613    

Background:  Circulating CD4⁺ T-cell count is a strong predictor of the short-term risk for opportunistic infection in HIV infection. Potent antiretroviral therapy partially restores CD4^– T-cell counts, but there are no direct comparisons of the extent of this restoration in different parts of the world.

Methods:  Differences in total, first phase (week 0 to 8) and second phase (week 8 to 48) changes of CD4⁺ cell counts in response to antiretroviral treatment were assessed across 4 regions of the world—South Africa, Europe, and North and South America. The primary study sample consists of 301 treatment-naïve patients who achieved complete follow-up for 48 weeks in a randomized clinical trial of 3 different doses of atazanavir vs nelfinavir plus didanosine and stavudine (BMS-424-007). The main endpoint was magnitude of CD4 cell increase across regions; multivariable logistic regression analysis was used to assess the effect of covariates including demographics, drug use, treatment regimen, baseline plasma HIV RNA, virologic response, and baseline CD4 and CD8 cell counts.

Results:  Median 48-week CD4 cell increases in Africans (n = 105), Europeans (n = 97), North Americans (n = 45), and South Americans (n = 54) were 109, 261, 162, and 204 (p < 0.001). The bulk of the difference was attributable to smaller median second phase (week 8 to 48) changes in the African patients (37 cells vs 141, 74, and 110 cell increases in the Europeans, North and South Americans, respectively). Gender was also associated with CD4 cell response, with women having larger CD4 increases. Complete virologic responses tended to be more common in Africans and no co-variate other than region explained the lower magnitude CD4 cell gains in this group. An identical analysis conducted on data from 4 other clinical trials including these populations revealed a similar trend towards smaller CD4 increases in Africa in all but 2 of the studies; in all cases, virologic responses were comparable across geographic regions.

Conclusions:  Despite better virologic responses to antiretroviral therapies, South African patients experienced smaller CD4 cell-count increases. The mechanism for these attenuated responses remains to be elucidated; differences in viral clades, co-infections, immune activation, or unrecognized biases are unproven possibilities. These differences may have important implications regarding standards for treatment initiation and expectations for immune restoration in South Africa.

Keywords: CD4 Restoration; Antiretroviral Therapy; Geographic Differences