To compare two different routes of intermittently administered rhIL-2 with a highly active antiretroviral regimen (HAART) to HAART alone. The comparison is based on the following: proportion of patients achieving at least 50-percent increase in CD4 counts above prerandomization baseline values after 1 year of rhIL-2 and the rate of change in CD4 counts. To compare the safety and tolerance of these regimens and their effect on quality of life. To assess the effects of rhIL-2 when combined with HAART on changes in immune cell phenotypes and function and on HIV viral load and the rate of antiviral drug resistance development.

The poor responsiveness of late stage HIV-infected patients to rhIL-2 is thought to occur because of low T cell regenerative capacity and high viral burden. If means were available to effectively suppress virus replication, the indigenous immune restorative responses of the host may be further stimulated and enhanced by rhIL-2. The use of protease inhibitors with nucleoside-analogue combination regimens appears to be most effective in controlling virus replication. High-dose intermittent rhIL-2 administered either intravenously or subcutaneously has been shown to be effective in inducing CD4 responses in a number of studies.

The poor responsiveness of late stage HIV-infected patients to rhIL-2 is thought to occur because of low T cell regenerative capacity and high viral burden. If means were available to effectively suppress virus replication, the indigenous immune restorative responses of the host may be further stimulated and enhanced by rhIL-2. The use of protease inhibitors with nucleoside-analogue combination regimens appears to be most effective in controlling virus replication. High-dose intermittent rhIL-2 administered either intravenously or subcutaneously has been shown to be effective in inducing CD4 responses in a number of studies.

All patients receive HAART consisting of two nucleoside analogues (at least one of which is new to the patient) and the protease inhibitor indinavir for 12 weeks. At Week 10, an HIV RNA is done. Patients with more than 5,000 copies/ml are taken off the study. Patients with 5,000 copies/ml or less continue on the study and are stratified by their antiretroviral combination, and randomized to one of three treatment arms. Patients on Arm I continue HAART alone for an additional 72 weeks. Patients on Arm II continue HAART plus intermittent rhIL-2 by continuous intravenous (CIV) administration for an additional 72 weeks (9 courses). Patients who achieve both at least a 25-percent increase and at least a 100-cell increase in CD4 count above prerandomization baseline switch to subcutaneously administered rhIL-2 after 3 courses (24 weeks) or 6 courses (48 weeks) of CIV therapy for the remainder of their treatment course. Patients on Arm III continue HAART plus subcutaneous rhIL-2 for an additional 72 weeks (9 courses). [AS PER AMENDMENT 10/31/97: When protocol ACTG 928 is activated, patients randomized on ACTG 328 are stratified based on whether or not they have registered on ACTG 928.] [AS PER AMENDMENT 10/30/98: For the 4 weeks immediately prior to randomization, patients must receive continuous HAART therapy. Exceptions may be made by the protocol chair for short (i.e., less than 48 hours) interruptions of antiretroviral therapy during this period. Also per this amendment, two substudies have been added: A5033s and A5046s. Patients who choose to enroll in substudy A5046s must do so at Week 60 of ACTG 328. However, if the patient is already beyond Week 60 but not beyond Week 84 of ACTG 328 when A5046s becomes available, he/she is still encouraged to participate in A5046s. For such patients, ACTG 328 treatment is extended beyond Week 84 until the patient completes 24 weeks of A5046s.] [AS PER AMENDMENT 4/23/99: Patients who choose to enroll in substudy A5046s must do so at or after Week 64 of ACTG 328 and no later than Week 104.] [AS PER AMENDMENT 9/16/99: Patients choosing to enroll in A5046s must do so at or after Week 64 and no later than Week 128.] [AS PER AMENDMENT 1/22/99: A 16-week treatment extension is provided for patients who have reached Week 84 of ACTG 328 and intend to enroll in either A5051 or A5046s (which are currently not available).] [AS PER AMENDMENT 4/23/99: Therapy is extended for 24 weeks beyond the original 84 weeks. Study treatment for patients awaiting co-enrollment into A5046s continues for up to 20 weeks until co-enrollment and then until completion of A5046s (24 weeks). Study treatment continues for 24 weeks beyond Week 84 for patients who are awaiting enrollment in A5051 or who have chosen not to participate in A5046s.] [AS PER AMENDMENT 6/3/99: A 24-week treatment extension has been added for patients who have reached Week 100 and intend to enroll in A5046s.] [AS PER AMENDMENT 9/16/99: A treatment extension has been added for up to 44 weeks beyond Week 84 and through completion of A5046s (24 weeks) for on-study/on-study treatment patients who intend to co-enroll in A5046s. Patients who intend to enroll in A5051 or A5046s and remain on ACTG 328 study treatment are eligible for the treatment extension from Week 84 to Week 108. Patients who intend to co-enroll in A5046s and remain on ACTG 328 study treatment are eligible for the treatment extension from Week 108 to Week 128. All patients who are awaiting co-enrollment in A5046s must continue to receive their assigned ACTG 328 treatment regimen. Study treatment for ACTG 328 patients awaiting co-enrollment into A5046s will continue for up to 44 weeks and until completion of A5046s (24 weeks). Each patient is followed every 8 weeks for the duration of the treatment extension period to monitor for safety and to collect cells and plasma for storage. No substudy evaluations are performed during the 24-week extension.]

Inclusion Criteria

Concurrent Medication:

Required:

Patients must be able to initiate one of the following nucleoside analogue regimens of which at least one of the drugs must be new to the patient:

• ZDV + 3TC, ZDV + ddI, d4T + 3TC, or d4T + ddI. [AS PER AMENDMENT 4/5/00: ddI is contraindicated in patients with serum amylase or lipase values greater than 1.5 x ULN or who have a history of pancreatitis. ddI should also be used with extreme caution and only if clinically indicated in patients with known risk factors. For more information, see package insert.]

Allowed:

• Prophylaxis for Pneumocystis carinii pneumonia and other opportunistic infections according to current CDC recommendations. Prophylaxis, once started, should be continued despite increases in CD4 counts during the course of the study.
• Any standard regimen for an opportunistic infection.
• Maintenance therapy for opportunistic infections that develop on study treatment is permitted according to standard medical care; except for foscarnet during rhIL-2 administration and rifabutin and rifampin.
• Maintenance therapy with <= 1000 mg/day of acyclovir is permitted for recurrent genital herpes.
• Erythropoietin and filgrastim (G-CSF) are permitted when clinically indicated.
• Antibiotics for bacterial infections are permitted as clinically indicated.
• Medications for symptomatic treatment such as antipyretics and analgesics are permitted. Ibuprofen and acetaminophen are the preferred agents.
• Any elective standard immunizations, e.g., flu shot, should be given no less than 4 weeks prior to any blood draw for HIV RNA by bDNA assay.
• Topical corticosteroid use provided applied to a site separate from any skin test or rhIL-2 injection site (if frequent therapy is required for large surface area, protocol chair must be contacted).

Concurrent Treatment:

Allowed:

• Local radiation therapy.

Patients must have:

• Prior or current documentation of HIV seropositivity by any licensed ELISA and confirmation by either Western blot, positive HIV antigen, positive HIV culture, or a second positive antibody test by a method other than ELISA.
• CD4+ cell count 50 to 300 cells/mm3 [AS PER AMENDMENT 10/31/97: 50 to 350 cells/mm3] once within 30 days prior to study entry, as measured in an ACTG-certified laboratory.
• Ability to initiate one of the following nucleoside analogue regimens of which one of the drugs must be new to the patient:
• zidovudine (ZDV) + lamivudine (3TC), ZDV + didanosine (ddI), stavudine (d4T) + 3TC, or d4T + ddI. [AS PER AMENDMENT 10/31/97: A nucleoside analogue is considered "new" if it was never taken previously or if prior exposure to the nucleoside analogue was for less than 30 days and occurred more than 3 months prior to entry, with the exception of 3TC. For 3TC exposure to be considered "new", prior exposure must have been less than 1 week and must have occurred more than 3 months prior to entry.]

Patients must have the following conditions for the randomization step of the study:

• Completion of 11 weeks of HAART.
• HIV RNA of 5,000 copies/ml or less within approximately 1 week (Week 10) prior to randomization at week 11.

Exclusion Criteria

Co-existing Condition:

Patients with the following symptoms and conditions are excluded:

• Any active AIDS-defining illness by the CDC case definition with the exception of minimal (less than 10 lesions) cutaneous Kaposi's sarcoma.
• Significant cardiac insufficiency (NYHA grade 2). Patients with isolated hypertension controlled by antihypertensive medication but no cardiac disease are eligible.
• Untreated thyroid disease (treated and stable hyperthyroidism or hypothyroidism for at least 4 weeks prior to study entry is permitted).
• Sensitivity to albumin or allergy to albumin.

Concurrent Medication:

Excluded:

• Concurrent treatment with investigational antiretroviral agents other than FDA-sanctioned treatment IND drugs.
• Treatment for active cardiac disease, with the following medications: anti-anginal agents such as nitrates, calcium channel blockers, beta blockers, antiarrhythmics including digitalis and afterload reducers.
• Patients with malignancy requiring treatment with systemic or local cytotoxic chemotherapy.

Prohibited medications:

• interferons, interleukins (other than the study rhIL-2), sargramostim, dinitrochlorobenzene, thymosin alpha 1, thymopentin, inosiplex, polyribonucleoside, ditiocarb sodium, thalidomide, rifabutin, rifampin, midazolam, triazolam, oral ketoconazole, cisapride, terfenadine, astemizole, any investigational drug, therapy with foscarnet, systemic corticosteroids (systemic corticosteroids for <= 21 days are permitted for treatment of Pneumocystis carinii pneumonia; for other indications, contact the Protocol Chair), and other protease inhibitors. [AS PER AMENDMENT 4/5/00: St. John's wort is also excluded.]

Patients with the following prior conditions are excluded:

• History of autoimmune disease, including inflammatory bowel disease and psoriasis (stable autoimmune thyroid disease is permitted).
• Clinically significant CNS disease or seizures that have been active within 1 year prior to entry.

Prior Medication:

Excluded:

• Use of any known immunomodulatory therapy within 4 weeks prior to study entry including but not limited to drugs such as systemic corticosteroids; interferons; interleukins; thalidomide; sargramostim (GM-CSF); dinitrochlorobenzene (DNCB); thymosin alpha 1 (thymosin alpha); thymopentin; inosiplex (Isoprinosine); polyribonucleoside (Ampligen); ditiocarb sodium (Imuthiol).
• Any prior systemic treatment with rhIL-2.
• Any prior treatment with any protease inhibitor. [AS PER AMENDMENT 10/31/97: More than 14 days total prior treatment with any protease inhibitor.]
• Use of rifampin or rifabutin within 7 days prior to study entry.
• Use of cisapride (Propulsid), terfenadine (Seldane), astemizole (Hismanal), midazolam (Versed), triazolam (Halcion), ketoconazole (Nizoral), [itraconazole (Sporanox) AS PER AMENDMENT 10/31/97], or delavirdine within 2 weeks prior to study entry.

Active alcohol or substance abuse that, in the opinion of the investigator, will seriously compromise the patient's ability to adhere to the demands of the study.