Background: There is little clinical data on the response by women to anti-retroviral therapy (ART). In addition, cohort studies suggest that women have more aggressive HIV infection than men. Since ART is not dosed by weight, there may be gender differences in efficacy because of differences in drug levels achieved.

Methods: The Parkland HIV Database was queried for patients ART-naïve patients (pts) treated with Nelfinavir (NLF) containing regimens. This observational database contains demographic, pharmacy data, as well as CD4 counts and HIV-1 plasma RNA (VL). Baseline weights were abstracted from the pts’ medical record. Women subjects were paired with male subjects in a 1:2 match. The primary endpoint was the time to treatment failure as measured Kaplan-Meier analysis defined as two consecutive VL > 400 copies/ ml or discontinuation of NLF. Secondary endpoints were the proportion of pts with VL < 400 copies at 1 and 2 yrs on intention to treat analysis (ITT). Effect of weight on outcome was determined by identifying a cut point by a forward regression model. Potential confounding variables were analyzed by Cox proportional hazards model..

Results: 162 anti-retroviral naïve women treated with NLF containing regimen were matched by CD4 cell count and VL to 324 ART naïve men treated with NLF. Women on average weighed less than men (151 lbs vs 166 lbs, p < 0.01). Time to treatment failure was longer in women compared to men (295 days vs 236 days, p < 0.01). Percentage of patients with VL < 400 copies/ml was greater in women than men at 1 (48% vs 34%, p < 0.01) and 2 yrs (36% vs 25%, p = 0.01) on ITT. Forward regression identified a baseline weight of 180 lbs as being a significant predictor of failure. Cox model adjusting for baseline differences in the 2 groups demonstrated decreased risk of failure in women (OR = 0.75, p = 0.046) and increased risk of failure for individuals with baseline weight > 180 lbs (OR = 1.4, p = 0.04)

Conclusions: Anti-retroviral naïve women treated with NLF containing regimens had a better virologic response than men. This is only partially explained by differences in baseline weight. The role of gender should be further explored. Weight-based dosing of ART should also be examined.