PrEP demonstration project showed superior adherence with tenofovir alafenamide/emtricitabine compared to tenofovir disoproxil fumarate/emtricitabine in a sample of partnered sexual minority men

Abstract

Sexual minority men (SMM) remain at high risk of HIV infection in the United States, and for those in relationships, dyadic functioning may contextualize prevention decisions. Pre-exposure prophylaxis (PrEP) for HIV prevention was previously limited to tenofovir disoproxil fumarate/emtricitabine (TDF/FTC) until the FDA approved tenofovir alafenamide/FTC (TAF/FTC) for PrEP in 2019. Data regarding substance use, sexual behavior, holding an active PrEP prescription, and type of PrEP regimen (TDF/FTC versus TAF/FTC) were analyzed from a sample of 421 partnered SMM. The majority of the sample on PrEP reported a TDF/FTC prescription as opposed to TAF/FTC. However, SMM report better adherence to TAF/FTC than TDF/FTC in multivariable models when comparing self-reported adherence. Novelty of TAF/FTC, treatment fatigue with TDF/FTC, and/or a belief in TAF/FTC’s superior efficacy and mitigated side effects may be contributing. More studies using objective adherence metrics and surveys are needed.

RESUMEN

Los hombres de minorías sexuales (SMM, por sus siglas en inglés) permanecen en alto riesgo de infección por VIH en los Estados Unidos y, para quienes están en una relación romántico, el funcionamiento diádico puede contextualizar las decisiones de prevención del VIH. La profilaxis previa a la exposición (PrEP) para la prevención del VIH se limitaba a tenofovir disoproxil fumarato / emtricitabina (TDF / FTC) hasta que la FDA aprobó tenofovir alafenamida / FTC (TAF / FTC) para PrEP en 2019. Datos sobre el uso de sustancias, comportamientos sexuales, teniendo una prescripción de PrEP activa, y el tipo de régimen de PrEP (TDF / FTC versus TAF / FTC) se analizaron de una muestra de 421 SMM asociados. La mayoría de la muestra que toma PrEP indicó teniendo una prescripción de TDF / FTC en lugar de TAF / FTC. Sin embargo, comparando la adherencia auto-informada, SMM indica mejor adherencia a TAF / FTC que TDF / FTC en modelos multivariables. La novedad de TAF / FTC, la fatiga del tratamiento con TDF / FTC y/o la creencia en la eficacia superior de TAF / FTC y los efectos secundarios mitigados pueden contribuir a la mejor adherencia a TAF / FTC. Se necesitan más estudios que utilicen métricas y encuestas de adherencia objetivas.

INTRODUCTION

Uptake of biomedical prevention strategies among sexual minority men (SMM) in relationships is an essential component of the prevention strategy of Ending the HIV Epidemic (1). SMM comprised 69% of new HIV cases in 2018 in the U.S (2). As many as two-thirds of new HIV infections in this population are transmitted between main—rather than casual—partners (3). The Centers for Disease Control and Prevention recommends pre-exposure prophylaxis (PrEP) for SMM in serodiscordant or non-monogamous relationships (4).

Options for PrEP for HIV prevention in the U.S. have been limited to tenofovir disoproxil fumarate/emtricitabine (TDF/FTC) since 2012 (5–7); however, in 2019, the Food and Drug Administration (FDA) approved tenofovir alafenamide/emtricitabine (TAF/FTC) for PrEP for non-vaginal sex (8). TAF/FTC is marketed as a PrEP option with a greater renal and bone safety profile than TDF/FTC (9, 10). TAF/FTC may also have a more rapid onset than TDF/FTC, potentially increasing prevention potency (11). However, the side effects and efficacy of TAF/FTC are likely only marginally improved when compared to those of TDF/FTC (8).

As TDF/FTC, until recently, has been the only PrEP regimen available to SMM, little research has examined correlates of uptake or adherence to TDF/FTC versus TAF/FTC. Previous research with SMM in relationships has indicated that PrEP uptake is inhibited, for at least some, by concerns that their partners will perceive a lack of fidelity, trust, or commitment to the relationship (12, 13). By contrast, research has also shown that PrEP use in relationships may serve as a conduit for open communication regarding sexual agreements and, in so doing, bolster emotional closeness (14). While research has highlighted the role of relationship dynamics in PrEP uptake and adherence among SMM in relationships, no research has been conducted which may lay a foundation for future studies on how such dynamics may influence the uptake and adherence of either TDF/FTC or TAF/FTC among partnered SMM (13–16). The current study investigated whether differences in uptake or adherence to the two oral PrEP options exist among a sample of partnered SMM (18 years of age and over) and examined the behavioral predictors of PrEP adherence with each regimen.

METHODS

Data was analyzed from an English-language online screening questionnaire for SMM used to determine eligibility for a randomized controlled trial of a PrEP adherence intervention. Online advertisements across social media platforms and dating apps targeted SMM in the New York City metropolitan area. The Institutional Review Board at CUNY Hunter College approved procedures and participants’ free, and informed consent was obtained. Data collection from the questionnaires occurred between January 21, 2020 to March 2, 2020.

Participants

Eligibility criteria included being 18 years of age or older; a sexual minority (i.e., gay, bisexual, queer); cis-male with a cis-main male partner; sex with at least one casual partner in the past 30 days; and drug use in the past 30 days.

Measures

Demographics.

Participants reported age, education, income, race and ethnicity, relationship length, HIV status, and the HIV status of their main partner.

Drug Use.

Participants were asked a series of questions regarding specific drugs use in the previous 30 days with a dichotomous response format (yes/no) for each type of drug. The types included: Cannabis, cocaine/crack, methyl enedioxy -methamphetamine (MDMA), gamma-hydroxybutyrate (GHB), ketamine (K), and methamphetamines. Further, MDMA, GHB, and K were combined into one variable labeled “club drug use” in a dichotomous variable (at least one was used vs. none were use).

Sexual Behaviors

Condomless sex in the past 90 days.

In separate questions, participants reported on the occurrence of anal sex with their main partner or a casual partner. Participants were then asked to report the number of condomless anal sex (CAS) events with their main and casual partners separately.

Biomedical HIV prevention

PrEP Use.

First, participants reported whether or not they had an active PrEP prescription. Then those with an active prescription indicated whether it was for TDF/FTC or TAF/FTC. This information was used to create a 3-category variable to indicate PrEP status (Not on PrEP, TDF/FTC, and TAF/FTC).

Adherence.

Participants who indicated that they had an active PrEP prescription then reported the number of days that they missed taking their daily PrEP as prescribed, in the past 30 days. This yielded a count variable ranging from 0 to 30.

Statistical Analysis

Bivariate analysis tested associations between demographic variables, substance use, and sexual behaviors with 3-category PrEP uptake (Not on PrEP, TDF/FTC, and TAF/FTC). Next, a negative binomial regression model using the count variable (number of doses missed) examined the predictive utility of PrEP-type in the context of demographic and behavioral covariates.

RESULTS

A total of 622 screeners were accessed between the dates of January 21, 2020 to March 2, 2020. 201 screeners were incomplete. Thus, there were 421 complete screeners with no missing data. All items on the survey were forced responses. Of the 421 participants, 254 (60%) indicated no active PrEP prescription, and 167 (39.7%) indicated having a PrEP prescription with 119 (28.3%) having TDF/FTC and 48 having a TAF/FTC prescription. Table 1 shows bivariate associations between participant characteristics and the three PrEP uptake categories (Not on PrEP, TDF/FTC, and TAF/FTC). The average number of PrEP doses missed was higher among men on TDF/FTC (2.30 days missed with standard deviation (SD) 4.71) relative to number of doses missed (0.65 days with SD 1.14) among participants on TAF/FTC (t(165) = 2.41, p = 0.017 for difference). Lower income (χ2(2) = 8.52, p = 0.016) was also significantly associated with men not on PrEP are more likely to earn less that $40,000 compared to those on TDF/FTC. Men on TAF/FTC do not differ significantly from either of the other groups with respect to income. Being on PrEP was positively associated with casual partner CAS with 73.9% of those on TDF/FTC reporting CAS and 72.9% of those on TAF/FTC reporting CAS compared to 34.3% of those not on PrEP (χ2(2) = 62.58, p < 0.001).

Table I:

Summary of Demographic Characteristics

Overall			Not on PrEP		TDF/FTC		TAF/FTC		Test Statistic	P Value
	n = 421	(%)	n = 254	(%)	n =119	(%)	n = 48	(%)
Age									χ2= 17.06, df = 10	.054
18–24	42	(9.9)	32	(12.6)	6	(5.0)	4	(8.3)
25–29	95	(22.5)	62	(24.4)	23	(19.3)	9	(18.8)
30–34	81	(19.1)	42	(16.5)	30	(25.2)	9	(18.8)
35–30	72	(17.0)	38	(15.0)	27	(22.7)	7	(14.6)
40–44	44	(10.4)	23	(9.1)	16	(13.4)	5	(10.4)
45+	89	(21.0)	57	(22.4)	17	(14.3)	14	(29.2)
Education									χ2= 2.47, df = 2	.290
< 4 year	141	(33.5)	91	(35.8)	33	(27.7)	17	(35.4)
≥ 4 year	280	(66.5)	163	(64.2)	86	(72.3)	31	(64.6)
Income									χ2= 8.52*, df = 2	.016
< 40k	169	(40.1)	116a	(45.7)	36b	(30.3)	17a,b	(35.4)
≥ 40k	252	(59.9)	138a	(54.3)	83b	(69.7)	31a,b	(64.6)
Race and Ethnicity									χ2= 6.792, df = 6	.341
Black	77	(18.3)	49	(19.3)	23	(19.3)	5	(10.4)
White	197	(46.8)	111	(43.7)	57	(47.9)	29	(60.4)
Latinx	96	(22.8)	64	(25.2)	25	(21.0)	7	(14.6)
Other	51	(12.1)	30	(11.8)	14	(11.8)	7	(14.6)
Drug Use (past 30 days)
Cannabis	196	(46.6)	119	(46.9)	55	(46.2)	22	(45.8)	χ2= 0.02, df = 2	.988
Cocaine/Crack	32	(7.6)	15	(5.9)	13	(10.9)	4	(8.3)	χ2= 2.94, df = 2	.229
Club Drugs (GHB/K/MDMA)	64	15.2	30	(11.8)	25	(21.0)	9	(18.8)	χ2= 5.84, df = 2	.054
Methamphetamines	33	(7.8)	23	(9.1)	9	(7.6)	1	(2.1)	χ2= 2.73, df = 2	.255
Condomless sex (past 90 days)
Main Partner CAS	254	(60.3)	147	(57.9)	78	(65.5)	29	(60.4)	χ2= 1.99, df = 2	.369
Casual Partners CAS	210	(49.9)	87a	(34.3)	88b	(73.9)	35b	(72.9)	χ2= 62.58**, df = 2	<.001
Partner HIV Status
Negative	363	(86.2)	222	(87.4)	100	(84.0)	41	(85.4)	χ2= .80, df = 2	.669
Relationship Length
> 2 years	279	(66.3)	166	(65.4)	78	(65.5)	35	(72.9)	χ2= 1.07, df = 2	.585
	M	SD	M	SD	M	SD	M	SD
PrEP Adherence (range 0–28)††	0.72	2.7	-	-	2.30	4.71	0.65	1.14	t = 2.41**, df = 165	.017

^*p < 0.05;

^**p < 0.01;

^a-bWithin rows, cells with different superscripts differ at p < .05;

^†Days missed a dose in past 30 days

In multivariable negative binomial regression models, PrEP type was associated with adherence (Table II). Specifically, the use of TAF/FTC was associated with significantly better adherence than the use of TDF/FTC (RR = 0.29; CI 95% 0.15, 0.55; p < 0.01). In addition, greater income (RR = 0.42; CI 95% 0.21, 0.86; p = 0.017), using club drugs (RR = 0.35; CI 95% 0.16, 0.76; p = 0.008), and being White (versus Black) (RR = 5.69; CI 95% 2.27, 14.30; p < 0.001) were associated with fewer missed doses for both TAF/FTC and TDF/FTC. Methamphetamine use was associated with more missed doses for both types of PrEP (RR = 5.38; CI 95% 1.94 14.96; p = 0.001).

Table II.

Negative Binomial PrEP Adherence Model (n = 169)

	B	CI	Rate Ratios	CI	P Value
Age
18–24 (ref)
25–29	−0.06	−1.32, 1.20	0.94	0.26, 3.30	.921
30–34	0.29	−1.02, 1.62	1.34	0.35, 5.06	.660
35–30	−0.21	−1.59, 1.16	0.80	0.20, 3.21	.762
40–44	−1.20	−2.69, 0.31	0.30	0.07, 1.36	.120
45+	−0.73	−2.18, 0.73	0.48	0.11, 2.07	.328
Education
≥ 4 year (ref: <4 year)	0.08	−0.58, 0.75	1.09	0.56, 2.12	.795
Income
≥ 40k (ref: < 40k)	−0.86	−1.57, −0.16	0.42*	0.21, 0.86	.017
Race and Ethnicity
Black (ref)
White	1.74	0.82, 2.66	5.69**	2.27, 14.30	<.001
Latinx	0.73	−0.27, 1.72	2.07	0.76, 5.58	.152
Other	0.51	−0.42, 1.44	1.66	0.65, 4.22	.283
Drug Use (past 30 days)
Cannabis	0.32	−0.27, 0.91	1.37	0.76, 2.46	.295
Cocaine/Crack	0.14	−0.71, 0.99	1.15	0.49, 2.69	.748
Club Drugs (MDMA/GHB/K)	−1.04	−1.82, −0.30	0.35**	0.16, 0.76	.008
Methamphetamines	1.68	0.66, 2.71	5.38**	1.94 14.96	.001
Condomless sex (past 90 days)
Main Partners	−0.17	−0.78, 0.45	0.84	0.45, 1.58	.603
Casual Partners	−0.07	−0.73 0.56	0.93	0.48, 1.81	.837
Type of PrEP
TAF/FTC (ref: TDF/FTC)	−1.24	−1.88, −.059	0.29**	0.15, 0.55	<.001
Main Partner HIV Status
Negative (ref: Positive)	−0.04	−0.79, 0.75	0.98	0.45, 2.04	.920
Relationship Length
>2 years (ref: ≤ 2 years)	0.19	−0.57, 0.96	1.22	0.56, 2.62	.615

^*p<0.05;

^**p <0.01

DISCUSSION

Although the FDA had only recently approved TAF/FTC for PrEP relative to our analysis, these findings suggest that the use of TAF/FTC is associated with better adherence than TDF/FTC in multivariable models. Findings also demonstrate that, while the majority of our sample had active TDF/FTC prescriptions as opposed to TAF/FTC for PrEP, there were no notable demographic differences between those prescribed the different types of oral PrEP. Further research regarding PrEP uptake and awareness as these two options become mainstream, along with new options for injectable ART for prevention, is indicated (17).

A number of factors may contribute to our finding of better adherence to TAF/FTC compared to that of TDF/FTC. Suboptimal adherence to TDF/FTC, when compared to TAF/FTC, may be a function of HIV prevention fatigue with the longer-available option (18, 19). Since TAF/FTC for PrEP only received FDA approval in the latter half of 2019, those SMM with active TAF/FTC prescriptions may have started those regimens more recently than those prescribed TDF/FTC (8). Therefore, TAF/FTC’s novelty may mitigate the effect of prevention fatigue and, in so doing, drive more optimal adherence to this oral PrEP regimen (18, 19).

Preliminary findings from Gilead Sciences’ DISCOVER trial (published in abstract form only) have publicized TAF/FTC’s superior efficacy and lower side effect profile in comparison with those of TDF/FTC (8, 10, 11). In light of these abstract proceedings, SMM’s belief in the DISCOVER trial’s results may, therefore, serve to motivate better adherence to TAF/FTC than to TDF/FTC (9, 11). Recent research has shown that suboptimal PrEP adherence is informed by SMM’s concern regarding medication side effects (20), although the comparable benefits of TAF/FTC over TDF/FTC are still being examined (21). Future research by our group that includes surveys and objective adherence metrics will more explicitly examine motivations behind our findings.

Relationship dynamics—particularly sexual agreements—may also contribute to the aforementioned finding on differences in HIV prevention (22). Research has found that seronegative SMM in non-monogamous relationships may appraise their own HIV infection risk as increased due to sexual interactions with casual partners (12). Further research has also indicated that PrEP use is more valued and positively perceived in the context of non-monogamous relationships among SMM than in those that are monogamous (16, 23). Considering the novelty and publicized health benefits of TAF/FTC, findings suggest that a prescription to this option for PrEP may indicate greater awareness of one’s heightened HIV-risk. Thus, those with TAF/FTC prescriptions may be optimally adherent to this regimen. By contrast, for those among the sample with TDF/FTC prescriptions, suboptimal PrEP adherence may be a function of periods during which sexual contact with casual partners and subsequent HIV-risk is considered low as a result of changing relationships over time.

Findings regarding extant differences in TAF/FTC and TDF/FTC adherence may also be evidence of adherence drivers founded at the health care provider-level. Research has shown— prior to the approval of TAF/FTC for PrEP—that primary care providers have cited TDF/FTC’s side effects, as well as challenges in monitoring adherence, as barriers to prescribing PrEP to patients who would otherwise stand to benefit from such a regimen (24, 25). As such, our findings may indicate that—for participants who changed from TDF/FTC to TAF/FTC regimens—health care providers may have encouraged such a regimen transition, citing a given patient’s adverse side effects to TDF/FTC, but otherwise optimal PrEP adherence. Conversely, our findings on suboptimal TDF/FTC adherence may indicate that providers would be reluctant to encourage such a change in regimen should a patient already struggle with adherence to TDF/FTC.

We also found factors associated with decreased adherence to PrEP in general (regardless of regimen) that have been reported in other studies, including methamphetamine use and being of lower income. However, our findings showed that participants who identified as Black were more adherent to PrEP than those who identified as White. This finding may indicate a shift in PrEP awareness and adherence among Black SMM and deviates notably from previous research showing lower adherence among minority populations (26, 27). Further research is needed to identify the social constructs that may explain this difference.

Findings are limited by a lack of supplementary data on the study cohort. Data on whether those participants with an active TAF/FTC prescription had switched from a TDF/FTC regimen prior to their participation in the study was not collected. Qualitative data on the motivation for such a transition, should it have been made by participants, may speak directly to the drivers of more optimal adherence to TAF/FTC versus TDF/FTC seen in our results. The current study also could not distinguish suboptimal PrEP adherence from on-demand PrEP regimens, which research has shown to be highly effective in the prevention of HIV (in the study of TDF/FTC regimens) (28). Thus, future research on TAF/FTC versus TDF/FTC adherence ought to control for on-demand PrEP regimens so as not to conflate such regimens with non-adherence. Additionally, participants were recruited though social media and mobile dating apps and, therefore, may not represent the general population of partnered SMM. Finally, we collected adherence data via self-report, which can be limited compared to more objective metrics, such as drug levels.

CONCLUSIONS

Overall, our findings are intriguing and suggest that adherence to TAF/FTC is better among SMM than adherence to TDF/FTC for PrEP. As HIV prevention options expand for SMM (17), further research on the behavioral factors that influence PrEP uptake and adherence to different regimens would inform future iterations of biomedical HIV prevention options and adherence interventions for SMM. Future research on perceptions of PrEP efficacy and side effects would be especially meaningful as multiple drug options expand to the market.