This 'pragmatic' trial is a 2X2 open randomized study of patients in advanced HIV disease who have failed on conventional Highly Active Antiretroviral Therapy (HAART) regimens including all three classes of anti-HIV drugs. The first randomization will allocate patients to an intended 3-month antiretroviral drug-free period (ARDFP) or No ARDFP. The second randomization will allocate patients to Mega-ART (5+ drugs) or to Standard-ART (up to 4 drugs). The total study duration is 6.5 years with 5 years of intake and 1.5 year (minimum) of follow-up; median duration of patient follow-up is about 4 years. The target sample size is 390 patients and will provide 75% power to detect a 30% reduction in the hazard rate for the primary endpoint with mega-ART. Sixty-four sites will be participating in the trial--24 VA, 19 UK and 21 Canada.

Primary Hypothesis:

Compared to patients in Standard Antiretroviral Therapy (ART), patients in Mega-ART assuming full compliance, will experience a 30% reduction in the hazard of reaching a clinical endpoint (AIDS event or death).

Secondary Hypotheses:

Time to development of a new, non-HIV related serious adverse event, health related quality of life, the incidence of grade 3 or 4 clinical or laboratory adverse events and changes in virological and immunological markers (CD4 cell count, viral load, resistance profiles) will vary between the different treatment strategies.

Interventions:

Eligible patients will be randomized to one of four treatment strategy arms:

1. No Antiretroviral Drug-Free Period (No ARDFP) and Standard-ART
2. No Antiretroviral Drug-Free Period (No ARDFP) and Mega-ART
3. Antiretroviral Drug-Free Period (ARDFP) and Standard-ART
4. Antiretroviral Drug-Free Period (ARDFP) and Mega-ART

Note: The 'first' randomization will be ARDFP vs No ARDFP. Patients randomized to No ARDFP will receive their 'second' randomization at the same time. However, patients randomized to an Antiretroviral Drug Free Period (ARDFP) will receive their 'second' randomized assignment (Standard or Mega-ART) at the end of the ARDFP.

Note: All Serious Adverse Events were coded using the MedDRA coding dictionary; other (not serious) Adverse Events were collected as part of the study but were not coded using MedDRA or any other standardized coding dictionary.

This is the first trial of a Tri-National collaboration effort between the UK MRC, the Canadian CIHR and the VA CSP. The OPTIMA Trial was reviewed and approved by CSEC on October 12, 2000. The pre-kickoff meeting was held on March 21, 2001 in Washington, DC. The VA study kickoff meeting was held in Dallas, TX on May 16-18, 2001 and the Canadian kickoff was held in Toronto on May 29, 2001. The UK will have individual site initiation. As of October 17, 2005 there have been 357 patients enrolled in OPTIMA, at 64 sites in the three countries (279 in the VA, 41 in Canada and 37 in the UK). To date there are 64 sites actively participating in the study (24 in the VA, 19 in UK and 21 in Canada). Last date of patient follow-up was December 31, 2007.

Inclusion Criteria:

• Ability to provide informed consent
• Age of 18 years or more
• Serologic or virologic diagnosis of HIV infection
• Failure of at least two different multi-drug regimens that include drugs of all 3 classes that the patient can tolerate or laboratory evidence of resistance to drugs in each of the 3 classes
• Had at least 3 months of current ART and are still on treatment
• Two most recent results (which can include screening) on current ART of CD4 count less than or equal to 300 cells/mm3 or less than or equal to 15%, and a plasma viral load greater than or equal to 5,000 copies/ml (Roche Amplicor, v1.0), or greater than or equal to 2,500 copies/ml (by bDNA: Bayer v3.0/Chiron v3.0 or PCR:Roche Amplicor Monitor/COBAS v1.5)

Exclusion Criteria:

• Pregnancy, breast-feeding or planned pregnancy
• Likelihood of poor protocol follow-up or if Mega-Art is not feasible (due to significant intolerance of many ARV drugs)
• Serious, uncontrolled major opportunistic infection (OI) within 14 days of screening
• Likelihood of early death due to non-HIV disease