PrEP use and HIV seroconversion rates in adolescent girls and young women from Kenya and South Africa: the POWER demonstration project

Abstract

Introduction: HIV incidence remains high among African adolescent girls and young women (AGYW). The primary objective of this study is to assess pre-exposure prophylaxis (PrEP) initiation, use, persistence and HIV acquisition among African AGYW offered PrEP in order to inform PrEP scale-up.

Methods: POWER was a prospective implementation science evaluation of PrEP delivery for sexually active HIV-negative AGYW ages 16-25 in family planning clinics in Kisumu, Kenya and youth and primary healthcare clinics in Cape Town and Johannesburg, South Africa. Follow-up visits occurred at month 1 and quarterly for up to 36 months. PrEP users were defined based on the month 1 refill. PrEP persistence through month 6 was assessed using Kaplan-Meier survival analysis among AGYW with a month 1 visit, defining non-persistence as an ≥15 day gap in PrEP availability for daily dosing. PrEP execution was evaluated in a subset with PrEP supply from the prior visit sufficient for daily dosing by measuring blood tenofovir diphosphate (TFV-DP) levels.

Results: From June 2017 to September 2020, 2550 AGYW were enrolled (1000 in Kisumu, 787 in Cape Town and 763 in Johannesburg). Median age was 21 years, 66% had a sexual partner of unknown HIV status, and 29% had chlamydia and 10% gonorrhoea. Overall, 2397 (94%) initiated PrEP and 749 (31%) had a refill at 1 month. Of AGYW who could reach 6 months of post-PrEP initiation follow-up, 128/646 (20%) persisted with PrEP for 6 months and an additional 92/646 (14%) had a gap and restarted PrEP. TFV-DP levels indicated that 47% (91/193) took an average of ≥4 doses/week. Sixteen HIV seroconversions were observed (incidence 2.2 per 100 person-years, 95% CI 1.2, 3.5); 13 (81%) seroconverters either did not have PrEP dispensed in the study interval prior to seroconversion or TFV-DP levels indicated <4 doses/week in the prior 6 weeks.

Conclusions: In this study of PrEP integration with primary care and reproductive health services for African AGYW, demand for PrEP was high. Although PrEP use decreased in the first months, an important fraction used PrEP through 6 months. Strategies are needed to simplify PrEP delivery, support adherence and offer long-acting PrEP options to improve persistence and HIV protection.

Trial registration: ClinicalTrials.gov NCT03490058.

Keywords: Africa; HIV; adherence; persistence; pre-exposure prophylaxis; young women.

Figure 1

PrEP initiation and persistence beyond 6 months among women who initiated PrEP. The flow chart depicts PrEP initiation and persistence among the 2550 women who enrolled in the POWER cohort. PrEP persistence was assessed based on attendance of the 1 month visit for a PrEP refill, and persistence through 6 months among those reached at least 6 months of post‐PrEP initiation follow‐up time. PrEP non‐persistence was defined as a gap of ≥15 days between the refill return date and the next PrEP refill dispensation, assuming daily dosing prior to the next PrEP dispensation date. PrEP persistence beyond 6 months was defined as persistent use through 6 months and PrEP refill at the visit 6 month visit.

Figure 2

PrEP persistence among women who had PrEP persistence beyond 1 month. The cumulative probability of a PrEP non‐persistence failure event within 6 months of PrEP initiation was modeled using the Kaplan–Meier estimator of the survival function. The 749 women who had a PrEP refill at 1 month are included in the analysis. The vertical drop at each point indicates PrEP non‐persistence “events,” defined as a gap of ≥15 days between refills that would provide sufficient PrEP supply for daily dosing.