AstraZeneca Cherry-picking Data Email

8/7/2019 AstraZeneca Cherry-picking Data Email

1/9

EXHIBIT 14


2/9

From: Tumas John JASent: 5/11/1999 1:20:48 PMTo: Murray Michael MF; Davies Diane DE MMCCi Lawrence Richard RA;Price Anna AC; Hough Nick NW ; Jones Martin AM - PHMSi Litherland SteveS; Rak Ihor H,; Raniwalla Joher J; Tumas John JACC: Goldstein Jeffrey JMSubject: RE: Small ReviewDear All,Thank you to those who have commented on this review - I will collatethe commentsand pass them on to Dr. Small.A Dr. Small has contacted Jeff andexplainedthat the schedule for this review has been moved up.A. We will now haveto sendour comments to her tomorrow (~iednesday).A Therefore, if anyone has anyfurthercomments, please pass them along.

Best regards,John

From: A Tumas John JASenL: A Wednesday, May as, 1999 9:47 AMTo: A A A Hurray Michael MF; Davies Diane DE - MMCC; Lawrence RichardRA; PriceAnna AC; Hough Nick N~i; Jones Martin AM - PHMS; Litherland Steve Si RakIhorIW; Raniwalla Joher JCc: A . A . A Goldstein Jeff.rey JMSubject: A A. A. A A A Small ReviewImportance: A A A HighDea.!;All,Attached is a dra f.t review of quetiapine by Dr ..Joyce Small.A You mayrecall,Dr. Small was the lead investigator for Trial 8, high dose, low dosequetiapinevs placebo.Dr. Small has asked that we review the document to ensure that the mostcurrentinformation is included.A May I ask that comments be ..limited to thisrequestand any inaccuracies found.A She is not looking fox; editing support andwouldlike to keep Zeneca' s influence on this r evi.ew rrunLrneL.Please return any comments to me by Thursday, May 13.Best regards,John

1 SQ1 ED01906715


3/9

&It;>

2 SQ1ED01906715


4/9

From: Hough Nick NWs ent.: 5/10/1999 9:06:26 AMT o: Murra y Mi.ch ae L t4F; Davies Dd ane DE - l-lHC:-;i L aw re nc e R ic ha rd RA;Price A nna A C; jones Hartin AM - P!-lHS;Litherland Steve 5; Rak Tho r !tv;Ran i.vra Ll .a J ohe r J: 'I'uma s John JA .cer G olds te in J ef frey ~ > 1B e c :Subject: RE; Small ReviewJohn,here are my comments on 'Small', some of ;.!hichreflect my usualconcerns, ie selective use of QUEST, over LookLnq I",hat. bappened in study14 etc; however there are also some onvious positive messages that couldbe added:* elderly data should be based on 52 weeks if possible* selective inclusion of QUEST data and not COSTA R means that thispaper is unsuitable for 'promotio!1al purposes' - this paper goes evenfurther than the visual aid ('data display' approach?) since the author

actually makes a comparative statement - 'with advantages fo.rQTP ondepression ratings and the CGI'* therefore, not 'approvable' for inter national pr omotional purposes selective inclusion of QUEST data is in conflict with our publicationpolicy since we have no plans to de anything with COSTA R study 14 , the head to head comparison against haloperidolunfortunately resulted in a 'p-value' in favour of HA L on the positivesympton scale: therefore it is n01: possible to say that relief ofp osi ti ve s ymp tom s a ppea rs comparab L e t. o s t andazd neur c Lep t.ics* 1st para under 'neurological effects' tends to switch back and forthbetween the HA L and CPZ comparative data and doesn't flow logicallytherefore'" could perha ps i ncl ude the CLOG to U'J'P switch data wrt to weight gain' rev er sa l' / i mpr oved d ia beti c. sy mpto ms ??* need medical check on what is said wtt ECGI Q Tc i nt er va ls* under therapeutic potential, 2nd para - 'studies of theseapplicati ons' ...thi s is wr itt.en a s though somechtnq has been describedimmedi ately pr evi ousl y relevant to thi s state~ent??Hope this.is helpful ,Cheers,Nick> - - - - - - - - - ->From: Tumas John JA>Sent: 05 May 1999 14:47>To: Murray Hichael MF; Davies Diane DE - M.MCC; Lawr erice R ichard Rl'.;Pr ice A nna ACi Hough Nick NW; Jones Martin A M - PHMS; Litherland SteveS; Rak Ihor IW; Raniwalla Joher J>Cc: Goldstein Jeffrey J I > 1>Subject: Small Review> Im po rt an ce : H ig h>>Dear A ll,>>A ttached is a draft review of quetiapine by Dr. Joyce Small. You mayrecall, Dr. Small was the lead investigator for Trial 8, high dose, lowdose quetia pine vs .placebo.

Confldential 1 SQ1 ED00099916


5/9

\" . . '

>>Dr. Small has asked that we review the document to ensure that the mostcurrent information is included. Hay I ask that comments be limited tothis request and any inaccuracies found. She is not looking for editingsupport and would li>ke to keep Zensea's influence on this reviewmir.Jimal.>>Please return any corrunents t.ome by Thursday, fay 13.>> Be st r eg ard s,>>John;,>File: QUETIAPI.NE.doc>

Confidential 2 SQtEQOOO9-99.16


6/9

UnknownFrom:Sent:To:Cc:

Subject:Attachments:

Gavin Jim JPWednesday, December DB , 1999 12:32 PMDe Vriese GeertHoldsworth Debbie D;Tumas John JA;Tugend Georgia GL;Czupryna Michael MJ;GormanAndrew AP;Wilkie Alison AM;Litherland Steve S;Murray Michael MF;Rak Ihor IW;OwensJudith J;O'Brien Shawn SP;Denerley Paul PM;Goldstein Jeffrey JMRE: 2 EPS Abstracts for APAjamapubs.pdf

Thanks for this Geert. If I could add my own thoughts in advance of the GPT tomorrow ...Certainly any progress on the(selective) use of data from COSTAR would be particularly appreciated, as I'm currently getting mixed messages onwhether we use the EPS data from this trial.I was interested to hear that we are discussiing the recent JAMA article on the reporting of Clinicaltrials (link attached).This article concerns me as it highlights what appears to be an increasing scepticism among journal editors with regardsto certain aspects of company-sponsored publications. Janssen have had their fingers burned in the past in this regard,and are consequently cited every time such an editorial appears, something that presumably irritates the hell out of them.Quite apart from any ethical considerations, if they thought we were publishing positive data vs risperidone from QUESTwhile results from a second trial were being buried, they'd be onto it in a flash. Selectively using (for example) the EPSdata from COSTAR is pushing it too far in my opinion, and might prove extremely damaging in the long run (and you canbet Janssen would push it), and would destroy our current high standing in the publishing community.

~ll:Jj amapubs . pd f (112

KB )RegardsJim

From:Sent:To:Subject:FYIFrom:Sent:To:Cc:Subject:Kendra,John,

Owens Judith J08 December 199909:24GavinJimJPFW: 2 EPS Abstracts for APA

DeVriese Geert08 December 1999 08:42Baker Kendra; Tumas John JAScanlon Rose Ann RA; Denerley Paul PM; Owens Judith JRE: 2 EPS Abstracts for APA

REDACTED

From:Sent:To:Cc:Subject:

Baker Kendra07 December 199922:49Owens Judith J; DeVriese GeertTumas John JA; Scanlon Rose Ann RA; DenerleyPaul PMFW: 2 EPS Abstracts for APAPRIVILEGED AND CONFIDENTIAL

1

CONFIDENTIALAZSER 12916364


7/9

Best regards,Ken.drCt / 'BcUc.erAttorneyLegal DepartmentAstraZenecaTel . (302) 886-4233 Fax: (302) [email protected]

From:Sent:To:Subject:

Scanlon RoseAnn RATuesday, December 07, 19992:33 PMBaker, KendraFW: 2 EPS Abstra.cts for APAREDACTED

Rose Ann ScanlonAssistant General CounselAstraZenecaTelephone: 302 886 4009Fax: 302 886 8221

From:Sent:To:Subject:

From:Sent:To:Cc:

Subject:

Denerley Paul PMDecember 07, 1999 10:24 AMScanlon Rose Ann RAFW: 2 EPS Abstracts for APA

Tumas John JAMonday, Deoember 06, 1999 11:45 PMOwens Judith J; Jones Martin AM - PHMS; Litherland Steve S; Gavin Jim JPHoldsworth Debbie D; Tugend Georgia GL; Czupryna Michaet MJ; Gorman Andrew AP; Wilkie Atison AM; Murray MichaelMF; Rak Ihor IW; O'Brien Shawn SP; Denerley Paul PM; Gotdstein Jeffrey JM; Woods Paul PB; Holdsworth Debbie 0; DeVriese Geert; Shadwell Pamela PGRE: 2 EPS Abstracts for APAPlease allow me to join the fray.There has been a precedent set regarding "cherry picking" of data. This would be the recent Vell iganpresentations of cognitive function data from Trial 15 (one of the buried trials). Thus far, I am not aware of anyrepercussions regarding interest in the unreported data.That does not mean that we should continue to advocate this practice. There is growing pressure from outsidethe industry to provide access to all data resulting from clinical trials conducted by industry. Thus far, we haveburied Trials 15, 31, 56, and are now considering COSTAR.The larger issue is how do we face the outside world when they begin to criticize us for suppressing data. One

2

CONFIDENTIALAZSER12916365
mailto:[email protected]:[email protected]


8/9

could say that ourcompetitors indulge in this practice. However, until now, I believe we have been looked uponby the outside world favorably with regard to ethical behavior. We must decide if we wish to continue to enjoythis distinction.The reporting of the COSTAR results wil l not be easy. We must find a way to diminish the negative findings.But, in my opinion, we cannot hide them.Best regards,John

From:Sent:To:cc:

Gavin Jim JPMonday, December 06, 1999 1 :59 PMOwens Judith J; Jones Martin AM - PHMS; Litherland Steve SHoldsworth Debbie 0; Tumas John JA; Tugend Georgia GL; Czupryna Michael MJ; Gorman Andrew AP; Wilkie AlisonAM: Murray Michael MF: Rak Ihor IW; O'8rien Shawn SP: Denerley Paul PM; Goldstein Jeffrey JM: Woods Paul P8;Holdsworth Debbie 0; De Vriese Geert; Shadwell Pamela PGRE: 2 EPS Abstracts for APAUbJect :

Steve's comments are pertinent, as the EPS abstracts (for the APA) and the Scourge of EPS review bothemanate from the eCNP symposium, and as such represent a potential transition of COSTAR data from a"closed" mtg to a public forum. Coming in late to the debate, the only directive Ihave on QUEST/COSTAR(contained in a document compiled by Ihor & Martin inAugust) suggested using them "as clinicallyappropriate", but independently.I believe the newly-formed Commercial Support Team will be considering looking at potential ways of usingCOSTAR. With regards to the present outputs however, a short-term solution (given the impending APAdeadline) is to avoid reference to COSTAR inthe proposed APA abstract. Whether or not we discuss it ineither the poster or the review subsequently will need to decided by the team, with reference to howwewould then need to approach the efficacy story.RegardsJim

Subject:

Lither land Steve S06 December 1999 11 :51Owens Judith J; Jones Martin AM - PHMSHoldsworth Debbie D: Tumas John JA; Tuqend Georgia GL: Czupryna Michael MJ; Gorman Andrew AP; WilkieAlisen AM; Gavin Jim JP: Murray Michael MF; Rak Ihor IW; O'Brien Shawn SP; Denerley Paul PM; GoldsteinJeffrey JM; Woods Paul PB: Holdsworth Debbie D; De Vriese Geer!RE: 2 EPS Abstracts for APA

From:Sent:To:Cc:

Martin has drawn our altention to an enduring problem which requires resolution as soon as possible. should we publish COSTAR? The disadvantages are obvious, not Jeastthat we provide theopposition with potentially damaging data when they calculate p values re the primary efficacyendpoint i f not, can we extract some information and use this to support our messages? The following isscheduled to appear in Clear Vision (proceedings of the ECNP EPS meeting):A second study comparing flexible dosing of risperidone (6-10 mg daily) and quetiapine (300-600mg daily) reported that over 10 weeks significantly more risperidone patients (31.4%) thanquetiapine patients (14.1%)10 my draft 30.4 and 13.1% ; need to check experienced EPS orakathisia (30.4% and 16.6 15.4 in Mk doc%, respectively) (p


9/9

outside the Company (I know that we have had some feedback but I understand this was conflicting anduncoordinated) .Steve

Subject:

Jones Martin AM - PHMS06 December 1999 10:55Owens Judith JHoldsworth Debbie D; Tumas John JA; Tugend Georgia GL; Czupryna Michael MJ; Gorman Andrew AP ;Wilkie Alison AM; Gavin Jim JP; Litherland Steve S; Murray Michael MF; Rak Ihor IW; O'Brien Shawn SP;Denerley Paul PM; Goldstein Jeffrey JMRE: 2 EPS Abstracts for APA

From:Sent:To:Cc:

JudithI have no real comments on the Juncos abstract, but am concemed about Tandon's.In Tandon's results section, he refers to a randomised comparative study. This study is COSTAR.think that we are s1ill not comfortable about communicating the overall results of this study. Whilstthis data may have been presented orally in London, I think this abstract would be the first time wehave put anything 'down on paper'. Are we sure that this we can present the EPS data in isolationgiven the nature of the other results? Will we not create a desire for further information about thestudy? Can we not refer to published (non-comparative) data for risperidone, as we must be doingthis for olanzapine? Should we be looking at the ziprasidone data too? They seem to have dose-response effect as well.Martin

From:Sent:To:

Owens Judith J02 December 1999 17:14Wilkie Alison AM; Gavin Jim JP; Ulherland Steve S; Murray Michael MF; Rak thor IW; Jones Martin AM-PHMS; O'Brien Shawn SP; Denerley Paul PM; Goldstein Jeffrey JMHoldsworth Debbie D; Turnas John JA; Tugend Georgia GL; Czupryna Michael MJ; Gorman Andrew AP2 EPS Abstracts for APAHigh

Cc:Subject:Importance:Dear AllPlease find attached, for your review, 2 EPS abstracts that are intended for submission to APA.The abstracts are based on presentations at the AstraZeneca symposium 'CLEAR VISION - Afresh look at EPS' held during this year's ECNP.Please return any comments you may have by midday (UK time) Monday 6 December.Kind regardsJudi1hFile: Juncos abstracLdocFile: Tandon abstract.docJudith OWensExt: 24164llF34 Mereside

4

CONFIDENTIALAZSER 12916367

Documents

AstraZeneca Cherry-picking Data Email