Side Effects: A Prosecutor, a Whistleblower, and a Bestselling Antidepressant on Trial

by Alison Bass

  The Glaxo attorneys said no way.

  "They didn't want to give us any money," Firestein said. Finally, both sides settled on a figure of $2.5 million. The money would go into the state coffers. Firestein finished drafting the assurance agreement and consent order that had to be filed with the Southern District Court of New York.

  On Thursday, August 26, the New York attorney general's office announced it had settled its legal action against GlaxoSmithKline. Spitzer hailed the settlement as "transformational in that it will provide doctors and patients access to the clinical testing data necessary to make informed judgments." He predicted that other drug companies would follow GlaxoSmithKline's lead in posting their studies online. His prediction wasn't entirely based on wishful thinking: Earlier in the summer, Eli Lilly had announced that it would create a Web site on which it planned to post the clinical results of already approved drugs. Johnson and Johnson and Forest Labs, which were under scrutiny by the AG's office, had made similar noises, as had Merck.

  COMPARED TO THE barrage of coverage that greeted the initial filing of the AG's lawsuit, news of the settlement was muted, perhaps because it came at a time when so many reporters were on vacation. The New York Times and the Wall Street Journal reported the news in short pieces buried inside; other major papers didn't mention it at all.

  The day the settlement was announced, Firestein was sitting on her usual perch in Stark's office when Conway strode in. He had a cup of soda in his hand.

  "Great job!" he said, raising the cup in the air. "We did it!"

  Earlier in the summer, Eliot Spitzer had publicly congratulated Firestein for her work on the case. He had given her (along with fifteen other attorneys in the office) the prestigious Lefkowitz Award at a special ceremony in June. The award was named after Louis Lefkowitz, a revered former New York State attorney general who had served from 1957 to 1978. "It's the highest award you can get in this office," Spitzer explained. "You get it for doing something that exemplifies the values of this office and the public-spiritedness of the litigation we're supposed to bring on behalf of the people of New York"

  At the ceremony in the New York State Supreme Court building on Foley Square, Spitzer told the assembled attorneys that the case against GlaxoSmithKline epitomized everything the New York State attorney general's office stood for. "I don't usually single out specific litigation because the litigation everybody here does is important," he said. "But this case is really, really important because it has the potential to change an entire industry's practices. This case is not about money; it is about protecting people's health."

  Spitzer called Firestein up to the front of the hall, shook hands with her, and gave her the Lefkowitz certificate in a frame. There was a burst of applause, and Firestein, her face flushed red with a mixture of embarrassment and pleasure, walked back to her seat.

  Yet now that the case was finally settled, Firestein felt oddly let down, almost bereft. She was glad the case was over and happy that the AG's office had prevailed, but some part of her mourned the excitement, the prolonged high she'd been on in preparing for this day. Over the past seven or eight months, she'd developed a real sense of camaraderie with Stark and even Conway and Baker. It was the kind of fellowship that developed from working around the clock on something you fervently believed in, with colleagues who had become as wedded to its success as you were. Firestein found it hard to accept that all that heady anticipation was now over. Finis.

  Stark talked about the four of them celebrating by going out to lunch at her favorite kosher restaurant, a few blocks down on Broadway. But for some reason or other, the lunch never happened. As Joe Baker recalled, "We were all too busy working on the next case"

  THE PUBLIC SPOTLIGHT, of course, didn't shut off overnight. GlaxoSmithKline's public reaction to the settlement was oddly unrepentant, almost dismissive, as if the summer's legal wrangling had all been a tempest in a teapot. In a press release dated August 26, the drug company said that "although GlaxoSmithKline believes the charges made in that litigation by the Attorney General are unfounded, the company has agreed to pay the State of New York $2.5 million to avoid the high costs and time required to defend itself in protracted litigation." The release noted that the company's plans for a clinical registry had been under way "for some months" and quoted Mark Werner, senior vice president for U.S. Legal Operations, as saying, "We are pleased that the Attorney General believes the clinical trial register we have been developing will provide useful information to the medical and scientific community."

  The company's disingenuous response so angered Spitzer that the following week, he lashed out at GlaxoSmithKline. "The arrogance of the [Glaxo] commentary is offensive and problematic," he said in an interview published in the Financial Times of London on August 31. "We are going to be watching them with an eagle's eye to see that they have abided by the terms of the settlement."

  By then, New York's attorney general was no longer the only lawmaker watching the pharmaceutical industry with an eagle eye. A congressional subcommittee on oversight and investigations had decided to take the matter into its own hands.

  artin Teicher's office at McLean Hospital was tucked away in a hard-to-find annex behind the newer and more imposing Mailman Research Center. There was no sign on his door, and first-time visitors would walk right past it on their way to the Developmental Biopsychiatry Research Program (the official name for Teicher's research group). The group office resembled a standard doctor's waiting area, with a sofa and chair soothingly upholstered in a dark blue and green leafy pattern, a coffee table, an assortment of magazines, and the requisite New England seascape. Teicher's private office was a more modest affair, with barely enough room for a desk, a small leather couch, two thin bookcases, a file cabinet, and two chairs. On one wall, a small portrait of a frightened little boy caught the eye. The painting of a sad clown hung in another corner, and a number of framed diplomas were displayed above his desk. A mounted Westminster clock chimed to mark the passing hour. The melodious clock was the only hint of Teicher's abiding love affair with music. When he turned his sights on a medical career as a young man, music became a sideline he enjoyed whenever he could squeeze it into his busy life. The year he turned fifty, though, Teicher decided to make some changes in his life. He remarried and began devoting more time to his music. He went to a folk music festival in Texas and took a songwriting course there. By the summer of 2004, his marriage was foundering (he and his second wife would divorce in 2005), and Teicher would spend hours writing songs about love and hurt and betrayal. (He eventually posted three of the blues tunes he had written and sung himself on myspace.com, where visitors to the site could listen to his songs and see a picture of Marty playing the guitar in shades and an open-collared white shirt.)

  The blues, however, were far from Teicher's thoughts when he returned to his office on the afternoon of August 20, 2004. There, he found a message from his assistant Cindy McGreenery: a reporter for the Baltimore Sun wanted to interview him. In the past few months, as the controversy over the SSRI antidepressants heated up and the FDA finally acknowledged that yes, there was an increased risk of suicidal behaviors from antidepressants, Teicher had suddenly become a soughtafter source. Reporters from all over the country wanted to talk to the man who had first raised that possibility fourteen years ago. But this time, Teicher was careful to consult with the hospital's medical chief and PR director, and he didn't take all or even most of the press calls. He would only speak to reporters from newspapers that he knew and trusted, such as the Boston Globe, the Baltimore Sun, and the New York Times. At the age of fifty-three, with a receding hairline and a closely cropped salt-and-pepper beard, Teicher had become, finally, a cautious man.

  The first media calls back in February 2004 had caught him by surprise. Immersed as he was in a completely different field of study-the biological effects of child abuse on brain development -Teicher simply hadn't been paying much attention to the renewed debate over the SSRI antidepressants
and their link to suicidal behaviors in children. But after the first press inquiry, he started browsing the FDA Web site and reading the news coverage more closely. He found himself cheering on Andrew Mosholder, the FDA official whose initial reanalysis had so panicked his superiors that they prevented him from testifying at the February hearing. Months later, Teicher was gratified to see that Mosholder's conclusions had finally been confirmed by another FDA reviewer and published on the agency's Web site. According to that August 20 posting, FDA analyst Tarek A. Hammad had reviewed clinical data submitted to the agency for nine of the newer antidepressants: Prozac, Paxil, Zoloft, Celexa, Effexor, Remeron, Luvox, Wellbutrin, and Serzone. Hammad had found that out of every one hundred children and teenagers in those trials, two to three experienced increases in suicidal thoughts and behaviors as a result of treatment with the drug. The increase in suicidal gestures among participants who took antidepressants was 1.8 times, or nearly double, the rate of suicidal thoughts and behaviors among those taking sugar pills.

  Despite such clear-cut findings, most of the comments posted on the psychiatric Web sites Teicher browsed sounded disturbingly dismissive. Some comments implied that the difference between the antidepressants and placebo on safety and efficacy grounds was not that significant; so what was all the fuss about?

  Teicher knew this wasn't true. He also knew that the real risk of suicidal ideation among patients who took antidepressants was probably much higher than the trial data indicated. Not only had the adult and pediatric trials of SSRIs like Prozac, Zoloft, and Paxil specifically excluded patients who had been hospitalized for depression, but in a number of cases, children who responded poorly to the antidepressants had dropped out of the studies and had not been followed to see if they attempted to kill themselves. In serving as an expert witness against Eli Lilly back in the 1990s, Teicher had seen evidence in unsealed legal documents that several drugmakers had included as "placebo suicides" deaths that had occurred after patients had been taken off other meds to participate in these trials. So their deaths should not have been coded as suicides from placebo, but rather as deaths that occurred during the washout period before the trials began. It was the contention of experts like David Healy, among others, that if the drug companies had designed their studies to more accurately reflect the number of suicides and suicide attempts among participants, the suicidal risks from the SSRIs would have been far greater than that reported by the FDA.

  In the same posting in which it announced Tarek Hammad's findings, the federal agency announced that it would be holding an advisory panel hearing on September 13 and 14 to consider mandating stronger warnings for antidepressants prescribed to minors. Even so, Teicher found it hard to believe that any real change would occur. So when he called back Julie Bell, the reporter from the Baltimore Sun, he couldn't hide his skepticism. "People are going to equivocate and they're going to do nothing," he told Bell.

  A fast-moving juggernaut of events would soon prove him wrong. On August 26, the New York State attorney general's office settled its case with G1axoSmithKline. Two weeks later, on September 8, the International Committee of Medical Journal Editors announced its plan to require that researchers register key aspects of the initial protocols for their studies on a public Web site if they wanted to be published in a leading medical journal. In identical editorials that ran in twelve top medical journals that month, the editors declared that this new policy would become effective starting in July 2005.

  Before the media had time to absorb this development, Congress grabbed the spotlight. On September 9, the U.S. House Energy and Commerce's Subcommittee on Oversight and Investigations held a full-day hearing at which legislators, Democratic and Republican alike, excoriated the pharmaceutical industry and the FDA for withholding "vital information" about the safety and effectiveness of drugs for children and adolescents.

  "What we are seeing is that the pharmaceutical industry has systematically misled physicians and patients by suppressing important data on their drugs," said Rep. Henry Waxman (D-Calif.), a member of the House subcommittee. "At the same time, the industry has encouraged physicians to use drugs that are ineffective and possibly even dangerous in one of the most vulnerable population groups, children. The industry has reaped literally billions of dollars in the process."

  Waxman then announced that he and Rep. Edward Markey (D-Mass.) were planning to introduce a bill that would impose penalties on drug companies that did not register their clinical trials and the outcomes on a public Web site run by the National Institutes of Health (http://www.clinicaltrials.gov).

  During the September 9 hearing, several legislators mentioned the New York AG's case as a catalyst for their concerns. One lawmaker Rep. Peter Deutsch (D-Fla.)-even brought up a few of the damaging internal Glaxo memos that Rose Firestein had unearthed. Referring to GlaxoSmithKline's 2001 memo to its sales reps lauding Paxil's "REMARKABLE efficacy," Deutsch wanted to know whether David Wheadon, GlaxoSmithKline's senior vice president for regulatory affairs, thought Paxil did in fact have "remarkable" safety and efficacy. Wheadon, who had been corralled for the hearing along with top officials from six other drug companies, did some serious backpedaling. "We, in the case of Paxil, have not been able to discern a significant effect versus placebo ... ," he acknowledged.

  Earlier in the day, Rep. Joe Barton (R-Tex.), chairman of the Energy and Commerce Committee, had lashed out at the FDA for stonewalling his requests for information. "The conduct by the FDA has only reinforced my past sentiments that the Food and Drug Administration really stands for Foot Dragging and Alibis," Barton said.

  The FDA had dropped the ball at other key times as well, Barton charged. Even though the agency had been required by law since 2002 to make summaries of all pediatric clinical trials available to the public, it had not posted the summaries of the antidepressant trials on its Web site until very recently. Of those postings, "almost all of them were made just three weeks ago after I made a personal phone call to an individual at the FDA," Barton said. "Why did it take so long for the FDA to do its job?" he asked.

  The press pounced on another disturbing revelation from the hearing. In a bid to divert attention from themselves, executives from three drug companies testified that the FDA had urged them not to disclose on their SSRI product labels any information that might indicate problems with the drugs' safety. Wyeth Pharmaceuticals senior vice president Joseph Camardo said that the FDA actually reversed Wyeth's decision to amend its label for Effexor to include a warning that the antidepressant was associated with increased hostility and suicidal thinking among children.

  "Why would the FDA require a company to remove stronger labeling?" demanded an incredulous Rep. Greg Walden (R-Ore.).

  Dr. Janet Woodcock, the FDA's deputy commissioner for operations, was the government official on the hot seat that day. At first she tried to dodge Walden's question, replying that the FDA's job is to make the language in the drug labels "factual and based on the scientific data that is available." When Walden noted that the agency's own analysts had found scientific evidence of an increased risk of suicidal behaviors among study participants who had been prescribed the antidepressants, Woodcock responded that the FDA would be taking up those findings at its advisory meeting the following week. But that is all she would say. The head of the FDA's Center for Drug Evaluation and Research (which oversees drug labeling) was not about to venture a public opinion on the need for stronger labels just yet. Janet Woodcock had been in Washington long enough to know that top government officials who spoke out of turn didn't last long.

  IN FIVE DAYS, however, Woodcock and other FDA officials were singing a different tune. On September 14, the FDA's own advisory panel (a joint gathering of its psychopharmacological drugs advisory and pediatric advisory committees) decided that physicians and patients should be warned in the strongest possible terms that antidepressants did increase the risk of suicidal thoughts and behaviors in children and adolescents. Panel members voted 15-8 in favor of the FD
A's requiring that all nine antidepressants that had been reviewed carry "black box" warnings, so-called because the warnings are written in boldface type, surrounded by a black border, on the information sheet that physicians review when prescribing the drugs.

  TEICHER WAS DELIGHTED to read the news on the New York Times Web site the next morning. He found the panel's decision to include Prozac in its black box recommendations especially satisfying. He had been worried that Eli Lilly's blockbuster drug would be given a pass. Prozac was, after all, the only antidepressant officially approved for use in children and adolescents. Its patent, of course, had lapsed in February 2001, allowing generic versions of fluoxetine to compete with the brand. Teicher couldn't help wondering: was it just a coincidence that shortly after the patent on Prozac expired, Eli Lilly laid off his ex-wife, Bev, from her research job in Indianapolis? Bev was now back in the Boston area after four years in the Midwest. By the time she returned, his son was almost through college and his daughter was finishing up her senior year of high school. They no longer needed their father to drive them around.

  Teicher knew he would not soon forget what Eli Lilly and its various agents had done to him. But his concern that Prozac be included in the black box warnings was based strictly on scientific grounds. He believed that the published results of the Prozac pediatric trials had been flawed, in much the same way the Keller Paxil study had been. And he was not alone. In a 2004 article for the British Medical Journal, Australian researchers argued that the pediatric trials of Prozac done by Graham Emslie and his coauthors had not found the drug more effective than placebo on the study's primary outcome measures, only on the secondary outcomes.