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

by Alison Bass

  Howard had envisioned Tangredi-Hannon as a tall, stern woman dressed in a dark suit, her hair caught in a tight bun. Instead, a small woman with short brown hair and expressive eyes waved her into the plush leather passenger seat. As Tangredi-Hannon skillfully negotiated the curves of Route 195 high above Providence, she wasted no time on chitchat.

  "God, you work for Marty Keller. That can't be easy," she said.

  Howard laughed. She felt immediately comfortable with this woman.

  "Oh, I hardly ever see him," she said. "I've only met with him twice."

  Keller enjoyed an elegantly updated suite of offices in the Sawyer Building, a Victorian brick building across campus from the psychiatry department's more dilapidated research quarters in Duncan. He communicated with Howard and the other support staff mostly by fax. It was not uncommon for Peg Ciarlone, his secretary, and Howard to fax documents back and forth twenty, thirty times a day. But even in absentia, Keller's presence cast a huge shadow on his research staff.

  DR. MARTIN KELLER had come to Brown in 1989 from Massachusetts General Hospital and Harvard Medical School, where he had a reputation for pulling in substantial grant money for multisite research studies. As head of psychiatry at Brown, Keller was tasked with building an academic research department where none had existed before. That meant bringing together often-feuding research groups from the six Rhode Island hospitals affiliated with the university's school of medicine. By the time Howard joined his staff in the fall of 1994, Keller had largely succeeded in his mission-with a few bumps along the way. He had assembled a drug research empire that rivaled anything his former Harvard colleagues could boast of, bringing in millions of dollars each year from federal research institutes and pharmaceutical companies to study mental illness and promising drug treatments. While Keller's single-minded focus on raising money endeared him to the dean of Brown's medical school and other university officials, the way he conducted business did not sit so well with some of his colleagues. A few years earlier, Jonathan Cole, then head of psychopharmacology for McLean Hospital, had sat in on a committee chaired by Keller, who was at Brown by then. Keller was principal investigator of a $1.4 million Harvard-Brown study funded by the Upjohn Company to study people who had been diagnosed with anxiety at twelve medical centers and see how they did, on and off certain medications. Cole took an instant dislike to the way Keller presided over the committee meetings. As the senior McLean psychiatrist would later recall, Keller ran the meetings "like an emperor." He would give orders and then zip off in the middle of the meeting to take phone calls. "He raised a hell of a lot of money from Upjohn and was happy to plan the study, but he didn't seem particularly interested in making sure it went all right," Cole said.

  A few years later, Howard saw her boss display the same kind of cavalier behavior toward his purported colleagues. In June 1995, the research heads from the six hospitals affiliated with the psychiatry department gathered one afternoon to pick the winners of Brown's annual internal competition for research grants. The meeting was the culmination of a three-month-long process to decide which research proposals should receive coveted seed money from the university and its affiliated hospitals, grants ranging from $25,000 to $100,000. The psychiatric chiefs from Butler, Rhode Island, Bradley, and Miriam hospitals, along with other senior researchers in the department, had devoted hours to weighing the fifty applications, with the understanding that the ten top-ranked proposals would receive funding. As Keller's liaison to the group, Howard sat in on every one of those meetings, and it had fallen to her to submit the top-ranked proposals to Keller, by fax of course. As Brown's chief of psychiatry, he held the purse strings for almost all the research conducted at Brown and its affiliates. He thus wielded the final vote on grants. Yet he had not attended a single meeting, even though the group, known as ECOR (Executive Committee on Research), met every few weeks in a small conference room adjoining his office suite.

  On this day, however, Keller made an appearance. As the group sat chatting among themselves, he swept in, wearing an exquisitely tailored suit. He flung the faxed list of proposals across the conference table and, striding over to the whiteboard, started scrawling down research ers' names. Howard recognized several of these new names: they were researchers at Butler Hospital, where Keller was psychiatric chief. None of them had been included in the committee's top-ranked list. She also noticed that the researcher whose proposal the group had ranked number one was not even on Keller's list.

  At first, a baffled silence permeated the room, and then one of the research heads piped up.

  "I don't understand. You haven't included Gary Epstein's proposal," said Ron Seifer, research chief at Bradley Hospital. Seifer was referring to the most favored proposal, which belonged to a young researcher at the Veterans Administration who was doing creative work in molecular genetics.

  Charlie Marotta, a neuroscientist and the director of the Psychiatric Research Division at Brown, nodded. "Epstein is doing some very important research," he said. "He's putting in unbelievably long hours."

  "We're not going to fund anyone from the VA," Keller responded. "The VA's got its own money."

  Dave Abrams, the research head at Miriam, one of the Brownaffiliated hospitals, spoke up next. He was a compact British-born scientist with a nationally recognized reputation in behavioral research. In 2004, he would leave Brown to become the director of the Office of Behavioral and Social Sciences Research for the National Institutes of Health.

  "I don't see Bess Marcus's name on the board," Abrams said, referring to another young researcher who worked with him at Miriam Hospital and also ranked in the top ten. "She's doing some very valuable research on the preventative benefits of physical activity."

  Keller stopped writing on the board. He turned around.

  "Look, the dollars have got to be divided up proportionately to the hospitals that put money into the pool," he said. "I realize this is different from how we've done it before, but this is the way it's going to be from now on."

  The research chiefs looked at one another. Finally Abrams said, "It would have been helpful if we had known this before we spent all our time rating the candidates according to merit. This was a very laborintensive process."

  Keller shrugged. "I realize that," he said. "But the rules have changed"

  A few minutes later, he left the room.

  In the silence that followed, Howard could feel the anger rising among the men around the table. They waited until Keller's footsteps had receded.

  "We really got screwed over here," Seifer said.

  Marotta, who had worked with Keller at Harvard, jerked his head in agreement.

  "If I had known this, I would have advised my people to not even apply," he said bitterly. "This is just the way it was at Harvard"

  Now, GLANCING AT the woman steering the car beside her, Howard wondered how much Alice Tangredi-Hannon knew about the enmity between Keller and the hospital chiefs. Probably a good deal. She struck Howard as politically savvy. As the decaying textile mills of New Bedford flashed by, Tangredi-Hannon chattered about the house she and her husband had just finished building in Marion. It was the dream getaway that they had worked hard toward all their lives.

  "We don't have kids," she said, as if to explain why she was spending so much money on a house. "We have cats"

  She paused, biting her lower lip. "Or rather, we have one cat. My Persian died last week. It was horrible!"

  Howard knew exactly how her companion felt.

  "I'm so sorry. It's awful to lose a pet," she said. "They're like part of your family."

  Tangredi-Hannon nodded but kept her eyes on the road.

  "I love cats" Howard paused, wondering what the other woman would think of her pet menagerie. "We have three cats and two dogs. Maria loves animals."

  "Maria?"

  "That's my daughter. I adopted her let's see-seven years ago"

  "How wonderful!" Tangredi-Hannon's voice radiated warmth. "She must bring you a lot
of joy."

  Howard paused. She wasn't quite ready to tell the research director about her daughter's illness and its dizzying ups and downs.

  "Yes, she does," Howard said. "She's the most important thing in my life."

  BEFORE HOWARD KNEW IT, they had turned off Route 6 onto the half-dirt, half-paved road that led to her house. She felt a twinge of embarrassment at how the cottage must look to Tangredi-Hannon. The roof had been patched, and the ivy, honeysuckle vines, and wild roses creeping up the sides did a good job of masking the sagging walls. But the lilies, white yarrow, and hollyhocks in the front yard looked wild and untended. One of her dogs was batting playfully with a small kitten in the sand-eroded driveway.

  Tangredi-Hannon cut the engine and turned toward Howard. "Do you need a ride to work tomorrow? I'd be happy to give you one"

  "Oh, that's okay. I can rent a car from the garage down the street. They give me a good rate"

  Tangredi-Hannon looked appalled.

  "Oh no, I can't let you do that! I enjoyed your company. I'd be happy to drive you until your car is fixed. Really."

  HER FIRST FEW MONTHS at Brown, Howard had been buried beneath an avalanche of paperwork. Not only was she new to the department, but the woman she reported to, Carolyn O'Sullivan, the department administrator, had arrived only a few months before and was still trying to get a handle on things. Sometimes Howard felt as if she were on an archaeological dig, sifting through layers of reports, memos, and faxes detailing all the research Martin Keller had his hand in. There was the ongoing study of the psychobiology of depression, funded by the National Institute of Mental Health (NIMH), which Keller had landed while still at Mass General; a longitudinal study of eating disorders, also funded by the NIMH; a study of anxiety disorders, funded first by Upjohn and now NIMH; and a more recent five-year, NIMH-sponsored study of whether lithium could prevent the recurrence of manic depression in adolescents. In 1993, the pharmaceutical giant Pfizer had given Keller more than $1 million to coordinate a six-year, multicenter study of Zoloft in treating chronic depression. And then there was the clinical trial funded by another drugmaker, SmithKline Beecham, to compare the effectiveness of Paxil with placebo and an older tricyclic antidepressant in adolescents. As one of six sites, Brown had been awarded an initial $800,000 in 1993 for the Paxil study, and Keller was principal investigator for that one too.

  Keller, it was clear, had jumped aboard the fast-moving juggernaut of drug companies eager to capitalize on the success of Prozac. After the FDA dismissed the Prozac-suicide link and approved Zoloft and Paxil in quick succession, other pharmaceuticals hurried to develop their own SSRI products. Meanwhile, the makers of Prozac, Zoloft, and Paxil, recognizing the potential for even greater market share, began positioning their drugs for use in other populations and disorders (besides depression). Lilly, Pfizer, and SmithKline Beecham all had their sights on the under-eighteen market, and by the mid-1990s, all three had embarked on what the FDA called "the gold standard" of drug research: comparative studies that randomly assigned young patients (without their knowledge) to one of two groups, with one group taking the new drug and the other taking a sugar pill or placebo. The clinical trials Marty Keller was leading on Paxil and Zoloft also randomly assigned patients to a third "blinded" group for comparison's sake: those taking imipramine, an older tricyclic antidepressant. For some new medicines, the FDA only required proof of efficacy over an older drug; but with most medications, and especially with psychoactive drugs like Prozac and Paxil, the FDA wanted proof of efficacy over placebo. The human mind, after all, is a tricky thing, and past research had revealed a surprisingly high rate of response to placebo among people with depression and other mental illnesses. Perhaps, some experts speculated, the symptoms of depression were alleviated just by virtue of being involved in an intensive clinical trial, with all the extra support and attention that such studies provide. To win FDA approval, the new drug had to work significantly better than placebo in two different randomized studies.

  The federal agency would soon give drug companies another incentive to test these drugs on children and adolescents. In 1997, Congress would enact the FDA Modernization Act, which promised that drug companies conducting pediatric clinical trials would get an additional six months of patent protection from generic competition, whether or not the drugs they studied were found to be effective for use in children. Over the next decade, such extended patent exclusivity would earn the SSRI makers millions of dollars in additional revenue.

  Doing such studies served another important purpose. If the results of the trials were positive, the drug companies would encourage researchers to present their findings at medical conferences and publish them in respected medical journals. This kind of publicity was priceless: it convinced many physicians that the drugs being studied were safe and effective enough to prescribe for non-FDA-approved uses. It was this very strategy that would permit drug companies like Pfizer, Lilly, GlaxoSmithKline, and Forest Labs (the maker of Celexa) to vastly extend their market share, as physicians flocked to prescribe the SSRIs for a host of off-label uses, ranging from depression in children to obsessive-compulsive disorder, generalized anxiety disorder, and social phobia. And indeed, as soon as the results were in on the Paxil trial that Keller had spearheaded, its pharmaceutical sponsor would hatch plans to submit the multisite study to one of the most prestigious medical journals around: the Journal of the American Medical Association (JAMA).

  In 1995, however, Brown was still enrolling participants in the Paxil study, and Howard found herself spending more and more time in the department's dusty file room on the second floor of Duncan, reading closely as she sifted through documents and filed them away. Her file room explorations and conversations with co-workers convinced her that Keller was playing fast and loose with the protocols for the Paxil study and another clinical trial. She had noticed, for instance, some discrepancies in the documentation for the Paxil trial. Participants in such studies frequently develop adverse side effects from taking a new drug, some of them serious enough to require hospitalization. Such side effects could range from suicidal thoughts and hostile behaviors to dizziness, nausea, headaches, and chest pain. According to FDA research guidelines, research investigators had to document each and every one of these "serious adverse events"-known in the trade as SAEs-in memos to both the drug company and the research site's Institutional Review Board. And they had to make an educated guess as to whether that adverse event was caused by the drug under study.

  Howard suspected that researchers on the Paxil trial were not accurately coding these adverse events. According to a trail of faxed memos, at least two adolescent patients enrolled in the Paxil trial had been yanked from the study after threatening or attempting suicide. Yet on several memos submitted to Brown's Institutional Review Board, Keller wrote that these teenage girls "withdrew during the acute phase for reasons of noncompliance." Howard wondered how the girls' suicidal behavior could be considered noncompliant. Wouldn't it be more accurate to describe their behavior as an adverse effect of the drug? Having read so much of the literature on depression herself, Howard knew that miscoding even a few suicidal patients in this kind of study could skew the results.

  She had also stumbled across inconsistencies in yet another Kellerheaded study. This one involved taking manic-depressive adolescents off lithium and then putting them back on it to see if the drug prevented the recurrence of their disorder. It seemed as if the researchers were having trouble recruiting and keeping teenagers in the study. No surprise there, Howard thought: what parents want their kids to be yanked off meds that are helping them? Yet the number of participants listed in the grant renewal proposal didn't seem to reflect the actual number of teenagers in the lithium study.

  Howard had also unearthed copies of several invoices to the Massachusetts Department of Mental Health requesting payment for a study of chronically ill schizophrenic patients at the Corrigan Mental Health Center, a state-funded hospital in Fall River, Massach
usetts. These invoices baffled her. She couldn't find any documentation for the Corrigan study itself informed consent protocols or memos indicating who was involved in the study or how long it had been ongoing. Theo Manschreck, a schizophrenia researcher from McGill University, had arrived at Brown a few months after Howard, apparently to do research at Corrigan Mental Health Center. But the payment invoices to the Department of Mental Health in Massachusetts, which apparently had a contract with Brown for the research at Corrigan, indicated that the study had started three years earlier. From what Howard could divine, Keller first obtained $70,000 from the Massachusetts mental health agency in 1992 and slightly higher amounts each of the succeeding three years. The invoices indicated that Brown had received more than $200,000 thus far from the Commonwealth of Massachusetts for research at Corrigan Mental Health Center.

  Whenever Howard asked Carolyn O'Sullivan about Corrigan, the harried administrator would say, "I don't know anything about it. Just send the invoices over to Peg"

  Once, after a meeting of ECOR, when some of the hospital chiefs hung around to exchange the latest gossip, Howard had turned to one doctor and asked if he knew anything about Brown's involvement in a schizophrenia study at Corrigan. No, he said, Brown isn't affiliated with Corrigan. Another psychiatrist said much the same when she asked him about it a few weeks later. He had never heard of Corrigan Mental Health Center, he said.

  By the time Howard's car broke down that Tuesday in September, she had become convinced that the psychiatry department at Brown was getting paid for research it wasn't doing. She also suspected that Keller had submitted falsified invoices to Massachusetts state officials. There were names of researchers on those invoices who Howard knew had nothing to do with a state hospital in Fall River.

  Howard had discussed her concerns with several co-workers but had been advised not to say anything. The last time one of Keller's employees had spoken out-about his boss's apparent double-billing on travel expenses-the whistleblower had been demoted and moved against his will to another department at the university. An internal audit had confirmed the double-billing: Keller seemed to have made a practice of getting reimbursed by both Brown and his pharmaceutical company sponsor for the same trips. State police, who were also investigating Keller's billing practices at the time, acknowledged that Brown had uncovered evidence of overbilling by Keller that amounted to several thousand dollars. In the end, the state police investigation was dropped at the request of Brown officials, and Keller was required to refund the university just under one thousand dollars. But that didn't help the whistleblower's career at Brown. He left under a cloud soon afterward.