Nucleoside Analogue Drug Combinations

ACTG 175

Results from ACTG 175, a large multicenter study in the U.S. that compared two combinations of nucleoside antiretroviral drugs with two monotherapies, indicate that the treatment of moderately immuno-compromised people with HIV (i.e., a CD4 count between 200 and 500) delayed progression to AIDS and death. Moreover, the results indicate that zidovudine (AZT) monotherapy did not provide as much benefit overall as either combination therapy or monotherapy with didanosine (ddI). The results of ACTG 175, which was sponsored by the National Institutes of Health (NIH), were announced on September 14, 1995, and presented the following week at a large international meeting in San Francisco.

The study began in late 1991, lasted for three years, and enrolled nearly 2500 patients, of which 1000 were AZT-naive, that is, had not used antiretroviral therapy before entering the trial. The balance was AZT-experienced, with almost half of these patients having taken AZT for more than two years before entering the study. Patients were randomized to one of four drug regimens in which they were given standard doses of zidovudine (AZT, made by Glaxo Wellcome) plus zalcitabine (ddC, made by Hoffmann-LaRoche); AZT plus didanosine (ddI, made by Bristol Meyers Squibb); AZT monotherapy; or ddI monotherapy. Not surprisingly, the AZT-experienced group generally did better by switching from AZT monotherapy to another of the regimens being tested. For these patients, switching to ddI or adding ddI to AZT delayed the time of progression to AIDS or death; however, switching to AZT plus ddC did not demonstrate as much benefit. In contrast, AZT-naive patients who were randomized to the AZT plus ddC regimen showed the most clinical benefit, although AZT plus ddI and ddI alone were also better than AZT monotherapy.

The analyses and interpretations of ACTG 175 -- especially those comparing the different treatment groups -- are complicated by the fact that most patients were not taking their originally assigned drug regimens at the conclusion of the three-year study. Of the 2467 enrollees, two-thirds were known to have discontinued their original regimens before the end of the study, including almost one-fifth who were "lost to follow-up," i.e., the status of each was unknown at the end of the study. Of the 67% who were known to have changed regimens, 20% (13% of the 2467 enrollees) did so, as called for by the protocol, when their CD4 count fell below 50% of its entry value or when, in a few enrollees, an AIDS defining condition occurred. An additional 3% of the total discontinued their drug regimens, as the protocol instructed, because of severe or worsening toxicity that was potentially related to the drugs being studied. The remaining patients who discontinued their originally assigned therapies did so for a variety of reasons. These reasons included mild-to-moderate side effects, CD4 loss, or perceived clinical deterioration.

The most important information from this study is a demonstration for the first time that the survival of moderately immunocompromised patients infected with HIV can be prolonged with antiretroviral treatment. The results strongly suggest that for AZT-naive patients, it is better to initiate antiretroviral drug therapy with AZT plus ddC, AZT plus ddI, or ddI alone, than with AZT monotherapy. For AZT-experienced patients, this study indicates that it is better to switch from AZT to ddI monotherapy or to add ddI to AZT, than to continue taking AZT monotherapy. However, these results should not be interpreted as recommending that any of these regimens can be taken indefinitely or that they are best for everyone.

The Delta Trial

On September 26, the preliminary results of another large multicenter trial conducted in Europe and Australia were announced. The Delta trial began in 1992 and enrolled 3300 patients with an entry CD4 count between 50 and 350. About two-thirds of these patients were AZT-naive and one-third were AZT-experienced, with three or more months of prior AZT therapy. The same regimens as in ACTG 175 were compared, except for ddI monotherapy, which was not studied. The trial was stopped prematurely, after an average follow-up time of 26 months, because of a highly significant difference in the proportion of AZT-naive patients who died on AZT monotherapy (17%) compared to either of the combinations (on AZT plus ddI, 10%; on AZT plus ddC, 12%). Progression to AIDS was also slower in the patients receiving one of the combination therapies.

In contrast to ACTG 175, however, patients in the Delta trial who were AZT-experienced (63% of them had taken AZT for more than one year prior to enrollment) experienced a rate of progression to AIDS and death that was similar across all three drug regimens. (In ACTG 175, AZT-experienced patients on either ddI alone or on AZT plus ddI did better in this respect than those who continued on AZT alone.) It may be that adding ddI to AZT in a patient population that has more advanced disease (as is the case in the Delta trial) is of less benefit than making the same change at an earlier stage of the disease. More detailed analyses, further patient follow-up, and a closer comparison of the two studies will have to be performed to understand these differences.

The results of a second NIH-sponsored study, conducted by the Terry Beirn Community Programs for Clinical Research on AIDS (CPCRA 007), will be available early next year. This trial compares the same three regimens as the Delta trial, but in an even more advanced patient population (CD4 under 200 at entry). Since most of these patients (80%) were AZT-experienced at enrollment, this study should provide important information to help reconcile the different outcomes in this group of patients between ACTG 175 and the Delta trial.

In the meantime, it can be concluded that the Delta trial, like ACTG 175, clearly shows the superiority of both AZT plus ddI and AZT plus ddC over AZT alone in patients who as yet have not taken antiretroviral drugs. In addition, didanosine (ddI) alone, as ACTG 175 suggests, may be as effective as either of the two above-mentioned combinations. However, it would be inappropriate to conclude from either study that any combination therapy is better than any monotherapy. One can only conclude that these two particular combinations, AZT plus ddI and AZT plus ddC, are better than AZT monotherapy, particularly in AZT-naive patients.

AZT plus 3TC Studies

A very important unanswered question is whether any of the promising new drugs, which were not available when ACTG 175 and the Delta trial were designed, will be of even greater benefit to patients than those drugs and drug combinations that were tested. A case in point is the combination of AZT and lamivudine (3TC), another nucleoside analogue made by Glaxo Wellcome. The results of four studies of the combination of these two drugs, AZT and 3TC, compared to AZT alone, were first announced late last year. Two of these studies were conducted in Europe and two in the U.S., in either AZT-naive or AZT-experienced patients. In all four studies, the combination of AZT plus 3TC was superior to AZT alone in decreasing viral load, i.e., the amount of HIV in the plasma, and in improving CD4 counts. The response in terms of these laboratory markers to the combination therapy was greater and longer lasting in AZT-naive patients than in AZT- experienced patients.

These studies were relatively short (six months of comparing AZT alone with the combination, after which all patients were offered the combination) and small (fewer than 100 patients in each treatment group). Unlike ACTG 175, they were not designed to determine whether changes in laboratory markers would translate into clinical benefit, i.e., rate of disease progression and survival. Although desirable effects on laboratory markers were more pronounced and sustained than had been seen with any other nucleoside analogue alone or in combination with AZT, an overall assessment of risks and benefits of new drugs and combinations intended for the long-term treatment of HIV disease requires large long-term studies in which clinical outcome, i.e., the effect of the treatment on progression of disease, is carefully monitored and evaluated.

Lamivudine in combination with AZT was recently approved by the FDA for initial therapy and for AZT- experienced patients. Over 35,000 patients participated in Glaxo Wellcome's expanded access program, which ran for two years prior to FDA approval.

Non-Nucleoside Reverse Transcriptase Inhibitors (NNRTIs)

All the nucleoside analogues discussed above limit viral multiplication by inhibiting a crucial viral enzyme, reverse transcriptase, and by competing with the natural nucleosides for incorporation into the DNA of the virus's genetic material. There is, in addition, another class of drugs called non-nucleoside reverse-transcriptase inhibitors, which are chemically very diverse, but very specific inhibitors of this enzyme. Unfortunately, HIV quickly develops resistance to these drugs, which limits the duration of their antiretroviral activity, especially when they are given alone.

Three non-nucleoside reverse-transcriptase inhibitors are currently in clinical trials (Behringer Ingelheim's nevirapine, Janssen's loviride, and Upjohn's delavirdine); within the next year or so, they may be FDA-approved and available by prescription in this country. amfAR-supported Community-Based Clinical Trials Network units are actively participating in studies of NNRTIs.

Protease Inhibitors

A very promising, potent new class of antiretro-viral drugs, which should help prolong life and reduce morbidity in HIV-infected people, is that of protease (or proteinase) inhibitors, which inhibit the activity of protease, another enzyme of HIV that is critical for its multiplication. Viral resistance to these drugs -- four of which are currently in controlled clinical trials -- develops over time. Nevertheless, the existence of this third class of drugs opens many additional opportunities for new drug combinations and hence, new treatment options.

The first of the protease inhibitors to be tested in humans, saquinavir (Invirase, made by Hoffmann-LaRoche), is the weakest, due largely to its poor absorption by the gastrointestinal tract after oral administration. However, when given in combination with certain nucleoside analogues, e.g., AZT and ddC, antiretroviral activity is improved, as is immune status. Indinavir (Crixovan, made by Merck) was the second protease inhibitor to enter clinical trials. It is a potent inhibitor of HIV, especially at higher doses, and can be combined with nucleoside analogues, such as AZT. The third protease inhibitor, ritonavir (ABT-538, made by Abbott), is in advanced clinical development. It is also very potent, although not as well tolerated as the others.

A randomized trial in patients with advanced clinical disease that compares ritonavir to a placebo control (both added to "standard of care" antiretroviral drugs) is currently being conducted. amfAR's Community-Based Clinical Trials Network units are also participating in the trials of protease inhibitors. Saquinavir was recently approved by the FDA for use in combination with nucleoside drugs. Both indinavir and ritonavir are available in limited quantities to patients with advanced disease and no other treatment choices. Eligible patients are being selected by a lottery system to participate in the expanded access programs for these latter two proteases. Both pharmaceutical companies are expected to file new drug applications with the FDA in 1996, seeking marketing approval for their protease inhibitors. A fourth protease inhibitor, AG1343 (Viracept, made by Agouron), is not quite as far along in its development; but to date, it appears to be very active, as well as safe.

In addition to the possibility of combining protease inhibitors with nucleoside analogues, certain protease inhibitors may be combined with each other to improve the extent and duration of the antiretroviral activity of both drugs. For example, the combination of saquinavir and ritonavir, which so far has been tested only in animals, improves the blood levels of saquinavir manyfold, due to an intriguing pharmacologic effect of ritonavir. Ritonavir inhibits a certain class of important liver enzymes, which metabolize, or break down, saquinavir in the body. Thus, the co-administration of ritonavir with saquinavir results in higher levels of saquinavir and may compensate, at least in part, for saquinavir/ritonavir poor absorption. Higher levels of saquinavir have more powerful antiviral activity, which may be augmented further by the concomitant antiretroviral activity of ritonavir. This combination and others soon will be tested in humans, with potentially exciting results.

Resistance to Antiretroviral Drugs

The development of resistance to drugs administered over long periods of time is a problem with all antiretrovirals, although more so with some than with others. The challenge of minimizing the effects of viral resistance -- which renders drugs less effective as antiretroviral treatments -- may be addressed by several strategies that appear promising. One strategy is to prolong the time before the virus develops resistance by administering two or more drugs in combination and also administering higher doses of certain of the drugs (some are too toxic to be taken in higher doses). Another strategy, designed for the longer term, is still largely theoretical because no evaluative studies have been conducted as yet. This strategy is based on the use of specific laboratory markers to determine the best time to discontinue one drug regimen whose efficacy is waning and switch to a more active regimen before clinical deterioration occurs.

A concern raised not only by protease inhibitors, but also by certain nucleoside analogues, is cross-resistance. Cross-resistance means that when HIV has developed resistance to one drug, it is found to be resistant to other drugs of the same class, despite the fact that HIV has not been exposed to them. Early laboratory studies of protease inhibitors suggested that cross-resistance might be a serious problem with this particular class of drugs. However, subsequent studies indicate that cross-resistance is unlikely to be a problem with all of them. Thus, the protease inhibitors, several more of which are in earlier stages of development than those mentioned above, hold great promise for improving on the antiretroviral effects and clinical benefits of the nucleoside analogues, bringing closer the day when HIV disease becomes truly manageable.

Management Strategy Trials

It must be remembered that not all patients respond well to treatment with antiretroviral drugs, particularly the nucleoside analogues. For example, some patients who take AZT plus 3TC have no detectable changes in viral load or in CD4 counts; or, if there are changes, they are very short-lived. We have much more to learn about how best to use these drugs and -- because individuals respond differently to the same drugs -- how best to customize treatment for each patient. Given the number of new antiretroviral drugs currently under investigation and the larger number of potential drug combinations that look promising and should be studied in humans, it becomes increasingly important to think about new ways to design such studies and conduct them in the community setting -- the real world of patient care and treatment. Community-based trials can enroll the large numbers and the diversity of patients that need to be studied.

What now must be done is a large collaborative trial in which the most promising drugs and drug combinations are compared in a variety of patients who change regimens when a predetermined "switch point" is reached (such as a rise in viral load or a fall in CD4 count) before disease progression occurs. Through its Department of Clinical Research and Information, amfAR is exploring the feasibility of conducting such a trial in the community setting.