CROI: HIV Resistance Linked to Hepatitis Drug

LOS ANGELES, March 1 -- A drug for patients infected with hepatitis B virus, an agent that was thought to be inactive against HIV for those co-infected with that virus, may not be so innocuous with HIV patients after all, investigators suggested here.

The drug, called Entecavir (Baraclude), taken by HIV-positive patients who are not on HIV therapy, appears to inhibit HIV replication, according to Chloe Thio, M.D., of Johns Hopkins. It also appears to lead to the emergence of HIV resistance mutations that may compromise future HIV therapy.

The finding comes from three case studies, followed by sensitive laboratory testing, Dr. Thio told an oral abstract session at the Conference on Retroviruses and Opportunistic Infections.

The case reports were enough that the FDA and Bristol-Myers Squibb, the drug's maker, notified health care professionals earlier this week that the possibility of creating HIV resistance "cannot be excluded based on current information."

Commenting during the question period after the presentation, Richard Colonno, Ph.D., a Bristol-Myers Squibb scientist, said the company tested the drug extensively for 10 years before its approval last March.

"We've assayed the compound against HIV over that 10-year-period probably dozens of times," said Dr. Colonno. "In no case did we see any activity against HIV below micromolar concentrations."

In its most recent guidelines, the American Association for the Study of Liver Diseases said the drug should be used in co-infected patients not currently being treated for HIV.

But that opinion may now have to reviewed, according to Margaret Shuhart, M.D., of the University of Washington in Seattle, who is the chair of the society's practice guidelines committee.

She cautioned that the report is still only an oral abstract and the data need to be reviewed carefully before the guidelines are changed. "In the end it is an n of 1," she said.

Dr. Shuhart was referring to the fact that detailed information was presented on only one patient, a 31-year-old man, although some details were given on two others.

Dr. Thio said the report began with the observation of two patients whose HIV viral load dropped by at least a factor of 10, simultaneously with the administration of entecavir, although neither was being treated for the HIV infection.

"That made me wonder if there was activity against HIV," she said. The researcher later found a third patient with the same pattern.

The researchers used a single cycle infectivity assay to test the drug against HIV and compared its effect to lamivudine (3TC, Epivir) and zidovudine (AZT).

All three drugs inhibited viral replication to below 50%, but entecavir reached the 50% level 500 times faster than AZT and 50 times faster than 3TC. Oddly, however, while the other two drugs increased their effect with increasing doses, entecavir's effect flattened out at higher concentrations.

"It's a very unusual dose-response curve," Dr. Thio admitted, adding that she and colleagues don't yet know what to make of it.

The researchers also analyzed HIV isolates from the first patient (testing of the others is under way) and found that the virus showed no sign of mutation at the beginning of entecavir treatment.

But at four months, 60% of HIV isolates showed the M184V mutation in the HIV reverse transcriptase enzyme and by six months almost all showed the mutation.

The M184V mutation confers resistance to lamivudine and emtricitabine (Emtriva).

In the infectivity assay, Dr. Thio said, entecavir showed no effect against the mutated virus.

The three lines of evidence together are convincing, Dr. Thio told reporters later. "These are very robust findings," she said.

Dr. Shuhart said the report is "provocative," but probably won't change her practice. Faced with a co-infected patient needing HBV treatment, "I think I would use it, and I would monitor the patient carefully," Dr. Shuhart said.