Dr. Joe Eron from the University of North Carolina, Chapel Hill, began the discussion of antiretroviral therapy by reviewing our current understanding of the virology of HIV infection. Viral replication is a dynamic process, with billions of virions being produced per day. The destruction of CD4 cells is directly linked with HIV replication and viral production. These findings are at the core of the rationale for early, aggressive therapy:

1. Early intervention limits genetic diversity: Because HIV is so highly mutagenic, early treatment limits the emergence of multiple HIV variants or mutants, that may compromise response to therapy.
2. Early intervention preserves immune function: We know from studies in cancer patients who have received chemotherapy that the ability to regenerate CD4 cells diminishes with age. Those CD4 cells that do regenerate with effective antiretroviral therapy tend to be the cells that still exist, and are preferentially memory cells. When patients with advanced disease who have lost their naive CD4 cells are treated with antiretroviral therapy, the rise in absolute CD4 cell number may not reflect true restoration of immune function.
3. Aggressive combination therapy delays the development of resistance: This requires not only combination therapy, but therapy with synergistic drugs that do not select for cross-resistant mutations. Drug levels must be high enough to maintain constant inhibition of replication, which requires an understanding of drug interactions between antiretroviral agents. Successful therapy with highly suppressive regimens also involves strict adherence to treatment regimens.
4. Early aggressive therapy provides the potential for eradication: Although the possibility of HIV eradication is still highly speculative, it is generally believed that if it is achievable at all, it will be in patients who are treated with highly suppressive therapy early in the course of their HIV disease.
5. Early aggressive therapy also treats tissue reservoirs: There is now good evidence from a number of studies that reducing plasma viral load also reduces viral load in lymphoid tissue and in semen. There are also some data suggesting that viral load in the cerebrospinal fluid is correlated with plasma viral lode.

The correlation between the effect of a treatment on viral load and clinical benefit has now been firmly established. In ACTG 175, the landmark study demonstrating the superiority of combination nucleoside therapy over AZT monotherapy, after 8 weeks of therapy, for every one log (10-fold) reduction in viral load there was a 65% reduction in disease progression or death and a 70% reduction in death. In an Upjohn-sponsored trial of delavirdine, a one log reduction in viral load was associated with an 80% reduction in disease progression.

Dr. Eron discussed the multitude of potential regimens for initial therapy, which include:

• 2 nucleoside reverse transcriptase inhibitors (NRTIs) + 1 protease inhibitor (PI)
• 2 NRTIs + 1 non-nucleoside reverse transcriptase inhibitor (NNRTI)
• 2 NRTIs
• 4 drugs: (2 NRTIs + 2 PIs or 2 NRTIs + 1 PI + 1 NNRTI, etc.)

From his initial comments, it is clear that more aggressive regimens (regimens containing a PI, for example) are preferred under optimal conditions. However, Dr. Eron concluded with some words of caution:

• Compliance will dictate who responds and who doesn't. Patients who are likely to be non-compliant may not be better off with a more aggressive regimen.
• HIV disease is usually a long-term disease, especially for those who have low viral loads and high CD4 counts. Starting out with early, aggressive therapy involves a serious commitment to long-term polypharmacy, with all its "complexity and chaos," for patients who might have done fine on their own for a very long time.
• We really don't know how to treat "triple drug" failures. It may turn out that when it comes to using protease inhibitors, "the first shot is the best shot." While new drugs, such as abacavir (1592), are in the pipeline, it is not at all clear how effective they will be in patients with resistance to multiple agents.
• The widespread use of early, aggressive therapy may lead to the transmission of resistant HIV, especially in cases where the therapy is not completely effective.

Dr. Robert Murphy, from Northwestern University Medical School in Chicago, concluded the session on antiretroviral therapy, by discussing data from a number of recent antiretroviral therapy trials.

Protease Inhibitor Trials:

Merck 035: At 68 weeks, over approximately 85% of patients taking AZT, 3TC, and indinavir had viral loads less than 500 copies/mL, and approximately 75% were undetectable by the ultrasensitive assay, with viral loads less than 50 copies/mL.

ACTG 320: This landmark study involved over 1000 patients with CD4 counts less than 200 cells/mm3 (median 86) who were randomized to receive either AZT/3TC or AZT/3TC/indinavir. The study had to be stopped early because there was a significant benefit to triple-drug therapy. Patients taking AZT/3TC/indinavir were less likely to progress (defined as the development of a new AIDS-related event or death) or to die, compared with those taking AZT/3TC. Not surprisingly, this difference was most pronounced in patients who enrolled with CD4 counts less than 50cells/mm3. Fortunately , this may well be the last large antiretroviral trial relying on clinical endpoints (disease progression or death); as changes in viral load and CD4 count are now considered to be acceptable endpoints for clinical trials.

Merck 039: This was a trial comparing AZT/3TC with AZT/3TC/indinavir in patients with advanced disease (CD4 less than 50 cells/mm3) and prior AZT experience, but no experience with either 3TC or protease inhibitors. Those on triple therapy had 70-80 cell increases in CD4 count and over 2 log (100-fold) viral load reduction. Over 65% achieved undetectable viral loads.

NNRTI Trials:

Dr. Murphy first reviewed drug interactions with the NNRTIs. Nevirapine induces the CYP3A system, leading to decreased levels of protease inhibitors, rifampin, and oral contraceptives. Delavirdine inhibits CYP3A, leading to increased levels of those drugs. DMP 266 appears to be an inducer of CYP3A. Drug interactions with loviride are well defined yet. Nevirapine significantly lowers levels of saquinavir and indinavir. It has been suggested that the dose of indinavir be increased to 1000 mg when it is coadministered with nevirapine. Because saquinavir levels are already low, it should probably not be given in combination with nevirapine, unless it is being combined with a drug that significantly increases saquinavir levels, such as ritonavir. Protease inhibitors do not alter nevirapine levels.

Nevirapine/indinavir/3TC: A small, open-label trial using this combination involved patients heavily pretreated with nucleosides analogs, including 3TC. None had taken nevirapine or protease inhibitors They had advanced disease (CD4 less than 50 cells/mm3) and viral loads over 500 copies/mL. Patients received indinavir, nevirapine, and 3TC, all at standard doses. At 16 weeks, the median rise in CD4 count was 75 cells/mm3, the median decrease in viral load was 3.12 logs, and 56% achieved undetectable viral loads ( Delavirdine/saquinavir: A small trial using this combination demonstrated a 4 to 5 fold increase in saquinavir levels. However, marked increases in liver function tests were seen, and therefore a larger trial was not approved. This combination should be used with caution. Delavirdine also increases levels of other protease inhibitors, especially indinavir and nelfinavir.

Dual Protease Inhibitor Therapy:

Ritonavir/Saquinavir: Ritonavir is a strong inhibitor of the cytochrome p450 system, leading to marked increases in the levels and half-lives of other protease inhibitors. Dr. Murphy gave an update on the ritonavir/saquinavir trail. Although the doses of the two agents varied initially, most participants are now taking 400 mg of each drug twice daily. The results remain impressive at 48 weeks, with viral loads remaining undetectable in most patients. Few have added nucleoside analogs, even though that option was open to them after 24 weeks.

Other PI/PI combinations being studied include nelfinavir/saquinavir (nelfinavir increases saquinavir levels, but to a lesser degree than ritonavir), and ritonavir in combination with either indinavir or nelfinavir (which may allow for twice daily dosing of both PIs).

Nucleoside Analog Therapy:

d4T/ddI/3TC: In a small study by Leibowitch and colleagues, 20 patients, both experienced and naive to nucleoside analogs, took open-label d4T/ddI/3TC. At 24 weeks, viral load reduction was 2.4 to 2.6 logs, and CD4 increases of 200 cells/mm3 were noted. 80% achieved undetectable viral loads. Those patients who had never taken antiretroviral therapy had better responses.

Once daily ddI: Because ddI has a long intracellular half-life, once-daily dosing may be as effective as twice-daily dosing and may be associated with less neurotoxicity. Once daily dosing is common in Europe. While the French typically give 300 mg daily, studies in the U.S. will probably look at 400 mg daily. ddI/d4T/hydroxyurea: Hirschel and colleagues compared ddI/d4T with ddI/d4T/hydroxyurea in 141 patients, 80% of whom were naive to antiretroviral therapy. The results were as follows:

Of note, hydroxyurea lowers absolute lymphocyte count, which can lead to a lower absolute CD4 count than would be expected based on the CD4 percent. Thus, in patients on hydroxyurea, the CD4 percent may be a more accurate reflection of immune function than the absolute count.

Dr. Murphy noted that despite the fact that 11 drugs are now approved for the treatment of HIV disease, treatment options remain limited. Examples of obstacles that prevent us from using the full complement of drug combinations include:

Dr. Murphy gave examples of unsuccessful treatment strategies, including:

• monotherapy
• alternating monotherapy
• adding single drugs to failing regimens
• use of inadequate combinations
• use of combinations that don't reduce viral loads effectively
• "patient unfriendly" regimens

He then gave examples of a number of novel regimens, including:

1. Once daily therapy (ddI, nevirapine)
2. Double protease inhibitor combinations
3. Triple nucleoside analog combinations
4. PI/NNRTI combinations
5. Hydroxyurea-containing regimens

Treatment strategies that are becoming increasingly accepted include:

1. Initiate therapy with potent triple-drug regimens

2. Change to an all new potent triple-drug regimen in the event of treatment failure

3. Utilize novel approaches when options are limited.

He discussed the "Chinese menu approach" to choosing antiretroviral therapy. An appropriate regimen might include one drug from column 1, one drug from column two, and one or two drugs from column 3:

Future therapies, drugs that may become available in 1997 or 1998, include:

Nucleoside Analogs Reverse Transcriptase Inhibitors

abacavir (1592U89)
MKC 442

Non-nucleoside Analog Reverse Transcriptase Inhibitors

DMP266
loviride
HBY097

Protease Inhibitors

141W94/VX-478
ABT 378

Integrase Inhibitors

AR 177

Nucleotide Analogs

adefovir