Part V: Forest Leveraged the False Results of MD-18 to Obtain an Adolescent Indication for Lexapro

Part I | Part II | Part III | Part IV | Part V | Part VI

In 2004, Forest sent a request to the FDA inquiring whether “a positive study with escitalopram using a conventional acute treatment design . . .  along with the previous positive study with citalopram (Study CIT-MD-18) be adequate to support an indication for acute treatment in pediatric patients aged 12 – 17 years?”[257]  Relying on the false claim that MD-18 was positive, Forest wanted to know whether an additional positive study in adolescents would be enough to obtain an adolescent indication for Lexapro.  In response, the FDA stated: “We believe that one additional positive acute treatment study of adolescents in addition to Study CIT-MD-18 would support a claim for the acute treatment of adolescents with MDD.”[258]  Thus, if Forest could obtain a positive adolescent clinical trial for Lexapro, the FDA agreed to give Lexapro an adolescent indication.  This promise, however, was based on a belief that MD-18 was a positive study.  As Dr. Laughren explained:

  1. Okay. All right. So my understanding based on the response from the FDA is that if Forest could produce a positive double-blind, placebo-controlled clinical trial with Lexapro in children aged 12 to 17, it would then agree to provide an indication for Lexapro for that age group.
  2. Yes, that’s — that is what it’s saying. I mean, of course, it would — you know, it would have to be reviewed. It’s subject to review by FDA. But in principle, yes, that is what this letter says.

. . .

  1. Okay. If MD-18 was negative — okay, just assume that for a second — would the FDA have made this agreement? . . .
  2. No. I don’t — I don’t believe so. That would be my impression that – that we would not have — have reached that agreement.[259]

Ultimately, Forest was able to obtain a statistically positive result in MD-32, as discussed above.  Forest then, in 2008, submitted the results of MD-32 and MD-18 as part of a supplemental new drug application seeking an adolescent (12-17) indication for Lexapro.  Since the FDA had already promised to approve the application, its review of the data was barebones.

However, before delving into the FDA’s review of this supplemental application, it is important briefly to discuss the FDA’s original review of MD-18, again.  MD-18 involved both children (7-11) and adolescents (12-17).[260]  However, the study was not powered to measure just children or adolescents.[261]  And, after getting the results, Forest conducted several statistical analyses of the data and stated, repeatedly, that “[s]imilar effects were seen in the children and adolescent subgroups, as evidenced by the lack of a treatment-by-age group interaction[,]” that “the magnitude of the treatment effect was similar in the child and adolescent subgroups[,]” and that “[t]he magnitude of the mean citalopram-placebo differences on the efficacy ratings was numerically higher in the adolescents than in the children, but no significant treatment-by-age group interactions were observed, indicating that the citalopram treatment effect was not age dependent.”[262]  And, this lack of treatment-by-age difference was seen across every primary and secondary endpoint.

Despite this, when Dr. Laughren prepared his memorandum in 2002—when both Forest and the FDA knew the FDA would not be approving a pediatric indication for Celexa[263]—he noted that the observed treatment effect in the study was primarily coming from the adolescent subgroup.[264]  When Dr. Laughren was asked about this during his deposition, he explained “[i]t’s something that I — that I generally do.  I — you know, I explore a little bit more.”[265]  It was, as he put it, an “exploratory” analysis and was not a prespecified hypothesis.[266] Thus, there was no statistical analysis done on the data to determine whether the difference between children and adolescents was statistically significant:

  1. And so just based on what you said here, do you know whether or not the differences observed here were statistically significant or not?
  2. I — I don’t. And again, from my standpoint, it — it wouldn’t be that important. Because a P-value, whether it met that usual threshold of statistical significance would not be particularly relevant for something that wasn’t — that wasn’t being prespecified and tested.I mean — and you could do that. You could say if you make it on the overall analysis, then you get to — you have another 0.05 to look first at — at adolescents, and if you win there, then you get to look at — but it wasn’t done that way.[267]

It is also important to note, as discussed above, that when MD-18 was reviewed by the FDA, there was no statistical review done on the study[268] and Dr. Hearst, quite literally, copied and pasted his entire MD-18 analysis from Forest’s final study report for MD-18.

However, when MD-18 was resubmitted to the FDA in 2008 as part of Forest’s application for an adolescent indication for Lexapro—an application the FDA had already agreed to approve before Forest even started enrolling patients in MD-32—the FDA did not conduct any significant rereview of the data, but relied exclusively on Dr. Hearst’s and Dr. Laughren’s prior reviews of MD-18.

Specifically, the primary review of the supplemental application was done by Dr. Roberta Glass,[269] the team leader review was done by Dr. Ni Khin,[270] the statistical review was done by George Kordzakhia,[271] and the overall review and final approval was issued by Dr. Laughren.[272]

Dr. Glass, as the primary reviewer, did not conduct any in-depth analysis of MD-18.  Rather, Dr. Glass assumed that MD-18 was already positive, as she specifically noted that “[t]he sponsor reached an agreement with FDA that a pediatric claim for escitalopram, an isomeric version of citalopram, could be obtained with the support of one positive pediatric study in escitalopram in addition to the one positive study in citalopram.”[273]  According to Dr. Laughren, Dr. Glass was focused only on MD-32, not MD-18:

  1. All right. So it appears that Dr. Glass is operating off of the fact that Study MD-18 was positive and that they just had to look at whether or not there was an additional positive study for adolescents with Lexapro; is that right?

. . .

  1. That’s correct.[274]

This superficial approach to reviewing MD-18 is reflected in her report.  Dr. Glass simply copied and quoted Dr. Laughren’s analysis, which included data from the unblinded patients, stating: “The study is positive for the primary efficacy variable of change from baseline of the CDRS-R total Score (p=0.038). . . . As Dr. Laughren noted in his memo of 9/16/02, ‘…it appears that the positive results for this trial are coming largely from the adolescent subgroup.’”[275]  Indeed, Dr. Glass copied and pasted Dr. Laughren’s exploratory analysis into her review.[276] And, like Dr. Laughren, there is no statistical analysis of the difference between children and adolescents or any discussion about the fact the study report specifically stated that the differences were not driven by age.  Dr. Glass also never mentioned or discussed the effect of the unblinding—indeed, there is no indication she was even aware of it or its significance.  When Dr. Laughren was shown this data during his deposition, he agreed:

  1. It appears that Dr. Glass is relying on your exploratory analysis of the different effects observed in the pediatric and adolescent subgroup in your memo of September 16th, 2002.
  2. That’s correct.
  1. And indeed, she has pasted the results on the next page. It says “Summary of Primary Efficacy Variable for Study 18 by Age Subgroups,” and it says — literally says: “Extracted from memorandum by Laughren, September 16, 2002.” Do you see that?
  2. I do.

. . .

  1. It does not appear that she did a comprehensive clinical review of MD-18 at this point; is that right?

. . .

  1. That’s likely the case, yes.[277]

And, Dr. Khin likewise conducted a superficial review of MD-18—again, predicated on the fact that the FDA had already determined that MD-18 was positive.  Dr. Khin explained: “In this review cycle, our review of efficacy was focused on the positive results from one placebo controlled short-term study (study SCT-MD-32) in our evaluation of the efficacy and safety of escitalopram in the acute treatment of MDD in adolescents.”[278] Dr. Laughren explained:

  1. Would it be fair to say that they had marching orders at this point in their review that Study MD-18 was positive, just look at 32 and tell us if that’s also positive?
  2. I — I don’t — I don’t know that I would call that marching orders. . . . I think there was — there was that understanding that, you know, we had already looked at — at 18 and made a judgment that it was a positive study. I mean, certainly no one instructed them not to look at 18. . .
  1. . . . [T]hey appeared at least to have been relying upon the agreement that the FDA reached with Forest in 2004.
  2. I think that’s fair.[279]

Consistent with that approach, Dr. Khin specifically stated that she relied on Dr. Hearst’s (the FDA reviewer who copied and pasted from Forest’s final study report) and Dr. Laughren’s reviews: “I would refer to the clinical review by Dr. Earl Hearst dated 9/12/02 and a memorandum by Dr. Thomas Laughren dated 9/16/02 regarding their reviews of materials submitted under supplemental NDA for citalopram on 04/18/2002. I will briefly summarize their interpretation of results from the Study 18 . . . below.”[280]  Then, Dr. Khin proceeded to copy and paste Dr. Laughren’s exploratory analysis from 2002—the very analysis that was never subjected to any statistical analysis.[281]  And, she made her conclusion relying on the Hearst and Laughren analyses: “Based on prior clinical review by Dr. Hearst and Dr. Laughren’s memo, we should be able to count on positive efficacy results from citalopram study 18 in the same aged population for acute treatment of MDD.”[282]  Dr. Laughren confirmed this:

  1. So it appears that Dr. K[h]in is relying heavily, if not exclusively, on Dr. Hearst and yourself’s analysis of Study MD-18.
  2. That’s correct. Now, of course, this is the team leader review. It’s not the primary review.[283]

This reliance on Dr. Laughren’s exploratory review was raised during his deposition as well:

  1. When you prepared your memo for CD – for MD-18, and you did this exploratory analysis dividing the adolescents from the children, did you anticipate that that being — that was going to be used to support an indication for a different drug in adolescents? . . .
  2. I — I doubt that I was thinking ahead that far.
  1. Fair enough. In retrospect, it seems that that’s exactly what happened.
  2. That’s true.[284]

Finally, Dr. Kordzakhia conducted the FDA’s statistics review for the supplemental application.[285]  But, that statistics review, like Dr. Glass’s and Dr. Khin’s reviews, was limited to MD-32.[286]  This means, for a second time, MD-18 evaded any meaningful statistical review.  Dr. Laughren admitted this was not typical:

  1. Is that typical for a pivotal trial that’s going to be used to support indication to have just not been given any statistical review? . . .
  2. It’s prob- — it’s probably not typical.

. . .

  1. Do you think that probably would have been helpful, particularly since you’re using a particular subgroup of an exploratory analyses that you did in your review of the study? . . .
  2. In — in retrospect, I think I — I would have preferred that.[287]

Ultimately, Dr. Laughren approved the supplemental application and Lexapro was indicated for adolescent depression in March 2009.[288] And, subsequently, Forest continued to aggressively market and sell Lexapro for use in adolescents throughout the United States, albeit now it was “legal.”   Dr. Laughren, however, admitted that, if MD-18 was a negative study, he would not have approved the adolescent indication for Lexapro:

  1. If MD-18 was in fact negative, would you ever have approved Lexapro for use in adolescents? . . .
  2. I mean, if — if — if you couldn’t rely on 18 as a source of evidence, then you would’ve only had one source of evidence for Lexapro. So the answer is this is speculation, but I — I would not have recommended approving it.

. . .

  1. You’re the one who ultimately did approve it, right?
  2. Because I — I considered Study 18 a reasonable source of evidence.
  1. No, I know. And I’m just saying it’s not speculation because you’re actually the one who ultimately signed off finally on Lexapro’s approval for adolescents, right?
  2. Yes. . . .
  1. And you’re saying you wouldn’t have approved it if there was only one study, positive Study 32, right? . . .
  2. That’s correct.[289]

And, while Dr. Laughren defended his decision to approve Lexapro for adolescents during his deposition, he admitted that the FDA, even today, could re-review this data and conclude that MD-18 was negative and, thus, rescind the Lexapro approval for adolescents:

  1. Do you agree, though, Doctor, that a reasonable regulatory person at the FDA could come to a different conclusion about the positive results of MD-18? . . .
  2. It — this is always a matter of judgment. So the answer would be, yes, different people looking at the same dataset can reach a different conclusion.

When one considers that Forest’s corporate representative admitted, under oath, that MD-18 is negative when the unblinded patients are removed,[291] and the mountain of evidence showing these patients were, in fact, unblinded,[292] the scope of the fraud comes into view—by misleading the FDA about the unblinding in MD-18 using “masterful euphemisms”[293] and deliberately misleading language, Forest bamboozled the FDA into approving Lexapro for use in adolescents.  And, now, the man responsible for allowing this mischarge of science—indeed, the godfather of all modern antidepressants,[294] Dr. Laughren—started a company called Psychopharm Consulting, where he touts having “29 years of experience at the FDA in assisting pharmaceutical companies with psychiatric drug development programs,” and “hope[s] to continue in this effort as an independent consultant.”[295]

Part VI: Forest Used the False Assertion that MD-18 Was Positive and the Fda’s Approval for Lexapro to Negotiate Reduced Penalties in Usao Case

When the USAO negotiated Forest’s criminal plea, civil settlement, and CIA, Forest was able to use the “fact” that MD-18 was positive and that the FDA had approved, in March 2009, the adolescent use of Lexapro.  As it turns out, however, those two material facts, which clearly limited the ability of the United States to fully prosecute Forest’s off-label promotion, were based on outright fraud and deception.   Under the terms of the criminal plea, the USAO retains the right to reopen its prosecution if the plea agreement was based on or involved any falsehoods.[296]  The information and evidence set forth in this memorandum strongly support the reopening of the USAO’s prosecution of Forest, to hold Forest accountable for the fraud perpetrated on the FDA, the USAO, physicians, parents, and the medical / scientific community.

Part I | Part II | Part III | Part IV | Part V | Part VI


[257] Exh. 68, Letter from FDA re. Forest’s Questions at 1.

[258] Id.

[259] Exh. 8, 2017 Depo. of T. Laughren at 368:24-369:9.

[260] Exh. 9, Excerpts of Study MD-18 Rpt. at pg. 38.

[261] Exh. 21, Email re. Notes from conference call Oct 4 (w/attachment) at 2 “Note that study was not powered to look at differences within the two subgroups (children and adolescents).”

[262] Exh. 9, Excerpts of Study MD-18 Rpt. at pgs. 69, 72, 83.

[263] Exh. 8, 2017 Depo. of T. Laughren at 94:23-95:4 (“Q. Would it be fair to say then that when you stated here that the agreement between the sponsor and FDA that these trials were negative was referring to negative in the sense that it wouldn’t be sufficient to secure a pediatric indication? A. That’s – that’s the way I interpret that, yes.”).

[264] Exh. 52, T. Laughren, Memo. re. Recommendation for Non-Approval at 3 (“[I]t appears that the positive results for this trial overall are coming largely from the adolescent Subgroup.”).

[265] Exh. 8, 2017 Depo. of T. Laughren at 283:5-8.

[266] Id. at 284:10-21.

[267] Id. at 284:24-285:14.

[268] Exh. 52, T. Laughren, Memo. re. Recommendation for Non-Approval at 1 (“Since there was agreement between the sponsor and FDA that these trials were negative, there was no need for a statistics review of the efficacy data.”).

[269] Exh. 74, Roberta Glass, Clinical Review, FDA.

[270] Exh. 75, Ni Khin, Memorandum re. Recommendation of Approval Action for Lexapro, FDA.

[271] Exh. 76, George Kordzakhia, Statistical Review and Evaluation, FDA at 1, 3 (indicating that the statistical review were done on MD-32)

[272] Exh. 77, T. Laughren, Recommendation for Approval Action for Lexapro, FDA; Exh. 49, Lexapro Approval Letter for Adolescents at 3.

[273] Exh. 74, Glass supra note 269 at 8; see also id. at 9 (“On November 16, 2004, the Division confirms that one additional positive acute treatment study with escitalopram in adolescents, in addition to Study CIT-MD-18, is adequate evidence to support a labeling claim that escitalopram is an effective acute treatment of MDD in adolescents. Thus, Study SCT-MD-32 in adolescent patients was initiated in February 2005.”).

[274] Exh. 8, 2017 Depo. of T. Laughren at 381:6-12.

[275] Exh. 74, Glass supra note 269 at 22.

[276] Id. at 23.

[277] Exh. 8, 2017 Depo. T. Laughren at 382:13-383:16 (emphasis added).

[278] Exh. 74, Khin supra note 270 at 2.

[279] Exh. 8, 2017 Depo. of T. Laughren at 387:5-388:2.

[280] Exh. 74, Khin supra note 270 at 3.

[281] Id. at 6 (“Summary of primary efficacy results by age group for Study CIT-MD-18 – LOCF (data extracted from Dr. Laughren’s memo dated 9/16/2002)”; see id. at 5 (“Please see Table 2 in section 5.1.3 regarding summary of primary efficacy results by age group for Study CIT-MD-18 (LOCF data extracted from Dr. Laughren’s memo dated 9/16/2002)”).

[282] Id. at 6-7.

[283] Exh. 8, 2017 Depo. of T. Laughren at 389:12-18.

[284] Id. at 392:12-393:1 (emphasis added).

[285] Exh. 76, Khin supra note 270.

[286] Id. at 1, 3 (indicating that the statistical review were done on MD-32).

[287] Exh. 8, 2017 Depo. of T. Laughren at 384:4-385:7.

[288] Exh. 49, Lexapro Approval Letter for Adolescents at 3, at 1-2.

[289] Exh. 8, 2017 Depo. of T. Laughren at 401:15-402:16 (emphasis added).

[290] Id. at 402:18-403:2 (emphasis added).

[291] Exh. 12, 2016 Depo. of S. Closter (Forest’s Rule 30(b)(6) Corporate Representative) at 294:10-295:20 (“If they were removed from the study, I understand that the result would have been negative.”).

[292] Exh. 16, Email re. CIT-18 FAX to Investigational sites (w/ attachment) at 2 (“[D]ispensing these tablets would automatically unblind the study.”); Exh. 18, Email responses re. Letter to FDA for CIT-18 at 1 (“[T]he integrity of the blind was unmiatakenly violated.”); Exh. 20, Email re. CIT-MD-18 at 1 (“We need to generate Tables . . . excluding the 9 patients who were unblinded at the beginning of the study.”); Exh. 21, Email re. Notes from conference call Oct 4 (w/attachment) at 2 (“[S]ome citalopram tables were not blinded. The 9 patients who received unblinded medication were included in the main analyses[.]”); Exh. 11, 2016 Depo. W. Heydorn at 155:2-24, 157:18-21, 218:6-13 (“Q. So with respect to the nine patients who received the pink tablets, the study was unblinded with respect to them automatically, correct? . . . THE WITNESS: I guess yes. . . . Q. So they were unblinded as well, correct? . . .  THE WITNESS: With respect to those patients, I would assume so. . .  Q. Now, having seen this e-mail from Dr. Flicker and the fax from Dr. Tiseo, would you agree that the patients who were subject to the dispensing error were actually unblinded? . . .THE WITNESS: I don’t know for a fact, but that’s the implication from these letters, yes.”); Exh. 46, J. Jureidini Expert Report at 5 (showing that the results from the unblinded Celexa patients were 50-60% greater in than blinded Celexa patients).

[293] Exh. 18, Email responses re. Letter to FDA for CIT-18 at 1.

[294] Exh. 8, 2017 Depo. T. Laughren at 27:23-28:7 (“Would it be fair to say that during your time at the FDA, you were involved in some capacity with the approval or review of all of those SSRIs? A. Every one of them[.]”).

[295] Exh. 78, LinkedIn Profile for T. Laughren & PsychoPharm Consulting, at 1 (emphasis added).

[296] Exh. 2, Plea Agreement at 6.