Abstract
Abstract Background Integrase strand transfer inhibitor (INSTI)-based antiretroviral therapy (ART) is recommended first-line HIV treatment. We recently demonstrated increased weight gain associated with INSTI use among women living with HIV (WLH) enrolled in the Women’s Interagency HIV Study (WIHS), raising concern for cardiometabolic consequences. We, therefore, evaluated the effects of INSTI use on lipids, insulin resistance, and glycemic control in WLH. Methods Data from 2008 to 2017 were analyzed from WLH enrolled in WIHS. Women who switched to or added an INSTI to ART (SWAD group) were compared with women who remained on non-INSTI ART (STAY group). Outcomes included changes in fasting total cholesterol (TC), low-density lipoprotein (LDL), high-density lipoprotein (HDL), triglycerides (TG), and glucose; hemoglobin A1c; and incident insulin resistance (defined as homeostatic model assessment of insulin resistance [HOMA] score ≥2). Outcomes were measured 6–12 months before and 6–18 months after INSTI switch/add in the SWAD group with comparable time points in the STAY group. Linear regression models compared change over time in each outcome by SWAD/STAY group, adjusted for age, race, WIHS site, income, smoking status, statin use, and ART regimen at baseline. Results In total, 881 WIHS participants (182 SWAD and 699 STAY) were followed for a mean 1.8 (±1.1) years. Mean age was 49 (±8.8) years, BMI was 31 (±8.2) kg/m2, and 49% were Black. At baseline, SWAD vs. STAY was more likely to report NNRTI (vs. PI)-based ART and statin use (both P < 0.0001), but all baseline lipid and glucose variables were similar. Compared with STAY, the SWAD group experienced significantly greater decreases in HDL (−2.4 vs. +0.09 mg/dL, P = 0.03) and trended toward greater decreases in TC (−2.6 vs. −2.4 mg/dL, P = 0.07) at follow-up, without significant differences in TG or LDL. The SWAD group had significantly greater increases in A1c (+0.08% vs. −0.05%, P = 0.01) but trended toward lower incidence of insulin resistance (19% vs. 32%, P = 0.05). Conclusion Despite reported increases in weight, INSTI use was associated with only modest changes in lipid measurements and glycemic control during short-term follow-up of WLH compared with non-INSTI ART. Research is needed to elucidate long-term cardiometabolic effects. Disclosures Anandi N. Sheth, MD, MS, Gilead Sciences, Inc.: Research Grant.