AIDS TREATMENT NEWS Issue #211, November 18,1994
John S. James

It turned out that some of the early data supporting the convergent combination approach was in part the result of laboratory error; but even so, the idea remained worth testing. (AIDS TREATMENT NEWS covered these developments in issue #170, March 5, 1993, and issue #181, August 20, 1993.) A clinical trial was quickly organized, and it began in May 1993; it was fully enrolled by July. The trial lasted 48 weeks, then data analysis began, and a preliminary report was released this week. Much more analysis is still to be done, however, and some of the more interesting results may come later.

This study enrolled a total of 398 subjects. They had T- helper counts less than or equal to 350; the median count at entry was 138. All had at least six months of prior treatment with AZT, ddI, or ddC; the median prior treatment was 25 months. These volunteers were randomly assigned to either receive AZT plus ddI, or AZT plus ddI plus nevirapine, an experimental HIV treatment. All three drugs target the reverse transcriptase enzyme of HIV; when used alone, nevirapine has a very strong anti-HIV effect at first, but then the effect diminishes as the virus develops resistance to the drug.

At 48 weeks, the main preliminary results were:

* T-helper counts were 25 percent higher for patients assigned to the triple combination, compared to those assigned to AZT plus ddI.

* The infectivity of PBMCs (peripheral blood mononuclear cells) was lowered by 50 percent more in the triple combination arm than in the AZT plus ddI control arm. [Note: viral measures were only done on a subset of 198 of the 398 subjects.]

* HIV RNA showed a significant difference in favor of the triple combination at first, but according to the November 17 executive summary, the difference was not statistically significant at 48 weeks. [However, after the executive summary was finalized, it was learned that the reason the PBMC virology but not HIV RNA virology remained statistically significant at 48 weeks may have been because of the different ways these numbers were collected and analyzed.]

* There was no indication of a survival or disease- progression benefit. Seventeen percent of those on triple combination died or experienced HIV disease progression (as defined by the study), vs. 14 percent of those on AZT plus ddI. This difference, which was not statistically significant, was in the wrong direction. While the study was not designed to detect clinical benefit in death or disease progression, and did not have enough patients to do so reliably, still it seems unlikely, in view of these results, that there is any substantial clinical benefit of the triple combination.

* There were more side effects from the triple combination. The biggest difference between the groups was in skin rash, with severe rashes affecting 8 percent of those on triple combination, vs. two percent of those on AZT plus ddI. The overall rate of severe adverse reactions was high: for signs and symptoms, 30 percent on triple combination vs. 24 percent on the two-drug combination; for hematology and chemistry abnormalities, 30 percent on triple combination vs. 26 percent in the AZT plus ddI control arm. These overall rates were not significantly different between the arms.

This study collected important data, and there is much more to be learned from it. For example, it is possible that certain subsets of patients may have responded differently to the treatments -- those with less advanced disease, for example, or those who began the study with or without AZT- resistant or ddI-resistant viruses.

Comment

This study is likely to be cited to argue that viral load and other blood tests cannot be trusted to tell what drugs are working. We think it is too early to draw that conclusion. The viral load differences appear to be marginal; with plasma HIV RNA, whether the difference is even statistically significant at 48 weeks seems to depend on how you look at the data -- and usually it is easy to get statistical significance using this measure, even with far fewer patients, provided that there is any substantial effect to measure. Also, it is likely that many patients had AZT- resistant viruses before they started, so they were never effectively on a triple combination; and nevirapine resistance can develop quite rapidly. If the side effects of this drug balanced a relatively minor virological benefit, then the difference between clinical progression and viral- load changes could easily be explained -- with no need to postulate that viral load does not show drug benefit. Further study of the existing data may help to explain what was actually going on in this trial.

The main lesson we draw from this example is that we need more flexible research strategies, with many small trials which can be designed and run rapidly, to quickly check out promising leads which suggest the possibility of big gains in viral-load reductions, in measures of immune health, or in other indicators of activity. Then the larger, confirmatory trials could be better focused, because they would be based on data from human experience, not just on laboratory studies.

941118
ATN21103

AEGiS presents published material, reprinted with permission and neither endorses nor opposes any material. All information contained on this website, including information relating to health conditions, products, and treatments, is for informational purposes only. It is often presented in summary or aggregate form. It is not meant to be a substitute for the advice provided by your own physician or other medical professionals. Always discuss treatment options with a doctor who specializes in treating HIV.

Copyright ©1980, 1994. AEGiS. All materials appearing on AEGiS are protected by copyright as a collective work or compilation under U.S. copyright and other laws and are the property of AEGiS, or the party credited as the provider of the content. .