Changes in kidney function among MSM initiating on-demand TDF/ FTC for HIV prep

Background: Pre-exposure prophylaxis (PrEP) with TDF/FTC is recommended for HIV prevention. Daily PrEP with TDF/FTC is associated with a small but statistically significant decrease in estimated glomerular filtration rate (eGFR) similar to HIV-infected patients on TDF. We wished to assess whether on demand TDF/FTC based-PrEP could minimize the risk of eGFR reduction among MSM. Methods: We used data from the randomized double-blind placebo-controlled ANRS-IPERGAY trial conducted among HIV-uninfected MSM with creatinine clearance >60mL/min. eGFR was assessed using CKD-EPI equation at enrolment, months 1, 2 and every 2 months thereafter. We evaluated the mean decline slope of eGFR change from baseline and the occurrence of eGFR <70mL/ min/1.73m2 in the placebo and on-demand TDF/FTC groups. We also determined risk factors for eGFR <70mL/min/1.73m2 in all patients initiating TDF/FTC included in the blind or the open-label extension phases of the study. Results: During the blind phase, 201 participants were randomized to placebo and 199 to on demand TDF-FTC. Participants on TDF/FTC took a median number of 15 pills/month (IQR 11 to 21). The mean eGFR at baseline was 106mL/ min/1.73m2. During a median follow up of 9.3 months, the mean decline slope of eGFR was-0.13 and-0.07 mL/min/1.73m2 per month in the TDF/FTC and placebo group, respectively (P=0.27). The cumulative proportion of patients with an eGFR <70mL/min/1.73m2 at 12 months was higher on TDF-FTC: 8% [95%CI 4-13%] than placebo: 3% [CI 0-6%], P=0.04. Compared to placebo, the risk of eGFR <70mL/min/1.73m2 did not increase significantly in patients who took <15 pills/month: HR 1.75 [CI 0.65-4.7%] as compared to those using ≥15 pills/month: HR 2.54 [CI 1.07-6.04%]. Including both phases, 389 participants initiated on demand TDF/FTC with a median follow up of 19.1 months. Small but significant decline in eGFR occurred over time (mean slope:-0.09mL/ min/1.73m2 per month, P<0.01). Only 2 participants had persistent eGFR <60mL/min/1.73m2 and 3 discontinued TDF/FTC for kidney function decline. The cumulative proportion of eGFR <70mL/min/1.73m2 from baseline was 14% [9-18%] at 24 months. Factors associated with eGFR <70mL/min/1.73m2 were high pill use (HR 1.9 [CI 1.03-3.49%], P=0.04), age > 40 years (P<0.01) and low eGFR at baseline (P<0.01). Conclusion: On demand PrEP with TDF/FTC is associated with limited and nonclinically relevant eGFR decline, especially in young participants, those with low pill use and high baseline eGFR.