UCSF-Harvard team publishes major HIV therapy study over objections of sponsor; compound shows no ef

By Jennifer O'Brien

Researchers led by University of California, San Francisco and Harvard School
of Public Health are reporting the largest randomized trial among HIV-1
infected persons conducted during the 1990s.  The compound tested in the trial,
intended to boost the immune response of HIV patients, had no effect on slowing
disease progression, the research team reports.

The researchers and their colleagues submitted their paper to JAMA, where it is
published in the November 1 issue, despite the attempt by the company
sponsoring the trial to block publication.

The company has filed an action with the American Arbitration Association
against the University of California and lead author James O. Kahn, MD, UCSF
associate professor of medicine in the Positive Health Program at San Francisco
General Hospital Medical Center, to block publication. The study, established
in 1995, was led by Kahn and senior author Stephen W. Lagakos, PhD, Professor
of Biostatistics at the Harvard School of Public Health.

The company’s action against the University of California and Kahn continues,
and includes a demand for $7 million to $10 million from Kahn and the
University, which the company contends would be the financial damage to the
company from publication of the findings.

The University of California has filed a counterclaim, arguing that the
agreements between the company and the University give the University the right
to publish the data, and that the company has wrongfully withheld the final
data from Kahn.

The decision to publish the study results affirms the University of California’
s right to publish unfavorable or neutral, industry-sponsored research

The trial was sponsored and funded by a private pharmaceutical company which
entered into agreements with UCSF (Kahn) and Harvard (Lagakos) in September
1995 to allow Kahn to act as the study chair and Lagakos to act as statistician
for a nationwide team of investigators in evaluating the clinical benefit of
the drug.

The drug investigated, HIV-1 Immunogen (known as Remune), is a disabled,
inactivated form of HIV.  It was administered in small doses to patients with
the hope that it would act as an immune stimulator, prompting the immune system
to ramp up its response against HIV proteins, and thus reduce the rate of
disease progression.

The intention was for the compound to supplement standard anti-retroviral
therapies.  These approved medications, including classes such as reverse
transcriptase inhibitors and protease inhibitors, lead to a decrease in HIV in
the blood and to an increase in the numbers of CD4+ T cells—white blood
cells that fight infection—and thus contribute to longer patient survival.
However, despite the enormous improvements these therapies have provided, the
risk for eventual HIV disease progression remains high.

Mild viral immunizations, with products such as the compound studied in this
trial, have been considered a low cost, well tolerated, potentially good
solution for augmenting anti-retroviral therapies, but their efficacy has
remained unclear. (Efforts to develop a vaccination that would wipe out or
prevent HIV infection altogether are also underway at many institutions,
including UCSF and Harvard). The current trial was intended to clarify the
efficacy of a specific compound developed from a disabled, inactivated form of

The trial involved 2,527 HIV-infected patients at 77 medical centers in the
United States. Most of the patients were receiving anti-retroviral therapies,
and none had ever developed the major characteristics associated with HIV’s
progression to AIDS, including low counts of CD4+ T-cells, and potentially
lethal opportunistic infections, such as pneumocystis pneumonia.

Half of the patients received the experimental drug, and half received a
placebo. Neither the patients, nor their physicians, knew which they received,
making this a so-called randomized, double-blind study.

The primary measure for the drug’s efficacy was HIV clinical progression or
death. The secondary measures were the quantity of the virus in the blood, CD4+
T-cell counts, and body weight.

The trial, begun in 1996, was intended to continue until all patients had been
followed for two and a half years. However, the study was terminated five
months early, in May 1999, after the second of three scheduled interim analyses
of the data by an independent Data Safety and Monitoring Board (DSMB) showed
that the compound had no clinical benefit and that it was unlikely any clinical
benefit would be shown by extending the study. The study leadership, which held
the responsibility for making the final interpretation of the results, agreed
with the DSMB recommendation, and the trial was stopped.

“Not only was there no indication of improvement, but the patients receiving
the drug and those receiving the placebo were so evenly balanced that there was
very little chance that continuing the study would yield a significant
difference between the two groups,” says Kahn.

Lagakos concurs. “It was clear the drug had failed to improve the conditions of
patients and that it was highly unlikely that any benefit would appear in the
final months of the trial.”

Following the termination of the trial, participating patients were seen for
their final scheduled clinical visits. The team did not receive data from these
final visits, but the researchers estimate that they did have 95 percent of the
confirmed clinical progressions that would have been available in the final

Still, in order to prepare a paper on their research findings, the researchers
sought the final data from the company that manufactures the drug and which
funded the study. However, the company refused to provide the final data unless
the researchers agreed to the inclusion of additional analyses specified by the
company. In addition, the company demanded the authority to approve all
publications. Because these conditions were unacceptable to the team, Kahn and
his co-authors prepared a paper based on the interim data analysis used by the

“Full disclosure and academic freedom is a nonnegotiable item with us.” says
Zach Hall, PhD, UCSF vice chancellor of research affairs. “That’s what we stand
for. It’s one of our core values. The University must vigorously defend the
right of its faculty to publish their results.”

“This is not just for issues of academic rights. We are doing this on behalf of
the more than 2,500 patients and our colleagues at the 77 sites. This is part
of our public service and responsibility to fully and fairly report on our
findings,” says Kahn.

In addition to maintaining University of California’s right to publish the
data, the University is also seeking an order that the data be turned over to
the study team.

“The final set of data would include information about how patients responded
to the drug and how patients developed responses while receiving different
anti-retroviral therapies, and different doses. Therefore, the full data set
could provide important information about the effectiveness of other HIV
treatments and could be very valuable as we plan other studies,” says Kahn.

The data also could have provided more information on any adverse reactions to
the drug, as assessed through the final patient visit and laboratory
evaluation, he says.

Moreover, notes Lagakos, the researchers were unable to obtain from the company
the addresses of the site co-investigators who participated in the trial, so
the manuscript and data could not be shared with them.

“We are publishing this data on behalf of the more than 2,500 patients and the
researchers at the 77 sites participating in this study, who were not allowed
to see the data as analyzed by the study team,” says Kahn. “Some of those
patients would have benefited from knowing that the drug failed—they could
have moved into other experimental drug trials.”

“This was a study that showed a product did not work,” he says. “It’s often
difficult to get studies of ineffective experimental drugs published in a high
profile journal.  The fact that we believed in the importance to publish this
study is an indication that we are informing other individuals who may need
these results in order to make an informed choice whether to participate in
other clinical studies involving this product.”

The study co-chairs were Dr. Kenneth Mayer, MD, a professor of medicine at
Memorial Hospital of Rhode Island and Brown University, and Dr. Henry Murray,
MD, a professor of medicine at Henry Weil Medical College of Cornell
University. The other co-author of the study was Deborah Weng Cherng, MS, a
biostatistician at the Harvard School of Public Health.