It is well established that adherence to daily tenofovir disoproxil fumarate (TDF)—emtricitabine (FTC) for HIV pre-exposure prophylaxis (PrEP) is a powerful predictor of efficacy.1–5 It has also been shown that objective pharmacologic measures of adherence are critical to understand PrEP outcomes.6,7 With the recent approval of tenofovir alafenamide (TAF)/FTC as PrEP in the United States, similar objective pharmacologic adherence measures are needed to understand PrEP outcomes with TAF/FTC.
Various pharmacologic measures have been used to assess adherence to TDF/FTC across research and clinical settings.7 Each of these measures has advantages and disadvantages. Tenofovir (TFV) and emtricitabine (FTC) concentrations in plasma, urine, and saliva are pharmacologic measures that can be used to indicate recent drug ingestion (ie, adherence). These moieties have short half-lives (∼15 hours) and do not appreciably accumulate from first dose to steady state.8–12 In addition, urine and saliva concentrations exhibit high variability.10,13 These characteristics lead to the main disadvantage for these moieties which is the lack of information on adherence leading up to the most recent dose, and the related issue of “white coat” adherence, where patients/participants are consistently nonadherent but take a dose immediately before their study or clinic visit.14 The rate of “white coat” dosing varies by PrEP study from “not common” to up to 34%, indicating a need to better define this phenomenon.15–18 Advantages of urine, saliva, and plasma concentrations include their ease of collection and confirmation of recent drug ingestion and the potential application to point-of-care testing.19
Other pharmacologic moieties exhibit longer half-lives and provide information about averaged adherence over time, including intracellular tenofovir-diphosphate (TFV-DP) in peripheral blood mononuclear cells (PBMCs) and red blood cells (RBCs), as well as TFV in hair.7,14,20,21 In PBMC, TFV-DP has a half-life of approximately 3–5 days.12,22 Steady-state concentrations in PBMC are 5- to 8-fold higher than that of a single dose and reflect averaged dosing over the previous 1–2 weeks.11,12 However, specialized processing of PBMC is time consuming and not widely available, and results are dependent on consistent and accurate cell counts.7 For these reasons, TFV-DP in RBC, measured using dried blood spots (DBS), was developed, validated, and has since become widely used as an adherence biomarker.22 After TDF/FTC dosing, TFV-DP had a 17-day half-life in DBS, which translates to a 25-fold accumulation from first dose to steady state.22,23 This degree of accumulation provides a large dynamic concentration range that can differentiate gradients of adherence over the preceding 1–3 months.22,23 TFV in hair exhibits similar characteristics.20 The main limitation of these moieties is they provide a measure of the average adherence over the preceding 1–3 months, not patterns of dosing (ie, starting and stopping doses). However, for DBS, FTC-TP is measured simultaneously with TFV-DP, and it has a much shorter half-life, such that quantifiable FTC-TP reflects a recent dose much like plasma TFV.23 The use of these two measures in combination allows for interpretation of both cumulative and recent TDF/FTC dosing.
The DOT-DBS study established dose proportionality and adherence benchmarks for TFV-DP in DBS after TDF/FTC dosing. Rounded 25th percentiles of TFV-DP were used for adherence benchmarks representing 2, 4, and 7 TDF/FTC doses/wk.23 These benchmarks were 350, 700, and 1250 fmol/punch, and they have been used to assess PrEP adherence and outcomes across multiple clinical trials for TDF/FTC.24–27 A value of ≥700 fmol/punch (ie, ≥4 doses/wk on average) has been associated with high PrEP efficacy across multiple studies in MSM.7 This relationship is consistent with the IPERGAY trial which showed 86% efficacy for 3.75 doses/wk on average, and TFV-DP in PBMC from iPrEx, which suggested 90% efficacy for approximately 3 doses/wk on average.28,29
However, despite this proven predictive value, these adherence thresholds with TDF/FTC23 cannot be used to interpret cumulative TAF/FTC adherence because of the unique pharmacology of TAF. TAF is a novel prodrug of TFV, shown to exhibit more efficient loading of target cells (PBMC) relative to TDF, because of its specificity for cathepsin A. This results in ∼7-fold higher concentrations of TFV-DP in PBMC and 90% lower plasma TFV concentrations.30,31 Regarding RBCs and DBS, a pilot study showed that TFV-DP in DBS after TAF was approximately 1/7th of steady-state concentrations measured after TDF (224 fmol/punch versus 1560 fmol/punch).32 Thus, prospective research is needed to establish expected TFV-DP in DBS after TAF/FTC, which was the motivation for this study.
The goals of the TAF-DBS study were (1) to describe the pharmacokinetics of TFV-DP and FTC-TP in RBCs, measured in DBS, after 33%, 67%, and 100% of daily TAF/FTC dosing, (2) to assess dose proportionality of TFV-DP in DBS, and (3) to establish adherence interpretations based on steady-state TFV-DP and FTC-TP concentrations.
TAF-DBS was conducted at the University of Colorado Anschutz Medical Campus. The study was approved by the Colorado Multiple Institutional Review Board and registered on clinicaltrials.gov (NCT02962739). All participants provided written informed consent and were made aware of the benefits and risks of participating in the study.
Study participants were adults without HIV aged 18–59 years who were deemed able to comply with study procedures, including directly observed dosing. Because participants were given less than daily dosing, individuals at high risk of HIV acquisition were excluded. Other exclusion criteria included pregnant women, a positive HIV or Hepatitis B virus (HBV) screening, estimated glomerular filtration rate (eGFR) < 60 mL/minute/1.73 m2 (Modified Diet in Renal Disease equation) to minimize variability in renal function, urine protein ≥2+, history of nontraumatic bone fractures, medical conditions that alter RBC kinetics, or any conditions that, in the opinion of the investigators, might interfere with the study.
The study design was modeled after that of the DOT-DBS study with TDF/FTC.23 A 36-week prospective, randomized, crossover study was conducted. Participants were given 25 mg TAF in combination with 200 mg FTC, as Descovy (Gilead Sciences, Foster City, CA),33 and assigned to 2 different 12-week dosing regimens of either 33%, 67%, or 100% of daily dosing, separated by a 12-week washout period (see Figure 1, Supplemental Digital Content, https://links.lww.com/QAI/B455). The 33% arm was defined as dosing on day 1 followed by skipped doses on days 2 and 3, repeated for 12 weeks; the 67% arm was dosing on days 1 and 2, followed by a skipped dose on day 3, repeated for 12 weeks; 100% daily dosing was dosing every day for 12 weeks. All doses were directly observed by study personnel either in person, by live video streaming, or with a timestamped video. Doses could be taken at any time within the 24-hour dosing day. Study personnel recorded the date and time of each dose and whether the dose was taken within 2 hours of a meal.
Blood was collected in an EDTA tube for PBMC, DBS, and plasma analysis at baseline (predose) and approximately 4 hours after the first dose. Blood was subsequently collected once weekly during the dosing regimens and washout period without regard to time or day since the last dose (ie, convenience sampling). An additional DBS sample was collected through fingerstick at weeks 1 and 12. Urine and hair were collected at various intervals. Analysis of urine, hair, PBMC, and plasma concentrations is not discussed in this report. Complete blood count, complete metabolic profile, phosphorus, and lipase were collected by standard venipuncture at screening and every 4 weeks while on TAF/FTC. HBV surface antigen (HbsAg) was collected at screening, and HIV antibody was collected before starting and at the completion of each regimen.
For DBS, 25 µL of whole blood was spotted 5 times onto a Whatman 903 protein saver card. Fingerstick DBS were collected using a single use lancet (SurgiLance 2.8 mm, 21 G). The first drop of blood was wiped away using gauze, and subsequent blood drops were wicked onto the Whatman 903 card, 1 drop per collection spot. Cards were air-dried at room temperature for at least 3 hours (and up to overnight) and stored at −80°C until analysis.
Two 7-mm punches were taken from DBS for the quantification of TFV-DP and FTC-TP, which was performed using a previously validated methodology.34 A modification to this method was developed and validated for the extraction to account for the lower TFV-DP concentrations in DBS with TAF versus TDF. The 7-mm punches were extracted with 2 mL of methanol:water creating a lysed cellular matrix. This lysate was then assayed, which comprised the “sample.” For the comparisons of fingerstick versus paired venipuncture, smaller punches were used to accommodate the smaller blood spots available (one to two 3-mm punches or one 7-mm punch). We validated that the punch size/number of punches (ie, RBC number) in the “sample” gave the same normalized TFV-DP value. Results were then normalized to two 7-mm punches, if applicable. The assay was linear from 25 to 6000 fmol/sample for TFV-DP and 0.1–200 pmol/sample for FTC-TP.
All reported data are presented with means, medians, and ranges, unless otherwise specified.
Dose proportionality was assessed with a power model (on the natural log scale) based on observed TFV-DP in DBS at weeks 12 and 36 (after 12 weeks of dosing), ln(Yijk) = µ + Si + Pj + βlog(Dk) + εijk. Yijk is TFV-DP in DBS for the ith subject (i = 1, …35), jth period (j = 1 and 2), and kth dose (k = 33%, 67%, and 100%), µ is the overall mean, Si is a random subject effect, Pj is the period effect, and εijk is random error.35 Dose proportionality dictates that β equals 1 and was assumed if the 90% confidence interval (CI) was within 0.8 and 1.25. This was assessed for both the as treated (including only observed ingested doses of TAF/FTC) and per randomization (intent-to-treat) populations.
The effects of demographic variables were evaluated with all available concentration data (α = 0.05). The following variables were assessed as predictors of TFV-DP in DBS: sex, race (categorized as African American or Non-African American), age, hematocrit, weight, eGFR, and average doses within 2 hours of meals. A mixed effects model with tensor product of natural b-spline transformation of study day, dose proportion (calculated at each sampling time), and study arm was used to model logTFV-DP concentration as a smooth, nonlinear function of day, dose proportion, and study arm. Percent differences were calculated for paired fingerstick and pipetted samples, and the coefficient of determination was generated with linear regression.
Thirty-eight participants were enrolled in the study. Characteristics of 37 participants are listed in Table 1. One participant was noncompliant with study dosing and was not included. Thirty-five completed at least 1 dosing regimen, and 34 completed 2 regimens (1 was withdrawn at week 32 due to a traumatic leg fracture unrelated to study drug). One withdrew from the study before completion of their first regimen after a panic attack unrelated to the study, and one was removed from the study after week 9 because they were no longer following study procedures.
Observed concentrations of TFV-DP in DBS at steady state (weeks 12 and 36) among those receiving 33%, 67%, and 100% daily dosing are shown in Table 2. Figure 1 displays observed and fitted concentrations for each dosing frequency. The mean (95% CI) half-life for TFV-DP in DBS during the washout period was 20.8 (19.3 to 21.3) days. Median concentrations approached steady state at week 10, as indicated by less than 10% difference between weeks 10 and 12 at each dosing frequency. For a given participant, 95% of week 12 estimates were within of −12.7% and 21.0% of their averaged week 10 and 11 concentrations.
TFV-DP concentrations in DBS increased linearly with TAF/FTC dosing frequency. The estimated slope, β, (90% CI) was 1.14 (1.07 to 1.21) in the as treated group, in which only ingested doses were included. This was similar to the estimate when assessed by randomization arm (ie, intent-to-treat population): β = 1.13 (90% CI: 1.06 to 1.20). The relationship was also maintained after separate adjustment for age, race, sex, weight, eGFR, hematocrit, average doses near a meal, and treatment period, of which all had 90% CIs within 1.06 and 1.21.
The median (range) percentage of prescribed doses taken was 100 (89.4 to 104.8) %. One participant randomized to 33% dosing took an additional dose on a day they were not supposed to dose. Median (range) percentage of doses that were taken 2 hours before or after a meal was 94 (31 to 100) %. Univariate and multivariate analyses assessed differences by sex, race, age, weight, hematocrit, eGFR, and average doses near meals, all based on biological plausibility and previous observations for TDF. In univariate analyses, only race, age, and weight were significant predictors of TFV-DP in DBS. TFV-DP concentrations in DBS were 33% lower (95% CI −45.65% to −17.6%) among black participants, 2.1% higher (95% CI: 0.66% to 3.55%) with each year increase in age, and 1.1% lower (95% CI −1.52% to −0.58%) with each 1-kg increase in weight. However, after adjustment for sex, race, age, weight, hematocrit, eGFR, and doses taken near a meal, only weight was a significant predictor of TFV-DP in DBS, with a 0.9% decrease (95% CI −1.36% to −0.30%) in TFV-DP per 1-kg increase in weight (P = 0.003).
Fitted TFV-DP concentrations for 2, 4, and 7 doses/wk among men and women are shown in Table 3. The rounded 25th percentiles were 450, 950, and 1800 fmol/punches for 2, 4, and 7 doses/wk, respectively. These percentiles were used to construct adherence interpretations as follows: <450, <2 doses/wk; 450–949, 2–3 doses/wk; 950–1799, 4–6 doses/wk; and ≥1800, 7 doses week. These thresholds were consistent for men and women, as the 25th percentile concentrations for each sex were within 9% of the overall population.
Finally, the predictor and outcome were reversed to estimate the range of doses associated with TFV-DP concentrations. Figure 2 shows the estimated doses that are consistent with concentrations of 450, 950, and 1800 fmol/punches. For example, among those with a concentration of 450 fmol/punches, we estimate 75% would be taking less than 1.5 doses/wk. With a concentration of 950 fmol/punches, the estimated median [interquartile range (IQR)] is 3.9 (2.9–4.9) doses/wk. Finally, among those with a concentration of 1800 fmol/punches, an estimated 75% would be taking more than 5.6 doses/wk.
All participants had quantifiable FTC-TP in DBS at weeks 12 and 36. Median (IQR) concentrations were 1.86 (1.24–2.58), 2.53 (2.07–3.30), and 4.51 (3.30–4.97) pmol/punches for 33%, 67%, and 100% daily dosing, respectively. At week 13, 1 week into the washout, 6 (50%), 11 (92%), and 11 (100%) participants had quantifiable FTC-TP in the 33%, 67%, and 100% arms. At week 14 (ie, 2 weeks after the last FTC dose), 0 (0%), 1 (8%), and 0 (0%) participants in the 33%, 67%, and 100% arms, respectively, had quantifiable FTC-TP in DBS.
TFV-DP by fingerstick was lower compared with venipuncture by a median (IQR) of 9.0% (2.9%–19.8%), whereas FTC-TP was lower by 17.8% (9.3%–21.2%), respectively. However, the coefficient of determination (r2) was high (>0.94) for both comparisons (see Figure 2, Supplemental Digital Content, https://links.lww.com/QAI/B455).
This was a prospective, randomized, crossover, directly observed dosing study among adults without HIV that was designed to provide adherence interpretations for TFV-DP and FTC-TP in DBS after PrEP with F-TAF. TFV-DP in DBS was found to increase linearly with TAF/FTC dosing, with a half-life of 20.8 days. Similar to previous studies with TDF/FTC, proposed adherence categories were defined as the 25th percentiles of each dosing group: <450 fmol/punches, <2 doses/wk; 450–949 fmol/punches, 2–3 doses/wk; 950–1799 fmol/punches, 4–6 doses/wk; and ≥1800 fmol/punches, 7 doses/wk. These categories were the same in men and women, but research will be needed to apply these categories in special populations such as pregnant and postpartum women and adolescents. The clinical utility of these adherence interpretations is to help estimate the average adherence (ie, number of TAF/FTC doses/wk), given a TFV-DP concentration in DBS. As shown in Figure 2, several dosing frequencies are possible with each concentration, but most dose frequencies fall near the appropriate interpretation, and a clear distinction is present for very high (7 doses/wk) versus very low (<2 doses/wk) dosing. In combination with efficacy data for TAF/FTC as PrEP, these adherence thresholds are useful in defining pharmacokinetic forgiveness and the relationship between adherence and efficacy.
The pharmacokinetics of TFV-DP after TAF/FTC were similar to TDF/FTC, but with some notable differences. The slope for TFV-DP versus dose was 1.14 (1.07 to 1.21) for TAF versus 1.04 (0.97 to 1.11) for TDF.23 Technically, TAF passed dose-proportionality criteria (90% CI between 0.8 to 1.25), but concentrations increased more than dose (discussed further below). In addition, the half-life of TFV-DP in DBS was slightly longer for TAF versus TDF (20.8 days versus 17 days). The reason for this slightly longer half-life is not entirely clear, but another study showed differences in drug distribution and a slower rate of TFV-DP elimination from PBMC after TAF versus TDF dosing.36 Another difference was TFV-DP in DBS was similar between men and women after TAF, whereas TFV-DP in DBS was higher in women than men after TDF. The reason may be related to a higher body weight difference between the sexes in the TDF study compared with this one (19 kg versus 10 kg).
However, the biggest difference was much lower TFV-DP in RBCs/DBS with TAF versus TDF. To account for this, more blood was analyzed for TAF to provide a similar TFV-DP range compared with TDF, to facilitate adherence interpretation across a gradient of dosing. If less blood was assayed with TAF, the TFV-DP range would have been compressed by the lower limit of the assay, and TFV-DP would have become unquantifiable earlier than with TDF. With the methodological adjustment, approximately 11 times more RBCs were quantified from two 7-mm punches for TAF/FTC, compared with one 3-mm punch used for TDF/FTC.23 Assaying this much more blood led to comparable TFV-DP ranges and similar time to unquantifiable TFV-DP concentrations for TAF and TDF. However, there are important implications. For example, collecting enough blood for two 7-mm punches will be challenging by fingerstick. Smaller punches can be assayed and normalized to two 7-mm punches, but more unquantifiable samples will result. In addition, for a patient switching from TDF to TAF, it could take up to 5 months for the TFV-DP in DBS from TDF to washout before the concentrations could be attributed to TAF, assuming the patient started from TDF steady state. The differences in DBS concentrations between TAF and TDF likely stem from the lack of cathepsin A in RBCs, which is mostly found in lysosomes and is the primary enzyme responsible for the conversion of TAF to TFV within cells.37 It is possible that plasma TFV supplies RBCs for conversion to TFV-DP for TAF. Indeed, plasma TFV after TAF was approximately 1/10th that after TDF.38 Additional research is warranted to understand RBC uptake and phosphorylation of TFV and TFV prodrugs.
FTC-TP in DBS was quantifiable in 100% of the participants in the 100% dosing arm after 1 week of washout, but fell to 92% and 50% in the 67% arm and 33% arms, respectively. The lower quantifiable rates in the 67% and 33% arms were explained by 1.8- and 2.4-fold lower steady-state concentrations in these arms, respectively. The FTC-TP concentrations in this study were higher and remained quantifiable longer than FTC-TP in the F/TDF study, as more blood was analyzed.17,23 This enabled an estimation of dosing consistency in the preceding week. For example, as FTC-TP approached 4.5 pmol/punch (the 100% dosing median), dosing could be interpreted as more consistent in the preceding week. If FTC-TP was below the limit of quantitation while TFV-DP was quantifiable, this would suggest remote dosing with a recent dosing holiday of up to a week. One caveat with the FTC-TP concentration is it can be influenced by FTC in plasma while the blood tube sits before spotting.34 This study was conducted under controlled conditions enabling strong dose–concentration relationships, but in less-controlled settings where blood may sit for long periods before spotting, the most conservative approach would be to consider FTC-TP a dichotomous dosing biomarker (ie, quantifiable or below the limit of quantification representing a recent dose was ingested or was not ingested, respectively). Taken together, the simultaneous measurement of FTC-TP and TFV-DP in DBS informs both recent dosing and cumulative adherence to TAF/FTC.2
TFV-DP in DBS has been used across multiple PrEP studies and thousands of participants to estimate adherence for both TDF/FTC and TAF/FTC.24–27 Most recently, in the DISCOVER study,25 which included over 5000 MSM and TGW comparing TAF/FTC versus TDF/FTC as PrEP, TFV-DP in DBS was used to assess adherence in both arms. DISCOVER used the adherence thresholds proposed in this communication with the exception of using 900 fmol/punches rather than the 950 fmol/punches for ≥4 doses/wk, which was based on preliminary data from this study.39 Only 2 (of 22) HIV infections occurred in DISCOVER participants with TFV-DP associated with ≥2 doses/wk, 1 at 2–3 doses/wk in the TAF arm and 1 at ≥4 doses/wk in the TDF arm. The remaining 20 HIV infections occurred either at baseline or among those with low (<2 doses/wk) or undetectable TFV-DP in DBS.25
Age, weight, and race were significant predictors of TFV-DP in DBS in this study, but only weight was significant in multivariable analyses. Additional research is needed to study the effects of race, as this study had mostly whites. In addition, previously published data23,40 have demonstrated that TFV-DP in DBS during TDF-based therapy differs in persons living with HIV. Because of this, the adherence categories defined here for TAF/FTC should not be extrapolated for use among persons living with HIV to assess ART adherence, and adherence:concentration benchmarks must be separately established in this population.
This study had many strengths, including the controlled study design with directly observed dosing. However, there were also limitations. Although dose-proportionality was achieved, the concentration–dose slope was greater than 1. This observation was caused by TFV-DP being higher in the second dosing regimen (weeks 24 through 36) among those receiving 100% TAF/FTC compared with those receiving 100% in the first regimen (weeks 0 through 12). Concentrations were similar in the first and second dosing periods among those receiving 33% and 67% dosing. We considered the possible effect of carryover from the first regimen as median (range) residual concentrations at the start of the second regimen (week 24) were 76 (BLQ-262 fmol/punches). We subtracted these residual concentrations from the concentrations in second regimen, but this was negligible by week 36. Thus, the higher concentrations in the second dose group with 100% dosing could not be explained by carryover from the first regimen. In addition, this was not explained by controlling for race, sex, and weight. Whether this difference is due to chance or an underlying biological mechanism (eg, differences in kinase or cathepsin A expression/function or unloading/reloading of aging red blood cells) is not known at this time. We suggest this needs to be replicated before accepted as biology. Another limitation was the study population consisted of mostly white, healthy adults, which limits the ability to quantify the influence of comorbid conditions and demographic characteristics on TFV-DP in DBS. Finally, similar to what was observed with TDF/FTC, there was some overlap in TFV-DP in DBS among participants receiving 67% dosing with both 33% and 100% dosing (depicted in Figure 2), which could result in misclassification of adherence. However, the same method of adherence interpretations has been used successfully over the past 6 years for TDF-FTC.24–27 In addition, no overlap was observed between the 33% and 100% arms, making misclassification unlikely among those who are poorly versus highly adherent to TAF/FTC.
In conclusion, TFV-DP in DBS increased linearly with TAF/FTC adherence/dosing frequency, with a long half-life of approximately 21 days. FTC-TP informed dosing in the preceding week. Generally, the collection and processing of DBS is convenient, with samples being stable for 5 days after spotting at room temperature.34 These factors facilitate the use of TFV-DP and FTC-TP in DBS as cumulative and recent adherence measures for TAF/FTC to interpret study outcomes, with promise of application to clinical settings to assess medication adherence for HIV prevention and treatment.
The authors thank the study participants, the personnel at the Colorado Antiviral Pharmacology Laboratory, and the staff at the University of Colorado Clinical and Translational Research Center for their support and contributions to this work.
1. Baeten JM, Donnell D, Ndase P, et al. Antiretroviral prophylaxis for HIV prevention in heterosexual men and women. N Engl J Med. 2012;367:399–410.
2. Grant RM, Lama JR, Anderson PL, et al. Preexposure chemoprophylaxis for HIV prevention in men who have sex with men. N Engl J Med. 2010;363:2587–2599.
3. Marrazzo JM, Ramjee G, Richardson BA, et al. Tenofovir-based preexposure prophylaxis for HIV infection among African women. N Engl J Med. 2015;372:509–518.
4. Thigpen MC, Kebaabetswe PM, Paxton LA, et al. Antiretroviral preexposure prophylaxis for heterosexual HIV transmission in Botswana. N Engl J Med. 2012;367:423–434.
5. Van Damme L, Corneli A, Ahmed K, et al. Preexposure prophylaxis for HIV infection among African women. N Engl J Med. 2012;367:411–422.
6. Fonner VA, Dalglish SL, Kennedy CE, et al. Effectiveness and safety of oral HIV preexposure prophylaxis for all populations. AIDS. 2016;30:1973–1983.
7. Brooks KM, Anderson PL. Pharmacologic-based methods of adherence assessment in HIV prevention. Clin Pharmacol Ther. 2018;104:1056–1059.
8. Koenig HC, Mounzer K, Daughtridge GW, et al. Urine assay for tenofovir to monitor adherence in real time to TDF/FTC as PrEP. HIV Med. 2017;18:412–418.
9. Fonsart J, Saragosti S, Taouk M, et al. Single-dose pharmacokinetics and pharmacodynamics of oral tenofovir and emtricitabine in blood, saliva and rectal tissue: a sub-study of the ANRS IPERGAY trial. J Antimicrob Chemother. 2017;72:478–485.
10. de Lastours V, Fonsart J, Burlacu R, et al. Concentrations of tenofovir and emtricitabine in saliva: implications for preexposure prophylaxis of oral HIV acquisition. Antimicrob Agents Chemother. 2011;55:4905–4907.
11. Hendrix CW, Andrade A, Bumpus NN, et al. Dose frequency ranging pharmacokinetic study of tenofovir-emtricitabine after directly observed dosing in healthy volunteers to establish adherence benchmarks (HPTN 066). AIDS Res Hum Retroviruses. 2016;32:32–43.
12. Seifert SM, Chen X, Meditz AL, et al. Intracellular tenofovir and emtricitabine anabolites in genital, rectal, and blood compartments from first dose to steady state. AIDS Res Hum Retroviruses. 2016;32:981–991.
13. Drain PK, Kubiak RW, Siriprakaisil O, et al. Urine tenofovir concentrations correlate with plasma and relates to TDF adherence: a randomized directly-observed pharmacokinetic trial (target study). Clin Infect Dis. 2019. doi: 10.1093/cid/ciz645.
14. Castillo-Mancilla JR, Haberer JE. Adherence measurements in HIV: new advancements in pharmacologic methods and real-time monitoring. Curr HIV/AIDS Rep. 2018;15:49–59.
15. Bekker L-G, Roux S, Sebastien E, et al. Daily and non-daily pre-exposure prophylaxis in African women (HPTN 067/ADAPT Cape Town Trial): a randomised, open-label, phase 2 trial. Lancet HIV. 2018;5:e68–e78.
16. Corneli AL, Deese J, Wang M, et al. FEM-PrEP: adherence patterns and factors associated with adherence to a daily oral study product for pre-exposure prophylaxis. J Acquir Immune Defic Syndr. 2014;66:324–331.
17. Castillo-Mancilla J, Seifert S, Campbell K, et al. Emtricitabine-triphosphate in dried blood spots as a marker of recent dosing. Antimicrob Agents Chemother. 2016;60:6692–6697.
18. Koss CA, Bacchetti P, Hillier SL, et al. Differences in cumulative exposure and adherence to tenofovir in the VOICE, iPrEx OLE, and PrEP demo studies as determined via hair concentrations. AIDS Res Hum Retroviruses. 2017;33:778–783.
19. Gandhi M, Wang G, King R, et al. Development and validation of the first point-of-care assay to objectively monitor adherence to HIV treatment and prevention in real-time in routine settings. AIDS. 2020;34:255–260.
20. Liu AY, Yang Q, Huang Y, et al. Strong relationship between oral dose and tenofovir hair levels in a randomized trial: hair as a potential adherence measure for pre-exposure prophylaxis (PrEP). PLoS One. 2014;9:e83736.
21. Adams JL, Sykes C, Menezes P, et al. Tenofovir diphosphate and emtricitabine triphosphate concentrations in blood cells compared with isolated peripheral blood mononuclear cells: a new measure of antiretroviral adherence? J Acquir Immune Defic Syndr. 2013;62:260–266.
22. Castillo-Mancilla JR, Zheng JH, Rower JE, et al. Tenofovir, emtricitabine, and tenofovir diphosphate in dried blood spots for determining recent and cumulative drug exposure. AIDS Res Hum Retroviruses. 2013;29:384–390.
23. Anderson PL, Liu AY, Castillo-Mancilla JR, et al. Intracellular tenofovir-diphosphate and emtricitabine-triphosphate in dried blood spots following directly observed therapy. Antimicrob Agents Chemother. 2018;62:pii: e01710-17.
24. Liu AY, Cohen SE, Vittinghoff E, et al. Preexposure prophylaxis for HIV infection integrated with municipal- and community-based sexual health services. JAMA Intern Med. 2016;176:75–84.
25. Hare CB, Coll J, Ruane P, et al. The phase 3 Discover study: daily F/TAF or F/TDF for HIV preexposure prophylaxis . Conference on Retroviruses and Opportunistic Infections; March 4–7, 2019; Seattle, Washington.
26. Hosek S, Landovitz R, Kapogiannis B, et al. Antiretroviral pre-exposure prophylaxis (PrEP) for adolescent men who have sex with men (MSM) ages 15-17 years in the United States. JAMA Pediatr. 2017;171:1063–1071.
27. Grant RM, Anderson PL, McMahan V, et al. Uptake of pre-exposure prophylaxis, sexual practices, and HIV incidence in men and transgender women who have sex with men: a cohort study. Lancet Infect Dis. 2014;14:820–829.
28. Molina JM, Capitant C, Spire B, et al. On-demand preexposure prophylaxis in men at high risk for HIV-1 infection. N Engl J Med. 2015;373:2237–2246.
29. Anderson PL, Glidden DV, Liu A, et al. Emtricitabine-tenofovir concentrations and pre-exposure prophylaxis efficacy in men who have sex with men. Sci Transl Med. 2012;4:151ra125.
30. Ray AS, Fordyce MW, Hitchcock MJ. Tenofovir alafenamide: a novel prodrug of tenofovir for the treatment of Human Immunodeficiency Virus. Antivir Res. 2016;125:63–70.
31. Yager J, Brooks K, Castillo-Mancilla J, et al. Tenofovir diphosphate in PBMC following increasing TAF vs TDF dosing under directly observed therapy . 20th International Workshop on Clinical Pharmacology of HIV, Hepatitis, and Other Antiviral Drugs; May 14–16, 2019; Noordwijk, the Netherlands.
32. Castillo-Mancilla J, Coyle RP, Zheng J-H, et al. Tenofovir diphosphate arising from TAF IS quantifiable IN dried blood spots . Conference on Retroviruses and Opportunistic Infections; February 13–16, 2017; Seattle, WA.
33. Descovy (emtricitabine/tenofovir alafenamide) [package insert]. Foster City, CA: Gilead Sciences, Inc; 2019.
34. Zheng JH, Rower C, McAllister K, et al. Application of an intracellular assay for determination of tenofovir-diphosphate and emtricitabine-triphosphate from erythrocytes using dried blood spots. J Pharm Biomed Anal. 2016;122:16–20.
35. Sethuraman VS, Leonov S, Squassante L, et al. Sample size calculation for the Power Model for dose proportionality studies. Pharm Stat. 2007;6:35–41.
36. Greene SA, Chen J, Prince HMA, et al. Population modeling highlights drug disposition differences between tenofovir alafenamide and tenofovir disoproxil fumarate in the blood and semen. Clin Pharmacol Ther. 2019;106:821–830.
37. Callebaut C, Stepan G, Tian Y, et al. In vitro virology profile of tenofovir alafenamide, a novel oral prodrug of tenofovir with improved antiviral activity compared to that of tenofovir disoproxil fumarate. Antimicrob Agents Chemother. 2015;59:5909–5916.
38. Podany AT, Bares SH, Havens J, et al. Plasma and intracellular pharmacokinetics of tenofovir in patients switched from tenofovir disoproxil fumarate to tenofovir alafenamide. AIDS. 2018;32:761–765.
39. Yager J, Castillo-Mancilla J, Ibrahim M, et al. Tenofovir-diphosphate in dried blood spots following escalating TAF/FTC dosing [poster 463]. Conference on Retroviruses and Opportunistic Infections; March 4–7, 2019; Seattle, WA.
40. Castillo-Mancilla J, Morrow M, Coyle RP, et al. Tenofovir diphosphate in dried blood spots is strongly associated with viral suppression in individuals with HIV infection. Clin Infect Dis. 2019;68:1335–1342.