Treatment Responses to Integrase Strand-transfer Inhibitor-containing Antiretroviral Regimens in Combination With Short-course Rifapentine-based Regimens for Latent Tuberculosis Infection Among People With HIV

Abstract

Background

Real-world experience with combinations of short-course rifapentine-based regimens and integrase strand-transfer inhibitor–containing antiretroviral therapy (ART) in management of latent tuberculous infection (LTBI) is limited among people with human immunodeficiency virus (PWH).

Methods

From August 2019 to October 2022, PWH receiving 3 months of weekly rifapentine plus isoniazid (3HP) or 1 month of daily rifapentine plus isoniazid (1HP) in combination with ART were included. The primary outcome was virologic response within 12 months after LTBI treatment; secondary outcomes included treatment completion rate and safety of LTBI regimens.

Results

During the study period, 479 PWH (94.6% male; median age, 43 years) were included: 142 received 1HP and bictegravir (BIC)-containing regimens (1HP/BIC group), 46 1HP and dolutegravir (DTG)-containing regimens (1HP/DTG group), 38 3HP and BIC-containing regimens (3HP/BIC group), 214 3HP and DTG-containing regimens (3HP/DTG group), 17 1HP and other ART regimens (1HP/others group), and 22 3HP/other ART regimens (3HP/others group). In the intention-to-treat analysis, the proportions of PWH maintaining plasma HIV-1 RNA <200 copies/mL within 12 months after LTBI treatment completion were 96.5% (1HP/BIC), 100% (1HP/DTG), 100% (3HP/BIC), 95.8% (3HP/DTG), 100% (1HP/others), and 100% (3HP/others). The overall completion rates were >80% for all treatment groups, whereas >50% of the included PWH experienced any adverse event. LTBI regimens and ART combinations were not associated with virologic response and completion rate.

Conclusions

Combinations of short-course rifapentine-based regimens and integrase strand-transfer inhibitor-containing ART maintained viral suppression for most PWH within 12 months of LTBI treatment completion with low rates of grade 3 or higher adverse events.

In 2021, there were an estimated 703 000 incident tuberculosis (TB) cases with 187 000 deaths among people with human immunodeficiency virus (HIV; PWH) [1]. PWH are at an elevated risk of developing active TB disease and reactivation of latent TB infection (LTBI) because of their weakened immune systems. To address this issue, the World Health Organization strongly recommends that PWH undergo systematic testing and receive treatment for LTBI [2]. Despite global progress in TB prevention for PWH, additional efforts are required to achieve the goal of delivering TB preventive treatment to 90% of PWH by 2025 [1].

Obstacles to managing LTBI among PWH include motivation issues, low confidence of clinicians, adverse reactions, and lengthy treatment duration [3]. Our study on the TB care cascade within a national program for managing LTBI among PWH showed that the major gap in care engagement involved the initiation of LTBI treatment [4]. Given that single-tablet integrase strand-transfer inhibitor (InSTI)-containing antiretroviral therapy (ART) is the preferred first-line HIV treatment, healthcare providers may hesitate to initiate LTBI treatment because of concerns about limited clinical data on the effectiveness of InSTI-based ART, adverse effects, and drug interactions when coadministering it with LTBI treatments.

Two primary forms of LTBI treatment for presumed drug-susceptible infection are isoniazid for at least 6 months (isoniazid preventive therapy [IPT]) and regimens containing a rifamycin (rifampicin or rifapentine). Although IPT has been the most widely used form of LTBI treatment, the shorter duration of rifamycin-based regimens provides a notable advantage [5]. The updated guidelines prefer IPT and short-course, rifamycin-based, 3- or 4-month regimens for LTBI treatment [2, 6], with an alternative option of 1 month of daily rifapentine plus isoniazid (1HP) [2]. However, rifamycins are potent inducers of hepatic drug-metabolizing enzymes, leading to a significant increase in the clearance of antiretroviral agents [7]. Therefore, guidelines suggest rifamycin-based regimens should be prescribed to PWH who are on ART when drug interactions allow. Based on a phase 1/2 trial conducted among PWH, no dose adjustment is required for dolutegravir (DTG) when coadministered with 3 months of weekly rifapentine plus isoniazid (3HP) [8]. Pharmacokinetic and randomized studies showed that 92% of PWH on DTG-containing regimen remained virologically suppressed when coadministered with 1HP [9, 10]. Although pharmacokinetic studies discouraged coadministering once-daily bictegravir (BIC) with rifamycins due to a substantial reduction in BIC trough concentrations [11, 12], a prospective study demonstrated that HIV viral suppression could be reachieved in PWH concurrently receiving BIC-containing regimen and 1HP, even with transient viremia observed during LTBI treatment [13]. These findings suggest that short-course rifapentine-based regimens for LTBI treatment may not compromise the HIV control in most PWH who are on DTG- or BIC-containing ART after completion of LTBI treatment.

With the recent ART scale-up and widespread InSTI use [14], more research on the coadministration of short-course rifapentine-based regimens and InSTI-containing ART is warranted. In this study, we aimed to evaluate the virologic response and tolerability among PWH who concurrently received short-course rifapentine-based regimens and BIC or DTG-containing ART.

METHODS

Study Setting and Population

Considering that Taiwan reported a TB incidence rate of 30.1 cases per 100 000 people in 2021, the Taiwan LTBI program has been extended to include PWH since 2019 [15]. The designated hospitals for HIV care provide PWH with free-of-charge interferon-gamma release assay (IGRA). Those testing positive or indeterminate (low mitogen response) for IGRA are counseled to undergo directly observed therapy (DOT) for LTBI treatment after ruling out active TB by clinical symptomatology and chest radiography. The LTBI regimens consist of 1HP, 3HP, 3HR (3 months of daily rifampicin plus isoniazid), 4R (4 months of daily rifampicin), or 9H (9 months of daily isoniazid) [4]. For those eligible for LTBI treatment, the regimen is determined after shared decision-making between PWH and their treating physicians. Because the coadministration of rifapentine or rifampicin may lead to decreased plasma concentrations of antiretroviral drugs and require dose adjustment, physicians informed PWH of the effect of rifamycins on antiretroviral concentrations. Following the recommendations of the Taiwan Centers for Disease Control, physicians provided the option of either switching their antiretroviral regimens or continuing with their current ones, with close monitoring of virologic response during LTBI treatment. This decision was based on the available pharmacokinetic and clinical data [16]. PWH were required to provide written informed consent for both IGRA testing and LTBI treatment per the regulations implemented by the Taiwan Centers for Disease Control. During 2019–2022, the rates of LTBI screening and treatment uptake were more than 70% in the LTBI program.

This multicenter, retrospective cohort study was carried out at 16 designated hospitals located in various downtown areas around Taiwan between August 2019 and October 2022 to include PWH aged 20 years or older who had been receiving ART along with 1HP or 3HP. The study did not influence the decision-making process regarding the selection of ART and LTBI treatment. PWH with a history of LTBI or TB treatment were excluded. HIV case managers used in-person or smart-phone video-based DOT to oversee treatment adherence and obtain information on any adverse event (AE) encountered. Clinical assessment and laboratory investigations were conducted every 2 to 4 weeks during the first month, and subsequently every 4 weeks until the end of LTBI treatment. Monitoring of plasma HIV RNA load (PVL) and CD4 count was carried out every 3 to 6 months per the national HIV treatment guidelines [17].

This study was approved by the Research Ethics Committee or institutional review boards of the participating hospitals and the requirement for obtaining informed consent to gather clinical data for subsequent anonymous analysis was waived (Supplementary Material).

Outcomes

The primary outcome was virologic response (PVL <200 copies/mL) within 12 months after LTBI treatment, which was evaluated by means of intention-to-treat (ITT) and per-protocol (PP) analyses. The time point selected for evaluating virologic response within the 12-month time frame was close to the end of the follow-up period. In the ITT analysis, PWH who did not continue follow-up for virologic response to ART were classified as having virologic failure, whereas those who did not complete LTBI treatment but achieved PVL <200 copies/mL were still considered to have achieved virologic responses. PWH who did not complete LTBI treatment or follow-up were excluded from the PP analysis. The secondary outcomes were virologic response during LTBI treatment and at 3 to 6 months after LTBI treatment, completion rate, as well as safety of LTBI regimens. Treatment completion was defined as receiving at least 80% of doses within 120% of the scheduled time, and the reasons for discontinuation or treatment modification were recorded. AEs were documented and graded [18], and an AE was classified as serious if it led to death, a life-threatening event, hospitalization, and persistent or significant disability or incapacity [19].

Laboratory Investigations

QuantiFERON-TB Gold In-Tube assay was initially used for LTBI screening, which was replaced with the QuantiFERON-TB Gold Plus assay (Qiagen, Germantown, MD, USA) in late 2021 [20, 21]. IGRA was conducted using the same kit at each participating hospital. The results were interpreted according to the instructions of the manufacturer. The pooled estimates of sensitivity and specificity were 91.4% and 98.7% for the QuantiFERON-TB Gold In-Tube assay and 91.4% and 97.8% for the QuantiFERON-TB Gold Plus assay, respectively [22].

Statistical Analyses

The χ² test or Fisher exact test was used to analyze categorical variables, whereas the Wilcoxon-Mann-Whitney test was used to compare continuous variables. Logistic regression was used to determine the factors associated with virologic response and AEs. Variables that had a P value <.05 in univariable analyses were incorporated in the maximum model for multivariable analyses, and a backward selection was implemented to identify the final model. All statistical tests were two-sided, and variables with a P value <.05 were deemed significant. All statistical analyses were performed using STATA software version 17.0 (Stata Corporation, College Station, TX).

RESULTS

Clinical Characteristics of the Included PWH

During the study period, 479 PWH received 1HP or 3HP treatment after testing positive (467, 97.5%) or indeterminate (12, 2.5%) for IGRA; they were predominantly male (94.6%) with a median age of 43 years (Table 1). The main HIV transmission route was male-to-male sexual contact (65.3%), followed by illicit drug use (25.5%) and heterosexual contact (9.4%). The included PWH had a median CD4 count of 650 cells/mm³, with 97.9% having achieved PVL <200 copies/mL before LTBI treatment. Among them, 205 (42.8%) and 274 (57.2%) PWH received 1HP and 3HP treatment, respectively. Compared with PWH receiving 1HP, those receiving 3HP were more likely to be people who inject drugs, have anti-hepatitis C virus positivity, and initiate LTBI treatment while incarcerated; additionally, they were less likely to have hepatitis B surface antigen positivity and cerebrovascular diseases.

Although 440 PWH (91.9%) had been receiving BIC or DTG-containing ART, the remaining 39 (8.1%) were on different ART regimens. Overall, 142 PWH received 1HP concomitantly with BIC-containing regimens (1HP/BIC group), 46 1HP and DTG-containing regimens (1HP/DTG group), 38 3HP and BIC-containing regimens and (3HP/BIC group), 214 3HP and DTG-containing regimens (3HP/DTG group), 17 1HP/other ART regimens (1HP/others group), and 22 3HP/other ART regimens (3HP/others group).

Virologic Outcomes

The median interval between completion/discontinuation of LTBI treatment and first virologic follow-up was 3.0 months (interquartile range [IQR], 2.5–4.9) for the 1HP group and 4.2 months (IQR, 2.0–10.0) for the 3HP group. In the ITT analysis, the rates of PVL <200 copies/mL within 12 months after LTBI treatment completion were 96.5% (137/142) for the 1HP/BIC group, 100% (46/46) 1HP/DTG group, 100% (38/38) 3HP/BIC group, 95.8% (205/214) 3HP/DTG group, 100% (17/17) 1HP/others group, and 100% (22/22) 3HP/others group (Figure 1, Supplementary Table 1). After excluding 41 PWH who did not complete LTBI treatment and 10 who were lost to follow-up, the PP analysis showed that the rates of maintaining PVL <200 copies/mL within 12 months after LTBI treatment were similarly high for all study groups: 100% (131/131) for the 1HP/BIC group, 100% (40/40) 1HP/DTG group, 100% (37/37) 3HP/BIC group, 99.5% (183/184) 3HP/DTG group, 100% (14/14) 1HP/others group, and 100% (22/22) 3HP/others group. In multivariable analysis, the factor associated with failure to maintain PVL <200 copies/mL within 12 months after LTBI treatment in the ITT analysis was incarceration (adjusted odds ratio [AOR], 12.67; 95% confidence interval [CI], 2.92–54.98) (Table 2). However, LTBI regimens and concurrent ART were not associated with virologic response.

Additionally, more than 90% of PWH maintained PVL <50 copies/mL within 12 months after LTBI treatment. In both the ITT and PP analyses, PWH receiving 1HP and 3HP had similar virologic responses, 95.6% for 1HP and 94.5% 3HP in the ITT analysis (P = .604), and 98.4% for 1HP and 97.5% 3HP in the PP analysis (P = .575). Furthermore, more than 90% of included PWH maintained virologic responses (PVL <50 or 200 copies/mL) during LTBI treatment and 3 to 6 months after LTBI treatment in the PP analysis (Supplementary Table 1); there was no significant difference in virologic responses among all groups. The detailed information of PWH failing to maintain PVL <50 copies/mL during and after LTBI treatment in the PP analysis is summarized in Supplementary Tables 2 and 3. Transient increases in viral load were observed during LTBI treatment in 17 (6.1%) PWH, which did not lead to subsequent treatment failure. Only 1 receiving 3HP and DTG-containing regimen had PVL >200 copies/mL because of poor adherence.

Completion Rate and AEs

The completion rates were similarly high for the 1HP and 3HP groups (92.7% vs 90.5%, P = .401) (Table 3). Among 41 PWH who did not complete LTBI treatment, 34 (82.9%) were due to AEs leading to discontinuations and 7 (17.1%) regimen modifications. The median time to discontinuation was 21 days (IQR, 12–57 days). In multivariable analysis, treatment noncompletion was associated with older age (AOR, 1.04; 95% CI, 1.01–1.08) and male sex (AOR, 0.27; 95% CI, .10–.77) (Supplementary Table 4). No TB cases were reported during the study period.

The rates of PWH who encountered any AE were comparable between the 1HP and 3HP groups (60.0% vs 55.1%, P = .284) (Table 3). The most reported AEs were flu-like symptoms (39.2%) (Figure 2). More dermatologic AEs (17.6% vs 9.1%) and new hepatic events (17.1% vs 10.6%) occurred in the 1HP group than in the 3HP group. The AEs were mainly of grade 1 (39.2%) or 2 (12.9%) in severity, and only 1 individual had grade 4 hepatotoxicity related to acute hepatitis C. Grade 3 AEs, predominantly flu-like symptoms, were more frequently reported in the 1HP group than in the 3HP group (Supplementary Table 5). In multivariable analysis, the factors predicting the occurrence of AEs were incarceration (AOR, 4.17; 95% CI, 1.36–12.80) and concurrent use of 3HP and BIC-containing regimen (AOR, 0.11; 95% CI, .03–.34) (Supplementary Table 4).

DISCUSSION

In this study evaluating the virologic response and tolerability among PWH receiving short-course rifapentine-based regimens, >95% of PWH receiving BIC or DTG-containing regimens remained virally suppressed after completion of LTBI treatment, though transient viremia was observed during the LTBI treatment course. Although AEs were common, most AEs were tolerable and treatment completion rates were similarly high in the treatment combination groups. More dermatologic AEs and new hepatic events were reported in the 1HP group than in the 3HP group. LTBI regimens and ART were not associated with virologic response and completion rate.

Rifapentine coadministration decreases plasma concentrations of InSTIs by inducing cytochrome P450 3A and uridine diphosphate glucuronosyltransferase 1A1 [23]. According to pharmacokinetic data, dose adjustment of DTG was not needed in combination with weekly rifapentine, but once-daily bictegravir/emtricitabine/tenofovir alafenamide (BIC/FTC/TAF) alongside weekly rifapentine was not recommended because of a significant decrease in BIC trough concentrations (35%–85%) [8, 12]. However, prior studies demonstrated that PWH who received both 3HP and DTG-containing regimens, as well as those who received 1HP and BIC-containing regimens, were able to maintain viral suppression following LTBI treatment [8, 13]. This virologic outcome may be attributed to the higher genetic barrier of BIC and DTG as well as relatively short exposure durations to suboptimal concentrations. In a phase 1/2 trial, a 36% increase in DTG clearance was observed among 61 PWH concurrently receiving DTG-containing regimens and 3HP; however, only 1 had trough concentration below the 90% maximal inhibitory concentration for DTG and all participants maintained viral suppression during 3HP treatment [8]. In another prospective study of 48 PWH concurrently receiving 1HP and once-daily coformulated BIC/FTC/TAF, the proportion of maintaining BIC trough concentrations above the 95% effective concentration dropped to 37%. During LTBI treatment, 33% participants had transient viremia; however, all participants reachieved viral suppression 3 and 6 months after completion of 1HP [13]. Furthermore, intracellular tenofovir-DP levels remain high, although coadministrating FTC/TAF with rifampicin decreases TAF plasma exposure [16]. Our study observed that approximately 6% of PWH experienced transient viremia during LTBI treatment, but most reachieved viral suppression following LTBI treatment. Most PWH with failure to maintain virologic responses (71%) were people who inject drugs and/or had an incarceration history, which may contribute to reduced adherence. Although further pharmacokinetic and clinical studies are warranted to support the combination use of InSTI-containing ART and rifapentine, our findings suggest that both BIC- and DTG-containing regimens may be considered acceptable for coadministration with either 1HP or 3HP while closely monitoring the virologic response.

Short-course preventive therapy promotes the uptake and completion of treatment for at-risk populations with LTBI [24–26]. In a randomized trial enrolling PWH concurrently receiving 1HP and nonnucleoside reverse transcriptase inhibitor–containing regimens, a significantly higher completion rate was observed with 1HP compared with 9H (97% vs 90%) [27]. Despite a high incidence of AEs noted in our study, prior studies have shown a lower risk of AEs associated with short-course preventive therapy compared with IPT [5, 27]. Although the data comparing the safety of 1HP and 3HP were limited, a recent meta-analysis revealed that the proportion of grade 3 and 4 liver toxicity was higher in those receiving 1HP compared with 3HP, whereas the frequency of serious AEs was more common in those receiving 3HP [28]. Our study also showed that both 1HP and 3HP regimens were well tolerated among PWH; however, more PWH receiving 1HP experienced cutaneous reactions, liver toxicity, and grade 3 flu-like symptoms than those receiving 3HP. Although AEs may be reported more frequently among those receiving 1HP and experiencing incarceration as a result of close monitoring and may be reduced with the combination use of 3HP and BIC-containing ART, AEs still require enhanced surveillance, particularly for those with relevant underlying medical conditions.

Multilevel interventions, including improving patient–provider communication and patient literacy, providing IGRA testing, and using short-course rifapentine-based regimens, should be implemented to enhance optimization and reduce TB-related mortality [29, 30]. In our previous study evaluating the care cascade of LTBI treatment among PWH, the LTBI treatment initiation rate was 74% and the completion rate 92% [4]. Concerns regarding AEs and potential drug interactions continued to contribute to hesitancy in initiating LTBI treatment among PWH. Our study showed that treatment noncompletion was primarily attributed to AEs, particularly among older females. Therefore, close monitoring of adherence and AEs should be highlighted in the populations.

Our study has several limitations. First, the clinical characteristics were not balanced between the study groups in this retrospective observational study; nevertheless, LTBI regimens and concurrent ART were unrelated to virologic response and treatment completion after adjusting for covariates in logistic regression analyses. Second, AE-reporting bias might exist, though DOT was used to monitor AEs. Last, most of PVL measurements were performed at regular time intervals according to clinical care practices but not at fixed time points, and only 58% (279/479) of the included PWH underwent PVL testing during the LTBI treatment course. In addition, antiretroviral resistance testing was not performed for those with viremia. Although infrequent PVL measurements and the absence of resistance testing results may pose a risk of virologic failure, our study demonstrated that viral suppression could be achieved after completion of LTBI treatment in most of the included PWH.

In conclusion, we demonstrated that combinations of BIC or DTG-containing regimens with short-course rifapentine-based regimens may be considered acceptable for maintaining viral suppression, with exhibiting a favorable safety profile. The high completion rate observed with short-course rifapentine-based regimens for LTBI among PWH has the potential to enhance LTBI treatment outcomes in this population.

Supplementary Data

Supplementary materials are available at Clinical Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Author contributions. K. Y. L., C. Y. C., and C. C. H. managed and supervised the study. K. Y. L. and C. C. H. contributed to the study concept and design. K. Y. L., H. Y. S., C. J. Y., P. L. L., Y. T. L., N. Y. L., B. H. L., H. J. T., M. H. L., N. C. W., T. C. C., I. M. H., A. T. P., C. Y. L., C. S. T., C. Y. C., and C. C. H. were involved in collection and assembly of clinical data. K. Y. L., H. Y. S., and C. C. H. participated in data analysis. K. Y. L., C. Y. C., and C. C. H. undertook interpretation of the data and drafted the report. All authors reviewed and approved the final version of the report.

Acknowledgments. The authors thank all study members participating in the Taiwan HIV Study Group. The names of the members of the Taiwan HIV Study Group are listed here:

Chien-Ching Hung, Sui-Yuan Chang, Hsin-Yun Sun, Yu-Shan Huang, Kuan-Yin Lin, Guan-Jhou Chen, Pei-Ying Wu, Ling-Ya Chen, Hsi-Yen Chang, Wen-Chun Liu, and Yi-Ching Su (National Taiwan University Hospital); Ning-Chi Wang and Te-Yu Lin (Tri-Service General Hospital and National Defense Medical Center); Chia-Jui Yang and Mao-Song Tsai (Far Eastern Memorial Hospital); Mei-Hui Lee (Shuang Ho Hospital); Yi-Chieh Lee (Lotung Poh-Ai Hospital); Shu-Hsing Cheng and Chien-Yu Cheng (Tao-Yuan General Hospital, Ministry of Health and Welfare); Bo-Huang Liou (Hsinchu MacKay Memorial Hospital); Sung-Hsi Huang, Yi-Chia Huang, and An-Ting Peng (National Taiwan University Hospital Hsin-Chu Branch); Yuan-Ti Lee, Yu-Lin Lee, and Chia-Chun Lin (Chung Shan Medical University Hospital); Shih-Ping Lin, Chia-Yin Hsieh, and Hsiu-Wen Wang (Taichung Veterans General Hospital); Mao-Wang Ho (China Medical University Hospital); Chun-Eng Liu and Ing-Moi Hii (Changhua Christian Hospital); Chi-Ying Lin (National Taiwan University Hospital Yunlin Branch); Hung-Jen Tang (Chi Mei Medical Center); Nan-Yao Lee and Chin-Shiang Tsai (National Cheng Kung University Hospital); Tung-Che Hung (Ditmanson Medical Foundation Chia-Yi Christian Hospital); Yen-Hsu Chen, Po-Liang Lu, and Chun-Yuan Lee (Kaohsiung Medical University Hospital); Tun-Chieh Chen (Kaohsiung Municipal Ta-Tung Hospital); Hung-Chin Tsai and Hsi-Hsun Lin (Kaohsiung Veterans General Hospital); Chen-Hsiang Lee (Kaohsiung Chang Gung Memorial Hospital).

Financial support. No financial support or grants were received for this study.

References

Author notes