Variations on a Theme of Sleaze

VARIATIONS ON A THEME OF SLEAZE

A few days ago I posted on corrupt reports in two medical journals, where key opinion leaders (KOLs) and corporate employees misrepresented the potential of Janssen’s atypical antipsychotic (AAP) drug risperidone for depression. One of these reports appeared in a general medical journal, Annals of Internal Medicine (AIM), which confirms the designs of the corporate marketers: by volume, treatment of depression now is centered in primary care. Considering the weak efficacy data, the dubious risk-benefit profile, and the inferiority of AAP drugs to other options, there is no justification for the broad and early adjunctive use of these agents for depression in primary care. The Eli Lilly Company, which markets the combination of olanzapine and fluoxetine in a single pill (Symbyax), has the same objective. Other companies are moving rapidly into this market space.

Combining AAP drugs with antidepressants takes us back to the bad old days of antidepressant-antipsychotic drug combinations like Triavil in the1960s and 1970s, when we learned that depressed patients are especially susceptible to a serious adverse event known as tardive dyskinesia (TD) caused by antipsychotic drugs. While the risk of TD is less than with the early antipsychotic agents, it is still unacceptably high with AAP drugs for patients who are unlikely to show meaningful clinical benefit, as I detailed earlier. In adult patients with schizophrenia the risk of TD with olanzapine treatment is about 2.5% at 1 year. In children and adolescents treated with AAP drugs for 6 months the risk is an alarming 6% (Wonodi I et al Movement Disorders 2007; 22: 1777). In patients with mood disorder, these figures are likely to be higher. And this is before one even begins to factor in the metabolic toxicity of AAP drugs (weight gain, obesity, insulin resistance, and Type II diabetes mellitus)! Patients are not well served when AAP drugs are pushed for treating depression in primary care.

Medical journals are not the only compromised medium. Continuing Medical Education (CME) is a second front in the campaign to expand the AAP drug market. The standard formula calls for corporate sponsorship channeled through an “unrestricted educational grant” to a medical education communications company (MECC). The MECC employs writers to prepare the “educational content,” and academic KOLs are recruited to deliver this content. The KOLs are chosen for their willingness to be “on message” for the corporate sponsor. If they go “off message” they know they will not be invited back. The talk of “unrestricted grants” is window dressing. The MECC also secures the imprimatur of a nationally accredited CME sponsor, typically an academic institution. The sponsor is paid to certify that the CME program meets the standards of the Accreditation Council on Continuing Medical Education (ACCME). Everybody turns a buck: the MECC and its staff are handsomely paid (CME is now a multi-billion dollar business); the KOLs are generously rewarded with honoraria and perquisites; the academic sponsor is well paid by the MECC; the ACCME receives dues from the academic sponsor; the audience obtains free CME credits rather than having to pay for these required educational experiences; and the corporate sponsor gets what it considers value for its marketing dollar.

ACCME standards include clear identification of off-label drug use; full and fair disclosure of clinical trial results, warts and all; and clinical guidance to the audience about risks, benefits, and treatment options. It is not an exaggeration to say that these standards are more often honored in the breach than in the observance. That is because CME events have been degraded to little more than thinly veiled advertising, built around promoting a product rather than around education. So corrupted has the process become that the Macy Foundation recently recommended that industry financing of CME be ended, “whether such support is provided directly or indirectly through subsidiary agencies.” See Daniel Carlat for more on this topic. Is this position alarmist? Consider the following examples of corruption in CME.

A widely advertised CME program appeared on-line 6 November 2007, titled Treatment-Refractory Depression: Is there a Role for Atypical Antipsychotics? Note the leading question and the product category focus. The program was developed by the MECC PeerView Institute for Medical Education, and it was sponsored by the Semel Institute for Neuroscience and Human Behavior at UCLA, which certified CME credits. Like the journal articles I discussed last week, this CME program is marked by a concatenation of deceits. The major messages were, augment earlier rather than later; AAP drugs are “an emerging therapeutic option” for augmentation (note the branding language); and AAP drugs are efficacious. None of these messages is based on credible evidence.

The first sleight of hand was the leadoff presentation, which featured the STAR*D study results concerning remission and response rates to adjunctive treatments or switching after various levels of treatment failure. The data naturally suggest unmet needs in treating depression (a favorite theme of marketers). The unspoken implication of this academic veneer is that the later studies described in the CME program were comparable to the STAR*D study in terms of case material, which is not so. Recruitment to STAR*D was explicitly different from recruitment to the usual experimercial sponsored by a drug company, where many cases come from contract research organizations, not from clinical referral streams. Moreover, STAR*D was a purely descriptive study that by design could not identify specific treatment effects.

Charles B. Nemeroff, MD, PhD from Emory University (yes, the same) discussed short term use of AAP drugs. His presentation is a model of being economical with the truth. Dr. Nemeroff has clearly mastered the art of accommodating his many corporate clients. He discussed 4 atypical antipsychotic drugs as augmenting agents for nonresponding depression. When discussing olanzapine he went beyond his short term remit to suggest that long term treatment is efficacious, yet he neglected to address the neurological or metabolic toxicity of long term olanzapine. When discussing risperidone he did not disclose that he was senior author of the major report he described and cited; he neglected to disclose the retractions he and Mark Rapaport from Cedars-Sinai Medical Center had been obliged to publish; he neglected to disclose that treatment with risperidone beyond 6 weeks was no more efficacious than placebo. That aspect was discussed by another speaker, who repeated Dr. Nemeroff’s now-retracted and discredited claims for significant long term preventive efficacy of risperidone in a subgroup of patients. Dr. Nemeroff made further claims about risperidone improving sexual function in patients receiving an SSRI antidepressant but he failed to disclose that beyond 6 weeks risperidone impaired sexual function in women who were receiving the SSRI; he backed up his claims about risperidone and sexual functioning by citing his publication in his own journal Neuropsychopharmacology that contained no data whatsoever on the matter; and he neglected to address the metabolic toxicity of risperidone that the corporation disclosed on ClinicalTrials.gov. He also falsely stated that the short term efficacy of risperidone in nonresponding depression was demonstrated in a controlled study, citing his own open-label study. These problems were called to the attention of the Semel Institute for Neuroscience and Human Behavior at UCLA. In response, the CME program was revised on January 11, 2008. Dr. Nemeroff’s material now contained a different citation that again contained no data concerning sexual side effects. They removed the claim that the short term efficacy of risperidone had been established in a controlled trial (although by then the problematic report of this very issue in AIM had been published for over 2 months). The inadequate discussion of the toxicity of risperidone in Dr. Nemeroff’s trial was unchanged, and another speaker continued to repeat the retracted and discredited claims of Dr. Nemeroff for significant long term preventive efficacy of risperidone in a subgroup of patients. No explanation of the changes made on January 11, 2008 in the on-line materials was provided by UCLA to CME readers who had studied the erroneous and biased material for more than 2 months. Other speakers made passing reference to the metabolic toxicity of AAP drugs but only in a perfunctory way that had no educational value, like what we see in direct-to-consumer advertising. Nobody mentioned the risk of TD.

As the Macy Foundation report makes clear, CME providers are expected to give learners guidance on the risk-benefit balance of new treatments. It is disingenuous of Dr. Nemeroff to talk up risperidone for short term treatment of these difficult depressions by exaggerating the benefit, downplaying the risks, avoiding comparison with alternative treatments, and glossing over the problem of longer term loss of efficacy. These are clear violations of ACCME principles.

When discussing aripiprazole for nonresponding depression, Dr. Nemeroff once again was economical with the truth. Note that Bristol-Myers Squibb, the marketer of aripiprazole, sponsored this PeerView/UCLA program. To document his claims about aripiprazole, Dr. Nemeroff cited one Abstract from the American Psychiatric Association meeting in May 2007. That does not meet ACCME standards of documentation for learners, most of whom would be unable to access the cited Abstract (not that it would tell them much even if they could). For some reason, Dr. Nemeroff did not inform learners that the complete report of the aripiprazole study had appeared in June 2007 (Berman RM et al. J Clin Psychiatry 2007;68: 843-853), fully 5 months before the CME event went on-line. From that readily available report it is clear that the Number Needed to Treat (NNT) for response with aripiprazole is 10, which compares unfavorably with a NNT of 4 for lithium, the best established augmenting option in placebo-controlled trials. A NNT of 10 means a clinician would need to treat 10 patients with aripiprazole before obtaining one remission that would not have occurred anyway with placebo. That does not constitute compelling clinical benefit. Dr. Nemeroff did not candidly discuss these troubling data. Dr Nemeroff provided his CME audience none of the remission or response data from the published aripiprazole study, though these data were readily available. These omissions of published, highly relevant information signify disrespect for his audience by Dr. Nemeroff, incompetence by the MECC, and failure of due diligence by the accrediting institution, UCLA, to ensure that accurate, balanced information and adequate documentation are provided. Likewise, no substantive risk-benefit analysis was provided to guide CME learners, and there was no meaningful discussion by Dr. Nemeroff of the metabolic toxicity of aripiprazole. The published report tells us that 7.1% of patients treated with aripiprazole gained more than 7% body weight, a very significant difference (p < 0.01) from the placebo treated patients (1.2%). Dr. Nemeroff did not share that information with the CME audience. Instead, he slyly minimized the appearance of the problem by showing a mean weight gain of only 2 kg with aripiprazole. In addition, the neuromotor toxicity of aripiprazole was remarkable (23.1% akathisia and 27.5% extrapyramidal symptoms). Dr. Nemeroff gave the CME audience no guidance about that problem or about its unblinding effect in the trial. Why do highly paid KOLs behave in this way? Do they think nobody will notice?

A final insult to CME learners in this program was the disclaimer that “The Semel Institute for Neuroscience and Human Behavior at UCLA is responsible for the selection of this report’s topics, the preparation of editorial content, and the distribution of this report” but “No responsibility is taken for errors or omissions in these reports.” Well, then, who is responsible? Why not UCLA, considering the ACCME standards and the fees UCLA received for sponsoring this CME activity through an “educational grant” from Bristol-Myers Squibb Company? Overall, it is difficult to avoid the impression that this so-called CME activity is a meretricious infomercial for Dr. Nemeroff’s corporate clients rather than a balanced educational event that aims to give practitioners considered guidance on a difficult clinical problem.

Will these revelations slow the marketing-inspired momentum for use of atypical antipsychotic drugs in depression? Not likely. Indeed, a new road show is right now getting under way, bringing the good news about atypical antipsychotic drugs in depression to CME audiences in Miami, San Francisco, Los Angeles, Chicago, Boston, and New York. Why now? Has some new insight been achieved that requires urgent communication to physicians? No. The road show has been launched now because aripiprazole was recently approved by the FDA for the secondary indication of adjunctive treatment in depression. It’s all about marketing. The faculty speakers are the usual suspects – KOLs and KOL wannabes who enjoy cozy or nepotistic relationships with the chairman. The funding is through another “educational grant” from the marketers of aripiprazole. The CME sponsor is an outfit in Texas that knows how the CME game is played. And the chairman of this new enterprise? Why, none other than the compromised Dr. Charles Nemeroff from Emory University. Why are we not surprised?