had seen in Doctor Beasley, Wheadon or 13 Heiligenstein? 14 A. Well, I mean I'm not sure you 15 would call it poor medical judgment, I think -- 16 you know, all of us would look at the same set of 17 data and we may reach different conclusions. 18 Q. I'm not talking about data, 19 I'm talking about their integrity as medical 20 doctors. 21 A. I never questioned their 22 integrity. 23 Q. I'm talking about their 24 ability to exercise their profession and their Page 110 1 subspecialty as psychiatrists, you respect their 2 judgment as psychiatrists? 3 A. I never questioned their 4 expertise. 5 Q. Would you feel uncomfortable 6 if you had a psychiatric problem in going to any 7 one of these individuals for advice concerning 8 your psychiatric problems? 9 A. I don't believe so. 10 Q. All right. Wasn't it part of 11 their duty, then, to exercise judgement as 12 psychiatrists, based on their medical knowledge 13 and background, concerning whether or not these 14 reported adverse events of suicide and hostility 15 were related to Prozac? 16 A. I'm sorry, could you repeat 17 the question? 18 MR. SMITH: Read it back. 19 (THE COURT REPORTER READ BACK THE 20 REQUESTED TESTIMONY.) 21 A. Well, within the limitations 22 of the data they had available, and many times 23 they did not have, despite their efforts, a great 24 deal of information, so it was not -- there was Page 111 1 not a whole lot of psychiatric judgment involved, 2 it was more a matter of, you know, 3 bureaucratically they were left by default to put 4 it into the possibly related category. I just 5 don't know enough about -- your implication, I 6 believe, is that they examined it and reached the 7 conclusion that they were reasonably, possibly 8 related. 9 Q. The reason I say that is 10 because Doctor Heiligenstein and Doctor Wheadon 11 and Doctor Beasley have all three testified 12 that's exactly what they did on many instances. 13 A. I can't judge that. All I'm 14 saying is there may have been some of these, and 15 the fuzziness may have been related to the fact 16 that they didn't have information. And if you 17 have the testimony from Doctor Beasley, you're 18 asking me to speculate about what was in their 19 mind, and I'm not the best person to do that. 20 Q. You're certainly not going to 21 give us sworn testimony today, Doctor Zerbe, are 22 you, that what they're doing is checking yes, 23 reasonably related only in instances where they 24 didn't have any information or scant information, Page 112 1 is it? 2 A. Well, I just don't know, I 3 don't know which ones were reasonably, possibly -- 4 Q. Why would you make such a 5 statement, that you suspect or that there were 6 many instances where we didn't have any 7 information so they may have put it in that 8 category of yes, reasonably related? 9 MR. LORE: I don't know that that's 10 what he stated, Paul. 11 MR. SMITH: If it's not, he can tell 12 me. 13 A. I mean, again, if you've got 14 Doctor Beasley and that group saying that they 15 looked at the cases and they thoroughly evaluated 16 them and reached the conclusion that they were 17 reasonably, possibly related, then that probably 18 is much more valid than my speculation about what 19 they looked at, I mean that's the only point that 20 I would make. 21 Q. Would that be a surprise to 22 you, that Doctor Wheadon has testified, for 23 instance, that he categorized at least eighty 24 reports of suicidality and hostility and violent Page 113 1 aggressive behavior as being reasonably, probably 2 related to the use of Prozac, would that surprise 3 you? 4 A. I guess it really wouldn't 5 surprise me. 6 Q. Doctor Wheadon was one of the 7 psychiatrists that was charged with the 8 responsibility of investigating these instances 9 of suicidality and hostility, wasn't he? 10 A. Doctor Wheadon was one of the 11 physicians, yes. 12 Q. Go with me to page two where 13 it speaks about page five of the verbatims. Do 14 you see that? 15 A. Yes. 16 Q. Right above the top of that, 17 it says page five under suicidal thinking in 18 clinical trials, you may want to note that the 19 trials were not intended to address issue of 20 suicidality. Also in paragraph two, patients 21 were excluded that were serious suicidal risk. 22 Do you see that? 23 A. Yes. 24 Q. That word is kind of fuzzy, Page 114 1 but it is risk. Do you agree, Doctor Zerbe, that 2 the clinical trials, as of September 14th, 1990, 3 done in connection with the use of Prozac were 4 not intended to address the issue of suicidality? 5 A. Well, not directly. They were 6 not intended to assess the suicidality, per se, 7 there were aspects of the trials that allowed an 8 evaluation of suicidality. 9 Q. Listen to my question. Do you 10 agree with Doctor Heiligenstein's statement there 11 that, quote, trials were not intended to address 12 issue of suicidality, end quote, that's a yes or 13 no? 14 A. Well, is it yes or no? I 15 agree only partially with that statement. There 16 are ways -- you know, as suicidality is part of 17 the disease, the trials were intended to assess 18 suicidality -- 19 Q. So you don't agree with it? 20 A. -- as part of the disease. 21 MR. LORE: No, he said he agreed with 22 part of it and disagreed with part of it. 23 Q. What part do you agree with 24 and what part do you disagree with? Page 115 1 A. If the implication of the 2 statement is that you can't assess suicidality, 3 then I disagree with it. If the implication is 4 that we had specific scales to look at 5 suicidality, per se, then I agree with it. Do 6 you understand what I'm saying? One is able to 7 assess suicidality as part of the disease of 8 depression from the trials that were done. 9 Q. Okay. And I believe you've 10 already testified in your deposition previously 11 that the HAM-D, item 3, suicidality item on that 12 scale was not a sophisticated scale and that 13 there were better scales to measure suicidality 14 in your opinion. 15 A. Yes, and I still agree with 16 that statement. 17 Q. My reading of this statement 18 is simply, Doctor Zerbe, that the trials were not 19 intended to assess suicidality. 20 A. Well, only -- I mean 21 suicidality is part of depression, they were 22 intended to address the issue of depression, and 23 therefore, in a sense, they were intended to 24 address suicidality as part of depression. Page 116 1 That's the problem, that's why I would disagree 2 with the statement. 3 Q. Then tell us, as a member of 4 senior management, if the trials were intended to 5 address suicidality, why there wasn't a better 6 determination and examination made of the issue 7 of suicide in connection with the clinical 8 trials? 9 MR. LORE: If you understand Mr. 10 Smith's question. 11 A. I think you're misstating what 12 I stated. My disagreement with this is that if 13 you consider suicidality as part of depression -- 14 Q. I don't consider it a part of 15 depression in every instance, there's millions of 16 people who are depressed who are not suicidal, 17 aren't there? 18 A. Yes. But are you going to let 19 me finish my general statement? 20 Q. Yes. 21 A. If you consider suicidality as 22 part of the spectrum of depression, okay, then 23 the trials were designed to thoroughly evaluate 24 depression and suicidality as part of it, in the Page 117 1 garden variety, you know. On the other hand, it 2 was not a trial designed to look at suicide, per 3 se, and if that's what the meaning of this 4 statement, then I agree with the author. If 5 they're excluding it, that the trials were 6 invalid for assessment of suicidality in some 7 maybe less refined way, but still a valid way, 8 then I disagree with the statement. 9 Q. Well, would you agree with me 10 that if it was intended to assess suicidality, 11 that the Prozac clinical trials certainly didn't 12 use a very sophisticated means to address the 13 issue of suicidality? 14 A. It was intended to address 15 suicidality as part of the spectrum of the 16 disease, and the sophistication was at the same 17 level as with any drug for the treatment of 18 depression. And it was a -- 19 Q. That's HAM-D, item 3? 20 A. That's HAM-D, item 3, and 21 adverse events, frequency of suicide, et cetera. 22 Q. Do you think the adverse event 23 frequency of suicide is important in examining 24 the issue of whether or not there's a Page 118 1 relationship to Prozac and suicidality? 2 A. It can be of importance, yes. 3 Q. And do you think it's 4 important to look at the instances where this 5 occurred during the clinical trial process? 6 A. Yes, and that was done. 7 Q. And do you think it's 8 important to look at instances post-marketing 9 where it was reported that it occurred in making 10 a determination whether or not the situation 11 exists? 12 A. All those are useful pieces of 13 information. 14 Q. That's scientifically useful. 15 A. There are pretty significant 16 limitations in spontaneous events. 17 Q. There's advantages and 18 disadvantages -- 19 A. Yes. 20 Q. -- to the post-marketing data, 21 is there not? 22 A. There's limited statistical 23 use that you can make in post-marketing data. 24 Q. But that doesn't mean it's not Page 119 1 valuable. 2 A. No, I didn't mean to suggest, 3 I just said there are limited -- you emphasized 4 scientific, and I just want to make sure that the 5 implication is that those don't weigh equally 6 scientifically. 7 Q. But they both play a part in 8 the analysis of the issue, don't they? 9 A. All three play a part, I think 10 we talked about. 11 Q. Those three being? 12 A. I'm trying to reconstruct them 13 myself, scales, events and trials, and 14 spontaneous events. 15 Q. All right. 16 (A SHORT LUNCH RECESS WAS TAKEN.) 17 (PLAINTIFFS' EXHIBIT NO. 17 WAS 18 MARKED FOR IDENTIFICATION AND 19 RECEIVED IN EVIDENCE.) 20 Q. Doctor Zerbe, Exhibit 17 is a 21 document dated July 30, 1984 and has to do with 22 minutes of a fluoxetine medical marketing staff 23 meeting which apparently occurred on July 24th, 24 1984, correct? Page 120 1 A. Yes. 2 Q. And the minutes were dictated 3 by -- or prepared by Rod Casavant? 4 A. Yes. 5 Q. In new products planning, 6 correct? 7 A. Yes. 8 Q. This seemed to be a fairly 9 early document in your career with Lilly. At 10 that time you were probably -- well, no, you were 11 the director of the neuroendocrine division at 12 Lilly, were you not? 13 A. Yes. 14 Q. You apparently had prepared 15 some questions in connection with a consultant 16 visit that occurred on that date, is that right? 17 A. Yes. I don't remember the 18 details, but that's what it says here if you're 19 referring to bullet point two, list of questions 20 R. Zerbe drafted. 21 Q. Yes. 22 A. Yes. 23 Q. Do you recall drafting those 24 questions? Page 121 1 A. I really don't. 2 Q. Apparently Lilly had some 3 psychiatrists coming in to discuss 4 antidepressants in general, Prozac specifically, 5 in connection with future clinical trials and 6 marketing of Prozac, is that right? 7 A. That's right. 8 Q. And you had listed some 9 questions. The first question you raise is 10 strengths and weaknesses of fluoxetine; key 11 unanswered questions resulting from -- and I 12 don't know what that is, do you recall what that 13 is, blacked out? 14 A. I could only speculate. 15 Q. All right, what is it? 16 A. I'm sorry, what is the 17 speculation? 18 Q. Uh-huh. 19 A. This may be related in some 20 way to a conference that was held in, I think, 21 June of '84, which was -- there were concerns 22 about how well this conference went in terms of 23 the clarity with which the data was conveyed, and 24 that may be what it's related to. Page 122 1 Q. Is that the Montecatini -- 2 A. Montecatini conference. 3 Q. And what were the problems 4 that were raised in the Montecatini conference? 5 A. I did not attend the 6 conference itself, so everything that I had was 7 secondhand. And as I recall, some of the major 8 issues related to efficacy, how effective was the 9 drug, and that's kind of what I recall. I don't 10 recall safety issues being a dominant part part 11 of it, but again, that's just recollection and 12 that's secondhand information. 13 Q. Do you recall that suicide or 14 suicidality was raised in the Montecatini 15 conference in any context? 16 A. I don't recall that it was 17 raised in any context. 18 Q. The reason we raised that is 19 that there were some BGA officials there, and if 20 you'll recall one of the earlier documents we 21 gave you had to do with the BGA's impression that 22 Prozac was activating with a side effect profile 23 of suicides and suicide attempts. Do you recall 24 that? Page 123 1 A. I recall the discussion of 2 that in the previous deposition. 3 Q. Do you recall there being a 4 sense that that problem was raised at that time 5 in Montecatini? 6 A. That's not the way I would 7 have remembered it, no, I don't think -- I don't 8 recall that being the issue at Montecatini. 9 Q. Obviously these are questions 10 that you had written out. 11 A. I'm not sure how extensively 12 written they were, they may have just been a 13 bullet point like this, I'm not sure. But I was 14 probably the author of it, yes. 15 Q. An additional item that you 16 list is -- or question that you list to be -- and 17 I assume these are questions to be posed to the 18 consultants, is that correct? 19 A. That's the way I read the 20 document. 21 Q. And this was -- there at the 22 center of the page in Exhibit 17, you posed the 23 question of usefulness/future potential of 24 combination drugs. Page 124 1 A. Where are we now, I'm sorry? 2 Q. Right in the center of the 3 page, bullet point -- 4 A. Oh, okay. 5 Q. All right. Says 6 usefulness/future potential of combination drugs 7 (i.e. antidepressant and tranquilizer) in 8 unipolar depression. Do you see that? 9 A. Yes. 10 Q. Do you recall where it was 11 that -- or what it was that you had seen that 12 raised that potential question in your mind as to 13 whether or not it would be useful or whether 14 there would be any potential in combining 15 antidepressants and tranquilizers? 16 A. I really don't know the origin 17 of that, exactly who raised it or on what it was 18 based. I think at this particular point there 19 was a lot of exploration of alternative 20 approaches, and I'm not sure this was any more 21 than that kind of thing, to explore, consider 22 combinations as you would with other type 23 medications. 24 Q. The question is why were you -- Page 125 1 and this is apparently a question that you had 2 raised -- why were you raising that issue? 3 A. Well, I don't think in my 4 authoring of the list of questions I was 5 necessarily any more than a sort of scribe or 6 consolidator of the issues that surrounded the 7 product at that particular point, so I don't 8 recall this being a particular issue from my 9 perspective, if that's your question. 10 Q. All right. 11 A. What I was doing was serving 12 as a central point to get all of the questions on 13 the table and, you know, kind of taking the lead 14 from that perspective. But that wasn't 15 particularly my question. 16 Q. All right. But it was a 17 question that was being raised by the scientists 18 in general? 19 A. It was raised by someone, 20 whether it was the scientists or the marketing 21 component, I don't know. 22 Q. What would be the marketing 23 value of that, combining an antidepressant such 24 as Prozac with a tranquilizer? Page 126 1 A. I think there was a view, 2 perhaps at one point, that the sedative effects 3 of the tricyclics, which weren't present with 4 fluoxetine early in the treatment, were perceived 5 by physicians as something positive. 6 Q. All right. 7 A. And that by combining a 8 tranquilizer, one could mimic those early effects 9 of the tricyclics. 10 Q. All right. 11 A. I think that was the marketing 12 angle. 13 Q. Was it also a scientific 14 consideration also? 15 A. Well, I mean certainly 16 scientifically one could mimic the effects of 17 tricyclics, but I don't think from a medical 18 perspective it was viewed as something that was 19 very critical or very important. 20 Q. The reason I ask that is 21 because early on in the clinical trials, back as 22 early as 1978, Doctor I. H. Slater, one of the 23 original medical monitors in connection with this 24 drug, reflected that there was an issue early on Page 127 1 as to whether or not it might be of benefit to 2 have Prozac plus a tranquilizer or benzodiazepine 3 together, marketing squuzed together in one pill -- 4 that's s-q-u-u-z-e-d -- you know, combined in one 5 pill, back as early as '78, '79 when Doctor 6 Slater was analyzing it. Had you heard that 7 raised before? 8 A. I had never heard that Doctor 9 Slater had proposed that, no. 10 Q. Regardless of whether he 11 proposed it or who specifically proposed it, had 12 you ever heard that mentioned as a potential from 13 a medical standpoint, that it would have some 14 benefits to add something to fluoxetine 15 hydrochloride that has a tranquilizing effect to 16 get -- 17 A. I don't -- it's tough to sort 18 of categorically say that I never had heard it 19 because it was an issue that bubbled up from time 20 to time with regard to the benefits of the 21 sedation of tricyclics, but I don't remember any 22 specific conversation in which it was really 23 considered a serious proposal. 24 Q. But it was raised as a Page 128 1 consideration in this meeting by you as a list of 2 questions -- 3 A. Right. 4 Q. -- by you for these 5 consultants to review? 6 A. It was one of the issues that 7 was on somebody's mind and we wanted to capture, 8 I think, all of those issues. 9 Q. And the feeling was that since 10 Prozac didn't have a generally sedating effect, 11 such as the tricyclic antidepressants did, that 12 it might be of some benefit to combine a 13 tranquilizer to get that initial tranquilizing, 14 sedating effect since Prozac didn't possess that 15 property, would that be accurate? 16 A. I think the question that came 17 up was, was that something of importance, and 18 that's what was being proposed as a question. 19 Q. Was the thought to actually 20 add with the Prozac in the preparation some 21 tranquilizing agent or was it to prescribe an 22 additional pill that you would take along with 23 Prozac? 24 A. I think the proposal that's Page 129 1 being discussed here is more some combination 2 medication. 3 Q. You mean in one pill? 4 A. In one pill. 5 Q. All right. 6 A. I mean that is a 7 consideration. Again, my recollection was it was 8 more a consideration from the marketing 9 perspective than it was the medical perspective. 10 Q. If that had been done, let's 11 say you stuck a quantity of benzodiazepine in 12 with fluoxetine hydrochloride, could it have 13 still been marketed as a specific serotonin 14 reuptake inhibitor? 15 A. Well, it's very difficult for 16 me to answer that categorically. My sense would 17 be that you could not have marketed it. I'm not 18 sure that you could have marketed it at all, if 19 you know what I mean, that combination 20 medications are very difficult to get approved. 21 You have to have very sound scientific rationale, 22 and I don't know that that scientific rationale 23 for a combination ever really existed. 24 Q. In other words, if you Page 130 1 combined Prozac with a tranquilizer you would 2 lose the purity that Prozac possesses as a 3 specific compound? 4 A. That's -- you would no longer 5 have a specific compound, you would have a 6 combination therapy, and the FDA is -- 7 particularly the FDA is very reluctant to approve 8 combination therapies. 9 Q. It would have made it much 10 more difficult to get Prozac approved by the FDA 11 if you had combined a tranquilizer, wouldn't it? 12 A. Well, I'm not sure -- that 13 isn't quite what I was saying. The objective was 14 not related to Prozac approval, per se, what I'm 15 saying is -- 16 Q. I didn't say that. 17 A. But what I'm saying is whether 18 one could ever get a combination approved or not 19 I think is questionable, in my opinion. 20 Q. Well, there are combination 21 drugs that are approved for use? 22 A. There are. 23 Q. Give us some examples of some 24 well-known combination drugs. Page 131 1 A. Levodopa, which is L-dopa and 2 Carbodopa, it makes a lot of sense 3 pharmacologically because of the way it affects 4 the metabolism of the one component. And some 5 older drugs are combinations, over-the-counter 6 drugs. But in general, the FDA is reluctant to 7 combine combinations, combine drugs, because of 8 essentially the proliferation of different 9 combinations that one can have and the complexity 10 in assessing the pharmacology and the adverse 11 events of each of the components. So in general 12 you have to demonstrate that both are effective 13 on their own, plus the combination adds more than 14 just a, you know, an additive value, it's 15 synergistic, and that's pretty tough to 16 demonstrate. 17 Q. It would have made it tough to 18 get Prozac approved if you had demonstrated that? 19 A. You're saying it would have 20 been tough to get Prozac approved, and what I'm 21 saying is that's not the point. The point is not 22 Prozac approval, one would have had Prozac 23 approved and it would have gone along presumably 24 the same way, as a single agent. What I'm saying Page 132 1 is it would have been difficult to ever get a 2 combination approved, so it wouldn't have 3 necessarily been a very viable option. Do you 4 understand what I'm saying? 5 Q. Yes. But if I went into a 6 drugstore today, and I went into that section of 7 the drugstore today that said co-remedy 8 preparations -- 9 A. Yes. 10 Q. -- wouldn't ninety-five 11 percent of those co-remedy preparations that I 12 see there be combination drugs? 13 A. A very high percentage of 14 them, but they're very old drugs. 15 Q. But a hundred percent of them 16 have been approved by the Food and Drug 17 Administration, right? 18 A. Some of them -- actually, I 19 think -- I could be wrong, but many of them were 20 grandfathered, they existed, they were on the 21 marketplace before the current drug regulations 22 came into place, so whether they were actually 23 formally approved by the FDA or whether they were 24 simply marketed and grandfathered, I don't know. Page 133 1 I think many of the things, like cold 2 preparations, were grandfathered. 3 Q. Yes, but there's new cold 4 preparations every day that hit the market, and 5 you know that, that are combinations of 6 ingredients that may have or have not been 7 grandfathered in, and they're approved. 8 A. Well, we're getting into an 9 area probably that I'm not the best expert. But 10 my perception would be that virtually all of 11 those ingredients that are marketed in those 12 combinations are things that have been approved 13 previously and are put together in some different 14 way and marketed in a new way. Some previously 15 approved ingredient, I guess is the way I would 16 say it. But you're -- I'm probably not the best 17 person to answer all those questions on cold 18 preparations. The fact is, combination 19 medications as prescription drugs are pretty 20 tough to get approved. 21 Q. But that doesn't mean that 22 they're not approved every day, does it, Doctor? 23 A. I can assure you they aren't 24 approved every day, but -- Page 134 1 Q. Well, regularly. If we go 2 into our pharmacy and we see a new cold 3 preparation, regularly. 4 A. You're talking about 5 over-the-counter medications. 6 Q. Uh-huh. 7 A. I'm talking about approval of 8 prescription medications, new prescription 9 medications, in combination. 10 Q. Regardless, we put them in our 11 mouth and swallow them, and they have a physical 12 effect on our body, don't they? 13 A. Well, I guess I'm confused 14 about exactly what question you're asking. 15 Q. I'm confused at the 16 distinction you're drawing between prescription 17 and nonprescription medication when it doesn't 18 make any difference, the FDA has to approve it, 19 isn't that right? 20 A. Yes. 21 Q. If I want to combine an 22 antihistamine and aspirin and bill that as a cold 23 preparation, I must get FDA approval of that, 24 shouldn't I? Page 135 1 A. Yes. 2 Q. In other words, if I want to 3 market it legally, right? 4 A. Yes. 5 Q. And the claim will be made by 6 me, if I'm saying that this would be a good 7 preparation, it's been demonstrated that aspirin 8 is effective for certain uses in indications in 9 colds, and antihistamines are effective, and a 10 combination thereof will, as you say, have a 11 synergistic effect and make an even better cold 12 medication, correct? 13 A. I have to -- state it again as 14 a question, I want to be sure I answer this 15 correctly. 16 Q. Well, if I combine aspirin and 17 antihistamine that hasn't ever been combined 18 before, say a particular type of antihistamine, 19 I'm going to have to get FDA approval of that, 20 aren't I? 21 A. Yes. 22 Q. And to do that I'm going to 23 have to establish that the two primary 24 ingredients, aspirin and antihistamine, are safe Page 136 1 and efficacious. 2 A. Yes. 3 Q. And that the new compound, the 4 combination of those two, is safe and 5 efficacious, correct? 6 A. That's right. 7 Q. Can I use existing clinical 8 trial data on the existing preparations? 9 A. You could use it only to 10 demonstrate the safety and efficacy of the 11 individual elements, but you cannot use them -- 12 you can't extrapolate to the combination as being 13 either additive nor synergistic. 14 Q. So I'll have to have 15 additional clinical trials -- 16 A. Yes. 17 Q. -- combined to demonstrate the 18 efficacy of the two products in combination, is 19 that right? 20 A. That's right, generally that's 21 right. 22 Q. And that's one of the things 23 that you were saying makes it difficult to get a 24 combination drug approved? Page 137 1 A. That's right. 2 Q. Now, if I were a physician and 3 a patient came to me, I could certainly prescribe 4 a combination of medications to treat a 5 particular illness, could I not? 6 A. Yes. 7 Q. For instance, if a patient 8 came to me with a cold, I could prescribe aspirin 9 as one ingredient, and a specific antihistamine 10 as another ingredient, right? 11 A. Yes. 12 Q. Two pills? 13 A. Yes. 14 Q. And there's no prohibition 15 against a physician prescribing those two 16 ingredients in combination, is there? 17 A. That's correct. 18 Q. And the physician doesn't have 19 to go out and do a clinical trial to demonstrate 20 efficacy of the combination of those two 21 ingredients, does he? 22 A. No. 23 Q. And that combination of 24 ingredients that the physician may have Page 138 1 prescribed for that patient may be more 2 efficacious in treating that cold that that 3 patient had, right? 4 A. Yes. 5 Q. Because the patient may have 6 had a number of symptoms, the patient may have 7 had a headache and achiness on one hand which 8 could be alleviated by the aspirin, correct? 9 A. Yes. 10 Q. And the patient may have had a 11 runny nose and sneezing which could be alleviated 12 with the antihistamine, correct? 13 A. That's right. 14 Q. Doctors do that every day, 15 don't they? 16 A. Yes. 17 Q. Combine medications for 18 treatment of a particular illness that has a 19 variety of symptoms, don't they? 20 A. Yes. 21 Q. And don't pharmaceutical firms 22 in many products encourage a physician to use 23 their product with another product to alleviate a 24 particular symptomatology? Page 139 1 A. I think you're making a very 2 general statement in which I would not 3 necessarily agree with. I don't think that's a 4 frequent strategy with regard to the desired 5 state for a pharmaceutical company, I think you 6 prefer to have a drug that met the need as a 7 single agent. 8 Q. Wouldn't you want the 9 physician to know that this agent can also be 10 used in combination with other agents to treat 11 other conditions? 12 A. Well, I think it's always 13 desirable to know and understand those kinds of 14 things, potential combinations, if there were 15 some problem, particularly with regard to adverse 16 events, that you shouldn't combine them. 17 Q. Maybe I can give you an 18 example that just popped into my head. I have 19 kidney stones and on occasion I have to go to the 20 doctor -- I mean go to the hospital, the pain is 21 so severe, and the physician will prescribe an IV 22 injection of Demerol, and he'll also prescribe in 23 that same injection another preparation or 24 another medication called Zestril -- Page 140 1 A. Uh-huh. 2 Q. -- because I'm semi-allergic 3 to Demerol, Demerol makes me nauseated. And 4 apparently that's a known adverse event -- 5 A. Uh-huh. 6 Q. -- or reaction that some 7 people have to Demerol. 8 A. Uh-huh. 9 Q. So the physician prescribes 10 Zestril, which -- is that an antiemetic or is 11 that just something that -- 12 A. It can serve in that purpose, 13 yes. 14 Q. Are you familiar of physicians 15 prescribing those two medications together? 16 A. Yes, Demerol and Zestril are 17 given together sometimes. 18 Q. And the reason for that is? 19 A. I'm not sure it's as much 20 related to the side effects as the synergistic 21 effects of the two with regard to the relief of 22 pain and, you know, muscle relaxation and those 23 kinds of things. But to be honest with you, I'm 24 not an expert on Demerol and Zestril Page 141 1 combinations. 2 Q. Neither am I. 3 A. You're a user. 4 Q. I'm an expert of kidney stone 5 pain, and I had Demerol without Zestril and it's 6 made me vomit every time, and after the first 7 time my physician said I'll give you some Zestril 8 to alleviate that. Right now, they always 9 prescribe it in that manner, that's not 10 unreasonable medical practice, is it? 11 A. No. 12 Q. All right. 13 A. But I think your question was 14 more related to pharmaceutical companies actually 15 making the suggestion, or whatever, that those be 16 given in combination, and I think that's a 17 different question. 18 Q. Well, would it be true, or is, 19 in the package insert or the prescribing 20 information or the literature on Demerol, is 21 there a recommendation that it be given along 22 with other medications if the patient complains 23 of nausea or to reduce the chance that the 24 patient may become nauseated? Page 142 1 A. I don't know what the Demerol 2 package insert says with regard to concomitant 3 use of Zestril. 4 Q. Doesn't Lilly make Demerol? 5 A. Not that I know of. 6 MR. LORE: Now to my knowledge, 7 Counsel. 8 Q. Darvon? 9 A. Darvon, yes. 10 Q. Do you recall whether or not 11 Lilly suggested use of Zestril or an antiemetic 12 or some type of other medication when 13 administering Darvon if a patient experiences 14 nausea? 15 A. I don't believe it does, I 16 don't think there's anything in the package 17 insert about that. 18 Q. Okay. So to get back to that, 19 to this combination of using Prozac and a 20 tranquilizer, the purpose of doing that would be, 21 in some instances, to reduce the anxiety and 22 insomnia that by 1984 had already been 23 demonstrated to occur in some patients? 24 A. Well, again, you're extending Page 143 1 this statement a lot further than I would have. 2 My recollection of the discussion about the 3 combination was not related to treating any side 4 effect of Prozac, but was related to treating 5 anxiety associated with the disease early on. 6 Q. All right. Are you sure about 7 that? 8 A. Excuse me? 9 Q. Do you have a definite 10 recollection that in making that recommendation 11 you weren't suggesting that you would have less 12 anxiety, activating-related side effects by 13 prescribing or combining tranquilizers with 14 Prozac? 15 A. I'm going to have to ask you 16 to repeat the question. 17 Q. I'm sorry, I was behind you 18 when I directed that. 19 (THE COURT REPORTER READ BACK THE 20 REQUESTED TESTIMONY.) 21 A. I'm not sure. 22 MR. LORE: I object to the form as not 23 understandable. 24 (THE COURT REPORTER READ BACK THE Page 144 1 REQUESTED TESTIMONY.) 2 MR. LORE: I object again. This time 3 I'm sure that I don't understand it, a couple of 4 double negatives and all sorts of things that are 5 in there, however if you understand the question, 6 go ahead and answer it. 7 A. Well, the only -- the question 8 I would have is what do you mean by definite 9 recollection. I think I already stated that my 10 recollection was that this was related to 11 treatment of the disease and not any specific 12 side effects of Prozac. But that's just a 13 recollection. 14 Q. I thought you stated earlier 15 that your recollection was that this was an idea 16 raised by marketing that would make the Prozac 17 more attractive since it didn't have a sedating 18 effect as did the tricyclics? 19 A. Well, maybe we're talking past 20 each other. That, I think, is exactly another 21 form of what I said in response to your most 22 recent question. Let me try again. It's my 23 recollection that this was a suggestion, and 24 again, a recollection that it was primarily Page 145 1 driven by the marketing area, and it was meant to 2 address the concerns that Prozac did not have a 3 sedating effect, and in some patients the 4 sedating effect was important in the treatment of 5 the disease. It was not intended to be a 6 treatment of any side effect of Prozac, per se, 7 it was -- it was intended to be considered for 8 treatment of some pharmacologic activity -- for 9 providing some pharmacologic activity that was 10 not provided by Prozac. Do you understand what 11 I'm saying? 12 Q. Uh-huh, I think so. What 13 pharmacologically about Prozac makes it not a 14 sedating antidepressant? 15 A. Well, I'm probably -- I'm not 16 the best person to address that question in terms 17 of the detailed pharmacology and why sedation 18 would not occur. I'll leave it at that. 19 Q. You used -- the reason I 20 addressed it to you is because that's a statement 21 you made, that Prozac, by virtue of its 22 pharmacology, is not a sedating antidepressant. 23 And my question to you is, what specifically 24 about Prozac is it that makes it not a sedating Page 146 1 antidepressant? 2 A. Well, let me try again. I'm 3 not probably the best person to describe that 4 aspect of pharmacology. What I said was -- I'm 5 not sure I said exactly what you had said, but 6 what I said was that it does not have the same 7 sedative effects that tricyclics have. 8 Q. All right. Then I'll ask you 9 this question: What is it about the tricyclics 10 pharmacologically that gives them a sedating 11 property? 12 A. I think it's related to the 13 anticholinergic effects that the tricyclics have. 14 Q. What do you mean by 15 anticholinergic effects? 16 A. Blocks cholinergic receptors. 17 Q. What are cholinergic 18 receptors? 19 A. They're another type of 20 receptor that responds to acetylcholine or 21 acetebocholine. 22 Q. So what you're saying is that 23 Prozac and the tricyclics work on different sets 24 of receptors? Page 147 1 A. What I'm saying is that Prozac 2 does not have the same spectrum of effects on 3 different receptors that tricyclics do. Prozac 4 is much more specific for the serotonin reuptake 5 mechanism, whereas tricyclics have a wider 6 variety of effects on different neurotransmitters 7 and reuptake mechanisms. 8 Q. Which makes them more likely 9 to produce sedation, and makes Prozac less likely 10 to produce sedation? 11 A. That's right. 12 Q. By virtue of the fact that the 13 tricyclics affect a variety of receptors, and 14 Prozac only affects the serotonin reuptake 15 receptors, is that correct? 16 A. Yes. 17 Q. And both produce separate -- 18 in that instance produce separate and distinct 19 physiological responses by virtue of separate and 20 distinct actions on different receptors, is that 21 what you're saying? 22 A. That's right. 23 (PLAINTIFFS' EXHIBIT NO. 18 WAS 24 MARKED FOR IDENTIFICATION AND Page 148 1 RECEIVED IN EVIDENCE.) 2 Q. Exhibit 18 is a document dated 3 May 19, 1989 authored by one of the Lilly 4 psychiatrists, Charles M. Beasley, M.D., is it 5 not? 6 A. Yes. 7 Q. You are copied in on that 8 document? 9 A. Yes. 10 Q. And it appears to be a report 11 on an American Psychiatric Association trip that 12 Doctor Beasley made, is that right? 13 A. That's right. 14 Q. He says attached are 15 fluoxetine-related abstracts. Does that mean 16 that he has gone to the American Psychiatric 17 Association and he's received their abstracts of 18 studies in connection with Prozac? 19 A. Generally what would happen, 20 and I can't speak for this case, I don't know the 21 background specifically, but generally what would 22 happen is they publish a book of the abstracts, 23 and Doctor Beasley probably simply copied the 24 abstracts that he's referring to or that were Page 149 1 related to Prozac and attached them to the memo 2 so that everyone would have the opportunity to 3 see those. 4 Q. So this would have been 5 literature that was presented at the American 6 Psychiatric Association convention or conference, 7 is that right? 8 A. Probably would have been in 9 the abstract book, yes. 10 Q. But I'm not sure what an 11 abstract and an abstract book is. 12 A. Well, scientists send in 13 abstracts to be presented at the meeting, and 14 then all of those are published generally in a 15 program, an abstract book, and probably that's 16 what Doctor Beasley has done. 17 Q. Okay. So at these 18 conventions, there's a program and there's 19 various speakers, is that right? 20 A. Yes. 21 Q. And those speakers either send 22 outlines or summaries of their speech, and that's 23 contained in this abstract book, is that what 24 you're saying? Page 150 1 A. They submit an abstract of 2 what they plan to present prior to the meeting, 3 and then at the time of the meeting the abstract 4 is published as part of a larger book so that the 5 scientists that attend the meeting can hear the 6 talk and can read the abstract and kind of put 7 all of the information together. 8 Q. Okay. Sounds like when I go 9 to a convention or seminar and I make 10 presentations at conventions and seminars, they 11 always make me do research and an outline and 12 things of that nature as something to present 13 along with my speech. Is this similar to what's 14 going on with the American Psychiatric 15 Association? 16 A. They get similar, yes. It 17 would be published probably in a book in advance 18 and you wouldn't hand it out. There are 19 similarities. It's kind of a synopsis of what 20 you intend to present. 21 Q. Oh, no, these are published, 22 this is a handbook that you get with the seminar. 23 A. It sounds similar. 24 Q. And you can get some valuable Page 151 1 information by simply looking at the handbook, 2 you don't have to hear the speeches sometimes. 3 Can the same be said? 4 A. I think the same could be 5 said, yes. 6 Q. Apparently Doctor Beasley has 7 received those, and this, I guess, would have 8 been the 1989 meeting of the American Psychiatric 9 Association, is that right? 10 A. Yes, I assume. 11 Q. He says overall these 12 abstracts are relatively positive except in NR40 13 and perhaps NR215, correct? 14 A. Yes. 15 Q. Do you have any idea what 16 those deal with? 17 A. Do I have any idea, yes, I 18 would say they're probably abstract numbers. I 19 don't know what the NR would stand for. I assume 20 it's some code that the American Psychiatric 21 Association uses to designate the abstracts. 22 Q. All right. Apparently there 23 would be an abstract book, and in that abstract 24 book one of the abstracts would either have a Page 152 1 page number or be designated as abstract NR40? 2 A. That's the way I would 3 understand it. 4 Q. It goes on to say with regard 5 to mania, we are presently running as follows: 6 Fluoxetine sixty-six slash sixty-six hundred and 7 eighty-three, paren, point oh nine nine percent, 8 close paren, versus TCAs, three dash seven 9 thirty-four, paren, point forty-one percent, 10 close paren. Do you see that? 11 A. Yes, I see it. 12 Q. Do you know to what he's 13 referring there? 14 A. I really don't. 15 Q. Obviously, it would appear 16 that he's speaking of adverse events in 17 connection with reports of mania, does it not? 18 MR. LORE: He's already said that he 19 doesn't know what it stands for. If you can give 20 him any further answer, go ahead. 21 A. I don't know how you get these 22 figures from adverse events -- I mean they are 23 some incidence figures, but I don't know where, 24 for example, the sixty-six eighty-three would Page 153 1 come from, it couldn't be spontaneous events, so 2 I really don't know what it is. 3 Q. You don't have any idea what 4 figures he's speaking of in connection with mania 5 where you have a percentage occurrence twice as 6 high -- over twice as high in Prozac as in 7 tricyclic antidepressants, correct? 8 A. I think you would have to 9 address the question to Doctor Beasley. It seems 10 completely out of context with regard to a trip 11 report, that's the thing that's confusing to me. 12 Q. He goes on to say, of course 13 separating mania from euphoria from nonmanic 14 activation, paren, increase in psychomotor 15 activity level is not an easy clinical task. To 16 some extent, the blank abstract NR215 and the 17 blank abstract NR267 are in conflict. He 18 mentions abstract NR215 up in the first sentence. 19 Could he be speaking of some studies that were 20 done and presented as papers in this conference 21 back in 1989? 22 MR. LORE: I think he already answered 23 that question, Paul, he doesn't know. Go ahead 24 and answer if you have a better response. Page 154 1 A. I don't really -- I think you 2 would have to ask Doctor Beasley what he was 3 referring to in the memo. 4 Q. All right. But at least 5 there's some instance where mania is being 6 reported twice as much in connection with people 7 using Prozac than those people using tricyclic 8 antidepressants, isn't it? 9 A. From the extrapolation of 10 this, it's impossible to tell. I don't know what 11 he's talking about, it could be successful 12 treatment of mania for all I know, I don't know. 13 Q. Okay. Well, he does mention 14 NR215 as not being positive, doesn't he, in the 15 first sentence? 16 A. It says and perhaps NR215. 17 But again, what that means, I don't know. 18 Q. All right. We'll ask Doctor 19 Beasley about that. 20 (DISCUSSION OFF THE RECORD.) 21 (PLAINTIFFS' EXHIBIT NO. 19 WAS 22 MARKED FOR IDENTIFICATION AND 23 RECEIVED IN EVIDENCE.) 24 Q. Doctor Zerbe, Exhibit 19 is a Page 155 1 document dated November 13, 1985 authored by 2 Doctor Leigh Thompson, is it not? 3 A. Yes. 4 Q. You're not favored with a copy 5 of this document, but you're mentioned in the 6 document. Do you see where you're mentioned? 7 A. Yes. 8 Q. Have you seen this document 9 before? 10 A. I don't recall seeing it. 11 Q. Have you reviewed any 12 documents in preparation for your deposition here 13 today? 14 A. No. 15 Q. About the middle of the middle 16 paragraph of the first page, it says the second 17 additional -- addition addressed the question of 18 drug interaction. The assessor was concerned 19 about the drug interaction data presented in the 20 registration package, correct? 21 A. Yes, sir. 22 Q. And I didn't mention this, and 23 identified this document. This apparently has to 24 do with registering Prozac in England, does it Page 156 1 not? 2 A. Yes. 3 Q. And at the time Prozac had not 4 been approved for marketing in England, is that 5 right? 6 A. Yes. 7 Q. That paragraph goes on to say 8 Doctor Zerbe learned from Doctor Lemberger that 9 new data showed the half-life of elimination of 10 Diazepam to be doubled when a single dose of 11 Diazepam is given to patients receiving the 12 eighth daily dose of fluoxetine. This led the UK 13 to change the drug interaction statement to 14 include the sentence, quote, although there is no 15 clinical evidence of any untoward effects when 16 Flatine, which is Prozac in England, is 17 coadministered with benzodiazepines, there are 18 limited pharmacokinetic data to suggest that the 19 half-life of Diazepam may be prolonged, end 20 quote, is that correct? 21 A. Yes, sir. 22 Q. Now do you recall your 23 discussion with Doctor Lemberger at the Lilly 24 Clinic concerning this extension of the effects Page 157 1 of Diazepam? 2 A. I don't recall the specific 3 conversation. I know this was a -- this was an 4 observation, but I don't recall the specific 5 discussion with Doctor Lemberger. 6 Q. Now Diazepams are 7 tranquilizers, and specifically there's a 8 tranquilizer by the trade name of Valium, is it 9 not? 10 A. Yes -- well, it's 11 benzodiazepines, that whole family of drugs. 12 Q. They're tranquilizers. 13 A. Yes, they're minor 14 tranquilizers, they aren't like the more powerful 15 tranquilizers like Thorazine. 16 Q. Well, they have a -- what's 17 that fancy word, anxiolytic effect? 18 A. Anti-anxiety, anxiolytic. 19 Q. Anxiolytic effect on humans, 20 is that right? 21 A. Yes. 22 Q. And they do that by virtue of 23 affecting brain chemistry in humans, do they not? 24 A. Well, I'm not sure what you Page 158 1 mean by affecting brain chemistry, but they 2 affect the brain, yes. 3 Q. And the manner in which they 4 affect the brain is they act on receptor sites, 5 correct? 6 A. Yes. 7 Q. Different receptor sites than 8 those acted on by Prozac, correct? 9 A. Yes. 10 Q. What receptor sites are 11 affected by benzodiazepines? 12 A. There's what is called a 13 benzodiazepine receptor. 14 Q. All right. Is that a specific 15 receptor designed to interact with only 16 benzodiazepines? 17 A. Well, it's hard to say only 18 benzodiazepines, but it was a receptor site that 19 was identified as a result of the binding of 20 benzodiazepines to it. There may be other things 21 that bind to it, but benzodiazepines are the 22 thing that led to its identification and their 23 belief to be the mechanism for the benzodiazepine 24 actions. Page 159 1 Q. Well, is Valium a 2 benzodiazepine a specific reuptake inhibitor? 3 A. No. It doesn't have to do 4 with benzodiazepine reuptake, it has to do with 5 the benzodiazepine binding to the receptor and 6 activating the receptor. So it's different than 7 Prozac and its mechanism on serotonin receptors 8 or serotonin reuptake. 9 Q. Do benzodiazepines affect 10 dopamine, epinephrine or other neurotransmitters? 11 A. If there are such effects, 12 they're minor, I mean the primary pharmacologic 13 effect is through the benzodiazepine receptor. 14 Q. If Prozac affects only the 15 serotonin reuptake receptors, how is it that it 16 prolongs the half-life of benzodiazepines which 17 apparently have specific receptors, Doctor Zerbe? 18 A. Well, it has to do with a 19 completely different issue, and that is the 20 metabolism -- 21 Q. Okay -- 22 A. -- of the benzodiazepines. 23 And that's not related to the benzodiazepine 24 receptor, that's related to other metabolic Page 160 1 enzymes in the liver that break down 2 benzodiazepines, and those are affected by 3 fluoxetine. 4 Q. All right. So by that can we 5 say that fluoxetine affects other 6 neurotransmitters indirectly by virtue of 7 extending or shortening their metabolites? 8 A. Could you restate the 9 question? 10 (THE COURT REPORTER READ BACK THE 11 REQUESTED TESTIMONY.) 12 A. Well, benzodiazepines are not 13 really neurotransmitters, benzodiazepines are 14 drugs that interact with a specific receptor in 15 the brain, so I'm not sure that the statement is 16 quite correct. It certainly is true that 17 fluoxetine can affect the metabolism of other 18 pharmacologic agents. 19 Q. Okay. And by that, could it 20 be said that indirectly that fluoxetine affects 21 other -- neurotransmission of other 22 neurotransmitters even though it's a specific 23 serotonin reuptake inhibitor? 24 A. I don't believe you can make Page 161 1 that extrapolation. 2 Q. All right. Who would know 3 that? 4 A. Doctor Fuller would probably 5 be a good -- 6 Q. You're not the expert in that, 7 are you? 8 A. No, Doctor Fuller would 9 probably be better able to tell you whether 10 fluoxetine affects other neurotransmitters. 11 Q. Did you know that cocaine is a 12 serotonin reuptake inhibitor? 13 A. I know cocaine has effects, I 14 think, on a number of CNS neurotransmitters. 15 Q. Including reuptake of 16 serotonin? 17 A. I didn't know that 18 specifically, but that doesn't surprise me. 19 (PLAINTIFFS' EXHIBIT NO. 20 WAS 20 MARKED FOR IDENTIFICATION AND 21 RECEIVED IN EVIDENCE.) 22 Q. Exhibit 20 is a document dated 23 February 7, 1990 concerning Prozac safety reports 24 authored by Doctor Leigh Thompson, correct? Page 162 1 A. Yes. 2 Q. And we were talking earlier 3 about the great efforts that were going on in 4 connection with reviewing these reports of Prozac 5 and suicide and hostility, weren't we? 6 A. Yes. 7 Q. And Exhibit 20 is a reflection 8 of Doctor Leigh Thompson's emphasis that these 9 reports of suicide and hostility be handled 10 thoroughly, correct? 11 A. Yes. 12 Q. He says that -- and you're 13 copied in on this. Do you recall getting this 14 memo? 15 A. I don't recall this specific 16 memo. 17 Q. It appears that he's almost 18 frantic in this memo to me, doesn't it to you? 19 A. Well, you had to know Leigh, I 20 think, to put that into perspective. I mean 21 frantic isn't the way I would quite characterize 22 it. 23 Q. How would you characterize it? 24 A. I think this is a Page 163 1 demonstration of Leigh's commitment and the way 2 that he would express that. To some people it 3 might sound frantic, to these of us who worked 4 with him a lot, it's just a demonstration of how 5 important it was to him. 6 Q. All right. He says that 7 anything that happens in the UK can threaten this 8 drug, correct? 9 A. I think I saw that, but I 10 would want to read it. 11 Q. Third line from the top. 12 A. Anything that happens in the 13 UK can threaten this drug in the U.S. and 14 worldwide. 15 Q. What was occurring in the UK 16 at this time? 17 A. My recollection of this, and 18 the reason I think that Patrick Keohane was 19 copied -- Patrick Keohane was the medical 20 director in the UK at the time, I believe -- was 21 that Patrick Keohane had expressed some concern 22 about resources necessary to process adverse 23 events, and we simply reiterated the commitment 24 to see that even if other things are not Page 164 1 addressed, that Prozac adverse events are being 2 properly handled. That's my recollection of this -- 3 the background of this memo. 4 Q. All right. Doctor Thompson 5 says the FDA is very, very skitterish, does he 6 not? 7 A. Yes. 8 Q. Would you characterize the 9 Food and Drug Administration as being very, very 10 skitterish in February of 1990? 11 A. I probably wouldn't have used 12 those terms. 13 Q. Would you have used one very 14 as opposed to two veries? 15 A. The way -- it's a matter of 16 tone. I probably would have said they're very 17 concerned about it. Skitterish may be a bit 18 more, but then I didn't have the benefit of 19 talking directly to Doctor Leber. 20 Q. Is skitterish a John Hopkins 21 term or Cleveland Clinic term, which is that? 22 A. I don't know. 23 Q. All right. Well, he doesn't 24 even have a high school diploma, so he wouldn't Page 165 1 know. 2 A. He doesn't need one. 3 Q. He says every significant 4 event about Prozac has been a show stopper on the 5 twelfth floor. Meetings immediately with Earl, 6 Mel, et cetera, correct? 7 A. Yes. 8 Q. Had you been to some of those 9 meetings on the twelfth floor at that time? 10 A. I can't remember at that 11 specific time. I think I probably had been at 12 some meetings, but I don't remember any specific 13 meetings at that time. It's hard to put the time 14 perspective in appropriate position. 15 Q. At some point, you did become 16 a member of, what was it, Tuesday morning or 17 Thursday morning group of upper management that 18 met to discuss Prozac-related problems at least 19 once a week? 20 A. We met. I'm not sure it was 21 at least once a week, but we met not infrequently 22 on a regular basis. 23 Q. And there were senior 24 management members at that meeting -- at those Page 166 1 meetings, were there not? 2 A. Many times there were, yes. 3 Q. Many times Doctor Mel Perelman 4 was there? 5 A. Yes. 6 Q. Many times -- almost all the 7 time you and Doctor Leigh Thompson were there? 8 A. If we were in town. 9 Q. And Doctor Wood -- Mr. Wood 10 was even there on some occasions, was he not? 11 A. Occasionally. 12 Q. Doctor Herr was there on 13 occasion? 14 A. Yes. 15 Q. And these are, as we say in 16 Texas, big dogs at Eli Lilly and Company, are 17 they not? 18 A. Yes. 19 (DISCUSSION OFF THE RECORD.) 20 (PLAINTIFFS' EXHIBIT NO. 21 WAS 21 MARKED FOR IDENTIFICATION AND 22 RECEIVED IN EVIDENCE.) 23 Q. Doctor Zerbe, Exhibit 21 is 24 another document authored on the same day by Page 167 1 Doctor Leigh Thompson, is it not? 2 A. Yes. 3 Q. If you will look at it, in 4 fact it's dated two minutes later than Exhibit 5 20. 6 A. Yes, it is. 7 Q. You're not copied on this 8 memo. Have you seen this memo before? 9 A. I don't recall seeing it. 10 Q. Do you think this is the type 11 of memo you would recall seeing had it been 12 copied to you? 13 A. Had I been the primary -- 14 well, had I been the primary recipient at the 15 time, I certainly would have taken note of it. 16 Ten years later, eight years later, however long 17 ago it was, I may not have recalled this specific 18 memo. 19 Q. Doctor Leigh Thompson, two 20 minutes after talking about there can't be a 21 fumble of minor proportions and that everything 22 that happens with Prozac is a twelfth-floor show 23 stopper and that the FDA is very, very 24 skitterish, says as follows, I quote: I am Page 168 1 concerned about reports I get RE UK attitude 2 toward Prozac safety. Leber suggested a few 3 minutes ago that we use the CSM data base to 4 compare Prozac aggression and suicidal ideation 5 with other antidepressants in the UK. Although 6 he is a fan of Prozac and believes a lot of this 7 is garbage, he is clearly a political creature 8 and will have to respond to the pressures. I 9 hope Patrick realizes that Lilly can go down the 10 tubes if we lose Prozac, and just one event in 11 the UK can cost us that. You know my prejudice 12 about Patrick, but if I hear one more problem 13 about not covering Prozac safety in the UK, 14 Allen, I'm going to really be up in arms, signed 15 Leigh, end quote, correct? 16 A. Yes. 17 Q. Apparently you were right, 18 that what you and Doctor Thompson were concerned 19 about was resources to follow up with UK safety 20 reports, is that right? 21 A. I think that's what is 22 suggested here. 23 Q. Doctor Thompson indicates a 24 prejudice about Patrick Keohane, is that right? Page 169 1 A. Keohane. 2 Q. Keohane. What is his 3 prejudice about him? 4 A. Probably you would be best -- 5 it would probably be a question best addressed to 6 Doctor Thompson, and not me, exactly what his 7 prejudice is. 8 Q. Has he ever expressed anything 9 to you that would lead you to give us some 10 impression as to what his prejudice is, Doctor 11 Zerbe? 12 A. I can't remember any specific 13 situations that he has cited. I think there may 14 have been some personality conflict, but I don't 15 really know the specifics of what his, quote, 16 prejudice was against him, unless it's related to 17 the personality conflicts. But you could 18 probably address he and Doctor Weinstein about 19 what they're referring to there. 20 Q. Do you agree or did you agree 21 on February 7th, 1990 -- let me address that, 22 February 7, 1990. Did you agree that Lilly could -- 23 can go down the tubes if we lose Prozac? 24 A. No, I don't think I would have Page 170 1 put it quite in those terms at that point. 2 There's no question that Prozac was a critical 3 drug for the company, but going down the tubes 4 isn't quite the way I would have characterized 5 it. 6 Q. How would you have 7 characterized it had it been up to you to express 8 the importance of Prozac to Eli Lilly and 9 Company? 10 A. Well, I would have said 11 something probably more like, you know, this is a 12 key for growth in the future for the company. It 13 doesn't mean the company would end if we didn't 14 have it, but it means that we would be in a much 15 less favorable position. 16 Q. Have you heard the term 17 block-buster drug? 18 A. Yes. 19 Q. Would you agree that Prozac 20 was, and is, a block-buster drug for Eli Lilly 21 and Company? 22 A. Yes. 23 Q. Are you familiar with its 24 standing as how it's rated in numbers of Page 171 1 prescriptions as compared to other medications in 2 the United States? 3 A. I don't know the latest 4 figures, but I do know it's very high in terms of 5 the volume of sales and money that it generates 6 for the corporation. 7 Q. It says in this same memo that 8 Leber, meaning Paul Leber, head of the division 9 of neuropharmacology at the United States Food 10 and Drug Administration, correct? 11 A. Yes. 12 Q. Suggested a few minutes ago we 13 using the CSM data base to compare Prozac 14 aggression and suicidal ideation with other 15 antidepressants in the UK. Do you see that? 16 A. Yes. 17 Q. That never was done, was it? 18 A. Actually the CSN green card 19 system -- I guess is the CSM system; it's the 20 green card, yellow card system, whatever that is, 21 where patients that are given the prescription 22 are followed -- there was, as I recall, a report, 23 a safety report, that did come out of that much 24 later and did look at all types of adverse Page 172 1 events. So I think that that actually did 2 happen, but it was done at a later time. 3 Q. Are you talking about Inman's -- 4 A. Yes, Inman's. 5 Q. -- surveillance -- not 6 surveillance, but -- I guess it was sort of a 7 surveillance. 8 A. Yes. 9 Q. That's not what Leber's 10 suggesting here, is he? 11 A. I think the Inman data would 12 have included that analysis. It talks about the 13 outcomes regardless of what type of outcomes or 14 adverse events. 15 Q. Read this with me. 16 A. Okay. 17 Q. What Doctor Leber was 18 suggesting was using the CSM data base to compare 19 Prozac aggression and suicidal ideation with 20 other antidepressants in the UK. It's a 21 completely different thing than a surveillance 22 study, isn't it, this doesn't say anything about 23 a surveillance study, does it? 24 A. I don't know if it isn't a Page 173 1 surveillance study, the Inman data, I don't know 2 what CSM data base he's referring to. I think 3 that must be the Inman data base unless you have 4 information to the contrary, because that's the 5 only data base that would allow that kind of 6 assessment. 7 Q. Doesn't the CSM have their own 8 data base? 9 A. You mean in terms of adverse 10 events? 11 Q. Yes. 12 A. Yes, but -- 13 Q. That's what I'm talking about 14 and that's what he's talking about too, isn't it? 15 A. Well, I don't know. To me, 16 that would not -- one would not be able to 17 address the question of aggression and suicidal 18 ideation based on a spontaneous event adverse 19 event data base, I don't believe. 20 Q. Why? 21 A. Well, because you don't -- you 22 don't have a denominator. That's one of the 23 chronic problems with any spontaneous event data 24 base. Page 174 1 Q. That's what Leber is 2 suggesting that be done, though, isn't he? 3 A. Again, my reading of this 4 would be in reference to the Inman data base, but 5 maybe -- 6 Q. It doesn't say a thing about 7 the Inman data base, does it? 8 MR. LORE: That's what his reading is, 9 Paul, and now you're arguing with him. And he's 10 given you his answer, and you don't like his 11 answer, but that's the answer he's given. 12 MR. SMITH: I certainly don't because 13 it doesn't say Inman on here, it says CSM data 14 base. 15 Q. There is a separate CSM data 16 base, is there not, for adverse events that occur 17 in the United Kingdom? 18 A. CSM maintains -- or the UK 19 regulatory agency maintains information about 20 spontaneous events. 21 Q. As does the United States Food 22 and Drug Administration. 23 A. Right. 24 Q. And when we talk about the Page 175 1 U.S. FDA spontaneous reporting system, we're 2 talking about something very specific, aren't we? 3 A. Yes. 4 Q. All right. Now, when we're 5 talking about the English CSM data base, I was of 6 the assumption that what we would be talking 7 about would be the British equivalent or 8 counterpart to the FDA spontaneous reporting 9 system. Am I wrong? 10 A. I believe you're wrong. You 11 may have to ask Doctor Thompson about 12 specifically what he meant. 13 Q. Well, it would probably be 14 best to ask Doctor Leber about what he meant, 15 wouldn't it, because it was Doctor Leber that 16 made that suggestion, wasn't it? 17 A. Maybe Doctor Leber may or may 18 not even remember the conversation in the way 19 that Doctor Thompson has conveyed it. With 20 regard to the specific memo, in putting the facts 21 together I would have thought that Doctor 22 Thompson was referring to the Inman data base, 23 which would have been much more suitable for 24 addressing this question than any spontaneous Page 176 1 event data base, whether it was created by the 2 CSM or the FDA. 3 Q. Because that was a 4 postmarketing surveillance study? 5 A. Well, because it had no 6 denominator and was affected by so many other 7 factors than simply the incidence of those 8 events, the preference for reporting on new 9 products versus old products, et cetera. So it 10 really would have been very difficult to 11 interpret, and I think Doctor Thompson would have 12 also -- would have also interpreted Doctor 13 Leber's suggestion in that way. But that's 14 speculation on my part, I think you would have to 15 ask Doctor Thompson what he meant in his memo. 16 Q. Just so I'm clear, we have 17 information that a Doctor Inman did a 18 surveillance study in England, correct? 19 A. Yes -- I'm sorry, I don't know 20 what information you have, but my recollection 21 was that Doctor Inman did a surveillance of 22 fluoxetine post-marketing in the UK. 23 Q. All right. And a surveillance 24 study is kind of a fifty/fifty deal, isn't it, Page 177 1 it's sort of in between a clinical trial and 2 post-marketing spontaneous report system data, is 3 that right? 4 A. Basically, yes. 5 Q. And it gives you kind of a 6 third arm? 7 A. It's another way of addressing 8 the question, yes. 9 Q. And Inman was something that 10 was done independently, wasn't it? 11 A. Yes. 12 Q. And was that of some 13 particular benefit to Lilly? 14 A. Well, the Inman system for 15 collecting those adverse events that occur with 16 medications is particularly useful because it's 17 not completely dependent upon spontaneous 18 reporting. 19 Q. Okay. It doesn't depend on 20 spontaneous reporting, right, and it's something 21 independent from the manufacturer. 22 A. Yes. 23 Q. You don't have a reporting 24 bias, right? What was Inman paid by Lilly to do Page 178 1 the surveillance study -- Inman was paid, wasn't 2 he? 3 A. I don't think Inman was paid 4 specifically to do the surveillance study, but 5 Inman has been supported by contributions, as I 6 recall, both before and after the fluoxetine 7 surveillance program. And I think his choice of 8 programs to do surveillance projects on is his, 9 it's not related to company sponsorship. 10 MR. SMITH: Let's take a break. 11 (A SHORT RECESS WAS TAKEN.) 12 (PLAINTIFFS' EXHIBIT NO. 22 WAS 13 MARKED FOR IDENTIFICATION AND 14 RECEIVED IN EVIDENCE.) 15 Q. Exhibit 22 is a document dated 16 September 12th, 1990 authored by Doctor Leigh 17 Thompson, is it not? 18 A. Well, there are two messages 19 here, part of it's authored by, it looks like Max 20 Talbott, and the other is by Leigh Thompson. 21 Q. But they go hand in hand, do 22 they not? 23 A. Yes. 24 Q. And the document authored by Page 179 1 Leigh Thompson is the first one in order, 2 correct? 3 A. That's correct. 4 Q. And Max is just following up 5 with some additional information on processing, 6 is he not? 7 A. Yes. 8 Q. In that portion authored by 9 Doctor Leigh Thompson, he designates his memo as 10 urgent, does he not? 11 A. Yes. 12 Q. You're copied in on the memo 13 or the memo is directed in part to you? 14 A. Yes. 15 Q. As well as psychiatrists at 16 Lilly, correct? 17 A. I'm not sure any of the 18 primary copies are psychiatrists. 19 Q. Well, Doctor Allen Weinstein 20 is a psychiatrist, isn't he? 21 A. No. 22 Q. All right. It's also directed 23 to Doctor Mel Perelman, is it not -- 24 A. Yes. Page 180 1 Q. -- president of Lilly Research 2 Labs at the time? 3 A. Yes. 4 Q. Correct? 5 A. Yes. 6 Q. He says that he has spoken on 7 the telephone with Doctor Paul Leber at the FDA 8 concerning a meeting with Doctor Bob Temple with 9 the FDA in the next couple of days, correct? 10 A. Yes. 11 Q. And that he wanted to be 12 brought up-to-date concerning suicidality and 13 Prozac, correct? 14 A. Yes, that Doctor Temple wanted 15 to be brought up to speed. 16 Q. Yes. Doctor Temple was what 17 with the FDA? 18 A. Doctor Temple was the head of 19 one of the two new drug divisions at the FDA. 20 Q. Doctor Temple was superior to 21 even Doctor Paul Leber? 22 A. That's correct. 23 Q. At the time? 24 A. Yes. Page 181 1 Q. It mentions Doctor Carl Peck 2 also, does it not? 3 A. I didn't see that, but yes, he 4 did. 5 Q. And Doctor Peck was in between 6 Doctor Temple and Doctor Leber? 7 A. No, I don't believe so, Doctor 8 Peck was Doctor Temple's boss. 9 Q. All right. Doctor Thompson 10 says, I think this means that he, that is Leber, 11 is being pushed by Temple, paren, and from Peck's 12 comments yesterday, at least Peck is concerned, 13 close paren, to change the label, right? 14 A. Yes. 15 Q. So there was concern that the 16 FDA would require some label changes in 17 connection with Prozac and suicidality? 18 A. There was concern about the 19 FDA forcing some label change, yes. 20 Q. And in the last paragraph he 21 says, I am now very concerned that Temple et al 22 may force a label change even before we get there 23 on 25 September or, next worse, have this a fait 24 d'acompli when we arrive, correct? Page 182 1 A. Yes. 2 Q. Then he goes on to say -- 3 Doctor Thompson goes on to say, that report must 4 move swiftly through approval and to Doctor 5 Leber's hands, he is our defender, end quote, 6 correct? 7 A. Yes. 8 Q. Was it your experience that 9 Doctor Leber was Lilly's defender at the FDA? 10 A. Well, it depends on how you 11 characterize defender. I think Doctor Leber 12 could be relied upon to look at data objectively, 13 but I never felt like he had any -- that he gave 14 us a particularly easy ride through all of this. 15 He was very demanding that we maintain high 16 scientific standards and address his questions, 17 and I think he maintained a very balanced view. 18 Q. So do you disagree with Doctor 19 Thompson when he says he is our defender? 20 A. Again, I think you would have 21 to ask Doctor Thompson exactly what he meant by 22 our defender. Knowing Doctor Thompson, my guess 23 is that he would summarize that the same way that 24 I have summarized it, and that is that he was Page 183 1 very balanced and tough, and as long as we had 2 the correct scientific information, he would 3 support us. 4 Q. I guess, being a lawyer, when 5 I see the term defender, I see somebody like Mr. 6 Lore sitting beside you as Lilly's defender in 7 this case. I see him as an advocate for Lilly, 8 doing what he can to put forth Lilly's position 9 in connection with the subject. Is my 10 interpretation of Doctor Leber being a defender 11 of Lilly or Prozac not analogous to my analogy to 12 Mr. Lore or Mr. Myers or Mr. Freeman, even, of 13 being Lilly's defender? 14 MR. LORE: I object, I think he's 15 already answered your question. You can go ahead 16 if you want to say something further. 17 A. I would say that's not at all 18 the experience that I had. I think Doctor Leber 19 was, again, very balanced, scientific, objective, 20 and very demanding of Lilly. 21 Q. So you deny that Doctor Leber, 22 as an employee of the Food and Drug 23 Administration, was an advocate for Lilly or 24 Prozac? Page 184 1 A. I deny that, yes, or I 2 disagree with that statement. 3 (PLAINTIFFS' EXHIBIT NO. 23 WAS 4 MARKED FOR IDENTIFICATION AND 5 RECEIVED IN EVIDENCE.) 6 Q. Exhibit 23 is a document dated 7 October 2, 1990 authored by Doctor Leigh Thompson 8 directed to you, is it not? 9 A. Yes. 10 Q. Concerning materials for a 11 November 10th Prozac symposium, correct? 12 A. Yes. 13 Q. He describes the symposium as 14 being data for this Washington symposium. Do you 15 know what Washington symposium was occurring in 16 November, 1990? 17 A. I think he's referring to the 18 November 10th Prozac symposium which is, I think, 19 the Washington symposium, and it was an annual or 20 every-other-year scientific review, CNS review, 21 and there was a section in that review on 22 clinical -- new clinical observations, as I 23 recall. 24 Q. Was this something that was Page 185 1 sponsored by the Food and Drug Administration or 2 required by the Food and Drug Administration? 3 A. No, I don't believe so. The 4 one I'm thinking of definitely was not. 5 Q. Who put that symposium on? 6 A. The one I'm referring to was 7 supported by Lilly. 8 Q. All right. And would that be 9 a situation where health care providers and 10 people in the pharmaceutical industry would be 11 invited to attend this symposium on an 12 educational basis? 13 A. Yes. 14 Q. He talks about you getting 15 some data for him in connection with the 16 symposium, correct? 17 A. Yes. 18 Q. And data on Prozac experience, 19 correct? 20 A. Yes. 21 Q. Then he says in the next to 22 the last paragraph, then the question is what to 23 do with the big numbers on suicidality. If the 24 report numbers are shown next to those for Page 186 1 nausea, they seem small. Should I get into the 2 total count at all, paren, I'm afraid someone 3 will ask for sure, close paren, or should I leave 4 this to David to introduce, question mark, end 5 quote. Did I read that accurately? 6 A. Yes, I believe so. 7 Q. Do you recall getting this 8 memo? 9 A. I don't recall this specific 10 memo. 11 Q. What were the numbers on 12 suicidality in October, 1990? 13 A. I don't know exactly what 14 those numbers were, but -- 15 Q. I mean generally. 16 A. We were, at that time, getting 17 many reports of suicidality and suicidal ideation 18 because of all the publicity that was surrounding 19 Prozac at the time. 20 Q. Now, let me stop you a second, 21 Doctor Zerbe. In your answer you give a direct 22 answer, and then you give a qualifier, and you 23 say we were getting many reports of suicides and 24 suicidality, which is a factual statement, is it Page 187 1 not? 2 A. Yes. 3 Q. Then you say because of the 4 publicity and media attention. What information 5 do you have or did you have then that supports 6 your contention that the reason you were getting 7 this number of reports of suicidality was because 8 of any particular media attention? 9 A. Well, there were a number of 10 things that were done to try to address that 11 question. First of all, it's not an unusual 12 observation, it's been observed with other 13 compounds, in fact it's been published that you 14 can see blurbs of activity associated with 15 publicity. 16 Q. Tell me specifically what 17 publications you have in mind that support that 18 position, that it's known in the scientific 19 literature that you get -- if you get media 20 publicity you're going to have additional reports 21 of a particular adverse event in the spontaneous 22 reporting system. 23 A. The paper that comes to mind, 24 and I can't quote the citation, per se, but there Page 188 1 was a paper published by Webber or Weber, or 2 something, looking at frequency of adverse 3 events, particularly with new compounds, and what 4 things might influence that. 5 Q. All right. 6 A. And I don't know what the 7 citation is, but -- 8 Q. When was that reported? 9 A. I don't know the exact date, 10 it would have been a few years before all of this 11 came up. 12 Q. So that wouldn't bear directly 13 on the relationship of Prozac and suicide, Prozac 14 wasn't even discussed in that article, was it? 15 A. No, no. 16 Q. What other facts did you have 17 then or do you have now that indicate your 18 qualifier to your factual answer that we were 19 getting a high number of reports of suicidality 20 by virtue of the media attention? You mentioned 21 the one article. 22 A. Well, we looked at the time 23 course of those events related to publicity, and 24 there was a fairly dramatic burst of activity Page 189 1 following the publicity. There's no reason to 2 believe that if this were a real event, 3 uninfluenced by publicity, that the shape of that 4 curve would have been any different with regard 5 to the sales before and after the publicity 6 related to Doctor Teicher's paper. So if it were 7 a real event and people's behavior was not being 8 impacted by the publicity, the curves should have 9 been parallel to the sales. 10 Q. Okay. Do you know how many 11 people read Doctor Teicher's article? 12 A. Well, I'm not -- I don't know 13 how many people read Doctor Teicher's article, 14 but that's not really the point. It wasn't how 15 many people read Doctor Teicher's article, it was 16 related to the publicity that flowed from Doctor 17 Teicher's article, from the activities of the 18 Church of Scientology, and from the Trial Lawyers 19 group, and their appearances on T.V. and so 20 forth. 21 Q. Wait a second. You all were 22 appearing on T.V. too, weren't you? 23 A. Well, our responses did not 24 help that issue at all, it didn't decrease, all Page 190 1 we were doing was responding, but you know it's 2 publicity nonetheless, and there still were 3 people that were reporting probably even as a 4 result of what we were saying. It's an awareness 5 of it, it's a greater frequency of reporting as a 6 result of increased concern about the issue, good 7 or bad. 8 Q. Isn't it a fact, Doctor Zerbe, 9 that ninety-five percent of the reports to the 10 spontaneous reporting system were made by Lilly 11 to the FDA concerning suicidality? 12 A. Well, I mean, we were made 13 aware of those and subsequently reported those, 14 as we were obligated to do, and routinely did. 15 Q. So if I looked at the total 16 number of 1639s on suicidality, which in fact I 17 have done, I would see ninety-five percent of the 18 person that generated this report to the FDA as 19 being Eli Lilly and Company, wouldn't I, you 20 agree with that, don't you? 21 A. Probably, yes. 22 Q. And the next question becomes, 23 where did Lilly get those reports, doesn't it? 24 A. Those came from either people Page 191 1 calling the company directly or some Lilly 2 employee being made aware of such incidents when 3 they were informed by a physician or somebody 4 they called on, or whatever. 5 Q. The majority of those reports 6 came from Lilly's own detail people, didn't they? 7 A. I imagine that's true, I don't 8 have those figures, but -- 9 Q. You wouldn't dispute that the 10 majority of the reports of suicidality in 11 connection with Prozac and violent aggressive 12 behavior came from reports by Lilly to the FDA -- 13 came from reports by Lilly to the FDA that Lilly 14 had learned about by virtue of their contact with 15 physicians through their detail people, through 16 their sales people, right? 17 MR. LORE: If you know, Doctor. 18 A. I think that most reports -- I 19 don't know about Prozac specifically, but most 20 reports came through such contacts. 21 Q. Of the detail people talking 22 to physicians, that they were selling Prozac to, 23 right? 24 A. They were detailing it, yes. Page 192 1 Q. Okay. It didn't come -- those 2 reports didn't come from the media, they came 3 from Lilly themselves by virtue of their contacts 4 with their market, that is the doctors of this 5 country? 6 A. Well, Lilly was only serving 7 as the reporter, the reports came from 8 physicians. 9 Q. Indeed. 10 A. And physicians, just like 11 everybody else, are more aware of, as a result of 12 publicity, more aware of such events. 13 Q. But those physicians were 14 reporting to you instances where they had had 15 patients that had experienced this problem, 16 correct? 17 A. Well, maybe, maybe not. Most 18 likely that's the case, but not necessarily. 19 Q. You don't file a 1639, Doctor 20 Zerbe, in a situation where a physician reports 21 to your detail person, hey, by the way, do you 22 know anything about Prozac and suicidality, I've 23 been reading a lot about it in the paper. 24 A. Well -- Page 193 1 Q. You don't make a 1639 in that 2 situation, do you? 3 A. You have to have an 4 identifiable patient, -- 5 Q. Right. 6 A. -- an event, and a drug. 7 Q. Patients have to have reported 8 this, and a patient has to have reported it to a 9 physician, and the detail person has to report it 10 to Lilly, and then Lilly reports it to the Food 11 and Drug Administration, correct? 12 A. Right. 13 Q. And that's where you get these 14 large number of reports? 15 A. But the awareness of that and 16 the interest in discussing it, the changes in the 17 numbers still could be influenced by publicity. 18 Q. To a large extent? 19 A. To a very large extent. 20 Q. Okay. Tell me of any instance 21 that you become aware of where a physician has 22 told you, I have had a patient coming in to me 23 who has claimed they are now suicidal after 24 taking Prozac, and they say that the reason they Page 194 1 even began thinking about suicide was not because 2 they took Prozac, but because of the vast media 3 reports or seeing some particular lawyer on T.V.? 4 A. Are you asking me to cite a 5 specific patient case? 6 Q. Yes. 7 A. I can't do that, but there 8 were patients, there clearly were descriptions of 9 patients that I received secondhand that were 10 influenced by the publicity -- 11 Q. How many? 12 A. -- stopped their medication as 13 a result of the publicity, and had problems 14 associated with that. Now how many, I don't 15 know, I don't know the details, but -- I can't 16 cite specific cases, but there were such cases 17 that were reported to me. 18 Q. You see, I don't understand it 19 and maybe you can help me, Doctor Zerbe. When 20 you tell me the vast reason or the reason for 21 this vast increase in reports of adverse events 22 in connection with suicidality and Prozac has to 23 do with media publicity and lawyers going on 24 T.V., you make that statement, and yet I don't Page 195 1 see it when the report stems from a patient 2 talking to their physician, and the physician 3 reporting it to a Lilly detail person. 4 A. Well, I think -- first of all, 5 again, it's my opinion that it can impact the 6 patient reporting incidents to the physician, it 7 can impact the types of questions the physician 8 asks the patient, they're now concerned about 9 that and they'll ask the patient and the 10 depressed patient may say well, yes, I did have 11 some ideation, it may or may not be related to 12 the drug, but did have some ideation. So it 13 influences the patient, it influences the 14 questions the physician asks, it influences the 15 response that the physician has to the sales rep. 16 All of those things are influenced by that 17 publicity and increase the rate at which such 18 events are reported. 19 Q. Are you saying that maybe the 20 physician is reporting or is making a more 21 detailed and serious inquiry of their patient as 22 to the existence or nonexistence of this 23 particular phenomena in their patient? 24 A. I'm saying that could be an Page 196 1 explanation. 2 Q. And of course didn't occur 3 during the clinical trials. 4 A. No. 5 Q. All right. 6 A. No, wait, that's not true that 7 it didn't occur during clinical trials. Suicide, 8 suicidal ideation was routinely evaluated in the 9 clinical trials. 10 Q. Only by virtue of the item 11 three on the Hamilton depression scale? 12 A. And any adverse events that 13 may have been related outside of that, if 14 somebody had suicidal ideation as an adverse 15 event. 16 Q. But that wouldn't have been 17 elicited from the patient. The only elicitation 18 of the patient in the clinical trials as to 19 whether or not they're experiencing any suicidal 20 ideation or suicidality or violent aggressive 21 thoughts or behavior is by virtue of the item 22 three of the Hamilton depression scale. 23 A. That was the main item. It 24 could be reported in other ways, but elicitation Page 197 1 was through the Hamilton depression scale. 2 Q. Now, what you're saying, it 3 sounded like to me, is now in the post-marketing 4 experience, since the physicians are more aware 5 that this may be something that is related to the 6 drug, that the physicians are asking their 7 patients about it when before they weren't. 8 A. Well, first of all, I'm not 9 saying before that none of them asked, I'm not 10 implying that at all, some did. What I'm saying 11 is they would more consistently ask, perhaps, 12 they might dwell on it more, they might, as a 13 result of the publicity, reach conclusions about 14 something the patient said, they may follow up 15 more aggressively on the item because of the 16 publicity. 17 Q. Do you think that's 18 necessarily bad? 19 A. I didn't say it was 20 necessarily bad, but what I thought we were 21 explaining was the change in the frequency of the 22 reporting. 23 Q. We are, we are. Why are you 24 all of a sudden seeing -- if you say it's all of Page 198 1 a sudden, what accounts for this increase in the 2 number of reports? And you've indicated that 3 there's an article that was published that 4 indicated -- or publicity -- you're liabel to 5 have more reports. 6 MR. LORE: And he's also indicated 7 that after this publicity, although the sales 8 went up, the number of complaints went down. And 9 he also said that, Paul, so don't leave that out. 10 MR. SMITH: No, what he said was that 11 there wasn't a corresponding increase in the 12 number of adverse events reported corresponding 13 to the number of increase in sales. 14 MR. LORE: I thought that's what I 15 said. 16 MR. SMITH: No, it's not. 17 A. Let me restate -- 18 MR. LORE: Restate, I'm sorry. 19 A. What I said was that the 20 change in the slope -- I think you do make 21 another very good point that I didn't make, but 22 the change -- 23 MR. SMITH: See, he didn't make the 24 point by his own admission. Page 199 1 A. Let me make it now. The 2 change in the slope during the publicity was much 3 greater for the reporting of suicide and suicide 4 related events than was the growth in sales. The 5 point, however, can also be made that following 6 the diminishment, if that's a word, the 7 diminishing of the publicity, the rates first 8 leveled off, and then did appear to start 9 actually declining despite the fact that sales 10 continued to increase. And this phenomenon is 11 not unique to Prozac, it's been observed with 12 other drugs and publicity associated with it. 13 Q. But you indicated to me that 14 physicians are being more careful in their 15 examination of patients to see if this phenomena 16 is occurring? 17 MR. LORE: I don't think that he said 18 that, I don't think that that's what he 19 testified. 20 Q. They're questioning their 21 patients more carefully. 22 A. That -- as ways that this 23 could occur, maybe they're questioning them more 24 carefully, maybe they're following up on events Page 200 1 more, maybe they're just reporting things more 2 frequently, just because it's at the top of their 3 mind and in the minds of the sales reps. 4 Q. Let's be fair to everybody 5 concerned, Doctor Zerbe, would we do that for 6 this occasion? In fact what was also occurring 7 was Lilly was writing the physicians and telling 8 them there wasn't any association between Prozac 9 and suicidality, weren't they? 10 A. Lilly was -- Lilly responded 11 in a letter to physicians, I don't know exactly 12 when or how it relates to this time frame, 13 summarizing the data that we had, and it was very 14 carefully worded to be a balanced actual 15 statement. 16 Q. Oh, the conclusion of it was 17 that there was no association between Prozac and 18 suicidality, wasn't there? 19 A. That's right. 20 Q. Okay. So Lilly is giving the 21 doctors as much attention about this subject as 22 the media is? 23 A. No -- 24 Q. Lilly wrote every doctor that Page 201 1 they knew of that prescribed Prozac, didn't they? 2 A. I think those things occurred 3 in a completely different time frame. 4 Q. Tell me who got the Dear 5 Doctor letters explaining Lilly's position in 6 connection with Prozac and suicidality and 7 violent aggressive behavior? 8 A. You mean who received it? 9 Q. Yes, which doctors. 10 A. I don't know the distribution 11 of those. 12 Q. How were the doctors selected 13 to receive these letters? 14 A. I was not involved in that 15 selection. 16 Q. Would it be accurate to say 17 that every member of the American Psychiatric 18 Association received a letter? 19 A. I don't know. 20 Q. Would it be accurate to say 21 that every member of the American Medical 22 Association received the letter? 23 A. I don't know, I really don't 24 know. Page 202 1 Q. How many letters were sent? 2 A. I think there were many 3 letters sent. 4 Q. Thousands? 5 A. I imagine thousands. 6 Q. My question simply is, where 7 did you get the mailing list? 8 A. I don't know, I don't know 9 where the mailing list came from. 10 Q. Did you get them from Lilly's 11 detail people? 12 MR. LORE: He's already said he 13 doesn't know, Paul. 14 A. I don't know. 15 Q. Does Lilly maintain some type 16 of data on doctors that they call on to make them 17 aware of their product? 18 A. You know, I really don't know 19 the mechanism, but I think it is fair to 20 emphasize that, on your point, that those events 21 occurred in very different time frames. The 22 occurrence of the publicity preceded, and was 23 much more closely correlated, with the increased 24 frequency of events than the letter that went to Page 203 1 physicians, which came later. 2 Q. I think the facts will bear 3 out that the Teicher article was published in 4 February, 1990, all right, and by June, 1990, 5 Lilly had a letter to all the doctors saying 6 there was no association with Prozac and 7 suicidality, all right? 8 A. I don't know the dates. 9 Q. Do you dispute that, does that 10 sound reasonable to you? 11 A. There was a -- well, I just 12 don't know the dates. That sounds earlier than I 13 had remembered it, but it still -- the point 14 still stands, that frequency of adverse events 15 preceded by a substantial amount that letter. 16 Q. Let's go back to Exhibit 23, 17 Doctor Thompson's statement. Then the question 18 is what to do with the big numbers on 19 suicidality. What did you do about it? 20 A. Well, what he, I think, is 21 asking is how should that question be handled, 22 how should the numbers in adverse events be 23 handled with regard to suicidality. 24 Q. Yes, because it goes on to say Page 204 1 the numbers are shown next to those for nausea, 2 they seem small. 3 A. To be honest, I don't recall 4 how it was presented. The issue of suicidality 5 was presented at that conference, as I recall, 6 and what he's referring to here is David Wheadon 7 made a presentation on the data that was 8 available at that time as I believe. 9 Q. Do you know how the big 10 numbers on suicide were presented at that 11 symposium? 12 A. I really don't, I just don't 13 remember. 14 Q. Were they presented next to 15 nausea so they would seem small? 16 A. I don't remember, I really 17 don't remember. 18 Q. Were the numbers on 19 suicidality in October of 1990 big? 20 A. Well, big is a relative term. 21 They were small next to nausea, but they were -- 22 Q. Big next to rash? 23 A. Well, in spontaneous or 24 clinical trials? Page 205 1 Q. Either. 2 A. In the clinical trials they 3 were small next to rash. In spontaneous, they 4 would have been at least closer in terms of size, 5 but probably still much smaller. 6 Q. How about insomnia? 7 A. I mean you're asking me to 8 speculate, we could go on forever. Probably the 9 facts should speak for themselves. 10 Q. The reason is because you said 11 I don't know whether the numbers are big or not, 12 it's relative. I'm trying to get some kind of -- 13 A. I was just making a general 14 statement about big is a relative term in trying 15 to answer that question. 16 Q. Isn't big a real big term if 17 there is causality and one person commits suicide 18 as a result of using this drug, it's going to be 19 big however you cut it, isn't it? 20 A. Well, the causality question 21 is a completely separate issue. Patients who are 22 depressed commit suicide, so there are going to 23 be patients that commit suicide in that trial. 24 The question is does it have anything to do with Page 206 1 the treatment, and there's not any evidence that 2 that was the case. 3 Q. If there is -- well, I don't 4 know if there's evidence that that's the case, 5 your own scientists have said it's possibly, 6 reasonably related in a number of instances. So 7 that raises a question. Now, if in one of those 8 instances it was correct in the use of Prozac was 9 related to the patient's suicide, is that going 10 to be a big number, even if it's just one? 11 A. We're always concerned about 12 the safety of any patient. 13 Q. If one person loses their life 14 as a result of using Prozac, is that an 15 acceptable risk? 16 A. Well, if in general the 17 benefit is greater than that risk, it's a wise 18 decision to use the medication. I'm not sure 19 that I understand your point. 20 Q. Is Prozac safe in your 21 judgment? 22 A. Yes. 23 Q. If Prozac caused one person to 24 commit suicide, would it be safe in your Page 207 1 judgment? 2 A. On balance, if it prevented 3 five hundred suicides, then the balance is 4 favorable in terms of the risk/benefit. It's 5 like saying -- to turn it around, it's like 6 saying if one person dies in an airplane crash, 7 should all airplanes be banned? 8 Q. Do you think -- do you know of 9 any instance, any factual instance yourself that 10 indicates that anybody's life has been saved as a 11 result of Prozac? 12 A. That's very hard to assess. 13 What you can say is that their depression can be 14 successfully treated with Prozac. 15 Q. Wait a minute. Prozac has 16 never been shown to be any more efficacious in 17 the treatment of depression than any other 18 antidepressant, has it? 19 A. That wasn't the question, the 20 question was whether it had saved any lives. 21 Q. It's not any more likely to 22 alleviate depression than any other 23 antidepressant, is it? 24 A. There's not reason to believe Page 208 1 that it's superior from an efficacy standpoint, 2 but there's no reason to believe that it's 3 inferior from a safety standpoint. 4 Q. I didn't say it was, I didn't 5 say it was inferior as far as efficacy. You're 6 just saying, you know, if Prozac can alleviate 7 depression, then it may save lives. But other 8 antidepressants alleviate depression, and Prozac 9 is no more efficacious in alleviating depression 10 than any other antidepressant, is it? 11 A. I think I answered that. 12 Q. That it's not? 13 A. That there's no evidence that 14 it's superior. I can't say that unequivocally in 15 all studies it's never been demonstrated to be 16 superior, but in general, it's roughly the same 17 from an efficacy standpoint. 18 Q. The balance is in some studies 19 it was shown more efficacious and other studies 20 it was shown less efficacious. 21 A. That's right. But your 22 question is was there any evidence, and there may 23 be some evidence, but in general, efficacy is 24 similar. Page 209 1 Q. Right. So, my question again 2 is, with that in mind, since Prozac generally is 3 not any more efficacious in relieving depression 4 than any other antidepressant, how can you make 5 the statement that Prozac may have prevented some 6 individuals from committing suicide? 7 A. I think that statement is very 8 easy to make. Your question was did it save any 9 lives, and my argument is yes, it did, because it 10 successfully treated depression. 11 Q. But other medications 12 successfully treated depression. 13 A. Sure, and other medications 14 have saved lives too. 15 Q. So, do you know of any 16 instance of any individual that would have 17 committed suicide but for the use of Prozac? 18 A. That's -- I mean, that's a 19 very difficult question. How do you know the 20 person would have committed suicide when you 21 prevented the suicide from occurring? All you 22 can do is approach it from a study standpoint. 23 Q. Do you know of any instance 24 where the taking -- Page 210 1 A. I don't know of any instance 2 personally. 3 Q. -- of Prozac prevented an 4 individual from committing suicide? 5 A. One hears anecdotes regarding 6 Prozac and how it saved people's lives. 7 Q. One hears anecdotes regarding 8 Prozac and how it ruined people's lives, too. 9 A. And I think those are both in 10 the same category in terms of validity. 11 Q. Okay. So what did you do with 12 the big numbers on suicide? 13 A. I don't remember how the 14 numbers were presented, Doctor Thompson might be 15 able to tell you from his recollections how that 16 was handled. 17 (PLAINTIFFS' EXHIBIT NO. 24 WAS 18 MARKED FOR IDENTIFICATION AND 19 RECEIVED IN EVIDENCE.) 20 Q. Exhibit 24 is a document dated 21 June 11, 1991 directed to you, authored by Doctor 22 Leigh Thompson, is it not? 23 A. Yes. 24 Q. Is it your handwriting? Page 211 1 A. No, it doesn't look like my 2 handwriting. 3 Q. It may have come from somebody 4 else's files? 5 A. Yes, that's my guess, it did 6 come from somebody else's file. 7 Q. Do you recall receiving this 8 note from Doctor Thompson? 9 A. Again, only -- I don't recall 10 the specific memo, it would be -- I recall the 11 general tone, the background. 12 Q. Is it in general, this Doctor 13 Thompson's comments, in preparation for the 14 advisory committee meeting that ended up 15 occurring in September of 1991? 16 A. That appears to be the nature 17 of the memo. 18 Q. In the first paragraph, he 19 says what do we have already done on epidemiology 20 that could be referenced in an executive summary. 21 What would an executive summary be? 22 A. Generally, in preparation for 23 things like the advisory committee, you would 24 submit a document, and it would be -- at the Page 212 1 beginning you would have like an executive 2 summary which would summarize the key points. 3 Q. And in fact there was a 4 document submitted by Lilly to the advisory 5 committee that did have a summary at the outset 6 summarizing data and conclusions, did it not? 7 A. It was submitted to the FDA, 8 who then provided to the members of the advisory 9 committee. 10 Q. Doctor Thompson then goes on 11 and says, I think you have a white paper prepared 12 on incidence outcome, frequency of acts, and 13 suicide, but do we also have included the new 14 stuff we found on the increased incidence of 15 violence, end quote. Do you see that? 16 A. I see it, yes. 17 Q. Then there's a note out there 18 that says JHH, Prozac only, is that right? 19 A. That's what it appears to be 20 to me. 21 Q. And then does it say -- what's 22 it say under that, antidep patients? 23 A. That's what it looks like to 24 me. Page 213 1 Q. Or antidepressant patients? 2 A. Yes, that's what it looks like 3 to me. 4 Q. What data -- are you aware of 5 any new stuff that indicated an increased 6 incidence of violence? 7 A. Well, I tell you, it's very 8 difficult to know exactly what Leigh is referring 9 to here. 10 Q. I'm simply asking you, Doctor 11 Zerbe -- 12 A. In my opinion -- 13 Q. No, were you aware of any 14 stuff that had been found on an increased 15 incidence of violence? 16 A. Not related to Prozac. 17 Q. All right. Now, additionally, 18 were you aware that there was an effort made to 19 get the names of the advisory committee members 20 to know what their background was, what they had 21 written, and things of that nature? 22 A. I don't think there was ever 23 really such an effort going beyond just 24 understanding the names and the institutions, I Page 214 1 don't think there was ever a systematic review of 2 the background of those members of the advisory 3 committee. 4 Q. How do you know? 5 A. I don't recall that happening. 6 I don't ever remember seeing any summary that 7 said here are the people and here's the -- and I 8 was in a position that that would have made sense 9 for me to know that had I done it because we were 10 the ones designing the presentation. 11 Q. Well, and you had some of your 12 investigators on the advisory committee, didn't 13 you? 14 A. I don't think there were 15 investigators, only one, maybe, but that was all 16 disclosed, that's a matter of public record. 17 Q. Did you disclose the fact that 18 Doctor Dunner had been one of your investigators? 19 A. I think that was disclosed at 20 the advisory committee. 21 Q. And he was granted a waiver? 22 A. I don't know how it was 23 handled, but my only point was it was disclosed. 24 Q. The fact is, Doctor David Page 215 1 Dunner, a psychiatrist in Seatle, Washington, was 2 on the advisory panel, wasn't he? 3 A. He was -- yes, he was a member 4 of the advisory committee, yes. 5 Q. And the fact is that he had 6 done clinical studies examining Prozac for Lilly, 7 hadn't he? 8 A. I believe he did have such 9 studies. 10 Q. And he had been paid by Lilly 11 to do such studies? 12 A. If he did the studies, I 13 imagine he was paid to do those, yes. 14 Q. Did you know, in fact, that 15 some of his studies made up protocol twenty-seven 16 which was in fact one of the pivotal studies, you 17 knew that, didn't you? 18 A. That may be the case. But 19 again, I think that was all disclosed at the 20 advisory committee meeting. 21 Q. Was it? 22 A. My recollection was at the 23 beginning of the meeting there was a -- every 24 member of the advisory committee had to -- I'm Page 216 1 sorry, that there was a reading of all potential 2 conflicts of interest, and that was made as part 3 of the transcript. In addition, I don't even 4 think Doctor Dunner stayed to the end of the 5 meeting, that was my recollection. 6 Q. He voted, didn't he? 7 A. I think he may have only voted 8 on one point, and he wasn't there during most of 9 the discussion. 10 Q. Maybe he didn't even need to 11 come. 12 A. Maybe he didn't. 13 Q. Probably wasn't any need for 14 him to stay there? 15 MR. LORE: That's not a question. 16 MR. SMITH: That's a side comment, 17 that's not a question. 18 Q. Well, Doctor Thompson, in 19 Exhibit 24, says what do we know about the 20 advisory committee members. What have they 21 written recently, will there be any change in 22 membership this summer, what did they say about 23 Xanax and Halcion, what did they say about 24 Wellbutrin, can we get clues from transcripts on Page 217 1 their thinking, doesn't it? 2 A. Yes. 3 (PLAINTIFFS' EXHIBIT NO. 25 WAS 4 MARKED FOR IDENTIFICATION AND 5 RECEIVED IN EVIDENCE.) 6 Q. Exhibit 25 is a document 7 authored by Doctor Leigh Thompson dated July 16, 8 1994 is it not? 9 A. Yes. 10 Q. It's directed to you and 11 others? 12 A. Yes. 13 Q. And concerns label changes on 14 Prozac, does it not? 15 A. Yes. 16 Q. Specifically Doctor Leigh 17 Thompson has had a conversation with Doctor Paul 18 Leber at the Food and Drug Administration, has he 19 not? 20 A. Yes. 21 Q. And the subject was aplastic 22 anemia and labeling, wasn't it? 23 A. Yes. 24 Q. Doctor Thompson says I told Page 218 1 him that sending a physician to sites, we have 2 discovered that several cases were in fact in 3 error, and others had better alternative cause. 4 His reply was, quote, in quote, is there a single 5 case in which there is not a better alternative, 6 end quote. I replied I thought there was, but 7 needed to check in detail and we wanted to meet 8 with his staff and discuss the addition. He said 9 that he would be happy to meet, but he could not 10 this week, before we effected the change but that 11 his general philosophy was if there were a single 12 case that looked like it could be related to 13 Prozac that it belonged in the label and we could 14 discuss location, correct? 15 A. Yes. 16 Q. End quote. Does that mean 17 that if there's a case that is related to Prozac, 18 that it should be in the Prozac labeling? 19 A. Well, the way I read this 20 memo? 21 Q. Uh-huh. 22 A. He's talking specifically 23 about aplastic anemia. 24 Q. No question about it. Page 219 1 A. And if the conclusion was that 2 there was a case that didn't have another 3 alternative, it should be someplace in the label. 4 Q. Well, isn't that a general 5 expression of the FDA policy concerning whether 6 or not something should or should not be in the 7 label? 8 A. I don't believe I considered 9 it a general position, I think aplastic anemia is 10 one of those types of events that if you have a 11 single case, it should be in the label. But 12 every single adverse event would not end up in 13 the label. 14 Q. Well, Doctor Thompson goes in 15 and says I told him that we were happy with the 16 proposed location in the post-introduction 17 reports, but we were concerned about adding every 18 event that could happen to four million patients, 19 end quote, correct? 20 A. Yes. 21 Q. Are you saying that I can't 22 draw an analogy between aplastic anemia and 23 suicidality -- 24 A. That's exactly what I'm Page 220 1 saying. 2 Q. -- in connection with whether 3 or not it should be in the label, and where in 4 the label it should be? 5 A. I don't think it's valid to 6 draw such an analogy. 7 (A SHORT RECESS WAS TAKEN.) 8 (PLAINTIFFS' EXHIBIT NO. 26 WAS 9 MARKED FOR IDENTIFICATION AND 10 RECEIVED IN EVIDENCE.) 11 Q. Doctor Zerbe, Exhibit Number 12 26 apparently is reflective of a meeting 13 concerning fluoxetine in the UK authored by 14 Doctor -- or Mr. A.D.J. Brockwell, correct? 15 A. Yes. 16 Q. You're copied on this memo, 17 correct? 18 A. Yes. 19 Q. It has to do with questions 20 that had been raised by the CSM, which is what, 21 the British counterpart to the U.S. FDA? 22 A. Yes. 23 Q. What is it, Committee for 24 Safety in Medicine? Page 221 1 A. Yes. 2 Q. Apparently in 1985, on the 3 bottom of page two of this document, the CSM 4 raised a question, question number five being, 5 quote, clarification of comparative groups is 6 also required. Dose relationships are difficult 7 to demonstrate in the early clinical studies 8 because of the dose escalation that was used. 9 Preliminary data from the fixed-dose studies, 10 however, does suggest a dose relationship for 11 some of the adverse events, end quote, correct? 12 A. Yes. 13 Q. And the action there was 14 pending clarification, correct? 15 A. Yes. 16 Q. Do you recall this document or 17 this memo? 18 A. No, I really don't. 19 Q. Do you recall that the British 20 government's review of the data had indicated 21 that -- indicated that there might be a 22 suggestion of a dose relationship for some of the 23 adverse events? 24 A. I don't think that's what this Page 222 1 memo says, but I don't recall that at this 2 particular point. 3 Q. Did I read it incorrectly? 4 A. The way I read it, it says the 5 dose relationships are difficult to demonstrate 6 in early clinical studies because of the dose 7 escalation that was used. That was the CSM 8 piece. And then I think what David Brockwell is 9 saying is preliminary data from the fixed-dose 10 study, however, does suggest a dose relationship 11 for some of the adverse events. It's almost like -- 12 the way I read it is it's almost like an action 13 item that -- the issue of dose, we've got some 14 preliminary data to address that point, that's 15 kind of the way I read it. I don't think the CSM 16 would have even had preliminary data from the 17 fixed-dose study, that's why I question whether 18 it's the CSM asking this. 19 Q. It's all phrased under CSM, 20 question number five, isn't it? 21 A. I understand what you're 22 saying, I'm just telling you how I read this 23 statement or read this -- yes, it's a statement. 24 Q. Okay. Do you dispute that Page 223 1 there was preliminary data from the fixed-dose 2 study that suggests a dose relationship for some 3 of the adverse events? 4 A. No -- well, I don't know about 5 the preliminary study, but I don't dispute the 6 fact that in that fixed-dose study there was a 7 dose response relationship in some adverse 8 events. 9 Q. Okay. Could that generally be 10 said concerning other studies that had been done 11 comparing different dosage of Prozac, that in 12 higher dosage you saw in some adverse events 13 higher incidence of adverse events at higher 14 versus lower dosage? 15 A. Well, I think until the 16 fixed-dose study was done, it was difficult to 17 demonstrate that clearly, that dose-response 18 relationship in terms of efficacy or adverse 19 events. 20 Q. Okay. I'm talking about 21 adverse events. 22 A. Right. 23 Q. Would it be accurate to say as 24 a general proposition, in some studies, there was Page 224 1 a finding that there were more adverse events of 2 some kinds depending upon higher dosage of Prozac 3 used? 4 A. Well, I hate to be difficult 5 at this hour, but I don't think that necessarily 6 those data existed about dose-response 7 relationships and adverse events until the 8 fixed-dose study occurred. 9 Q. Okay. 10 A. There was no more than a 11 suggestion that at higher doses there were higher 12 adverse events. But the study hadn't really been 13 done. 14 Q. Now once the fixed-dose study 15 was done, that suggestion turned into a reality, 16 didn't it, it was indeed demonstrated that there 17 were some adverse events that occurred more 18 frequently on higher doses? 19 A. Yes. 20 Q. Okay. Also, one other exhibit 21 in connection with what happened in the United 22 Kingdom. 23 (PLAINTIFFS' EXHIBIT NO. 27 WAS 24 MARKED FOR IDENTIFICATION AND Page 225 1 RECEIVED IN EVIDENCE.) 2 Q. It might help you if you look 3 up in the right-hand corner, the handwritten, it 4 says communicated verbally by somebody, assistant 5 secretary to CSM, March 22, '85. 6 A. Okay. 7 Q. Exhibit 27 appears to be some 8 finding made by the CSM in connection with the 9 original application of Prozac, does it not? 10 A. Yes. 11 Q. And finding number five was, 12 quote, the incidence of adverse reactions in the 13 light of the lack of evidence on efficacy is 14 unacceptable. The company's analysis of the 15 adverse effect was inadequate and should include 16 adverse reaction rates in comparative groups, 17 correct? 18 A. Yes. 19 Q. Number six says, the data 20 sheet should be amended to the satisfaction of 21 the secretary, comment: The design of the 22 clinical studies and statistical analysis makes 23 assessment of the clinical value of fluoxetine 24 extremely difficult, correct? Page 226 1 A. Yes. 2 Q. Again, as we said earlier, in 3 1985 wouldn't it be accurate to state that the 4 British government had before them that data that 5 was submitted to the Food and Drug Administration 6 to support efficacy and safety of Prozac? 7 A. Well, again, I don't know 8 exactly what they had in front of them. The sort 9 of -- the data would basically have been the 10 same, but how it was presented may have led to 11 some of the questions that resulted here. 12 Q. But you don't have any 13 evidence to know that for a fact at all, do you? 14 A. No. 15 Q. The data may have been 16 presented exactly to the CSM in England as it was 17 presented to the FDA in the United States, as far 18 as you know? 19 A. It may have been, although I 20 doubt it was presented exactly, but it may have 21 been. 22 Q. It's supposed to have had the 23 same data? 24 A. The same fundamental data, Page 227 1 yes, but it could have been presented a different 2 way, some analysis may have been omitted at their 3 request, whatever. My only point is that it may 4 not have been identical in every way. 5 Q. Could we say that when Lilly 6 submitted their data to the British government 7 prior to 1985 in support of their application for 8 registration of Prozac in England, that they did 9 so in a manner that they thought would be 10 acceptable to the British government to support 11 Lilly's claims of efficacy and safety in 12 connection with Prozac? 13 A. That was the intent, yes. 14 (DISCUSSION OFF THE RECORD.) 15 (PLAINTIFFS' EXHIBIT NO. 28 WAS 16 MARKED FOR IDENTIFICATION AND 17 RECEIVED IN EVIDENCE.) 18 Q. Exhibit 28 is a two-part 19 document, and I stapled the first page to it just 20 to give you an idea of when it was transmitted -- 21 well, you are copied on page one of the memo 22 authored by Doctor Lange, are you not? 23 A. Yes. 24 Q. Or Mr. Lange? Page 228 1 A. Yes. 2 Q. Then page two and three have 3 to do with registering Prozac in the United 4 Kingdom, England in a ten milligram formulation, 5 correct? 6 A. Yes. 7 Q. Ten milligram solid 8 formulation, is that right? 9 A. Yes. 10 Q. Now, as a little background, 11 Lilly had for sometime prior to March 21, 1991 12 manufactured Prozac in a liquid form, had they 13 not? 14 A. I don't know the exact date 15 for liquid. My recollection would have been it 16 was later than 1991. 17 Q. Okay. Well, you may be right. 18 Then are you saying at the time that this 19 document was authored in March of 1991, that 20 there may not have been anything in existence 21 other than the twenty milligram solid 22 formulation? 23 A. That's right. 24 Q. Now, let me ask you this also: Page 229 1 Is twenty milligrams of Prozac twice as much as 2 ten milligrams of Prozac, in quantity? 3 A. I'm looking for the trick in 4 the question, but yes, it's twice as much. 5 Q. Is it twice as potent as ten 6 milligrams? 7 A. I'm not sure that I really -- 8 it's tough to answer the question, I don't know 9 what you mean by twice as potent. 10 Q. Should it have twice as much 11 potential to bind to serotonin reuptake receptors 12 to increase presence of serotonin in the synaptic 13 cleft? 14 A. I don't think you can reach 15 that simple conclusion from what you've said. 16 Q. Why can't I get there? 17 A. Well, because the binding to 18 it may be affected by the concentrations that are 19 already there, so you don't actually get twice as 20 much binding, you may only get a quarter of as 21 much binding. And there certainly is a limit to 22 that linear area in relationship, so I don't 23 think you can say that twenty has twice as much 24 of an effect as ten or, et cetera. Page 230 1 Q. Can you say forty has twice as 2 much as twenty? 3 A. No, you really can't reach 4 that conclusion, I don't believe. 5 Q. Can you say eighty has twice 6 as much as forty, binding properties? 7 A. Well, again, I'm not sure what 8 you mean by binding properties. 9 Q. I thought what Prozac did was 10 it got up there in the receptors that ordinarily 11 would reuptake serotonin from the synaptic cleft 12 and competed there for the serotonin molecules, 13 and made it less likely that the serotonin would 14 be reuptake at that receptor site. In other 15 words, Prozac actually occupies a space on the 16 serotonin reuptake receptor. Is that incorrect? 17 A. Well, I don't know -- in 18 general, that's not incorrect. 19 Q. All right. And I'm not saying -- 20 A. But what I'm saying is that's 21 not necessarily a linear relationship. 22 Q. That's what I'm trying -- of 23 course maybe these questions should be directed 24 to Doctor Fuller or somebody that's more familiar Page 231 1 with neuropsychopharmacology or biology than 2 yourself, but as a general proposition could it 3 be said that you should have an increase, an 4 increase in serotonin reuptake greater at twenty 5 than at ten? 6 A. Well, in general there tends 7 to be dose-response relationships. 8 Q. Okay. 9 A. In general, the higher dose 10 produces more pharmacologic effects. 11 Q. Okay. 12 A. But that's not true in all 13 cases nor with all medications. 14 Q. We're talking about Prozac and 15 what has been established in the Prozac clinical 16 trial experience. 17 A. Okay. That wasn't clear, I 18 thought you were talking more generally in 19 pharmacology. But I think the general statement 20 still holds within the dose range of twenty, 21 forty, sixty, at least based on the fixed-dose 22 study clinically that there's more of an effect, 23 not necessarily twice as much, but more. 24 Q. Was there any biological study Page 232 1 done to measure whether or not it was twice the 2 effect? 3 A. I don't believe so. 4 Q. Could that be done? 5 A. I don't believe that could be 6 done directly. One might look at indirect 7 markers, but the validity of those in evaluation 8 of brain effects -- 9 MR. SMITH: We're beginning to get 10 into something that you like, Bart, and you're 11 leaving. 12 MR. BROWN: I know. 13 Q. I cut you off and I didn't 14 mean to. 15 A. I didn't finish the statement, 16 but to be honest with you, I don't know where I 17 was, you'll have to repeat it. 18 (THE COURT REPORTER READ BACK THE 19 REQUESTED TESTIMONY.) 20 A. I think I can finish the 21 statement by simply saying that it wouldn't 22 necessarily -- there's no valid mechanism by 23 which to look at the CNS effects on those 24 receptors in humans. Page 233 1 Q. But the hypothesis is that 2 more should be more, and less should be less? 3 A. Within some limits, I mean 4 dose ranges aren't infinite, there's a limit to 5 extrapolation of that. 6 Q. All right. But within limits 7 it could be said more equals more effect, less 8 equals less effect? 9 A. In general. 10 Q. All right. Apparently there 11 was discussion of whether to have solid forms of 12 Prozac in twenty milligrams, correct? 13 A. Yes. 14 Q. And there was a discussion in 15 connection with both the regulatory and the 16 psychiatric standpoint. 17 A. Yes. 18 Q. It says we work closely with 19 the Medicines Control Agency, MCA, in late '90, 20 '91. In discussions with myself and David 21 Wheadon in November, Doctor Sue Wood, head of MCA 22 safety monitoring, expressed three MCA priorities 23 for Prozac, right? 24 A. Yes, sir. Page 234 1 Q. Item C was introduction of 2 lower Prozac dose, correct? 3 A. Yes. 4 Q. So can I assume that in March 5 of 1991, Prozac had been approved in England but 6 it was only available in the twenty milligram 7 solid dosage form? 8 A. Yes. 9 Q. It said the need for C, which 10 is introduction of lower Prozac dose, has been 11 strongly reenforced by Doctor Wood and others in 12 her group on several subsequent occasions, 13 including the suggestion that I provide a written 14 statement of Lilly policy in timing in this area. 15 And then he puts in parentheses, not done, close 16 paren, correct? 17 A. Yes. 18 Q. It says they believe that this 19 is necessary because alternate-day dosing 20 recommended in some patients may reduce 21 compliance, correct? 22 A. Yes. 23 Q. So is alternate-day dosing a 24 situation where the physician says take twenty Page 235 1 milligrams on Monday, Wednesday and Friday? 2 A. Every other day, basically 3 every other day. Whether that's how we 4 recommended it, I don't know, but every other 5 day. 6 Q. And from a medical standpoint, 7 there's some drawbacks to that, is there not? 8 A. There are drawbacks from the 9 perspective of compliance, yes. 10 Q. When you say from the 11 perspective of compliance, what do you mean? 12 A. It means more difficult for 13 patients to remember to take their dose every 14 other day as opposed to every day. 15 Q. So the physician is going to 16 be better assured that the patient is consuming 17 their medication as directed if they direct them 18 to do it on a more regular basis, which would be 19 every day, take it every morning? 20 A. I think, again, in general, 21 yes. It's probably better to get compliance 22 every day, once a day, than every other day. 23 Q. All right. Item B there is 24 that the five milligram data interpreted as Page 236 1 showing efficacy, correct? 2 A. Yes. 3 Q. Do you agree that the five 4 milligram data showed efficacy? 5 A. In the trials done, five 6 milligrams was statistically better than placebo. 7 Q. Therefore efficacious in 8 treating depression? 9 A. Yes. 10 Q. C, it says relatively large 11 numbers of Lilly supplied events, and UK yellow 12 card reports may be reduced if ten milligram dose 13 used in some patients, correct? 14 A. That's what it says, yes. 15 Q. Now that goes back to the 16 yellow cards that you were talking about earlier, 17 which is a notice in England of an adverse event 18 in connection with Prozac, right? 19 A. Yes, it's a more formal 20 follow-up system than spontaneous adverse events. 21 Q. Is there a requirement for a 22 yellow card analysis in England that there be 23 some investigation as to causal relationship 24 between the event complained of and the drug? Page 237 1 A. I really don't remember how 2 that's worded within the yellow card system, 3 whether it's any event or whether it's only 4 causally -- events believed to be causally 5 related, I don't know, I don't remember that. 6 Q. But here, the regulatory 7 people at the CSM, the British equivalent to the 8 FDA, were finding that or believed that if you 9 reduce the dosage, you would reduce the number of 10 reported adverse events, correct? 11 A. They say that that may -- that 12 may decrease the dose -- I'm sorry, that may 13 decrease the frequency of adverse events by 14 decreasing the dose. 15 Q. To ten milligrams? 16 A. To ten milligrams. 17 Q. D says Committee on Safety of 18 Medicines members appears to have a position that 19 ten milligram dose would be effective, correct? 20 A. Yes. 21 Q. Would that be based on the 22 five milligram data that had been submitted to 23 them? 24 A. I can only speculate about Page 238 1 what they base that on. But that's the dominant -- 2 that's the only real good efficacy data that's 3 available. 4 Q. Because there never was any 5 studies undertaken to examine ten milligrams, was 6 there? 7 A. I don't believe there were 8 studies at ten milligrams. 9 Q. It had been intended, hadn't 10 it, but was not done? 11 A. I'm not sure what you mean by 12 it had been intended. I don't know who intended 13 to do it. To my knowledge there was no decision 14 to do ten milligrams, that got reversed. I just 15 wasn't aware of that. 16 Q. Didn't you know that Lilly had 17 made up clinical trial kits containing ten 18 milligrams to be used in the studies, clinical 19 trials? 20 A. I don't recall that, but that 21 may be the case. 22 Q. And that Mr. Wood, chairman of 23 the board, chief executive officer, made the 24 decision not to do ten milligrams? Page 239 1 A. I think that may have occurred 2 when I was in England, I just don't recall that 3 discussion with Mr. Wood or -- 4 Q. Do you recall any discussion 5 with Mr. Wood concerning dosage? 6 A. I don't recall any discussion 7 with Mr. Wood regarding dosage. 8 Q. Mr. Wood generally never liked 9 the idea of a ten milligram dose, did he? 10 MR. LORE: Only if you know. 11 A. I really don't know what his 12 position was. 13 Q. As far as you know? 14 A. As far as I know? 15 Q. Right. 16 A. Again, I never had any 17 discussions with Mr. Wood about it. 18 Q. Did you ever have any 19 discussion with Mr. Wood about Prozac at all? 20 A. I don't think ever directly, 21 no. I made presentations, but that was all. 22 (PLAINTIFFS' EXHIBIT NO. 29 WAS 23 MARKED FOR IDENTIFICATION AND 24 RECEIVED IN EVIDENCE.) Page 240 1 Q. Exhibit 29 indicates that Mr. 2 Wood had made a decision to amend the Prozac 3 fixed low-dose protocol and exclude the ten 4 milligram fluoxetine dosage regimen, did he not? 5 A. That's what the memo says. 6 Q. It says with his -- with this 7 exclusion, please revise the CT assembly by 8 removing all patient kits that contain ten 9 milligram. This will include 160 kits with 10 bottles having CT No. 8033-5A. The procedure has 11 been discussed with Max Talbott for regulatory 12 adherence. We're still scheduled to start the 13 study on 6-23-85. We will need CT release by 14 that date. If this is a problem, please call. 15 Thank you for your cooperation, Melissa Humbert, 16 right? 17 A. That's what it says. 18 Q. So does that indicate to you, 19 what it says, that Mr. Wood had decided not to do 20 a ten milligram study, but that the study had 21 intended to include ten milligram materials, in 22 fact those materials were already packed into 23 some of the clinical trial kits and had to be 24 removed? Page 241 1 A. That's what the memo says. 2 Q. You don't read that a 3 different way than I read it? 4 A. No. 5 Q. Did you know that ten 6 milligram had been intended to be analyzed at one 7 point? 8 A. I know there was a lot of 9 discussion about the doses in the second 10 fixed-dose trial. To be honest with you, I 11 didn't realize that the decision -- that the 12 material had been ordered to be included in it. 13 For all I know, they ordered extra CT kits not 14 having reached a conclusion about what doses were 15 going to be. 16 Q. That's not what that says, is 17 it, it says that Mr. Wood has decided not to do 18 it, doesn't it? 19 A. Well, I don't think that's 20 incompatible with what I said. What I said was 21 it's possible that the decision was still pending 22 about what would be included. 23 Q. What evidence do you have that 24 there was still some question about what should Page 242 1 be included? 2 A. Well, what I'm saying is it 3 would not be unusual, unheard of, with regard to 4 issues like this to make up more kits than you 5 intend to use, simply because you have the 6 flexibility. And the one thing that makes me 7 think this may be the case, this is speculation, 8 but the very short time frame that they have to 9 actually start the study. So it may be that they 10 actually, you know, packed a whole set of kits, 11 which then some decision will be made pending 12 that decision and discussion. 13 Q. Do you know if they had two 14 and a half milligram packs made up? 15 A. To my knowledge, they didn't, 16 but I don't know. I didn't know they had the ten 17 milligram kits made up. 18 Q. That's my point. How could 19 you speculate on this when you didn't even know 20 they had ten milligram kits made up? 21 A. I'm just speculating based on 22 what information you've given me. 23 Q. Actually, to be honest, you're 24 surprised to see that, aren't you? Page 243 1 A. Yes, I am, I didn't realize 2 that they had actually packed the kits. 3 Q. You didn't realize they had 4 actually packed the kits, and you didn't realize 5 that the chairman of the board and chief 6 executive officer of Eli Lilly and Company would 7 be having direct input into a scientific decision 8 in connection with dosages administered in the 9 clinical trial, did you? 10 MR. LORE: I object, that's not what 11 he said, that's not what he just said. He just 12 said that he was surprised the package -- 13 MR. SMITH: Well, I'm giving him some 14 other things to be surprised about. 15 Q. That's surprising to you, 16 isn't it, that the chairman of the board and 17 chief executive officer would have direct input 18 into what dosage to examine in a particular 19 clinical trial? 20 A. Well, again, without knowing 21 the background, it's tough to say, you know, how 22 Mr. Wood's name got involved in it. You probably 23 need to talk to Melissa Humbert to find out 24 exactly what the background was. Page 244 1 Q. I tell you what, I talked to 2 Mr. Wood about it, and he said he didn't make any 3 decision in any clinical trials, that's what he 4 paid you all to do. That makes sense, doesn't 5 it? 6 A. Sure does. 7 Q. He said he didn't like the 8 idea of switching to -- not switching to, but 9 adding to a solid -- going to a solid ten 10 milligram dosage in addition to twenty milligram, 11 he didn't think it made sense. But he never -- 12 he stated -- he testified under oath, he didn't 13 make any decisions in connection with any 14 scientific matters because he's not a scientist, 15 and that's what he pays you all to do, correct? 16 MR. LORE: That's not a question. I 17 object -- 18 MR. SMITH: I had a correct at the end 19 of it. 20 MR. LORE: -- to the references you're 21 making and what you're representing Mr. Wood 22 said, Paul, I'm not sure he said anything close 23 to that. 24 MR. SMITH: He did say that, I swear Page 245 1 he said that, he said it under oath 2 MR. LORE: What is your question? 3 Q. My question is: Aren't you 4 surprised that the chairman of the board would be 5 having a direct input into a direct scientific 6 question on a direct clinical trial? 7 A. I mean in some ways I'm 8 surprised, in some ways I'm not surprised. I 9 don't know whether that's really what Melissa 10 Humbert -- 11 Q. I don't either, all I can see 12 is what she wrote there in the English language. 13 A. Okay. I'm somewhat surprised 14 that this would have been ascribed to Mr. Wood, 15 he would not have routinely been making decisions 16 about clinical trials. 17 Q. He testified he didn't make 18 decisions about clinical trials. It made sense 19 to me at the time because he's not a scientist, 20 is he, as far as -- 21 A. No. 22 Q. -- he testified concerning his 23 educational background? 24 A. No. Page 246 1 Q. Back to Exhibit 28, the 2 original Irwin stuff. Do you know then why, in 3 point D on page one, under regulatory, the 4 committee on Safety of Medicines appears to have 5 a position that a ten milligram dose would be 6 effective? 7 A. Well, I don't know what they 8 base that on. 9 Q. Point E says they, meaning the 10 CSM, the English equivalent to the United States 11 FDA, forced the dose of Sertraline and Paroxatine 12 down based on Prozac perception in the UK. Do 13 you see that? 14 A. I see it. 15 Q. What is Sertraline and 16 Paroxatine? 17 A. They're two other serotonin 18 reuptake inhibitors. 19 Q. Were you aware in March, 1991 20 that the United Kingdom had forced the dosage of 21 other serotonin specific reuptake inhibitors down 22 based on a perception of Prozac? 23 A. I was not aware of that. 24 Q. Now, the memo goes on to say Page 247 1 in connection with why would they accept ten 2 milligrams as effective, they say the MCA 3 currently would accept available five milligram 4 data to support a ten milligram form being 5 available. Do you see that? 6 A. Yes. 7 Q. Were there any other instances 8 that you know of in which Prozac was accepted for 9 use in a higher dosage based on efficacy data 10 conducted on lower dosages? 11 A. I'm not aware of any other 12 such situation. 13 Q. Doctor Keohane also indicates 14 the psychiatrists wanted the lower doses, doesn't 15 he, on page two? 16 A. Is that a question? 17 Q. Doesn't he, on page two, 18 indicate that psychiatrists in some instances are 19 requesting a lower dosage than the current 20 available twenty milligram dosage? 21 A. Yes. 22 Q. And that the liquid 23 formulation of twenty milligrams per five 24 milliliters was not seen by the clinicians as a Page 248 1 good option? 2 A. That's what's stated here, 3 yes. 4 Q. And that they preferred a 5 solid ten milligram form? 6 A. Yes. 7 Q. The conclusion of Doctor 8 Keohane's report is strong regulatory and 9 significant customer interest in a ten milligram 10 solid form. MCA likely to press hard now, 11 correct? 12 A. That's what it says, yes. 13 (PLAINTIFFS' EXHIBIT NO. 30 WAS 14 MARKED FOR IDENTIFICATION AND 15 RECEIVED IN EVIDENCE.) 16 Q. In connection with the five 17 milligram and ten milligram preparation, Doctor 18 Leigh Thompson, the chief scientific officer of 19 Eli Lilly, wrote a memo of June 11, 1991 which is 20 reflected in Exhibit 29, is it not? 21 A. Yes. 22 Q. I'm sorry, Exhibit 30. 23 Exhibit 30 is to Charles B. Sampson, is it not? 24 A. Yes. Page 249 1 Q. He's a statistician, is he 2 not? 3 A. Yes. 4 Q. You, however, are copied on 5 that document, are you not? 6 A. Yes. 7 Q. Were you aware of this 8 situation in connection with ten milligram, the 9 ten milligram issue that had been to be presented 10 to the board? 11 A. I don't recall the specific 12 board presentation on ten milligrams, but I was 13 aware that ten milligrams was an issue. 14 Q. Do you recall seeing this 15 memo? 16 A. I don't recall the specific 17 memo. 18 Q. Was it your habit, though, 19 Doctor Zerbe, to read those copies -- those memos 20 to others directed by Doctor Leigh Thompson on 21 which you were copied? 22 A. Yes. 23 Q. Was Doctor Charles Sampson a 24 Ph.D under your general supervision or was he in Page 250 1 a different -- 2 A. No, he was under my 3 supervision. 4 Q. And had he done statistical 5 work for you, specifically in the past? 6 A. I don't know whether we worked 7 on any specific projects, but he and I had worked 8 together a great deal. 9 Q. And you respect his judgment 10 as a statistician or his ability as a 11 statistician? 12 A. Yes. 13 Q. Do statisticians exercise 14 judgment? 15 A. Oh, I think so, yes. 16 Q. And they exercise judgment 17 primarily in the data they select to analyze? 18 A. I wouldn't have termed it 19 exactly that way. 20 Q. Or in the particular 21 statistical method they use to analyze particular 22 data? 23 A. Yes. 24 Q. You're going to have to use Page 251 1 judgment in only one or two ways, either in the 2 data they choose to analyze or the method, the 3 statistical method they use to analyze a 4 particular set of data. 5 A. And they also use their 6 judgment to help assist in the final 7 interpretation of the data. 8 Q. Yes, as opposed to comparing 9 various sets of data, a statistician can provide 10 some judgment in those three areas? 11 A. If one worked at it, you could 12 probably identify others, but in general, that's 13 true. 14 Q. Now, this memo says Charlie, 15 let me ask for your help -- let me ask you for 16 help in putting together two slides for the 17 board. I've got to say something about ten 18 milligram, both in regard to attributes and to 19 logistics of when we will file in the U.S. 20 International filing is a big, big problem, end 21 quote. Let me stop there. What was the big, big 22 problem in connection with international filing 23 of ten milligram solid form preparations? 24 A. I don't know, maybe Doctor Page 252 1 Thompson, you know, would be better able to 2 answer that question than I would, but ten 3 milligrams was a problem because we were 4 concerned about whether there would be sufficient 5 efficacy to sustain the, you know, efficacy of -- 6 the perception of the efficacy of the drug. 7 Q. Okay. Well, I think that's 8 what Doctor Thompson is addressing in the next 9 paragraph, isn't he? 10 A. Right. 11 Q. Because he said I don't think 12 we have any ten milligram efficacy data. 13 A. Right. 14 Q. We do have the Wernicke study 15 of five milligram versus twenty and forty. Some 16 people have massaged those data to make five 17 milligram look not quite as good as twenty. I'm 18 not sure I know how the massage was done, but I 19 probably need a slide to show whatever it is that 20 makes it look less good. I would probably also 21 like a slide that shows a more global view of how 22 it works. Do we have a display by week of the 23 HAM-D total changes? As I recall, five milligram 24 actually went down a little faster than the Page 253 1 others, and ended up about the same. Help me, if 2 you will, on that to show on ten milligram to the 3 board, end quote, correct? 4 A. That's what the memo says, 5 yes. 6 Q. What he's saying is, is he 7 not, that there was a Wernicke study examining 8 five versus twenty versus forty. Everybody knew 9 that, correct? 10 A. Right. 11 Q. And that somebody had massaged 12 that data to make the five milligram in support 13 of efficacy of twenty milligram not as 14 efficacious as twenty, right? 15 A. Well, I don't know what he 16 means by massaged, but that's what the memo says. 17 Q. What do you -- when you use -- 18 when you see the term some people have massaged 19 these data, how do you interpret massage? 20 A. Well, I think, knowing Leigh, 21 what Leigh probably meant is that it was 22 carefully analyzed and evaluated from multiple 23 perspectives. It carries a more negative 24 connotation read in a way such as this than I Page 254 1 think the situation would have warranted. 2 Q. Well, I guess the reason it 3 has such a negative connotation is what it goes 4 on to say. You can stop, maybe, where he said 5 massaged data, that might have been enough, but 6 he says I don't know -- I'm not sure how the 7 massage was done, but I probably need a slide to 8 show whatever it is that makes it look less good. 9 What he's saying, isn't it, what is it about the 10 massage that made the five milligram look less 11 good than the twenty milligram, isn't it? 12 A. That's what he says, and he 13 probably is the best one to answer the question 14 about exactly what he was implying as a result of 15 that statement. 16 Q. Well, I have, and he says I 17 was implying that that data should have been 18 massaged, and I was implying a scientific 19 analysis in the massage. But there was a 20 scientific analysis in the massage to make the 21 five milligram look not as good as the twenty 22 milligram, correct? 23 MR. LORE: Again, that's -- 24 MR. SMITH: That's what it says. Page 255 1 MR. LORE: I object to Mr. Smith's 2 representation of what Doctor Thompson testified 3 to about this memorandum, and I also object 4 because you're asking this witness to agree with 5 you as to what that representation was, and 6 that's going to require him to speculate to what 7 Doctor Thompson meant, and he's already said he 8 doesn't know, that you need to ask Doctor 9 Thompson. 10 MR. SMITH: Well, I have. See, I was 11 there and you weren't, Mr. Lore. And you weren't 12 at Doctor Beasley's. I was there and asked him 13 about it. What he said about it is reflected in 14 the transcript, and that's going to be the best 15 evidence. 16 MR. LORE: Right. 17 MR. SMITH: I'm trying to relate to 18 you what my recollection is of what he said, and 19 I'm trying to do that in as accurate a way as I 20 can, you know, limited by my brain resources. 21 MR. LORE: But are you asking him to 22 agree with you or disagree with you? 23 MR. SMITH: I'm asking him if he knows 24 that there was a massage that occurred that made Page 256 1 five milligram look less efficacious than twenty 2 milligram. 3 A. Well, let me state it the way 4 I would have stated it, ignoring the word massage 5 and all of the implications. There was a 6 detailed analysis of the dose range, and analysis 7 from a psychiatrist's standpoint about the 8 optimum to risk/benefit, I guess you would say, 9 with regard to the use of the appropriate dose. 10 And the conclusion, there was, and I think there 11 was -- I think there was concensus among the 12 psychiatrists that the best overall balance was 13 the twenty milligram dose. That's the way I 14 recall the discussion, the analysis, and the 15 conclusions. 16 Q. But by virtue of what he says, 17 there apparently was a scientific basis to 18 believe that the five milligram was about as 19 efficacious, if not equally efficacious? 20 A. Again, my recollection was 21 that the -- that five milligrams was 22 statistically efficacious. However, one doesn't 23 reach the conclusion about dosage recommendation 24 purely on efficacy, you also look at adverse Page 257 1 event profile and so forth, and the magnitude of 2 the efficacy, and reach a clinical judgment about 3 the best overall benefit for the most patients. 4 Q. Well, he goes on to say, as I 5 recall, five milligram actually went down a 6 little faster than the others, and ended up about 7 the same. And he's talking about HAM-D scores, 8 isn't he? 9 MR. LORE: If you know. 10 A. I don't know what he was 11 talking about, nor do I recall those data. 12 Q. That's what it says, is HAM-D 13 scores. He said I probably need a slide to show 14 whatever it is that makes it look less good. I 15 probably would also like a slide that shows a 16 more global view of how it works. We do have a 17 display by week of the HAM-D total changes, and 18 as I recall, five milligram actually went down a 19 little faster than the others and ended up about 20 the same. Help me, if you will, on that to show 21 ten milligram to the board, correct? 22 A. That's what that says, yes. 23 Q. So I take it that when he says 24 five milligram actually went down a little Page 258 1 faster, he's talking about decreases in HAM-D 2 scores, which indicates improvement, right? 3 A. That's the way I would read 4 this statement as well. But, I mean, he's the 5 person to ask probably about exactly -- 6 Q. Well, we did, and he did. And 7 his analysis of the data was that the HAM-D 8 scores ended up about the same, between five and 9 twenty, correct? That's what it says here. 10 A. Oh, the same -- that the end, 11 they ended up about the same, that's what the 12 memo says. 13 (PLAINTIFFS' EXHIBIT NO. 31 WAS 14 MARKED FOR IDENTIFICATION AND 15 RECEIVED IN EVIDENCE.) 16 Q. Exhibit 31 is a document 17 authored by one of the Lilly psychiatrists, 18 Doctor John Heiligenstein, dated November 7, 19 1990, is it not? 20 A. Yes. 21 Q. And it has to do with his trip 22 to a meeting in Vienna, correct? 23 A. Yes. 24 Q. And it's a meeting of Page 259 1 psychiatrists in Vienna, is it not? 2 A. I don't know -- I mean I can't 3 remember exactly what the Vienna meeting was. 4 Q. Would you send him to Vienna 5 to attend a meeting of proctologists? 6 A. I doubt it. 7 Q. You're copied in on this memo, 8 are you not? 9 A. Yes. 10 Q. He says he would like to share 11 the sense of what he had heard. Clearly, very 12 clearly, there's a need for a lower dose in 13 pulvule form. Five milligram would be most 14 appropriate, although at the child psychiatry 15 meeting in Chicago, two point five milligram was 16 suggested. The liquid formulation will only 17 partially address this need, correct? 18 A. That's what the memo says. 19 Q. Did you discuss this with 20 Doctor Heiligenstein? 21 A. I don't remember discussing 22 the Vienna meeting, per se, but again, the issue 23 of dose was discussed a number of times. 24 Q. Were you recommending a move Page 260 1 to a ten milligram or five milligram dose? 2 A. My recommendation was that we 3 provide a ten milligram dose. But that was not 4 based on a recommended ten milligram dosage, if 5 you understand what I mean. That is that the 6 dose range was still twenty to eighty milligrams, 7 but we needed a ten milligram dosage form because 8 we had certain situations where lower doses were 9 called for in the package insert, things like, 10 you know, renal disease, and that it was 11 important to provide that alternative. Others 12 argued that we could meet that need by 13 every-other-day dosing or by the liquid dosage 14 form. But I really felt like it was important to 15 have a ten milligram dosage, for convenience, not 16 because I concluded that the dosage 17 recommendation was incorrect. 18 Q. Do you agree with several, 19 like in CSN, that if you had lower dosage of 20 Prozac or dosage in ten milligram form, you were 21 likely to see a lowering of adverse events 22 experienced with the drug? 23 A. Well, going back to that memo, 24 if we could find it, regarding the CSM, I agreed Page 261 1 with, I guess, point number one -- or A, which 2 was alternate-day dosing recommended in some 3 patients may reduce compliance. I think that was 4 true. I don't think the overall balance between 5 risk and benefit of the five milligrams was 6 superior to the twenty milligrams, and I think 7 the overall assessment, again, by the 8 psychiatrists with regard to that point, was that 9 the twenty milligram dosage was the appropriate 10 dosage, dosage, okay, I think they agreed that a 11 ten milligram solid dose form would provide 12 convenience. 13 Q. Well, is it your opinion, 14 based on your knowledge of the clinical trial 15 data in the post-marketing experience, that 16 twenty milligram is more likely to be efficacious 17 than ten milligram in Prozac? 18 A. The data, as we reviewed them, 19 suggested that ten milligrams was somewhat more 20 effective and did not have any higher incidence 21 of adverse events than five milligrams in the 22 study. 23 MR. LORE: I'm sorry, you said ten 24 milligrams, did you mean twenty? Page 262 1 THE WITNESS: I meant twenty, I'm 2 sorry. 3 Q. I was going to say, I didn't 4 know that you ever studied ten milligram. 5 A. No, twenty milligrams versus 6 five milligrams, I'm sorry. 7 Q. Give me that answer again if 8 you can. 9 A. The assessment of the data, as 10 I recall them -- and Doctor Tollefson and I 11 talked about this quite a bit, so Doctor 12 Tollefson may be an appropriate person to address 13 that question to. The overall assessment was 14 that the twenty milligram dose provided the 15 optimal balance of efficacy and adverse events as 16 compared to five milligrams, and that -- again, 17 my recollection was that it was slightly more 18 efficacious, and the frequency of adverse events 19 was no higher. 20 Q. Well, how do you comport that 21 with Doctor Leigh Thompson's analysis, that five 22 milligram actually, on HAM-D scores, started off 23 better at reducing HAM-D scores and ended up 24 about the same? Page 263 1 A. Well, again, I think Doctor 2 Thompson in his statement, and you've asked him, 3 so I don't know the details, but the way I read 4 that memo, it was based on his recollection. The 5 discussions I had with Doctor Tollefson, when we 6 looked at this issue to try to determine a 7 position, was based on fairly extensive analysis 8 by Doctor Tollefson. And on balance, I sided 9 with Doctor Tollefson with regard to the dosage 10 recommendation, the analysis. Sided may not be 11 the right word, but I would at this point favor 12 that as a much more detailed analysis than Doctor 13 Thompson's recollection. That's the way I would 14 put it together. 15 Q. Okay. Well, would you say 16 that twenty milligram and five milligram are 17 equally efficacious? 18 A. Well, again, my recollection 19 of the data from the discussions with Doctor 20 Tollefson was that twenty milligrams was slightly 21 more effective without causing any higher 22 frequency of adverse events. 23 Q. All right. Let me ask it this 24 way: Was there a statistical -- statistically Page 264 1 significant difference in efficacy between five 2 milligrams and twenty milligrams in the Wernicke 3 clinical trial? 4 A. I don't believe there was a 5 statistical difference. 6 Q. Was there a statistical 7 difference in adverse events between five 8 milligram and twenty milligram in the Wernicke 9 trial? 10 A. I don't believe there was a 11 statistical difference, although there may have 12 been. One looks at so many adverse events, there 13 may have been some, there was a slight 14 statistical difference. 15 Q. Back to Exhibit 31. Point two 16 of Doctor Heiligenstein's memo says, we have not 17 established the lowest effective dose, and if we 18 listen to the prescribers, some effort should be 19 made to do this. If not, MDs will continue to 20 dissolve pulvules and will anxiously await new 21 medications with greater flexibility in dosing, 22 correct? 23 A. That's what the memo says. 24 Q. Do you agree that at least as Page 265 1 of June, 1993, when you left Lilly, that Lilly 2 had not established the lowest effective dose? 3 A. I really don't agree with the 4 way that John's framed this. Lowest effective 5 dose can be interpreted in different ways by 6 different people. There may be doses lower that 7 one can demonstrate statistical benefit -- with 8 statistical efficacy, but that doesn't mean that 9 that's an appropriate dosage recommendation. One 10 has to consider the balance between the degree of 11 efficacy and the degree of side effects to come 12 up with that recommendation. And it is my belief 13 that that was thoroughly evaluated and that 14 twenty milligrams was, based on the data that was 15 available, the best recommendation. 16 Q. And you weren't aware of any 17 massage that had occurred in that data to make 18 twenty milligram look better than five? 19 A. I was not aware of any massage 20 if the implication of that was that in some way 21 the data were twisted, turned or misrepresented. 22 There was an extensive analysis of the data. 23 Q. How was the decision made to 24 market a ten milligram solid pulvule? Page 266 1 A. I think -- well, I think the 2 final decision was made in one of those Tuesday 3 morning sessions. 4 Q. All right. Do you recall 5 specifically the decision being made to go ahead 6 and go with ten milligrams? 7 A. I don't remember the specific 8 discussion about it, I don't remember exactly 9 when that took place. 10 Q. Did you have input into the 11 decision, regardless of when it occurred? 12 A. Well, I mean input in the way 13 I think I would have participated in those 14 sessions. We would have been asked from a 15 clinical perspective our assessment, and we would 16 have provided that. 17 Q. You're not a psychiatrist, but 18 based on your involvement in the clinical trial 19 process, the post-marketing experience with 20 Prozac, and based on the fact that you supervised 21 psychiatrists who were in even more directly 22 involved in this, would you have any opinion with 23 respect to whether or not a twenty milligram 24 dosage of Prozac is more likely to alleviate Page 267 1 depression than five? 2 A. Well, again, my impression of 3 the data was that the twenty milligram dose was 4 somewhat better, not statistically, but somewhat 5 better than the five milligram dose in terms of 6 efficacy. And that was based on an analysis done 7 primarily by Doctor Tollefson. 8 Q. Well, the five versus twenty 9 versus forty versus sixty was done back in 1985 10 by Doctor Wernicke, wasn't it? 11 A. Well, but Doctor -- the 12 question of dosage came up repeatedly. When the 13 discussion, and I think even the data, from the 14 original fixed-dose study and the subsequent 15 fixed-dose study came out, I don't think I was 16 even in the country, I didn't participate in that 17 discussion. But it did come up after I returned, 18 and we again looked at the information. 19 Q. When would that have been? 20 A. I think it would have been in -- 21 it must have been in 1991, '92. 22 Q. Would that have been in 23 connection with the recommendation and the urging 24 of the British at the CSM? Page 268 1 A. Well, again, it was a 2 continually surfacing issue, and I think every 3 time it came up we looked at it again. We looked 4 at dose at that time. I'm not sure I was ever 5 involved in the discussion at the time of the CSM 6 about the data, the five milligram dose at that 7 particular time, but the analysis that I recall 8 that in my mind sort of put the issue to bed was 9 the one that Doctor Tollefson did. I trusted his 10 psychiatric judgment and his knowledge of the 11 pharmacology, and it was my view that that was a 12 reasonable conclusion. 13 Q. But that was an analysis of an 14 older trial that had been done back in 1985? 15 A. Yes, it wasn't a new trial. 16 Q. And didn't even examine ten 17 milligram -- 18 A. That's right. 19 Q. -- dosage? 20 A. I thought your original 21 question was twenty and five. 22 Q. Right, it was. But the 23 analysis that you relied on by Doctor Tollefson 24 was done based on the old five, twenty, forty Page 269 1 data done in '85 by Doctor Wernicke? 2 A. Yes. 3 Q. There weren't any new studies 4 done? 5 A. No. 6 Q. Why did the issue of dosage 7 continue to come up? 8 A. I mean it's hard to speculate. 9 I think frequently new medications, the dosage 10 question, continues to surface. 11 Q. But this medication had been 12 around since 1987 now, going on six years. 13 A. I don't know, maybe it was 14 because the drug was perceived to be, I don't 15 know, very effective at the higher doses, and 16 people thought they could get by with lower 17 doses, maybe they had some clinical experience 18 with individual cases with patients that 19 anecdotally got better. I don't know, it's hard 20 to speculate exactly why. I suppose there were 21 people that still weren't convinced. We probably 22 did a more thorough evaluation of dose than 23 anybody had ever done, and I suppose that even 24 opened up more questions. Page 270 1 Q. Well, your thorough evaluation 2 showed no statistical differences in efficacy 3 between something that was twenty-five percent by 4 weight as large, didn't it? 5 A. That's right. 6 Q. Pretty potent medication, 7 isn't it? 8 A. Well, it was effective, yes. 9 (PLAINTIFFS' EXHIBIT NO. 32 WAS 10 MARKED FOR IDENTIFICATION AND 11 RECEIVED IN EVIDENCE.) 12 Q. Exhibit 32 is a document out 13 of Doctor Leigh Thompson's file? 14 A. Uh-huh. 15 Q. And he had a little difficulty 16 in identifying exactly what it was, and that's 17 the reason I'm asking you some questions about 18 it. Do you know who Bob Hunt is? 19 A. Yes. 20 Q. Who is he? 21 A. Bob Hunt worked in the DEU and 22 was -- I don't know the exact title, but he was 23 responsible for -- one of the people responsible 24 for following Prozac adverse events. Page 271 1 Q. Okay. The first page of this 2 document says explanations of output, right? 3 A. Right. 4 Q. And Mr. Hunt is the DEU expert 5 or DEN expert -- this is probably output -- from 6 the DEN system, is it not? 7 A. That's what I would interpret 8 it to be. 9 Q. Turn to page three. It says 10 Prozac spontaneous reports annual, 3 June -- I 11 think that's '91. It says from Gary Martindale, 12 DEN query, is that right? 13 A. Yes. 14 Q. Who is Gary Martindale? 15 A. Well, Gary -- well, I think 16 Gary may have been in the systems area or the 17 statistics area, and probably was doing this 18 query for Bob Hunt. But that's speculation. 19 Q. On the next page, at the 20 bottom of the page, there's a Sarah McKenzie. Do 21 you know who that is? 22 A. I believe Sarah was also in 23 DEU. 24 Q. Do you have any reason to Page 272 1 doubt the authenticity of any of this data that 2 they reported to you? 3 A. You mean whether this was 4 truly generated from DEN, the summary? 5 Q. From the DEN system and 6 whether they're accurately reflected here on 7 these pages. 8 A. I don't. 9 Q. Have you requested information 10 from Mr. Hunt, Mr. Martindale or Ms. McKenzie in 11 your work at Lilly? 12 A. I had associations with all of 13 them from time to time. 14 Q. Did you find them to be 15 reliable and were sources of accurate data? 16 A. I think they did their best, 17 yes. 18 Q. Well, was the DEN system, in 19 your judgment, complete and updated accurately or 20 completed and updated periodically to reflect 21 accurately spontaneous adverse events as well as 22 clinical trial data in connection with Prozac? 23 A. Yes. 24 Q. Have you ever seen these Page 273 1 figures before? 2 A. I don't recall ever seeing 3 this document. 4 MR. SMITH: Thank you, Doctor Zerbe. 5 THE WITNESS: Thank you. 6 MR. LORE: No questions. 7 MR. OLTMAN: No questions. 8 (THE WITNESS WAS EXCUSED.) Page 274 1 COMMONWEALTH OF KENTUCKY ) 2 : ss COUNTY OF JEFFERSON ) 3 4 I, MARY KATHLEEN NOLD, A NOTARY PUBLIC IN 5 AND FOR THE STATE OF KENTUCKY AT LARGE, DO HEREBY 6 CERTIFY THAT THE FOREGOING TESTIMONY OF 7 DR. ROBERT ZERBE 8 WAS TAKEN BEFORE ME AT THE TIME AND PLACE AS 9 STATED IN THE CAPTION; THAT THE WITNESS WAS FIRST 10 DULY SWORN TO TELL THE TRUTH, THE WHOLE TRUTH, 11 AND NOTHING BUT THE TRUTH; THAT THE SAID 12 PROCEEDINGS WERE TAKEN DOWN BY ME IN STENOGRAPHIC 13 NOTES AND AFTERWARDS TRANSCRIBED UNDER MY 14 DIRECTION; THAT IT IS A TRUE, COMPLETE AND 15 CORRECT TRANSCRIPT OF THE SAID PROCEEDINGS SO 16 HAD; THAT THE APPEARANCES WERE AS STATED IN THE 17 CAPTION. 18 WITNESS MY SIGNATURE THIS THE 23RD DAY OF 19 AUGUST, 1994. 20 MY COMMISSION EXPIRES MARCH 10, 1996. 21 22 23 _________________________ MARY KATHLEEN NOLD 24 COURT REPORTER AND NOTARY PUBLIC Page 275 1 E R R A T A S H E E T 2 3 STATE OF ) : SS 4 COUNTY OF ) 5 6 I, DR. ROBERT ZERBE, THE UNDERSIGNED 7 DEPONENT, HAVE THIS DATE READ THE FOREGOING PAGES 8 OF MY DEPOSITION AND WITH THE CHANGES NOTED 9 BELOW, IF ANY, THESE PAGES CONSTITUTE A TRUE AND 10 ACCURATE TRANSCRIPTION OF MY DEPOSITION GIVEN ON 11 THE 9TH DAY OF AUGUST, 1994 AT THE TIME AND PLACE 12 STATED THEREIN. 13 PAGE NO. LINE NO. CHANGE REASON Page 276 1 PAGE NO. LINE NO. CHANGE REASON 2 3 4 5 6 7 8 _____________________________ 9 DR. ROBERT ZERBE 10 SWORN TO AND SUBSCRIBED BEFORE ME THIS 11 _____ DAY OF __________, 1994. 12 _____________________________ NOTARY PUBLIC, STATE OF 13 AT LARGE Page 277 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Page 279 1 CONTINUING EXAMINATIONBY MR. SMITH:...............11 2 COMMONWEALTH.....................................264 3 PLAINTIFFS' EXHIBIT NO. 11........................34 4 PLAINTIFFS' EXHIBIT NO. 12........................46 5 PLAINTIFFS' EXHIBIT NO. 13........................50 6 PLAINTIFFS' EXHIBIT NO. 14........................62 7 PLAINTIFFS' EXHIBIT NO. 15........................72 8 PLAINTIFFS' EXHIBIT NO. 16........................85 9 PLAINTIFFS' EXHIBIT NO. 17.......................116 10 PLAINTIFFS' EXHIBIT NO. 18.......................143 11 PLAINTIFFS' EXHIBIT NO. 19.......................150 12 PLAINTIFFS' EXHIBIT NO. 20.......................156 13 PLAINTIFFS' EXHIBIT NO. 21.......................161 14 PLAINTIFFS' EXHIBIT NO. 22.......................172 15 PLAINTIFFS' EXHIBIT NO. 23.......................178 16 PLAINTIFFS' EXHIBIT NO. 24.......................203 17 PLAINTIFFS' EXHIBIT NO. 25.......................209 18 PLAINTIFFS' EXHIBIT NO. 26 ......................212 19 PLAINTIFFS' EXHIBIT NO. 27.......................217 20 PLAINTIFFS' EXHIBIT NO. 28.......................219 21 PLAINTIFFS' EXHIBIT NO. 29.......................231 22 PLAINTIFFS' EXHIBIT NO. 30.......................239 23 PLAINTIFFS' EXHIBIT NO. 31.......................249 24 PLAINTIFFS' EXHIBIT NO. 32.......................260 Page 280 1 CONTINUING EXAMINATIONBY MR. SMITH:...11 2 COMMONWEALTH.......275 3 PLAINTIFFS' EXHIBIT NO. 11......35 4 PLAINTIFFS' EXHIBIT NO. 12......47 5 PLAINTIFFS' EXHIBIT NO. 13...52 6 PLAINTIFFS' EXHIBIT NO. 14...64 7 PLAINTIFFS' EXHIBIT NO. 15...74 8 PLAINTIFFS' EXHIBIT NO. 16...88 9 PLAINTIFFS' EXHIBIT NO. 17.....120 10 PLAINTIFFS' EXHIBIT NO. 18.....148 11 PLAINTIFFS' EXHIBIT NO. 19.....155 12 PLAINTIFFS' EXHIBIT NO. 20.....162 13 PLAINTIFFS' EXHIBIT NO. 21.....167 14 PLAINTIFFS' EXHIBIT NO. 22.....179 15 PLAINTIFFS' EXHIBIT NO. 23.....185 16 PLAINTIFFS' EXHIBIT NO. 24.....211 17 PLAINTIFFS' EXHIBIT NO. 25.....218 18 PLAINTIFFS' EXHIBIT NO. 26 ....221 19 PLAINTIFFS' EXHIBIT NO. 27.....225 20 PLAINTIFFS' EXHIBIT NO. 28.....228 21 PLAINTIFFS' EXHIBIT NO. 29.....240 22 PLAINTIFFS' EXHIBIT NO. 30.....249 23 PLAINTIFFS' EXHIBIT NO. 31.....259 24 PLAINTIFFS' EXHIBIT NO. 32.....271 Page 281