Effect of tenofovir-based HIV pre-exposure prophylaxis against HBV infection in men who have sex with men

Abstract

Background and Aims: 

Men who have sex with men (MSM) are vulnerable to contracting HBV as a sexually transmitted infection. We evaluated the incidence of HBV infection (HBI) and the prophylactic effect of tenofovir-based pre-exposure prophylaxis (PrEP) on HBI in an MSM cohort.

Methods and Results: 

MSM who were older than 16 years were enrolled from January 2018 and followed up until June 2021 and tested for HIV, bacterial sexually transmitted infections, and HBsAg/ HBsAb and HBcAb every 3 months based on inclusion criteria, including HBsAg, HBcAb, HBsAb, and HIV negativity at enrollment. HBI was defined as seroconversion of HBsAg or HBcAb status. The log-rank test was used to evaluate the prophylactic effect of PrEP against HBI. As a substudy, individuals excluded from the main study due to HBs Ab positivity were evaluated for HBI incidence. Among 1577 MSM, 786 participants (546 PrEP nonusers, 131 daily PrEP users, and 109 event-driven PrEP users) met the criteria and were included. The annual incidence of HBV among PrEP nonusers (3.8%, 21 infections, with 559.5 person-years) was significantly higher (p = 0.018, log-rank test) than that among daily PrEP users [0.77%, 1 infection (admitted nonadherence), with 129.3 person-years] and event-driven PrEP users (no infection with 93.8 person-years). Although the incidence of HBI and HIV infection decreased with PrEP use, the incidence of other sexually transmitted infections was higher in both daily and event-driven PrEP users. The annual incidence of HBV among HBsAb-positive and HBcAb-negative PrEP nonusers was 1.8% (3 infections, with 167.5 person-years).

Conclusions: 

Tenofovir-based PrEP prevented HBI among MSM in a real-world setting.

INTRODUCTION

In western countries and in Japan, men who have sex with men (MSM) are at great risk of HBV exposure through sexual intercourse.^1,2 The worldwide implementation of HBV vaccination has decreased the incidence and prevalence of HBV infection over the last 2 decades.³ However, those who are nonresponders to HBV vaccination, comprising an estimated 5% to 10% of all vaccinated individuals, remain at risk of HBV infection.⁴ In Japan, universal HBV vaccination was unavailable until 2016, and MSMs remain vulnerable to HBV infection.⁵ The prevalence of HBV infection among MSM with HIV infection in Japan is estimated to be ~10%, but limited information is available on the epidemiology of HBV infection among MSM without HIV infection.⁶

Antiretroviral therapy containing tenofovir disoproxil fumarate (TDF) 300 mg/emtricitabine (FTC) 200 mg has a prophylactic effect against HBV infection among MSM with HIV infection.^6–8 Furthermore, daily HIV pre-exposure prophylaxis (PrEP) with TDF/FTC has been implemented in many countries and regions as a promising strategy for HIV prevention.^9–14 Theoretically, PrEP could have a prophylactic effect against HIV and HBV, given the evidence on TDF/FTC use among individuals with HIV infection. However, the most important concern with PrEP is the suboptimal adherence rate, which undermines the prophylactic efficacy against HIV infection.^10,15 Assuming that adherence to PrEP (prophylaxis for asymptomatic people) is lower than that to antiretroviral therapy for HIV infection (treatment for patients), the prophylactic efficacy of PrEP against HBV infection in a clinical setting is of great interest. Besides daily PrEP, “on-demand” PrEP (a double dose of TDF/FTC taken 2–24 h before each sexual intercourse, followed by 2 single doses of TDF/FTC taken 24 and 48 h after the first drug intake), also known as event-driven PrEP (ED-PrEP), is an evidence-based regimen among MSM and is used in various countries and regions.^16,17 However, the prophylactic effect of this regimen against HBV infection has not been evaluated previously.

We aimed to assess the incidence of HBV infection and the prophylactic effect of PrEP against HBV infection among MSM without HIV infection living in Tokyo and its suburbs. Moreover, we evaluated the prophylactic effect of PrEP according to the dosing regimen while examining its associated risk factors, including sexual activity and sexually transmitted infection (STI) prevalence and incidence.

METHODS

Study population and design

In January 2017, we established an HIV-negative MSM cohort at a sexual health clinic (SHC) at the National Center for Global Health and Medicine in Tokyo, Japan, to provide HIV, bacterial STI, and viral hepatitis tests (HAV, HBV, and HCV) every 3 months. MSM without HIV infection, aged 17 years and older, who provided written informed consent were included in the SHC study (UMIN000035304). Participants were tested for HIV infection, syphilis, pharyngeal and rectal Chlamydia trachomatis and Neisseria gonorrhoeae infections, and HBsAg and HBs, HBc, HCV, and HAV-IgG antibodies every 3 months for free. Among the patients included in the SHC between January 2018 and December 2020, those who fulfilled the inclusion criteria were enrolled and followed up until June 2021. The inclusion criteria were HBsAg negative (<0.05 IU/mL), HBs and HBc antibody negative [<10.0 mIU/L and<0.5 S/CO (CLIA method), respectively], and HIV-negative status at study enrollment.

The SHC recommends individuals without previous HBV and HAV exposure be vaccinated against HBV and HAV at their own expense. TDF/FTC was provided, including STI testing every 3 months free of charge through a clinical trial of daily PrEP (UMIN000031040). Individuals who missed participating in the clinical trial because of the limited number of participants but were willing to use PrEP could purchase generic TDF/FTC or tenofovir alafenamide (TAF)/FTC online at their own expense. These “self-PrEP” users chose daily or on-demand PrEP based on their preference. MSM who did and did not use PrEP were prospectively followed up for incident HBV infection and other STIs every 3 months during the follow-up period.

As a substudy, those who have excluded from the main study due to HBsAb and/or HBcAb positivity, were screened for incident HBV infection for HBs antibody-positive and HBc antibody-negative participants and for their HBV reactivation for HBc antibody-positive participants.

This study was approved by the Human Research Ethics Committee of the National Center for Global Health and Medicine (NCGM-G-002091-05 and NCGM-G-002333-02). All participants provided written informed consent for study participation and for the research-based use and publication of their clinical and laboratory data. This study was conducted in accordance with the principles of the Declaration of Helsinki.

Outcomes

Incident HBV infection was defined as seroconversion to HBcAb or HBsAg positive during the follow-up period. A single and transient HBcAb positivity without any HBsAb-positive seroconversion was regarded as a false positive. Participants who acquired HBV or HIV infection were censored. Moreover, individuals with HBs antibody levels ≥10 mIU/mL (CLIA method) were similarly censored because they had acquired effective protection against HBV infection. The incidence of HBV infection was calculated according to the PrEP status.

Measurements

The cutoff values of HBsAg, HBsAb, and total HBcAb tests using CLIA were 0.05 IU/mL, 10.0 mIU/mL, and 1.00 S/CO, respectively; BML Inc., Tokyo, Japan) were used. For HAV and HCV detection, the HAV-IgG and HCV antibody (CLIA and CLEIA, cutoff values: 1.00 S/CO and 1.0 mIU/mL, respectively; BML Inc., Tokyo, Japan) were used. Every 3 months, the time of screening for other STIs, cohort members underwent HIV, syphilis, and pharyngeal and rectal C. trachomatis and N. gonorrhoeae testing using the fourth-generation antigen/antibody by HISCL (Sysmex Co., Tokyo, Japan), quantitative rapid plasma reagin (RPR)/Treponema pallidum antibody hemagglutination (TPHA; Sekisui Medical Co., Tokyo, Japan/Fujirebio Co., Tokyo, Japan), and transcription-mediated amplification (Bio Medical Laboratories Inc., Tokyo, Japan), respectively.

Participants with HBsAg-positive conversion were tested for HBV DNA, HBV serotype, HBe antigen/antibody, and aspartate aminotransferase/alanine aminotransferase levels.

Statistical analysis

The chi-squared or Fisher exact test was used to analyze categorical variables. The independent t test was used to compare means, and in case of asymmetry, the Mann-Whitney U or Kruskal-Wallis test was used. Statistical significance was defined as a 2-sided p < 0.05. The log-rank test was used to evaluate the prophylactic effect of PrEP against HBV infection. Cox proportional hazards regression was used to evaluate the prophylactic effect of PrEP against HBV infection and other factors, including sexual activity. All statistical analyses were performed using Stata version 16 (StataCorp, College Station, TX).

RESULTS

Figure 1 illustrates the flowchart of the enrollment of MSM in this study. A total of 1577 MSM were included in the cohort until December 2020, of whom 791 were excluded from the study in accordance with the exclusion criteria (41 due to HIV infection, 22 due to HBsAg positivity, 219 due to HBsAb positivity, 336 due to HBcAb positivity at enrollment, and 173 due to the lack of relevant data). Thus, 786 participants (546 PrEP nonusers and 240 PrEP users, including 131 daily PrEP users and 109 ED-PrEP users) were followed up with HIV and STI testing every 3 months during the observational period. Table 1 shows the baseline characteristics of MSM with and without PrEP use.

FIGURE 1: Flowchart of men who have sex with men with and without pre-exposure prophylaxis. Abbreviations: MSM, men who have sex with men; PrEP, pre-exposure prophylaxis.

TABLE 1 - Baseline differences between PrEP users and nonusers

		PrEP use
Characteristics	No PrEP use (n=546)	All (n=240)	Daily (n=131)	Event-driven (n=109)	p a (PrEP nonusers vs. users)
Age (y)	32.1 (8.9)	36.3 (9.3)	36.1 (9.4)	36.5 (9.2)	<0.001
History of syphilis	95 (17.4)	42 (17.5)	22 (16.8)	20 (18.3)	0.973
Active syphilis	29 (5.3)	10 (4.2)	8 (6.1)	2 (1.8)	0.496
Chlamydial infection	89 (16.3)	34 (14.2)	21 (16.0)	13 (11.9)	0.448
Gonorrheal infection	51 (9.3)	13 (5.4)	6 (4.6)	7 (6.4)	0.064
Any bacterial STI	134 (24.5)	51 (21.3)	30 (22.9)	21 (19.3)	0.316
HAV antibody+	46 (9.1)	34 (16.3)	17 (16.7)	17 (15.9)	0.005
HCV antibody+	0	1 (0.5)	1 (1.0)	0	0.289
No. sexual partners, mean (SD)	12.3 (39.9)	11.5 (14.7)	11.5 (14.5)	11.5 (15.0)	0.77
Average rate of condom use	68.3 (31.9)	65.7 (30.2)	68.4 (29.2)	62.3 (31.1)	0.285
Illicit drug abuse ever	59 (10.9)	29 (12.1)	12 (9.2)	17 (15.6)	0.64

Note: The independent t test was used to compare means, and in case of asymmetry, the Mann-Whitney or Kruskal-Wallis test was used.

^aChi-squared or Fisher exact test was applied for categorical variables.

Abbreviations: PrEP, pre-exposure prophylaxis; STI, sexually transmitted infection.

The average age was significantly lower among PrEP nonusers than in PrEP users (32.1 vs. 36.3 y, p <0.001). There was no significant intergroup difference in STI status at enrollment, except for HAV antibody positivity (9.1 vs. 16.3%, p = 0.005). With regard to the details of the illicit drugs, most of them were inhaled alkyl nitrites or marijuana and little methamphetamine were used. Figure 2 illustrates that the Kaplan-Meier curve identified a significant increase in HBV incidence among PrEP nonusers (p = 0.018, log-rank test; 21 infections, annual incidence 3.8%, in 546 PrEP nonusers, with 559.5 person-years) compared with PrEP users [1 infection, annual incidence 0.45%, in 240 PrEP users, with 223.1 person-years (1 infection, annual incidence 0.77% in 131 daily PrEP users, with 129.3 person-years; 0 infection in 109 ED-PrEP users, with 93.8 person-years)]. We confirmed that the HBV-infected case among daily PrEP users had 0 adherence. A total of 130 (23.8%) PrEP nonusers (median follow-up time, 201.5 d)and 68 (28.3%) PrEP users (median follow-up time, 161 d) were censored because of HBsAb levels ≥10 mIU/mL (p = 0.178).

FIGURE 2: Kaplan–Meir curve to estimates incidence of HBV infection among MSM with and without PrEP. Abbreviations: ED, event-driven; MSM, men who have sex with men; PrEP, pre-exposure prophylaxis.

Figure 3 showed that HBV and HIV incidence decreased with PrEP use. In contrast, the incidence of other STIs, including syphilis, chlamydiosis, and gonorrhea showed tendencies to increase with PrEP use though there were no significant intergroup differences in STIs prevalence at enrollment. Noteworthily, not only the daily PrEP but also the ED-PrEP users showed a higher incidence of all STIs than the PrEP nonusers in this study population, although ED-PrEP was sometimes referred as to an option for MSM who have sex less frequently.^18–20 The results demonstrated that HBV infection was prevented in both the daily PrEP users and the ED-PrEP users whose risk of STI and HBV was high. Also, these users did significantly worse with the average rate of condom use after enrollment than PrEP nonusers (65.6% in PrEP nonusers and 48.1% in PrEP users, p < 0.001), whereas there were no significant differences at enrollment. Those increased risks in the PrEP users reflect so-called “risk compensation.”

FIGURE 3: Incidence rate of HBV, HIV, and bacterial STIs with and without PrEP use. Abbreviations: PrEP, pre-exposure prophylaxis; STI, sexually transmitted infection.

Table 2 identifies the preventive and risk factors estimated by the Cox proportional hazards regression. We were unable to perform multivariable analysis due to an insufficient number of events for a statistically sound analysis. The univariate analysis identified age, per year increase (HR: 0.94, 95% CI, 0.883–0.996, p = 0.037) and PrEP use (HR: 0.13, 95% CI, 0.017–0.938, p = 0.043) as significant preventive factors, and bacterial STIs at enrollment (HR: 3.36, 95% CI, 1.455–7.758, p = 0.005) and the number of sexual partners at enrollment (HR: 1.01, 95% CI, 1.001–1.010, p = 0.011) as significant risk factors for HBV infection.

TABLE 2 - Factors associated with acute HBV infection in the univariate analyses

	Univariate analysis
Variables	HR	95% CI	p a
Age, per year increase	0.94	0.883–0.996	0.037
Use of PrEP	0.13	0.017–0.938	0.043
Bacterial STI at enrollment	3.36	1.455–7.758	0.005
No. sexual partners at enrollment	1.01	1.001–1.010	0.011
Average rate of condom use at enrollment	1	0.988–1.015	0.855
Illicit drug abuse ever	0.78	0.183–3.353	0.742

^ap was calculated using logistic regression.

Abbreviations: PrEP, pre-exposure prophylaxis; STI, sexually transmitted infection.

The details of the 22 participants who developed HBV infection are shown in Table 3. The only case of HBV infection among the daily PrEP users was a participant who did not take any TDF/FTC. He had severe hepatitis and was hospitalized. None of the participants with HBV infection developed fulminant hepatitis; 2 (9.1%) progressed to chronic HBV infection, of whom 1 was taking steroids after renal transplantation. The median interval from the last negative test to the first positive test for HBV serology was 94.5 days (56–630 d). Twenty participants achieved HBsAg clearance, and the interval range was 56–630 days. Nine cases in which genotypes were successfully identified were all genotype A. The prevalence and incidence of bacterial STIs were 11 (50%) and 7 (31.8%), respectively. The details of HBV serology are shown in Table 4. Among the 13 cases with HBsAg positive, 1 case was asymptomatic and underwent steroid therapy after renal transplantation. Of 21 cases who developed HBV infection in the PrEP nonusers, 9 (42.9%) asymptomatic cases with HBs Ag negative were observed.

TABLE 3 - Characteristics of the 22 cases with acute HBV infection

										Bacterial STI
Case	Age (y)	PrEP	Observational period (d)	Interval from last negative to first positive test (d)	HBsAg/HBsAb/HBcAb	Genotype/HBV DNA (log IU/mL)	HBsAg status	T-Bil/ALT/PT-INR	Symptom	Prevalence	Incidence
1	25	0	175	91	−/+/+	NA	Clear	NA	No	No	Yes
2	22	Daily PrEP (adherence 0%)	77	77	+/−/+	A/7.5	Clear Aa	3.4/3728/1.74	Yes	No	Yes
3	22	0	1095	98	+/−/+	A/4.9	Clear	1.8/1389/0.95	Yes	Yes	Yes
4	36	0	465	101	+/−/−	Ae/5.0	Clear	5.6/2842/1.09	Yes	No	No
5	39	0	168	84	−/−/+	NA	Clear	NA	No	Yes	No
6	18	0	105	105	−/−/+	ND	Clear	3.1/3102/1.19	Yes	Yes	No
7	27	0	59	59	+/−/+	A/8.3	Clear	2.7/2145/1.16	Yes	No	No
8	30	0	472	98	−/+/+	NA	Clear	NA	No	Yes	No
9	55	0	336	56	−/+/+	NA	Clear	NA	No	No	No
10	34	0	172	84	+/−/+	Ae/2.8	Clear	NA	Yes	No	No
11	30	0	203	98	+/−/+	Ae/3.4	Clear A/ETVb	NA	Yes	Yes	No
12	22	0	304	183	−/+/+	ND/1.0	Clear	NA	No	Yes	Yes
13	23	0	64	64	+/−/−	ND/3.5	Clear A/HAVc	4.9/3729/1.42	Yes	No	No
14	31	0	84	84	+/−/+	Ae/8.7	Clear	16.8/3956/1.02	Yes	No	No
15	36	0	382	183	+/−/+	A/6.5	Clear	18.4/3935/1.27	Yes	Yes	No
16	24	0	630	630	−/+/+	NA	Clear	NA	No	Yes	No
17	27	0	221	178	−/+/+	NA	Clear	NA	No	Yes	No
18	24	0	410	84	+/−/−	NA/6.1	Chronic/postrenal transplantation Steroid use	NA	No	No	Yes
19	23	0	214	63	+/−/+	A/8.9	Chronic	3.7/1849/1.31	Yes	Yes	Yes
20	31	0	273	112	+/−/−	NA/6.5	Clear	10.9/4193/1.29	Yes	No	No
21	19	0	91	91	−/−/+	NA	Clear	NA	No	No	Yes
22	36	0	196	112	−/+/+	NA	Clear	NA	No	Yes	No

^aHospitalized.

^bEntecavir was used.

^cCoinfection with acute HAV.

Abbreviations: ALT, alanine aminotransferase (U/L); NA, not available; ND, not detected; PrEP, pre-exposure prophylaxis; PT-INR, prothrombin time-international normalized ratio; STI, sexually transmitted infection; T-Bil, total bilirubin (mg/dL).

TABLE 4 - Detail of HBV serology according to PrEP use

	HBs Ag not detected (n)	HBs Ag detected (n)
Daily PrEP use
Anti-HBc Ab SC with hepatitis	0	1
Anti-HBc Ab SC without hepatitis	0	0
No anti-HBc Ab SC	130	0
Event-driven PrEP use
Anti-HBc Ab SC with hepatitis	0	0
Anti-HBc Ab SC without hepatitis	0	0
No anti-HBc Ab SC	109	0
No PrEP use
Anti-HBc Ab SC with hepatitis	0	11
Anti-HBc Ab SC without hepatitis	9	1
No anti-HBc SC	525	0

Abbreviations: Ab, antibody; PrEP, pre-exposure prophylaxis; SC, seroconversion.

As a substudy, we evaluated the incidence of HBV infection among MSM who were HBsAb positive and HBcAb negative and did not use PrEP. Three (1.8%/y, 95% CI, 0.37–5.2%/y) asymptomatic incident HBV infections occurred among 144 MSM with 167.5 person-years of follow-up. The HBsAb titers before the infection in these 3 participants were 910.8, 47.0, and 98.1 mIU/mL, respectively. None of them was HBsAg positive. No reactivation of HBV infection was observed among 220 (283.1 person-years) HBsAb-positive and HBcAb-positive participants (0%/y, 95% CI, 0–1.30%/y) or among 23 (21.0 person-years) HBsAb-negative and HBcAb-positive participants (0%/y, 95% CI, 0–16.8%/y).

DISCUSSION

This study showed that tenofovir-based PrEP prevented HBV infection in HIV-negative MSM in a real-world setting. Notably, both daily PrEP and ED-PrEP use showed a preventive effect despite the higher risk of STIs compared with MSM who did not use PrEP. The prophylactic effect of PrEP was ~90% based on the incidence rate ratio, which is equivalent to the prophylactic effect reported in the previous studies among MSM with HIV infection.^6–8 This study illustrates the preventive effect of PrEP against HBV infection in a real-world clinical setting, wherein PrEP adherence is considered suboptimal compared with antiretroviral therapy. As PrEP adherence varies drastically by country and area,^21,22 the results may not be universally applicable. However, if PrEP adherence is optimal, ED-PrEP could prevent HBV infection, although further prospective studies are required to confirm this aspect. We previously reported that, in a feasibility study in Tokyo, the adherence to PrEP was as high as 95%.²³ The reason for this high adherence rate is possibly attributable to the limited access to PrEP and increased awareness of PrEP among MSM in Japan. As of March 2022, PrEP had not been approved in Japan and there was no public provision for PrEP. Thus, MSM who seriously seek PrEP purchase generic drugs through private clinics or the Internet at their own expense. However, if adherence deteriorates, the prophylactic effect against HBV infection will be undermined. In fact, a daily PrEP user in the study who confessed to the absence of adherence was infected with HBV and was hospitalized for severe acute HBV infection.

Interestingly, the incidence of STI during the observational period was unrelated to the incidence of HBV infection, which can be explained by the preventive effect of PrEP against HBV infection and previously reported “risk compensation.”^24,25 In this cohort, both daily PrEP and ED-PrEP were associated with a decreased incidence of HIV and HBV infection and an increased incidence of other STIs. Although ED-PrEP were sometimes thought to be used among MSM with less frequent sexual intercourse,^18–20 ED-PrEP users in our study cohort had a higher risk of bacterial STIs than PrEP nonusers. This suggests that not only daily PrEP use but also ED-PrEP use had a prophylactic effect against HBV infection. None of the study participants had frequent switches between daily and ED-PrEP use. Only 13.9% of the PrEP users in the cohort experienced switches of PrEP.

One of the strengths of this study is the prospective nature of the cohort study, which enables the estimation of the concise incidence rate of HBV and other STIs, including asymptomatic infections, due to regular quarterly testing to ascertain HBV serostatus. Previous evidence on the prophylactic effect of TDF/FTC against HBV infection is limited to retrospective data, although they covered a longer time period than that in our study. Despite the shorter follow-up period, this study demonstrated the preventive effect of PrEP in a real-world clinical setting in a prospective cohort. Furthermore, this study illustrated the details of cases of acute HBV infection, including asymptomatic infection, which has rarely been reported in a research study. Although the 22 cases of HBV infection are inadequate to provide the exact incidence or clearance rate of acute HBV infection, we can extrapolate from epidemiological data on acute HBV infection among the MSM population, where the genotype is dominantly Ae.²⁶ In this regard, the chronicity rate of HBsAg in the current study was 9.1%, which was lower than the 36.4% rate among cases with acute HBV infections with genotype A in a previous Japanese study.²⁷ Furthermore, a strength of the study is that prophylactic HBV vaccination was also provided in the substudy. The incidence of HBV infection was 1.8%/y among participants who were HBsAb positive and HBcAb negative, which was 47.4% lower than the incidence in PrEP nonusers (3.8%/y). Although this incidence may seem higher than generally expected and that in the PrEP users, all 3 cases of HBV infection were asymptomatic, which suggests that HBV vaccination may have prevented symptomatic hepatitis. Given that the only case HBV infection in the daily PrEP was in a participant who did not take PrEP and developed severe hepatitis, vaccination is still important among PrEP users, as it does not require adherence.

Nonetheless, this study has several limitations. First, it was an observational study that could not control the sample size for comparisons; however, although the sample size of PrEP users was smaller than that of PrEP nonusers, the higher incidence of bacterial STIs in the PrEP user group than among PrEP nonusers, compensated for the lower incidence of HBV infection in the PrEP users. Second, this was a single-center study, which makes it difficult to generalize the results. However, services for PrEP follow-up tests and STI tests for MSM had been limited, except for the SHC, in Japan until a private clinic started providing generic PrEP medication and follow-up tests without national health insurance coverage in May 2020. The number of MSM enrolled in the SHC to seek free STI tests for MSM reached almost 1600 by December 2020. Thus, the results of this study are likely to be representative of the epidemiology of STIs among MSM in the Tokyo metropolitan area. Third, not all PrEP users provided complete data on adherence to PrEP. Our previous PrEP feasibility study with 124 MSM in SHC revealed a >95% adherence rate.²⁵ A similarly high adherence rate might be expected in the remainder of the PrEP users as well, as the provision of PrEP has been quite limited, whereas the awareness of PrEP has rapidly increased in Japan. However, evidence on the level of adherence required to prevent acute HBV infection is of interest, and further studies are required. Fourth, the difference in the protective effect of TDF and TAF was not evaluated in the study due to the limited sample size. Of the daily PrEP users, 57.4% used generic TAF/FTC, and all the ED-PrEP users used generic TDF/FTC. We believe that it is acceptable not to distinguish them in the analysis, given that the daily use of TDF or TAF are both theoretically effective against HBV.

In conclusion, the present study demonstrated the prophylactic effect of tenofovir-based PrEP against HBV infection among MSM in a real-world setting. Nonetheless, further studies are required in other settings, where PrEP adherence may be suboptimal.