Pediatric Antidepressant Role Not Settled, FDA Says; Pfizer Defends Zoloft

A pooled analysis of two Zoloft clinical trials in children suggests the product is effective for pediatric depression, Pfizer said during a Sept. 9 House Energy & Commerce/Oversight Subcommittee hearing.

Although neither of the two placebo-controlled trials of Zoloft (sertraline) met the efficacy endpoints in pediatric major depressive disorder, a pooled analysis of the data showed a statistically significant result, Pfizer VP-U.S. Medical for Psychiatry & Neurology Cathryn Clary, MD, said.

The Sept. 9 hearing illustrated the extent to which the antidepressants controversy has moved beyond the question of the drugs' role in the pediatric population. The regulatory status of pediatric indications for antidepressants was treated almost as an afterthought during the hearing (see chart: " ¹ Rep. Barton's Rx For Pediatric Depression ").

The focus of the hearing was on reporting of negative clinical trial data, and questions from subcommittee members addressed off-label use, clinical trial disclosure and pediatric exclusivity standards (see ² (Also see "Off-Label Regulation Is One Theme Of House Antidepressant Hearing" - Pink Sheet, 20 Sep, 2004.) ).

As a practical matter, the extensive publicity surrounding claims that antidepressants increase pediatric suicidality is likely to have a greater impact on use of the products than FDA's ultimate decision about indications.

However, FDA Acting Deputy Commissioner for Operations Janet Woodcock, MD, stressed that the "jury is still out" on the role of the class as a whole.

"We think the jury is still out on the effectiveness of these drugs. Some of the studies were not conclusive. It is very common for effective drugs, when they're tested in adult depression, to show no effect," Woodcock said during the hearing.

"As a public health agency, FDA must weigh the possibility of an increased risk of suicidality in young patients taking these drugs against the known risk of suicide in patients whose depression goes untreated," Woodcock said.

Woodcock noted the issue would be addressed at a joint meeting of the Pediatric and Psychopharmacologic Drugs Advisory Committees Sept. 13-14. The advisory committee recommended a "black box" warning for all antidepressants about the risk of pediatric suicide.

During the advisory committee, Pfizer noted that "there was no significant difference between active treatment, in this case sertraline, and placebo, but both groups showed improvement from baseline to endpoint." The company says it has no plans for further studies of MDD in children.

In opening remarks, Energy & Commerce Committee Chair Joe Barton (R-Tex.) included the Zoloft studies among the many pediatric depression trials that "showed no efficacy," but the company argued that they in fact suggest Zoloft may be effective in children.

"Before the studies were completed, and before the blind was broken...Pfizer decided to pool the two studies, in order to have sufficient power to show a difference, if one existed," Clary explained.

"When the two studies were pooled, thus providing an adequate sample size to test if there were true differences, the treatment effect of Zoloft compared to placebo...was shown to be statistically significant," Clary said. She acknowledged that the effect was "modest in size."

In designing the two multicenter trials, Pfizer powered the studies on a previous single-site trial of fluoxetine.

"To address the concern that the random variability in such a study might be greater than in a small well-controlled single-site study, the two studies were designed to be identical in order to allow a combined analysis," Clary said.

"We believe that the pooled analysis was the most valid, scientifically accurate way to show the data," Clary said. However, FDA did not grant a pediatric MDD indication because "Pfizer had not fulfilled the requirement for two positive trials," Clary explained.

Although Pfizer was specifically asked to do two pediatric MDD trials for Zoloft under a formal written request issued by FDA, it was not clear that the company would need two positive studies for the MDD indication, Clary said.

"It was clear in the written request that we were to perform two studies. What was not clear was that two studies need to be positive in order to receive approval," she said.

Zoloft was approved in October 1997 for pediatric obsessive compulsive disorder based on one study. FDA cited "the general view that OCD is essentially the same disorder in adults and children" in accepting the single trial.

Rep. Greg Walden (R-Ore.) quoted from a sample FDA written request provided by the agency that states, "we believe that a pediatric depression claim for any antidepressant already approved in adult depression would need to be supported by two independent adequately well-controlled clinical trials in pediatric depression."

Clary maintained that Pfizer may have received a less specific request. "My experience...was that the actual content of these letters have been changing and they're becoming more specific," Clary said.

Walden also quoted FDA's evaluation of the MDD data, by reviewer Andrew Mosholder, MD: "'The sponsor proposed pooling the data from the two trials to yield a statistically significant result on the basis the trials were conducted under identical protocols. This, however, would be a major departure from our usual policies discouraging pooling of efficacy data.'"

"We of course accepted" the agency's conclusion on the pediatric depression data, Clary said.

Pfizer received a six-month pediatric exclusivity extension for Zoloft in 2002. Pfizer and FDA agreed, however, that the data did not need to be reflected in labeling.

"FDA chose not to describe the studies in the label," Clary said, quoting from an agency letter to Pfizer: "Given the fact that negative trials are frequently seen, even for antidepressant drugs that we know are effective, we do not feel that it would be useful to describe these negative trials in labeling, since this may be misinterpreted as evidence that Zoloft does not work in this population."

The only antidepressant approved for use in children is Lilly's Prozac . That fact gave Lilly the easiest task at the hearing. Prozac pediatric data are posted on the FDA website under the agency's general disclosure provisions for approved applications.

Aside from the two studies to support the approval of Prozac, the only other antidepressant with a positive study in pediatric MDD is Forest's Celexa (citalopram). A U.S. trial sponsored by Forest was positive; a European trial by marketing partner Lundbeck was negative.

Walden asked Forest Exec VP-Scientific Affairs Lawrence Olanoff why it "took over a year...for you to mention publicly that one of the studies was negative?"

"We didn't put as much emphasis on the European study for a number of reasons, the main reason being that we didn't find the study particularly informative," Olanoff said.

He noted the high placebo response rate, imbalances between the active and placebo groups, and less restrictive enrollment criteria for the European trial.

Unlike all the other pediatric MDD antidepressant trials, the European study of Celexa "allowed for the enrollment of patients without regards to a past history of hospitalization due to psychiatric illnesses, past history of suicide-related events, or a history of a failure to respond to other antidepressants," Olanoff said.

"We couldn't provide useful information on a negative study. We also knew it is very difficult to publish negative studies," Olanoff added.

The other manufacturers testifying before the House committee - GlaxoSmithKline ( Paxil ), Wyeth ( Effexor ), Bristol-Myers Squibb ( Serzone ), Organon ( Remeron ) - highlighted their efforts to report negative clinical studies of their drugs in children.