Structured Treatment Interruptions

The benefit of structured treatment interruptions (STIs) for HIV has been the subject of a lot of debate. Interrupting antiretroviral treatment is an enticing idea to many people living with HIV and to healthcare providers. STIs would limit drug exposure and therefore decrease drug toxicity, drug costs, and possibly decrease drug resistance and failure rates. It was thought that STIs would also lead to "autovaccination" and boost anti-HIV immunity, as a result of repeated episodes of recurrent HIV in the blood (viremia), which in turn would result in persistent low HIV viral loads (VL) even off therapy. Unfortunately, small studies have resulted in disappointing results. Autovaccination, although it may be helpful in acute HIV infection, does not appear to improve HIV-related immunity in the chronically infected patient. More importantly, STIs resulted in a risk of significant declines in CD4 cell counts.

Bernard Hirscel presented the results of one of the largest cohorts of chronically infected HIV+ people (133) who have undergone STIs in the Swiss and Spanish Intermittent Treatment Trial (SSITT) (Abstract 528). Individuals were on stable HAART therapy (no NNRTIs used) with a suppressed virus load for at least 6 months and a CD4+ cell count above 300. Participants underwent four cycles in which HAART was alternately stopped for 2 weeks and then restarted for 8 weeks. Treatment was then discontinued indefinitely at week 40. Participants were allowed to restart HAART if pre-determined thresholds for virus load and CD4+ cell counts were exceeded. Viral "rebound" was defined as a load above 200 copies/mL and viral "response" was defined as a load below 5000 copies at 52 weeks.

Eighty-eight of 133 participants had a rebound after the first 2 weeks of interrupted treatment. Only 17 percent of the participants responded at week 52 which declined to 11 percent by week 96. Forty-nine patients remained off HAART after a median of 44 weeks. CD4 cell counts declined by 175 cells/mm3 during the first 12 weeks off treatment and then stabilized. In all patients except one, the viral load was promptly resuppressed when the previous HAART regimen was resumed. HIV-specific immunity increased by week 52, but did not correlate with response. This study, along with many others, disproved the benefit of autovaccination, since the patients with greater anti-HIV immune responses actually had higher viral rebounds.

The authors concluded that a number of patients with HIV infection on suppressive HAART could be safely managed off HAART for several months, but that STIs alone are rarely enough to maintain a low viral load when HAART is stopped. It is clear that STIs were not effective in controlling HIV infection in the majority of patients. However, the measure of success is dependent on the goals of treatment. The patients in this study were able to be off HAART for several months with prompt resuppression of their HIV after resuming HAART, making drug holidays a viable option to diminish side effects and costs. However, these patients all suffered a substantial drop in their CD4+ cell counts.

Jens Lundgren, from the EuroSIDA cohort study, reported new and interesting data on the risk of developing a new AIDS-defining event after interrupting or stopping HAART (Abstract 48). Of 5,385 patients who initiated HAART, 776 patients (20 percent) interrupted treatment for a multitude of reasons, although 518 of these patients re-initiated therapy sometime later. These drug interruptions were not scheduled or structured. The rate of HAART interruption was 21 percent for every 5 years and was more frequent in women and IV drug users. During the first 3 months off HAART there was a median drop in the CD4+ cell count of 30 cells, and an increase in the HIV viral load of 1 log (a 10-fold increase). Significant progression of clinical disease off HAART increased by more than five times, especially in patients with low CD4+ cell counts (i.e., people with a CD4+ cell count below 50 had an 80 percent risk of developing an AIDS-defining illness off therapy compared to only a 30 percent chance on HAART). It appears that HAART has a beneficial clinical effect not explained by the CD4 cell count and the HIV viral load even when treatment is failing. This data also suggests that STIs are not clinically safe in patients with low CD4 cell counts.

Other prospective, controlled studies are in progress. These are looking at different STI schedule strategies (such as determining interruptions by the calendar, by a threshold CD4 cell count, by viral load, or by comparing long and short interruptions) and evaluating STI in combination with additional immune stimulatory maneuvers (such as vaccines and cytokines). These studies will help to determine if STIs are a reasonable strategy to reduce long-term exposure to antiretrovirals, thereby decreasing associated costs and side effects.