Background Two trials of acyclovir (ACV) 400 mg twice daily as daily suppressive therapy against herpes simplex virus type 2 (HSV-2) proved ineffective for prevention of HIV acquisition. Explanations for this lack of efficacy are unclear. In one of these trials, HPTN 039, ACV was modestly effective in reducing genital ulcers due to HSV-2; however, there was less reduction in the frequency of genital ulcers and higher HSV-2 DNA quantity in breakthrough lesions in women from African sites than in men from US sites. Pharmacokinetic differences of ACV have been proposed as one explanatory variable for these findings. Methods Sixty-eight HIV-negative, HSV-2 seropositive women participated in a pharmacokinetic study of ACV after completion of HPTN 039 (a phase III, randomised clinical trial of daily acyclovir 400 mg p.o. twice daily). Following a single oral dose of 400 mg of acyclovir, blood was collected over an 8 h period. An LC/MS/MSbased assay determined ACV concentrations. PK parameters were estimated using non-compartmental methods. Results Sixty-six African women had complete PK data for evaluation. Mean (range) age was 36 (21e54) years and weight was 70 (40e129) kg. The geometric mean (95% CI) for PK parameter estimates were: Cmax 0.31 ug/ml (0.28, 0.34), AUC0-inf 1.59 ug*hr/ml (1.43, 1.76), Tmax 1.56 h (1.40, 1.80), and half-life 2.8 h (2.5, 3.0). This Cmax was lower than 8 comparable single dose ACV PK studies in non-African populations, mean 46% lower (range 28%e 59% lower, all p values <0.006). Similarly, AUC was lower than all other studies, mean 38% lower (range 26%e62%, all p values <0.001). In some studies, Tmax was earlier and the half-life was shorter. Subject weight did not explain the differences. Conclusion Acyclovir exposure in black African women was lower than in comparable ACV studies of non-African populations. These statistically significant differences in drug exposure (Cmax and AUC) may be clinically significant and partly explain the modest effects of ACV on HSV-2 recurrence in these African women. Results The epidemiological impact of PrEP is largely driven by programme characteristicsdcoverage, prioritisation strategy and time to scale updas well as individual's adherence behaviour. If PrEP is prioritised to key groups, it could be a cost-effective way to avert infection and save lives (up to 8% less new infections with 5% coverage). Across all our scenarios the estimated highest cost per DALY gained (US$2755) is below the WHO recommended threshold for costeffective interventions for the region (