HIV Drug Holidays Get New Support

GENEVA, Aug. 4 -- The notion of HIV-treatment interruption -- consigned not long ago to the ash heap of failed ideas -- should be revisited, according to international investigators.

In the so-called Staccato trial, HIV patients were randomized to either continuous highly active anti-retroviral therapy (HAART) or to treatment interruptions when their immune system was sufficiently robust, according to Bernard Hirschel, M.D., of Geneva University Hospital.

The study found that treatment interruption reduced drug use without affecting control of the virus and without the development of resistance mutations, Dr. Hirschel and colleagues reported in the Aug. 5 issue of The Lancet.

The researchers randomized 430 patients in a 1:2 ratio to continued therapy or scheduled treatment interruptions. To be eligible for the study, patients had to have a CD4 count greater than 350 cells per microliter of blood.

Treatment interruption was guided by the CD4 count -- if it dropped below 350 cells per microliter, HAART was restarted and continued until the count was again above 350. Fifty percent of patients in the treatment interruption group had restarted HAART by 18 weeks after randomization.

After a median of 21.9 months of randomized treatment:

• Patients in the treatment interruption group used anti-retroviral drugs on 37.5% of days in the study, compared with 99% of days for those on continuous treatment.
• 90.5% of patients in the treatment interruption group reached a viral RNA load of less than 50 copies per milliliter of blood, compared with 91.8% in the continued treatment group. The difference was not statistically significant.
• There were no AIDS-defining events in either group.
• Diarrhea and neuropathy were more frequent with continuous treatment, while oral and genital candidiasis was more frequent with treatment interruption.
• Ten patients -- or 2.3% -- developed resistance mutations, but there was no significant difference between groups.

The finding contrasts starkly with the SMART study, another large international trial evaluating treatment interruption, which was stopped in January because patients in the intermittent treatment arm had twice the risk of disease progression as those on continuous therapy. They also had an increased risk of major complications, such as cardiovascular, liver, and kidney disease.

The rates seen in the SMART study would have produced 17 AIDS-defining events or deaths in the Staccato treatment interruption group, Dr. Hirschel and colleagues noted, but no such events -- and only one death, from colon cancer -- were actually seen.

"It would be wrong to conclude that SMART means the end of all HIV treatment interruption," Dr. Hirschel and colleagues said. However, "such a striking discrepancy is unlikely due to chance and requires explanation," they said.

Part of the explanation may be that the CD4 treatment guideline in Staccato was 350 cells per microliter of blood, compared with 250 cells in SMART, which would tend to reduce the time spent with low CD4 counts, the researchers said.

Another factor that may be involved, they said, is that patients in SMART had a median 72 months on HAART before they started the treatment interruption program. By contrast, Staccato patients had a median 16 months. Also, the type of HAART used was different between the trials, "but the relation between those differences, and the apparently different outcome, is not obvious," Dr. Hirschel and colleagues said.

The bottom line, the researchers said, is that "Staccato's results provide reassurance about the one risk that was widely feared when these trials began -- development of resistance and loss of efficacy of treatment."

In fact, they suggested, if the starting CD4 count were sufficiently high -- more than 500 cells per microliter of blood -- "treatment interruptions lasting many months" might be safe, with a consequent saving in drug costs and reduction in drug-related side effects.