TCPR: Dr. Ghaemi, Lamictal (lamotrigine) has become a very popular medication in psychiatry, but there remains some confusion about when to use it. What is its official FDA-approved indication?
Dr. Ghaemi: Lamictal is approved for delaying the time to relapse to depressive and manic episodes in bipolar type I disorder. This is based on two good randomized clinical trials. They were both 18-month studies. One of them began with the patients being depressed, and the other one began with the patients being manic. In the first study, patients with acute bipolar depression were all given Lamictal open-label, and patients who responded to Lamictal (about half of the subjects) were then enrolled in the double-blind part of the study, in which they were randomized to Lamictal, lithium, or placebo (Calabrese JR, et al., J Clin Psychiatry 2003 Sep;64(9):1013-24).
TCPR: And what were the results of that study?
Dr. Ghaemi: Compared to placebo, Lamictal and lithium prolonged the time to a recurrence of depression and mania.
TCPR: I know you have some reservations about the methodology used in the Lamictal studies – what are they?
Dr. Ghaemi: Patients were eligible for the study only if they had tolerated and responded to Lamictal, meaning that the study sample was enriched with patients who were likely to do well on that agent.
TCPR: In other words, the deck was stacked in favor of Lamictal. Is that an unusual study design?
Dr. Ghaemi: Actually, it isn’t unusual. All maintenance studies in bipolar disorder these days are done this way. For example, both the Zyprexa studies and the Abilify studies, both of which got FDA approval, used the same method: patients were preselected based on whether their acute mood episode responded to the drug of interest, then they were enrolled in double-blind trials comparing that drug with placebo.
TCPR: Why is this design so popular?
Dr. Ghaemi: There are two rationales behind it. First, if you want to maximize the likelihood that the drug will be better than placebo, then this would be one way of doing it. There were previous studies in which this approach was not taken, like the Depakote maintenance study (Bowden CL et al., Arch Gen Psych 2000;57:481-489). In that case the study was not enriched with Depakote responders; people were allowed to be in the study even if they hadn’t taken Depakote before. They were enrolled if their bipolar disorder was in remission, and then they were randomized to Depakote, lithium or placebo, and followed for a year. In the end, Depakote and lithium were not better than placebo, and so the FDA did not give Depakote an indication for maintenance treatment of bipolar disorder. The second rationale is that this design more accurately reflects clinical practice. Psychiatrists don’t wait until people are in remission before trying a medication; instead, they start these drugs during an acute episode of depression or mania, and then they need to decide whether or not to continue patients on the drug long-term.
TCPR: So what are the problems with this design?
Dr. Ghaemi: The disadvantages are two-fold. One is that even if the data are valid they do not generalize to all patients. In other words, what studies like this show is not that you can put anybody on Zyprexa or Abilify or Lamictal and they will have long-term benefits in bipolar disorder. What they show is that if somebody does well with those drugs for an acute episode, they might do well with those drugs long-term. The second critique is that this design does not fairly compare Lamictal with lithium because the patients were preselected to be tolerant of and responsive to Lamictal but they were not preselected to be tolerant of and responsive to lithium. So it shouldn’t be a surprise that in some of the analyses, Lamictal is better tolerated and more effective than lithium.
TCPR: So, according to your reading of the data, what is the clinical situation that we should be using Lamictal for?
Dr. Ghaemi: It should help more than placebo with the prevention of both mania and depression, although probably more effectively for depression than mania. This benefit is best seen in persons who initially tolerate and appear to benefit from Lamictal acutely.
TCPR: It seems to me that many clinicians have jumped to the conclusion that since Lamictal works to prevent depression, therefore it must also be effective for treating an acute episode of depression.
Dr. Ghaemi: And that makes intuitive sense, but you can never assume that just because a drug is effective for prevention of depression that it is also effective for acute treatment, or vice versa. I call it “the happily ever after fallacy,” this idea that if you get better acutely, you are going to stay better forever if you just stay on the same drug. For instance, quetiapine works for acute bipolar depression but we don’t have data to show that it prevents it. And there is extensive literature on lithium for depression prevention in bipolar disorder but much less definitive literature for acute efficacy.
TCPR: What is the actual evidence with regard to Lamictal for treating acute depression, either bipolar depression or unipolar depression?
Dr. Ghaemi: Five studies have been conducted testing Lamictal for acute bipolar depression and two for unipolar depression. Each had a similar design with adequate numbers of subjects (about 200 per study), and every single one of these studies was negative, meaning that Lamictal performed no better than placebo.
TCPR: That’s surprising. Have these studies all been published?
Dr. Ghaemi: Only one was published, and that was a study in which a secondary analysis showed that Lamictal was more effective than placebo (on the MADRS depression scale), but there was no separation on the primary outcome (the Hamilton depression scale) (Calabrese JR et al., J Clin Psychiatry 1999;60:79-88).
TCPR: How do you know about the results of the other studies if they were never published?
Dr. Ghaemi: Because GlaxoSmithKline was legally required to post all of their studies, positive or negative, on their website (see http://ctr.gsk.co.uk/Summary/lamotrigine/studylist.asp). This was a result of an earlier suit accusing the company of suppressing data that indicated a significant suicide risk among children taking Paxil.
TCPR: Does this data mean that Lamictal is actually not an effective antidepressant?
Dr. Ghaemi: I think that is the obvious conclusion, at least acutely. When there are that many wellpowered studies, and all are negative, this usually means the drug doesn’t work. However, there is one other interpretation if one wanted to be more charitable, which is based on the fact that Lamictal requires a very slow titration – most of these studies were about two months long and it takes a month just to get to 100 mg, which would be an average dose. Maybe Lamictal didn’t have enough time to show a benefit, and maybe a 12 or 16 week study would have shown a separation from placebo. The problem is that the average major depression episode in bipolar disorder gets better naturally in three to six months. So there is no way you could prove it; you couldn’t make the study longer and show that Lamictal is better than placebo because the placebo recovery would be very high. So either Lamictal is an ineffective antidepressant, or it is effective but irrelevant, because by the time it would work the present episode would disappear naturally.
TCPR: My sense is that very few people know about these negative data, is that true?
Dr. Ghaemi: It is true. I first came across this data about a year ago, and I actually serve on the national advisory board for Lamictal, so I am privy to more data than most psychiatrists.
TCPR: I think most psychiatrists would assume that somebody being paid by the company to be on an advisory board would be biased in favor of the product, or at least unlikely to blow the whistle on negative studies. Are you unusual in that sense?
Dr. Ghaemi: I think I am in the minority. My experience in the few advisory boards that I have been on is that it is not uncommon for there to be one or two people in that room who are pretty critical of the company’s approach about interpreting or marketing their data. So to some extent the companies are willing to hear criticism, but usually it is a minority view that ultimately doesn’t have much impact on what they decide to do.
TCPR: My understanding is that you wrote up your discovery of the negative Lamictal data and submitted the paper to some journals. What has been the response?
Dr. Ghaemi: I first submitted it to JAMA because I knew that they were sympathetic to this kind of critique. Their reaction was, “We already publish many papers like this; this is old news; there is nothing new here.” They recommended that I send it to a psychiatric journal. So then I sent it to the American Journal of Psychiatry, but they rejected it as well, saying that they were doubtful that this type of negative publication bias was common among other companies marketing medications for bipolar disorder.
TCPR: Do you think there is much suppressed negative data about other drugs?
Dr. Ghaemi: It’s very hard to get this information. Companies are not required to disclose it. And if they do publish it, they will sometimes delay publication for two or three years, and then publish it in an obscure journal that is less likely to be read.