Efficacy of dolutegravir/lamivudine in adults with HIV‐1 and isolated reactive hepatitis B core antibody: Insights from Phase 3/3b study results

Abstract

Objectives

As two‐drug HIV‐1 antiretroviral therapy (ART) regimens become standard, data gaps remain concerning their safety and efficacy among people with isolated reactive hepatitis B core antibodies (anti‐HBc). Here, we assessed the safety and efficacy of dolutegravir/lamivudine (DTG/3TC) versus three‐ or four‐drug regimens (3DRs/4DRs) among people with HIV‐1 and isolated reactive anti‐HBc.

Methods

Data were evaluated from GEMINI‐1, GEMINI‐2, STAT, TANGO and SALSA at various time points, including Weeks 48, 96, 144 and 196. Participants had hepatitis B virus (HBV) exposure, isolated reactive anti‐HBc serology, no HBV infection and were naive to ART (GEMINI‐1/‐2, STAT) or virologically suppressed before switching (TANGO, SALSA). All studies except STAT had a 3DR/4DR comparator.

Results

Of 2798 participants, 76 (DTG/3TC, n = 44; 3DR/4DR, n = 32) had isolated reactive anti‐HBc. Across studies, no instances of HBV reactivation occurred, and liver chemistry elevations were infrequent and generally mild. Among those naive to ART at baseline and receiving DTG/3TC, 78% (18/23) achieved HIV‐1 RNA <50 copies/mL at Week 144 in GEMINI‐1/‐2, and 60% (3/5) achieved HIV‐1 RNA <50 copies/mL at Week 48 in STAT. Among participants with virologic suppression at baseline who switched from 3DR/4DR to DTG/3TC, none experienced HIV‐1 RNA ≥50 copies/mL at the end of TANGO (Week 196) or SALSA (Week 48). HIV‐1 suppression rates were comparable to those of 3DRs/4DRs.

Conclusions

This analysis provides new information about DTG/3TC treatment for people with HIV‐1 and isolated reactive anti‐HBc. There were high rates of HIV‐1 suppression with DTG/3TC, no cases of HBV reactivation and no new safety signals identified.

INTRODUCTION

Simplification to two‐drug regimens (2DRs) for HIV‐1 treatment may be advantageous over three‐ or four‐drug regimens (3DRs/4DRs) to minimize chronic drug exposure, mitigate potential drug–drug interactions and provide overall lower costs, without compromising efficacy [1]. However, tenofovir‐sparing regimens raise specific questions in people with prior hepatitis B virus (HBV) exposure [2].

In Phase 3/3b studies, dolutegravir/lamivudine (DTG/3TC) demonstrated durable efficacy, a high barrier to resistance and a favourable safety profile among participants naive to antiretroviral therapy (ART) through 48 weeks in STAT and 144 weeks in GEMINI‐1/‐2, as well as among participants who were virologically suppressed by previous ART through 48 weeks in SALSA and 196 weeks in TANGO [3, 4, 5, 6]. The efficacy of DTG/3TC in people with HIV‐1 and prior HBV exposure has not been specifically evaluated in clinical trials.

Chronic HBV, defined as presence of hepatitis B surface antigens (HBsAg) for ≥6 months, affects ~300 million people globally. Moreover, ~10% of the ~40 million people with HIV‐1 have HBV co‐infection [2, 7, 8], including 5% to 15% of those in the United States [9]. In ~90% of the cases, acute HBV is cleared spontaneously, leaving reactive hepatitis B core antibodies (anti‐HBc) in some individuals due to waning immunity [8]. A subset of individuals may develop isolated reactive anti‐HBc, which can be associated with low‐level HBV viraemia (i.e., occult HBV infection) [8]. It is estimated that 17–40% of people with HIV‐1 have isolated reactive anti‐HBc serologies [10].

Current guidelines recommend 3DRs for people with HIV‐1 and HBV (i.e., HBsAg positive). Because tenofovir alafenamide (TAF), tenofovir disoproxil fumarate (TDF), emtricitabine (FTC) and 3TC are active against both HIV‐1 and active HBV, an ART regimen for people with HIV‐1 and HBV should include TAF or TDF plus 3TC or FTC as the nucleoside reverse transcriptase inhibitor backbone [9]. Moreover, withdrawal of HBV‐active agents as part of an ART regimen can result in HBV reactivation among people with HIV‐1 and prior HBV exposure [9]. The safety of tenofovir‐sparing 2DRs for HIV‐1 treatment in people with isolated reactive anti‐HBc remains unclear given the potential concerns regarding HBV reactivation and hepatic sequelae [11, 12]. In addition to understanding safety, limited data are available evaluating HIV‐1 suppression rates in patients with isolated reactive anti‐HBc receiving DTG/3TC. A retrospective study showed lower HIV‐1 suppression rates (69% vs. 88%) in people with isolated reactive anti‐HBc (i.e., past HBV exposure) versus non‐reactive anti‐HBc (i.e., no HBV exposure) after switching to DTG/3TC, but it did not account for confounding factors such as age and nadir CD4+ cell count [13].

The aim of our analysis was to evaluate the safety and efficacy of DTG/3TC in participants with HIV‐1 and isolated reactive anti‐HBc in Phase 3/3b studies that were conducted in patient populations who were treatment‐naive or virologically suppressed.

MATERIALS AND METHODS

Pooled analysis

This analysis pooled data from five Phase 3/3b studies evaluating DTG + 3TC (multi‐tablet; GEMINI‐1 and GEMINI‐2) or DTG/3TC (single‐tablet; STAT, TANGO, SALSA) versus 3DRs or 4DRs. Participants with evidence of past HBV exposure (reactive anti‐HBc) with no evidence of active HBV infection (non‐reactive HBsAg and non‐reactive HBV‐DNA) and no evidence of HBV immunity (hepatitis B surface antibodies [anti‐HBs]) were eligible for all studies except STAT, in which participants were enrolled before receipt of baseline laboratory results (including HBV serology test results). Analysis definitions, including HBV criteria and HIV‐1 outcomes, are presented in Table S1.

Study designs

Full study designs have been previously published (Table S2) [3, 4, 5, 6].

• GEMINI‐1/GEMINI‐2: Two identical Phase 3, randomized, double‐blind, multi‐centre studies evaluating DTG + 3TC versus DTG + TDF/FTC in adults naive to ART. Results are reported at Weeks 48, 96 and 144 [6].

• STAT: Phase 3b, open‐label, single‐arm, multi‐centre study evaluating DTG/3TC in adults naive to ART in a test‐and‐treat care model, in which treatment was initiated before analysis of baseline laboratory results. Participants had confirmed HIV‐1 diagnoses within 14 days of screening, no history of hepatic or renal impairment, and no known or suspected HBV co‐infection. Upon receipt of baseline laboratory results, participants with chronic HBV infection were evaluated for potential treatment modification. Results are reported at Week 48 [5].

• TANGO: Phase 3, randomized, open‐label, multi‐centre study comparing switching to DTG/3TC versus continuing TAF‐based regimen in adults with HIV‐1 suppression. Participants who switched to DTG/3TC on Day 1 are referred to as the early‐switch group and those who continued TAF‐based regimen and maintained HIV‐1 suppression at Week 144 and then switched to DTG/3TC at Week 148 are referred to as the late‐switch group. Results are reported at Weeks 48, 96, 144 and 196 [4].

• SALSA: Phase 3, randomized, open‐label, multi‐centre study comparing switching to DTG/3TC versus continuing three‐ or four‐drug current antiretroviral regimen (CAR) in adults with HIV‐1 suppression. Results are reported at Week 48 [3].

Evaluated endpoints

• HBV reactivation and HBV‐related adverse events (AEs).

  • HBV reactivation was defined as a change in HBV‐DNA from negative at baseline (≤2000 IU/mL, including non‐detectable) to positive post‐baseline (>2000 IU/mL) and/or a change in HBsAg from negative at baseline (non‐reactive) to positive post‐baseline (reactive).
  • HBV‐related AE preferred terms included macrocytic anaemia, eosinophilia, lymphadenopathy, lymphocytosis, neutropenia and splenomegaly.

• Liver chemistry tests.

• HIV‐1 virologic outcomes.

Ethics

This is a secondary analysis of previously approved studies conducted in accordance with the Declaration of Helsinki. All participants provided written informed consent for the original studies.

RESULTS

Baseline characteristics

Among 2798 participants across the five studies, 76 (3%) had an isolated reactive anti‐HBc: 3% (46/1433) in GEMINI‐1/‐2, 4% (5/131) in STAT, 2% (13/741) in TANGO and 2% (12/49) in SALSA. Forty‐four and 32 participants received DTG/3TC or 3DR/4DR, respectively. Demographics and baseline characteristics were generally similar across groups (Table S3). HBV vaccination data were not consistently reported across studies.

HBV‐associated AEs

No instances of HBV reactivation occurred during any of the studies. HBV‐associated AEs were infrequent, and participants did not have an increased risk of HBV‐associated AEs while receiving DTG/3TC. In GEMINI‐1/‐2, five participants receiving DTG + 3TC had HBV‐associated AEs of any grade, including eosinophilia (n = 1), lymphadenopathy (n = 2), lymphocytosis (n = 1), macrocytic anaemia (n = 1) and splenomegaly (n = 1). No HBV‐associated AEs were reported in STAT. In TANGO and SALSA, one participant receiving DTG/3TC had an HBV‐associated AE (eosinophilia), as did one receiving CAR (neutropenia).

Liver chemistry

Liver chemistry elevations were infrequent and generally mild (Table 1). In GEMINI‐1/‐2, Grades 1 and 2 elevations were comparable across participants receiving DTG + 3TC versus DTG + TDF/FTC (Grade 1: 26% [6/23] vs. 17% [4/23]; Grade 2: 9% [2/23] vs. 9% [2/23]). No liver chemistry test elevations were reported in STAT. In TANGO and SALSA, Grades 1 and 2 elevations were infrequent and comparable between DTG/3TC versus CAR (Grade 1: 13% [2/16] vs. 11% [1/9]; Grade 2: 6% [1/16] vs. 11% [1/9]). Grade 4 elevations were observed in 9% (2/23) of participants receiving DTG + 3TC in GEMINI‐1/‐2, but none were reported in STAT, TANGO or SALSA. Across all studies, one participant receiving DTG + 3TC in GEMINI‐2 had acute hepatitis E virus (HEV), with liver enzyme elevations that met liver‐stopping criteria at Week 144.

TABLE 1.

Summary of maximum post‐baseline emergent liver chemistry test elevations through end of study. ^a

Parameter, n (%)	GEMINI‐1/‐2		STAT	TANGO/SALSA b
	DTG + 3TC (N = 23)	DTG + TDF/FTC (N = 23)	DTG/3TC (N = 5)	DTG/3TC (N = 16)	CAR (N = 9)
Grade 1	6 (26)	4 (17)	0	2 (13)	1 (11)
Grade 2	2 (9)	2 (9)	0	1 (6)	1 (11)
Grade 3	0	1 (4)	0	0	0
Elevated serum or plasma AST	0	1 (4)	0	0	0
Grade 4	2 (9)	0	0	0	0
Elevated serum or plasma AST	2 (9)	0	0	0	0
Elevated serum or plasma ALT	1 (4)	0	0	0	0

Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; CAR, current antiretroviral regimen; DTG, dolutegravir; FTC, emtricitabine; 3TC, lamivudine; TDF, tenofovir disoproxil fumarate.

^a Includes serum or plasma albumin, alkaline phosphatase, ALT, AST, bilirubin and protein.

^b All Grade 1 elevations were reported in TANGO, and all Grade 2 elevations were reported in SALSA.

HIV‐1 virologic outcomes in participants naive to ART

In GEMINI‐1/‐2, rates of HIV‐1 suppression with DTG + 3TC were similar to those of DTG + TDF/FTC at Week 48 and remained consistent through Week 144. At Week 144, 78% (18/23) of participants receiving DTG + 3TC and 74% (17/23) receiving DTG + TDF/FTC achieved HIV‐1 RNA <50 copies/mL. Evaluating a more stringent virologic threshold of target not detected (TND), HIV‐1 suppression rates remained high and similar between groups (Figure 1a). In STAT, 3/5 (60%) participants receiving DTG/3TC achieved HIV‐1 RNA <50 copies/mL at Week 48. Of the two participants who did not achieve HIV‐1 RNA <50 copies/mL at Week 48, one discontinued and one had a change in ART (both for non–efficacy‐related reasons) and were classified via Snapshot analysis as having HIV‐1 RNA ≥50 copies/mL despite having post‐baseline HIV‐1 RNA <40 copies/mL.

FIGURE 1.

HIV‐1 virologic outcomes in participants with isolated reactive anti‐HBc who were (a) naive to ART at baseline or (b) virologically suppressed before switch to DTG/3TC. ^aIn GEMINI‐1/‐2, 1 participant discontinued DTG + 3TC for non–efficacy‐related reasons by Week 48; by Week 96, three participants had discontinued DTG + 3TC for non–efficacy‐related reasons. ^bData for HIV‐1 RNA <40 copies/mL and TND were not available for STAT. ^cOf the two participants in STAT with HIV‐1 RNA ≥50 copies/mL, one discontinued for non–efficacy‐related reasons (decision by physician; non‐compliance with study procedures) with HIV‐1 RNA ≥50 copies/mL and one had a change in ART that was not efficacy related (decision by participant or proxy due to incarceration). By Week 48, their HIV‐1 RNA measurements were <40 copies/mL. This post hoc analysis was completed using a Snapshot analysis; therefore, any changes in ART classified these participants as having HIV‐1 RNA ≥50 copies/mL, irrespective of the reason. ^dIn TANGO, one participant discontinued DTG/3TC before Week 48 for non–efficacy‐related reasons. ^eIn TANGO, one participant discontinued CAR before Week 144 for non–efficacy‐related reasons. ART, antiretroviral therapy; CAR, current antiretroviral regimen; c/mL, copies/mL; DTG, dolutegravir; FTC, emtricitabine; 3TC, lamivudine; TDF, tenofovir disoproxil fumarate; TND, target not detected; W, week.

HIV‐1 virologic outcomes in participants who were virologically suppressed

No participants experienced HIV‐1 RNA ≥50 copies/mL at the end of TANGO (Week 196) or SALSA (Week 48). Switching to DTG/3TC versus continuing 3DRs or 4DRs resulted in similar rates of HIV‐1 suppression. In TANGO, 89% (8/9) of participants receiving DTG/3TC and 75% (3/4) receiving CAR maintained HIV‐1 RNA <50 copies/mL at Week 144. HIV‐1 suppression rates using the TND measure were similar (Figure 1b).

DISCUSSION

This analysis highlights the safety and efficacy of DTG/3TC in people with HIV‐1 and isolated reactive anti‐HBc who are initiating ART or in the setting of stable switch. Across multiple Phase 3/3b studies, DTG/3TC was not associated with an increased risk of HBV‐associated AEs; there were no cases of HBV reactivation, and liver enzyme elevations were infrequent. DTG/3TC demonstrated high rates of HIV‐1 suppression for ~4 years among people with isolated reactive anti‐HBc who were either naive to ART or virologically suppressed before switching to DTG/3TC.

The use of DTG/3TC for HIV‐1 treatment in people with isolated reactive anti‐HBc has not previously been studied in clinical trials and required further evaluation given the high prevalence of this clinical scenario in this patient population. Additionally, the use of DTG/3TC in this population was deemed a clinical decision‐making challenge with concerns of HBV reactivation in the absence of treatment with dual agents with anti‐HBV activity [14]. Current guidelines recommend regular monitoring of transaminases and/or HBV‐DNA for people who have isolated reactive anti‐HBc when using TAF‐ or TDF‐sparing ART regimens [9, 15]. Although these recommendations are precautionary, findings from this analysis provide reassuring evidence that DTG/3TC effectively suppresses HIV‐1 without an increase in HBV‐related AEs. This study also reinforces the importance of careful interpretation of a patient's hepatitis serologies when initiating or selecting ART, including consideration of the recency of available test results. People with HIV‐1 and an unknown or non‐immune HBV status require HBV screening before use of an agent that does not contain TAF or TDF plus 3TC or FTC [9]. Vaccination should be initiated in individuals without evidence of chronic HBV infection and who are non‐immune to HBV [9].

Our results align with prior findings demonstrating no cases of HBV reactivation and low rates of liver enzyme abnormalities among people who had isolated reactive anti‐HBc when receiving DTG/3TC or other 2DRs that do not include a second agent for HBV [16]. The protective role of 3TC in suppressing HBV activity may explain the absence of HBV reactivation in our analysis, even without a second anti‐HBV agent. Across studies, Grade 3 or 4 liver enzyme elevations were rare. No elevations were observed in any participants from the STAT study when DTG/3TC was initiated in the absence of baseline hepatitis serologies. One participant in GEMINI‐2 receiving DTG + 3TC experienced liver enzyme elevations related to acute HEV that met liver‐stopping criteria per study protocol; however, this event was deemed unrelated to DTG + 3TC.

Although no cases of HBV reactivation were observed in this analysis, prior studies have reported reactivation among people with isolated reactive anti‐HBc receiving 2DRs. Findings from Salpini et al. showed that switching to regimens without TDF or TAF resulted in HBV reactivation in 40% of individuals with isolated reactive anti‐HBc, including those with occult HBV infection and low nadir CD4+ cell count [17]. Welford et al. reported HBV reactivation in 8% (3/38) of people with isolated reactive anti‐HBc after switching to cabotegravir + rilpivirine; however, these reactivations were not associated with elevated liver enzymes or clinical complications, suggesting low‐level HBV activity may remain manageable in the absence of symptoms [18].

Our analysis was limited by the small number of individuals with isolated reactive anti‐HBc. Outcomes were not assessed among participants whose hepatitis serologies were seropositive for anti‐HBc and anti‐HBs, as this population was considered out of scope for this analysis. Additionally, longer‐term outcomes beyond the study durations and the inability to assess vaccination status in all individuals limit the full assessment of long‐term safety and efficacy of DTG/3TC. Future trials should evaluate HBV reactivation risk and clinical outcomes in patients positive for anti‐HBc/anti‐HBs to gain further knowledge about treatment strategies.

CONCLUSIONS

Among people with HIV‐1 and isolated reactive anti‐HBc across five Phase 3/3b studies of DTG/3TC, no HBV reactivation was observed and liver‐related AEs were rare. DTG/3TC demonstrated high rates of HIV‐1 suppression that were comparable to 3DRs or 4DRs, including under stringent TND measures. These findings provide useful insight for clinicians considering DTG/3TC as a treatment option for individuals with HIV‐1 and isolated reactive anti‐HBc. Further research is needed to expand on these findings and optimize clinical management in this unique population.

AUTHOR CONTRIBUTIONS

DF, AD‐M, RAG, RB and BJ contributed to the conception of the study. DF, RAG and BJ contributed to the design of the study. PJ and RAG contributed to the acquisition of data. DF, PJ, RAG and BJ contributed to the analysis of data. DF, JS, ETO, AD‐M, PJ, RAG, SDA, CMP, MU and BJ contributed to the interpretation of data. All authors contributed to drafting the manuscript and to critically revising the manuscript for important intellectual content. All authors have read and approved the final manuscript for publication.

CONFLICT OF INTEREST STATEMENT

DF, ETO, AD‐M, RAG, SDA, RB, CMP, MU and BJ are employees of ViiV Healthcare or GSK and may own stock in GSK. JS has served as the principal investigator for clinical trials, for which his institution received grants, and has received honoraria from AbbVie, Gilead, Janssen, Merck, Theratechnologies and ViiV Healthcare. PJ was a complementary worker on behalf of GSK at the time of the analysis and owns stock in GSK.

ETHICS STATEMENT

Study protocols were reviewed and approved by ethics committees or institutional review boards, and all studies were conducted in accordance with the Declaration of Helsinki. All participants provided written informed consent before study initiation.

Supporting information

Table S1. Analysis definitions [1, 2].

Table S2. Study designs and populations.

Table S3. Demographics and baseline characteristics.

DATA AVAILABILITY STATEMENT

Anonymized individual participant data and study documents can be requested for further research from www.clinicalstudydatarequest.com.