Reconciling counterfactual placebo HIV incidence rate estimates for an active-controlled pre-exposure prophylaxis trial, PrEPVacc

Reconciling counterfactual placebo HIV incidence rate estimates for an active-controlled pre-exposure prophylaxis trial, PrEPVacc | Research Square Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Reconciling counterfactual placebo HIV incidence rate estimates for an active-controlled pre-exposure prophylaxis trial, PrEPVacc Sheila Kansiime, Christian Holm Hansen, Henry Bern, Julie Fox, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6344793/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 5 You are reading this latest preprint version Abstract Background Estimating counterfactual placebo HIV incidence rates (the rate in a placebo arm if it had been included) is critically important for the interpretation of active-controlled HIV pre-exposure prophylaxis (PrEP) trials. The PrEPVacc trial (NCT04066881) enabled this estimation using three approaches. In this paper, we contrast the resulting estimates and explore the potential biases of each. Methods The PrEPVacc HIV prevention trial conducted in Uganda, Tanzania and South Africa included a non-inferiority comparison of two daily oral PrEP regimens—Emtricitabine/tenofovir (TDF/FTC) and Tenofovir alafenamide/Emtricitabine (TAF/FTC) administered for a target duration of 26 weeks (PrEP component). The trial also included a concurrent placebo-controlled, three-arm HIV vaccine trial. The analyses described in this paper were aimed at estimating counterfactual placebo HIV incidence for the active-controlled PrEP component of the trial. Estimates generated using multivariable Poisson regression models and data from a pre-trial registration cohort have been published previously. In this paper we generate estimates using data from the post-PrEP component of the trial and the adherence-efficacy relationship of TDF/FTC and comprehensively review the potential biases of each approach. Results Participants in the PrEP component of the PrEPVacc trial were predominantly female (87%), with 53% aged ≥ 25 years, 60% identifying as sex workers, and 26% diagnosed with a sexually transmitted infection at baseline. Point estimates from the different approaches ranged from 0.7 to 1.9/100 person-years. All three counterfactual placebo incidence estimation approaches had considerable biases, including uncertainties around calendar trends in the pre-trial registration cohort and whether vaccine effects observed in the post-PrEP period of the PrEPVacc trial applied similarly during the PrEP component. The adherence-efficacy approach was particularly imprecise and unreliable due to the small number of HIV infections (two) observed in the TDF/FTC arm and the imprecision of the risk reduction estimates attributed to the observed level of adherence (Tenofovir Diphosphate levels). Conclusion Obtaining a precise estimate of counterfactual placebo HIV incidence in the PrEPVacc trial was challenging. Further development and standardisation of counterfactual placebo HIV incidence estimation approaches would simplify analyses of future HIV pre-exposure prophylaxis trials. Pre-Exposure prophylaxis Counterfactual placebo HIV incidence Clinical trial design non-inferiority TDF/FTC TAF/FTC Figures Figure 1 Figure 2 Figure 3 Introduction Pre-exposure prophylaxis (PrEP) is a highly efficacious HIV prevention intervention [ 1 ]. Trials of novel PrEP interventions are now designed as active-controlled and require large sample sizes to demonstrate non-inferiority in comparison to the current standard of care, oral Emtricitabine/tenofovir (TDF/FTC) [ 2 , 3 ]. Future trials assessing non-inferiority of new interventions to injectable PrEP will likely require even larger sample sizes. Trials of other HIV prevention interventions, such as vaccines or monoclonal antibodies, can include placebo arms while providing oral PrEP to all participants. However, these trials also require large sample sizes, especially if PrEP uptake is high among participants [ 4 ]. An alternative to conducting a large-scale non-inferiority trial is to base active-controlled PrEP evaluations on comparisons with an estimated counterfactual placebo HIV incidence rate [ 2 , 3 , 5 ] also referred to as a hypothetical placebo incidence rate or background HIV incidence rate. Estimated counterfactual placebo incidence rates have gained increasing acceptance with the expectation that they will provide an approach to utilising limited person-years of follow-up, or fewer observed HIV infections more efficiently [ 3 , 6 ]. Several methods have been proposed for estimating counterfactual placebo incidence rates, that can be broadly grouped into three categories: (i) utilization of external controls ( estimating HIV incidence from historical or concurrent cohorts, utilising HIV risk indicators, risk scores, and correlations with other sexually transmitted infections (STIs) among others); (ii) inferring incidence based on the estimated effectiveness of TDF/FTC as a comparator in an active-controlled trial; and (iii) using recency assays to estimate HIV incidence shortly before screening. Beyond HIV prevention, external controls (usually historical cohorts) have been used in clinical trials for decades and are considered acceptable by regulatory bodies, provided comparability to trial participants and data reliability can be demonstrated [ 7 – 12 ]. The PrEPVacc trial was designed to use an averted infection ratio (AIR) framework to compare oral TAF/ FTC, a novel form of PrEP, to oral TDF/ FTC. The AIR approach requires smaller sample sizes than traditional incidence rate ratios but requires a reliable estimate of the counterfactual placebo HIV incidence rate [ 13 ]. The AIR contrasts the difference between the counterfactual placebo incidence and the incidence in an experimental arm, to the difference between the counterfactual placebo incidence and that in the standard intervention (control) arm. In the PrEPVacc trial, three approaches could be used to estimate the counterfactual placebo HIV incidence rate for the PrEP component. These included a) Using HIV incidence data from a pre-trial registration cohort and utilising statistical modelling to adjust for any calendar time trends and differences between the cohort and the trial population in terms of important predictors of HIV incidence; (b) Using HIV incidence data from the post-PrEP component of the PrEPVacc trial, where PrEP was promoted but not an investigational product, while accounting for potential vaccine effects, and (c) Estimating incidence based on the observed adherence data in the PrEPVacc PrEP trial component and the known effectiveness of TDF/FTC. In this paper, we estimate the counterfactual placebo HIV incidence rate using approaches (b) and (c), alongside approach (a) that we published previously, incorporating sensitivity analyses for each. We then consider the potential biases associated with each approach. Materials and methods Study design and population The PrEPVacc study, which included a pre-trial registration cohort and an HIV prevention trial, was conducted at four study sites in three countries: Masaka, Uganda; Mbeya and Dar es Salaam, Tanzania; and Durban, South Africa. Enrolment in the registration cohort took place between July 2018 and October 2022, and enrolment in the PrEPVacc trial took place between December 2020 and March 2023. Individuals were eligible to participate in the registration cohort if they were HIV-uninfected adults (18–45 years), at high-risk of HIV infection and consented to the study procedures. Participants were followed for up to 3 years or until trial enrolment. Once enrolled in the cohort participants received HIV testing every 3 months, risk reduction counselling and referral for PrEP. Additional details can be found elsewhere [ 14 ]. For the trial, participants were considered eligible if they were HIV-uninfected adults aged 18–40 years at the time of enrolment, willing and able to provide informed consent, willing and able to comply with the visit schedule and provide blood, urine and other samples, available for at least 82 weeks from screening and, if female, willing to use a highly effective method of contraception. A comprehensive list of inclusion and exclusion criteria can be found elsewhere [ 17 ]. At trial enrolment (week 0), participants were randomised 1:1 to one of two oral PrEP regimens: once daily TDF/FTC or TAF/FTC, for a target duration of 26 weeks. They were also concurrently randomised 1:1:1 to one of two HIV vaccine regimens or saline placebo, with vaccinations administered at 4 time-points at weeks 0, 4, 24 and 48 (Fig. 1 ). The primary endpoint for the analysis of both the vaccine and PrEP interventions was confirmed incident HIV infection. HIV infections diagnosed before or at 2 weeks after the third injection visit (week 26) contributed to the primary PrEP analysis, while infections occurring thereafter contributed to the primary vaccine analysis (Fig. 1 ). Follow-up HIV testing was conducted with a trial-specific HIV testing algorithm. Bio-Rad Genscreen™ ULTRA HIV Ag-Ab assay was used for screening; positive results were confirmed by an HIV-1 RNA PCR assay. Details can be found elsewhere [ 14 , 15 ]. During the post-PrEP component of the PrEPVacc trial (beyond week 26 or 2 weeks after the third vaccination), participants were encouraged to access PrEP through local providers. Participants were followed for a minimum target of 74 weeks from enrolment, or until the end of the trial. The rationale for incorporating both the PrEP and vaccine in the PrEPVacc trial design was that the clinical protective effect of the vaccines was not anticipated until after the third vaccination, when participants were expected to reach peak immunogenicity. PrEP would provide protection from HIV infection until this time point. Statistical analyses Data management and analysis were conducted in Open Clinica (Community Edition) and Stata version 18.0 (College Station, Texas, USA) respectively. In the PrEPVacc trial, the approaches described in more detail below could be utilised to estimate the counterfactual placebo HIV incidence rate. a) Utilising incidence data from the observational pre-trial registration cohort Results from this analysis have been published previously [ 16 ]. We developed a multivariable Poisson regression model to assess the associations between HIV incidence in the registration cohort and baseline predictors (socio-demographic and behavioural variables) as well as time-related predictors (calendar time, follow-up duration). We then used the estimated regression coefficients and the participant characteristics from the active-controlled PrEP trial to predict the counterfactual placebo incidence rate. Confidence intervals (95% CIs) were estimated using bootstrapping methods resampling participants with replacement from the registration cohort. Sensitivity analyses were also conducted to evaluate the effect of calendar period. b) Utilising incidence data from the post-PrEP component of the PrEPVacc trial (beyond week 26) Similar to the analyses conducted to predict the counterfactual placebo HIV incidence rate using the pre-trial registration cohort data, we utilised multivariable Poisson regression models to adjust for differences between the active-controlled trial dataset (up to week 26) and the external control dataset (beyond week 26) and predict the counterfactual placebo HIV incidence for the active-controlled trial. Data from all participants previously enrolled in the PrEPvacc PrEP trial component (both PrEP arms) were included. Forward stepwise model building was used to derive the final multi-variable model, with predictors added to the model if they had a P-value < 0.2. The allocated trial vaccine arm, PrEP arm, and age, were included a priori. Associations between HIV incidence and other variables (socio-demographic and behavioural) were reported. In this paper, we assumed the vaccine effect remained constant throughout the trial. (This assumption is reviewed and justified in the Discussion section). We also investigated the possibility of HIV risk reduction due to continued PrEP use beyond week 26, by conducting a sensitivity analysis limited to follow-up periods after the date of the last reported PrEP use for each participant. Additionally, we considered the possibility of a calendar period effect and reported analyses that incorporated this factor. c) Estimating HIV incidence based on observed adherence and known effectiveness of TDF/FTC. Evaluations of PrEP adherence in the PrEPVacc trial, utilising various adherence measures, will be published separately. Here, we utilised estimated adherence based on dried blood spot data (Tenofovir Diphosphate levels, which estimate average number of pills per week over the last 6-weeks) at week 8, and referenced literature [17–20], to estimate the HIV incidence risk reduction attributable to TDF/FTC during the PrEPVacc PrEP trial component. We adopted HIV risk reduction estimates from the IPREX, HPTN083 and HPTN084 trials [17], incorporating sensitivity analyses informed by each of these trials. We also considered what the HIV incidence risk reduction attributable to TDF/FTC would have been assuming 85% protection when TFV-DP is detectable, as reported by a systematic review by Fonner et al [18]. (This is a broad assumption as various definitions of “detectable drug” were used across the trials in the systematic review). The overall estimates of risk reduction in the PrEPVacc trial were calculated as weighted averages of the risk reduction effects corresponding to different adherence categories: undetectable, less than 2 pills per week, 2–3 pills per week and greater or equal to 4 pills per week. We then utilised these HIV incidence risk reduction estimates and the observed HIV incidence rate in the PrEPVacc active-controlled TDF/FTC arm to infer the counterfactual placebo HIV incidence rate (i.e., the incidence rate in the absence of PrEP adherence). The counterfactual placebo incidence was calculated as: Observed incidence in the TDF/FTC arm/(1 - estimated risk reduction as per the literature). The confidence interval (CI) for the counterfactual placebo incidence was then obtained based on the CI of the observed incidence in the TDF/ FTC arm, (treating the risk reduction as fixed). Finally, we assessed the potential biases of each of the counterfactual placebo incidence rate estimation approaches. Results Participant characteristics Participants (n = 1380) in the PrEPVacc PrEP component (week 0 to week 26) were predominantly from the Mbeya (31%), Masaka (30%) and Verulam (26%) study sites. Approximately 13% were from the Dar es Salaam site. The cohort was 87% female, with 53% aged ≥ 25 years. Additionally, 41% were single, 60% were sex workers, 76% reported engaging in transactional condomless sex in the last 3 months, and 26% were diagnosed with an STI at baseline (Table 1 ). Table 1 Participant characteristics in the PrEP and post-PrEP components of the PrEPVacc trial, along with HIV incidence and covariates in the post-PrEP phase Factor PrEP component (Up to week 26)* Post-PrEP component (Beyond week 26)* Level N (%) Person-years(%) §§ N (%) Person-years(%) § HIV infections IR/100 PY (95 CI) aIRR /100 PY (95 CI) ¥ P-values 1380 (100) 709.2 (100) 1297 (100) 1648.4 26 1.6 (1.1–2.3) Site Dar es salaam 178 (13) 92.4 (13) 171 (13) 236.9 (14) 3 1.3 (0.4–3.9) 0.42 (0.05–3.79) 0.165 Masaka 412 (30) 214.0 (30) 398 (31) 586.7 (36) 5 0.9 (0.4–2.0) Ref Mbeya 433 (31) 212.6 (30) 385 (30) 413.3 (25) 10 2.4 (1.3–4.5) 2.29 (0.73–7.17) Verulam 357 (26) 190.2 (27) 343 (26) 411.5 (25) 8 1.9 (1.0–3.9) 2.70 (0.69–10.53) Calendar year** Dec 2020/2021 415 (30) 123.2 (17) 92 (7) 22.5 (1) 0 0 N/A 0.810 2022 1222 (89) 451.4 (64) 803 (62) 376.4 (23) 5 1.3 (0.6–3.2) Ref 2023 517 (37) 134.6 (19) 1270 (98) 1061.8 (64) 18 1.7 (1.1–2.7) 1.61(0.50–5.21) 2024 0 (0) 0 (0) 1029 (79) 187.7 (11) 3 1.6 (0.5–5.0) 1.01 (0.15–6.94) Time in follow-up since trial enrolment (years) ** § First 1380 (100) 709.2 (99) 1297 (100) 593.8 (36) 10 1.7 (0.9–3.1) Ref 0.983 Second 0 (0) 0 (0) 1207 (93) 877.1 (53) 14 1.6 (0.9–2.7) 0.94 (0.37–2.36) ≥Third 0 (0) 0 (0) 459 (35) 177.5 (11) 2 1.1 (0.3–4.5) 0.86 (0.16–4.75) Sex at birth Male 180 (13) 95.3 (13) 175 (14) 288.1 (17) 1 0.3 (0.0–2.5) Ref 0.099 Female 1200 (87) 613.9 (87) 1122 (87) 1360.3 (83) 25 1.8 (1.2–2.7) 5.64 (0.72–44.09) Categorised age min-24 647 (47) 326.2 (46) 593 (46) 750.0 (45) 14 1.9 (1.1–3.2) Ref 0.741 25 or higher 733 (53) 383.0 (54) 704 (54) 898.4 (55) 12 1.3 (0.8–2.4) 0.87 (0.38–2.00) Marital Status Single 564 (41) 282.7 (40) 515 (40) 672.0 (41) 15 2.2 (1.3–3.7) Ref 0.154 Married/cohabiting/in a relationship 547 (40) 287.9 (41) 527 (41) 646.0 (39) 10 1.5 (0.8–2.9) 0.88 (0.27–2.93) Divorced/separated/widowed 269 (19) 138.6 (20) 255 (20) 330.4 (20) 1 0.3 (0.0–2.1) 0.13 (0.02–1.03) Education Primary or lower 544 (39) 276.5 (39) 508 (39) 685.4 (42) 8 1.2 (0.6–2.3) Ref 0.975 Secondary or higher 836 (61) 432.7 (61) 789(61) 963.0 (58) 18 1.9 (1.2–3.0) 0.98 (0.35–2.79) Religion Christian 1055 (76) 540.1 (76) 986 (76) 1215.2 (74) 22 1.8 (1.2–2.7) Ref 0.306 Muslim/ other 325 (24) 169.1 (24) 311 (24) 433.2 (26) 4 0.9 (0.3–2.5) 0.43 (0.08–2.18) Categorised occupation Sex worker 826 (60) 418.6 (59) 766 (59) 920.6 (56) 17 1.8 (1.1–3.0) Ref 0.752 Salon/bar/lodge worker (not sex worker) 75 (5) 37.5 (5) 70 (5) 87.2 (5) 0 N/A N/A Subsistence fisheries 17 (1) 9.1 (1) 17 (1) 38.9 (2) 0 N/A N/A Unemployed 277 (20) 145.8 (21) 267 (21) 320.0 (19) 8 2.5 (1.3–5.0) 3.33 (0.19–58.47) Other 185 (13) 98.2 (14) 177 (14) 281.6 (17) 1 0.4 (0.1–2.5) 0.47 (0.04–5.89) Number in household 5 or lower 985 (71) 504.7 (71) 922 (71) 1185.6 (72) 21 1.8 (1.2–2.7) Ref 0.070 6 or higher 395 (29) 204.5 (29) 375 (29) 462.8 (28) 5 1.1 (0.5–2.6) 0.34 (0.10–1.09) Number of Rooms 3 or lower 1080 (78) 553.4 (78) 1013 (78) 1301.2 (79) 23 1.8 (1.2–2.7) Ref 0.155 4 or higher 300 (22) 155.8 (22) 284 (22) 347.2 (21) 3 0.9 (0.3–2.7) 0.31 (0.06–1.56) Number of partners last 3m 5 or less 522 (38) 276.1 (39) 500 (39) 649.0 (39) 11 1. 7 (0.9–3.1) Ref 0.679 6 or higher 858 (62) 433.1 (61) 797 (61) 999.4 (61) 15 1.5 (0.9–2.5) 0.76 (0.21–2.75) Days since last condomless sex 2 or less 645 (47) 330.8 (47) 612 (47) 770.7 (47) 13 1.7 (1.0–2.9) Ref 0.664 3–6 380 (28) 194.2 (27) 350 (27) 450.5 (27) 5 1.1 (0.5–2.7) 0.79 (0.27–2.33) 7 or higher 355 (26) 184.2 (26) 335 (26) 427.2 (26) 8 1.9 (0.9–3.7) 1.33 (0.49–3.61) Transactional condomless sex last 3m No 339 (25) 181.0 (26) 328 (25) 425.2 (26) 6 1.4 (0.6–3.1) Ref 0.202 Yes 1041 (75) 528.2 (75) 969 (75) 1223.2 (74) 20 1.6 (1.1–2.5) 2.65 (0.59–11.9) Sex after recreational drugs last 3m No 1210 (88) 620.4 (87) 1136 (88) 1432.5 (87) 24 1.7 (1.1–2.5) Ref 0.936 Yes 170 (12) 88.8 (13) 161 (12) 215.9 (13) 2 0.9 (0.2–3.7) 0.94 (0.21–4.18) Sex while drunk last 3m No 518 (38) 264.0 (37) 487 (38) 621.6 (38) 9 1.4 (0.8–2.8) Ref 0.795 Yes 862 (62) 445.2 (63) 810 (63) 1026.8 (62) 17 1.7 (1.0–2.7) 0.89 (0.38–2.08) Syphilis No 1327 (96) 681.9.3 (96) 1248 (96) 1578.7 (96) 25 1.6 (1.1–2.4) Ref 0.969 Yes 52 (4) 26.9 (4) 48 (4) 69.2 (4) 1 1.4 (0.2–10.3) 1.04 (0.14–7.82) Gonorrhoea No 1195 (92) 619.0 (87) 1127 (92) 1467.9 (89) 24 1.6 (1.1–2.4) Ref Yes 108 (8) 55.4 (8) 101 (8) 125.2 (8) 0 N/A N/A Chlamydia No 1049 (81) 543.7 (77) 987 (80) 1280.5 (78) 16 1.2 (0.8–2.0) Ref 0.155 Yes 254 (20) 130.7 (18) 241 (20) 312.5 (19) 8 2.6 (1.3–5.1) 1.88 (0.83–4.80) Any sexually transmitted infection No 1028 (74) 528.4 (75) 966 (75) 1217.5 (74) 17 1.4 (0.9–2.2) Ref 0.350 Yes 352 (26) 180.8 (25) 331 (25) 430.9 (26) 9 2.1 (1.1–4.0) 1.49 (0.65–3.42) IR- Incidence rate; aIRR- Adjusted incidence rate ratio; Ref- Reference *Week 26 implies Visit 9. Participants were allowed to attend visit 9 (target week 26) up until week 63. Study PrEP was dispensed until visit 9. **A participant could be included in more than one category for the variables time in follow-up since trial enrolment (years) and calendar period . For example, they could contribute person-years to both the first and second years of follow-up or to multiple calendar years (e.g., 2020 and 2021). Hence, the total percentages for these variables exceed 100%. § During the first year of follow-up, there was substantial continued PrEP use beyond visit 9. §§ Percentages are proportions of person-years ¥ Multivariable analyses were adjusted for vaccine arm, PrEP arm, study site, gender, age categories, and other variables with a P-value < 0.2 (marital status, number in household and chlamydia diagnosis at baseline). Other variables were returned one at a time and their multivariable aIRR and p-values estimated. N/A: Not applicable. The estimated counterfactual placebo HIV incidence rates Results from the approach utilising the pre-trial registration cohort data have been previously published [ 21 ]. When calendar period effects were incorporated in the prediction model, counterfactual placebo HIV incidence rates ranged from 1.2 to 1.9 per 100 person-years. In contrast, analyses without the calendar period effect predicted a counterfactual placebo incidence rate of 3.1 (2.1–4.4) per 100 person-years. PrEP uptake in the registration cohort was low, with 8% of participants ever starting PrEP. Sensitivity analyses excluding participants who had ever started PrEP predicted similar counterfactual placebo HIV incidence rates as those including all participants [ 21 ]. Tables 1 – 3 summarise results relevant to the estimation of the counterfactual placebo HIV incidence rate. Table 1 compares participants contributing person-years up to week 26 with those contributing person-years beyond week 26 in the PrEPVacc trial. It also presents the observed HIV incidence rate beyond week 26 and the effects of various covariates on HIV incidence. Table 2 presents analyses that utilise the adherence-efficacy relationship of TDF/FTC to predict the counterfactual placebo HIV incidence. Table 3 and Fig. 3 present the varying estimates of the counterfactual placebo HIV incidence rate from the different approaches under varying assumptions. Table 2 ) Utilising the adherence-efficacy relationship of TDF/FTC to predict the counterfactual placebo HIV incidence Estimated adherence in the TDF/FTC arm based on dried blood spot results (PrEPVacc) ¥ Estimated HIV incidence reduction (%) § by adherence category (BLQ; <2 ; 2–3 ; ≥4) (%) Observed HIV incidence rate (per 100 person-years) in the PrEP component of the PrEPVacc trial (TDF/FTC arm); person-years; events Estimated counterfactual placebo HIV incidence (per 100 person-years) (In reference to: IPREX; HPTN083; HPTN 084; 85% protection) (% of participants by pill adherence category: BLQ; <2; 2–3; ≥4 pills per week) Ref: IPREX* (0%; 0%; 76%; 96%) Ref: HPTN083* (0%; 49%; 99%; 99%) Ref: HPTN084* (0%; 16%; 80%; 88%) Assuming 85% protection if TFV-DP is detectable** Males 8; 45; 18; 29 42% 69% 47% 78% N/A; 48.8; 0 N/A Females 27; 56; 13; 4 14% 44% 23% 62% 0.65; 310.8; 2 0.8; 1.2; 0.8; 1.7 Overall 24; 54; 14; 8 18% 48% 27% 65% 0.56; 359.6; 2 0.7; 1.1; 0.8; 1.6*** ¥ Adherence (TFV-DP) measured in a random sample (n = 290; 41%) at week 8. BLQ: Below limit of quantification; Ref: Reference *IPREX was a placebo-controlled trial among males at high risk of HIV [ 24 ]. HPTN083 and HPTN084 were active-controlled trials conducted among males at high risk of HIV and females at high risk of HIV in SSA, respectively. Estimates of risk reduction attributed to TDF/FTC were inferred by comparing HIV incidence between higher and lower adherence categories. As a result, these estimates are susceptible to confounding effects in the reference trials. [ 17 ] **A systematic review of placebo controlled TDF/FTC trials concluded that participants with detectable levels of drug had 85% protection[ 18 ]. §Assuming no confounding factors for the association between HIV incidence and PrEP adherence. ***Confidence intervals are presented in Table 3 . Table 3 ) Sensitivity analyses: Varying estimates of the counterfactual placebo HIV incidence under different assumptions Alternative assumptions Expected incidence rate (per 100 pY) in the PrEP trial by site and overall; expected cases Comments Using pre-trial registration cohort data Dar es Salaam Masaka Verulam Mbeya Overall 1. Calendar period effect assumed to be site-specific 0.7; 0.6 1.4; 3.0 3.4; 6.4 0.5; 1.1 1.6 (0.6–4.2); 11.3 - The assumptions made about the calendar period effect substantially influence the counterfactual incidence estimate. 2. Calendar period effect assumed to be region specific (Regions: East Africa, South Africa) 0.5; 0.5 0.8; 1.7 3.3; 6.2 0.9; 1.9 1.5 (0.5–3.1); 10.6 3. Calendar period effect assumed to be uniform across sites 0.8; 0.7 1.1; 2.3 1.6; 3.0 1.3; 2.8 1.2 (0.4–2.6 ); 8.4 4. Registration cohort data limited to 2020 onwards and calendar period effect assumed site-specific 1.4; 1.3 3.5; 7.4 1.6; 3.0 0.9; 1.9 1.9 (0.6–7.9); 13.4 Using data from the post-PrEP component of the PrEPVacc trial - The observed lower incidence rate in 2022 (Table 1 ) results in lower estimates if a calendar effect (ie change in HIV incidence for a 1-unit increase in calendar year) is included in the prediction model. Of note, there was weak evidence (P = 0.810) of a calendar effect. - Vaccine effect assumed constant across the PrEPVacc trial (week 0 to week 74 or trial end). 5. Overall unadjusted estimate 1.3; 1.2 0.9; 1.9 2.0; 3.8 2.5; 5.3 1.6 (1.1–2.4); 11.3 6. Estimates generated using a multi-variable model adjusted for vaccine effect, PrEP arm, study site, sex, age, marital status, household size, and baseline chlamydia diagnosis 0.5; 0.5 1.0; 2.1 2.0; 3.8 2.7; 5.8 1.7 (1.0–2.4); 12.0 7. Overall unadjusted estimate, with adjustment for a calendar effect 0.3; 0.3 0.6; 1.3 1.4; 2.6 1.8; 3.8 1.2 (0.6–1.9); 8.5 8. Overall unadjusted incidence rate after the most recent date PrEP was taken. 1.3; 1.2 0.8; 1.7 2.0; 3.8 2.4; 5.1 1.6 (1.0–2.2); 11.3 Using the adherence-efficacy relationship of TDF/FTC - The risk reduction assumed for a given level of adherence substantially influences the counterfactual placebo incidence estimate. - The reference trials (IPREX, HPTN083, HPTN084) provided widely varying estimates of risk reduction when adherence was < 2 pills/week. - The uncertainty (reflected in the confidence bounds) of the risk reduction estimates from the previous trials and the random error in the adherence distribution in the PrEPVacc trial (Supplementary table 1 ) were not incorporated in these analyses. 9. Unadjusted, in reference to IPREX N/A N/A N/A N/A 0.7 (0.2–2.8); 4.9 10. Unadjusted, in reference to HPTN083 N/A N/A N/A N/A 1.1 (0.3–4.4); 7.8 11. Unadjusted, in reference to HPTN084 N/A N/A N/A N/A 0.8 (0.2–3.2); 5.6 12. Assuming 85% protection for all with detectable drug N/A N/A N/A N/A 1.6 (0.4–6.6); 11.3 The approach using the PrEPVacc trial HIV incidence data beyond week 26 (Table 1 ) provided counterfactual placebo HIV incidence point estimates ranging from 1.2 to 1.7 per 100 person-years (Table 3 ). PrEPVacc trial participants contributing person-years up to week 26 were generally comparable to those who went on to contribute person-yearsbeyond week 26 in terms of study site, gender, age, occupation and, HIV risk indicators with differences in proportions of no more than 7% (Table 1 ). The main difference was in the calendar period, with 81% of person-years up to week 26 occuring 2022 or earlier, while 75% of person-years beyond week 26 were from 2023 onwards. However, there was little evidence of a calendar period effect on HIV incidence in the data beyond week 26 (P = 0.810). Using the adherence-efficacy relationship of TDF/FTC provided counterfactual placebo HIV incidence point estimates ranging from 0.7 to 1.6 per 100 person-years (Table 2 ). Among 290 participants in the TDF/FTC arm who were randomly sampled and assessed at week 8 (41% of participants in the arm), 76% had detectable levels of Tenofovir-Diphosphate (TFV-DP). Of these, 54% had TFV-DP levels equivalent to < 2 tablets per week, 14% had levels equivalent to 2 to 3 tablets per week, and 8% had levels equivalent to 4 or more tablets per week (Table 2 ; Fig. 2 ; Supplementary table 1 ). Risk reduction estimates based on adherence data at week 8 ranged from 18% to 65% (Table 2 ). These estimates were then applied to the observed HIV incidence in the TDF/FTC arm (two infections, 0.56 per 100 person-years (95% CI: 0.14–2.22)) to generate the counterfactual placebo incidence estimates (Table 2 ). TFV-DP levels from dried blood spots were reassessed at week 26 in a small subset of participants (n = 39) who had been previously assessed at week 8. Of those reassessed, 51% had detectable TFV-DP levels of drug; 38% had TFV-DP levels equivalent to < 2 tablets per week; 13% had levels equivalent to 2 to 3 tablets per week, and none had TFV-DP levels equivalent to 4 or more tablets per week (Supplementary table 1 ). Detailed analyses of adherence in the PrEPVacc trial will be published separately. Approximately, 88% of participants in the PrEP component reported taking PrEP before at least 3 of 5 assessed sex acts, and 89% had a medicine possession ratio ≥ 75% at week 26. Adherence as per TFV-DP levels in DBS at week 8 varied by socio-demographic and behavioural variables including gender, study site (Fig. 2 ), age, marital status, and occupation. The potential biases of the various counterfactual placebo incidence estimation approaches have been summarised in Table 4 . Table 4 Potential biases of the different counterfactual placebo incidence estimation approaches Counterfactual placebo incidence estimation approach Potential biases/limitations Using pre-trial registration cohort data - Model choice for prediction could introduce error. - Due to the registration cohort largely occurring before the trial, extrapolations needed to be made utilizing observed calendar trends, which introduced uncertainty. - COVID-19 disruptions in the registration cohort introduced uncertainty regarding the reliability of observed calendar trends. - Unmeasured or uncontrolled differences between participants in the registration cohort and trial could introduce bias. - Trial participants may behave differently than cohort participants. - The start of follow-up for person-years from the registration cohort to include in the model was unclear. Specifically, whether to include all person-years or only the more recent person-years. If restricted to recent person-years, the appropriate cut-off remained uncertain. - The approach does not consider the possibility of a vaccine effect during the PrEP component of the PrEPVacc trial. It predicts counterfactual placebo HIV incidence based on the assumption that all participants were in the vaccine placebo arm, yet PrEP trial participants were distributed across both active and placebo vaccine arms Using PrEPVacc trial data beyond week 26 - We assumed constancy of the vaccine effect across the trial, yet it is possible that the vaccine effect could have varied over time and multiple injections. - The small number of person-years limited multi-variable modelling, however participants are broadly expected to be similar in the two datasets (up to week 26 Vs beyond week 26), if loss to follow up (LTFU) is low. - It is unclear how best to adjust for continued PrEP use beyond week 26 which could result in underestimation of the counterfactual placebo incidence estimate. - A decline in incidence over time in follow-up could result in underestimating the counterfactual placebo incidence. Using the adherence-efficacy relationship of TDF/FTC - The estimates of expected risk reduction at a given level of adherence are imprecise [ 17 , 20 ], and it is unclear how to incorporate the uncertainty (reflected in the confidence bounds) of the risk reduction estimates in the counterfactual placebo HIV incidence rates. - The populations in the reference trials (IPREX, HPTN083, HPTN084) differed considerably from ours and the expected risk reduction at a given level of adherence from these trials may not be generalizable to our study population. - Dried blood spots were only tested for a random sample (~ 41%) of participants at week 8 and in a subset (n = 39) of them at week 26. It is questionable whether the data are representative of overall adherence. Also, it is unclear how other adherence data (such as: self-reported adherence around condomless sex acts and urine tenofovir tests) can be incorporated in these analyses. - The random error in the adherence distribution in the PrEPVacc trial was not incorporated in these analyses. - Due to the limited number of events in the TDF/FTC arm of the PrEP component of the PrEPVacc trial, minimal or no confounding is assumed in the association between adherence to TDF/FTC and HIV incidence, yet the validity of this assumption is questionable. - With few events observed, the estimate of the HIV incidence rate in the TDF/FTC arm is imprecise and consequently the estimate of the counterfactual placebo HIV incidence rate from the approach is also imprecise. - Counterfactual placebo incidence could not be estimated at site level for sites with zero observed infections during the PrEP phase. * Counterfactual placebo incidence - What would the incidence rate have been had participants in the active-controlled PrEP trial not received any PrEP? Discussion The point estimates of the counterfactual placebo HIV incidence rates ranged from 0.7 to 1.9 per 100 person-years, lower than the observed incidence rate in the pre-trial registration cohort of 2.9 per 100 person-years [ 16 ]. The results of the PrEP component of the PrEPVacc trial, based on the averted infections ratio (AIR), which incorporates an estimate of the counterfactual placebo HIV incidence rate, will be published separately. Each approach used to estimate the counterfactual placebo HIV incidence rate had several potential biases or limitations, which we discuss below. The pre-trial registration cohort data comprised the largest dataset, with over 3600 person-years of follow-up [ 16 ], which enabled multi-variable modelling to adjust for demographic, behavioural risk and calendar period differences between the registration cohort and trial participants. However, this approach had several limitations. These included potential bias introduced by model choice, the need to extrapolate based on observed calendar trends due to differences in calendar periods between the registration cohort and the trial, which was complicated by COVID-19 disruptions [ 16 ], the possibility of unmeasured or uncontrolled differences between participants (and their risk behaviours) in the registration cohort and trial which could introduce bias, and the fact that this approach did not account for any vaccine effects present in the trial. Other limitations of the registration cohort approach have been described elsewhere [ 16 ]. Using PrEPVacc trial data beyond week 26 was a promising approach, as participants in the PrEP component (up to week 26) were more similar to, and had a closer temporal overalp in follow-up with, those contributing data beyond week 26 compared to participants in the registration cohort. However, the limited number of person-years available restricted the ability to perform robust multi-variable modelling, which could have further adjusted for differences in participant characteristics between those followed up to week 26 and those beyond 26. The key limitation of this approach was the uncertainty about whether the observed vaccine effect after the PrEP component of the PrEPVacc trial [ 22 ] was similar to the vaccine effect during the PrEP phase. Prior to the conduct of the trial, a protective vaccine effect was not anticipated until after week 26, when participants were expected to reach peak immunogenicity. However, in the PrEPVacc vaccine trial, more infections were observed in the active trial vaccine arms than in the placebo [ 22 ]. It is possible that the vaccine effect could have varied over time as participants received multiple injections. In this paper, we assumed the vaccine effect remained constant throughout the trial (from week 0 to week 74, or the trial’s end), because investigating the vaccine effect over time was beyond the scope of this paper. This assumption will be explored in more detail in the PrEPVacc main trial vaccine efficacy paper. Additional concerns with this approach included uncertainty regarding how best to adjust for continued PrEP use beyond week 26, and the potential decline in HIV incidence over time during follow-up, both of which could result in an underestimation of the counterfactual placebo HIV incidence rate. Using the adherence-efficacy relationship of TDF/FTC has the advantage of referencing empirical risk reduction estimates from the literature; however, the expected risk reduction estimates at various adherence levels, especially for lower adherence categories (< 2 pills per week), are imprecise. The estimates utilised in our study were based on data from the IPREX, HPTN083 and HPTN084 trials [ 17 , 23 , 24 ]. However, the study populations in these trials differed considerably from ours, raising concerns about the generalisability of the risk reduction estimates to our study population. The risk reduction estimates from HPTN083 and HPTN084 were inferred by contrasting incidence in higher adherence and lower adherence categories. This implies that the estimates may be susceptible to confounding effects in the reference trials. Other key limitations of this approach included the inability to adjust for confounders of the association between adherence to TDF/FTC and HIV incidence, due to the limited number of observed infections (two) in the TDF/FTC arm of our trial. This is particularly important as adherence varied considerably across socio-demographic and behavioural variables. Similarly, due to the small number of events, the estimate of the HIV incidence rate in the TDF/FTC arm is imprecise, which in turn leads to an imprecise estimate of the counterfactual placebo HIV incidence rate from this approach. For simplicity, we did not incorporate the uncertainty (reflected in the confidence bounds) of the risk reduction estimates from previous trials, and the random error in the adherence distribution in the PrEPVacc trial (Supplementary Table 1). Incorporating both would have further widened the confidence intervals of the counterfactual placebo HIV incidence estimates. Additionally, dried blood spots were tested only for a random sample (~ 41%) of participants at week 8, this is a higher proportion than previous PrEP trials, however it is unclear how representative these results were of overall adherence across the PrEPVacc trial. Week 26 results from a sub-set of participants showed a decline in adherence over time. Also, other adherence data, such as self-reported adherence to event-driven use and white coat dosing, were not incorporated, yet could potentially have impacted HIV incidence estimates. The analyses presented in this paper had other limitations. Firstly, the impact of COVID-19 disruptions on counterfactual placebo HIV incidence estimation was not anticipated during trial planning. These disruptions had a substantial impact, potentially influencing the accurancy of the pre-trial registration cohort approach. It is possible that this approach would have performed better in the absence of COVID-19-related disruptions. Secondly, there is currently no standard approach to estimating a counterfactual placebo HIV incidence rate, despite increasing recognition of its importance. This is a key challenge that needs to be addressed to ensure objectivity and replicability of results in future HIV prevention trials, when decisions need to be made about model choice, accounting for calendar period effects, defining criteria for selection of external controls, determining appropriate dataset cut-offs (especially when external control datasets include data from much earlier calendar periods), and estimating risk reduction at varying levels of adherence, among other factors. This demonstrates the novelty of our analyses. To our knowledge, this is only the second HIV PrEP effectiveness trial to incorporate a counterfactual placebo incidence rate in its primary analyses. The PURPOSE trials [ 25 ] were the first, while earlier trials incorporated counterfactual placebo incidence estimates only in secondary analyses, that did not influence the primary trial conclusions [ 26 – 28 ]. In conclusion, estimating the counterfactual placebo incidence rate for the PrEPVacc trial was challenging. Further development and standardisation of approaches for counterfactual placebo HIV incidence estimation could help mitigate these challenges in future trials. Abbreviations aIRR Adjusted incidence rate ratio AIR Averted infections ratio AIDS Acquired immunodeficiency syndrome CI Confidence interval COVID-19 Coronavirus disease DBS Dried blood spots HIV Human immunodeficiency virus LSHTM London School of Hygiene and Tropical Medicine MRC/UVRI Medical Research Council / Uganda Virus Research Institute MPR Medicine possession ratio PrEP Pre-exposure prophylaxis PrEPVacc A Phase IIb three-arm, two-stage HIV prophylactic vaccine trial with a second randomisation to compare TAF/FTC to TDF/FTC as pre-exposure prophylaxis SSA Sub-Saharan Africa STI Sexually transmitted infections TAF/FTC Emtricitabine/tenofovir alafenamide (TAF/FTC) TDF/FTC Emtricitabine and tenofovir disoproxil fumarate (TDF/FTC) TX Texas UK United kindgdom Declarations Ethical approval Ethical approval for the PrEPVacc trial was obtained from national research ethics committees at all the study sites (including: The Uganda National Council for Science and Technology (ref: 2717), the National Institute for Medical Research (Tanzania) (ref: R.8a/ VolIX/3333) and the South African Medical Research Council (ref: EC020-11/2019). This study also obtained approval from the London School of Hygiene and Tropical Medicine ethics review committee (ref: 21317) and the UVRI Research Ethics Committee (GC/127/18/03/637). Informed consent Informed consent to participate and for publication, was obtained from all the study participants. Availability of data and materials The datasets generated and analyzed during the current study are available in the LSHTM repository, [PERSISTENT WEB LINK TO DATASETS TO BE ADDED] Competing interests The authors declare that they have no competing interests. Funding PrEPVacc is funded by The Second European & Developing Countries Clinical Trials Partnership (EDCTP2); (Grant reference: RIA2016 - 1644). RH received funding from the UK Medical Research Council (MRC) and the UK Foreign, Common wealth and Development Office (FCDO) under the MRC/FCDO Concordat agreement and is also part of the EDCTP2 programme supported by the European Union. Grant Ref: MR/R010161/1. SMc and DD received funding from the UK Medical Research Council.   Grant Ref: MC_UU_00004/03. Authors' contributions Contributed to the conceptualization and design of the study: SK, CH, ER, HB, JF, SMc, DD, RH Acquisition of data and study conduct: SK, CH, ER, HB, JF, SMc, DD Contributed to the analysis and interpretation of the data: SK, CH, ER, HB, JF, SMc, DD, RH Writing, reviewing and editing: SK, CH, ER, HB, JF, SMc, DD, RH. All authors have read and approved the final manuscript. Acknowledgements We thank all the participants and staff involved in the PrEPVacc study: South African Medical Research Council, Durban, South Africa: Glenda Gray, Nishanta Singh, Zakir Gaffoor, Neetha Morar, Thandiwe Sithole, Kubashni Woeber, Samantha Siva, Eldi-nah Hwengwere; Africa Health Research Institute, Durban, South Africa: Rujeko Samanthia Chidawanyika, Nteboheleng Mahapa, Phindile Khanyile, Janet Seeley; National Institute for Medical Research-Mbeya Medical Research Centre, Mbeya, Tanzania; Lucas Maganga, Wiston William, Emmanuel Kapesa, Elizabeth Danstan,  Doreen Pamba, Marco Missanga Amani Kway, Abisai Kisinda, Lilian Njovu, Lwitiho Sudi, Revocatus Kunambi, Jane Ambindwile; Muhimbili University of Health and Allied Sciences, Dar es Salaam, Tanzania: Said Aboud, Patricia Munseri, Eligius Lyamuya, Frank Msafiri, Agricola Joachim, Edith Tarimo, Diana Faini Tumaini Nagu, Deus Buma, Muhammad Bakari; MRC/UVRI and LSHTM Uganda Research Unit, Entebbe, Uganda: Pontiano Kaleebu, Freddie Mukasa Kibengo, Ayoub Kakande, Jennifer Serwanga, Janet Seeley, Rachael Kawuma, Sylvia Kusemererwa, Anita Kabarambi, Martin Onyango, Jonathan Kitonsa, Sylvia Masawi, Vincent Basajja, Tobias Vudriko, Peter Hughes, Florence Nambaziira, Shamim Nabukenya, Gertrude Mutonyi, Rita Nakiboneka, Susan Mugaba, Sarah Nakamanya, Denis Kibuuka, Bernadette Kalanzi; Instituto Nacional de Saúde, Maputo, Mozambique: Ilesh Jani, Edna Viegas , Isabel Remane, Odete Bule, Edna Nhacule, Patricia Ramgi, Raquel Chissumba, Eduardo Namalango, Yolanda Manganhe , Carmelia Massingue, Igor Capitine, Jorge Ribeiro; Imperial College London, UK: Jonathan Weber, Cherry Kingsley, Thomas E. Miller, Sarah Joseph, Claire Wenden; Eurovacc Foundation, Lausanne, Switzerland: Giuseppe Pantaleo, Song Ding; Gilead, Foster City, CA, USA: James S. Rooney, Alex Kintu, Iera Monasterio, Lisette Bassi; Ludwig-Maximilians-University of Munich, Munich, Germany: Arne Kroidl; The Military HIV Research Program, Bethesda, MD, USA: Merlin Robb; Karolinska Institutet, Stockholm, Sweden: Charlotta Nilsson; Global Solutions for Infectious Diseases, San Francisco, CA, USA: Carter Lee, Vineeta Gulati; CONRAD, Arlington, VA, USA: Gustavo Doncel, Jessica Maitz, Annie Thurman; MRC Clinical Trials Unit at UCL, London, UK: Sheena McCormack, Simona Salomone, Angela Crook, David Dunn, Henry Bern, Aminata Sy, Liz Brod-nicki, Julie Fox; International AIDS Vaccine Initiative, New York, USA: Kundai Chinyenze, Jacqueline Musau, Mabela Matsoso, Mary Amondi, Ansuya Naidoo, Paramesh Chetty, Anne Gumbe. For the PrEPVacc trial, AIDSVAX® B/E vaccine is provided by Global Solutions for Infectious Diseases (GSID), USA. TDF/ FTC and TAF/ FTC are provided free of charge for participants in the trial by Gilead Inc. DNA-HIV-PT123 vaccine is provided by the EuroVacc Foundation. CN54gp140 formulated with MPLA adjuvant is provided by Imperial College London, UK. MVA-CMDR is provided by the Walter Reed Army Institute of Research (WRAIR), USA. This study is also made possible by the support of the American People through the U.S. President’s Emergency Plan for AIDS Relief (PEPFAR) through United States Agency for International Development (USAID). The contents of this publication are the sole responsibility of the PrEPVacc Team and do not necessarily reflect the views of PEPFAR, USAID, or the United States Government. References WHO. Updated WHO guidance on pre-exposure prophylaxis (PrEP) . 2023. Cutrell A, Donnell D, Dunn DT, et al. HIV prevention trial design in an era of effective pre-exposure prophylaxis. HIV Clin Trials. 2017;18(5–6):177–88. Donnell D, et al. Study design approaches for future active-controlled HIV prevention trials. Stat Commun Infect Dis. 2024;15(1):20230002. Prudden HJ, et al. Perspectives on Design Approaches for HIV Prevention Efficacy Trials. AIDS Res Hum Retroviruses. 2023;40(5):301–7. Glidden DV, et al. Using the adherence-efficacy relationship of emtricitabine and tenofovir disoproxil fumarate to calculate background hiv incidence: a secondary analysis of a randomized, controlled trial. J Int AIDS Soc. 2021;24(5):e25744. Bekker L-G et al. Twice-Yearly Lenacapavir or Daily F/TAF for HIV Prevention in Cisgender Women. New England Journal of Medicine. 0(0). Subramaniam D, et al. A Framework for the Use and Likelihood of Regulatory Acceptance of Single-Arm Trials. Therapeutic Innovation & Regulatory Science; 2024. Weberpals J, Wang SV. External Controls to Study Treatment Effects in Rare Diseases: Challenges and Future Directions. n/a(n/a): Clinical Pharmacology & Therapeutics; 2024. Sola-Morales O, et al. Effectively Leveraging RWD for External Controls: A Systematic Literature Review of Regulatory and HTA Decisions. Volume 114. Clinical Pharmacology & Therapeutics; 2023. pp. 325–55. 2. Jahanshahi M, et al. The Use of External Controls in FDA Regulatory Decision Making. Ther Innov Regul Sci. 2021;55(5):1019–35. U.S. Department of Health and Human Services Food and, Administration D. https : //www.fda.gov/regulatory-information/search-fda-guidance-documents/considerations-design-and-conduct-externally-controlled-trials-drug-and-biological-products. 2024. Rippin G et al. External Comparator Cohort studies - clarification of terminology. Front Drug Saf Regul, 2024. 3. Dunn DT, et al. The averted infections ratio: a novel measure of effectiveness of experimental HIV pre-exposure prophylaxis agents. Lancet HIV. 2018;5(6):e329–34. MRC/UVRI and LSHTM Uganda Research Unit. A Combination Efficacy Study in Africa of Two DNA-MVA-Env Protein or DNA-Env Protein HIV-1 Vaccine Regimens With PrEP (PrEPVacc). 2019, Clinical Trials.gov. Joseph S, Kaleebu P, Ruzagira E, et al. OC 8491 PREPVACC: A PHASE III, MAMS ADAPTIVE PROPHYLACTIC HIV VACCINE TRIAL WITH A SECOND RANDOMISATION TO COMPARE F/TAF WITH TDF/FTC PREP. BMJ Global Health; 2019. Kansiime S et al. Challenges in estimating the counterfactual placebo HIV incidence rate from a registration cohort: The PrEPVacc trial. Clinical Trials, 2024: p. 17407745241304721. Anderson PL, Marzinke MA, Glidden DV. Updating the Adherence–Response for Oral Emtricitabine/Tenofovir Disoproxil Fumarate for Human Immunodeficiency Virus Pre-Exposure Prophylaxis Among Cisgender Women. Clin Infect Dis. 2023;76(10):1850–3. Fonner VA et al. Effectiveness and safety of oral HIV preexposure prophylaxis for all populations. AIDS, 2016. 30(12). Glidden DV, Das M, Dunn DT. Using the adherence-efficacy relationship of emtricitabine and tenofovir disoproxil fumarate to calculate background hiv incidence: a secondary analysis of a randomized, controlled trial. 2021. 24(5): p. e25744. Moore M, Stansfield S. Efficacy estimates of oral pre-exposure prophylaxis for HIV prevention in cisgender women with partial adherence. 2023. 29(11): pp. 2748–52. Kansiime S, et al. Developing HIV risk prediction tools in four African settings. Tropical Med Int Health. 2023;28(9):720–30. PrEPVacc. News Release: HIV vaccines tested in PrEPVacc fail to reduce infections. Editor: PrEPVacc; 2024. Anderson PL, Marzinke MA, Glidden DV. Updating the Adherence-Response for Oral Emtricitabine/Tenofovir Disoproxil Fumarate for Human Immunodeficiency Virus Pre-Exposure Prophylaxis Among Cisgender Women. Clin Infect Dis. 2023;76(10):1850–3. Anderson PL, et al. Emtricitabine-tenofovir concentrations and pre-exposure prophylaxis efficacy in men who have sex with men. Sci Transl Med. 2012;4(151):151ra125. Bekker L-G, et al. Twice-Yearly Lenacapavir or Daily F/TAF for HIV Prevention in Cisgender Women. N Engl J Med. 2024;391(13):1179–92. Glidden DV, Stirrup OT, Dunn DT. A Bayesian averted infection framework for PrEP trials with low numbers of HIV infections: application to the results of the DISCOVER trial. Lancet HIV. 2020;7(11):e791–6. Donnell D et al. Counterfactual estimation of efficacy against placebo for novel PrEP agents using external trial data: example of injectable cabotegravir and oral PrEP in women. 2023. 26(6): p. e26118. Zhu Y, et al. Estimating counterfactual placebo HIV incidence in HIV prevention trials without placebo arms based on markers of HIV exposure. Clin Trails. 2023;21(1):114–23. Supplementary Files EQUATORNetworkReportingChecklist.docx Supplementarytable1.docx Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Major revision 10 Aug, 2026 Reviewers agreed at journal 17 Mar, 2026 Reviewers invited by journal 17 Mar, 2026 Editor assigned by journal 06 Jul, 2025 First submitted to journal 31 Mar, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information

Related documents

• Evaluating adherence in an active-controlled HIV pre-exposure prophylaxis trial (PrEPVacc) to inform the estimation of HIV incidence in a counterfactual placebo arm#403281
• PrEP-LANA (Pre-exposure Prophylaxis- Long-acting Novel Antiviral Agent)#804000
• Complexity of Adherence Challenges: Understanding Syndemic Factors Affecting HIV Treatment Adherence During Treatment Initiation in Cape Town, South Africa.#182471
• Implementing Tobacco Use Treatment in HIV Clinics Vietnam#1102246
• A Prospective, Observational Study of Pregnancy Outcomes Among Women Exposed to Truvada for PrEP Indication#992362