Steve Deeks reported, I think, key findings from his study of treatment interruption. Although resistance in plasma was not detected after interrupting therapy in 16/18 patients, in 4 of 8 patients studied resistance was detected in the patients PBMCs (peripheral blood mononuclear cells. This suggests a risk of therapy interruption. As a result, he suggests that the safety of strategic therapy interruptions remains to be determined. Following is a review of his talk.

The objective of this study was to evaluate the virologic and immunologic effects of discontinuing antiretroviral therapy in patients experiencing long-term virologic failure of a protease inhibitor based regimen by looking at viral replication, drug resistance, replication capacity, peripheral CD4 T cell counts. A secondary objective was to determine if resistance persists in PBMCs after shift to wild-type in plasma.

Included in the study were patients who had documented virologic failure of a protease regimen (>500 copies/ml) for at least one year, had a VL >2500 copies/ml at screening for this study, had been on a stable regimen for >16 weeks, and had a willingness to discontinue therapy. Patients were randomized to STI or continued therapy. After baseline patients came to office visit every week for 12 weeks at which time they restarted therapy and visited every 4 weeks. At each visit starting at baseline HIV-RNA (bDNA), CD4/CD8, and phenotypic resistance testing were performed. At baseline and week 12 (when restrarting therapy) PBMC co-culture and virus replication (fitness) capacity tests were performed. Patients were encouraged to restart therapy when >1 log VL increase or 50% loss of CD4s occurred.

Eighteen patients were randomized to discontinue therapy. At baseline their CD4s were 245 (range 104-307), HIV-RNA, 4.6 log--40,000 copies/ml (range 4.0-5.0 log ),average time on therapy-- 36 months (range 30-39), virologic failure (months) --31 (range 28-33), PI resistance (fold change, Virologic assay)) --56 (23-79).

As a consequence of stopping therapy (baseline to 12 weeks), there was a

• median drop in CD4s of 94 cells (range -28 to -126)
• median increase in VL of +0.82 log (range 0.34-0.92 log),
• 17 patients had high-level PI resistance when starting study, and 16 reverted to being fully PI susceptible (susceptible was defined as <2.5 fold resistance or loss of susceptibility)
• mean time for this switch to occur was 8.5 weeks (range 2-15 weeks)
• NRTI resistance persisted often at much reduced levels, after reversion to a fully PI susceptible virus in 7 patients
• CD4 count typically declined as VL increased subsequent to the shift in phenotypic susceptibility in the 16 patients. Prior to the shift there was a slow increase in VL starting about 4 weeks and a slow decrease in CD4. After the shift there was a "dramatic" increase in VL and decrease in CD4. Deeks suggested that the slow increase in VL prior to the shift to PI phenotypic susceptibility reflected residual antiviral activity. The more quick increase after the shift reflected the increased replication capacity
• Replication fitness was measured by an assay and increased from a median 22.3% to 67.1% (p=0.004)
• Although resistance was undetected in plasma, resistant virus identical to baseline was cultured from PBMCs 12 to 36 weeks after therapy discontinuation in 4 of 8 patients showing phenotypic reversion. In other words, resistant virus can persist in PBMC after wild-type switch. Deeks said this has implications for therapy interruptions. The safety and efficacy of STIs as a therapeutic strategy remains to be determined
• Deeks concluded that despite high-level resistance while experiencing virologic failure on therapy, there is continued immunologic and virologic benefit