AR therapy is rapidly becoming the standard of care for the treatment of HIV infection. AR therapy provides the best opportunity for maximizing viral suppression, reducing toxicity and delaying the emergence of resistant strains. The newest class of AR agents, the HIV protease inhibitors, exhibits the most potent anti-HIV effects described to date. This study will compare 2 protease inhibitors, NFV and RTV for efficacy and safety in a population with advanced HIV disease, who are taking various background nucleoside therapies.

Eligible patients will be randomized either to NFV plus background AR nucleoside therapy or to RTV plus background AR nucleoside therapy. Background AR therapy may also be no background therapy, although use of protease inhibitors as monotherapy is not recommended unless there is no alternative. Patients will be allowed to cross over to the alternate protease inhibitor if they reach a primary study endpoint. Data will be collected every 4 months.

[AS PER AMENDMENT 10/2/97: Patients assigned to the NFV arm who develop a significant intolerance may switch to RTV or IDV; those assigned to the RTV arm who develop a significant intolerance are encouraged to switch to IDV (NFV allowed if IDV contraindicated). Switchover for intolerance is strongly discouraged during the first 4 weeks of follow-up. Patients initially assigned to NFV therapy who experience disease progression may switch to RTV; if RTV is not tolerated, patients may switch to IDV. Because of the cross-resistance between RTV and IDV, patients who progress on RTV should switch to NFV.] [AS PER AMENDMENT 12/15/98: Patients originally assigned to NFV who experience poor virologic control or disease progression should change to RTV or IDV or enroll in the PIP protocol (CPCRA 057). Conversely, patients originally assigned to RTV should change to NFV or enroll in the PIP protocol (such patients continue to be followed on this study). Because of cross-resistance between RTV and IDV, change from RTV to IDV is discouraged. Determination of poor virologic control or disease progression is at the discretion of the patient's clinician. Change in background antiretroviral therapy should occur at the same time that the protease inhibitor is changed for poor virologic control or progression; the choice of new background antiretroviral agents is at the discretion of the clinician.] Randomization is stratified by clinical site.] [AS PER AMENDMENT 3/11/98: SUBSTUDY CPCRA 045: At least 600 patients (>= 400 from CPCRA sites and >= 200 from CTN sites) will be enrolled in the substudy. These patients will have a plasma sample collected for HIV RNA and genotypic resistance within 30 days prior to randomization, at the 1-month visit, and at the q-4-month study visits thereafter until the end of the study. CD4+ cell counts will be done at the 1-month visit and at the q-4-month study visits until the end of the study. A subset of patients will also have immunophenotyping of CD4+ and CD8+ cell TCR V-beta clones carried out before and during treatment. Another subset of patients at selected sites will have viral cultures performed for phenotypic drug sensitivity testing.

Initially, specimens for 50 randomly chosen patients in the group originally assigned RTV will be identified for resistance testing. Of this group, specimens for those who have received RTV/IDV for more than 1 month will be analyzed for genotypic resistance to obtain an estimate of the rate of resistance development and to estimate the risk of disease progression associated with resistance to RTV/ZDV. Based on these estimates, determination will be made of the total number of patients and specimens in both treatment groups in order to address the primary objective of comparing genotypic resistance in the two groups.]

Inclusion Criteria

Concurrent Medication:

• Background AR nucleoside therapy is required, although background AR therapy may also be no background therapy. However, the use of protease inhibitors is not recommended as monotherapy unless there is no other alternative. Therefore, patients who are not on AR treatment may be enrolled at the discretion of the clinician.

Allowed:

• Saquinavir.

Patients must have:

• Documented HIV infection.
• A CD4+ cell count <= 100/mm3 within 3 months prior to the study. [AS PER AMENDMENT 3/11/98: CD4+ cell count <= 200/mm3 any time prior to entry].
• Parental consent if patient is < 18 years old.

Prior Medication:

Allowed:

• Saquinavir (SQV).

Exclusion Criteria

Co-existing Condition:

Patients with the following symptoms or conditions are excluded:

• Stage 2 or greater AIDS dementia complex.
• [AS PER AMENDMENT 10/2/97: Any acute disease or condition that would, in the physician's judgement, contraindicate starting NFV or RTV.]
• Known hypersensitivity to RTV or any of its ingredients (for patients assigned to RTV therapy).

Concurrent Medication:

Excluded:

• Concomitant use of protease inhibitors.
• Concomitant treatments that cannot be discontinued, and in the physician's judgement, should not be taken with NFV or RTV.

AS PER AMENDMENT 10/2/97:

• For patients randomized to NFV:
• Concomitant therapy with terfenadine, astemizole, cisapride, triazolam, midazolam, ergot derivatives, amiodarone, quinidine, or rifampin.

For patients randomized to IDV:

• Concomitant therapy with terfenadine, astemizole, cisapride, triazolam, midazolam, and rifampin.

Patients with any of the following prior symptoms are excluded:

AS PER AMENDMENT 10/2/97:

• History of clinically significant hypersensitivity reaction to any component of NFV tablets (for patients assigned to NFV therapy).

Prior Medication:

Excluded:

• Prior use of protease inhibitors except SQV.

[AS PER AMENDMENT 10/2/97:

• Prior use of IDV for more than 4 weeks or other protease inhibitors (except SQV) for any prior duration.]