1 NO. 90-CI-6033 JEFFERSON CIRCUIT COURT 2 DIVISION ONE (1) 3 *-*-*-*-* 4 JOYCE FENTRESS, ET AL. PLAINTIFFS 5 6 VS. DEPOSITION FOR PLAINTIFFS 7 8 SHEA COMMUNICATIONS, ET AL. DEFENDANTS 9 10 11 *-*-*-*-* 12 13 14 DEPONENT: GARY D. TOLLEFSON, MD 15 16 DATE: JULY 26, 1994 17 18 REPORTER: MARY KATHLEEN NOLD 19 20 *-*-*-*-* 21 22 KENTUCKIANA REPORTERS 730 WEST MAIN STREET, SUITE 250 23 LOUISVILLE, KENTUCKY 40202 (502) 589-2273 24 1 1 UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF INDIANA 2 INDIANAPOLIS DIVISION 3 IN RE ELI LILLY AND COMPANY ) Prozac Products Liability ) MDL Docket No. 907 4 Litigation ) 5 *-*-*-*-* 6 NO. 91-02496-A 7 JACKIE LYNN BIFFLE, ET AL ) IN THE DISTRICT ) COURT OF 8 V. ) DALLAS COUNTY, TEXAS ) 9 ELI LILLY & COMPANY AND ) 14TH JUDICIAL DISTA PRODUCTS COMPANY ) DISTRICT 10 *-*-*-*-* 11 NO. 92-14775-E 12 RICHARD HAROLD CROSSETT, JR., ) IN THE 13 CHAD H. CROSSETT, AMY MICHELLE ) DISTRICT CROSSETT AND KRISTEN ANN CROSSETT,) COURT OF 14 INDIVIDUALLY AND AS SURVIVORS OF ) AND ON BEHALF OF THE ESTATE OF ) 15 JOCQUETTA ANN CROSSETT, DECEASED ) ) 16 V. ) DALLAS COUNTY, ) TEXAS 17 ELI LILLY & COMPANY, DISTA ) PRODUCTS COMPANY, TEXAS ) 18 PSYCHIATRIC COMPANY, INC. ) D/B/A HCA WILLOW PARK ) 101st JUDICIAL 19 HOSPITAL, JAMES K. WITSCHY, M.D., ) DISTRICT AND DOUG BELLAMY, ED.D ) 20 *-*-*-*-* 21 22 23 24 2 1 NO. A-921,405-C 2 MARIA GUADALUPE REVES ) IN THE INDIVIDUALLY AND AS NEXT ) DISTRICT COURT 3 FRIEND OF GRANT JULIAN REVES ) OF A MINOR CHILD, AND ON BEHALF ) 4 OF THE ESTATE OF CHRISTIAN ) MARIE REVES, DECEASED ) 5 ) V. ) ORANGE COUNTY, 6 ) TEXAS ELI LILLY & COMPANY, DISTA ) 7 PRODUCTS COMPANY, RAVIKUMAR ) KANNEGANTI, M.D., HOSPITAL ) 8 CORPORATION OF AMERICA, A ) TENNESSEE CORPORATION, HEALTH ) 9 SERVICES ACQUISITION CORP., ) A DELAWARE CORPORATION, ) 10 HCA PSYCHIATRIC COMPANY, A ) DELAWARE CORPORATION, TEXAS ) 11 PSYCHIATRIC CO., INC., A/K/A ) AND/OR D/B/A HCA BEAUMONT ) 12 NEUROLOGICAL HOSPITAL, AND HCA) HEALTH SERVICES OF TEXAS, INC.) 128TH JUDICIAL 13 A/K/A AND/OR BEAUMONT ) DISTRICT NEUROLOGICAL HOSPITAL ) 14 *-*-*-*-* 15 16 17 18 19 20 21 22 23 24 3 1 IN THE CIRCUIT COURT OF COOK COUNTY, ILLINOIS COUNTY DEPARTMENT - LAW DIVISION 2 RENATO DI SILVESTRO, Individually) 3 and as Special Administrator of ) the Estate of JOHN DI SILVESTRO, ) 4 Deceased, ) ) 5 Plaintiff, ) ) 6 v. ) No. 91-l-7881 ) 7 ROBERT L. NELSON, et al., ) ) 8 Defendants, ) ) 9 GEORGE MELNICK, M.D., and PETER ) FINK, M.D. ) 10 ) RESPONDENTS IN DISCOVERY.) 11 *-*-*-*-* 12 SUPERIOR COURT OF THE STATE OF CALIFORNIA 13 FOR THE COUNTY OF LOS ANGELES 14 DR. MARIUS SAINES, etc., et al., ) Case No.: ) SC 008331 15 ) Plaintiffs, ) 16 ) vs. ) 17 ) ELI LILLY & COMPANY, a corporation;) 18 DISTA PRODUCTS COMPANY, a Division ) of Eli Lilly & Company; and DOBS 1-) 19 100, Inclusive, ) ) 20 Defendants. ) ___________________________________) 21 *-*-*-*-* 22 23 24 4 1 NO. 93-8792-D 2 DAVID KUNG, DALE KUNG COHEN ) IN THE DISTRICT ROBERT KUNG, AND TIMOTHY KUNG, ) COURT OF 3 INDIVIDUALLY AND AS SURVIVORS ) AND STATUTORY BENEFICIARIES ) 4 OF MAY YUN KUNG, DECEASED ) ) 5 VS. ) DALLAS, COUNTY ) TEXAS 6 ELI LILLY AND COMPANY, DISTA ) PRODUCTS COMPANY, AND MONIQUE ) 7 KUNKLE, PH.D. ) 8 *-*-*-*-* 9 IN THE DISTRICT COURT OF JOHNSON COUNTY, KANSAS CIVIL COURT DEPARTMENT 10 EUGENE HUSLIG, AS ADMINISTRATOR ) 11 AND EXECUTOR AND ON BEHALF OF ) THE ESTATE OF DEBORAH G. WEATHERS ) 12 HUSLIG, DECEASED, AND AS SURVIVING ) HUSBAND AND HEIR AT LAW OF DEBORAH ) 13 G. WEATHERS HUSLIG, DECEASED, ) AND IN HIS INDIVIDUAL CAPACITY AS ) 14 HUSBAND OF DEBORAH G. WEATHERS ) HUSLIG, DECEASED, AND RONALD C. ) 15 WEATHERS, SON OF DEBORAH G. ) WEATHERS HUSLIG, DECEASED, ) CASE NO.: 16 ) 94 C 192 PLAINTIFFS, ) 17 ) COURT NO. 7 VS. ) CHAPTER 60 18 ) MARY L. BILLINGSLEY, EXECUTOR OF ) 19 THE ESTATE OF THAD BILLINGSLEY, ) M.D., DECEASED D/B/A THE BENESSERE ) 20 CENTER, SUSAN C. JOHNSON, PH.D., ) BILLINGSLEY ENTERPRISES, INC., ) 21 F/K/A THAD H. BILLINGSLEY, M.D. ) CHARTERED, D/B/A THE BENESSERE ) 22 CENTER, ELI LILLY AND COMPANY, ) AND DISTA PRODUCTS COMPANY, ) 23 ) DEFENDANTS. ) 24 5 1 CAUSE NO. 93-04911-A 2 LINDA JILL WELCH, CARLINDA WELCH REX, CONNAN ROSS WELCH 3 AND CHAD MICHAEL WELCH, INDIVIDUALLY AND AS SURVIVORS 4 AND STATUTORY BENEFICIARIES OF CARL EUGENE WELCH, DECEASED PLAINTIFFS 5 V. 6 ELI LILLY AND COMPANY, DISTA 7 PRODUCTS COMPANY, NOE NEAVES, M.D., AND MINITH-MEIER 8 CLINIC, P.A. DEFENDANTS 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 6 1 I N D E X 2 WITNESS: GARY D. TOLLEFSON, MD 3 DIRECT EXAMINATION BY MS. ZETTLER.............. 10 4 EXAMINATION BY MR. SMITH...................... 164 5 WITNESS EXCUSED............................... 287 6 7 8 TOLLEFSON EXHIBIT NO. 1........................ 85 9 TOLLEFSON EXHIBIT NO. 2........................ 99 10 TOLLEFSON EXHIBIT NO. 3....................... 146 11 TOLLEFSON EXHIBIT NO. 4....................... 224 12 TOLLEFSON EXHIBIT NO. 5....................... 239 13 TOLLEFSON EXHIBIT NO. 6....................... 245 14 TOLLEFSON EXHIBIT NO. 7....................... 257 15 TOLLEFSON EXHIBIT NO. 8....................... 268 16 TOLLEFSON EXHIBIT NO. 9....................... 272 17 TOLLEFSON EXHIBIT NO. 10...................... 277 18 19 20 21 22 23 24 7 1 THE FOLLOWING DEPOSITION 2 OF GARY D. TOLLEFSON, MD, WAS TAKEN AT THE OFFICES 3 OF BAKER & DANIELS, 300 NORTH MERIDIAN STREET, 4 SUITE 270, INDIANAPOLIS, INDIANA, 46204, ON JULY 5 26, 1994; SAID DEPOSITION TAKEN PURSUANT TO NOTICE 6 IN ACCORDANCE WITH THE RULES OF CIVIL PROCEDURE. 7 *-*-*-*-* 8 9 A P P E A R A N C E S 10 11 NANCY ZETTLER COUNSEL FOR PLAINTIFFS 12 1405 WEST NORWELL LANE SCHAUMBURG, ILLINOIS 60193 13 14 PAUL SMITH COUNSEL FOR PLAINTIFFS 15 745 CAMPBELL CENTER 2 8115 NORTH CENTRAL EXPRESSWAY 16 DALLAS, TEXAS 75206 17 JOE FREEMAN 18 LAWRENCE J. MEYERS COUNSEL FOR ELI LILLY AND COMPANY 19 FREEMAN & HAWKINS 4000 ONE PEACHTREE CENTER 20 303 PEACHTREE STREET, N.E. ATLANTA, GEORGIA 30308-3243 21 22 JOHN BRENNAN COUNSEL FOR ELI LILLY AND COMPANY 23 FOUR GATEWAY CENTER 100 MULBEWRRY STREET 24 NEWARK, NEW JERSEY 07102-4096 8 1 APPEARANCES CONTINUED 2 BEATRICE M. SMITH COUNSEL FOR BEAUMONT NEUROLOGICAL HOSPITAL 3 FRIEND & ASSOCIATES LLP 1301 MCKINNEY, #2900 4 HOUSTON, TEXAS 77010 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 9 1 MR. FREEMAN: Let me just 2 state at this time a little brief stipulation. 3 This is the deposition of GARY TOLLEFSON, MD, 4 taken on behalf of the plaintiffs for purposes of 5 discovery and use at the trial on any of the cases 6 heretofore recited. The deposition of Doctor 7 Tollefson is taken by agreement of counsel and by 8 notice in the offices of Baker and Daniel in 9 Indianapolis. 10 Objection will be made at 11 this time as to any leading questions that Doctor 12 Tollefson's own counsel may put to him or any 13 objection that any lawyer may have to the 14 witness's response to the questions propounded. 15 All other objections will be reserved until the 16 time of court hearing. 17 * * * * * 18 GARY D. TOLLEFSON, MD, 19 called by Plaintiffs, after having been first duly 20 sworn, was examined and deposed as follows: 21 22 DIRECT EXAMINATION 23 BY MS. ZETTLER: 24 Q Doctor Tollefson, if you 10 1 could state your full name for the record, 2 please. 3 A Gary Dennis Tollefson. 4 Q And you are a medical 5 doctor? 6 A That is correct. 7 Q Can you give us your 8 current address? 9 A xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx 10 xxxxxxxxxxxxxxxxxxxxxxxxx 11 Q And how long have you 12 lived at that address? 13 A xxxxxxxxxxx. 14 Q Do you have any current 15 plans to move in the near future? 16 A I do not. 17 Q What is your date of 18 birth? 19 A 2/1/51. 20 Q Are you married? 21 A Yes, I am. 22 Q How long have you been 23 married? 24 A Approximately four years. 11 1 Q Do you have any children? 2 A Yes. 3 Q Can you give us their 4 names and ages, please? 5 A xxxxxxxxxxxxxxxxxxxxxx 6 xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx 7 Q Is this your second 8 marriage? 9 A Yes. 10 Q Would you give us a 11 little bit of your educational background starting 12 after high school, please? 13 A When I left high school, 14 I entered the undergraduate program at the 15 University of Minnesota, Minneapolis, Minnesota, 16 completed a four-year degree, psychology, BA summa 17 cum laude. Then I went on to the University of 18 Minnesota Medical School, completed my medical 19 degree that campus, then went on and obtained a 20 Ph.D. in psychopharmacology, also at the 21 University of Minnesota as part of the graduate 22 school. 23 Q Also at the University of 24 Minnesota? 12 1 A Yes. 2 Q What year did you 3 complete your undergrad work? 4 A Undergraduate work was 5 completed in 1973. 6 Q Your graduate or 7 undergraduate? 8 A Undergraduate. 9 Q I'm sorry, okay. And 10 when did you start medical school? 11 A 1973. 12 Q When did you complete 13 medical school? 14 A 1976. 15 Q And when did you achieve 16 your Ph.D.? 17 A January 1980. 18 Q Did you do an internship 19 and residency as part of your medical school 20 education? 21 A Yes, I did. 22 Q Can you tell us where you 23 did those, please? 24 A Internship was done at a 13 1 teaching affiliate of the University of Minnesota, 2 in St. Paul, Minnesota, St. Paul Ramsey Medical 3 Center. A residency in psychiatry, University of 4 Minnesota Hospitals. 5 Q Was your internship a 6 general rotating? 7 A Yes. 8 Q Are you Board certified, 9 sir? 10 A Yes, in psychiatry. 11 Q When did you become Board 12 certified? 13 A It was approximately 14 1980. 15 Q When did you start 16 working at Eli Lilly and Company? 17 A June 1991. 18 Q Have you worked 19 continuously for Lilly since that date? 20 A Yes. 21 Q Please give us your 22 employment history from when you started undergrad 23 to the time that you started working at Lilly, and 24 when I say employment history, I mean like 14 1 substantive, not odd jobs or waitering or things 2 of that nature. 3 A Well, that's -- in 4 summary, I had one employment relationship, it was 5 with St. Paul Ramsey Medical Center, where I began 6 my employment after completion of my residency, 7 and where I was until I chose to come to Eli Lilly 8 and Company. 9 Q What year did you begin 10 your residency? 11 A I began my residency in 12 1976. 13 Q So from 1976 until 19 -- 14 June of 1991, you worked for St. Paul Ramsey 15 Medical Center? 16 A No, that's not correct. 17 As a resident, I was at the University of 18 Minnesota Hospitals, it was a four-year 19 residency. Upon completion of the residency, I 20 began my employment relationship with St. Paul 21 Ramsey Medical Center. 22 Q Approximately 1980 to 23 1991, then? 24 A Yes, that's correct. 15 1 Q And what position did you 2 hold at St. Paul Ramsey Medical Center? 3 A Well, I began as an 4 instructor in both psychiatry and emergency 5 medicine, and worked on what was called the 6 consultation liaison psychiatric service. Also as 7 a member of the academic faculty at the University 8 of Minnesota, beginning as an assistant 9 professor. The time that I left, I was Chairman 10 of the Department of Psychiatry at St. Paul Ramsey 11 Medical Center and an Associate Professor, 12 Department of Psychiatry, at the University of 13 Minnesota. 14 Q Okay. Let's make sure 15 I've got this down right. You started as an 16 instructor at St. Paul Ramsey? 17 A Correct. 18 Q In what department? 19 A Psychiatry and emergency 20 medicine. 21 Q And did you then move 22 into the position as a consultation liaison for 23 psych services, or did you do that in conjunction 24 with your position as an instructor? 16 1 A In conjunction. 2 Q How long did that 3 position as consultation liaison last? 4 A Six years. 5 Q From 1980 to 1986? 6 A Approximately. 7 Q And then you joined the 8 faculty of the University of Minnesota Medical 9 School? 10 A No, I joined the faculty 11 in 1980. 12 Q And that was the 13 University of Minnesota? 14 A Correct. Those are 15 parallel appointments. 16 Q Okay. And when did you 17 become chairman of the Department of Psychiatry at 18 St. Paul Ramsey? 19 A 1987 or so. 20 Q And you continued in that 21 position until 1991, when you went to Eli Lilly? 22 A That's correct. 23 Q And then, I'm sorry, the 24 last one, you were also associate professor? 17 1 A I was promoted to 2 Associate Professor at the University of Minnesota 3 Hospitals. 4 Q Have you ever maintained 5 a private practice, Doctor? 6 A Actually during those 7 entire ten plus years at Ramsey, I was also a 8 member of Ramsey Clinic, which is a private 9 multispecialty practice in St. Paul. 10 Q And did you see patients, 11 did you in fact treat patients? 12 A Yes, during that entire 13 time. 14 Q Did you limit your 15 practice to psychiatry or did you treat patients 16 with other types of ailments? 17 A Principally psychiatry. 18 Q And how long were you 19 affiliated with the Ramsey Clinic? 20 A For those ten plus years, 21 from 1980 to assuming an employment relationship 22 with Eli Lilly. 23 Q Who did you first talk to 24 at Lilly before coming on board there? 18 1 MR. FREEMAN: Are you 2 speaking of an interview? 3 MS. ZETTLER: Yes. 4 A Doctor Leigh Thompson. 5 Q Did you approach Doctor 6 Thompson or did Doctor Thompson approach you? 7 A Regarding? 8 Q A job at Lilly. 9 A Doctor Thompson 10 approached me. 11 Q Had you known Doctor 12 Thompson prior to speaking with him about a job at 13 Lilly? 14 A I had met Doctor Thompson 15 on two occasions while consulting for Eli Lilly 16 and Company. 17 Q Prior to joining Lilly, 18 had you conducted any clinical trials on drugs 19 manufactured by Lilly? 20 A One. 21 Q Which drug? 22 A Prozac. 23 Q When did you conduct that 24 clinical trial? 19 1 A It began late in 1990 and 2 carried into '91, and that trial was not completed 3 at the time that I joined Eli Lilly and Company. 4 Q Other than that clinical 5 trial, had you conducted -- not just for Lilly, 6 but for any company, had you conducted any 7 clinical trials on psychotropic medications in the 8 past? 9 A Yes. 10 Q Can you give us an idea 11 of some of those drugs, if they are marketed? 12 A Buspar, Xanax, 13 Nimodipine. 14 Q Could you spell that, 15 please? 16 A N-I-M-O-D-I-P-I-N-E. 17 Trazodone. I believe that's it. 18 Q Besides these four drugs 19 and Prozac, have you conducted clinical trials on 20 any other medications? 21 A Yes. 22 Q What types of 23 medications? 24 A Antidepressants, 20 1 antianxiety agents, antipsychotic medications, and 2 putative therapies for Alzheimer's, other 3 dementias. 4 Q Besides Prozac or 5 fluoxetine hydrochloride, have you conducted 6 clinical trials on any other Lilly drugs? 7 A I have not. 8 Q When did you first start 9 conducting clinical trials in general? 10 A Approximately 1983. 11 Q And how did you get 12 involved in conducting clinical trials? 13 A My senior mentor at 14 St. Paul Ramsey Medical Center was a clinical 15 researcher who introduced me to clinical research 16 and allowed me to be a co-investigator. 17 Q And what was that 18 person's name? 19 A Doctor Vincente, 20 V-I-N-C-E-N-T-E, Tuason, T-U-A-S-O-N. 21 Q Is this Doctor Vincente 22 Tuason an MD? 23 A Yes. 24 Q Is Doctor Tuason still at 21 1 the University of Minnesota? 2 A No, he is not. 3 Q Where is he located now? 4 A University of New Mexico. 5 Q To your knowledge, had 6 Doctor Tuason conducted any clinical trials on 7 Prozac prior to your becoming involved in clinical 8 trials on Prozac? 9 A No. 10 Q Did Doctor Tuason 11 participate in the clinical trial on Prozac that 12 you participated in? 13 A No, he did not. 14 Q How did you become 15 involved in studying Prozac, did you approach 16 Lilly or did they approach you? 17 A I approached Lilly. 18 Q Who at Lilly did you talk 19 to? 20 A Doctor John 21 Heiligenstein. 22 Q And when did you first 23 approach Doctor Heiligenstein about conducting a 24 clinical trial on fluoxetine? 22 1 A In 1990. 2 Q Why did you talk to 3 Doctor Heiligenstein about conducting a clinical 4 trial on fluoxetine? 5 A I had a scientific 6 hypothesis that I proposed that I would be 7 interested in investigating. 8 Q You said earlier that you 9 had met Doctor Thompson on a couple of occasions 10 while consulting for Lilly, correct? 11 A Correct. 12 Q Did you begin consulting 13 for Lilly before or after you started this 14 clinical trial on Prozac? 15 A Shortly after. 16 Q Prior to contacting 17 Doctor Heiligenstein regarding the Prozac clinical 18 trial idea that you had in 1990, had you spoken 19 with anybody at Eli Lilly? 20 A Yes. 21 Q Who? 22 A I had a number of 23 interactions with a variety of people in the 24 marketing division, I could not provide you with a 23 1 comprehensive list of all those people. 2 Q Can you give me some of 3 the names? 4 A Charles Perry, Rick 5 Smith. Those are the two that come to mind. 6 Q When you said you had 7 spoken with Mr. Perry and Mr. Smith, among others, 8 in the marketing division at Lilly, in what 9 capacity did you speak with them? 10 A Providing ongoing, 11 continuing medical education programs. 12 Q They asked you to 13 participate? 14 A Yes. 15 Q When did you first -- 16 strike that. 17 Did you in fact 18 participate? 19 A Yes. 20 Q When did you first start 21 participating in ongoing medical education 22 programs? 23 A Say approximately 1988, 24 possibly 1989. 24 1 Q Prior to beginning your 2 participation in -- strike that. 3 Were these continuing 4 medical education programs sponsored by Lilly, to 5 the best of your knowledge? 6 A No. 7 Q Who were they sponsored 8 by? 9 A Various academic 10 institutions. 11 Q Why were Lilly marketing 12 people approaching you to participate then? 13 A They would ask, on behalf 14 of a university, whether or not I would be 15 interested in visiting that particular university 16 or that location to present. 17 Q What was the subject 18 matter of these continuing medical education 19 programs? 20 A Typically the recognition 21 and management of depression. 22 Q To your knowledge, was 23 Lilly providing funding in any manner for these 24 educational programs? 25 1 A I couldn't speak to each 2 and every one. I certainly know of circumstances 3 where unrestricted educational grants would be 4 provided at an institution to help defray costs 5 for putting on a symposium. 6 Q Is it your experience 7 throughout your professional years that drug 8 company personnel would approach potential 9 speakers for symposia or educational programs that 10 they were not at least in part sponsoring? 11 A Yes. 12 Q Okay, give me some 13 examples. 14 A I think it's common 15 practice for any number of pharmaceutical 16 companies, because of their national presence, to 17 speak to people in a certain region about would 18 they be interested in working with a faculty 19 member in another region. It's really a 20 convenience for communication, and I'm sure they 21 view it as a service. 22 Q The drug company views it 23 as a service? 24 A Yes. For example, since 26 1 you asked for one, the Upjohn Company has a 2 program called the Medical Sciences Liaison. The 3 MSL really is working with academic institutions 4 to help provide educational services, whether or 5 not they are directly involved sponsoring them or 6 not. They're really doing -- they're building a 7 relationship with a key customer. 8 Q Do you know of any 9 situation where the Upjohn Company has provided 10 educational services in situations where they are 11 not marketing a drug to treat a certain 12 affliction? 13 A Yes. 14 Q Give me some examples 15 there. 16 A Depression. 17 Q Any others? 18 A Like in general 19 psychosomatic illness, personality disorders. 20 Q To your knowledge, did 21 Eli Lilly participate in educational programs 22 regarding the diagnosis and treatment of 23 depression prior to studying Prozac? 24 A I have no idea. 27 1 Q You are aware that Lilly 2 is involved in educational -- medical educational 3 programs at this point in educating physicians and 4 the public to a certain extent as to the treatment 5 and diagnosis of depression, are you not? 6 A Yes. 7 Q And that is done through 8 the National Depression Screening Day, for 9 instance? 10 A That would be one 11 example. 12 Q As well as the National 13 Depression Awareness Campaign? 14 A Yes. 15 Q Tell us about the 16 clinical -- strike that. 17 Have you run more than 18 one clinical trial on Prozac? 19 A No, I have not. 20 Q Tell us about the 21 clinical trial that you ran on Prozac, starting 22 with your idea. 23 A The idea was that 24 selective serotonin uptake inhibitors, as a class 28 1 of medications, actually would be beneficial for 2 patients who were depressed and had baseline 3 agitation, relative to conventional antidepressant 4 medications. 5 Q Okay. Patients who are 6 depressed and had baseline agitation? 7 A Correct. 8 Q Relative to conventional 9 antidepressants, is that what you said? 10 A Yes. 11 Q When you say relative to 12 conventional antidepressants, are you speaking of 13 tricyclics, for instance? 14 A Yes. 15 Q Any others? 16 A No. 17 Q When you approached 18 Doctor Heiligenstein with this idea, what did he 19 say? 20 A He was interested in the 21 scientific hypothesis. 22 Q How did you come up with 23 that idea? 24 A Well, part of my job at 29 1 St. Paul Ramsey was in the psychopharmacology 2 division, and it was designed to come up with 3 creative or innovative study hypotheses to address 4 clinical concerns or clinical issues. Agitation 5 as a component of depression is a common 6 presentation that clinicians struggle with. 7 Q Had you prescribed Prozac 8 for depressed patients prior to that? 9 A Yes. 10 Q Had you seen patients who 11 had become agitated on Prozac? 12 A I'd seen patients who had 13 baseline agitation that during the natural history 14 of their disease, agitation or any other symptom 15 of the disease might tend to crescendo. 16 Q That's not answering my 17 question, Doctor. Had you seen any patients who 18 had become agitated while on Prozac? 19 A In a non-causally related 20 sense, yes. 21 Q How do you know it was 22 non-causally related? 23 A Because it was a present 24 feature at baseline and it was associated with 30 1 worsening of their depression in the context of a 2 failure to respond to intervention. 3 Q So it's your belief that 4 these people who became more agitated on the 5 Prozac became more agitated because the Prozac was 6 not working as opposed to the Prozac exacerbating 7 the agitation, correct? 8 A That's my opinion. 9 Q Since joining Eli Lilly 10 and Company, have you prescribed Prozac to any 11 patients? 12 A I have not. 13 Q Since working at Eli 14 Lilly, have you gleaned any information that has 15 changed your mind as to the causal relationship of 16 those patients that you have seen become more 17 agitated on Prozac? 18 A I have not. 19 Q Do you consider Prozac a 20 sedative type antidepressant? 21 A I do not. 22 Q Generally, tricyclics 23 tend to be more sedating, do they not? 24 A I think that may be a 31 1 generalization. There are some of the tricyclic 2 medications which are potent antihistaminic 3 compounds and in turn are very sedating. There 4 are others that are less so. 5 Q You said earlier that you 6 did conduct, at least in part, the clinical trial 7 that you proposed to Doctor Heiligenstein, 8 correct? 9 A Correct. 10 Q When did you begin that 11 clinical trial? 12 A Probably late 1990. 13 Q And was that clinical 14 trial structured to study the effect of fluoxetine 15 versus tricyclic antidepressants on depressed 16 patients who had baseline agitation? 17 A Yes. 18 Q Who developed the 19 protocol for that clinical trial? 20 A I did. 21 Q Did Lilly provide you 22 with any information that you used in the 23 preparation of the protocol? 24 A Clinical investigators' 32 1 brochure. 2 Q Anything else? 3 A No. 4 Q Did Lilly maintain final 5 approval of the protocol before you initiated it? 6 A Yes. 7 Q Did they make any changes 8 in your draft? 9 A Nothing substantive that 10 I recall. 11 Q Did they provide you with 12 a draft of one of their protocols to work from 13 when structuring your protocol? 14 A No. 15 Q Was this a single or 16 multicentered clinical trial? 17 A Two sites. 18 Q Where was the other site? 19 A Madison, Wisconsin. 20 Q I take it your site was 21 at the University of Minnesota? 22 A At St. Paul Ramsey 23 Medical Center, right. 24 Q Were you the principal 33 1 clinical investigator on that trial? 2 A Yes. 3 Q Was there a principal 4 clinical investigator at the Madison site? 5 A Yes. 6 Q Who was that? 7 A Doctor John Greist, 8 G-R-E-I-S-T. 9 Q Was this an open label or 10 a blinded study? 11 A Blinded. 12 Q What were the comparators 13 that were used? 14 A Imipramine. 15 Q Any others? 16 A No. 17 Q Was placebo used? 18 A No. 19 MR. FREEMAN: No 20 placebo? 21 THE WITNESS: No placebo. 22 Q (BY MS. ZETTLER) How 23 long was the treatment period? 24 A Six weeks. 34 1 Q What types of patients 2 were excluded from this study? 3 A Patients with primary 4 diagnoses other than major depression with 5 agitation, or any patients with a medical 6 contraindication to receiving antidepressant 7 pharmacotherapy. 8 Q Was this an inpatient or 9 outpatient study? 10 A Outpatient. 11 Q Were people who posed a 12 serious risk of suicide excluded? 13 A It was at the 14 investigator's discretion. 15 Q Is that what it stated in 16 the protocol? 17 A Yes. 18 Q Was there a criteria that 19 was used to decide whether or not a particular 20 patient posed a serious suicidal risk? 21 A Clinical judgment. 22 Q Did you in fact, as 23 principal investigator on the study at the 24 St. Paul site, exclude patients who you believed 35 1 were a serious suicidal risk? 2 A No. 3 Q Did you include people 4 that you considered a serious suicidal risk? 5 A There were patients we 6 included who were a suicide risk, yes. 7 Q How about serious 8 suicidal risk? 9 A I'd probably need your 10 definition of serious. 11 Q Well, Doctor, we've seen 12 a lot of protocols conducted -- from which trials 13 were conducted by Lilly where patients who posed a 14 serious suicidal risk were excluded, and we 15 haven't really gotten a straight definition of 16 what that means. So, my question is, were there 17 any patients that you excluded because you felt 18 that they posed too much of a risk of suicide to 19 be included in the trial? 20 A Only those that had 21 imminent and active suicidal intentions where 22 inpatient hospitalization was the treatment 23 intervention of choice, and if they were 24 inpatients, of course, they were excluded from 36 1 this protocol. 2 Q You said imminent active 3 suicidal ideation? 4 A Imminent intention -- 5 Q Intention, okay. 6 A -- to commit suicide. 7 Q And what kind of criteria 8 did you use to decide if the patient fit this 9 category? 10 A If the patient identified 11 a desire to commit an act of suicide, and often 12 whether or not they had a plan that had been put 13 together, so the method to that suicide. 14 Q Did you have a criteria 15 such as a certain rating on the Hamilton 16 Depression 3 question? 17 A No. 18 Q So a patient could have 19 rated a four on the HAMD-3, but if they did not 20 fit one of these other categories as you just 21 discussed, they may still be allowed on to the 22 trial? 23 A That would be possible. 24 Q Do you recall any 37 1 situation where that happened? 2 A I don't recall. 3 Q How about Doctor Greist; 4 to your knowledge did Doctor Greist exclude anyone 5 who he considered a serious suicidal risk from the 6 study? 7 A I don't know anything 8 about Doctor Greist's patient cohort. 9 Q Did you discuss the study 10 at all with Doctor Greist? 11 A The study, yes. 12 Q On how many occasions did 13 you discuss the study with Doctor Greist? 14 A Oh, more than two and 15 less than ten. 16 Q Have you given a 17 deposition before, Doctor? 18 A Yes. 19 Q On how many occasions? 20 A Two or three. 21 Q Have you given 22 depositions in connection with any of the Prozac 23 cases? 24 A No, I have not. 38 1 Q I just want to make sure 2 that we have the ground rules down. You're doing 3 a pretty good job so far, but -- 4 A Thank you. 5 Q You have to speak out 6 loud; obviously, your answers have to be verbal so 7 that the court reporter can take them down and so 8 that they can be recorded, okay? 9 A Sure. 10 Q Also, if there are any 11 questions that I ask you that you do not 12 understand, please let me know and I will rephrase 13 it so that you understand it, okay? 14 A That's fair. 15 Q Anytime you want to take 16 a break, let us know, we'll take a break; it's not 17 an endurance contest. 18 A All right. 19 Q Also, again, you've been 20 doing a pretty good job of it, but we can't talk 21 over each other. It makes it difficult for Kathy 22 to take it down, okay? So do your best to let me 23 finish my questions and I'll do my best to let you 24 finish your answers before I start a new question, 39 1 okay? 2 A I appreciate your 3 patience with me. 4 Q Did the protocol that you 5 developed allow for the administration of 6 concomitant medications in the clinical trial? 7 A Yes. 8 Q What concomitant 9 medications? 10 A Therapies that the 11 patient might have been taking prior to being 12 randomized into the study for concomitant medical 13 disorders, with the exclusion of any psychoactive 14 medication. 15 Q Was there a placebo 16 washout period in the study? 17 A There was a one-week 18 placebo run-in. 19 Q Did the protocol allow 20 for the concomitant use of, say, chloral hydrate 21 for insomnia at any time during the trial? 22 A That's a standard 23 inclusion for almost, I think, every study I've 24 been involved with, with all corporate sponsors. 40 1 So I don't recall specifically in this case, but I 2 would suspect that it was allowable. 3 Q Did you retain a copy of 4 your protocol in your files when you went to 5 Lilly? 6 A When I went to Lilly? 7 No. 8 Q Do you recall at any time 9 looking at the protocol for the agitated depressed 10 patient study -- if it's okay to call it that. 11 A Uh-huh. 12 Q -- while at Lilly? 13 A Would you repeat that? 14 Q Sure. Do you recall, 15 once you got to Lilly, ever looking at the 16 protocol that you had developed for the agitated 17 depressed patient study, for any reason? 18 A Not at Lilly. 19 Q Besides the possible 20 concomitant use of chloral hydrate for insomnia, 21 do you recall if the protocol allowed for, say, 22 the concomitant use of benzodiazepines for any 23 reason whatsoever during the clinical trial? 24 A It did not allow for 41 1 that. 2 Q Do you recall whether or 3 not the protocol allowed for any other 4 concomitant -- use of any other concomitant 5 medications besides the therapies the patients may 6 have been on prior to beginning the trial and the 7 chloral hydrate possibly? 8 A No others. 9 Q Was that study completed? 10 A Yes. 11 Q When was it completed? 12 A Early in 1992. 13 Q Was your portion of the 14 study completed when you went to Lilly? 15 A I'm not sure what you 16 mean by my portion. 17 Q Had you enrolled all the 18 patients you had wanted to enroll in your arm of 19 the two-site study? 20 A No. 21 Q Was there somebody at the 22 University of Minnesota who had taken over for you 23 when you went -- left to go to Lilly? 24 A Yes. 42 1 Q Who took over for you? 2 A Doctor Ronald Landbloom. 3 Q Can you spell the last 4 name, please? 5 A L-A-N-D-B-L-O-O-M. 6 Q He became principal 7 investigator on that arm of the study? 8 A Yes, he was a 9 co-investigator prior, and then became principal. 10 Q Did -- to your knowledge, 11 did Doctor Greist complete his arm of the study? 12 A Yes. 13 Q Do you recall -- strike 14 that. 15 Prior to going to Lilly, 16 while you were working on this study, do you 17 recall reporting any serious adverse events to 18 Lilly that occurred during the clinical trial? 19 A I don't remember any 20 serious events that occurred in the trial. 21 Q Do you recall whether or 22 not any of the patients discontinued the study 23 because of adverse events? 24 A Yes. 43 1 Q What was the planned 2 enrollment of the study? 3 A Approximately eighty 4 randomized patients. 5 Q Eighty per site or eighty 6 total? 7 A Eighty total. 8 Q And how about the 9 completion expectation? 10 A It really did not have a 11 completion expectation. 12 Q What types of scales were 13 used in the study? 14 A The principal rating 15 scale for mood was the Hamilton depression rating 16 scale. We created, from the research diagnostic 17 criteria, a rating instrument that we could use 18 for agitation. We used the Reynolds Suicide or 19 Suicidality Questionnaire. 20 Q Is that the Adult 21 Suicidal Ideation Questionnaire? 22 A Yes. And the Hamilton 23 rating scale for anxiety. 24 Q So the HAMD as well as 44 1 the HAMA? 2 A Correct. 3 Q Any others? 4 A I don't believe so. 5 Q Did you use any patient 6 self-rating scales? 7 A No. 8 Q Why not? 9 A Typically we did not use 10 subjective scales because they are poorly 11 validated; although each of the scales, of course, 12 that I mentioned involves patient report. 13 Q When you say it involves 14 patient report, it involves the rater's 15 interpretation of patient report, does it not? 16 A The scales all involve a 17 certain level of clinical sophistication in 18 analysis of a patient's response. 19 Q I'm sorry, I forgot, so 20 please help me out here, did you say this study 21 started in 1990? 22 A Yes. 23 Q Do you recall when in 24 1990? 45 1 A No, I do not. 2 Q Was it the early part of 3 1990 or the late part of 1990? 4 A I believe I said earlier 5 the latter part. 6 Q Whose idea was it to use 7 the Adult Suicidal Ideation Questionnaire in the 8 study? 9 A It was a suggestion of 10 Doctor Greist's. 11 Q Who recruited Doctor 12 Greist to work on this study? 13 A I did. 14 Q Why did you choose Doctor 15 Greist? 16 A He was both a personal 17 friend and a very respected research colleague. 18 Q Is he a psychiatrist? 19 A Yes. 20 Q Did you talk to Doctor 21 Greist about participating in the study before or 22 after approaching Lilly? 23 A After. 24 Q When you developed this 46 1 idea for the agitation study, did you intend it to 2 have more than one arm -- or, I'm sorry, more than 3 one site? 4 A If -- that was really 5 dependent on the number of subjects that would be 6 necessary to have adequate statistical power to 7 differentiate one arm from another arm. As it 8 turned out, the sample size calculated that we 9 needed would require a second investigator to 10 complete it in a timely way. 11 Q Did you have a goal for 12 completion, time-wise? 13 A No. 14 Q Did Lilly want you to 15 complete the study within a certain period of 16 time? 17 A No, no time frame was 18 mentioned. 19 Q What do you mean when you 20 say in a timely manner? 21 A We typically don't like 22 to have protocols drag on for more than a year and 23 a half or two years at our site because it 24 suggests we have somehow defined the study 47 1 population we want to look at incorrectly; we're 2 not finding any of those patients, essentially. 3 Q How do you generally -- 4 strike that. 5 Where do you pull your 6 patient pool from? 7 A While I was at St. Paul 8 Ramsey Medical Center? 9 Q Right. 10 A It's really quite 11 diverse. We had a number of HMO contracts, so it 12 was not infrequent that we would have patients 13 that were enrollees in HMO who would volunteer. 14 We also were involved with providing care to the 15 chronic and persistently mentally ill within 16 Ramsey County. So, either through our clinics or 17 the community medical health center we might 18 receive patient referrals. We also, within the 19 institution, had made individuals' physicians 20 aware of clinical trials in depression, so we 21 would receive referrals from family practice or 22 internal medicine as well. 23 Q To your knowledge, did 24 the center ever advertise for patients? 48 1 A I don't believe so, in 2 this study; although in the past, with other 3 sponsors, other protocols, we have used 4 advertisements, if we're bringing in a different 5 type of patient than typically would use our 6 facility. 7 Q To your knowledge, did 8 Doctor Greist advertise for patients in his part 9 of the study? 10 A I don't know. 11 Q Did the protocol prohibit 12 advertising? 13 A I don't believe so. 14 Q Besides Doctor 15 Heiligenstein, did you discuss your proposed study 16 with anybody else at Lilly? 17 A I did not. 18 Q Did you meet with Doctor 19 Heiligenstein prior to developing the protocol for 20 the study? 21 A No. 22 Q Did you go to Doctor 23 Heiligenstein with the protocol in hand? 24 A Essentially with a draft 49 1 protocol. 2 Q Did Doctor Heiligenstein 3 make any suggestions or changes on the protocol? 4 A Minor ones, not 5 substantial. 6 Q Can you give me some 7 ideas of what types of changes he made? 8 A No, I don't recall. As I 9 said, they were not of substance, more of format. 10 Q Did he provide you with 11 an informed consent -- 12 A No. 13 Q -- to use in the study? 14 A No. 15 Q Did he suggest rating 16 scales that you had either considered and not 17 included or that you had included and he wanted 18 taken out? 19 A No, I don't believe so. 20 Q Did you submit this 21 protocol to the FDA for approval? 22 A No. 23 Q To your knowledge, did 24 Lilly submit it to the FDA? 50 1 A That I don't know. 2 Q Did you need to get 3 approval from an institutional review board to 4 conduct this study? 5 A Yes. 6 Q Where is that review 7 board located? 8 A At St. Paul Ramsey 9 Medical Center. 10 Q Did they make any 11 suggestions about -- for changes in the protocol? 12 A I don't recall any. 13 Q When did you get the okay 14 from Lilly to conduct this study? 15 A It would have been 16 sometime, I'm thinking, in the latter half of 17 1990. 18 Q Okay, do you recall how 19 many months before the actual trial started? 20 A No. 21 Q More than two? 22 A Probably not, probably 23 less than. 24 Q Did Lilly sponsor the 51 1 study? 2 A They provided a grant for 3 the study. 4 Q Did they provide the 5 clinical trial materials? 6 A Yes. 7 Q Did they provide the 8 clinical reporting forms? 9 A Yes. 10 Q Did you use a 11 computerized reporting form or a hard copy 12 reporting form? 13 A Hard copy. 14 Q Did they provide -- 15 besides the clinical investigation manual, did 16 they provide you with any other documentation that 17 they wanted you to use during the study; for 18 instance, an adverse event dictionary, things of 19 that nature? 20 A They didn't provide 21 that. We typically used one that was common 22 amongst our sponsors for a variety of studies. 23 Q Which one was that? 24 A Co-Start. 52 1 Q Did you discuss using 2 Co-Start with Lilly? 3 A No. 4 Q Are you now familiar with 5 the ELECT Dictionary? 6 A No, we didn't use that in 7 any great degree at our site. 8 Q But, I mean, you are 9 aware that Lilly had its own dictionary based on 10 the Co-Start, correct? Now you are aware of that? 11 A Now that you've told me. 12 Q You've never heard of 13 ELECT before today? 14 A I've heard of it. I'm 15 not sure of its relationship relative to Co-Start. 16 Q Okay. Is it your 17 understanding that Lilly uses the Co-Start 18 dictionary now? 19 A Yes. 20 Q When did they start using 21 that dictionary? 22 A That I don't know. 23 Q How many patients did you 24 enroll in your -- at your site on the study? I'm 53 1 not talking about people who completed the study, 2 I'm talking about people who were initially 3 enrolled. 4 A Seventy to eighty. 5 Q And how many of those 6 patients completed the study? 7 A I don't know a precise 8 number. 9 Q Can you give me an 10 estimate? 11 A We had approximately half 12 of those randomized to imipramine who dropped out 13 due to adverse events, and approximately ten to 14 fifteen percent of those randomized to fluoxetine 15 who dropped for an adverse event. 16 Q What percentage of the 17 seventy to eighty patients that were enrolled in 18 the study were randomized to imipramine? 19 A It was equal, one to one 20 at the site. 21 Q And you said -- so 22 assuming it's seventy to eighty patients that you 23 enrolled, it was either thirty-five to forty 24 patients that were randomized to each arm of the 54 1 study, correct? 2 A That's correct. 3 Q Okay. And you say that 4 half of the thirty-five to forty patients 5 randomized to imipramine dropped out because of 6 adverse events? 7 A Approximately. 8 Q What types of adverse 9 events? 10 A Well, they ranged from 11 central nervous system events, such as anxiety or 12 excessive motor activity, nervousness, to 13 excessive sedation, constipation, light- 14 headedness, dry mouth, a variety of other 15 cholinergic phenomena. 16 Q Okay, and you listed CNS 17 events. When you say CNS, you mean central 18 nervous system, correct? 19 A Correct. 20 Q As anxiety, excessive 21 nervousness, excessive sedation, and were there 22 others? 23 A Light-headedness. And 24 others that would be non-CNS, I mentioned 55 1 constipation. 2 Q Okay. And I believe you 3 said ten to fifteen percent of the fluoxetine 4 patients dropped out because of adverse events, 5 correct? 6 A Correct. 7 Q And what types of adverse 8 events did they suffer that caused them to 9 discontinue? 10 A Similar events, far less 11 frequency. 12 Q So similar CNS events 13 such as anxiety, excessive nervousness, excessive 14 sedation? 15 A Pretty much consistent 16 with what's seen in the package insert for the 17 more frequent adverse events with the drug were 18 what we observed. 19 Q Did any of these ten to 20 fifteen percent of patients suffer from agitation 21 that caused them to drop out? 22 A No, other than their 23 baseline condition, which was a prerequisite for 24 getting into the study. 56 1 Q Well, did any of these 2 patients make up the group that you were talking 3 about earlier in which the drug didn't do anything 4 for them and they became more agitated as part of 5 their disease process? 6 A I would imagine there 7 were one or two that did not respond to therapy 8 that had a general worsening of symptoms, and that 9 general worsening might have included those 10 features. 11 Q Did any of the patients 12 in this study suffer from treatment emergent 13 psychosis? 14 A No. 15 Q Treatment emergent 16 suicidal ideation? 17 A No. 18 Q Worsening of suicidal 19 ideation? 20 A I'm sure that some of the 21 patients had an increase in suicidal ideation 22 rating scale scores. Again, that's part of the 23 failure to respond to treatment. That would be 24 expected. 57 1 Q Do you know if those 2 patients were fluoxetine or imipramine patients? 3 A There was statistically 4 more patients on imipramine that had that 5 experience than fluoxetine, but the numbers are 6 comparable. 7 Q Statistically more in 8 this study itself? 9 A Yes. 10 Q Has this study ever been 11 published? 12 A Yes. 13 Q Where? 14 A I believe it was the 15 Journal of Clinical Psychopharmacology, but I 16 don't remember the exact citation. We published 17 quite a few articles, and sometimes lost track of 18 which of the scientific publications which one 19 went into. 20 Q When was this study 21 completed as far as all the patients having been 22 completed in the study and the data being 23 analyzed? 24 A Study completion was 58 1 probably in the latter part of '91 into early '92, 2 and I would imagine that the data analysis 3 occurred in the three to six months thereafter. 4 Q Were you an author on the 5 article? 6 A Yes. 7 Q Did you review any of the 8 data that was collected in the study after you 9 came to Lilly? 10 A I reviewed it all. 11 Q Where were the 12 statistical analyses on the data done? 13 A At Lilly. 14 Q Is that something that 15 you agreed to prior to coming to Lilly? 16 A Yes. 17 Q Was -- strike that. 18 How was the grant paid on 19 that study, was it paid all in one lump sum at the 20 beginning or was it paid periodically? 21 A Periodically. 22 Q Was the final payment of 23 that grant conditioned on the production or the 24 writing of a manuscript for publication? 59 1 A I don't believe so, but I 2 don't recall. 3 MS. ZETTLER: Let's take 4 a break. 5 (SHORT BREAK TAKEN.) 6 Q (BY MS. ZETTLER) Doctor, 7 what were the results of the agitated depressed 8 patient study? 9 A We observed comparable 10 efficacy between fluoxetine and imipramine. There 11 were significant safety advantages in favor of 12 fluoxetine in that there were statistically 13 significantly more discontinuations due to adverse 14 events amongst those patients randomized to 15 imipramine. 16 Q Earlier you said that one 17 of the scales that was used in the study was 18 created by us. Who do you mean by us or -- 19 A Research staff at 20 St. Paul Ramsey Medical Center. 21 Q Besides yourself, who 22 worked on creating that scale? 23 A I shared it with Doctor 24 Ronald Landbloom, one of the co-investigators, for 60 1 any opinions that he might have. 2 Q Okay, could you say 3 that -- Doctor who, I'm sorry? 4 A Doctor Ronald Landbloom. 5 Q The one that took over 6 for you when you left? 7 A That's correct. 8 Q Anybody else? 9 A Not in the construction 10 of the scale, no. Once it was constructed, it was 11 shared with Doctor Greist as would-be protocol, 12 since he was the co-investigator. 13 Q So I make sure I 14 understand this, you created the scale and then 15 consulted with Doctor Landbloom for any advice 16 that he might have had regarding the scale? 17 A Yes. 18 Q Did he make any changes 19 or make any suggestions for changes on the scale? 20 A He did not. 21 Q How about Doctor Greist, 22 did he make any suggestions? 23 A No. 24 Q Did you validate that 61 1 scale before you used it in the clinical trial? 2 A No, this was an effort to 3 validate the scale. There are no existing scales 4 for agitation, but we chose the criteria, as I 5 mentioned earlier, from research diagnostic 6 criteria for agitated depression. So, in a sense, 7 it was not a new scale, it was taking established 8 and validated criteria and putting them into a 9 scaled format. 10 Q What about validating it 11 against the HAMA? 12 A Well, you're really 13 tapping into two totally different phenomena, so 14 there would be no reason to do that. We did look 15 at an individual item from the Hamilton, which is 16 the item that taps into psychomotor agitation, and 17 there was a strong correlation between the 18 psychomotor agitation item in the Hamilton and the 19 scale that we were employing. 20 Q When you say you were 21 looking for agitation, were you looking for 22 psychomotor agitation? 23 A Well, agitation as 24 described or defined by the research diagnostic 62 1 criteria, which could be motor, could be more 2 psychic, or it could be a combination of both. 3 Q Who were the author -- 4 who was the author or the authors of the research 5 diagnostic criteria for agitation that you based 6 your scale on? 7 A RDC criteria came out in 8 the '70s, and I think it was -- Renee Spitzer was 9 one of the primary authors, although it was a 10 multi-authored set of criteria. Spitzer and 11 Endicott, Jean Endicott were the primary authors, 12 however. 13 Q Were you able to validate 14 the scale that you created through the study? 15 A Yes. 16 Q Did you publish any 17 articles on the validation of the scale itself? 18 A It was included in the 19 master publication of the safety efficacy data. 20 Q When was that publication 21 published? 22 A Early '94. 23 Q Is Lilly using the 24 agitation scale that you developed in any other 63 1 clinical trials that's running on fluoxetine at 2 this time? 3 A None that I'm aware of. 4 Q Has Lilly used the scale 5 in any of its clinical trials on fluoxetine that 6 you're aware of? 7 A No. 8 Q Why not? 9 A I don't think that there 10 have been subsequent studies done specifically on 11 an agitated subpopulation in depression where that 12 type of scale would be useful. 13 Q How about the Adult 14 Suicidal Ideation Questionnaire, to your knowledge 15 has Lilly used it in any other clinical trials on 16 fluoxetine besides your study? 17 A I suspect, but I don't 18 know for sure. 19 Q Since coming to work at 20 Lilly, have you acted as clinical monitor on any 21 clinical trials conducted at Lilly? 22 A No. 23 Q Earlier you said you met 24 Doctor Thompson a couple of times prior to joining 64 1 Lilly when you acted as a consultant to Lilly, 2 correct? 3 A That's correct. 4 Q When was the first 5 occasion that you acted as a consultant to Lilly? 6 A I don't remember the 7 exact date. 8 Q Can you give me a year? 9 A 1990. 10 Q And how is it that you 11 became a consultant to Lilly? 12 A I was invited. 13 Q Who were you invited by? 14 A Doctor Thompson. 15 Q Did you know Doctor 16 Thompson before he contacted you about being a 17 consultant? 18 A No, I did not. 19 Q How did Doctor Thompson 20 first contact you about being a consultant? 21 A By telephone. 22 Q Had you known of Doctor 23 Thompson prior to his contacting you? 24 A No. 65 1 Q When Doctor Thompson 2 contacted you by phone, what did he say? 3 A He indicated that Lilly 4 would like to have me visit Indianapolis and serve 5 as a consultant, and to review part of the 6 clinical trial database. 7 Q Do you recall what time 8 of year in 1990 that he contacted you? 9 A No, I don't. 10 Q Did he tell you how he 11 got your name? 12 A No, he didn't mention it. 13 Q Did he tell you what 14 areas he wanted you to consult on? 15 A Wanted me to consult on 16 the hypothesis that antidepressants were or were 17 not associated with treatment emergent 18 suicidality. 19 Q Antidepressants in 20 general or fluoxetine? 21 A Antidepressant in 22 general, including fluoxetine. 23 Q As a consultant for Eli 24 Lilly, did he ever give you access to their 66 1 database on Aventyl? 2 A No. 3 Q Did you ask them why not? 4 A No. 5 Q Aventyl is another 6 antidepressant manufactured by Eli Lilly, correct? 7 A Correct. The database I 8 did review included many tricyclic antidepressants 9 quite similar in structure to Aventyl. 10 Q Those were the tricyclics 11 used in the fluoxetine clinical trials as 12 comparators, correct? 13 A Correct. 14 Q To your knowledge, has 15 Lilly ever run a clinical trial comparing Aventyl 16 and fluoxetine? 17 A I suspect that they have 18 looked at fluoxetine versus nortriptyline, which 19 is the generic name for Aventyl. 20 Q Do you know that for a 21 fact, or is that a guess? 22 A No, it's supposition. 23 Q Have you ever seen any 24 data in the fluoxetine clinical trial database 67 1 regarding studies comparing nortriptyline and 2 fluoxetine? 3 A I don't recall seeing 4 anything. 5 Q When Doctor Thompson 6 called you in 1990 to become a consultant on the 7 issue of suicidality and the use of 8 antidepressants, did you accept his offer? 9 A Yes. 10 Q Did he offer to 11 compensate you for your time? 12 A Yes. 13 Q How much did he offer you 14 to compensate you for your time as a consultant? 15 A An honorarium of seven 16 hundred and fifty dollars. 17 Q Seven hundred and fifty 18 dollars a day? 19 A I was only there for one 20 day. 21 Q Did he compensate you for 22 your time in reviewing the clinical trial 23 database? 24 A The first time that I was 68 1 at Eli Lilly and Company there was no significant 2 amount of data to review, so there was none. 3 Q Did this honorarium of 4 seven hundred and fifty dollars for one day 5 include expenses? 6 A Expenses were in addition 7 to that. 8 Q Travel expenses and 9 lodging? 10 A Travel only; I did not 11 stay overnight. 12 Q Before joining Lilly, you 13 consulted with Lilly on more than one occasion, 14 correct? 15 A Two. 16 Q This first occasion that 17 we've been talking about, you said there was a 18 meeting at Lilly? 19 A Yes. 20 Q Who was at that meeting? 21 A Doctor Thompson, Doctor 22 Masica, Doctor Beasley, Doctor Heiligenstein, 23 Doctor Wheadon, Doctor George Winokur, Doctor Jan 24 Fawcett. 69 1 Q Anybody else? 2 A There were other Lilly 3 people, I was not introduced to them. 4 Q Any other outside 5 consultants? 6 A Not that I recall. 7 Q How about Doctor David 8 Dunner? 9 A I don't remember if David 10 was there that day. 11 Q How about Doctor -- I'm 12 sorry, I didn't mean to cut you off. Did you 13 finish? 14 A I don't remember that 15 David was there that day. 16 Q Okay. 17 A It's possible. 18 Q How about Doctor Nemeroff? 19 A No. 20 Q How about Stuart 21 Montgomery? 22 A No. 23 Q How long did that first 24 meeting last? 70 1 A The entire working hours 2 of the day. 3 Q 9:00 to 5:00? 4 A 9:00 to 4:00, 9:00 to 5 4:30, something like that. 6 Q What was discussed during 7 that meeting? 8 A The fluoxetine database 9 and the concept, as I mentioned earlier, of is 10 there or is there not an association between 11 antidepressant pharmacotherapy and treatment 12 emergent suicidality. 13 Q Anything else? 14 A No. 15 Q What was said about the 16 fluoxetine database? 17 A I'm not sure what you 18 mean, what was said about it, I'm sorry. 19 Q You said you discussed 20 the fluoxetine database. 21 A Uh-huh. 22 Q And if could give me an 23 idea of what was discussed about the database, 24 other than the fact that it was there. 71 1 A The results of the 2 studies that went into the NDA, safety, efficacy, 3 protocol design, methodology. 4 Q When you say the results 5 of the studies that went into the NDA were 6 discussed, you mean the initial comparative 7 studies of fluoxetine, placebo, and imipramine, 8 among others? 9 A Correct. 10 Q During that initial 11 meeting, were you told that the German government 12 had raised the same issue as early as 1984? 13 A I don't recall. 14 Q Did Doctor -- did you 15 ever discuss with Doctor Winokur his review in the 16 mid '80s of the clinical data -- trial database as 17 it existed then to look at the issue of 18 suicidality and the use of fluoxetine? 19 A I didn't discuss it with 20 Doctor Winokur. 21 Q Prior to your becoming an 22 employee at Eli Lilly, were you ever made aware 23 that the German government had raised the issue 24 back in 1984? 72 1 A What issue? 2 Q The issue of whether the 3 use of fluoxetine was causally related to an 4 increase of suicide and suicidal ideation? 5 A I wasn't aware that it 6 had been -- it was a concern of the BGA. 7 Q Prior to your coming to 8 Lilly? 9 A Correct. 10 Q When did you first learn 11 that the German government had raised that issue? 12 A Probably during the first 13 six months of being at Lilly, I'm not sure of the 14 exact month. 15 Q Do you know before or 16 after the Drug Advisory Committee meeting in 17 September of 1991? 18 A Before. 19 Q You attended that 20 Advisory Committee meeting, did you not? 21 A Yes. 22 Q In fact, you spoke at 23 that meeting, correct? 24 A Correct. 73 1 Q To your knowledge, did 2 anybody at Lilly inform the members of that 3 committee, either during the meeting or otherwise, 4 that the German government had raised the issue of 5 whether or not fluoxetine causes or exacerbates 6 suicidal ideation and suicidal acts? 7 A I don't recall. It's 8 very possible. Our focus, as I mentioned earlier, 9 was on the US NDA studies. 10 Q Did you personally ever 11 inform anybody at the FDA at any time the German 12 government had raised the issue back in 1984? 13 A Did I personally, no. 14 Q What were you told at 15 that initial consultants meeting that you attended 16 about the results of the studies that went into 17 the NDA? 18 A Told the basic efficacy 19 data, the instruments that were used, the type of 20 study design in those particular studies, the 21 safety results. That's essentially it. 22 Q Did they talk about 23 numbers of suicide attempts? 24 A Yes. 74 1 Q What numbers did they 2 tell you occurred during those initial trials? 3 A I don't remember the 4 specific numbers. Suicides did occur in each and 5 every arm of the studies, including placebo. 6 Q In the initial NDA 7 studies? 8 A Yes. 9 Q Who told you that? 10 A I don't remember the 11 presenter of that specific information. 12 Q Did you ever see -- 13 A Probably Doctor Beasley. 14 Q Did you ever see any 15 written data to that effect? 16 A Yes. 17 Q When? 18 A When Doctor Beasley 19 published his draft manuscript. But I also 20 reviewed the raw data. 21 Q Okay, I'm talking about 22 at the time of the initial consultants meeting. 23 A Uh-huh. 24 Q Okay. Did you see a 75 1 draft of Doctor Beasley's article at that meeting? 2 A No, subsequent. 3 Q Did you see written 4 documentation of suicides occurring in the initial 5 NDA studies in the placebo arm? 6 A I'm not sure what you 7 mean by written documentation. You have to be 8 more specific. 9 Q Did you see 1639's, for 10 instance? 11 A No. 12 Q Did you see clinical 13 report forms? 14 A No. 15 Q Did you see final reports 16 that talked about suicides or suicide attempts 17 that occurred during the placebo arm of those 18 studies? 19 A I saw data that was 20 presented with graphic portrayal that very well 21 could have come from final reports. 22 Q But you don't know for 23 sure? 24 A I do not know for sure. 76 1 Q When you say you saw 2 graphic data, you mean slides? 3 A Slides and hard copy. 4 Q Were you given a hard 5 copy? 6 A I had a hard copy 7 available that day. 8 Q Did you take it with you? 9 A I did not. 10 Q Why not? 11 A It was the property of 12 Eli Lilly and Company. 13 Q So they wouldn't let you 14 take it with you? 15 A I didn't ask. 16 Q Did anybody take it with 17 them? 18 A That I don't know. 19 Q When you became an 20 employee of Eli Lilly, did you again review a hard 21 copy of that data that you were initially 22 presented with at the first meeting? 23 A Yes. 24 Q Did you keep a copy of 77 1 that in your file -- 2 A Yes. 3 Q -- as an employee? 4 A Yes. 5 Q Did you turn a copy of 6 that over to the legal department at Lilly? 7 A I would assume so. 8 Q What comparators -- 9 strike that. 10 In the initial studies 11 that went into the NDA, I believe it was your 12 testimony earlier that suicide attempts occurred 13 in the comparator arms of the studies, correct? 14 A All arms, including 15 placebo. 16 Q When you say all arms, 17 are you lumping comparators as a whole, or are you 18 talking about each individual comparator that was 19 used as an arm in each study? 20 A Comparators as a whole. 21 Q So you don't know if it 22 was suicides that occurred in, say, imipramine as 23 opposed to amitriptyline? 24 A No. Some of those sample 78 1 sizes may be too small to derive any meaningful 2 conclusions. 3 Q Are you aware of Protocol 4 27? 5 A No. 6 Q Are you aware of the 7 multicenter studies that Lilly conducted on 8 fluoxetine comparing fluoxetine, placebo, and 9 imipramine? 10 A Yes, as a group. 11 Q Are you aware of whether 12 or not suicide attempts occurred in the imipramine 13 arm of that multicenter study? 14 A No. 15 Q And it's your testimony 16 that as a result of the first meeting, you were 17 not asked to review any data from the clinical 18 trial database, correct? 19 A I reviewed the data that 20 was presented us, as I mentioned, in both slide 21 and hard copy format. 22 Q During the meeting? 23 A Yes. 24 Q You did not take any data 79 1 with you to review? 2 A That's correct. 3 Q What was the purpose of 4 the meeting, the first meeting? 5 A To explore or to 6 determine if there was a way to test or study the 7 hypothetical question; that is, is there or is 8 there not any association between antidepressant 9 pharmacotherapy in emergent suicidality. 10 Q Were you, Doctor Winokur 11 and Doctor Fawcett asked to give the Lilly group 12 suggestions on how that issue might be studied? 13 A Yes. 14 Q What kinds of suggestions 15 did you give them? 16 A Key suggestion was to 17 look at the database and to specifically assess 18 individuals who at baseline had either no or 19 minimal suicidal ideation, and then in turn at any 20 time during the clinical trial experienced the 21 emergence of significant suicidal ideation and/or 22 attempt. 23 Q And that's looking at the 24 HAMD-3? 80 1 A Item 3, zero to one to 2 three to four change at any time doing the course 3 of the trial. 4 Q Why did you determine 5 that it should be a change from zero to one to 6 three or four? 7 A We were trying to test 8 the hypothesis that have been forwarded by Teicher 9 and others, and that is that individuals who did 10 not have suicide evidence at baseline might during 11 the course of exposure to drug therapy experience 12 the phenomenon. 13 Q Did Lilly ask you to 14 limit the -- strike that. 15 Did Lilly ask you to 16 limit your recommendations to studying the 17 emergence of the Teicher phenomena? 18 A No. 19 Q Did the group make 20 suggestions on studying the issue of suicidal 21 ideation in general? 22 A It was discussed, 23 particularly the complexity and the challenge of 24 it, and the recognition of it as part of the 81 1 disease process. 2 Q Did this group make any 3 recommendations that Lilly actually conduct 4 clinical trials to look into the issue of whether 5 fluoxetine either initiated or exacerbated 6 suicidal ideation in general? 7 A That group did not. 8 Q What about violent 9 aggressive behavior, did the group discuss violent 10 aggressive behavior related to fluoxetine or 11 antidepressant use in general? 12 A No, I don't believe so. 13 Q As a consultant, were you 14 ever asked to review the database, make 15 recommendations or discuss the occurrence of 16 violent aggressive behavior and the use of 17 fluoxetine? 18 A No. 19 Q Were you given any 20 assignments, so to speak, after leaving the 21 meeting? 22 A No. 23 Q Were plans made at that 24 meeting to convene the group for another meeting? 82 1 A No. 2 Q Did you, Doctor Winokur 3 and Doctor Fawcett discuss your experience with 4 treating the depressed population with fluoxetine 5 that you had in your private practices? 6 A The depressed population 7 in general, we discussed a variety of modes of 8 therapy, including nondrug therapy. 9 Q Did Doctor Fawcett at 10 that meeting discuss the use of concomitant 11 sedatives in patients who were suffering from 12 anxiety on top of depression? 13 A I don't recall. 14 Q Are you aware that Doctor 15 Fawcett has published an article on the treatment 16 of suicidal patients who were suffering from 17 anxiety in which he recommends that when you give 18 an antidepressant that causes jitteriness, such as 19 fluoxetine, that a physician may want to prescribe 20 a concomitant benzodiazepine or other sedative 21 type of drug? 22 A Not precisely as you 23 stated, no. I think that his intention was a 24 little different. 83 1 Q What was his intention? 2 A I think what Doctor 3 Fawcett is saying is that patients that are being 4 treated for depression, may have, as part of that 5 syndrome, features of anxiety. Those features, if 6 they crescendo during the course of treatment, 7 might benefit by coadministration of an 8 anxiolytic. 9 Q Have you read that 10 article? 11 A I've read a number of 12 articles of Doctor Fawcett's, who of course has 13 written a significant number of publications in 14 the area of suicidality and affective disorders. 15 Q Okay, Doctor, I'm 16 interested in the article where he discusses the 17 coadministration of fluoxetine and benzodiazepines 18 in patients suffering from anxiety and 19 depression. Okay, have you read that particular 20 article? 21 A I would suspect he's 22 written more than one article on the issue in 23 general, so I'm not sure which one you're 24 referring to. It would be unlikely for him to 84 1 have only have written that once in his career. 2 Q So you're saying there's 3 other articles out there where he may have made 4 that recommendation? 5 A I'd say it's very 6 possible that he has discussed anxiety as part of 7 the depressive presentation. 8 MS. ZETTLER: Let me show 9 you what has been marked as Exhibit 1. 10 (TOLLEFSON EXHIBIT NO. 1 MARKED FOR 11 IDENTIFICATION.) 12 Q (BY MS. ZETTLER) Have 13 you seen this article before, Doctor? 14 A Yes. 15 Q It's an article written 16 by Doctor Fawcett, correct? 17 A He is listed as the first 18 author and only author. 19 Q Okay. And it's published 20 in the November 1990 issue of the Journal of 21 Clinical Psychiatry, correct? 22 A Correct. 23 Q At the bottom left-hand 24 corner of the first page it talks about this paper 85 1 having been presented at the 143rd Annual Meeting 2 of the American Psychiatric Association in New 3 York on May 12th through 17th of 1990, correct? 4 A Yes. 5 Q The title of the article 6 is "Targeting Treatment in Patients with Mixed 7 Symptoms of Anxiety and Depression," correct? 8 A That is correct. 9 Q If you look at the second 10 page of the article, it talks about treatment? 11 A Yes. 12 Q At the bottom of the page 13 it says combination therapy, correct? 14 A The last paragraph. 15 Q It says under that: 16 Suicide prevention is the first consideration in 17 the treatment of major depression with anxiety. 18 Severe anxiety symptoms must be addressed 19 immediately. Few of the antidepressants, 20 especially newer medications, have an early effect 21 on anxiety which may explain the low compliance 22 rate common to this whole therapeutic class. 23 Correct? 24 A I'd like a minute to read 86 1 it, if that's all right. 2 Q Sure, take your time. 3 A Okay, I see the sentence 4 that you're referring to. 5 Q Do you see it? 6 A Yes. 7 Q Okay. And he goes on to 8 say: Aggressive treatment with a benzodiazepine 9 anxiolytic is indicated for immediate relief of 10 anxiety in patients with major depression if they 11 manifest risk factors for suicide, if the anxiety 12 is severe, or if the antidepressant selected 13 causes jitteriness. Correct? 14 A That's what Doctor 15 Fawcett has said here. 16 Q Okay. And after 17 jitteriness, he has, paren, fluoxetine, close 18 paren, correct? 19 A As an example, that's 20 correct. 21 Q Or is without sedative 22 properties, paren, desipramine, imipramine and 23 bupropion, correct? 24 A Correct. 87 1 Q Did you ever discuss 2 Doctor Fawcett's opinion that fluoxetine causes 3 jitteriness? 4 A No, I don't think so. 5 Q At this meeting, did 6 Doctor Fawcett raise his belief that -- at least 7 in patients suffering from depression and anxiety, 8 that if you prescribe fluoxetine, you should 9 prescribe a concomitant benzodiazepine because of 10 the possible jitteriness associated with the 11 fluoxetine? 12 A I don't believe he would 13 say that. I don't think Doctor Fawcett in this 14 article is advocating prophylactic treatment of 15 jitteriness. 16 Q Well, he says right here 17 aggressive treatment with a benzodiazepine is 18 indicated for immediate relief of anxiety in 19 patients with major depression if they manifest 20 risk factors for suicide, one, correct? 21 A Uh-huh. 22 Q Two, if the anxiety is 23 severe? 24 A Uh-huh. 88 1 Q Three, if the 2 antidepressant selected causes jitteriness -- 3 A Uh-huh. 4 Q -- as in fluoxetine, 5 correct? 6 A That's correct. In other 7 words, he would treat patients who manifest the 8 event, not treat all patients prophylactically. 9 Q Well, he doesn't say that 10 there, does he, Doctor. He says -- 11 A That's the way I read it. 12 Q Well, he says here -- 13 A He says if. 14 Q Yes, if they manifest 15 risk factors for suicide, comma -- 16 A Uh-huh. 17 Q -- if the anxiety is 18 severe, comma -- 19 A Uh-huh. 20 Q -- or if the 21 antidepressant selected causes jitteriness -- 22 A Right. 23 Q -- fluoxetine. 24 A Right, if the patient 89 1 experiences jitteriness while on fluoxetine, in 2 Doctor Fawcett's article he would suggest 3 considering combination therapy. He does not say 4 for all patients receiving fluoxetine. 5 Q So you take that to mean 6 that Doctor Fawcett believes in some patients the 7 drug is going to cause jitteriness? 8 A Perhaps, yes, consistent 9 with the package insert. 10 Q Where does the package 11 insert say that it causes jitteriness? 12 A Jitteriness or related 13 event terms are listed in the package insert under 14 adverse events experienced during clinical trials. 15 Q What related terms to 16 jitteriness are listed in the package insert? 17 A I'd have to read the 18 package insert and identify them, I suppose, one 19 by one for you. 20 Q Agitation? 21 A It might or might not. 22 Q Nervousness? 23 A Might or might not. 24 Q Restlessness? 90 1 A Probably. 2 Q Anxiety? 3 A Possibly. 4 Q Tremors? 5 A Probably not. 6 Q Irritability? 7 A Possibly. All those are 8 vague, nonspecific. 9 Q That's how they're listed 10 in the package insert, Doctor. 11 A So they could or they 12 could not involve jitteriness. You have to ask 13 Doctor Fawcett what his specific definition of 14 jitteriness was, as he meant it in this article, 15 but that's how I would interpret it. 16 Q Has he ever discussed his 17 definition of jitteriness with you? 18 A I've not asked. 19 Q That's not the question. 20 Has he ever discussed -- 21 A No. 22 Q When was the next time 23 you were asked to be a consultant for Lilly? 24 A Three to six months after 91 1 the first visit. 2 Q Before or after you 3 became an employee? 4 A Before. 5 Q And I believe you said 6 you were again contacted by Doctor Thompson? 7 A I believe so. 8 Q How did he contact you 9 the second time? 10 A By phone. 11 Q When you were contacted 12 the second time to become a consultant, had you 13 already made your proposal to conduct the agitated 14 depressed patient clinical trial? 15 A I don't recall. 16 Q When Doctor Thompson 17 contacted you the second time, did he tell you 18 what he wanted you to consult on at Lilly? 19 A He wanted me to review 20 the data analyses that resulted from our first 21 visit when we recommended the zero to one, three 22 to four Item 3 assessment of the database. 23 Q Did you agree to do that? 24 A Yes. 92 1 Q Were you provided any 2 information to review? 3 A There was some background 4 information regarding the analyses. 5 Q Will you -- strike that. 6 Was another meeting held 7 with you and the other consultants? 8 A Yes. 9 Q Were you provided this 10 background information before the meeting? 11 A Yes. 12 Q How long before? 13 A I don't recall. 14 Q A day before, a couple of 15 weeks before? 16 A One to four weeks before. 17 Q Was that information -- 18 what form was that in? 19 A Written. 20 Q Do you recall how large a 21 document or documents it was? 22 A No. 23 Q Was it more than one 24 document? 93 1 A No. 2 Q Besides background on the 3 analysis that was done, was there any other 4 information that was provided to you before the 5 meeting? 6 A No. 7 Q Did they provide you with 8 the results of the analyses? 9 A Yes. 10 Q That was included in the 11 background information? 12 A (WITNESS MOVES HEAD UP 13 AND DOWN.) 14 Q You have to say yes or 15 no. 16 A Yes. 17 Q Did Doctor Thompson 18 discuss how you would be compensated for your time 19 as a consultant the second time? 20 A Yes. 21 Q Okay, and how were you 22 compensated? 23 A In a daily honorarium for 24 my time and travel at Lilly and for document 94 1 review. 2 Q What was the daily 3 honorarium? 4 A Seven hundred and fifty, 5 I believe. 6 Q How were you compensated 7 for your document review? 8 A I provided Doctor 9 Thompson with an estimate of the amount of time 10 that I took away from my clinical practice to 11 spend looking over the documents. 12 Q And how much time was 13 that? 14 A I don't recall, but it 15 was in the area of probably four to eight hours, 16 something like that. 17 Q And were you paid hourly 18 for that time? 19 A The four to eight hours 20 was for the second day, so I believe I was paid 21 for two days. 22 Q Two days' honorarium you 23 mean? 24 A Yes. 95 1 Q So seven hundred and 2 fifty dollars a day for two days? 3 A Times two approximately, 4 yes. 5 Q And you were also paid 6 your expenses, travel? 7 A Travel expenses, right. 8 Q How many days did you 9 spend at Lilly the second time? 10 A One or two. 11 Q Were you compensated for 12 the time it took -- strike that. 13 Were you given 14 information while you were at Lilly to review 15 also? 16 A I don't remember anything 17 in addition to the information provided prior to 18 the meeting. That material was the subject of our 19 discussion. 20 Q Who was in attendance at 21 the second meeting at Lilly? 22 A The same individuals that 23 I've mentioned to you before, and I believe there 24 was one additional consultant, Doctor Martin 96 1 Keller. 2 Q So it was Doctor Winokur, 3 Doctor Fawcett -- 4 A I don't recall if Doctor 5 Fawcett was at the second meeting or not. 6 Q Okay. And then Doctor 7 Keller? 8 A Martin Keller. 9 Q Where is Doctor Martin 10 Keller located? 11 A I believe that he's at 12 Brown University. 13 Q How about Ivan Miller, 14 was he at that meeting? 15 A I don't recall if he was. 16 Q Was Doctor Dunner there? 17 A I don't believe. 18 Q Did you ever consult with 19 Lilly on the issue of rechallenge protocol? 20 A No. 21 Q Did you ever review a 22 rechallenge protocol, as a consultant? 23 A As a consultant, no. 24 Q When were you -- strike 97 1 that. 2 When had you decided to 3 take employment at Lilly? 4 A April of '91. 5 Q Had you already decided 6 to move to Lilly when the second meeting, 7 consultant meeting occurred? 8 A I don't remember. 9 Q What was discussed at the 10 second meeting? 11 A The data that had been 12 provided us, which were the results of the 13 analysis that we had recommended to Lilly. 14 Q Was the issue of a 15 rechallenge study on patients who had become 16 suicidal on fluoxetine discussed at either of 17 these meetings? 18 A I don't recall that 19 discussion. 20 Q Any time before you 21 actually started at Eli Lilly, did you consult 22 with Lilly on a rechallenge protocol? 23 A No. 24 MS. ZETTLER: Let's go 98 1 off the record for a second. 2 (OFF-THE-RECORD DISCUSSION HELD.) 3 4 5 (TOLLEFSON EXHIBIT NO. 2 MARKED FOR 6 IDENTIFICATION.) 7 Q (BY MS. ZETTLER) Doctor, 8 after your attorney takes a look at this, I want 9 you to take a look at Exhibit No. 2, please. 10 MR. FREEMAN: Is this the 11 complete document? If it is, I don't have a 12 signature page. 13 MS. ZETTLER: Why don't 14 you let him go copy that while he's reading the 15 rest of it, because it will save some time that 16 way. 17 MR. FREEMAN: Do you want 18 to let him see this while I'm getting the last 19 page copied? You can ask him questions about the 20 first part. 21 MS. ZETTLER: Yes. 22 MR. FREEMAN: There's 23 Page 5. 24 Q (BY MS. ZETTLER) Okay, 99 1 have you had a chance to review Exhibit 2, Doctor? 2 A I've reviewed it. 3 Q Doctor, Exhibit 2 is a 4 letter written by you to Robert Zerbe at Lilly 5 dated May 3rd, 1991, correct? 6 A Correct. 7 Q And the subject matter of 8 that letter is rechallenge protocol and the 9 feasibility, correct? 10 A That's one of the topics 11 covered in the letter. 12 Q Other topics are 13 including scales to rate suicidality in ongoing or 14 to be conducted Phase 4 clinical trials and Phase 15 3 clinical trials on fluoxetine? 16 A Possible uses of 17 different instruments prospectively, and then 18 summary of a pending publication, or actually a 19 data paper that had been written at Ramsey. 20 Q At this time, May 3rd, 21 1991, or somewhere around that period, had there 22 been a group of consultants gathered by Lilly to 23 consult with Lilly on the issue of methods for 24 studying the issue of suicidal ideation and the 100 1 use of fluoxetine further? 2 A I believe so. 3 Q Were you one of those 4 consultants? 5 A Yes. 6 Q Were you considered an 7 outside consultant at this point or an in-house 8 consultant? 9 A By whom? 10 Q By Lilly. 11 A I don't know, you would 12 have to ask them. 13 Q How about you, what would 14 you consider yourself? 15 A Somewhere in between, 16 probably still an outside person, although this is 17 essentially within thirty days of my beginning at 18 Lilly. 19 Q You say at the beginning 20 of the letter it was a pleasure to see you on May 21 2nd; do you see that? 22 A Yes. 23 Q And was that a meeting of 24 the consultants on the rechallenge and other 101 1 methods of studying suicidality and use of 2 fluoxetine? 3 A I don't remember. I had 4 made a couple of other visits to Lilly as part of 5 being recruited, and it would have been one or the 6 other. I don't recall which was May 2nd, though. 7 Q Do you recall attending 8 meetings at Lilly prior to June 1st, I believe you 9 said, 1991, when you actually started your 10 employment with Lilly on the rechallenge issue? 11 A Judging by the date on 12 this letter, yes. 13 Q Okay. And those 14 meetings -- that meeting or those meetings that 15 you attended on the issue of studying suicidality, 16 apart from reviewing the database, were those 17 meetings separate and apart from the two 18 consultant meetings that we have already talked 19 about? 20 A I do not know 21 specifically, but based on the date of the letter, 22 I would assume that it was part of the second 23 consultants meeting where that was brought up. 24 Q So now your recollection 102 1 is that was discussed at the second consultants 2 meeting? 3 A I suspect that was the 4 case. 5 Q Does that refresh your 6 recollection as to who else besides Doctor 7 Winokur, possibly Doctor Fawcett and Doctor 8 Keller, were at that meeting? 9 A I met Doctor Miller on a 10 couple of occasions. It is possible that he was 11 at that consultants meeting. 12 Q Doctor Miller was 13 retained by Lilly to develop a revision of his 14 modified scale for suicidal ideation to use in a 15 potential rechallenge protocol, correct? 16 A I don't know whether or 17 not Doctor Miller proposed this to Lilly or Lilly 18 asked him to, but he -- the result was, yes, he 19 was going to modify his scale. 20 Q Okay. Are you aware that 21 Doctor Miller, working in conjunction with Lilly, 22 did in fact modify the scale? 23 A I know that he made 24 efforts to modify the scale, I'm not aware of the 103 1 final results of his modification. 2 Q Were you aware that 3 Doctor Miller has in fact completed a study to 4 attempt to validate that scale? 5 A Yes. 6 Q And that scale -- or that 7 study was sponsored by Lilly, correct? 8 A I believe so. 9 Q Do you know if Doctor 10 Miller was able to validate his scale? 11 A I honestly have not seen 12 the final results personally to know if he 13 successfully validated the scale or not. 14 Q Did you do any work at 15 all on that study? 16 A No. 17 Q Do you know if the data 18 from Doctor Miller's validation study has been 19 analyzed yet? 20 A I don't personally. I 21 have not seen it and I did not work with that 22 particular study. 23 Q Who was the clinical 24 monitor at Lilly on that study? 104 1 A I believe it was Doctor 2 Beasley at the time. 3 Q At the time that you 4 wrote the letter that comprises Exhibit 2, had you 5 been given a draft protocol, rechallenge protocol 6 to review? 7 A Can I have you repeat 8 that? 9 Q Sure. At the time, May 10 3rd, 1991, when you wrote this letter, had you 11 already seen a protocol or a draft of a protocol 12 for a rechallenge study on fluoxetine and suicidal 13 ideation? 14 A I believe so. 15 Q Had you seen any other 16 draft protocols for studies to look into the issue 17 of treatment emergent suicidal ideation and the 18 use of fluoxetine? 19 A I don't remember any 20 specific protocols to address things. 21 Q Had you seen any other 22 protocols developed by Lilly at this time dealing 23 with the issue of suicide, suicidal ideation and 24 the use of fluoxetine? 105 1 A I don't recall any 2 additional specific protocols. 3 Q At any time have you seen 4 a protocol developed by Lilly to study the issue 5 of violent aggressive behavior and the use of 6 fluoxetine as its main objective? 7 A I have. 8 Q When did you see such a 9 protocol? 10 A Well, Doctor 11 Heiligenstein conducted a retrospective protocol 12 analysis, and there are a series of so-called 13 exempt from IND study studies that Lilly has been 14 involved with outside external academic experts 15 exploring a variety of populations that may be 16 predisposed to aggressive or violent behaviors; 17 mentally retarded, borderline personality 18 disorder, to name a few. 19 Q Schizoaffective disorder? 20 A It's entirely possible. 21 Q Are you aware of any 22 schizoaffective disorder studies that have been 23 conducted at Lilly? 24 A Yes. That's not a 106 1 population that typically though is felt to be 2 predisposed to irrational violence. 3 Q Doctor Heiligenstein's 4 study on aggressive behavior and the use of 5 fluoxetine was a meta analysis, was it not? 6 A It was. 7 Q It was a preexisting 8 clinical trial data? 9 A It was a retrospective 10 review of existing clinical trial databases. 11 Q Similar to Doctor 12 Beasley's meta analysis on suicide? 13 A Correct, in over five 14 thousand patients. 15 Q Doctor Heiligenstein's 16 was not over five thousand patients, Doctor 17 Beasley's -- 18 A Doctor Beasley was. 19 Well, Doctor Heiligenstein's was drawn from the US 20 NDA database. I don't remember the exact number, 21 but I would assume that it's a similar number. 22 Q Heiligenstein's study was 23 drawn from the US IND database? 24 A Yes, NDA. 107 1 Q NDA? 2 A Or US IND, yes. 3 Q Was it based on the same 4 information that was discussed in the appendix to 5 the packet given to the Advisory Committee? 6 A Yes, I believe so. 7 Q Earlier you said that 8 schizoaffective disorder patients are not 9 predisposed to violent aggressive behavior, 10 correct? 11 A Correct. 12 Q What populations are 13 predisposed to violent aggressive behavior? 14 A Substance abuse, certain 15 personality disorders that are characterized by 16 poor impulse control, individuals with organic 17 brain injury. An example could be head injury, 18 mental retardation, dementia. Or other 19 individuals under the influence of other drugs of 20 abuse, I didn't mention that. 21 MS. ZETTLER: Why don't 22 we let her change the video. 23 (SHORT BREAK TAKEN.) 24 Q (BY MS. ZETTLER) Doctor, 108 1 just before the tape change we were talking about 2 patient populations that were predisposed to 3 violent aggressive behavior, okay? 4 A Okay. 5 Q And I've gotten some of 6 them down and want to make sure I get all of those 7 down that you listed. I have substance abuse, 8 personality disorders, organic brain injury, 9 dementia, and that's as far as I got. Would you 10 list the rest of them for me again, please? 11 A Personality disorders, 12 alcoholism, substance abuse, traumatic brain 13 injury. 14 Q Any others? 15 A There may be others, 16 those are the ones that are most commonly 17 associated with predisposing factors. 18 Q Are there others that are 19 less commonly associated? 20 A Oh, occasionally patients 21 who are delirious, that would be a possibility. 22 With the remaining psychiatric disorders, though, 23 the -- in most clinical experience, the rates are 24 no different than that of the general population 109 1 for externally directed violence or aggression. 2 Q Externally as opposed to 3 internally, like suicidal? 4 A Correct. 5 Q What about mania, 6 somebody who is suffering from mania or is manic, 7 are they a risk factor for violent aggressive 8 behavior towards others? 9 A I don't think they're 10 substantially at higher risk to be -- to 11 premeditate, for example, an aggressive or violent 12 act. They may be more impulse ridden, but in the 13 sense of any type of planned act of aggression or 14 violence, no, I don't think that they're at a 15 substantially higher risk. 16 Q We established earlier 17 that you spoke at the Advisory Committee meeting 18 in 1991, correct? 19 A Correct. 20 Q And your job at that 21 meeting, I believe, was to review the data gleaned 22 from the clinical trial database? 23 A Correct. 24 Q And you spoke of both the 110 1 issues of suicidal ideation and the use of 2 fluoxetine, as well as violent aggressive behavior 3 and the use of fluoxetine, correct? 4 A Correct, amongst other 5 things. 6 Q We have a copy of the 7 transcript from the Advisory Committee meeting, 8 and you spoke briefly about Lilly having 9 investigated possible trials or studies to conduct 10 on the issue of suicidality and the use of 11 fluoxetine, do you recall that? 12 A No, I'd probably need to 13 see the document. 14 Q This is going to be hard 15 to do with the video, but if you look at this 16 paragraph down here that I've marked. 17 A Okay. 18 Q At the bottom of the 19 second page that I showed you, you talk about 20 choosing to test the hypotheses generated by the 21 DEN system reports, do you recall that? 22 A I think my recollection 23 was that we were talking about a variety of ways 24 methodologically to test the hypothesis of whether 111 1 or not suicidality is emergent during the course 2 of antidepressant therapy. 3 Q Okay, and you say that 4 Lilly had chosen to test the hypothesis based on 5 the three methods of study that you'd talked about 6 earlier, one being prospective suicide to specific 7 trials? 8 A That was one of the three 9 mentioned. 10 Q Epidemiological 11 databases? 12 A Correct. 13 Q And detailed studies 14 drawn from controlled, double-blind clinical 15 trials, correct? 16 A That was the third, 17 correct. 18 Q Then you go on to say 19 that Lilly had already generated a series of 20 protocols to better characterize patients who 21 experience an increase in the suicidal ideation 22 during pharmacotherapy or other forms of 23 treatment, do you recall that? 24 A Who might, yes, 112 1 experience suicidal ideation. 2 Q Okay. But my point is 3 that you had already generated a series of 4 protocols as of the date of the PDAC? 5 A Protocols or ideas. 6 Q Okay. And you go on to 7 say the first of these protocols has already been 8 initiated, do you recall that? 9 A Yes. 10 Q What trial was that? 11 A I probably was thinking 12 of the agitation study. 13 Q That was a study that was 14 your idea prior to -- I believe your testimony was 15 prior to you becoming a consultant for Eli Lilly, 16 correct? 17 A Correct, it was 18 initiated. 19 Q But that wasn't a 20 protocol that was generated specifically to better 21 characterize patients who experienced an increase 22 in suicidal ideation during pharmacotherapy or 23 other forms of treatment, was it? 24 A That was one of the 113 1 secondary objectives. 2 Q At the time that you 3 developed it? 4 A Yes. You might recall I 5 mentioned that the Reynold's was one of the 6 questionnaires use in the protocol. 7 Q And that protocol was 8 your idea? 9 A Yes. 10 Q What was the results of 11 your -- of the secondary objective of your 12 agitated depressed study? 13 A It was comparable -- 14 well, let me start at the beginning, I guess. 15 There were no suicidal acts or suicides completed 16 during the course of the trial. The rates of 17 ideation numerically were similar, but as I recall 18 the analysis were slightly greater with imipramine 19 than fluoxetine, but the differences did not 20 achieve any statistical significance, and that the 21 majority of the patient populations had an 22 improvement in suicidal ideation rather than no 23 change. 24 Q How about those patients 114 1 who had dropped out of the study? 2 A I'm not sure what you 3 mean. 4 Q Did you look at their 5 ASIQ scores? 6 A Sure. The analyses are 7 always done based on the last patient visit rather 8 than a completer's analysis. So if a patient were 9 to drop out, at least to have received one 10 randomized dose of medication, whether they took 11 it or not, they would be included in the safety 12 analysis with the ASIQ. 13 Q Those patients who 14 dropped out in the fluoxetine arm of the study, 15 was there a time within the study when they 16 generally dropped out, like for instance within 17 the first couple of weeks of treatment? 18 A I don't recall specific 19 numbers, but with antidepressants in general, most 20 side effects occur very early in the course of 21 treatment and then tend to wane, so that the 22 majority of dropouts due to an adverse event are 23 likely to occur in the first week or two of 24 treatment. 115 1 Q Were there any other 2 studies that -- at the time of the PDAC that had 3 been initiated on the issue of characterizing 4 patients who experience an increase in suicidal 5 ideation during pharmacotherapy? 6 A There were a number of 7 hypothesis around. I would assume from the 8 earlier correspondence that you showed me that the 9 rechallenge protocol was one of those protocols 10 that had been discussed as a possible way to try 11 to address the question, and I suspect that there 12 were others. Off the top of my head, I don't 13 remember which ones and how many, but there were a 14 number of different possible ways to try to look 15 at the question that had been discussed. 16 Q Do you recall a trial 17 specifically with the objective of characterizing 18 patients who experienced an increase in suicidal 19 ideation? 20 A I remember a discussion; 21 that was one of the possible ideas that was 22 discussed, yes. 23 Q Do you recall seeing a 24 protocol regarding that idea? 116 1 A I remember seeing 2 documents, I don't remember if it was a finalized 3 protocol. I don't believe there was ever a 4 finalized protocol, but I think there were 5 probably draft documents outlining the concept of 6 such a study and how it might be conducted. 7 Q Did you ever see a 8 protocol for Doctor Miller's study? 9 MR. FREEMAN: For what? 10 MS. ZETTLER: To validate 11 his scale. 12 A It's possible. I don't 13 recall specifically reviewing it. Again, as I 14 stated earlier, I was not the clinical research 15 physician that monitored any of those trials. 16 Q What's a study considered 17 exempt from IND? 18 A Those are studies that 19 are being conducted in some way outside of the 20 conventional package insert, either in the sense 21 of the indication or the dosage or the 22 administration of the drug. 23 Q What regulatory 24 implications is it -- does a study that's exempt 117 1 from the IND have? 2 A Well, the investigator 3 needs to receive an exemption from the Food and 4 Drug Administration for such study. 5 Q Are those studies still 6 conducted under the IND? 7 A They can. They don't 8 have to be, but they may. More often than not 9 they are. 10 Q You said earlier that 11 you're aware of some IND exempt studies being 12 conducted to look at the issue of violent 13 aggressive behavior? 14 A Again patient populations 15 predisposed to. 16 Q Can you tell me about 17 some of those studies? 18 A No. 19 Q Why not? 20 A I typically wouldn't be 21 the one reviewing the specific protocols. There 22 is an exempt from IND committee that rates the 23 scientific merit of studies and then decides 24 whether or not that Lilly would be involved in any 118 1 way providing drug and placebo for such a study. 2 Q Who is on that committee? 3 A It's chaired by Doctor 4 David Goldstein. Typically there's been a 5 statistician that sits in to do analysis of the 6 proposal from a statistical viewpoint. Ben Rampey 7 has been one of the Lilly statisticians that 8 historically has been involved in that process. 9 It's not unusual to have one marketing person on 10 the committee. Jim Lancaster in the past has sat 11 on that committee. And I'm not sure if anyone 12 else routinely would be a sitting member. 13 Q Would Doctor Goldstein 14 then act as the medical monitor on those studies? 15 A No, on an exempt from IND 16 study Lilly typically doesn't function in the 17 conventional role as a medical monitor. That's 18 the whole idea behind the exemption. 19 Q Are these double-blind, 20 controlled studies, to your knowledge? 21 A They may, they may not 22 be. That's not a prerequisite. 23 Q Okay. Are they generally 24 pilot studies? 119 1 A Yes. Most of them are 2 testing various hypotheses that an investigator 3 has brought to Lilly. 4 Q Are you aware of any such 5 studies, meaning exempt IND studies, on patients 6 suffering from substance abuse -- 7 A Yes. 8 Q -- and the use of 9 fluoxetine? 10 A Yes. 11 Q How many such studies? 12 A I don't know the exact 13 number. I'm sure that there are in excess of 14 three or four protocols. 15 Q Are you specifically 16 aware of any of these studies as far as like who 17 is conducting the study, where it's being 18 conducted? 19 A No. 20 Q Are you aware of any such 21 studies that have been completed? 22 A I couldn't give you 23 investigators, but there are -- there is published 24 data on the use of fluoxetine in individuals with 120 1 substance abuse problems. 2 Q These IND exempt studies, 3 are any of them, to your knowledge, sponsored by 4 Lilly as far as grants are being provided or 5 clinical trial materials are being provided? 6 A Typically clinical trial 7 materials are provided. On occasion there may be 8 a small financial grant as well. 9 Q How about statistical 10 analyses, typically where is that done? 11 A At the investigator site. 12 Q Personality disorders, 13 are you aware of any of these IND exempt studies 14 being done on personality disorders? 15 A Yes. 16 Q Were you specifically 17 aware of any of them? 18 A I know of at least two in 19 borderline personality disorder, which is a 20 prototype of poor impulse control. 21 Q Could you tell me about 22 those two studies? 23 A I'm only aware of one 24 where the data has been published to date, and 121 1 that was published by a group in Cleveland in the 2 Archives of Psychiatry, where they showed that 3 therapy in fluoxetine was effective in reducing 4 impulse dyscontrol. 5 Q Was that a study where 6 Lilly participated either through a grant or by 7 providing clinical trial materials? 8 A Clinical trial materials, 9 yes. 10 Q Who at the Cleveland 11 group was involved in that study? 12 A I don't remember the lead 13 investigator. 14 Q Do you remember any of 15 the investigators? 16 A No. Again, as I had 17 said, that was not part of my routine daily 18 responsibilities to be part of that committee or 19 reviewing the proposals that came into that 20 committee. 21 Q Where in Cleveland was 22 this -- you know, at what site, Cleveland Clinic 23 or Case Western? 24 A I believe it was 122 1 conducted at Case Western. It would have been one 2 of those two for sure. 3 Q How about the other 4 personality disorder study that you're aware of, 5 who conducted that? 6 A That I don't know. 7 Q Do you know if the 8 results of that study have been published? 9 A I'm not aware of the 10 results of the study having been published. 11 Q Was that study conducted 12 here or outside the United States? 13 A Here, if you mean the 14 United States. 15 Q Right. 16 A Yes, in the United 17 States. 18 Q Do you know where? 19 A No. 20 Q Was that a study that was 21 also done through Lilly, either by grant or CT 22 material being provided, or both? 23 A All of the exempt from 24 IND studies are a request from investigators for 123 1 CT materials, so by definition they would all 2 involve Lilly providing at least CT materials. 3 Q Do you know if that study 4 has been completed? 5 A Which study? 6 Q The second personality 7 disorder study. 8 A No, I believe I said I do 9 not know if the results have been published or if 10 the study is completed. And the responsibility to 11 publish the exempt from IND studies is on the part 12 of the investigator. They have no obligation to 13 Lilly and Lilly has no follow-up on whether or not 14 they publish or not. Some of them, unfortunately, 15 do not publish. Some of them do not even complete 16 their studies. 17 Q Are you aware of any 18 situation where Lilly has asked that the 19 investigator not publish the results of a study? 20 A No. 21 MR. FREEMAN: In the 22 United States? 23 THE WITNESS: I'm not 24 aware of that happening anywhere in the world. 124 1 MR. FREEMAN: When you 2 get to a convenient spot, let's think about lunch. 3 MS. ZETTLER: Didn't you 4 eat breakfast this morning, Joe? 5 MR. FREEMAN: A big one, 6 6:00. 7 MS. ZETTLER: 6:00, 8 that's your problem, you get up too early. 9 Q (BY MS. ZETTLER) Tell me 10 about the schizoaffective disorders that has been 11 done at Lilly -- through Lilly? 12 A You had asked me if 13 schizoaffective disorder has ever been studied, I 14 know for a fact that it's being studied as part of 15 a current trial with another drug. I would have a 16 fairly high degree of suspicion, though, that 17 there is, amongst the many exempt from IND studies 18 with fluoxetine, one that would have been in 19 schizoaffective or related disorders. 20 Q The schizoaffective 21 disorder study that you were just speaking of, is 22 that a Lilly study? 23 A Yes. 24 Q Is that a study being 125 1 done with fluoxetine? 2 A No. 3 Q Is it being done with 4 another serotonin reuptake inhibitor? 5 A No. 6 Q Is it being done with 7 another Lilly-marketed drug? 8 A No. 9 Q Are you aware of any 10 studies that were performed by Lilly on 11 schizoaffective disorder and fluoxetine either as 12 a treatment regimen or as a safety profile? 13 A There are studies with 14 fluoxetine and related conditions. I do not 15 specifically recall a published study on 16 schizoaffective, but it would not surprise me. 17 Q Okay, how about 18 unpublished studies done by Lilly? 19 A No, I'm not aware. 20 Q How about the rest of 21 these categories that you gave us earlier, organic 22 brain injury, dementia, alcoholism, traumatic 23 brain injury, delirium. Are you aware of any IND 24 exempt studies that are being conducted on the use 126 1 of fluoxetine in any of those treatment groups? 2 A I'm aware that exempt 3 from IND studies are going on, for example, in 4 autism, mental retardation, some patients with 5 organic brain disorders. 6 Q Are you aware of any such 7 studies that have been published? 8 A There are studies, yes, 9 that have been published, not necessarily with any 10 Lilly involvement. 11 Q Okay, I'm concerned just 12 with the Lilly involved studies. Are you aware of 13 any such studies that Lilly has been involved in 14 that have been published? 15 A Yes. 16 Q Okay, what studies? 17 A I'm not in a position to 18 be able to cite the specific authors and articles, 19 but there are studies of fluoxetine and 20 alcoholism, substance abuse, including cocaine, 21 childhood autism, childhood mental retardation, 22 borderline personality disorders, a variety of 23 neurological disorders with organic brain 24 impairment. 127 1 Q When you say fluoxetine 2 and cocaine, are those published studies? 3 A There are. 4 Q Lilly involved published 5 studies? 6 A I suspect, I can't say 7 unequivocally. 8 Q Borderline personality 9 disorder, is that different than the personality 10 disorder studies we were talking about earlier? 11 A No. 12 Q So those are the same 13 studies? 14 A Yes. 15 MS. ZETTLER: Okay, Joe, 16 let's take a break for lunch. 17 (LUNCH BREAK TAKEN.) 18 Q (BY MS. ZETTLER) Doctor, 19 have you ever talked to Doctor Greist about 20 whether or not he has been asked to be an expert 21 witness in any of the Prozac cases? 22 A No, I haven't. 23 Q Have you ever met with 24 Doctor Greist to discuss the facts of any Prozac 128 1 case? 2 A No. 3 Q Are you aware of the 4 facts of any Prozac case? 5 A I'm sorry? 6 Q Are you aware of any of 7 the facts of any particular Prozac case? 8 A No, not in any detail. 9 Q Have you yourself been 10 asked by anybody at Eli Lilly or any of their 11 lawyers to act as an expert witness in the 12 Fentress versus Shea Communications case down in 13 Louisville, Kentucky? 14 A No. 15 Q Are you aware of the 16 Joseph Wesbecker incident? 17 A I'm aware of the 18 incident. 19 Q Okay, tell me what your 20 knowledge of that incident is. 21 A Just that Mr. Wesbecker 22 was involved in an aggressive event that ended 23 with violence, including his own death, and that 24 there have been linkages or allegations that 129 1 somehow it was related to exposure to fluoxetine. 2 Q Have you spoken with 3 anybody outside of Lilly who has been asked by 4 Lilly lawyers to be an expert witness in the 5 Fentress case that is surrounding the occurrence 6 on September 1980 -- September 14th, 1989 7 involving Joseph Wesbecker? 8 A No. 9 Q Do you know Robert 10 Woolson? 11 A No. 12 Q W-O-O-L-S-O-N? 13 A No. 14 Q Do you know Anthony 15 Rothchild? 16 A By name. 17 Q Okay, who is Doctor 18 Rothchild? 19 A If it's the same Doctor 20 Rothchild that I'm assuming, Doctor Rothchild is a 21 clinical psychiatrist and researcher who has -- my 22 only familiarity with him is he has published a 23 hypothetical linkage between an increase in 24 suicidality and neurological event term akathesia. 130 1 Q Is that the Rothchild and 2 Lock article? 3 A Yes. 4 Q Isn't that a rechallenge 5 article, reporting cases of three rechallenges of 6 patients who became suicidal while using 7 fluoxetine? 8 A I don't recall the 9 details of the article, but I believe there was a 10 rechallenge. 11 Q How about Robert 12 Granacher? 13 A No. 14 Q John Schwab? 15 A I believe there's a 16 Doctor Schwab who is -- or was chairman of 17 psychiatry at the University of Kentucky who has a 18 history of having worked in the consult liaison 19 area of psychiatry. 20 Q What is the consult 21 liaison area of psychiatry? 22 A That's a psychiatrist 23 that would provide a consulting opinion to 24 nonpsychiatrists who might have a patient that 131 1 they're providing care for, and in the context of 2 that care might believe that there could be a 3 psychiatric problem and bring in that individual 4 to provide consultation. 5 Q How about a Mr. James or 6 Dean James Fox? 7 A No. 8 Q Ed Winstil? 9 A No. 10 Q Ed Mercer? 11 A No. 12 Q Are you aware, Doctor 13 Tollefson, that you have been listed as an expert 14 witness in the Fentress versus Shea Communication 15 case? 16 A No. 17 Q Tell me about Doctor 18 Greist's background, as best you know it. 19 A I believe that he was 20 born in Indiana. He went to medical school at 21 Indiana University. I don't know really too much 22 about his post-medical training. Ever since I've 23 known Doctor Greist, he's been at the University 24 of Wisconsin, Madison, as faculty and professor of 132 1 psychiatry. 2 Q Do you know where he went 3 to medical school? 4 A No. 5 Q Does he specialize in any 6 particular area of psychiatry that you're aware 7 of? 8 A Well, I think his prime 9 interests are in anxiety, obsessive compulsive 10 disorder, and mood. 11 Q What types of articles 12 has -- when I say types, what subject matters has 13 Doctor Greist published on, that you know of? 14 A Those same topics, 15 depression, obsessive-compulsive disorder, social 16 phobia, suicidality and depression, bipolar 17 disease, manic depression. 18 Q How about violent 19 aggressive behavior as it relates to any 20 particular mental illness? 21 A I'm not aware of 22 anything. 23 Q You testified earlier 24 that it was Doctor Thompson who first contacted 133 1 you about taking a job at Lilly, correct? 2 A Yes. 3 Q And he had also contacted 4 you about acting as a consultant on at least two 5 occasions, correct? 6 A Yes. 7 Q When within the time 8 period that he asked you to be a consultant did he 9 offer you the position at Lilly? 10 A He did not offer me a 11 position at Lilly. 12 Q Okay, when did he 13 approach you about possibly taking employment with 14 Lilly? 15 A It would have been very 16 early in -- either at the very end of '90 or the 17 very beginning of the year in '91, I would think, 18 I don't remember the exact time. 19 Q How about in the context 20 of when he asked you to consult on those two 21 occasions, was it before he had asked you to 22 consult or after he had asked you to consult the 23 first time? 24 A After the -- 134 1 MR. FREEMAN: That he 2 talked to him about a job? 3 MS. ZETTLER: Right. 4 A After the first. 5 Q Okay. Before the second 6 time? 7 A Yes. 8 Q Had you accepted the job 9 or a job with Lilly before the second consultation 10 period? 11 MR. FREEMAN: Are you 12 relating that day to May 2nd or thereabouts? 13 MS. ZETTLER: The best 14 that he can remember. 15 A I think I had made a 16 decision to join Lilly at that time. 17 Q How did Doctor Thompson 18 approach you about whether or not you would be 19 interested in working with Lilly or for Lilly? 20 A He called me first and 21 indicated that Lilly had a very robust central 22 nervous system pipeline of products and that they 23 were looking for someone to come in and provide 24 some additional expertise in clinical 135 1 psychopharmacology to serve as essentially a 2 division leader in orchestrating those projects 3 for the new compounds in development. 4 Q Had you been looking for 5 other employment besides St. Paul Ramsey when 6 Doctor Thompson approached you? 7 A No. 8 Q Had other drug companies 9 approached you about going to work for them? 10 A Not with a formal 11 proposal. I had had some very informal 12 invitations on the part of the Upjohn Company to 13 look at a similar role with them. 14 Q Who else at Lilly did you 15 talk to after Doctor Thompson initially approached 16 you about working for Lilly? 17 A Regarding a potential 18 employment relationship? 19 Q Right. 20 A Doctor Mel Perelman. 21 Q Anybody else? 22 A I had interviews with a 23 number of individuals, but as far as actual -- I 24 suppose the only other person, I did speak with a 136 1 human resources representative, which is standard 2 procedure there, her name was -- I can't spell 3 it -- Joanna Engliophatic. 4 Q Anybody else from the 5 medical division? 6 A No, other than again, in 7 the process of some interviews, I did interview 8 with some individuals who were members of the 9 medical department, but it wasn't formally around 10 an employment relationship negotiation, it was 11 more of an interview. 12 Q Do you recall the names 13 of any of those other people? 14 A From the medical 15 division? 16 Q Right. 17 A Robert Zerbe, Allen 18 Weinstein. And I believe on one of the visits I 19 spoke with David Wheadon and John Heiligenstein. 20 Q How about Doctor Beasley? 21 A I think I said Doctor 22 Beasley, didn't I? 23 Q If you did, I missed it. 24 A Yes, Doctor Beasley and 137 1 Doctor -- I'm sorry, if I didn't, anyway -- oh, 2 Beasley, no. I said Wheadon and Heiligenstein. I 3 don't remember talking with Charles as an 4 interview at all. 5 Q What position were they 6 offering you? 7 A Executive Director 8 Clinical Investigation, Regulatory Affairs. 9 Q Is that the position that 10 you ended up accepting? 11 A Yes. 12 Q As Executive Director of 13 Clinical Investigation and Regulatory Affairs, 14 would you have come in as a superior to any of 15 those people that you just listed for us? 16 A Yes. 17 Q Who? 18 A Beasley we've talked 19 about, Wheadon, Heiligenstein. 20 Q How about Weinstein? 21 A No. 22 Q How about Doctor Masica? 23 A He was not in my -- when 24 I came to Lilly, he was no longer in my division, 138 1 so I didn't have any direct reporting relationship 2 with him. 3 Q What is Doctor Masica 4 doing at Lilly now? 5 A He works in the United 6 States Medical Organization. 7 MR. SMITH: Beg your 8 pardon? 9 THE WITNESS: He works in 10 the United States Medical Organization. 11 Q (BY MS. ZETTLER) And 12 where are you working now? 13 A Same place. 14 Q What is he doing within 15 the US Medical Organization? 16 A I don't know the specific 17 details. That group is really involved with 18 product support for agents that are currently 19 licensed, approved in the marketplace within the 20 United States, and interact very closely with 21 their marketing colleagues on a geographic or 22 regional basis. 23 Q How much contact do you 24 have on a daily or weekly basis with Doctor 139 1 Masica? 2 A None, other than if I see 3 him occasionally in the hallway to say hello. 4 Q Have you heard anything 5 about Doctor Masica's health? 6 A Yes. 7 Q Can you tell me what 8 you've heard? 9 A That he's had some 10 cardiovascular problems. 11 Q Is he currently on 12 medical leave? 13 A No. 14 Q When you first became 15 Executive Director of Clinical Investigation and 16 Regulatory Affairs, what were your 17 responsibilities related to fluoxetine? 18 A When I initially came, it 19 was primarily to be involved with the 20 postmarketing clinical investigation strategies, 21 the expert opinion leader programs or interactions 22 that we might set up for continuing medical 23 education, and any subsequent regulatory work that 24 we might do, for example, new indications for the 140 1 compound. Primarily it was a role to provide 2 mentorship and supervision to the clinical 3 research physicians, though, that worked within 4 the division. 5 Q When you say primarily 6 postmarketing clinical investigation strategies, 7 what do you mean? 8 A Clinical studies with 9 fluoxetine. 10 Q Deciding what clinical 11 studies to perform? 12 A I wish I was that 13 powerful I could make the decision. I had a voice 14 in it. I could make recommendations about things 15 I thought we should do or shouldn't do, based on 16 my scientific opinion. 17 Q What are opinion leaders? 18 A Academic physicians who 19 publish regularly and are recognized as thought 20 leaders in a particular discipline. 21 Q Is opinion leader a Lilly 22 term? 23 A No, I don't believe so, I 24 think it's a common term. 141 1 Q Is it a term that you 2 heard used prior to your employment at Lilly? 3 A Yes. 4 Q Is it a term that you had 5 heard used outside the drug industry context? 6 A Yes. 7 Q When did you first hear 8 that term? 9 A I have no idea. 10 Q Before you got your MD? 11 A Before I got my -- no, 12 sometime after. 13 Q After you started working 14 in clinical trials, clinical investigations? 15 A No, no, I would say 16 probably sometime during my residency or shortly 17 after completion of my residency. It would be a 18 guess. 19 Q Have you stayed in that 20 position as Executive Director of Clinical 21 Investigation and Regulatory Affairs the entire 22 time you've worked at Lilly? 23 A Yes. 24 Q Have your 142 1 responsibilities changed at all throughout the 2 time you've worked at Lilly? 3 A Broadened. 4 Q How have they broadened? 5 A They include all products 6 that are in development for central nervous system 7 indications, gastrointestinal or genitourinary. 8 Q Was there a time in your 9 employment at Lilly where you spent the majority 10 of your time working with fluoxetine? 11 A Perhaps the first four to 12 six months. 13 Q During that period of 14 time, how much of your time did you spend on 15 fluoxetine-related projects? 16 A Seventy, seventy-five. 17 Q Were you involved with 18 preparing for the 1991 Advisory Committee meeting? 19 A Yes. 20 Q What were your 21 responsibilities on that project? 22 A Primarily coordinate the 23 presentation itself, and to be the presenter. I 24 also did the background -- some of the background 143 1 literature search in the field. 2 Q Why is it they chose you 3 to be the presenter, if you know? 4 A I don't know. 5 Q Doctor, have you ever 6 conducted a prospective study in the use of 7 antidepressants and suicidal behavior? 8 A Yes. 9 Q When did you do that? 10 A Prior to joining Lilly, 11 probably circa 1990-ish. 12 Q Was that before or after 13 you became a consultant for Lilly? 14 A Before. 15 Q Can you tell us about 16 that study, Doctor? 17 A The primary study was the 18 one we referred to earlier this morning, the study 19 of fluoxetine versus imipramine in major 20 depression, baseline agitation. 21 Q Didn't you do a study 22 where you looked at a bunch of different classes 23 of antidepressants, including serotonin reuptake 24 inhibitors? 144 1 A Yes, that was a 2 retrospective analysis. 3 Q It was not a prospective 4 study? 5 A It was not. 6 Q Retrospective analysis of 7 what? 8 A Of a series of clinical 9 trials that had been conducted at St. Paul Ramsey 10 Medical Center with a variety of different 11 investigational antidepressant agents and 12 currently marketed or approved agents and placebo. 13 Q Okay. Were those studies 14 done in conjunction with one another? 15 A No, they were not. 16 Q Were the protocols 17 identical on those studies? 18 A Very similar. 19 Q What was the serotonin 20 reuptake inhibitor that you reviewed on that 21 study? 22 A Fluoxetine was one of the 23 serotonergically based compounds. 24 Q Any others? 145 1 A It gets into things that 2 are not approved. 3 MR. FREEMAN: Don't say 4 them. 5 Q Lilly compounds that are 6 not approved? 7 A No, no. 8 MS. ZETTLER: Let's go 9 off the record for a second. 10 (OFF-THE-RECORD DISCUSSION HELD.) 11 12 MS. ZETTLER: Doctor, 13 after your attorneys review this, I'd like you to 14 take a look at Exhibit 3, please. 15 (TOLLEFSON EXHIBIT NO. 3 MARKED FOR 16 IDENTIFICATION.) 17 Q (BY MS. ZETTLER) Have 18 you had a chance to review Exhibit 3? 19 A Yes. 20 Q Exhibit 3 is something 21 that was produced as part of your file, Doctor, 22 and it's a two-page exhibit which appears to be an 23 abstract of a study authored -- manuscript 24 authored by you, Doctor Landbloom and some others, 146 1 correct? 2 A Correct. 3 Q The title of the study, 4 at least as it appears on this exhibit, is 5 "Emergent Suicide During Treatment of Placebo 6 Controlled Double-Blind Study Employing Several 7 Distinct Antidepressants," correct? 8 A Correct. 9 Q Is this a study that you 10 were just telling me about where you -- 11 A Yes. 12 Q -- did a retrospective 13 analysis of studies that had been performed at 14 St. Ramsey? 15 A St. Paul Ramsey. 16 Q St. Paul Ramsey. 17 A Correct. 18 Q This doesn't really talk 19 about a retrospective analysis, does it, Doctor? 20 In fact, if you read this abstract alone, it 21 appears to be a double-blind controlled study that 22 was set up specifically for the purpose of 23 studying emergent suicide during treatment, does 24 it not? 147 1 A No. But again, you're 2 looking at the abstract, not the article. Any 3 scientific article has an extensive methodology 4 section that details the methodological 5 procedures. I'm not even sure this is the final 6 copy of an abstract that went into the 7 publication. It could well have been an earlier 8 draft. I don't know where it fits in the 9 chronology. 10 Q Why were you faxing this 11 to Eli Lilly in 1991? 12 A I don't recall that it 13 was faxed to Eli Lilly, but it may have been to 14 share -- 15 Q Well, if you look at the 16 first page, at the top, there's a fax line, and it 17 appears to be -- it says Ramsey with a date of 18 8-12-91, 10:15 AM. 19 A Uh-huh. 20 Q And a fax to number with 21 a 317 area code. 22 A Uh-huh. 23 Q Does that indicate to you 24 that you faxed that to Eli Lilly? 148 1 A I don't know if I did, 2 but it was faxed apparently from St. Paul. 3 Q It doesn't talk about 4 using fluoxetine in this study here, does it? 5 A Not by specific 6 reference, it says including serotonin selective 7 agents. 8 Q Let's look at the first 9 page. Is dioxepin a selective serotonin reuptake 10 inhibitor? 11 A It is a serotonin 12 reuptake inhibitor. 13 Q Is it a specific? 14 A No. 15 Q Is imipramine a specific 16 serotonin reuptake inhibitor? 17 A Once again it's a 18 serotonin reuptake inhibitor. 19 Q It's not selective, is 20 it? 21 A It depends on your 22 definition of selective. 23 Q Is it as selective as 24 fluoxetine, Doctor? 149 1 A No. 2 Q How about Glaxo's drug, 3 was it Fluparoxan? 4 A No. 5 Q What type of drug is 6 that? 7 A It's an noradrenergic 8 receptor blocker. 9 Q How about nefazodone? 10 A That is a serotonin 11 selective agent. 12 Q It is? 13 A Yes. 14 Q As selective as -- 15 A Yes, perhaps more so. 16 Q Is it more selective as a 17 serotonin reuptake inhibitor than fluoxetine? 18 A More selective as a 19 serotonergic agent than fluoxetine. 20 Q Is it marketed? 21 A It is approved by the 22 FDA. I'm not sure that the company is marketing 23 it at present. 24 Q How about desipramine? 150 1 A Mixed profile. 2 Q Not as selective as -- 3 A Correct. 4 Q -- fluoxetine, correct? 5 A Correct. 6 Q Doctor, you have to let 7 me finish my questions. Remember, we talked 8 earlier about Kathy not being able to take us both 9 down at the same time, okay? 10 A I apologize. 11 Q Okay. You said in the 12 beginning on the front page that financial support 13 for the protocols was received from Boots, McNeil, 14 Glaxo, Bristol-Meyers-Squibb and the Upjohn 15 Company, correct? 16 A Correct. 17 Q And these were studies 18 that had been conducted at St. Paul Ramsey? 19 A Correct. 20 Q And I believe your 21 testimony earlier is that at least you had not 22 conducted a clinical trial on fluoxetine prior to 23 the agitation study, correct? 24 A Funded by Lilly. 151 1 Q I believe my question was 2 whether you had performed any clinical trials on 3 fluoxetine. So now your testimony is that you had 4 conducted some clinical trials on fluoxetine that 5 were not funded by Lilly? 6 A Again, it depends on how 7 you want to define a clinical trial. We have done 8 work with fluoxetine observing clinical features 9 of patient response. At the time it was not part 10 of an organized clinical trial, nor was it funded 11 by Eli Lilly and Company. The drug was a 12 postmarketed drug, it was a common drug used in 13 our clinic. Part of our use of the drug includes 14 controlled observation. 15 Q Did you include a 16 double-blind controlled study of fluoxetine versus 17 placebo or another comparator as part of the 18 article that is abstracted in Exhibit 3? 19 A No. 20 Q Did you include any 21 fluoxetine patients as part of this article? 22 A Yes. 23 Q And these are the 24 patients that you observed in the clinical 152 1 practice who were on fluoxetine? 2 A Not in total, of course. 3 Q Besides clinical patients 4 that you observed on fluoxetine, did you include 5 any other patients in this article, on fluoxetine? 6 A I would have to have you 7 restate that. 8 Q Okay. I'm trying to get 9 an idea of what types of patients you included, 10 the fluoxetine patients that you've included in 11 this study that is reflected in Exhibit 3. 12 A Uh-huh. They were 13 nonresponders to the Glaxo sponsored trial with 14 fluparoxan, a very selective noradrenergic agent. 15 Nonresponders to an noradrenergic agent would 16 predict that a serotonergic agent might be the 17 next best alternative therapy. 18 Q So they were crossover 19 patients? 20 A No, crossed over from 21 either placebo or fluparoxan to fluoxetine, but 22 not the other direction. 23 Q Okay. Were the 24 patients -- strike that. 153 1 The nonresponders portion 2 of the study, was that double-blind and 3 controlled? 4 A No. 5 Q It was open label? 6 A Yes. 7 Q Any other fluoxetine 8 patients that you used? 9 A No, not in this draft. 10 Q In the study? 11 A Eventually, because we 12 had completed the fluoxetine agitation study that 13 we have talked about in detail, we did include 14 that database because of its similarity in 15 methodology to these previous clinical trials. 16 Q Did all of the other 17 studies reflected in this draft use the HAMD-AN3? 18 A I'm sorry? 19 Q I'm sorry, HAMD and HAMA 20 scales? 21 A All of them used the 22 HAMD. I doubt that they all used the HAMA. 23 Q Did all of these studies 24 or any of these studies use the Adult Suicidal 154 1 Ideation Questionnaire besides the agitation study 2 on fluoxetine? 3 A No. 4 Q Did any of these studies 5 use your agitation scale? 6 A No, but they were not 7 studies of specific agitated cohorts. 8 Q Was this paper ever 9 published, Doctor? 10 A It's under review for 11 publication. 12 Q By which publication? 13 A The journal is entitled 14 Depression. 15 Q When was it submitted? 16 A Approximately nine months 17 ago, six to nine months. 18 Q Has any form of this 19 article ever been published previous to submitting 20 it to the depression magazine? 21 A I don't believe so. 22 Q Has any form of this 23 article been presented at a seminar or compendium 24 or anything, meeting, professional meeting? 155 1 A It's very possible that 2 it was presented as part of a depression 3 discussion. I mean, it would be part of data that 4 I have collected, my own personal clinical 5 experience. 6 Q Who are the coauthors on 7 the article that was submitted to Depression? 8 A Ronald Landbloom is a 9 psychiatrist and staff member at Ramsey Clinic in 10 St. Paul Ramsey Medical Center, as is Barry 11 Rittberg. Melissa Pederson was clinical research 12 assistant. Sherrie Tollefson was the coordinator 13 of clinical research. Michael Luxenberg was a 14 statistician. Michael Godes is a psychiatric 15 nurse who worked in the clinical research section 16 part time. 17 Q Are those people all 18 still coauthors on the article as it's been 19 submitted to the Depression magazine? 20 A I believe so. 21 Q Who is Sherrie Tollefson? 22 A My spouse. 23 Q Do you discuss in the 24 manuscript as it's been submitted to the 156 1 Depression magazine the history of this trial? 2 A Trials, plural, yes. 3 Q And the fact that the 4 first fluoxetine patients who were included were 5 the patients who crossed over in the Glaxo study? 6 A As I said, this paper has 7 gone through several iterations. Those patients, 8 because they were not controlled, were not part of 9 a blinded trial, were not included in the final 10 draft of the paper. Instead, because the 11 agitation cohort had been completed, that was a 12 much better scientific database to use in the meta 13 analysis, so that was included. 14 Q Did you discuss the 15 differences between the agitation trial and these 16 other trials on the other drugs in the 17 publication? 18 A Yes, sure. Although it 19 has not been published yet, so I wouldn't refer to 20 it as a publication. 21 Q The manuscript. 22 A Yes. We haven't received 23 any response back from the editorial review board. 24 Q Is that unusual to submit 157 1 a manuscript and not hear from the editorial 2 review board for six to nine months? 3 A No, unfortunately, it's 4 not. 5 Q Have you submitted that 6 paper to any other journals or publications? 7 A No. 8 Q Doctor Beasley testified 9 at his deposition that he thought that you had 10 written an article on activation, sedation and the 11 lack of association with suicidality in fluoxetine 12 patients. He didn't think -- he didn't know if 13 that was the exact title, but he thought that that 14 was the subject matter. Do you recall such a 15 paper? 16 A Not specifically by that 17 title. We did do an analysis of adverse events 18 and whether or not there was a relationship 19 between change in baseline to during the course of 20 study of suicidal ideation relative to the 21 presence or emergence of any other adverse events. 22 Q Has that article been 23 published? 24 A Yes, but I don't remember 158 1 the journal. 2 Q When was it published? 3 A Probably around the first 4 of the year. 5 Q First of '94? 6 A Yes. 7 Q Do you remember the 8 title? 9 A Something to the effect 10 is there an association between treatment emergent 11 adverse events and suicidality during 12 pharmacotherapy of depression. 13 Q Is that a meta analysis 14 type study also? 15 A It was a retrospective 16 analysis of the fluoxetine database. 17 Q Did that database differ 18 at all from Doctor Beasley's database that he 19 looked at? 20 A It's possible it might 21 have included one or two additional newer studies 22 that had completed in the interim. 23 Q But the bulk of the -- 24 A Was the same. 159 1 Q -- data was the same as 2 Doctor Beasley's? 3 A That's correct. 4 Q When you were putting 5 together your agitation study, did you have the 6 benefit of Doctor Beasley's activation analysis 7 that he had done on the clinical trial database 8 prior to 1990? 9 A No. 10 Q Have you since seen that 11 analysis? 12 A Since my employment 13 relationship at Lilly I have seen the reprint. 14 Q When you say the reprint, 15 what do you mean? 16 A A copy of the article. 17 Q To the best of your 18 recollection, what does that article conclude? 19 A That patients may 20 experience a state of activation while receiving 21 fluoxetine as an adverse event, that it typically 22 occurs early in the course of events, that it 23 tends to wane over time, and that there's a modest 24 dose response relationship. 160 1 Q What is activation? 2 A Activation is really a 3 categorization or a cluster term for a number of 4 different events that I would consider to be 5 arousal related. 6 Q Do you recall what events 7 Doctor Beasley used in his cluster? 8 A Not specifically. 9 Q He used agitation, didn't 10 he? 11 A I'm sure that was one. 12 Q He used insomnia? 13 A Again, I said I don't 14 know the specific events that Charles used in 15 that. 16 Q How about nervousness? 17 A It's possible. 18 Q Do you recall what 19 percentage of patients Doctor Beasley found on his 20 retrospective analysis of the clinical trial 21 database would experience activation, as he 22 defined it? 23 A No, not precisely. 24 Q Somewhere around 161 1 seventeen percent, does that refresh your 2 recollection? 3 A That's in the ball park 4 at least. 5 Q Are you familiar with the 6 review of the clinical trial database on 7 fluoxetine to look at the incidence of violent 8 aggressive behavior with the use of fluoxetine? 9 A I'm familiar with it in 10 the sense of knowing it was conducted, yes. 11 Q Were you involved in that 12 project at all? 13 A Not directly. 14 Q Okay. Tell us what your 15 involvement was. 16 A I believe I reviewed a 17 copy of the manuscript that was authored from the 18 paper. I don't recall having any other 19 involvement in that project at all. 20 Q When you say the copy of 21 the manuscript that was authored, are you talking 22 about the manuscript that is attached to the 23 materials that was provided to the Advisory 24 Committee? 162 1 A Correct. 2 Q Did you make any changes 3 on that manuscript? 4 A I don't recall. If they 5 were, they would have been grammatic. I don't 6 remember any major scientific adjustments, if 7 that's what you're inferring. 8 Q Were you involved in 9 establishing the aggression cluster that was used 10 to analyze the -- or search the clinical trial 11 database? 12 A No. 13 Q Who was involved in doing 14 that? 15 A My understanding was that 16 John Heiligenstein. 17 Q Did you agree with the 18 terms that he chose to run in the clinical trial 19 database? 20 A It seemed representative. 21 Q Would you have added any 22 terms? 23 A None that I'm aware of. 24 Q Are you familiar with the 163 1 changes that the event term dictionaries used by 2 Lilly have -- what changes have occurred over the 3 years to the event terms dictionary that Lilly has 4 used with fluoxetine? 5 A No. 6 Q Have you ever had an 7 occasion to assign an event term to an adverse 8 event? 9 A Relative to which drug? 10 Q Fluoxetine. 11 A I don't believe so. 12 MS. ZETTLER: Let's take 13 a break. 14 (SHORT BREAK TAKEN.) 15 * * * * * 16 EXAMINATION 17 18 BY MR. SMITH: 19 Q Doctor Tollefson, my name 20 is Paul Smith. I represent several plaintiffs in 21 these cases involving Prozac and its relationship 22 to suicide and violent aggressive behavior. Do 23 you understand that? 24 A Yes. 164 1 Q My instruction to you 2 would be the same as that of Ms. Zettler, that is 3 if I ask you any question that you don't 4 understand, would you please let me know, sir? 5 A Sure. 6 Q I'll be glad to rephrase 7 it, if I can. 8 A All right, thank you. 9 Q Have you reviewed any 10 documents for your preparation here -- in 11 preparation for your deposition here today, Doctor 12 Tollefson? 13 A No. 14 Q Have you spoken with 15 anybody -- any employees of Lilly other than the 16 Lilly legal department concerning the substance of 17 any of their deposition testimony? 18 A No. 19 Q For instance, Doctor 20 Leigh Thompson gave his deposition last week. 21 Have you discussed Doctor Thompson's deposition? 22 A No. 23 Q Doctor Charles Beasley 24 gave his deposition several weeks ago, maybe 165 1 months ago. Have you discussed with Doctor 2 Beasley the subject of his deposition? 3 A No, I have not. 4 Q Have not discussed any of 5 the individuals who have given depositions here 6 what occurred in those depositions? 7 A I have not. 8 Q And you've not reviewed 9 any materials? 10 A No, sir. 11 Q Have you discussed the 12 deposition with Lilly lawyers, either in-house or 13 out-house? 14 MR. FREEMAN: Particular- 15 ly the out-house ones. 16 A Yes. 17 Q And with whom have you 18 met? 19 A Colleagues at the table, 20 and James Burns, Lilly counsel. 21 Q Mr. Burns is in-house 22 with Lilly? 23 A Correct. 24 Q Who from these other 166 1 firms have you met with? Have you met with 2 Mr. Larry Myers? 3 A No. 4 Q Mr. Joe Freeman? 5 A Not regarding -- well, 6 yes, I take that back, yesterday. 7 Q You've not given any 8 Prozac depositions before? 9 A No, sir. 10 Q In connection with this 11 analysis of studies that had been done concerning 12 antidepressant treatment and emergent suicidal 13 ideation, do I understand it that this publication 14 that's going to be submitted to Depression 15 magazine or publication is -- includes more than 16 the two hundred and twenty-six patients mentioned 17 in this abstract? What exhibit number is that? 18 A Exhibit 3 that you're 19 referring to? 20 Q Yes, sir. 21 A Yes. This particular 22 manuscript has been submitted for review, and it 23 does include one additional study, fluoxetine 24 versus imipramine in agitated depressed patients. 167 1 Q And how many Prozac 2 patients were on -- were in that study or 3 completed that study? I think you told us 4 earlier, did you not? 5 A There were approximately 6 forty randomized and thirty or so who completed. 7 Q Thirty fluoxetine 8 patients -- 9 A Correct. 10 Q -- completed the study? 11 And you reflected the results of that trial in the 12 publication that's under review at this time? 13 A Along with the other 14 trials mentioned in the publication, yes. 15 Q Now, as I understand it, 16 none of the two hundred and twenty-six patients 17 mentioned in Exhibit 3 were given Prozac during 18 the trials that were reviewed? 19 A That's correct. 20 Q So this paper that was 21 published earlier that's abstracted as Exhibit 3 22 wouldn't have anything to do with whether or not 23 Prozac resulted in suicidal ideation in depressed 24 individuals, would it? 168 1 A This paper was not 2 published previously. 3 Q All right. What happened 4 to it, it just -- 5 A I said it is currently 6 under review by the journal Depression. I thought 7 I understood your question to infer that this 8 abstract, Exhibit 3, had already been published. 9 Q All right. But the 10 article apparently had been written -- 11 A A draft of the article. 12 Q All right. And somebody 13 had to abstract that article in order to submit it 14 in an abstract form, correct? 15 A This was not submitted 16 for publication. 17 Q I understand that. 18 A All right. 19 Q But somebody sat down -- 20 you apparently, as the principal author, sat down 21 and wrote a paper describing this series of 22 studies that would be longer than one page, 23 correct? 24 A That is correct. 169 1 Q And where is that paper 2 contained? 3 A That paper is part of the 4 paper submitted to the journal Depression. 5 Q All right. But I thought 6 that that paper included updates on other patients 7 which did include patients taking Prozac? 8 A That paper that was 9 submitted to the journal Depression includes all 10 the data here plus one additional study, the 11 fluoxetine, imipramine study in agitated 12 depression. 13 Q So you had a paper that 14 was written in May of 1991, correct? 15 A A draft of it. 16 Q And I get that by virtue 17 of the -- 18 A A draft of it, yes. 19 Q And somebody abstracted 20 it to come up with this one-page piece of paper 21 that we've got in Exhibit 3, right? 22 A Exhibit 3, the abstract 23 is my abstract in one of the early drafts of the 24 manuscript. 170 1 Q How long was that early 2 draft, how many pages was it? 3 A I don't recall. I would 4 suspect it was fifteen, give or take five. 5 Q If I wanted to see that 6 draft to get a better idea of the objective, 7 results, methods, and conclusions that are 8 reflected here in the abstract, what would I have 9 to do? 10 A I don't know. I don't 11 have -- I don't know if I have a copy of it or 12 not. This was a very early draft. The process of 13 writing a scientific publication involves 14 iteration after iteration. I certainly -- you 15 know, I think it's routine practice to save every 16 iteration. 17 Q Did you submit that early 18 draft to any publications? 19 A I don't believe so. 20 Q This publication 21 Depression, what is that? 22 A It's a medical journal. 23 Q All right, and how long 24 has that medical journal been in existence? 171 1 A Approximately two, 2 possibly three years. 3 Q And where is it 4 published? 5 A I'm not sure. 6 Q By whom is it published? 7 A It would be speculating, 8 it may be L. Silver, but I don't remember the 9 medical publisher. 10 Q Well, how did you know to 11 submit it to them for publication, sir? 12 A Because I'm aware of the 13 journal. I don't worry about the publisher, I 14 worry about the journal. 15 Q Well, I mean, where did 16 you submit -- where did you mail this article to, 17 to get it to Depression? 18 A The editor of the 19 journal. 20 Q And where is he located, 21 or she? 22 A He is at Emory 23 University. 24 Q And who is the editor? 172 1 A Doctor Charles Nemeroff. 2 Q I'm sorry? 3 A Doctor Charles Nemeroff. 4 Q Can you spell that for 5 me? 6 A N-E-M-E-R-O-F-F. 7 Q And who is Doctor Charles 8 Nemeroff, other than being the editor? 9 A Professor and Chairman of 10 the Department of Psychiatry, Emory University. 11 He's also one of the leading scientific 12 investigators in the country. 13 Q And that's at -- in 14 Atlanta, Emory University in Atlanta? 15 A That's correct, Emory 16 University. 17 Q And did you speak with 18 him concerning the fact that you were going to 19 submit this to him? 20 A No, I did not. 21 Q Did you know that Doctor 22 Nemeroff was a Lilly consultant? 23 A I know that Doctor 24 Nemeroff has consulted scientifically for Lilly, 173 1 being one of the leading scientists in the 2 country, it would make sense to me that he would 3 be used by Lilly. 4 Q Did you know he 5 participated in the 1991 PDAC meeting? 6 A Yes. 7 Q In what capacity? 8 A Providing scientific 9 consultation on previous literature on 10 psychopharmacology, treatment of depression and 11 suicidality. 12 Q Was he a member of the 13 committee or an advisor to the committee? 14 A I don't believe he was 15 either. 16 Q Well, did Lilly hire him 17 to make a presentation to the committee on behalf 18 of antidepressant treatment? 19 A Yes. 20 Q And he presented data 21 that Lilly felt supported their position before 22 the PDAC? 23 A No, he presented his 24 scientific review of literature. 174 1 Q Have you called him and 2 asked him whether or not he intends to publish 3 this article that you've written? 4 A I've only written to 5 inquire about the status since it has become, as 6 we mentioned earlier, six to nine months, so I was 7 curious about the status of the paper. 8 Q And what kind of response 9 have you received? 10 A He indicated that he had 11 one of the reviewers that had had the paper for 12 over four months and had not yet reviewed it. 13 Q So you don't know whether 14 they're going to publish it, whether they're going 15 to suggest any changes or additions or deletions? 16 A That's correct. 17 Q Did he give you any idea 18 when you could expect to know something on that? 19 A No, he did not. 20 Q Go back to Exhibit 2, 21 which was the letter that you wrote to Doctor 22 Zerbe at Lilly after apparently you had agreed to 23 accept the job with Lilly, but while you were 24 still with St. Paul Ramsey Medical Center. Do you 175 1 have that in front of you? 2 A Yes, sir. 3 Q You indicated that one of 4 your first suggestions would be working 5 proactively with the Food and Drug Administration 6 for their consideration of adding a suicide 7 assessment tool to all new Phase 3 clinical drug 8 investigations, correct? 9 A That was one suggestion, 10 that is correct. 11 Q Has that been done, as 12 far as you know? 13 A As far as I know, it has 14 not. 15 Q It has not? 16 A Has not. 17 Q Do you know why it has 18 not? 19 A No. I think you would 20 probably have to ask the Food and Drug 21 Administration. 22 Q Well, do you know if 23 Lilly has asked the Food and Drug Administration 24 why that hadn't been done since you've been with 176 1 Lilly? 2 A Can I have you repeat 3 that, please? 4 MR. SMITH: Would you 5 read it back, please? 6 REPORTER: (READING) Do 7 you know if Lilly has asked the Food and Drug 8 Administration why that hadn't been done since you 9 have been with Lilly? 10 A Well, there are many 11 people at Lilly, so I can't tell you that no one 12 conceivably has asked. I'm not aware of anyone 13 who has asked. Of course, this was a suggestion 14 prior to an Advisory Committee that unanimously 15 found no substance in the association or the 16 relationship. 17 Q I get the impression that 18 some of the -- some of the ideas that were thrown 19 about as ways to investigate whether or not this 20 phenomena existed were discarded after the PDAC 21 meeting and its results. Is that accurate, it was 22 not felt necessary to pursue the investigation 23 specifically any further after that? 24 A That is not true. 177 1 Q All right. There 2 certainly is not the ongoing effort that is 3 suggested by your May 3rd, 1991 letter, is there, 4 by Lilly? 5 A No, there is. We 6 continue to assess suicidality as part of all 7 clinical trials in depression. 8 Q Well, Lilly is not using 9 a suicide assessment tool in all of its clinical 10 trials on Prozac right now, are they? 11 A They are. 12 Q Beg your pardon? 13 A They are. 14 Q Specific to -- 15 specifically designed to assess suicidality in all 16 trials? 17 A Specifically designed to 18 assess the signs and symptoms of depression, which 19 would include suicidality. 20 Q All right. Well, is 21 there anything specifically designed to address 22 suicidality, like the adult suicidal inventory 23 scale that you yourself have used in trials before 24 you joined Lilly? 178 1 A There's several scales 2 that have been purported to assess suicidality; 3 all of them have been standardized, though, in the 4 Hamilton, particularly with Item 3, the suicide 5 item, as a gold standard as a means to validate 6 them. None of them are in uniform use in clinical 7 investigation. 8 Q What I want to know is 9 what Lilly is doing now in their ongoing Prozac 10 clinical trials to examine whether or not there's 11 an emergence of suicidal ideation in patients 12 being treated with Prozac for any indication, all 13 right, sir? 14 A (WITNESS MOVES HEAD UP 15 AND DOWN.) 16 Q You are generally 17 responsible in that vein, as we sit here today, 18 are you not? 19 A Yes. 20 Q All right. You say that 21 currently the HAMD is being employed? 22 A Yes. 23 Q Specifically Item 3, 24 correct? 179 1 A All items in the Hamilton 2 are used; Item 3 is the one that is specific to 3 the question of suicidality. 4 Q Well, is there any other 5 question on the Hamilton depression scale that is 6 specific to the issue of suicidality? 7 A Only as it relates to the 8 disease depression. 9 Q I'm talking about 10 something to assess suicidality, Doctor 11 Tollefson. Is there anything in the HAMD, other 12 than in Item 3, specifically addressing the issue 13 of suicidality? 14 A No. 15 Q All right. But you say 16 that's the gold standard now in examining 17 suicidality in depressed individuals? 18 A I think I said that it 19 has been used as a gold standard in the attempt to 20 validate other instruments to assess suicidality. 21 Q All right. What other 22 instruments are currently being used by Lilly, 23 other than the HAMD Item 3, to assess suicidality 24 in patients being administered Prozac for any 180 1 indication? 2 A I think at this point 3 prospectively, the Hamilton is the only. 4 Q Have there ever been any 5 other standards or tests used other than the HAMD 6 Item 3 to assess suicidality in patients being 7 given Prozac for any indication? 8 A By Eli Lilly and 9 Company? 10 Q Yes, sir. 11 A Yes. 12 Q All right, tell me about 13 those. 14 A I think we've already 15 spoke about the Reynolds scale. 16 Q All right. When was the 17 Reynolds adult suicidality scale used? 18 A I mentioned that my 19 experience with it was in the study of Prozac 20 versus imipramine in agitated depression. 21 Q All right. Has anybody 22 else used it in any other trial, sir, at any other 23 time -- 24 A I don't know. 181 1 Q -- at Lilly? 2 A I don't know. 3 Q Don't you think you would 4 know about that if that were the case? 5 A Not necessarily. 6 Investigators may add a test scale, a laboratory, 7 as part of a study addendum that may not be part 8 of the core protocol, so it is possible. 9 Q Well, don't you try to 10 keep yourself relatively well advised of what's 11 going on in Prozac clinical trials worldwide at 12 this time? 13 A Relatively well advised, 14 but that's not my primary responsibility. 15 Q It's one of your 16 responsibilities, is it not, sir? 17 A Not primary. 18 Q But it's one of them? 19 A We have clinical research 20 physicians who monitor clinical trials. I 21 supervise a large group of clinical research 22 physicians who monitor the projects. 23 Q You supervise Doctor 24 Beasley, Heiligenstein, do you not? 182 1 A Yes. 2 Q Do you supervise Doctor 3 Pande, Atul Pande? 4 A I did before he left the 5 company. 6 Q All right. And do you 7 supervise -- well, you don't supervise Doctor 8 Masica? 9 A I do not. 10 Q Did Doctor Masica have 11 your job, or did you take over the function that 12 Doctor Masica had? 13 A When I first arrived at 14 Lilly, yes. 15 Q Do you know why Doctor 16 Masica was replaced with you? 17 A No. 18 Q Did anybody express to 19 you any dissatisfaction with the job that Doctor 20 Masica was doing in supervising the psychiatrists 21 there on board? 22 A No, quite the opposite. 23 Q Is Doctor Masica a 24 psychiatrist? 183 1 A He is not. 2 Q All right. Now, back to 3 where we were going. My question is: In your 4 time at Lilly, do you know of any other 5 investigator who at any time in studying 6 individuals being given Prozac for any indication 7 has employed the Reynolds adult suicidal inventory 8 scale other than yourself on that trial you did 9 before you came to Lilly with respect to agitated 10 depressed patients? 11 A Not specifically, no. 12 Q Well, do you know 13 generally of somebody at Lilly, some Lilly study 14 that would have used a scale to examine this 15 issue? 16 A Other than the Hamilton? 17 Q Yes, sir. 18 A No. 19 Q All right. Now, do you 20 know of any study, any protocol that call for any 21 examination that employed a specific scale to 22 examine whether or not patients being given Prozac 23 for any indication were becoming violent and 24 aggressive? 184 1 A The only one that comes 2 to mind would be the construction or modification 3 of the Beck scale by Miller, which you referred to 4 earlier today. 5 Q What's the name of that? 6 A I'm sorry? 7 Q What's the name of that? 8 A The Beck suicide scale, 9 I'm not sure if it has a different name, it's 10 authored by Aaron Beck. 11 Q No, I'm talking about 12 violent aggressive behavior. Does the Beck 13 suicide scale examine violent aggressive behavior? 14 A I'm sorry, I didn't 15 understand that that was your question. 16 Q All right, I may have not 17 made myself clear. 18 A No, to the best of my 19 knowledge, it does not. It certainly has not been 20 validated for that purpose. 21 Q So my question is: Has 22 there been any scale employed in any Lilly Prozac 23 studies to examine modifications in violent 24 aggressive behavior in patients taking Prozac for 185 1 any indication? 2 A All of the studies 3 include event terms for the observation of any of 4 those types of behaviors. They are not formally 5 scaled, they're just recorded as present or 6 absent. 7 Q I'm asking about scales. 8 For instance, you designed an agitation scale, did 9 you not? 10 A Correct. 11 Q I'm talking about a scale 12 designed to pick up emergence of violent 13 aggressive behavior in people taking Prozac. Are 14 there any scales being used by Lilly to examine 15 that issue? 16 A Just to assess the 17 presence or absence of. 18 Q And that's not a scale, 19 that's just one question? 20 A I consider it a scale, it 21 has two end points, either present or not present, 22 but it's not gradiated. 23 Q Yes, sir. In other 24 words, you don't have a base -- beginning baseline 186 1 and an ending end point? 2 A No, you do. 3 Q Present or absent? 4 A Correct. 5 Q It was either present or 6 absent at the beginning and present or absent at 7 the end? 8 A That's correct, or 9 anytime during the course of the trial. 10 Q All right, what's the 11 result of that in all of the Prozac studies? 12 A There are fewer instances 13 of a positive aggressive episode in patients 14 receiving fluoxetine than those receiving placebo. 15 Q There are fewer instances 16 of positive what, sir? 17 A Aggression event terms as 18 positive. 19 Q All right. Now, who did 20 that study? 21 A John Heiligenstein. 22 Q All right. Any other 23 studies other than that performed by Doctor 24 Heiligenstein? 187 1 A I don't believe so. 2 Q And there have been no 3 scales used to examine this issue? 4 A There are very few such 5 scales in existence. 6 Q I didn't ask you that. I 7 said there have been no scales used to examine 8 this issue? 9 A Not that I'm aware of. 10 Q When you wanted to 11 examine the issue of agitation, you designed a 12 scale, didn't you? 13 A I took criteria and tried 14 to assess degrees of agitation. 15 Q As I understand it from 16 your earlier testimony, you were able to submit 17 that to your colleagues to review that, and to 18 Lilly to review, and there wasn't any changes made 19 by anyone in that scale that you came up with, was 20 there? 21 A No, it was a pretty 22 obvious and simplistic approach. 23 Q All right. You say on 24 page two of your letter to Doctor Zerbe: It is my 188 1 general belief that the majority of consultants 2 felt that the current rechallenge protocol, while 3 scientifically or methodologically reasonable, 4 would be extremely difficult to carry out and not 5 likely come to fruition. Correct? 6 A Correct. 7 Q I believe that the group 8 in general endorsed my recommendations for a Food 9 and Drug Administration sponsored broad analysis 10 of the issue. Correct? 11 A Correct. 12 Q Do you know if there was 13 ever a broad analysis of the issue done by the 14 Food and Drug Administration or any group? 15 A I think they made an 16 exhaustive analysis as part of the Advisory 17 Committee hearing process. 18 Q Other than that, did they 19 do any study -- you're talking about the 20 rechallenge protocol in that paragraph, are you 21 not? 22 A Earlier in the paragraph. 23 Q All right. Well, isn't 24 that what you are talking about when you're 189 1 talking about a Food and Drug Administration broad 2 analysis of the issue? 3 A No, I wouldn't consider 4 rechallenge to be broad at all, it's highly 5 specific. 6 Q Well, you include that in 7 an analysis of the issue, do you not, in that one 8 paragraph? 9 A I'm sorry, I don't 10 understand your question. 11 Q Well, you include 12 rechallenge protocol as something that is 13 scientifically and methodologically reasonable, 14 don't you? 15 A Yes. 16 Q But you say it's 17 difficult to carry out? 18 A Yes. 19 Q And what you suggest is a 20 Food and Drug Administration sponsored broad 21 analysis of the issue? 22 A Correct. 23 Q Why were you suggesting 24 that the Food and Drug Administration do that? 190 1 A Because I think that the 2 feeling was that if -- and I'd emphasize 3 hypothetically -- if there were any relationship 4 between any drug and a particular adverse event, 5 in this case it was an extremely rare event, and 6 the only way to get a better feel for the 7 existence or the lack of existence of such a 8 relationship would be with a very large database. 9 Q All right. Well, a 10 larger database than that that Lilly had at the 11 time? 12 A Lilly's database was 13 primarily focused on fluoxetine. I think the 14 thought was that there would be value in looking 15 at other databases of similar size with other 16 chemical entities. 17 Q Well, are you talking 18 about putting it all together in, for instance, 19 the spontaneous reporting system? 20 A That's a completely 21 different mechanism to approach the question. It 22 would be one that could be used, but it is not 23 directly comparable to controlled clinical trials. 24 Q Well, the spontaneous 191 1 reporting system is beneficial in that it gets a 2 better picture of a broader range of patients than 3 does the clinical trial information, doesn't it? 4 A I would disagree with the 5 characterization of a better profile. It really 6 has a number of things it suffers from. It 7 clearly is a larger sample size, but it's also 8 extraordinarily uncontrolled and unable to 9 demonstrate causality. 10 Q Well, but it has a larger 11 amount of people being subject to the drug than 12 are put in any clinical trial? 13 A That is correct. 14 Q I mean, there is some 15 reason for the spontaneous reporting system, isn't 16 there, Doctor Tollefson? 17 A Of course. 18 Q And that reason is, is 19 that the United States government, through the 20 Food and Drug Administration, is of the belief 21 that the information supplied by the spontaneous 22 reporting system is of some value in analyzing the 23 safety of a drug, correct? 24 A I think that's true. I 192 1 think that they use it to observe for possible 2 trends. 3 Q All right. And just 4 because a product has been approved by the Food 5 and Drug Administration, that's not to mean that 6 that product is safe at all times for all people 7 thereafter, is it? 8 A That's correct. I know 9 of no product that's safe for all people. 10 Q And that's one of the 11 reasons that the Food and Drug Administration 12 requires, by law even, that manufacturers of a 13 drug submit to them any reports of adverse 14 experiences with their particular drug? 15 A I assume that's one of 16 the reasons, yes. 17 Q Well, do you think that's 18 a valid reason? 19 A Yes. 20 Q In other words, the 21 clinical trials have certain benefits in examining 22 the scientific question of the safety of the drug, 23 don't they? 24 A I think those are the 193 1 principal data. 2 Q Correct. I mean, 3 supposedly you're going to have investigators 4 looking at a drug potentially more careful, 5 potentially there's going to be better follow-up 6 by those investigators, correct? 7 A Circumstances are much 8 better controlled. 9 Q By virtue of that, there 10 is -- you're more likely to get a picture of a 11 patient that is not quite what the picture of the 12 patient that you would get if you looked at what 13 that patient's going to be in the general 14 population, correct? 15 A I think they are very 16 similar patients, they're just under less 17 controlled circumstances. 18 Q All right. And the Food 19 and Drug Administration is interested in what 20 comes up in those less-controlled circumstances, 21 aren't they? 22 A Yes. 23 Q And they're interested in 24 that because they're interested in the safety of 194 1 that drug, are they not? 2 A They're interested in 3 rare events, and that is a method to determine a 4 trend for a rare event. 5 Q Is it your testimony, 6 Doctor Tollefson, that the spontaneous reporting 7 system only picks up rare events in connection 8 with the safety of a drug, is that what your 9 testimony is here, sir? 10 A I think my testimony was 11 that's one of the reasons that there is a 12 spontaneous reporting system. 13 Q Does it only pick up rare 14 events? 15 A What's your definition of 16 rare? 17 Q You used the term, sir. 18 What's your definition of rare? 19 A I think things -- my 20 definition is things that might occur less than 21 one in ten thousand, for example. That's one of 22 the purposes for a spontaneous event reporting 23 system. 24 Q What's the other 195 1 purposes? 2 A Drug interactions -- 3 really, those would be the major purposes. 4 Q And you define a rare 5 event as an event that occurs only in one in ten 6 thousand individuals? 7 A Approximately. 8 Q Okay, if you gave that 9 medication to a million people and it occurred at 10 a rate of one in ten thousand people, how many 11 people would experience the event? 12 A It occurred exactly at 13 one in ten thousand? 14 Q Yes. 15 A You would have ten cases 16 per hundred thousand, one hundred cases per 17 million. 18 Q All right. And if that 19 drug were given to ten million people and that 20 event occurred at that rate, how many people would 21 be affected by that rare event? 22 A Approximately a thousand. 23 Q What if that were an 24 event that was life-threatening, would you 196 1 consider that to be an instance of an event that 2 should be reported that occurred in that 3 frequency, one in ten thousand? 4 A Of course. 5 Q Why? 6 A Because it would be a 7 serious adverse event. 8 Q And if it was given to 9 ten million people, it would cost a thousand 10 people their lives? 11 A Pardon me? 12 Q If it occurs in ten 13 million people, it could cost a thousand people 14 their lives, correct? 15 A Spontaneous event 16 reporting has nothing to do with causality. 17 Q No, but if it did in fact 18 was the cause -- if it in fact was the cause of 19 that rare event, then it would account for the 20 loss of one thousand lives if it were given to ten 21 million people, correct? 22 A If it were shown 23 unequivocally to be causal, yes. 24 Q What if it were only 197 1 causal in fifty percent of the people, so only 2 five hundred people were effected by this -- 3 A Uh-huh. 4 Q -- would that be a 5 serious event? 6 A Yes. 7 Q Would that be an event 8 that the physicians in this country should be 9 advised about? 10 A That would be up to the 11 FDA to determine. You would have to look at the 12 base rate of the disease. 13 Q Would Lilly want to 14 advise physicians of an event that occurred in 15 five hundred people out of ten million? 16 A If it was found to be 17 causal? 18 Q Yes. 19 A Sure, certainly. 20 Q Why? 21 A Pardon me? 22 Q Why? 23 A Because Lilly, and any 24 corporate sponsor I would think, would want to 198 1 provide a complete listing for the prescribing 2 physician of all causal adverse events. 3 Q What does Lilly do to 4 advise physicians of causal adverse events? 5 A It has a package insert. 6 Q Do all the adverse 7 reactions listed in the package insert have a 8 causal reaction to Prozac? 9 A No, they do not. 10 Q So what I wanted to know 11 is, what does Lilly do to advise physicians of the 12 reactions that are caused by Prozac? 13 A They're included in the 14 package insert if they're causal. 15 Q Which ones are those? 16 A There's a long list. I'd 17 have to have the package insert. I'd be happy to 18 read it to you. 19 Q Are they all -- 20 A All things that are not 21 considered to be postintroductory report and are 22 statistically significantly more common than were 23 seen with placebo may have causality associated 24 with them. 199 1 Q Doctor Thompson only 2 mentioned four: Nausea, tremor -- what are the 3 other two? 4 MS. ZETTLER: Agitation 5 and nervousness. 6 Q Agitation and 7 nervousness. Do you recall those? 8 A Yes. 9 Q Would you point them as 10 causal to Prozac in your review of the literature 11 and in your experience? 12 A They could be. 13 Q And I don't want to be 14 unfair with you, Doctor Thompson, as yourself, 15 says in any particular instance you can't 16 determine for sure whether or not a particular 17 reaction might or might not be caused by the drug, 18 correct, do you agree with that? 19 MR. FREEMAN: In a 20 particular patient I think he said. 21 MR. SMITH: Yes. 22 A To an extreme, I think 23 that's true. 24 Q What do you mean? 200 1 A I think one can have 2 varying degrees of suspicion if something is 3 causal or not, but one would be in a difficult 4 position with an N of one, one subject, to say it 5 was or it was not. 6 Q Turn to Page 4 of your 7 letter to Doctor Zerbe. In the fourth paragraph, 8 the last sentence, you say: I also have serious 9 questions how we will be able to maintain the 10 integrity of the blind given the fairly easily 11 identified side-effect profiles that would 12 differentiate desipramine and fluoxetine. 13 Correct? 14 A I have to apologize, I'm 15 not seeing that. You're on Page 4? 16 Q Page 4? 17 A And which paragraph? 18 Q The fourth paragraph, the 19 last sentence starting -- 20 A All right, hold on. 21 Q See it? 22 A I also have. 23 Q Yes. I also have serious 24 questions how we will be able to maintain the 201 1 integrity of the blind given the fairly easily 2 identified side-effect profiles that would 3 differentiate desipramine and fluoxetine. 4 Correct? 5 A I see that sentence. 6 Q All right. What are the 7 easily differentiated side effects between Prozac 8 and desipramine? 9 A The anticholinergic 10 profile of desipramine, which would include very 11 commonly dry mouth, light-headedness, 12 constipation, problems initiating urinary flow. 13 Q Urinary retention? 14 A Or just difficulty 15 initiating flow. 16 Q And none of those are 17 present generally speaking in Prozac treated 18 patients? 19 A Not significantly more so 20 than would be seen with placebo. 21 Q All right. And are there 22 particular easily identified side effects of 23 Prozac on the other hand that identify -- that are 24 related to Prozac? 202 1 A I think there are very 2 few. Some of the gastrointestinal upsets, the 3 patients might have, nausea, diarrhea, could be 4 signs the patient was on a selective serotonin 5 reuptake inhibitor. They certainly would not be 6 unique to fluoxetine. 7 Q All right. So your 8 concern here was, is that when the investigators 9 were doing these clinical trials, see dry mouth or 10 light-headedness or constipation or problems 11 instituting urinary flow, these are pretty clearly 12 known as side effects of tricyclic 13 antidepressants, correct? 14 A They're commonly reported 15 with tricyclics, yes. 16 Q Especially including 17 desipramine, correct? 18 A Yes, sir. 19 Q But you don't see those 20 in Prozac treated patients? 21 A You see them, but not as 22 high a frequency. 23 Q So therefore you're going 24 to -- your the investigator is -- your concern is 203 1 that the investigator is going to know which 2 patients are on which medication by virtue of what 3 you call easily identified side-effect profiles, 4 correct? 5 A Concern is not that they 6 would know, but that they might speculate that 7 they know. 8 Q All right. They might 9 speak with a degree of accuracy, too? 10 A Possibly. What that 11 degree is, I don't know. 12 Q You did that and could do 13 that when you were doing your clinical trials, 14 couldn't you, Doctor Tollefson? 15 A I did not do that; I 16 could have done that. 17 Q And in going back after 18 you got the results and after the trial was 19 finished, you could pretty well confirm who was on 20 what based on your knowledge of the side effects, 21 correct? 22 A We didn't do that, but I 23 suppose it would be theoretically possible if one 24 wanted to go back and look at individual case 204 1 records. 2 Q Right. 3 A But that's quite a 4 burdensome task. 5 Q Well, you were concerned 6 that the investigators might do this 7 subconsciously, not implying any ill motive to the 8 investigators, but it just -- it might be done 9 subconsciously? 10 A That is possible. 11 Q Why did you leave 12 St. Paul Ramsey to go with Lilly; were you 13 dissatisfied with St. Paul Ramsey? 14 A No, quite the opposite. 15 I sometimes wonder why I did leave, but -- I miss 16 Minneapolis and St. Paul. I enjoyed it very 17 much. But it was -- I had done what I basically 18 had been doing for about ten years and was 19 approaching forty and it was an opportune time to 20 try something different outside of the Twin Cities 21 and in a different facet of clinical investigation 22 and research, and there was a very attractive 23 pipeline of new compounds at Lilly that I thought 24 provided an opportunity for a very innovative 205 1 science. 2 Q Well, you were seeing 3 patients back at St. Paul Ramsey, were you not? 4 A Yes. 5 Q On a regular basis? 6 A Yes, sir. 7 Q On a daily basis? 8 A Most often, yes. 9 Q And that's not occurring 10 now, is it? 11 A That's correct. 12 Q Do you miss the patient 13 care? 14 A To a small degree. 15 Q Or were you getting 16 burned out for that? 17 A To a small degree, but 18 not to a tremendous degree where it's going to 19 drive me back to clinical practice in the next 20 week or two. 21 Q You think you're pretty 22 well finished with clinical practice? 23 A While I'm at Eli Lilly 24 and Company. If I were to do something different 206 1 in the future, I could certainly feel very 2 comfortable returning to patient care. 3 Q Well, do you have any 4 plans to do that, sir? 5 A Not at this time. I keep 6 my options open. 7 Q Well, are you 8 dissatisfied with your work at Lilly? 9 A Not at all. 10 Q In either one of the 11 consultants meetings that you had at Lilly, was it 12 discussed at that time that the Prozac clinical 13 database taken as a whole indicated that Prozac 14 was an activating medication? 15 A No. 16 Q It was after this 17 analysis of the database was made that it was 18 determined that Prozac was an activating 19 medication? 20 MR. FREEMAN: Objection, 21 you're stating a fact into the question that has 22 not been testified to by this witness as a fact, 23 when he has not testified that that is a fact, 24 that it is an activating medication. 207 1 MR. SMITH: Maybe I 2 misunderstood, and I'll certainly clear that up if 3 there is some misunderstanding. 4 Q (BY MR. SMITH) Was I not 5 wrong when -- did you testify that there was some 6 analysis made by Doctor Beasley or Heiligenstein 7 that -- or yourself, that Prozac was generally an 8 activating medication? 9 A No, I did not say that. 10 Q All right, what was the 11 finding in that connection? 12 A I think you were talking 13 about a study conducted by Doctor Beasley where he 14 classified a variety of side effects under a 15 category of activation and found that a minority 16 of patients receiving fluoxetine had that 17 experience. 18 Q And it was a greater 19 number of patients in Prozac-treated patients than 20 it was in placebo-treated patients, is that right? 21 A That's correct. 22 Q So you're more likely to 23 see some activation in the Prozac clinical trials 24 than in the placebo -- or in the Prozac patients 208 1 during the clinical trials than in placebo? 2 A That's correct. 3 Q All right. So from that, 4 since you had double blind clinical trials, since 5 you had placebo control and you saw the instance 6 of activation more readily in Prozac than placebo, 7 can you conclude from that that it's more likely 8 that you will see activation among Prozac treated 9 patients? 10 A Than? 11 Q Than patients that are 12 not receiving Prozac. 13 A No. 14 Q Why? 15 A If you're saying placebo, 16 that's one thing. If you're saying not being 17 treated, receiving a tricyclic, you know, I just 18 would have -- I'm not clear on what your 19 comparator is relative to -- 20 Q If the comparator is 21 placebo, they're not being treated with active 22 medication, are they? 23 A Okay, if you would like 24 to rephrase your question, help me understand 209 1 specific question relative Prozac versus placebo, 2 I could probably give you a better answer. 3 Q Okay. I'm not as good at 4 this as I should be, but the question is whether 5 or not -- the question Doctor Heiligenstein was 6 examining was whether or not Prozac was more 7 likely to be activating in patients receiving -- 8 in patients receiving Prozac than placebo, is that 9 right? 10 A No, it was Doctor 11 Beasley. 12 Q All right. Did Doctor 13 Beasley find a greater incidence of activation 14 among patients taking Prozac than placebo? 15 A Yes. 16 Q What is the significance 17 of that, sir? 18 A The significance is that 19 event terms that he mapped to activation are 20 likely part of the side-effect profile of Prozac 21 in some individuals. 22 Q All right. Can we say 23 then in some individuals that activation may be a 24 side-effect profile? 210 1 A Yes. 2 Q Was that discussed, the 3 fact that in some individuals there may be an 4 activation side-effect profile discussed at either 5 one of these consultant meetings that you heard? 6 A I think the entire 7 adverse event profile consistent with the package 8 insert of Prozac was discussed. 9 Q Was that discussed, sir? 10 A As part of that 11 discussion, yes. 12 Q Who mentioned that, that 13 you might see -- 14 A I don't recall. 15 Q -- activation? 16 A I don't recall. 17 Q I didn't see anything 18 about that in your examination or in your 19 testimony before the PDAC. Do you recall making 20 that statement? 21 A Which statement? 22 Q That in some patients 23 you're more likely to see an activation 24 side-effect profile than in placebo patients. 211 1 A I don't think that was 2 part of the presentation question. 3 Q All right. 4 A But clearly the FDA was 5 well aware of that, and it was part of their 6 overall discussion. They had an epidemiological 7 presentation on the adverse events through the 8 spontaneous we referred to earlier and in the 9 controlled clinical trial database. 10 Q I'm talking about 11 controlled clinical trials, right? 12 A I'm talking about both. 13 Q Because you've already 14 testified that as far as you're concerned, the 15 spontaneous reporting system is not nearly as good 16 as the clinical trial data, correct? 17 A No, I don't think I said 18 it wasn't nearly as good; I said that it's a 19 different method. 20 Q Different method? 21 A It's like green and 22 black. I mean there are two different colors, 23 those are two different approaches to the 24 assessment of potential adverse events. I 212 1 wouldn't want to say one is better, one is worse. 2 Q All right. 3 A They answer different 4 things. 5 Q Are both scientifically 6 valid methods of examining an issue? 7 A Yes, although I think 8 there are different scientific disciplines that 9 might endorse one or the other as favorable. 10 Q All right. Could it be 11 said that one may have an advantage over the 12 other, but the other may have an advantage over 13 the other? 14 A Yes. 15 Q In other words, that the 16 limitations of one might be the benefit of the 17 other, correct? 18 A In general, I think 19 that's fair. 20 Q And that it's 21 unreasonable to totally discount one from a 22 scientific standpoint and totally rely on the 23 other? 24 A If you use the word 213 1 totally, I would agree. 2 Q Absolutely rely on the 3 other, say I'm not going to look at the results of 4 the clinical trial database, I'm only going to 5 look at the results of the spontaneous reporting 6 system. That's no good, is it? 7 A I think it's best to take 8 them in concert. 9 Q Because they are both 10 scientifically reasonable methods for 11 investigating a question, correct? 12 A Yes. 13 (SHORT BREAK TAKEN.) 14 Q (BY MR. SMITH) Back to 15 the activation side-effect profile in some Prozac 16 patients. Was it reported to you in the 17 consultants meeting that the BGA was concerned 18 about the activation side-effect profile in some 19 Prozac patients? 20 A No, I don't remember any 21 specific reference to the BGA. 22 Q When did you first learn 23 that the BGA -- or did you ever learn that the BGA 24 was concerned about the activation side-effect 214 1 profile? 2 A When I joined the company 3 in June of 1991, that was not an issue in the 4 minds of the BGA, as I understood it. 5 Q But were you -- did you 6 understand that it had been raised as an issue -- 7 A No. 8 Q -- back in 1984? 9 A Not in '84, no. 10 Q At any time? 11 A I did not hear any 12 discussion regarding the BGA prior to my joining 13 the company. 14 Q No, I mean after you 15 joined the company, did you ever hear that back in 16 1984, June of '84, the BGA, the German equivalent 17 to the FDA had raised concerns about the side 18 effect activation profile of Prozac patients? 19 A No, I did not. 20 Apparently it was not an ongoing concern or it 21 would have been brought to my attention at that 22 time, though. 23 Q So it was not brought to 24 your attention? 215 1 A Not as a past event or a 2 current -- not as a previous concern or a current 3 concern. 4 Q And did you discuss this 5 with Doctor Zerbe in any way? 6 A I'm sorry, did I discuss 7 what? 8 Q This fact that the BGA 9 had raised the activation as an area of concern? 10 A No, I did not. 11 Q Did you know that the BGA 12 felt that this activation profile might have a 13 side-effect profile of suicides and suicide 14 attempts? 15 A No. 16 Q Do you know of any 17 psychiatric thinking or thinking among 18 psychiatrists that activation raises a side-effect 19 profile of suicide and suicide attempts? 20 A Teicher, for example, 21 hypothesized that association. It's not been ever 22 demonstrated in controlled clinical trials. There 23 have been hypothetical associations or linkages in 24 anecdotal case descriptions of such. 216 1 Q But this was raised back 2 in 1984 by the German government? 3 MR. FREEMAN: He's 4 already said he doesn't know anything about that 5 now. 6 MR. SMITH: I understand 7 he doesn't know anything about that. 8 Q (BY MR. SMITH) But I'm 9 wondering whether or not there was anything in the 10 psychiatric literature back in 1984 of which you 11 are aware that linked activation with a 12 side-effect profile of suicide and attempted 13 suicides? 14 A Relative to any one 15 particular drug -- 16 Q Yes. 17 A -- or just in general? 18 Q In general or to any one 19 particular drug? 20 A Not regarding the term 21 activation, no, and not regarding any specific 22 drug. 23 Q You say not regarding the 24 term activation; regarding any other term similar 217 1 to that? 2 A No, there's a long held 3 hypothesis that as patients improve from their 4 clinical depression -- and it is only a 5 hypothesis, that they first improve as far as 6 their motor function, and some people, Dietre and 7 Jarecki were two investigators for example that 8 popularized that concept that that might be a 9 period of time where people are more active, yet 10 their mood is not optimized yet during the course 11 of treatment. That was a hypothetical construct 12 I'm familiar with; I don't know that anyone has 13 ever replicated or validated it. That would be 14 the closest thing I can think of to your question. 15 Q But that question has 16 been raised as a potential hypothesis or a 17 potential explanation as to some suicides and 18 suicide attempts in individuals? 19 A Yes. 20 Q You're just saying you're 21 not aware of any clinical trials that have 22 validated that hypothesis, is that what you're 23 saying? 24 A That's what I'm saying. 218 1 Also, I don't think it's a contemporary 2 hypothesis. 3 Q Do you know of any 4 clinical trials that have been done to examine the 5 veracity of that hypothesis? 6 A Well, I think there have 7 been a number of trials that have done analyses of 8 different factors of depression rating scale 9 scores to see if one might improve before another, 10 and there isn't any consistent pattern of response 11 either time-wise or symptom-wise in individuals. 12 Q But that doesn't mean 13 that it doesn't occur, does it, sir? 14 A That what does not 15 occur? 16 Q That you have a situation 17 where an individual might have their motor 18 facilities improved at a greater speed than their 19 mood, therefore making it more likely that they 20 would attempt suicide or commit suicide? 21 A I don't know that there's 22 any reason to think that there's a linkage between 23 earlier motor improvement and suicide. I suspect 24 there may be patients where motor features are an 219 1 early response pattern, but the linkage between 2 the early motor improvement and suicide, I think, 3 is an unestablished connection. 4 Q But it hasn't proved to 5 be false, has it? 6 A I don't think anyone has 7 proven it to be true. 8 Q Or false? 9 A Well, by the number of 10 investigations failing to establish the 11 relationship, I think one would begin to think 12 there's a high probability that it's false, but it 13 has not been unequivocally demonstrated. It would 14 be essentially impossible. 15 Q All right. Is the theory 16 in that that normally you'll see, with depressed 17 individuals, a feeling of lethargy and inactivity, 18 and then that will raise and the person's mood 19 will not really raise, therefore you might have an 20 increase in suicidality by virtue of the activity 21 increasing before the mood increases; is that the 22 hypothesis? 23 A That was the hypothesis 24 of Dietre and Jarecki when they originally wrote 220 1 that paper. 2 Q When was that written? 3 A I believe it was in the 4 mid to latter '70s, but I don't know the exact 5 date. 6 Q And are you saying -- and 7 I'm not sure I understand. Are you saying there 8 have been clinical trials specifically designed to 9 examine that hypothesis as presented by those 10 authors that have proven that that hypothesis is 11 incorrect? 12 A I don't think that's what 13 I said. 14 Q All right. 15 A What I said was that 16 people have looked at response patterns to see if 17 indeed motor symptoms improve before other 18 symptoms, and that is not a consistent pattern of 19 improvement in the recovery process. 20 Q But that doesn't say it 21 doesn't occur in some individuals, does it? 22 MR. FREEMAN: Let him 23 finish his answer. 24 MS. SMITH: I'm sorry, I 221 1 thought you'd finished. Did you finish? 2 MR. FREEMAN: No, you cut 3 him off. I doubt the court reporter got it. Can 4 you read the last response? 5 REPORTER: (READING) What 6 I said was that people have looked at response 7 patterns to see if indeed motor symptoms improve 8 before other symptoms, and that's not a consistent 9 pattern of improvement in the recovery process. 10 MR. FREEMAN: Do you need 11 to add anything to that? 12 A I was going to say that 13 you look at any cluster of symptoms by chance in 14 an individual, one cluster will improve before 15 another. 16 Q And that's known? 17 A Pardon me? 18 Q And that's known? 19 A Yes. 20 Q All right. Were you ever 21 advised that in the spring or in early 1985 that a 22 Professor Herrmann had rendered an expert opinion 23 in connection with the registration of Prozac in 24 Germany? 222 1 A No. 2 Q Have you ever been 3 advised that Professor Herrmann, in that opinion, 4 found a greater incidence of suicide attempts of 5 individuals on Prozac than on imipramine? 6 A No. 7 Q And certainly you didn't 8 make the PDAC aware of that, since you didn't know 9 about it? 10 A I'm not aware of it, what 11 you're referring to, so I may have or I may not, 12 it wasn't -- 13 MR. FREEMAN: You're not 14 going to give him the twenty-one-page report? 15 MS. ZETTLER: When we get 16 it, we'll give it to him, Joe. 17 MR. SMITH: All I can 18 give him -- 19 MS. ZETTLER: We wanted 20 Doctor Herrmann's version, not Lilly's version. 21 MR. FREEMAN: Shoot, you 22 got the translation, you won't admit that you were 23 wrong. 24 MR. SMITH: Can you edit 223 1 out them -- what we call down in Texas, side bar 2 remarks? 3 MR. FREEMAN: No, no. 4 MR. SMITH: I'll make the 5 judge make him edit it out then. 6 (TOLLEFSON EXHIBIT NO. 4 MARKED FOR 7 IDENTIFICATION.) 8 Q (BY MR. SMITH) Review 9 Exhibit 4, Doctor Tollefson, so I can ask you some 10 questions about it. 11 A Okay. 12 Q Do you know on the bottom 13 of the page, it indicates that Professor Herrmann 14 left an opinion of twenty-one typewritten pages. 15 The essential points are summarized as follows, 16 and then those points are summarized on the last 17 two pages, correct? 18 A Presumably, I don't know. 19 Q And apparently Professor 20 Herrmann has reviewed, among other things, 21 Protocol No. 27, which is the US study, correct? 22 A I don't know the 23 reference Protocol No. 27, I'm sorry. 24 Q The study comparing 224 1 imipramine is a US study, it's a multicenter study 2 as far as you know, isn't it? 3 A There were trials with 4 imipramine done in the US and overseas. I'm not 5 sure where Protocol 27 was conducted. 6 Q Protocol 27 is a 7 multicenter study that was a pivotal study as far 8 as the FDA was concerned, comparing placebo, 9 imipramine, and Prozac. All right, does that help 10 you? 11 A No, but I'm not familiar 12 with Protocol 27 by that name and there were other 13 trials conducted against imipramine. The fact 14 that it included a placebo would suggest it was 15 done in the US, but I'm just saying I don't know 16 where it was conducted. 17 Q We've pretty well 18 established that through previous testimony of 19 Lilly employees up to date, okay. Can you take my 20 word for that? 21 A Well, you haven't told me 22 yet. Was it conducted in the US or not? 23 Q Yes. 24 A Thank you. 225 1 Q Although there might have 2 been a Canadian investigator. It was conducted in 3 the United States. What I'm pointing to is on 4 Page 2, Professor Herrmann says: Still not 5 resolved is the fact that suicide attempts have 6 been observed more frequently on fluoxetine as 7 compared to imipramine. Do you see that? 8 A Yes, I do. 9 Q Have you ever seen any 10 studies where suicide attempts were observed more 11 frequently on fluoxetine as compared to 12 imipramine? 13 A If one looks at 14 investigators within a study, one would expect to 15 see trends one way or the other, either in favor 16 of one compound or in favor of another, looking at 17 individual investigative sites. So in that sense, 18 yes. 19 Q Do you think that's what 20 Professor Herrmann is talking about here? 21 A It's quite possible. 22 Q Well, the reason I ask 23 that is that they go on to say: Only 24 epidemiologic data or literature on other 226 1 antidepressants may help to identifying whether it 2 happened by chance that incidence of suicide 3 attempts was abnormally high on fluoxetine or 4 abnormally low under comparators. 5 A Uh-huh. 6 Q That would indicate that 7 they're looking at the pooled data? 8 A No, not necessarily. It 9 sounds like they're looking at -- you've said an 10 individual study, not a pool of studies. 11 Q If you look under data 12 review, on the first page, it says: Analysis of 13 pooled studies fluoxetine versus imipramine versus 14 placebo, Protocol No. 27, submitted October 26th, 15 1984. 16 A Uh-huh. 17 Q So it doesn't look to me 18 like he's looking at individual studies based on 19 what's stated here as pooled studies, does it? 20 A Well, I mean, I'm not 21 sure what he's looking at. 22 Q All right. If he is 23 looking at pooled studies, then have you ever seen 24 any pooled studies where there were suicide 227 1 attempts observed more frequently on fluoxetine as 2 compared to imipramine? 3 A I've seen the database as 4 a whole where that is not the case. I would 5 suspect, if you were to take out individual 6 studies, or for that matter an isolated protocol, 7 it is possible, as he says here, perhaps by 8 chance, that you would see a trend one way or the 9 other. 10 Q All right. Well -- and 11 what is being said here is that suicide attempts 12 were abnormally high on fluoxetine or abnormally 13 low under comparators. Do you see that? 14 A Yes, I do. 15 Q Have you seen any data at 16 all in the Prozac clinical trial database to 17 indicate that suicide attempts were abnormally 18 high on fluoxetine and abnormally low under 19 comparators? 20 A Certainly not in the 21 database as a whole. 22 Q Are there some parts of 23 the database that indicate that? 24 A I don't know, but as I 228 1 said earlier, if you made random cuts in the 2 database, you could find trends one way or you 3 could very theoretically find trends the other 4 direction. So by chance the answer would be an 5 obvious yes, you could see it. 6 Q What are random cuts? 7 A If you had twenty 8 protocols and you chose to look at two out of the 9 twenty, by chance, the two that you choose might 10 show one thing. If you choose another two out of 11 the twenty, it might show an opposite 12 observation. The advantage of pooling all of it 13 together is to reduce the heterogeneity or the 14 variability and get a true estimate. 15 Q So to get the true 16 estimate in looking at the clinical trial data, 17 you should look at the entire database, is that 18 what you're saying? 19 A Of comparable studies. 20 Q All right. So you should 21 look at all double-blind placebo controlled 22 studies throughout the world? 23 A Assuming they're 24 comparable. If they're not comparable studies or 229 1 comparable patient populations, or the methodology 2 of the particular protocol differed substantially, 3 then it would not be good scientific principal to 4 lump them together or throw them together for an 5 analysis. They would have to meet certain rigors 6 of commonality from protocol to protocol in order 7 to be subjected to what's called the meta 8 analysis. 9 Q All right. Well, I 10 thought you just said that if you took separate 11 data cuts, you're going to get possibly something 12 that's not a true reflection of what the data as a 13 whole contains, is that right? 14 A No, I think it's a little 15 out of context with what I said. 16 Q How am I off context? 17 A I said that if you had a 18 pool of studies that did meet the criteria for 19 being pooled from meta analytic purposes, and then 20 you pooled only two of those, for example, twenty 21 studies, you might find observations that weren't 22 representative of the pool as a whole -- 23 Q Well, what have you -- 24 A -- by chance. 230 1 Q I'm sorry, I didn't mean 2 to cut you off. 3 A By chance. 4 Q What if you had 5 double-blind placebo controls in depression 6 conducted under basically the same protocol, would 7 you want to pool all of those studies regardless 8 of whether they were conducted in the United 9 States or in the entire world? 10 A It would depend on the 11 protocol. If they truly were the same protocol or 12 a very close facsimile of that protocol, then that 13 might be feasible. 14 Q There wouldn't be any 15 reason to use only data from one country if you 16 had data from many countries under the same 17 protocol for the same indication? 18 A I think the practice of 19 medicine is quite different in other countries, so 20 that the protocols rarely are identical, and 21 sometimes they're not even directly comparable 22 from country to country because of the practice 23 standards. 24 Q Well, give me an example 231 1 of that. 2 A Example might be using a 3 different diagnostic classification, it might be 4 on the issue of concomitant medication, it might 5 be in the issue of administering psych therapy 6 with drug therapy, it might have to do with the 7 frequency that patients are seen or the intensity 8 with which they are seen. Again, the art of doing 9 good clinical science is to reduce variables down 10 to the one that you're testing, and the 11 heterogeneity of a, quote, global database is that 12 you don't control many of those variables and you 13 begin -- you throw them all together and try to 14 pool them, they really aren't comparable studies. 15 MR. FREEMAN: Let the 16 record show that earlier the witness also showed 17 that in many parts of the world placebo studies 18 are not permitted. 19 MS. ZETTLER: Motion to 20 strike the side bar comments. 21 Q (BY MR. SMITH) Well, 22 you're not saying that those clinical trials 23 outside the United States have no clinical 24 significance to individuals living in the United 232 1 States, are you, Doctor Tollefson? 2 A I don't think I said 3 that. 4 Q Well, you didn't mean to 5 imply that either, did you? 6 A That they have no 7 significance? 8 Q Right. 9 A No, I don't think that 10 was the question we were talking about. We were 11 talking about statistical methodology in meta 12 analytic practice. 13 Q Okay. Well, is that 14 statistical methodology and meta analytic 15 practice, does that answer the question on 16 causation as to whether or not a particular 17 individual might have become suicidal on Prozac or 18 might have become violent and aggressive on 19 Prozac? 20 A I think all -- the vast 21 majority of scientists would agree that 22 double-blind controlled methodology, such as 23 you're citing, is the best way to test a 24 scientific hypothesis. 233 1 Q All right. So shouldn't 2 you use double-blind placebo controlled 3 methodology and use all of it? 4 A As long as the study 5 protocols are comparable that would permit the 6 pooling, yes. 7 Q All right. What if you 8 pooled bulimia, smoking cessation and obsessive 9 compulsive disorders, would that be appropriate -- 10 A I'm not sure. 11 Q -- in examining a 12 question? 13 A It depends on what the 14 question is and it depends on what you're pooling 15 them with. 16 Q Whether or not -- pooling 17 them together, pooling bulimia, smoking cessation 18 and OCD in examining the question as to whether or 19 not Prozac causes suicidality in individuals. 20 A Again, it would depend on 21 what the clinical question was. If one is trying 22 to do a comparison between mood disorders and 23 non-mood disorders, they would represent a 24 category of non-mood disorders. That is a common 234 1 denominator. 2 Q Well, isn't bulimia and 3 depression -- aren't they co-morbid conditions? 4 A They can be. They are 5 not necessarily seen together. 6 Q Isn't bulimia and 7 obsessive-compulsive -- I mean isn't depression 8 and obsessive-compulsive disorder co-morbid? 9 A Again, they again can 10 be. They are not always seen together. 11 Q Is what you're saying 12 that as long as you pool similar clinical protocols 13 designed to examine similar indications, that the 14 more data you pool, the more likely you are to get 15 an accurate reading as to what the data reveals? 16 A Not necessarily. 17 Q All right. So where am I 18 wrong? 19 A There are at least two 20 factors. One would be the number of studies, and 21 two would be the comparability of the studies. If 22 they indeed are all comparable, then the more that 23 you have, the better. If on the other token 24 you're adding studies to increase number, but 235 1 you're sacrificing the comparability of those 2 studies, you may then compromise validity. 3 Q No, I'm saying that you 4 have comparable studies, you're just getting 5 information from more studies? 6 A If they're truly 7 comparable, then you would include as many of 8 those comparable studies as you could. 9 Q You didn't have any 10 problem in pooling the data from those studies you 11 did in connection with this examination of the 12 issue of emergent suicide during treatment of 13 placebo double -- of placebo controlled-double 14 blind study employing several different 15 antidepressants, did you? 16 A Those were all controlled 17 trials at a single site using the same 18 investigative team, so they had a lot of 19 homogeneity so I did not. 20 Q And as long as you have 21 that homogeneity, the fact that there may be some 22 differences is not of particular importance to 23 you? 24 A It depends on what those 236 1 differences are. 2 Q Well, you didn't see any 3 differences in these studies to prevent pooling? 4 A That's correct. 5 Q All right. Why wasn't a 6 different statistical approach used in the 7 analysis of the US clinical trials as opposed to 8 that of the outside-US clinical trials? 9 A I'm not sure what you're 10 referring to. 11 Q All right. In one -- I'm 12 talking about in your presentation to the PDAC. 13 You used the Haenszel? 14 A Mantel-Haenszel Cochran. 15 Q Mantel-Haenszel in the 16 OUS studies, right? 17 A No, I believe that was 18 conducted in the US. 19 Q All right, why was that 20 used in the US to analyze the data in the United 21 States? 22 A That was not the only 23 method used. It was the method that was used in 24 the publication of the data in the British Medical 237 1 Journal. 2 Q Well, you used the 3 Mantel-Haenszel method of looking at the US IND 4 trials, right? 5 A That's one of the methods 6 used, yes. 7 Q Well, you showed a slide 8 using that? 9 A That's correct. 10 Q Why was that method 11 employed for looking at the US IND trials? 12 A It was a contemporary 13 analysis in the evolving field of meta analytic 14 statistical analyses that our statistical group 15 felt was the most conservative and favored 16 analysis to use in meta analyses at that time. 17 Q What is it, can you 18 explain it? 19 A No, I'm sorry, I'm not a 20 statistician. 21 Q What method -- 22 statistical method was used to analyze the OUS 23 data? 24 A To the best of my 238 1 knowledge, there was not an exclusive method used 2 for the OUS data only. 3 MR. SMITH: All right. 4 (TOLLEFSON EXHIBIT NO. 5 MARKED FOR 5 IDENTIFICATION.) 6 Q (BY MR. SMITH) Look at 7 Exhibit 5. Specifically, Doctor, look at that 8 fluoxetine and suicidal ideation article that you 9 wrote and then Doctor Teicher and associates 10 replied. You don't need to read that first 11 article, "Limitations on Fluoxetine" by Doctor 12 Hierholzer. 13 A Okay. 14 Q All right. What Exhibit 15 No. 5 is, is it not, Doctor, is an excerpt from 16 the American Journal of Psychiatry in December 17 1990 which reflects -- what is it, letters to the 18 editor by both yourself and Doctor Teicher? 19 A Correct. 20 Q And you've written to the 21 editor of the American Journal of Psychiatry 22 expressing your views, correct? 23 A Yes, I did. 24 Q And Doctor Teicher 239 1 responded to your letter? 2 A That's correct. 3 Q What is it, do the 4 publishers of these publications, when they 5 receive a letter such as yours, forward it on to 6 the author of the original publication or the 7 original paper for their comments? 8 A Yes. 9 Q And that's why they 10 called it Doctor Teicher and Associates Reply, I 11 guess? 12 A That's correct. 13 Q Down at the -- and you 14 are a Ph.D. psychopharmacologist as well MD 15 psychiatrist, are you not? 16 A Correct. 17 Q Doctor Teicher made some 18 statements at the end of the first paragraph of 19 his reply. Do you see where it says: First, the 20 presumed neurochemical selectivity of fluoxetine 21 is illusory? 22 A Yes. 23 Q Then he goes on to say: 24 Yes, this agent binds specifically to the 240 1 serotonin uptake port, but in doing so, it affects 2 serotonin turnover and postsynaptic receptors 3 located on other neurotransmitter systems. The 4 net effect is alteration in a number of 5 neurotransmitters, including dopamine and 6 norepinephrine. Correct? 7 A That's what he says. 8 Q All right. Do you agree 9 with that, sir? 10 A No. 11 Q All right. What is it 12 that -- in that statement that you disagree with? 13 A Well, first of all, the 14 first statement is not even referenced, so I'm not 15 sure what scientific basis Doctor Teicher and 16 colleagues have for the first statement. The 17 second statement that implicates dopamine with the 18 Baldessarini, et al., group has actually been 19 retested by that same group and they withdrew 20 their first observation, so it's an incorrect 21 observation. 22 Q Now I don't understand. 23 I didn't hear what you said about some group had 24 done something? 241 1 A Well, the second sentence 2 that you referred me to says the net effect is 3 alterations in a number of neurotransmitters, 4 including dopamine and norepinephrine. 5 Q Right. 6 A I was commenting that the 7 dopamine reference is by a group, Baldessarini, et 8 al., and that group has subsequently published 9 additional work saying that there is not an 10 observable affect with SSRI, serotonin uptake 11 inhibitors, and dopamine. 12 Q All right. So are they 13 saying that originally they thought there was 14 evidence that serotonin might effect levels of 15 norepinephrine -- 16 A Dopamine. 17 Q Dopamine, but they've 18 reanalyzed it and now they indicated that's not 19 correct? 20 A That is true. 21 Q What is that new 22 publication? 23 A I don't have the 24 citation. The same authors, and it was a follow-up 242 1 article to this Reference No. 2 that's cited here. 2 Q Was it published before 3 Doctor Teicher's reply in 1990? 4 A I don't know. 5 Q So could it have been 6 that as far as the citations presented by Doctor 7 Teicher, that what he said was accurate in 8 December of 1990? 9 MR. FREEMAN: Could have 10 been that the cite was accurate, because he's 11 already said -- 12 MR. SMITH: Yes, and that 13 there wasn't any evidence that they had withdrawn 14 it at the time that he made that statement. 15 A That's possible. 16 Q All right. Is it true 17 that fluoxetine binds specifically to the 18 serotonin uptake port, but in doing so it affects 19 serotonin turnover and postsynaptic receptors 20 located on other neurotransmitter systems? 21 A I'm not aware of any 22 definitive research saying that fluoxetine 23 caused -- or occupies postsynaptic receptors. 24 Q You say you're not aware 243 1 of any research that says fluoxetine occupies 2 postsynaptic receptors? 3 A That's correct. 4 Q In any system? 5 A That's correct. 6 Q Do you agree that his 7 statement -- do you agree to his statement: The 8 net effect is alteration in a number of 9 neurotransmitters, norepinephrine. If you omit 10 dopamine, is it true that fluoxetine affects 11 norepinephrine? 12 A I don't believe that 13 that's been demonstrated either. There is no 14 consensus that serotonin uptake inhibitors have 15 any effect on norepinephrine receptor systems. 16 Q Or dopamine? 17 A Or dopamine. 18 Q Do they have, in your 19 opinion, sir, effect on any other neurotransmitter 20 systems other than serotonin systems? 21 A I know of no evidence to 22 say that they do. 23 Q Okay. Do you know of any 24 evidence to say that they don't? 244 1 A They certainly have been 2 looked at in a variety of different models for 3 neurotransmitters, and they have not been shown to 4 have any substantial or significant effects on 5 other neurotransmitter systems, thus the term 6 selective serotonin uptake inhibitors. 7 Q Well, how do you define 8 substantial or significant? 9 A Anything that's 10 replicated, a consistent observation, one held by 11 a significant minority or greater of the 12 scientific audience. 13 MR. SMITH: All right. 14 (TOLLEFSON EXHIBIT NO. 6 MARKED FOR 15 IDENTIFICATION.) 16 Q (BY MR. SMITH) Can you 17 identify Exhibit No. 6, Doctor? 18 A I can't personally 19 identify it, the title on the document is Eli 20 Lilly and Company Spontaneous Reports of Adverse 21 Events, in parens Postmarketing, Through December 22 31, 1992. 23 Q All right. Is this 24 something generated in-house at Lilly? 245 1 A I don't recognize this 2 specific document, so I can't answer that. 3 Q Well, it was produced to 4 us as being out of your files. 5 A Mine, as in my personal 6 files? 7 Q Well, your files that you 8 maintained at your office at Lilly. Aren't you 9 turning your files over periodically to the legal 10 department? 11 A Sure, as many others are. 12 Q Yes, and this has been 13 produced to us as something out of your files? 14 A Okay. I would still 15 stand, I don't recognize it, but I'll assume that 16 it was produced. 17 Q That's understandable, 18 but you don't dispute that this is something that 19 was produced in-house at Lilly? 20 MR. FREEMAN: If you 21 know. 22 A I don't know. 23 Q Well, it's got a Pz 24 number on the lower left-handed corner. 246 1 A Uh-huh. 2 Q Which indicates that it 3 is something that's been maintained at Eli Lilly 4 and Company. 5 A Sure. 6 Q And marked for 7 identification. 8 A That doesn't tell me, 9 though, the source or the origin of the document, 10 it just says where it was housed. 11 Q Well, it says Eli Lilly 12 and Company Spontaneous Reports of Adverse 13 Events. 14 A Uh-huh. 15 Q Does Eli Lilly and 16 Company maintain data concerning spontaneous 17 reports of adverse events? 18 A Yes, it does. 19 Q Do they maintain data of 20 postmarketing adverse events reported to them 21 spontaneously for different periods of time? 22 A Yes. 23 Q It says here estimated 24 worldwide population patients exposed, eight 247 1 million, seven hundred and eighty-nine thousand, 2 correct? 3 A Yes. 4 Q Do you have any reason to 5 doubt the accuracy of the information reflected on 6 these tables? 7 MR. FREEMAN: As of what 8 date? 9 MR. SMITH: As of 10 December 31st, 1992, the date reflected on the 11 exhibit. 12 A I'm sorry, I guess not 13 knowing the exact origin, not knowing if this 14 specifically was validated, I can't speak to its 15 authenticity or its accuracy. I don't know if 16 this is a -- I don't know, A, that this is indeed 17 a report generated at Lilly. If so, if it had 18 been authenticated and validated as accurate, 19 because there is no -- 20 Q How would it have to be 21 authenticated and validated as accurate? 22 A It would have had a 23 review by the statistical group perhaps, or might 24 have had a review in this case by the drug 248 1 epidemiology network to validate the numbers. I'm 2 just saying I don't see a validation on here to 3 know this is a final validated document, so I 4 can't speak to that. 5 Q Well, have you seen 6 documents that are final validated documents? 7 A Well, our annual report 8 is a validated document that goes into the Food 9 and Drug Administration that summarizes 10 spontaneous adverse events. 11 Q Well, could this be part 12 of the annual report? 13 A It could be. 14 Q You were working for 15 Lilly in December 1992, weren't you? 16 A Yes, sir. 17 Q This is from your file. 18 Take my word for it. 19 A Fine, I believe you. 20 Q Do you know why it would 21 be in your file? 22 A Because I received it and 23 it was relating to Prozac potentially. 24 Q All right. Had you 249 1 received any data that was inaccurate with respect 2 to reflecting postmarketing adverse events while 3 you have been at Lilly? 4 A No, but all documents are 5 always validated for the accuracy of numbers 6 before -- 7 Q How do you validate it? 8 I mean, this is the first time this has ever come 9 up, Doctor Tollefson, with respect to numbers 10 reflected in documents that have been provided to 11 us. Now, if you're going to question that these 12 numbers are accurate, I need to know that so I can 13 make sure I'm not looking at unreliable or 14 unvalidated numbers. 15 A I'm not questioning their 16 accuracy, I'm just saying I don't know personally 17 that they've been validated. 18 Q Do you require some kind 19 of validation before you rely on numbers? 20 A Yes. 21 Q Who do you require to 22 validate your numbers? 23 A It depends on what the 24 numbers are. 250 1 Q Numbers such as this, 2 postmarketing spontaneous reports of adverse 3 events. 4 A The regulatory sciences 5 department, specifically the drug epidemiology 6 group. 7 Q Well, you mean the DEN 8 database? 9 A Individuals that are 10 responsible for that database. 11 Q Well, do you have any 12 reason to believe -- let's just take the first 13 item on this document, abdomen enlarged, 14 twenty-five. Do you have any reason to believe 15 that that number is inaccurate? 16 A No specific reason. 17 Q Do you have any general 18 reason? 19 A I think I tried to state 20 earlier I just wasn't sure the exact source of 21 this data or whether it had been validated. I 22 have no specific reason to believe that it's 23 inaccurate. 24 Q Do you have any question 251 1 about any particular numbers mentioned here? 2 A I don't have any 3 questions, no. 4 Q What do you do to 5 validate particular numbers with respect to 6 postmarketing adverse events? 7 A I don't do that, the drug 8 epidemiologic folks do. 9 Q How do they do it then? 10 A I would presume that if 11 there is a typed document, that they would make 12 sure that the numbers were consistent with their 13 database. 14 Q What would you have to 15 see before you would accept these as consistent 16 with the Lilly database? 17 A For example, the annual 18 report effective through December 31, 1992, where 19 I would put this next to it and compare numbers. 20 If they matched up -- 21 Q This may be part of the 22 annual report. 23 A It may, it may not. You 24 asked me what it would take to validate it, I 252 1 guess holding it against a validated document and 2 comparing numbers. 3 Q See, the problem is I've 4 never seen a validated document. I had always 5 taken these tables supplied by Eli Lilly and 6 Company to us as being accurate with respect to 7 the numbers of, for instance, postmarketing 8 events? 9 A They may very well be. 10 Q All right. But if 11 there's a problem with that, as far as you know, I 12 need to have it identified now. 13 A I'm not aware of any 14 problem. I'm just saying I don't know the source 15 of this. 16 Q Well, let me ask you 17 this: If you've got some question about the 18 validation of this and whether or not these 19 numbers are accurate, could you have, since you 20 have -- you are in charge of regulatory sciences, 21 are you not? 22 A No, I'm not. 23 Q What's your title? 24 A Executive Director, 253 1 Clinical Investigation and Regulatory Affairs. 2 Q Wouldn't that fall under 3 regulatory affairs? 4 A No, it does not. 5 Q Isn't the spontaneous 6 reporting of adverse events a regulatory affair, 7 Doctor Tollefson? 8 A It's the regulatory 9 science group, which is a different division. 10 Q Well, could you get this 11 validated for us? I mean, I'm interested to know, 12 Doctor Tollefson, if these numbers are accurate or 13 inaccurate? 14 A I know of no reason why 15 it couldn't be validated. 16 MR. SMITH: Do I need to 17 get a stipulation or do I need to prove it up? 18 I'm directing that to counsel. 19 MR. FREEMAN: I don't 20 know. 21 MR. BRENNAN: I don't 22 know what you want. This witness has said -- you 23 haven't been able to lay a foundation with this 24 document. I guess you can ask him about it, but 254 1 he says he doesn't know it, has never seen it 2 before and doesn't recognize it as a Lilly 3 document. I don't see what more we can do for you 4 sitting here about that document. 5 MR. SMITH: Well, the 6 problem is that it purportedly came from his file. 7 MR. BRENNAN: Lots of 8 things came from his file. Apparently so did 9 published material from other magazines. That 10 doesn't stand as an authentication of anything. 11 We produced what we were directed to produce. 12 A I could get something in 13 the mail and my secretary could put it in a file. 14 That doesn't guarantee I have seen it. 15 Q Oh, I understand that. 16 But you normally maintain things in your file even 17 though you -- I can understand that you haven't 18 seen this document. 19 A Uh-huh. 20 Q I'm just wondering if I 21 should question any particular number on any 22 particular page as reflected here. 23 MR. FREEMAN: He says he 24 doesn't know, so, I mean, that seems to me to be 255 1 the end of that. 2 MR. SMITH: Is that 3 correct you, you don't know whether any of these? 4 MR. FREEMAN: I don't 5 know. If I knew, I would tell you. 6 Q (BY MR. SMITH) Who has 7 access to the number of adverse events for a 8 particular body system and the nervous system at 9 any particular time? 10 A Relative to spontaneous 11 event reporting? 12 Q Yes. 13 A The drug epidemiology 14 group. 15 Q Who currently is the head 16 of the drug epidemiology group? 17 A Max Talbott. 18 Q Uh-huh. And as far as 19 you know, is the Lilly computer system able to 20 reflect how many reports of chest pain, for 21 instance, there is up through a particular period 22 of time? 23 A Yes. 24 Q How are they able to do 256 1 that? 2 A It's just a running total 3 of all those that are reported. 4 Q Because you keep -- Lilly 5 keeps records of those reports, correct? 6 A Exhaustive, extensive 7 records, yes. 8 Q And they keep it on their 9 computer, correct? 10 A They're computerized, 11 yes. 12 Q Is there anybody outside 13 of Lilly that has access to the DEN database? 14 A Not access to the 15 database, no. 16 Q Doesn't Lilly report 17 spontaneous events -- adverse events to the FDA 18 based on information collected in their DEN 19 database? 20 A Yes. 21 (TOLLEFSON EXHIBIT NO. 7 MARKED FOR 22 IDENTIFICATION.) 23 Q Exhibit 7 appears to be a 24 Lilly internal correspondence dated October 5th, 257 1 1992, correct? 2 A Correct. 3 Q It's to you from Phyllis 4 Donahue, correct? 5 A Correct. 6 Q Is that correct? 7 A Yes. 8 Q It's year-to-date DEN 9 suicide reports versus patient exposure, correct? 10 A It lists both factors, 11 yes. 12 Q Now, is this something 13 that is more validated, in your opinion, 14 concerning the accuracy of figures? 15 A Than? 16 Q Than Exhibit 6. I mean, 17 here's a report of a specific individual reporting 18 to you specific numbers on a specific particular 19 subject, correct? 20 A I would assume these are 21 valid numbers. 22 Q By virtue of you know 23 Phyllis Donahue? 24 A Correct. 258 1 Q And you asked her to do a 2 job? 3 A Uh-huh. 4 Q And you feel reasonably 5 sure that she's going to use her best efforts to 6 secure these figures, correct? 7 A Yes. 8 Q And that she's going to 9 do it in the best method available to her, 10 correct? 11 A That's correct. 12 Q And you know you got a 13 DEN system and a DEU division that makes an effort 14 to keep accurate reports of adverse events -- 15 A Yes, sir. 16 Q -- correct? And here's 17 an instance where she's reporting that to you? 18 A Yes. 19 Q All right. And she's 20 reporting to you in 1991 there were reports of one 21 thousand, five hundred and twenty-five suicides 22 and suicide attempts and suicide ideations, is 23 that correct? 24 A In 1991 or through 1991, 259 1 that's correct. 2 Q And then as of October 3 1st, 1992 there were five hundred and forty 4 reports in the DEN system? 5 MS. ZETTLER: Additional. 6 Q Correct? 7 A I heard two different 8 comments, which one am I correcting? 9 Q She's just jumping ahead 10 of me as to my next question. 11 A My interpretation is that 12 between January 1 and October 1 of 1992 there were 13 five hundred and forty. 14 Q Reports? 15 A Reports. 16 Q In addition to the one 17 thousand, five hundred and twenty-five that had 18 been received in 1991? 19 A Probably. 20 Q All right. Now, 21 Ms. Donahue -- and who is Ms. Donahue? 22 A She's a clinical 23 associate who works in the DEU, the epidemiology 24 unit. 260 1 Q All right. And it also 2 mentions an Anita Clark? 3 A A similar position. 4 Q All right. She says -- 5 Ms. Donahue says in 1991 there were six hundred 6 and seventy-nine thousand total US continuing 7 patients still on fluoxetine therapy at the end of 8 31 December 1991, correct? 9 A Uh-huh. 10 Q Is that a yes? 11 A Yes, sir. 12 Q Where would she have 13 gotten that information that there were, in 1991, 14 at the end of December 1991, there were six 15 hundred and seventy-nine thousand total US 16 continuing patients, and by that means -- she 17 means patients still on fluoxetine therapy? 18 A Probably from a database 19 survey. 20 Q And how would that 21 database survey work? 22 A There are outside 23 agencies who collect information on pharmaceutical 24 prescribing practices, and results of those 261 1 external surveys are available. 2 Q All right, who were some 3 of those outside sources that do that survey work? 4 A I'm not directly involved 5 with that, but one group that I know has been used 6 is called the IMS. I don't know specifically what 7 IMS stands for, but it does provide prescription 8 information. 9 Q All right. And then is 10 it your testimony that Lilly collects this from 11 maybe sources other than and including IMS? 12 A I'm not sure, collects 13 what? 14 Q Collects data on how many 15 people are taking Prozac at any particular time or 16 during any particular period of time. 17 A It's one of the few 18 sources to have an accurate estimate of the number 19 of patients taking medication, yes. 20 Q Is IMS? 21 A Services such as IMS. 22 Q That's what I'm asking, 23 is are there other services other than IMS that 24 provides this information to drug companies? 262 1 A Not working in that area, 2 I'm not aware of any additional, but it would not 3 surprise me if there are. 4 Q And you say that then is 5 entered into the Lilly database, those particular 6 numbers? 7 A No, I did not say that. 8 Q Where did Ms. Donahue get 9 this six hundred and seventy-nine thousand 10 figure? I thought you said we maintain it in our 11 database based on information provided to us from 12 IMS and other sources. 13 A I don't think I said 14 that. If I did, it was in error. 15 Q All right. 16 A We do not merge any of 17 this data into the DEN database. This is data 18 that is available separately and it was something 19 specifically I had requested. But there's not a 20 merger of adverse events, for example, with some 21 other database. 22 Q All right. All right. 23 So you specifically requested from Ms. Donahue 24 that in addition to finding out the number of 263 1 records of suicidality, that she also find out the 2 number of patients taking Prozac during those same 3 periods of time, is that right? 4 A That's right, I was 5 testing a hypothesis. 6 Q All right. And then she 7 gave you numbers also as of August 31st, 1992 of 8 US continuing patients totaling seven hundred and 9 twenty-seven thousand, correct? 10 A Correct. 11 Q Do I take it from that, 12 Doctor Tollefson, that she's saying according to 13 these IMS estimates that on December 31st there 14 were six hundred and seventy-nine thousand 15 patients who were continuing to take Prozac, 16 correct? 17 A Yes. 18 Q But by August 31st, 1992 19 that number had grown to seven hundred and 20 twenty-seven thousand? 21 A That's my interpretation 22 of it. 23 Q Well, did you ask her 24 what these numbers meant or did you know what 264 1 these numbers meant when you got them? 2 A Exactly what we just 3 said. 4 Q They are the number of 5 patients who are at that particular time on 6 continuing Prozac therapy, right? 7 A They're estimates. I 8 wouldn't want to misrepresent them as exact 9 numbers, but they're estimates. 10 Q Right, and you were using 11 those estimates to test a hypothesis? 12 A Yes. 13 Q And did you rely on those 14 estimates in testing your hypothesis? 15 A Yes. 16 Q Have you seen any other 17 more up-to-date figures on how many US continuing 18 patients there are with respect to Prozac? 19 A No, not continuing 20 patients. Exposures, yes; continuing patients, 21 no, I don't think I've seen a more recent number. 22 Q What was your reasoning 23 for requesting data or numbers of continuing 24 patients as opposed to just exposure, Doctor 265 1 Tollefson? 2 A I didn't specifically 3 request continuing patients, that's the way the 4 data was presented to me. 5 Q Okay. 6 A I'm not sure what the 7 definition is of continuing patients. I think my 8 understanding of it was at that point in time 9 those are the numbers of patients with exposures. 10 Q With exposures? 11 A (WITNESS MOVES HEAD UP 12 AND DOWN.) 13 Q I don't get that, because 14 if you compare that with Exhibit 6, total 15 worldwide population exposed is eight million, 16 seven hundred and eighty-nine thousand just two 17 months later. 18 A That's a cumulative 19 number. 20 Q Right. 21 A Worldwide cumulative. 22 Q Okay. So even though 23 there may have been, as of December 31, 1992, 24 almost nine million patients cumulatively who have 266 1 been exposed to Prozac, is what Ms. Donahue is 2 telling you in this memo that as of December 31st, 3 1991, in the United States there were six hundred 4 and seventy-nine thousand who were on continuing 5 Prozac? Do you see my -- where I'm going? 6 A I see what you're 7 saying. My understanding was that those were the 8 number of patients taking Prozac during that year. 9 Q All right. Taking one 10 particular Prozac pill or had maybe filled more 11 than one prescription? 12 A Probably who had refilled 13 at least one prescription, but I don't know that. 14 Q Who would know that; 15 would Ms. Donahue know that? 16 A Possibly. 17 Q What hypothesis were you 18 testing? 19 A Well, I was interested in 20 the possibility that media exposure and the 21 newness of fluoxetine in the marketplace were 22 factors that were influential in the reporting 23 rate of suicidal ideation, acts or attempts, 24 because those phenomena have been reported by 267 1 others to influence spontaneous event reporting. 2 So in order to look at that as even a concept, I 3 was interested in, as a numerator, the number of 4 acts or ideations in '91 versus year to date 5 in '92, but with a denominator of how many 6 patients were taking fluoxetine, because it was my 7 assumption that the prescribing prevalence of 8 fluoxetine had gone up from '91 to '92, and it was 9 my understanding that the number of events, as is 10 supported in the document had gone down, which 11 would suggest the ratio was decreasing. 12 Q So, the number of 13 prescriptions for Prozac appears to have gone up 14 from '91 to '92, based on this data? 15 A Right. 16 MR. SMITH: Okay. You 17 wanted to break? 18 MR. FREEMAN: If you do. 19 (SHORT BREAK TAKEN.) 20 (TOLLEFSON EXHIBIT NO. 8 MARKED FOR 21 IDENTIFICATION.) 22 Q (BY MR. SMITH) Doctor 23 Tollefson, can you identify Exhibit 8 for us, 24 Doctor? 268 1 A It's a messenger sent by 2 me August 12, 1991 to a number of Lilly personnel. 3 Q Who were you somewhat put 4 out with? 5 A Doctor Beasley. 6 Q All right. And what was 7 the reason that you were put out with Doctor 8 Beasley? 9 A Failure to have 10 communicated a decision to me. 11 Q What was the decision 12 that Doctor Beasley made that you wanted some 13 input into? 14 A I had had input into it, 15 we had reached a consensus. The decision that 16 came forth was one opposite from the group's 17 consensus and I was trying to understand the 18 process. 19 Q Okay, fill me in on what 20 the decision was. 21 A This had to do with the 22 meta analysis of the fluoxetine US IND database, 23 and we were looking at two different statistical 24 methods for conducting the meta analysis: One, 269 1 the chi-square, the second the Mantel-Haenszel 2 Cochran that we talked about earlier. 3 Q All right. And both of 4 them were used in the presentation to the FDA 5 Advisory Committee meeting, were they not? 6 A Yes. 7 Q And were not both of them 8 used in the meta analysis? 9 A Meta analyses with each 10 technique were conducted, yes. 11 Q All right, but you didn't 12 like it being conducted under the chi-square 13 analysis? 14 A No, that had been the 15 analysis that the group had agreed upon we would 16 use for the publication. 17 Q But then a different 18 method was used? 19 A The second method, yes. 20 Q And was that -- the 21 second method -- 22 A The Cochran 23 Mantel-Haenszel. 24 Q Was that the one that was 270 1 finally used at publication? 2 A Yes. 3 Q And what was your 4 complaint about use of that method? 5 A I didn't really have a 6 complaint about use of the method. In fact, it's 7 a much more conservative method than the 8 chi-square. I was concerned that the group, 9 particularly driven by Charles at that point, had 10 made a change in direction from what we had spent 11 a fair amount of time discussing as a group and 12 agreed on using the chi-square. 13 Q What were the pros for 14 the chi-square versus the cons of the chi-square? 15 A Both were acceptable 16 statistical methods. The chi-square is one that 17 had been around longer, people were more familiar 18 with it. The Cochran Mantel-Haenszel was a fairly 19 new technique, as I said much more conservative 20 technique for conducting a meta analysis, and for 21 some of our scientifically oriented statisticians 22 felt that sort of for cutting edge meta analysis 23 it was perhaps then a favored technique because it 24 was more conservative. 271 1 Q Did the use of the method 2 alter the differences in the statistical data? 3 A No, not -- 4 Q Or the interpretation, 5 did it alter the interpretation of the data? 6 A No, not really. 7 Q Well, did it some? 8 A No. It depends on what 9 question you're posing. Both analyses showed an 10 absence of an association between fluoxetine and 11 treatment emergent suicidality. The chi-square, 12 being a somewhat more liberal analysis, actually 13 showed some trend towards a positive protective 14 effect of Prozac. 15 MR. SMITH: All right. 16 (TOLLEFSON EXHIBIT NO. 9 MARKED FOR 17 IDENTIFICATION.) 18 Q (BY MR. SMITH) I believe 19 it was in Exhibit 3 or 4, your letter of March 20 3rd, before you joined Beasley -- joined Lilly, 21 you had written Doctor Zerbe -- 22 MR. SMITH: May 3rd. 23 Q May 3rd, and expressed to 24 him the fact that you felt the rechallenge 272 1 protocol was scientifically acceptable but you 2 didn't think it was practical, is that correct? 3 A That was my point at that 4 time, yes. 5 Q And on May -- what is the 6 date of that letter to Doctor Zerbe? 7 A May 3rd. 8 Q Were you aware that as 9 Exhibit 10 -- no, 9 reflects that on May 15th 10 there was a meeting with Doctor Zerbe and others 11 at the Food and Drug Administration to discuss 12 fluoxetine rechallenge protocol? 13 A No. That would have been 14 prior to my joining the company. 15 Q I understand that. Were 16 you aware that Lilly had agreed on May 13th, 1991 17 to proceed with a rechallenge study? 18 A No, I was not aware of 19 that. 20 Q Were you aware that in 21 bullet point one, two, three under one, that, 22 quote, we, meaning Lilly, agreed to have the 23 rechallenge protocol ready to go by September 1st, 24 1991 and provide data after the first quarter 273 1 which would provide information on six months of 2 experience, end quote? 3 A No, I wasn't aware that 4 there was an agreement. 5 Q All right. And the 6 rechallenge protocol wasn't ready to go by 7 September 1st, 1991, was it? 8 A There was a protocol, 9 yes, there was a protocol written. 10 Q Was it ready to go by 11 September 1st, 1991? 12 A Yes. 13 Q And was there data after 14 the first quarter which would provide information 15 on six months of experience? 16 A I'm not sure what the 17 reference here is to as far as what type of 18 information on six months of experience. I'm not 19 sure what the information is being referred to, 20 what kind of information. 21 Q Do you disagree that it 22 would be a reasonable interpretation of that 23 sentence that the rechallenge protocol would be 24 put into effect and that patients would be 274 1 enrolled and that data would be provided for the 2 first six months of experience? 3 A That could be an 4 interpretation, yes. 5 Q That didn't occur, did 6 it? 7 A Conduct of the study? 8 Q Yes. 9 A That's correct. 10 Q Is this the first time 11 you've seen that document? 12 A This is the first time 13 I've seen this document. Whether or not there was 14 any agreement or commitment to the FDA to conduct 15 this study was a point of contention. 16 Q A point of contention 17 between who? 18 A Whether or not there 19 actually was an agreement or a request to conduct 20 the study as opposed to discussing it in concept. 21 Q Well, who was in 22 contention about that? 23 A Various individuals 24 within the medical organization. 275 1 Q All right, some people 2 within the medical organization were saying that 3 you had agreed -- that Lilly had agreed to do a 4 rechallenge study and get it underway and others 5 were saying we hadn't really agreed to do that, we 6 didn't make that kind of commitment to the FDA? 7 A Nor had we been asked to 8 make that kind of commitment. 9 Q Who was on the pro and 10 who was on the con part of that? 11 A Doctor Zerbe was on the 12 side of believing that there had been a 13 commitment. Regulatory people in attendance at 14 the meeting such as Doctor Webber did not believe 15 that they had heard that commitment or a request 16 for such a commitment. 17 Q Reading this document, 18 one would believe that there had been that kind of 19 commitment though, wouldn't they, and that Doctor 20 Zerbe may have been right by virtue of Doctor 21 Kotsanos's characterizations of what occurred at 22 the meeting? 23 A At that point in time. 24 Q Yes, so would my answer 276 1 be yes? 2 A If this is an accurate 3 recollection of the meeting, then yes. 4 MR. SMITH: All right. 5 (TOLLEFSON EXHIBIT NO. 10 MARKED FOR 6 IDENTIFICATION.) 7 Q (BY MR. SMITH) While 8 they're looking at that, do you recall how it was 9 finally resolved, whether or not there had 10 actually been a commitment and whether there had 11 actually been a request by the FDA? 12 A Since the majority of 13 individuals felt that there had not been a formal 14 request and we had not heard anything from the FDA 15 during discussions about continuing or progressing 16 in a rechallenge study -- 17 Q So it just was never 18 done? 19 A The rechallenge study? 20 Q Right. 21 A There were -- there was a 22 review of the protocol that was written and some 23 very serious concerns about the ability to conduct 24 such a study, the ethics, the scientific 277 1 creditability of such a study where it was felt 2 that scientifically that particular design or that 3 concept would not yield meaningful or valid 4 results. 5 Q Well, was there ever a 6 formal decision made not to do the rechallenge 7 study, or did it just sort of drop? 8 A It was decided not to do 9 the study. 10 Q When was that decision 11 made? 12 A Sometime in the fall 13 of '91. 14 Q After the PDAC? 15 A I believe so. 16 Q And who made that 17 decision, sir? 18 A A group decision of 19 medical management. 20 Q Okay, who was in the 21 medical management group that made that decision? 22 A It would have been made 23 by a group -- myself, Doctor Zerbe, Doctor 24 Weinstein presumably, possibly Doctor Thompson. 278 1 Q Did anyone vote to 2 proceed with the rechallenge study in that group? 3 A No one voted to move 4 forward with the study. I think that Doctor Zerbe 5 raised the question of whether or not the FDA 6 expected such a study. 7 Q Did anyone ever formally 8 notify the Food and Drug Administration that the 9 rechallenge study would not be done? 10 A I don't know. 11 Q All right. Now look at 12 Exhibit 10. Exhibit No. 10 appears to be a 13 document dated July 2nd, 1991 authored by you, is 14 that correct, sir? 15 A Correct. 16 Q Directed to Mr. E.A. 17 West? 18 A Correct. 19 Q Who is Mr. E.A. West? 20 A Mr. West is a director in 21 corporate affairs. 22 Q And it's "Re: Louisville 23 Slugger," is that correct? 24 A Yes. 279 1 Q To what does Louisville 2 Slugger refer to? 3 A The newspaper. 4 Q Which newspaper? 5 A I don't remember if it's 6 the Courier -- I don't remember, one of the papers 7 in Louisville that had carried an account on 8 suicide and depression. 9 Q All right. Was the name 10 of the paper the Louisville Slugger? 11 A No. 12 Q Was there a study known 13 as the Louisville Slugger study? 14 A A study, no. 15 Q Was there a protocol by 16 the name of Louisville Slugger? 17 A No. 18 Q Was the article in the 19 Louisville paper about the Louisville Slugger? 20 A No. 21 Q Why do you use the term 22 Louisville Slugger here? 23 A It's the trade name for a 24 bat that's made in Louisville and Ed and I were 280 1 talking about sports and baseball, so it was just 2 picking out as a humorous introduction to a 3 letter. 4 Q But the subject of the 5 letter is not very humorous, is it, Doctor 6 Tollefson? 7 A No, no, it's very 8 serious. 9 Q The subject is suicide, 10 isn't it? 11 A Yes. 12 Q The subject is depression 13 suicide interface? 14 A Correct. 15 Q One subject is 16 perspective on tricyclic poisoning? 17 A Correct. 18 Q And it talks about 19 adverse events on fluoxetine versus conventional 20 antidepressants, correct? 21 A Correct. 22 Q Was this right? 23 A That is correct. 24 Q So what is the humor in 281 1 this? 2 A I don't think there's any 3 humor in the topic. The humor was in our common 4 interest. 5 Q Well, why would you refer 6 to these things in a common instance in baseball? 7 A Referring to the 8 newspaper. 9 Q The newspaper, why were 10 you referring to the Louisville newspaper? Was 11 the Louisville newspaper in July 1991 doing an 12 article on suicide or depression? 13 A I don't recall 14 specifically, but I would imagine it had something 15 to do either with an article or an inquiry. 16 Q An inquiry by the 17 Louisville newspaper? 18 A Possibly. 19 Q Do you recall doing 20 this -- writing this memo? 21 A Yes. 22 Q Could the subject of this 23 be Joseph Wesbecker? 24 A I don't think so. 282 1 Q Well, Joseph Wesbecker 2 lived in Louisville, Kentucky, didn't he? 3 A That may have been the 4 reason for the interest by the paper. 5 Q You see, I don't see any 6 reference to sports or bats in the subject of your 7 memo, Doctor Tollefson. 8 A Uh-huh, that's correct. 9 Q Could you be referring 10 to, when you use the term Louisville Slugger, the 11 tragedy that occurred at the Standard Gravure 12 plant on September 14th, 1989? 13 A Not specifically. 14 Q Were you generally? 15 A No. 16 Q You knew at the time that 17 there was litigation involving what occurred in 18 Louisville, Kentucky in September with Joseph 19 Wesbecker, did you not? 20 A I'm not sure that I was 21 aware there was litigation. I was aware of an 22 event that occurred there. 23 Q And you knew that event 24 was alleged to be related to Mr. Wesbecker being 283 1 on Prozac, don't you? 2 A Yes. 3 Q And you knew at the time 4 that eight people had lost their lives at the 5 Standard Gravure printing plant which was adjacent 6 to the Louisville Courier-Journal, did you not? 7 A No. 8 Q You knew eight people had 9 lost their lives in this incident in Louisville, 10 Kentucky, didn't you? 11 A Yes. 12 Q And that others had been 13 severely injured, didn't you? 14 A I was not aware of the 15 injuries. 16 Q Did you find any of that 17 situation as humorous? 18 A No, but that situation 19 didn't directly relate to this note. 20 Q I don't understand why 21 you characterized this entire note of suicide and 22 depression and tricyclic poisoning under the term 23 Louisville Slugger, Doctor Tollefson. Can you 24 explain that? 284 1 A No, it was just an 2 association that I had with Louisville. 3 Q Had you referred to -- 4 had you heard Mr. West refer to the situation at 5 Louisville as the Louisville Slugger incident? 6 A No, I had not. 7 Q Have you ever referred to 8 it, other than this memo, if you did, as the 9 Louisville Slugger incident? 10 A I didn't refer to it in 11 the memo or previous or subsequent as such. 12 Q Was the term Louisville 13 Slugger in the article that had been published or 14 was going to be published? 15 A I don't know if there was 16 an article to be published, but I would presume 17 not. 18 Q Did you and Mr. West have 19 some discussion concerning this prior to you 20 writing the memo? 21 A No, other than his 22 interest in the information contained. 23 Q All right. Well, did he 24 have an interest in Louisville Slugger bats? 285 1 A No, I said we had a 2 common interest in sports, and he had indicated to 3 me that he had -- I don't remember the specifics, 4 an inquiry or something from a Louisville paper, 5 or a correspondent at such a paper, and he was 6 asking me for some scientific information on 7 suicide, the depression interface, and adverse 8 events. 9 Q And so you sent it back 10 to him under the caption Louisville Slugger? 11 A Yes, not remembering the 12 name of the paper or anything specifically about 13 it. 14 MR. SMITH: Are you ready 15 to quit? 16 MR. FREEMAN: Yes. 17 MR. SMITH: Yes, but I 18 don't think we're going to come back tomorrow. I 19 think we're ready to stop with Doctor Tollefson, 20 if that's all right with you all? 21 MR. FREEMAN: Well, I 22 won't miss you. 23 MR. SMITH: I want it 24 known that I'm giving you back a day. 286 1 MR. FREEMAN: We 2 appreciate it. 3 (WITNESS EXCUSED.) 4 * * * * * 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 287 1 STATE OF KENTUCKY ) ) SS. 2 COUNTY OF JEFFERSON ) 3 I, MARY KATHLEEN NOLD, a Notary Public within 4 and for the State at Large aforesaid, do hereby 5 certify that the foregoing is a true, correct and 6 complete transcript of the deposition of GARY D. 7 TOLLEFSON, MD, taken at the time and place and for 8 the purpose set out in the caption hereof; that 9 the witness was duly sworn before giving said 10 deposition; that the said deposition was taken 11 down by me in stenotype and afterwards transcribed 12 on a computer under my direction; that the 13 appearances were as set out in the caption hereof; 14 and that a request was made by counsel that the 15 deposition be submitted to the witness for reading 16 and signature. 17 GIVEN my hand as notary aforesaid, this 18 __________ day of ___________________, 1994. 19 My commission expires March 10, 1998. 20 21 22 _______________________________________ MARY KATHLEEN NOLD 23 COURT REPORTER AND NOTARY PUBLIC STATE OF KENTUCKY AT LARGE 24 288 1 STATE OF___________________) ) SS: 2 COUNTY OF__________________) 3 I, GARY D. TOLLEFSON, MD, do hereby certify 4 that I have read the foregoing deposition given by 5 me on July 26, 1994, and that the answers 6 contained therein are true and correct to the best 7 of my knowledge and belief. 8 9 10 ______________________________ GARY D. TOLLEFSON, MD 11 12 13 ______________________________ (DATE) 14 15 16 Subscribed and sworn to before me this day by 17 GARY D. TOLLEFSON, MD. 18 My commission expires______________________. 19 20 ___________________________________ 21 NOTARY PUBLIC 22 23 ______________________________ 24 289 1 STATE OF____________________) ) SS: 2 COUNTY OF___________________) 3 I, GARY D. TOLLEFSON, MD, do here certify that 4 I have read the foregoing deposition given by me 5 on July 26, 1994, and that the answers contained 6 therein are true and correct to the best of my 7 knowledge and belief, with the following 8 corrections: 9 PAGE/LINE CORRECTION REASON FOR CORRECTION 10 __________________________________________________ 11 __________________________________________________ 12 __________________________________________________ 13 __________________________________________________ 14 __________________________________________________ 15 _________________________________ GARY D. TOLLEFSON, MD 16 _______________________________ (DATE) 17 18 19 Subscribed and sworn to before me this day by 20 GARY D. TOLLEFSON, MD. 21 My commission expires_____________________. 22 ________________________________ NOTARY PUBLIC 23 24 ________________________________ 290