The purpose of this study is to determine the safety and effectiveness of combining several anti-HIV drugs in order to decrease plasma viral load (level of HIV in the blood) in HIV-positive patients who have failed nelfinavir (NFV) treatment.

In order to determine the ability of a drug regimen to decrease viral load after drug treatment has failed, it is best to test a variety different of drug "cocktails" (drug regimens). The drug cocktails in this study include 2 new nucleoside reverse transcriptase inhibitors (NRTIs), efavirenz (an NNRTI, non-nucleoside reverse transcriptase inhibitor), and either 1 or 2 protease inhibitors. It is important to include multiple drugs from different groups in a drug cocktail since combinations containing fewer drugs are likely to fail.

To maximize the likelihood of a favorable response to salvage therapy, 4 or 5 drug regimens should be studied. Regimens containing fewer drugs, particularly those lacking a non-nucleoside reverse transcriptase inhibitor (NNRTI) such as efavirenz, are likely to result in an unacceptable rate of virological failure. Therefore, this study examines drug combinations which include two new nucleoside reverse transcriptase inhibitors (NRTIs), the NNRTI efavirenz, and either one or two protease inhibitors which are known not to produce cross-resistance to nelfinavir.

Patients are randomly selected to receive 1 of the following 4 treatment regimens:

Arm A: Ritonavir, saquinavir, efavirenz, and 2 new NRTIs. Arm B: Indinavir, efavirenz and 2 new NRTIs. Arm C: Amprenavir, efavirenz, and 2 new NRTIs. [AS PER AMENDMENT 3/22/00: Patients have the option to increase the APV dose or to add low-dose ritonavir. APV will continue to be provided by the study; ritonavir will not be provided by the study.] Arm D: Indinavir, amprenavir, efavirenz, and 2 new NRTIs. [AS PER AMENDMENT 6/28/99: All treatment regimens must include at least 1 new NRTI.] [AS PER AMENDMENT 3/22/00: ACTG 400 will continue to provide originally randomized study medications to all patients until approximately May 10, 2000, regardless of virologic response. Patients may also add antiretrovirals of their choice to this regimen (not provided by the study).] Clinical assessments are taken at Weeks 2, 4, 8, 12, 16, and every 8 weeks thereafter for the duration of the study. In addition, 2 substudies are being conducted: a drug-interaction substudy and a drug-exposure substudy. [AS PER AMENDMENT 3/22/00: Both substudies are closed to accrual and their pharmacokinetics assessments are discontinued.]

Inclusion Criteria

Patients may be eligible for this study if they:

• Are over 13 years old (need consent of parent or guardian if under 18).
• Are HIV-positive.
• Currently have virologic failure (more than 1,000 copies of HIV RNA per ml).
• Agree to abstinence or use of effective birth control during the study.
• Have been taking NFV for the past 12 weeks.

Exclusion Criteria

Patients will not be eligible for this study if they:

• Have a fever for 7 days or diarrhea for 30 days before study entry.
• Have a history of peripheral neuropathy within 60 days of study entry.
• Have hepatitis.
• Have any malignancy (cancer) other than minimal Kaposi's sarcoma.
• Are pregnant or breast-feeding.
• Are receiving radiation, chemotherapy, or any therapy for any illness within 14 days of study entry.
• Have taken amprenavir, saquinavir, indinavir or ritonavir for more than 7 days.
• Have received an HIV vaccine 30 days before study entry.
• Are receiving certain other medications.