Associations between medication adherence and virological outcomes (viral load suppression) in HIV treatment have been described, but the relationship between adherence and immunological response (CD4 cell recovery) is less well defined. In general, virological responses have been more closely associated with adherence than immunological responses.

CD4 T-cell counts were measured every 4 weeks from baseline to Week 8, every 8 weeks from Week 8 to Week 48, and every 12 weeks from Week 48 to Week 96.

For this analysis, time to immunological response was defined as the first of 2 consecutive CD4 cell counts at least 100 cells/mm³ above baseline after initiating lopinavir/ritonavir-based therapy.

Sensitivity analyses using alternative definitions of immune response were performed. MEMS monitors (electronic caps for medicine containers that record how often the bottle was opened) recorded and stored dosing histories for lopinavir/ritonavir.

Summary adherence statistics were computed for each inter-visit period during participation in the study. Subjects were post-stratified using timing adherence from baseline to Week 4 into four groups: <80%, 80% to 90%, 90% to <100%, and ≥100%.

Results

Of 228 subjects enrolled, 214 (94%) were considered evaluable. Demographic and baseline characteristics were similar for subjects in the once-daily and twice-daily dosing groups. However, subjects who received lopinavir/ritonavir once daily had significantly higher “timing adherence” from baseline to Week 4 compared with subjects who received lopinavir/ritonavir twice daily.

The median time to an immunological response with an increase in CD4 count of at least 100 cells/mm³ above baseline was similar for subjects receiving lopinavir/ritonavir once or twice daily (167 days vs 166 days; P=0.816).

There was no apparent difference between once-daily and twice-daily lopinavir/ritonavir with respect to the number of subjects censored from analysis (15.75% vs 21.84%; P=0.282). Censoring refers omission of subjects who did not achieve immunological response prior to premature discontinuation or completion of the study.

Longitudinal analysis suggested no difference over time in CD4 T-cell response profiles of subjects receiving lopinavir/ritonavir once or twice daily (treatment main effect, P=0.952; treatment-by-time interaction effect, P=0.231.

For the 4 Baseline to Week 4 “timing adherence” strata, the median times to immunological response were similar (P=0.684). There was a trend, however, toward more -- and earlier -- censoring with decreasing “timing adherence” (P <0.001).

Longitudinal analysis of CD4 cell counts also suggests no difference over time in the response profiles of subjects in the 4 “timing adherence” strata (“timing adherence” strata main effect, P=0.283; “timing adherence” strata-by-time interaction effect, P=0.207).

Sensitivity analyses using alternative definitions of immunological response were performed and showed similar results for the time to event analyses.

Immunological response defined by the first of 2 consecutive CD4 T-cell counts at least 50 cells/mm³ above baseline

Median time to immunological response was similar for subjects receiving lopinavir/ritonavir once versus twice daily (55 vs 50 days;

P=0.873), and there was no apparent difference between the once- and twice-daily lopinavir/ritonavir arms with respect to the number of subjects censored from analysis (8.66% vs 10.34%; P=0.812).

Median time to immunological response was similar for the 4 baseline to Week 4 “timing adherence” strata (P=0.336). However, there was a trend toward more -- and earlier -- censoring with decreasing “timing adherence” (P=0.001).

Immunological response defined by the first of 2 consecutive CD4 T-cell counts at least 200 cells/mm³ above baseline

Median time to immunological response was similar for subjects receiving lopinavir/ritonavir once versus twice daily (500 vs 592 days; P=0.292), and there was no apparent difference between the once- and twice-daily lopinavir/ritonavir arms with respect to the number of subjects censored from analysis (47.24% vs 57.47%; P=0.165).

Median time to immunological response was similar for the 4 baseline to Week 4 “timing adherence” strata (P=0.337), and there was no apparent difference between “timing adherence” strata with respect to the number of subjects censored from analysis (P=0.974).

Immunological response defined by the first of 2 consecutive CD4 T-cell counts at least 200 cells/mm³ above baseline for subjects withCD4 counts <200 cells/mm³ at baseline

Of the 214 subjects included in the primary analysis, 98 (45.8%) had baseline CD4 cell counts at or below 200 cells/mm³. Median time to immunological response was similar for subjects receiving lopinavir/ritonavir once versus twice daily (P=0.176), and there was no apparent difference between the once- and twice-daily lopinavir/ritonavir arms with respect to the number of subjects censored from analysis (48.28% vs 67.50%; P=0.066).

Median time to immunological response was similar for the 4 baseline to Week 4 “timing adherence” strata (P=0.623), and there was no apparent difference between “timing adherence” strata with respect to the number of subjects censored from analysis (P=0.974).

Conclusions

Based on these findings, the researchers concluded that time to immunological response and CD4 T-cell count profiles over time were not significantly different between subjects receiving lopinavir/ritonavir once or twice based.

Similarly, time to immunological response did not appear to differ based on adherence measures (from baseline to Week 4), though results may be confounded with censoring.

Abbott Laboratories, Abbott Park, IL.

10/26/07

Reference
RA Rode, TJ Podsadecki, GJ Hanna. CD4+ T-cell Response to Lopinavir/ritonavir-based Therapy in Antiretroviral-naïve HIV-1 Infected Subjects: The Role of Adherence. 11th European AIDS Clinical Society Conference. Madrid, Spain. October 24-27, 2007. Abstract (poster) P10.1/04.