CD4-guided structured treatment interruptions safe in patients at high risk of resistance

Patients at high risk of developing drug resistance can undergo CD4-guided structured treatment interruptions without developing major HIV drug resistance mutations, according to a small Thai study published electronically in Clinical Infectious Diseases on 4^th February.

Although the study was small, these findings suggest that CD4-guided treatment interruption may be a suitable alternative to continuous treatment in order to reduce the cost and side-effects of anti-HIV therapy, particularly in resource-limited settings.

A major concern of the use of structured treatment interruptions is the risk of resistance to antiretroviral drugs developing during repeated cycles of starting and stopping drug therapy. This may limit future treatment options.

To assess this in patients at high risk of drug resistance, investigators from the HIV Netherlands, Australia, Thailand Research Collaboration (HIV-NAT) recruited 23 HIV-positive Thai patients who had previously participated in the HIV-NAT 001 trial series that started in 1997. The previous trial consisted of a dual nucleoside analogue (NRTI) combination of AZT (zidovudine, Retrovir) and ddC (zalcitabine, Hivid) at standard or half dose for 48 weeks.

These patients went on to receive three years of highly active antiretroviral therapy (HAART) with ritonavir-boosted soft-gel saquinavir (Fortovase), with either AZT and 3TC (lamivudine, Epivir) or d4T (stavudine, Zerit) and ddI (didanosine, Videx / VidexEC). Patients maintaining viral loads below 50 copies/ml and CD4 cell counts above 350 cells/mm³ were eligible to enter the structured treatment interruption study.

The patients stopped taking their HAART regimen and CD4 cell count was monitored every two months. When a patient’s CD4 cell count fell to below 350 cells/mm³ or 30% of the CD4 cell count before treatment interruption, therapy was reinitiated until CD4 cell count rose above this threshold again. Over 48 weeks’ observation, the patients underwent between one and three interruptions, with a mean of 1.65 interruptions per patient.

The investigators carried out genotypic resistance tests on virus isolated from the patients when they switched from dual NRTI therapy to HAART, and again when HAART was stopped during the final interruption, a median of 32 days after stopping the antiretrovirals.

“Major HIV mutations were not induced through CD4 cell count-guided treatment interruptions in HIV-infected patients successfully treated with HAART after dual-NRTI therapy,” they conclude.

Samples were available for 20 (87%) of the patients at each time point. However, only eleven of the pre-HAART samples could be analysed, with the remaining nine having degraded or being from patients with viral loads below 400 copies/ml. Major mutations in the reverse transcriptase gene, which can cause drug resistance, were found in four (36%) of these samples. In contrast, minor mutations were found in five (45%) of the samples. Minor mutations do not cause resistance alone, but can contribute to drug resistance when present with other mutations.

Seventeen of the samples obtained after treatment interruption could be genotyped, nine (53%) of which contained resistance mutations. Only one (6%) of these was a major mutation and the remaining eight (47%) were minor.

Only one major drug resistance mutation occurred (T215Y) in a patient who did not have resistance mutations before treatment interruption. Conversely, the four patients who had pre-existing major mutations lost these mutations after structured treatment interruption.

“In this CD4 cell count-guided treatment-interruption trial, HIV mutations were frequently observed after structured treatment interruption in patients previously treated with dual-NRTI therapy before HAART,“ the authors comment. “However, only one of these mutations was a major mutation.”

The only mutations present in the protease gene were minor. In contrast to the expected increased in resistance mutations, their frequency was not increased during treatment interruptions (72 vs. 65%). Conversely, the frequency of mutations in the reverse transcriptase gene decreased significantly during treatment interruptions (72 vs. 12%, p

“The frequency of reverse-transcriptase gene mutations significantly decreased after structured treatment interruptions, and there was a trend toward a reduction in the number of pre-existing major mutations in the HIV reverse-transcriptase gene after structured treatment interruptions,” the authors explain. “Whether these mutations truly disappeared or whether we just measured overgrowth of wild-type virus is difficult to determine."

“Although we cannot conclude from our data that CD4 cell count-guided structured treatment interruption is beneficial for such patients, we did not find any harmful effects of structured treatment interruption on the virological outcome.”

After 96 weeks of CD4 cell count-guided treatment interruptions, all of the patients started therapy continuously for twelve weeks. At the end of this follow-up period, one patient in the structured treatment interruption group had developed virological failure, defined as a viral load above 500 copies/ml after three months of continuous HAART. This was found to be due to poor adherence.

All of the remaining patients had viral loads below 400 copies/ml at the end of follow-up, despite the presence of a major mutation in one patient. In a comparator group of 25 patients receiving HAART continuously for 96 weeks, one patient experienced virological failure.

“After 108 weeks of follow-up and with only one treatment failure, CD4 cell count-guided structured treatment interruption appears to be safe in this patient population,” the investigators conclude.

The study also contained a third treatment arm consisting of the same HAART regimen taken in a cycle one week on therapy, followed by a week’s interruption. However, due to a high rate of treatment failure (12 [46%] of 26 patients), this “week on, week off” arm was suspended after 72 weeks.

In an accompanying editorial, Roberto Arduino of The University of Texas Health Science Center at Houston calls for caution in the use of structured treatment interruptions until data from larger trials such as the SMART Study are available. “Given the established benefit of continuous HAART, we need much more definitive data on CD4 cell count-guided strategies before they should be recommended in clinical practice,” he warns.