The Body Covers: The 39th Annual Meeting of the Interscience Conference on Antimicrobial Agents and Chemotherapy
Managing HIV Disease for Life --
New Issues Challenges, and Opportunities
September 26, 1999
There are several key issues that continue to be the focus of debate in the desire to optimize the care for people with HIV. Controversies include the discussion of the optimal time to start treatment and the alternatives for creating effective combinations to use when starting. These were reviewed in this first talk.Abstract: New Strategies in the Treatment of HIV Disease
Over the past decade there have been studies done attempting to address the optimal time to start. In particular, there are studies that have initiated antivirals as soon as possible after seroconversion. The major justification for this is the observation that there are cytotoxic T cells that are particularly active in controlling HIV replication as seen in those who are long term slow progressors. Unfortunately, these cells are among the first to be destroyed soon after seroconversion. Initiating antivirals during this period, usually considered to be within the first six months after initial exposure, has been shown to preserve these cells. It remains unclear if preserving these cells through treatment will be sufficient to engender the same degree of protection from HIV as seen in the long term nonprogressors, but that is currently under evaluation. Nonetheless, most patients with HIV do not present for care during this phase of infection. After that initial period, there is less urgency in deciding when to initiate antivirals. However, given the observation that the nadir of viral load once on treatment correlates with durability from the regimen, there are data that still support starting earlier.
Yerly et al presented data that address this issue. (Yerly et al, Antiviral Therapy, 1999 resistance meeting.) Their study compared the results of initiating a standard triple antiviral combination in three cohorts: one who were seroconvertors, one who were chronically infected but had a CD4 count >500 cells, and one whose CD4 counts were below 500. They presented the results of achieving viral load suppression on an experimental assay with a sensitivity of 3 copies/ml. The results were that 75% of seroconvertors achieved below detection on this assay, while only 32% of the group with a CD4 count over 500 did so. However, only 8% who were treated with a CD4 count below 500 achieved this same threshold. These data suggest that there may be some benefit of additional durability to a regimen if initiating treatment in those with higher rather than lower CD4 counts.
As always, the plausible benefits of initiating as soon as reasonable have always been weighed against the concerns for cumulative toxicity of antivirals, both the immediate subjective side effects associated with treatment, as well as the increasing appreciation for metabolic disturbances in longer term use. In addition, there are data that reassure practitioners of the largely reversible decline of immune function, at least in several parameters that have been studied to date. Supportive data include the observation of not only the increases in CD4 cell counts when treatment is initiated even in advanced disease, but the ability to safely discontinue prophylaxis after rebounds in CD4 counts without evidence of illness thus far. However, there remain theoretical concerns of aspects to immune function that are not fully restored with treatment. This leads to justification for initiating antivirals before any deterioration to the immune system.
There have been proposals to randomize patients in large numbers to strategies of initiating "early" versus "delayed" treatment as a way to answer this question. While reasonable, it is important to note that treatment options continue to advance, including an increasing arsenal of medications with potential for improved profiles. In such a study, the "delayed" arm would have an advantage of potentially receiving a "better" regimen as they develop over time. However, it may still be that this improved regimen would perform even better had it been given earlier in infection. Thus, at least for these next several years, results of a clinical trial to answer this important question will not be available, and practitioners must continue to weigh the pros and cons in order to make this critically important decision.
Most of the regimens selected for initiation of highly active antiretroviral therapy (HAART) are based on the selection of two nucleoside reverse transcriptase inhibitors (NRTI) with some third agent, which has generally been either another nucleoside, a nonnucleoside (NNRTI), or a protease inhibitor (or dual PI). Each of the approaches has potential benefits as well as risks. Because of the timing of how medications were developed, studies that initially described the benefits HAART therapy have been based on using a PI with 2 NRTI's. There are several known advantages to this approach, including multiple studies demonstrating a reduction in clinical illness and death when this approach is used. Most of the PI's have data supporting a BID schedule with the exception of indinavir, and even it now has data that show it can be successful in a BID schedule when paired with a ritonavir. However, there are potential problems with some of the PI regimens. Acknowledging the preference on the part of at least some patients for taking the fewest possible number of pills, PI based regimens tend to have higher pill burdens when compared with the numbers required for regimens based on the other approaches. There are clear links to metabolic concerns when using most PI's, with increases in serum glucose as well as lipids reported, while the association to lipodystrophy has become more complex. Finally, for most of these regimens, there has been a somewhat lower response rate observed in the subset that has a higher baseline viral load. Recent advances in the use of PI's have included the use of ritonavir to improve the pharmacokinetic profile of the other PI's. The first study to validate this approach used the combination of ritonavir with saquinavir. Since then there are studies that support the use of ritonavir with indinavir as well. PK work presented at ICAAC 1999 suggests the potential for using ritonavir with a second PI that might permit once daily dosing. Pharmacokinetic data were presented suggesting the feasibility of using ritonavir with either saquinavir, indinavir, or amprenavir, although no clinical trial data are available for results using a once daily dosing schedule.
More recently, several studies have been done which support an approach using two NRTI's with an NNRTI. The advantages of these approaches include the data from two randomized studies (Atlantic and DuPont 006) supporting the comparability of these combinations with the results when using a PI. Results from one study suggest further that in a combination that used efavirenz with two nucleosides, there was even improved durability when compared to regimens based on indinavir. One small study (DuPont 043, Cohen, ECCMID 1999) also reported high levels of success in a cohort of 43 patients, with 97% of those on treatment at week 24 having a viral load below 50 copies. Further, this study verified the reports from study 006 of similar rates of suppression at viral loads over 5 log when compared to those with a viral load below 5 log. At this meeting are results from lymph node biopsies from a few patients on study 043, showing similar rates of viral clearance from nodal tissue compared with a PI treated cohort, again supporting the similarity of success using this approach.
The third option for creating an effective triple regimen has been that using triple nucleosides. There are two studies showing reasonable success rates when compared to other approaches. Study 3005, updated at 1999 ICAAC, reported comparable viral load suppression at week 48 for patients on a regimen using zidovudine, lamivudine and abacavir when compared to a combination using indinavir with AZT/3TC. However, those who started with a viral load over 5 log had a lower percent below 50 copies at week 48 than did those on the indinavir arm. The Atlantic study reported comparable results at week 48 in regimens that were based on the use of stavudine and once daily didanosine, added to either indinavir, nevirapine, or lamivudine. As with the study using abacavir, the subset of patients with a baseline viral load of about 50,000 or greater showed somewhat lower success rates in establishing viral suppression to below 50 copies in the triple nucleoside arm compared with the results in the other two arms. These studies combined inform us that a combination using triple nucleosides are an option to consider in those with a baseline viral load below 50-100,000 copies, but raise concerns as to their relative potency when compared with the other options.
The selection of which nucleoside pair to use has been another ongoing debate. Studies have generally shown comparable success rates regardless of which pair has been selected when used in a triple combination. However, an additional consideration is the impact of the medication choice on the success of subsequent options if the initial outcome is not lifelong viral suppression. While there have been conflicting data about in vitro predictions of sequences, an analysis of one recent clinical study sheds some light on this issue. Study ACTG 364 compared the initiation of nucleosides with either efavirenz, nelfinavir, or both in a nucleoside experienced cohort. A subset of these patients was further studied to explore if the nucleoside sequences used had any impact on the success of this switch to HAART. Katzenstein (Antiviral Therapy 1999, 4(suppl 1); 47) presented an analysis of patients who went from AZT plus 3TC to d4T plus ddI (n=40), compared to those on AZT plus either ddI or ddC who were switched to AZT plus 3TC (n=43). As patients had been on their initial nucleoside treatment for many months prior to this switch, there were multiple nucleoside mutations seen at the time of switching to this new HAART regimen. There were statistically more RT mutations observed in the group who were on zidovudine plus lamivudine versus those who were on ZDV with either didanosine or zalcitabine. This difference might help explain the difference seen in the likely of a failure of virologic suppression on the next regimen, where failure was defined as an RNA >2000 copies/mL. Those going from ZDV/3TC to d4T/ddI had a failure rate of 45%, whereas those who went from ZDV/ddI (or ddC) to ZDV/3TC had a lower failure rate of only 18%. There are important limitations to the applicability of these results to current approaches, particularly given current approaches that would switch soon after the time of viral rebound and limit nucleoside mutations. Nonetheless, these data lend some support to a rationale for regimens that use ddI (or ddC) instead of 3TC in initial combinations, as the next regimen using the alternative agent may be more successful when going in this direction.
There has been much work going on to increase the success of both initial combinations as well as the success if initial attempts are not completely successful. These approaches have included options of increasing the number of antivirals initially used to increase potency, as well as simplifying the regimen after establishing viral suppression. However, each these approaches have clear limitations. Increasing the number of initial medications has the burden of additional potential toxicity associated with the regimen, and it remains unclear that such a strategy is superior to the best of current triple therapy options. Similarly, results of studies that have attempted to simplify the number of agents after initial suppression have also been unsuccessful, with viral rebound seen after the regimens were altered to combinations using less agents. Strategies of substitutions have been more successful in maintaining viral suppression, and there are reports that at least some of the adverse events seen on PI based combinations can be altered when either an NNRTI or abacavir are used instead. However, the continued challenge is to define the optimal initial triple, rather than rely on approaches that compensate for problems with the initial choice. The strategy of "intensification" as a way to increase viral suppression if the initial therapy is not completely suppressive has also received some attention. While this approach does nothing to address problems ascribed to inadequate adherence, it is an alternative way to increase the potency of regimens that did not achieve maximal suppression initially.
Given that eradication of HIV appears unlikely based on current data, clinicians remain challenged by the urgency to maximize the use of current antiviral options in order to achieve long term success. There are multiple factors to consider when deciding when to initiate treatment in order to achieve maximal viral suppression. Similarly, there remain multiple factors to balance in deciding what to use for initial therapy in order to achieve both a high likelihood of virologic success on a regimen that is well tolerated, while leaving open potential future options.
Authored by: Calvin Cohen, M.D., Research Director, Community Research Initiative of New England
This article was provided by The Body PRO. Copyright © Body Health Resources Corporation. All rights reserved.