Summaries
Track B: Clinical Research, Treatment

In developing countries, tuberculosis (TB) is the commonest serious infection in the HIV-infected. TB is treated with four drugs; after two months, two of the drugs are stopped whereas the other two are continued for another four months. But HIV/TB co-infected patients also need three anti-HIV drugs. Can all of these be given together? Will the mix be toxic? Will HIV-treatment interfere with TB treatment and vice versa?

A presentation of Manosuthi from Thailand, MoOrB1013 used the most common TB treatment together with efavirenz, 3-TC, and d4T. In the average patient, TB treatment was started first, and HIV treatment added after approximately two months, and this was well tolerated. Based on theoretical concerns expressed in the past, there was concern that efavirenz would lose its efficacy, and would therefore have to be increased in dose from 600 to 800 mg per day. Manosuthi found that in their patients, who weighed 50 kg on average, that didn’t seem to be necessary; the 600 mg dose produced similar drug plasma levels as those measured in the past in patients who did not receive TB treatment. Moreover, increasing the efavirenz dose did not change the plasma levels significantly.

In conclusion, when you need to treat both TB and HIV:
(1) start with TB
(2) after two months, add efavirenz, 3-TC, and d4T
(3) don’t change usual dosages

Further work will probably be necessary to reassure those who think that the situation will be different in countries where the average body weight is greater.

Effective treatment for HIV equals multiple drugs (or does it ?) Eight years ago, the era of highly active anti-retroviral therapy started when several drugs, each with only short-time efficacy, were combined. The combination blocked development of resistance, and since then, all effective treatment combinations have used at least three drugs.

Many were surprised therefore, when Joe Gathe reported on patients treated with only Kaletra (ritonavir-boosted lopinavir) last November at ICAAC. Follow-up at that time was still short (24 weeks), and there was doubt as to whether the promising results would hold up. Well, they do. The 30 patients were quite sick when they started treatment, with a high mean viral load of 262,000 c/mL, and a low mean CD4 cell count of 169.5 cells/mm3. After 48 weeks on Kaletra only, two-thirds (20/30) had HIV RNA < 400 c/mL. Most of the 10 subjects who were no longer on LPV/r monotherapy changed therapy for non-virologic reasons. On average, CD4 cell counts increased substantially (317 cell/mm3). Selection of resistant genotypes did not occur. These provocative initial, non comparative data suggest that the strategy is now being tested in two randomized clinical trials, one of which uses LPV/r for maintenance treatment following induction with combined therapy. Potential advantages include sparing of multiple ART classes, limiting toxicity and decreasing cost. The pill burden (3 pills twice daily), however, is not less than what’s achievable with some of the more popular combination regimens.


WEDNESDAY
14 July

Interrupting HIV treatment

Consider your average successfully treated HIV-infected patient. He (or she) is in his forties; his CD4 cell count, although previously low, is now above 500. His risk of AIDS-related complications is close to zero, and his life expectancy probably normal, i.e. some forty years. Forty years of continuous anti-retroviral therapy? In view of the inconvenience, expense, and side effects of HAART, many patients eye such prolonged treatment with little enthusiasm, and would like to stop.

But stopping therapy could invite trouble. The virus will surely rebound and might become resistant ,and CD4 cells will decline again. Is it better to bear with continuous therapy, or to undergo scheduled treatment interruptions (STIs)? The short answer is that nobody knows, but that many try to find out, some of whom reported their findings today.

One approach which looked promising three years ago involved treating for one week, then stopping for a week, then treating again for a week, and so on. It was hoped that the treatment interruption would be too short for viral rebound to occur; no viral rebound means no CD4 count decline and no increase in contagiousness. However Jintanat Ananworanich (TuSy187, and AIDS 2003, 17:F33–F37) showed that with a protease inhibitor-based combination in Thailand, there was an unacceptably high rate of viral breakthrough. Using efavirenz, Mark Dybul (TuSy188) now reports that failures have also occurred in Uganda. The one week on, one week off approach is likely to be abandoned.

Many patients and physicians worry about the development of resistance during STIs. It is easy to stop, but when you need the treatment again, will it still work? Ananworanich (WeOrB1283) reported on patients who had their treatment interrupted as long the CD4 count exceeded 350. After 96 weeks of such CD4-guided therapy, they were again treated continuously. Results were reassuring: After 24 weeks of continuous therapy, 17/18 had a viral load below 50 copies per ml.

CD4-guided therapy implies measuring the CD4 count; such measures are not available everywhere. Hence the proposal to interrupt therapy for fixed periods. The French trial named “Window” (Marchou, WeOr B1285) enrolled 400 patients, who were randomized to either continuous therapy, or to a schedule of 8 weeks on, 8 weeks off therapy for 96 weeks. Final results will only be available at the end of 2005, but so far, things are going well, with only a moderate decrease in CD4 cells. At the start of “Window”, 30 percent of patients had viruses with one or another drug resistance mutation, with no increase during the trial. About 80 percent of patients had a viral load below 400 copies at the end of each 8-week treatment period. Patients with mutations reacted just as well to treatment, as patients with wild-type virus.

In summary, the safety data regarding resistance and the effect of re-treatment look encouraging. However, advantages and disadvantages of STIs can only be gauged by comparison with a continuously treated control group. Results from such comparative trials will have to await 2005, 2006, or even later: the largest such study, called SMART, will enroll 6000 patients and is expected to continue to the end of the decade.

Note: These reports were prepared by the Track B Rapporteurs Team.