Firm Disputes Handling of Drug Study
The complex entanglements between pharmaceutical companies and academic medical researchers have taken a bizarre twist in which a Carlsbad company is accusing UC San Francisco of suppressing favorable data about an anti-AIDS vaccine developed by the late Dr. Jonas Salk.
Dr. James O. Kahn of UC San Francisco and his colleagues report in today’s Journal of the American Medical Assn. that one of the largest AIDS trials ever, involving 2,527 patients at 77 U.S. medical centers, found that the therapeutic vaccine, called Remune, did not prevent progression to AIDS or death.
The Immune Response Corp., which manufactures the drug, argues, however, that a subset of data shows that Remune lowers levels of HIV in the blood--but that Kahn refused to publish the favorable data.
The company has initiated legal action against the university, seeking $7 million to $10 million in damages to its reputation because the data were not published.
Other researchers in the study, who agree that the data should have been included in the published report, have also written to the journal to protest that they were never allowed to see the paper before it was submitted and that their names were left off it.
The charges and countercharges seem likely to muddy the future of what was once viewed as a promising way to stimulate the body’s own immune response against the AIDS virus. And a resolution does not seem imminent.
“The product doesn’t work,” Kahn said. “Patients need to be informed. Colleagues need to be told that. We need to move on.”
“I’m stunned,” said Dr. John Turner of Graduate Hospital in Philadelphia, who provided many of the patients in the study. Remune “is not a cure. None of the drugs out there are a cure. But it is dramatically efficacious and useful in conjunction with other drugs.”
Remune, also known as HIV-1 immunogen, is essentially a killed virus that has been stripped of many of the proteins on its surface. Although it is called a vaccine, it is used to stimulate a beneficial immune response in patients who are already HIV-positive, not to prevent an infection.
The Kahn trial represents a case in which a carefully planned clinical trial was overwhelmed by unforeseen events. “The real world crashed in on this study,” said Dr. Daniel Pearce of the St. Luke Medical Group in San Diego, another study investigator.
When the study was conceived, the Food and Drug Administration required that potential AIDS drugs show an actual decrease in progression to AIDS or death before the drugs could be approved. Shortly after the study began, however, the FDA switched gears and allowed the use of what are called surrogate markers. That meant a drug could be approved if it simply lowered levels of the virus in the blood.
The disputed data involve blood virus levels, said Dr. Ronald B. Moss, the company’s medical director.
The bigger problem with the study was statistical. In any medical study, the smaller the effect you attempt to demonstrate, the larger the number of patients that must be studied.
In this case, the researchers designed the study to detect a 50% increase in survival produced by Remune on the assumption that the normal death rate for AIDS would continue at the 1995 rate of about 6% per year. Twenty-five hundred patients would have been enough to show such an effect.
Shortly after the study began in 1996, however, a new class of drugs called protease inhibitors was introduced and the death rate dropped below 1% per year.
At that death rate, said Dr. Dennis Carlo, president of Immune Response, detecting a 50% increase in survival would have required six to 10 times as many patients, a prohibitively large number.
But Kahn disagrees: “We had 106 people who progressed [to AIDS or death] during the study, and that gives us great confidence” that they would have seen a benefit if it were there. Half of those who progressed were in the group receiving Remune and half in the placebo group.
Based on those numbers, the study’s Data Safety Monitoring Board recommended in May 1999 that the study be halted prematurely because it could never prove any benefit. “The study was doomed from the start because the whole method of treating the disease changed during the trial,” said Dr. Robert “Chip” Schooley of the University of Colorado Health Sciences Center. “They recognized the futility of continuing,” he said.
That decision had an immediate effect on Immune Response. The company fired 43 employees--nearly a third of its staff--and many others were forced to take pay cuts.
But the company still saw a ray of hope in one subset of data on 250 patients in which blood virus levels were measured twice as often as in the rest of the patients. Moss said these data show that Remune lowered virus levels by about a factor of 10. He contends that this effect was masked in patients who were monitored less frequently because the patients changed drug regimens often.
Moss and Turner agree that this effect was discussed at a meeting of investigators last summer and that the majority of the researchers thought it should be included in the final report on the study.
But Kahn refused to include it. “They were trying to find some positive spin by dredging the data for something [good about their product], and that does not belong in the manuscript,” he said.
“If they want to report that data, let them publish their own paper,” he said.
When Kahn would not agree to include the disputed data, the company refused to provide him with the data from the final examination of each patient. He therefore published his report with incomplete data, although both sides agree that did not change any conclusions about the product’s efficacy.
That Kahn did not show a draft of the report to the study’s other researchers is virtually unprecedented.
Kahn says that the company would not provide him with the names and addresses of all the other investigators so that he could share the manuscript with them. Since he could not show it to all of them, he did not show it to any of them. “It’s unfair to selectively send it out to only certain people,” he said.
Moss said Kahn’s coauthor, Dr. Stephen Lagakos of Harvard Medical School, had all the information on names and addresses that Kahn requested.
In an editorial accompanying the paper, journal editor Dr. Catherine DeAngelis said the dispute illustrates what can happen when disagreement erupts between researchers and a sponsor who “has a proprietary interest in the findings.” The paper, she said, “stands on its own scientific merit.”
The original contract between UC San Francisco and Immune Response Corp. calls for binding arbitration in the event of a dispute, said UC lawyer Chris Patti. In September, the company demanded arbitration, he said, and the university is now choosing from a list of arbitrators provided by the American Arbitration Assn.
Whatever the arbitrator decides, this is probably not the final nail in the coffin for Remune, Schooley said. “It’s very safe, and we’ve had a very large experience with it over the last decade. It is seen as inducing several immune responses that you can measure in the laboratory. But the jury is still out about whether that immune response helps patients control the virus.”
