Background: A significant amount of HIV-related morbidity has been attributable to genital tract disease in women. It is established that tenofovir (TDF), emtricitabine (FTC), and raltegravir (RAL) have high genital concentrations while atazanavir/ritonavir (ATZ) has more modest levels. We examine if RAL vs ATZ-based regimen is superior with respect to improved cervical immune reconstitution and less immune activation in HIV+ women. Methodology: Peripheral blood (PB), cervical brush and lavage fluid (CVL), and cervical biopsies were collected from HIV+ women receiving TDF/FTC and either ATZ (n=19) or RAL (n=14). A history of CD4+ T-cells/mm3 (CD4) >= 300, and HIV RNA <48 copies/mL for a minimum of 6 months was required for entry. Cervical cells were analyzed by flow cytometry to determine %HLA-DR+CD38+(DR+38+) on CD4+ and CD8+T-cells. Sections of cervical biopsies were stained by immunohistochemistry and frequencies of CD3+CD4+, and CD3+CD8+cells per mm2 of tissue were determined using quantitative image analysis. RAL and ATZ trough concentrations in plasma and CVL were quantified with validated assays. T-tests, Fisher’s exact and linear regressions were used with log transformations as appropriate. Data are expressed as means or geometric means (CI). Results: RAL and ATZ groups did not differ by age (46 vs 43 years, p=0.5), race (p=1.0), CD4 nadir (309 vs 340, p=0.65), current CD4 (663 vs 758, p=0.4) or PB CD4:CD8 ratio (1.0 (0.84, 1.18) vs 1.03 (0.78, 1.37), p=0.8). Years since HIV diagnosis and years on current ARV was significantly lower in women taking RAL vs ATZ (8 vs 13; p=0.03 and 2 vs 5; p=0.01, respectively). Plasma RAL vs ATZ levels were 121 (44, 335) vs 620 (419, 918) ng/mL. CVL levels for RAL vs ATZ were 9 (4, 23) vs 4 (2, 8) ng/mL. Genital:plasma (G:P) concentration ratio was significantly higher for RAL (0.078 (0.034, 0.18)) vs ATZ (0.006 (0.004,0.011), p<0.001). Drug concentrations in plasma correlated with those in CVL (RAL ρ=0.63, p=0.02; ATZ ρ=0.54, p=0.03). RAL and ATZ groups did not differ by cervical %DR+38+CD4+T-cells (20 vs 16%, 0=0.4) or %DR+38+CD8+T-cells (23 vs 21%, p=0.7). Cervical CD4:CD8 T-cell ratio in RAL group 0.46 (0.33, 0.63) was not significantly different than ATZ group (0.52 (0.36, 0.74); p=0.6). In HIV- women, the CD4:CD8 T-cell ratio has been reported as 4:1. CD4:CD8 T-cell ratio was significantly higher in PB than cervical tissue in both groups (p<0.001). After adjusting for time on ARV, log(G:P drug ratio) was not a significant predictor of cervical CD4:CD8+ T-cell ratio or immune activation (p>= 0.2). Conclusions: Low cervical tissue CD4:CD8 ratio persists after years of ARV. Despite higher genital concentrations of RAL compared to ATZ, there was no significant difference in cervical CD4:CD8 ratio or immune activation between women on these regimens.