CD4+ T cell evolution and predictors of its trend before and after tenofovir/didanosine backbone in the presence of sustained undetectable HIV plasma viral load.

Background: Tenofovir with full-dose didanosine has been associated with paradoxical CD4 + T cell decrease despite virological suppression. We investigated whether tenofovir plus didanosine at a weight-adjusted dosage could be responsible for such an effect, and factors associated with CD4 + T cell count evolution under this combination.

Methods: This was a prospective observational multicohort study (Italian MASTER and Spanish Hospital Carlos III HIV cohorts). Patients with HIV plasma viral load suppression for >/= 6 months who switched to an antiretroviral combination including tenofovir plus didanosine were studied, as long as virological success was maintained. CD4 + T cell count variations over time (slopes) were compared before and after switching to tenofovir plus didanosine using linear mixed models and segmented regression analysis.

Results: Annual time-weighted CD4 + T cell count slope did not change significantly after the prescription of tenofovir plus didanosine: it was 14 cells/mm(3) [95% confidence interval (CI) - 7 to 35] from month - 24 to month - 12, 12 cells/mm(3) (95% CI - 14 to 38) from month - 12 to the time of switching, 30 cells/mm(3) (95% CI 5-55) from switching to month + 12 and 15 cells/mm(3) (95% CI - 8 to 39) from month + 12 to month + 24 after switching to tenofovir plus didanosine. No significant change in the slope of the segment after the switch to tenofovir plus didanosine-containing regimens when compared with the segment preceding the intervention was found (CD4 + T cell count slope change: 24 cells/mm(3); 95% CI - 10 to 58). Similar results were obtained using CD4 + T cell percentage over total lymphocytes. The significant independent predictors of lower CD4 + T cell count slope were older age (P = 0.006), lower nadir CD4 + T cell count (P < 0.001) and positive hepatitis C virus antibody (P = 0.03). Moreover, reduced estimated creatinine clearance was an additional independent predictor of lower CD4 + T cell count slope (P = 0.02), but only after excluding nadir CD4 + T cell count.

Conclusions: Tenofovir plus didanosine (weight-adjusted dosage) was not associated with paradoxical CD4 + T cell decrease in our patients maintaining undetectable HIV plasma viral load for a maximum of 24 months after switching. Several factors could explain variability in CD4 + T cell count evolution in these patients.