Efficacy and safety of TDF+3TC + EFV‑centric antiretroviral regimens for HIV/HBV co-infection: a systematic review and network meta‑analysis

Abstract

Objective

To evaluate the efficacy and safety of antiretroviral therapy(ART) regimens based on tenofovir(TDF) + lamivudine(3TC) + efavirenz(EFV) in patients co-infected with Human Immunodeficiency Virus (HIV) and Hepatitis B Virus (HBV), thereby providing evidence-based support for clinical treatment decision-making.

Methods

Relevant studies published from database inception to March 2025 were systematically retrieved from the CNKI, Wanfang, VIP, SinoMed, PubMed, Embase, Web of Science, and Cochrane Library databases. The revised Risk of Bias tool version 2 (ROB2) was used to assess study quality, and all meta-analyses were performed using Stata 18 software.

Results

A total of 31 randomized controlled trials (RCTs) involving 2124 participants and 12 treatment interventions were included, with four outcome measures evaluated. The network meta-analysis demonstrated that the TDF + 3TC + EFV triple therapy regimen was superior to the other evaluated regimens in achieving higher HBV DNA and HIV RNA negative conversion rates. In addition, TDF+ 3TC + EFV demonstrated greater efficacy than zidovudine(AZT) + 3TC + EFV in increasing CD4 + lymphocyte counts. The surface under the cumulative ranking curve (SUCRA) analysis indicated that TDF + 3TC + EFV ranked highest in terms of efficacy. Regarding safety, several studies reported various types of adverse events. However, no statistically significant differences were observed among the treatment groups.

Conclusion

Based on the currently available evidence, this network meta-analysis suggests that the TDF + 3TC + EFV triple regimen demonstrates superior efficacy across multiple outcome measures in patients with HIV/HBV co-infection. Nevertheless, owing to the inherent limitations of the included studies, further high-quality studies are required to confirm its role as a preferred therapeutic option.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12879-026-13806-5.

Introduction

The Joint United Nations Programme on HIV and AIDS estimated that by the end of 2024, approximately 40.8 million people were living with HIV globally, including 1.3 million new infections, and 31.6 million individuals were receiving ART [1]. A meta-analysis covering 87 countries and 836,644 individuals living with HIV showed that the global prevalence of HBV co-infection was approximately 8.4% [2]. The prevalence of co-infection is particularly high in West and Central Africa, the Middle East and North Africa, Asia and the Pacific, and sub-Saharan Africa [2]. The main risk factors for HBV co-infection among individuals living with HIV include alcohol consumption, higher income, and a history of smoking [3]. Studies have shown that co-infection with HIV and HBV significantly alters the natural history of HBV infection. Compared with individuals with HBV mono-infection, co-infected patients progress more rapidly to liver cirrhosis and hepatocellular carcinoma, have higher liver-related mortality, and often exhibit reduced treatment response [4, 5]. Therefore, antiviral treatment interventions for this population are of critical importance.

For individuals with HIV and HBV coinfection, treatment decisions depend not only on HBV viral load and the severity of biochemical or histological disease, but also on whether ART has been initiated. At present, TDF or its novel formulation, tenofovir alafenamide (TAF), is considered a first-line therapy for HBV due to its high potency and high barrier to resistance [6, 7]. When selecting an ART regimen, priority should be given to the inclusion of nucleoside reverse transcriptase inhibitors (NRTIs) with activity against HBV, such as 3TC or emtricitabine (FTC), in combination with TAF or TDF [8, 9]. According to guidelines from the World Health Organization (WHO), individuals with HIV and HBV co-infection should receive treatment targeting both viruses simultaneously by using an ART regimen that is effective against both viruses to reduce the risk of drug resistance [10]. A TDF-based regimen is recommended, typically consisting of TDF combined with 3TC or FTC, if there are no contraindications, along with a third agent such as EFV to prevent the emergence of HIV resistance mutations [10]. In addition, the Chinese Guidelines for the Diagnosis and Treatment of AIDS 2024 edition recommend that all individuals with HIV and HBV co-infection should initiate ART as early as possible, and that the regimen should include two agents with anti-HBV activity, with a preferred combination of TDF or TAF + 3TC or FTC [11]. These findings indicate that the implementation of combined antiviral therapy is of substantial clinical significance for individuals with HIV and HBV co-infection.

At present, evidence regarding the treatment of HIV and HBV co-infection is mostly derived from traditional meta-analyses, and systematic comparisons among different treatment regimens remain limited. Network meta-analysis enables a comprehensive evaluation of the efficacy and safety of multiple interventions within a single analytical framework, thereby providing more robust evidence for clinical decision making [12]. This study evaluated the efficacy and safety of the TDF+3TC + EFV regimen in the treatment of HIV/HBV coinfection compared with other antiretroviral therapy regimens, aiming to provide evidence for clinical treatment selection and related drug management strategies.

Methods

Registration

This study was conducted in accordance with the PRISMA guidelines for systematic reviews and meta-analyses and was registered in PROSPERO with the registration number CRD420251035951 [13]. The study protocol was designed and registered in advance to ensure transparency and reproducibility.

Literature search strategy

Relevant literature was systematically retrieved from databases, including CNKI, WanFang Data, VIP Database, SinoMed, PubMed, Embase, Web of Science, and the Cochrane Library, from inception to March 2025, with searches conducted in English and Chinese. The main search terms include: “tenofovir”, “tenofovir alafenamide”, “lamivudine”, “adefovir”, “zidovudine”, “hepatitis B”, “HBV”, “AIDS”, “HIV”, “acquired immune deficiency syndrome”, etc. The search strategy integrated Medical Subject Headings (MeSH) terms with free text terms and was further supplemented by manual searches to identify potentially overlooked studies. In addition, the reference lists of relevant reviews and original studies were manually screened to ensure the comprehensive inclusion of all relevant studies.

Study selection

This study included RCTs that met the following criteria: adult patients with HIV and HBV co-infection [14]. In the intervention group, patients received NRTI monotherapy or combination therapy, predominantly ART regimens based on TDF combined with 3TC. The control group received NRTI monotherapy, combination therapy, or placebo, specifically including regimens such as 3TC, TDF, and combination therapies of AZT, Adefovir (ADV), or stavudine (d4T) with TDF and 3TC. The primary outcome measures included efficacy indicators such as HBV DNA negative conversion and HIV RNA negative conversion. The secondary outcome measures included CD4 + lymphocyte count and the incidence of adverse reactions [14]. Data for all outcome measures were independently extracted from eligible original studies. If multiple time points were reported in a study, data from the prespecified primary time point were preferentially extracted.

The exclusion criteria primarily included non-randomized controlled trial designs, inability to access the full text, studies involving participants outside the target population or with duplicate data, interventions that did not meet predefined criteria, absence of relevant outcome measures, conference abstracts, and meta-analyses. Additionally, patients with HBV co-infected with viruses other than HIV, pregnant or lactating women, and those with other liver diseases, including autoimmune hepatitis, alcoholic liver disease, primary biliary cirrhosis, Wilson’s disease, or hepatocellular carcinoma, were excluded. Reports with incomplete information or insufficient data were also not included [15, 16].

Literature screening and data extraction

The titles and abstracts of all articles were reviewed using EndNote 21. Two researchers independently conducted an initial screening of the titles and abstracts, excluding studies that were clearly irrelevant. The full texts of the remaining articles were subsequently reviewed to assess eligibility according to the predefined inclusion and exclusion criteria. Any disagreements were resolved through consultation with a third researcher. For studies meeting the inclusion criteria, data were extracted using a standardized data extraction form. Extracted data comprised the first author, publication year, sample size, interventions, mean age, and outcome measures.

Risk of bias assessment

The quality of evidence was evaluated using the revised Risk of Bias 2 tool (ROB 2) [17]. The assessment was independently performed by two researchers (Zhang and Chen), followed by cross-checking of the results upon completion. If disagreements could not be resolved through discussion, a third researcher was consulted to reach a final decision.

Statistical analysis

All statistical analyses were conducted using STATA version 18.0 within a frequentist framework. Due to variations in measurement methods and outcome units across studies, effect sizes for continuous variables were expressed as standardized mean differences (SMDs) with 95% confidence intervals, whereas dichotomous variables were reported as risk ratios (RRs) with corresponding 95% confidence intervals [14, 18]. Heterogeneity among the included studies was evaluated using the I² statistic. When I² ≤ 50% or P > 0.10, heterogeneity was considered low, and a fixed effects model was used for data pooling. When I² > 50% or P ≤ 0.05, substantial heterogeneity was considered to be present, and a random effects model was employed for data synthesis [19]. In cases of substantial heterogeneity, random effects meta regression analysis was performed to investigate potential sources of heterogeneity and the associations between study-level covariates and treatment effects [20]. Multivariable meta regression analysis using the metareg command was conducted to examine the relationships between study characteristics and sources of heterogeneity. The following study characteristics were considered potential sources of heterogeneity: mean age, treatment duration, and geographic region. Subsequently, subgroup analysis and leave-one-out analysis were used for sensitivity analysis to explore potential sources of heterogeneity [21].

Stata 18 was used to generate a network diagram illustrating the comparative relationships among different interventions. When closed loops are present in the evidence network, inconsistency tests are required; however, such tests are omitted when a loop is informed by only a single study [22]. Meanwhile, pairwise comparisons among all interventions were presented using league tables and forest plots, and the surface under the cumulative ranking curve was used to rank the probabilities of outcome effects across interventions.

Results

Study selection

A preliminary search of the relevant literature identified 5,130 publications. After importing all records into EndNote 21, 2,667 duplicates were removed. Titles and abstracts were subsequently screened, resulting in the exclusion of 2,551 publications deemed irrelevant to the study topic. Full texts were further reviewed, and 85 publications that did not meet the inclusion criteria were excluded, resulting in a final dataset of 31 studies, including 5 in English and 26 in Chinese. The literature screening process is summarized in Fig. 1.

Fig. 1.

Flow diagram of the study selection process

Study characteristics

This study included 31 publications [23–53], of which one RCT compared three treatment regimens, whereas the remaining 30 trials compared two regimens each. The enrolled patients were primarily from China, and therefore, these findings are representative of a limited population. Detailed baseline characteristics are summarized in Table 1.

Table 1.

Characteristics of included studies

Included studies (First author, year of publication)	Region	Experimental group			Control group			Treatment Duration (months)	Outcomes
		Sample size	Mean age(y)	Intervention	Sample size	Mean age(y)	Intervention
Deng et al., 2019 [23]	China	40	46.12 ± 1.23	TDF+3TC	40	46.11 ± 1.22	3TC	NA	D
Zhang et al., 2018 [24]	China	30	42.62 ± 1.52	TDF+3TC	30	41.42 ± 1.25	3TC	12	D
Liu, 2017 [25]	China	60	31.2 ± 2. 3	TDF+3TC	60	31.3 ± 2.4	3TC	18	D
Da et al., 2019 [26]	China	40	39.5 ± 5.2	TDF+3TC	40	37.1 ± 5.6	3TC	12	B
Yin et al., 2017 [27]	China	45	39.82 ± 2.15	TDF+3TC	45	39.82 ± 2.15	3TC	12	B
Gao, 2016 [28]	China	35	36.2 ± 3.7	TDF+3TC	35	37.8 ± 3.6	3TC	12	B
Yang et al., 2018 [29]	China	34	33.0 ± 2.9	TDF+3TC	34	32.5 ± 2.5	3TC	12	D
Xia, 2018 [30]	China	30	33 ± 7	TDF+3TC	30	34 ± 8	3TC	12	B
Ding et al., 2020 [31]	China	18	42.6 ± 5.4	TDF+3TC	18	42.6 ± 5.4	3TC	6	B
Zhu, 2016 [32]	China	21	47.5 ± 6.9	TDF+3TC	21	45.5 ± 6.8	3TC	12	BD
Li, 2017 [33]	China	35	47.26 ± 10.31	TDF+3TC	35	48.12 ± 9.91	3TC	6	D
Yang, 2018 [34]	China	75	36.82 ± 2.31	TDF+3TC	75	38.74 ± 2.83	3TC	12	B
Sun, 2018 [35]	China	40	38.9 ± 6.0	TDF+3TC	40	38.2 ± 5.2	3TC	18	CD
Wang, 2020 [36]	China	35	38.96 ± 2.34	TDF+3TC	35	39.26 ± 2.16	3TC	12	B
Wu et al., 2020 [37]	China	39	43.2 ± 2.6	TDF+3TC	39	43.5 ± 2.3	3TC	12	BD
Huang et al., 2021 [38]	China	55	43.6 ± 5.3	TDF+3TC	55	43.5 ± 5.4	3TC	12	AD
Wang et al., 2023 [39]	China	20	38.93 ± 4.63	TDF+3TC	20	39.02 ± 4.74	3TC	6	B
Bai, 2018 [40]	China	20	35.87 ± 2.19	TDF+3TC	20	33.12 ± 1.96	TDF	12	B
Yin et al., 2019 [41]	China	37	37.0 ± 4.3	TDF+3TC	37	37.3 ± 4.3	TDF	12	D
Dore et al., 1999 [42]	CAN, AUS, EURO, ZAF	97	37	3TC	25	37	Placebo	12	A
Peters et al., 2006 [43]	US	27	NA	TDF	25	NA	ADV	12	D
Lu, 2021 [44]	China	34	51.38 ± 3.28	TDF+3TC + EFV	34	51.45 ± 3.31	TDF+3TC	6	ABD
Zhao et al., 2022 [45]	China	30	44.94 ± 6.38	TDF+3TC + EFV	30	45.71 ± 7.64	3TC + AZT	6	ACD
Luo et al., 2016 [46]	China	10	54.67 ± 2.22	TDF+3TC + EFV	10	54.12 ± 2.42	AZT+3TC + EFV	6	ABCD
Yan, 2023 [47]	China	38	36.47 ± 3.58	TDF+3TC + EFV	37	35.52 ± 3.62	AZT+3TC + EFV	3	ABCD
Cai, 2018 [48]	China	29	41.81 ± 7.49	TDF+3TC + EFV	29	42.12 ± 7.53	AZT+3TC + EFV	12	CD
Sarkar et al., 2020 [49]	India	33	36	TDF+3TC + EFV	32	36	3TC + ADV+AZT + EFV	30	C
Dore et al., 2004 [50]	LA, EURO, US	5	42	TDF+3TC + EFV	6	34	d4T+3TC + EFV	12	D
Wu, 2019 [51]	China	25	32.14 ± 3.25	TDF+3TC + EFV	25	31.25 ± 4.25	3TC + EFV	12	D
Zhou, 2016 [52]	China	45	35.13 ± 3.06	TDF+3TC + EFV	45	36.09 ± 3.12	3TC + EFV	12	B
Matthews et al., 2008 [53]	Thailand	13	37.5 ± 7.0	AZT+3TC + EFV	11	32.1 ± 8.0	AZT + TDF+EFV	12	C
Matthews et al., 2008 [53]	Thailand	13	37.5 ± 7.0	AZT+3TC + EFV	11	32.1 ± 8.0	TDF+3TC + EFV	12	C
Matthews et al., 2008 [53]	Thailand	11	37.5 ± 7.0	AZT + TDF+EFV	11	32.1 ± 8.0	TDF+3TC + EFV	12	C

Abbreviations: TDF, Tenofovir; 3TC, Lamivudine; EFV, Efavirenz; AZT, Zidovudine; ADV, Adefovir; d4T, Stavudine; A, HBV DNA Negative Conversion; B, HIV RNA Negative Conversion; C, CD4 + Lymphocyte Count; D, Adverse Reaction Rate; NA, not available

Risk of bias assessment

According to the RoB 2 assessment, 68% of the included studies were rated as having a low risk of bias, whereas 32% raised some concerns. The primary sources of concern regarding risk of bias were the randomization process and deviations from intended interventions. Detailed results of the RoB 2 assessment are presented in Fig. 2.

Fig. 2.

The quality assessment of included RCTs. (a) Risk of bias graph: review authors’ judgements about each risk of bias item presented as percentages across all included studies. (b) Risk of bias summary: review authors’ judgements about each risk of bias item for each included study

Primary outcome

Regarding the primary outcome measures, six studies reported HBV DNA negative conversion, while fourteen studies reported HIV RNA negative conversion. The corresponding network evidence plots are shown in Fig. 3a and b, respectively. In these figures, node size represents the number of participants included in each study, with larger nodes indicating larger sample sizes, whereas the thickness of connecting lines represents the number of studies within each comparison, with thicker lines indicating a greater number of included studies. The evidence plots indicate that no closed loops exist in the evidence structure, meaning that all comparisons among interventions are directly derived from original studies without forming indirect comparison loops. Therefore, inconsistency testing was not required, and a consistency model was directly applied for the network meta-analysis.

Fig. 3.

Network evidence plots. Abbreviations: TDF, Tenofovir; 3TC, Lamivudine; EFV, Efavirenz; AZT, Zidovudine; ADV, Adefovir; d4T, Stavudine

HBV DNA negative conversion

An overall heterogeneity analysis was performed across the six included studies on HBV DNA negative conversion, as shown in Fig. 4a. The heterogeneity analysis (I² =60.1%, P = 0.028) indicated moderate heterogeneity among the included studies; therefore, a random-effects model was applied for subsequent analysis.

Fig. 4.

Heterogeneity test. Abbreviations: TDF, Tenofovir; 3TC, Lamivudine; EFV, Efavirenz; AZT, Zidovudine; ADV, Adefovir; d4T, Stavudine

The results of the network meta-analysis indicate that, compared with other regimens, TDF + 3TC + EFV significantly improves HBV DNA negative conversion. TDF + 3TC is more effective than 3TC monotherapy [RR = 1.25, 95% CI (1.04, 1.50)] and placebo [RR = 5.50, 95% CI (1.42, 21.35)]in achieving HBV DNA negative conversion, while 3TC also demonstrates an advantage over placebo [RR = 4.40, 95% CI (1.15, 16.87)]. All other between-group comparisons showed no statistically significant differences. Detailed results are presented in Table 2; Fig. 5a.

Table 2.

Network meta-analysis of HBV DNA negative conversion

Abbreviations: TDF, Tenofovir; 3TC, Lamivudine; EFV, Efavirenz; AZT, Zidovudine; Values in red indicate statistical significance (P < 0.05, with the 95% confidence interval not including 1)

Fig. 5.

Forest plot of the network meta-analysis. Abbreviations: a, HBV DNA Negative Conversion; b, HIV RNA Negative Conversion; c, CD4 + Lymphocyte Count; d, Adverse Reaction Rate; TDF, Tenofovir; 3TC, Lamivudine; EFV, Efavirenz; AZT, Zidovudine; ADV, Adefovir; d4T, Stavudine

According to the SUCRA results presented in Fig. 6a, a larger area under the curve, approaching 100%, indicates greater treatment efficacy and a higher probability of being the optimal therapeutic option. The SUCRA values of the interventions, ranked from highest to lowest, were TDF + 3TC + EFV (99.6%), TDF + 3TC (76.3%), 3TC (48.8%), AZT + 3TC + EFV (39.7%), 3TC + AZT (33.4%), and placebo (2.2%).

Fig. 6.

SUCRA cumulative probability ranking plots. Abbreviations: TDF, Tenofovir; 3TC, Lamivudine; EFV, Efavirenz; AZT, Zidovudine; ADV, Adefovir; d4T, Stavudine

HIV RNA negative conversion

An overall heterogeneity test was conducted on the 14 included studies assessing HIV RNA negative conversion, as shown in Fig. 4b. The heterogeneity test (I² =0.0%, P = 0.920) indicated no significant heterogeneity across the included studies; therefore, a fixed effects model was applied for the analysis.

The results of the network meta-analysis indicated that TDF + 3TC + EFV was more effective in promoting HIV RNA negative conversion than other treatment regimens (P < 0.05). In addition, 3TC + EFV demonstrated superior efficacy compared with 3TC monotherapy [RR = 1.39, 95% CI (1.02, 1.89)]. TDF+3TC also showed significantly superior efficacy compared with 3TC monotherapy [RR = 1.29, 95% CI (1.19, 1.40)] and TDF monotherapy [RR = 1.50, 95% CI (1.02, 2.21)]. All other pairwise comparisons were not statistically significant. Detailed results are provided in Table 3; Fig. 5b.

Table 3.

Network meta-analysis of HIV RNA negative conversion

Abbreviations: TDF, Tenofovir; 3TC, Lamivudine; EFV, Efavirenz; AZT, Zidovudine; Values in red indicate statistical significance (P < 0.05, with the 95% confidence interval not including 1).

According to the SUCRA results (Fig. 6b), the regimens were ranked by SUCRA values from highest to lowest as follows: TDF+3TC + EFV (99.0%), 3TC + EFV (72.0%), TDF+3TC (64.6%), 3TC (29.2%), AZT+3TC + EFV (20.4%), and TDF (14.7%). Based on the available evidence, TDF+3TC + EFV demonstrated the highest cumulative probability of being the optimal regimen among all evaluated treatments.

Secondary outcome

Regarding secondary outcome indicators, five studies reported CD4 + lymphocyte count, while 18 studies reported adverse reaction rate, with the corresponding network evidence plots presented in Fig. 3c and d, respectively. In the CD4 + lymphocyte count evidence network shown in Fig. 3c, although a closed loop was identified, inconsistency testing was not conducted because the loop was derived entirely from a single study.

CD4 + lymphocyte count

An overall heterogeneity test was conducted on the five included studies assessing CD4 + lymphocyte count, as shown in Fig. 4c. The heterogeneity test (I² = 60.6% and P = 0.026) indicated moderate heterogeneity among the studies; therefore, a random effects model was applied for the analysis.

The results of the network meta-analysis (Table 4; Fig. 5b) showed that, in improving CD4 + lymphocyte count, TDF+3TC + EFV was superior to AZT+3TC + EFV [SMD = 0.90, 95% CI (0.44, 1.36)], while all other pairwise comparisons were not statistically significant.

Table 4.

Network meta-analysis of CD4 + lymphocyte count

Abbreviations: TDF, Tenofovir; 3TC, Lamivudine; EFV, Efavirenz; AZT, Zidovudine; ADV, Adefovir; Values in red indicate statistical significance (P < 0.05, with the 95% confidence interval not including 0).

According to the SUCRA results (Fig. 6c), the regimens were ranked by SUCRA values from highest to lowest as follows: TDF+3TC + EFV (87.6%), 3TC + AZT (74.5%), 3TC + ADV+AZT + EFV (45.7%), AZT + TDF+EFV (22.8%), and AZT+3TC + EFV (19.4%).

Adverse reaction rate

An overall heterogeneity test was conducted on the 18 included studies assessing adverse reaction rate, as shown in Fig. 4d. The heterogeneity test(I² = 5.6%, P = 0.388) indicated no significant heterogeneity among the studies. Therefore, a fixed effects model was applied for the analysis.

Eighteen RCTs reported the overall adverse reaction rate, involving nine treatment regimens. The results of the network meta-analysis showed that there were no statistically significant differences in the overall incidence of adverse reactions among all treatment regimens, as detailed in Table 5; Fig. 5d.

Table 5.

Network meta-analysis of adverse reaction rate

Abbreviations: TDF, Tenofovir; 3TC, Lamivudine; EFV, Efavirenz; AZT, Zidovudine; ADV, Adefovir; d4T, Stavudine; Values in red indicate statistical significance (P < 0.05, with the 95% confidence interval not including 1)

According to the SUCRA results (Fig. 6d), a larger SUCRA value indicates a higher probability of adverse reactions. The SUCRA ranking results for each intervention showed that the regimens ranked from highest to lowest as follows: 3TC + EFV (78.9%), AZT+3TC + EFV (76.1%), 3TC + AZT (71.9%), TDF+3TC + EFV (54.4%), d4T+3TC + EFV (52.8%), TDF+3TC (46.7%), 3TC (43.3%), ADV (15.3%), and TDF (10.5%).

Regression analysis and sensitivity analysis

Using HBV DNA negative conversion and CD4 + lymphocyte count as outcome measures, and mean age, treatment duration, and region as covariates, a meta-regression analysis was performed on the included studies. No sources of heterogeneity were identified. The results are presented in Tables 6 and 7. A leave-one-out method was applied to assess changes in heterogeneity after exclusion of individual studies. The results (Fig. 7) showed that exclusion of any single study did not significantly affect the heterogeneity of the two outcome indicators, suggesting that the results were robust. Subgroup analysis showed that the TDF + 3TC + EFV regimen demonstrated advantages over other control regimens in terms of HBV DNA negative conversion, HIV RNA negative conversion, and CD4 + lymphocyte count increase. However, no significant differences were observed among the treatment regimens in the total incidence of adverse reactions, with results shown in Fig. S1 to S4.

Table 6.

Meta-regression analysis of HBV DNA negative conversion

_ES	Coefficient	Std. err.	t	P > t	[95% conf.	interval]
age	-0.0007052	0.0268241	-0.03	0.981	-0.1161199	0.1147096
time	-0.033699	0.0459497	-0.73	0.54	-0.2314048	0.1640067
region	1.188237	0.8120068	1.46	0.281	-2.305546	4.682021
_cons	-0.4643902	1.664782	-0.28	0.806	-7.627369	6.698588

Table 7.

Meta-regression analysis of CD4 + lymphocyte count

_ES	Coefficient	Std. err.	t	P > t	[95% conf.]	[interval]
age	-0.03022	0.049414	-0.61	0.603	-0.24283	0.182397
time	0.020816	0.044777	0.46	0.688	-0.17184	0.213476
region	-0.81713	1.015714	-0.8	0.506	-5.1874	3.553128
_cons	2.791795	2.752635	1.01	0.417	-9.05184	14.63543

Fig. 7.

Sensitivity analysis. Abbreviations: a, HBV DNA Negative Conversion; b, CD4 + Lymphocyte Count

Publication bias

The results of the bias analysis for HBV DNA negative conversion rate, HIV RNA negative conversion rate, CD4 + lymphocyte count., and the incidence of adverse events are presented in Fig. 8, with each dot representing an individual study. The analysis indicated that the funnel plots in Fig. 8a and c comprised a limited number of direct comparisons; therefore, only a qualitative visual assessment was conducted without performing statistical tests such as the Egger test. Although the plots visually suggested no evident publication bias, the small number of studies implies that the symmetry of the funnel plots could be influenced by random factors; hence, this observation should be interpreted with caution. The funnel plot for the HIV RNA negative conversion rate (Fig. 8b) demonstrates that most study points are located in the upper region, indicating that the included studies exhibit relatively high precision. The scatter points are generally symmetrical, suggesting a low likelihood of publication bias. The funnel plot for the incidence of adverse events (Fig. 8d) indicates that the study points are generally symmetrically distributed around the zero line, although some points lie lower, which may suggest publication bias and indicate the presence of small-sample effects.

Fig. 8.

Comparison‑adjusted funnel plot. Abbreviations: a, HBV DNA Negative Conversion; b, HIV RNA Negative Conversion; c, CD4 + Lymphocyte Count; d, Adverse Reaction Rate; TDF, Tenofovir; 3TC, Lamivudine; EFV, Efavirenz; AZT, Zidovudine; ADV, Adefovir; d4T, Stavudine

Discussion

Patients with HIV and HBV co-infection typically exhibit frequent relapses, low cure rates, and an extended disease course, leading to an elevated risk of mortality. Therefore, strengthened management of both viruses is essential. At present, many commonly used ART regimens for HIV display potent anti-HBV activity, and clinical guidelines recommend TDF-containing ART regimens for individuals with HIV/HBV co-infection [54, 55]. Nevertheless, comparative clinical trials assessing different ART regimens in HIV/HBV co-infected individuals remain relatively scarce, and the results of existing studies are inconsistent; thus, a meta-analysis is warranted to provide a comprehensive evaluation.

This study included 31 RCTs and employed a network meta-analysis to evaluate the efficacy and safety of ART regimens based on TDF+3TC + EFV across four outcome measures, namely HBV DNA negative conversion, HIV RNA negative conversion, CD4 + lymphocyte count, and incidence of adverse reactions.

The network meta-analysis suggested that TDF+3TC + EFV may be superior to other evaluated regimens in improving HBV DNA negative conversion; however, this conclusion is based on limited evidence and should be interpreted with caution. Previous studies have indicated that, for HIV/HBV co-infection, combination therapy is more advantageous than monotherapy [15, 45], which is consistent with the findings of the present study. The primary focus of HBV treatment is the restoration of liver function; when liver function is normal, immediate antiviral therapy is generally unnecessary, even in the presence of high HBV DNA levels. A key mechanism of HBV-induced carcinogenesis is sustained high-level viral replication, which leads to the integration of viral DNA into the host hepatocyte genome, thereby activating proto-oncogenes and disrupting tumor suppressor gene function, ultimately promoting hepatocellular carcinoma development [56]. Consequently, clinical practice primarily aims to inhibit active HBV replication to prevent progression to cirrhosis and hepatocellular carcinoma. In addition to its activity against HIV-1, TDF specifically inhibits HBV reverse transcriptase: it is phosphorylated intracellularly to tenofovir diphosphate, which lacks a 3′-OH group, thereby terminating viral DNA chain elongation and effectively blocking viral replication [57–59]. Notably, some HIV and HBV co-infected patients maintain HBV control after discontinuing TDF, and studies suggest that post-treatment HBV virological rebound is more closely associated with uncontrolled HIV infection and HBeAg-positive status [60].

In terms of improving HIV RNA negative conversion rate, the TDF+3TC + EFV combination regimen was superior to the evaluated treatment regimens. The active metabolites of 3TC and TDF, generated through intracellular phosphorylation, inhibit viral reverse transcription by competitively binding to HIV reverse transcriptase, thereby replacing natural nucleosides [61]. A study of HIV and HBV co-infected patients further indicated that elevated intracellular concentrations of tenofovir diphosphate (TFV-DP) were significantly correlated with dual suppression of HIV RNA and HBV DNA, and long-term administration of the TDF/3TC regimen achieved high virological suppression rates (92.8% for HIV and 93.5% for HBV) [15, 62]. These results are consistent with the enhanced efficacy of the combination regimen observed in the present study. Moreover, EFV, a non-nucleoside reverse transcriptase inhibitor (NNRTI), exhibits potent activity against HIV-1, and even in the presence of single or multiple point mutations in viral reverse transcriptase, resistance to EFV remains relatively low [63].

HIV infection can result in substantial depletion of CD4 + lymphocytes and progressive impairment of immune function, thereby increasing the risk of opportunistic infections and malignant tumors and elevating mortality rates [64]. This study showed that, in terms of improving CD4 + lymphocyte count, the TDF+3TC + EFV combination regimen tended to be superior to other treatment regimens, suggesting that it may facilitate immune function recovery. Some studies have indicated that this regimen can increase patient body weight and CD4 + lymphocyte count in the short term; however, potential changes in laboratory indicators indicative of nephrotoxicity should be closely monitored [65]. This conclusion is consistent with the findings of the present study, indicating that the combination of these three drugs exerts a significant synergistic effect in suppressing viral replication and enhancing immune function. A lower median CD4 + lymphocyte count is additionally associated with suboptimal HBV suppression. In their study, the researchers observed that HIV-infected individuals with CD4 + lymphocyte counts below 200 cells/µL and HBV DNA testing adherence below 95% exhibited compromised HBV viral suppression [66].

Regarding safety, the network meta-analysis demonstrated no statistically significant differences in the overall incidence of adverse events among the treatment regimens. However, this finding should be interpreted cautiously. Owing to the limited sample sizes of the included studies, most comparisons yielded wide confidence intervals and were insufficient to reliably detect potential true differences. Furthermore, commonly reported adverse reactions across studies included nausea and vomiting, dizziness and headache, elevated blood lipid levels, and constipation, with occasional reports of drug rash, hypophosphatemia, and diarrhea. Hypophosphatemia may be attributable to TDF [54, 67]. TDF is eliminated through glomerular filtration and active proximal tubular secretion, which presents a potential risk of tubular toxicity. Drug accumulation within proximal tubular cells may result from elevated plasma drug concentrations, competition among transport proteins, or genetic polymorphisms [68]. Nonetheless, some reports have suggested that renal function may improve following the initiation of TDF-containing ART [69]. Common adverse effects of EFV include early self-limiting neuropsychiatric symptoms and hypersensitivity-associated rash. Long-term EFV use may increase the risk of suicidal behavior, induce neurocognitive impairment, and cause metabolic abnormalities [69]. Studies indicate that reducing the EFV dose from 600 mg to 400 mg may improve specific safety parameters, such as blood lipid levels, while preserving efficacy and adherence; however, for patients already experiencing neuropsychiatric symptoms, switching to an alternative agent rather than merely reducing the dose is recommended [70].

Given the following limitations, the findings of this network meta-analysis should be interpreted with caution. First, heterogeneity constitutes an inherent limitation of this analysis. Both HBV DNA negative conversion and CD4 + lymphocyte count showed substantial heterogeneity across the included studies. Although meta regression and sensitivity analyses indicated that region, mean age, and treatment duration were not sources of heterogeneity, other factors such as sample size, detection methods, and disease stage may still have influenced the accuracy of the results. Furthermore, when the number of studies included in a meta-regression analysis is small, the reliability of the results is compromised, which increases the risk of false positive or false negative findings; therefore, the results must be interpreted with caution. In addition, exclusion of any individual study did not result in significant changes in heterogeneity for either outcome. However, given the limited number of studies in some comparisons and concerns regarding the methodological quality of certain included studies, the findings of this network meta-analysis are subject to considerable uncertainty and should therefore be interpreted with caution.

In summary, the combination therapy of TDF + 3TC + EFV demonstrated potential advantages in improving overall clinical efficacy. However, owing to the aforementioned limitations, these findings should be interpreted with caution. Future research should focus on conducting large sample size, multicenter, double blind, high-quality RCTs to provide more robust evidence for the clinical management of patients with HIV and HBV co-infection.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

HIV

Human Immunodeficiency Virus

HBV

Hepatitis B Virus

TDF

Tenofovir

3TC

Lamivudine

EFV

Efavirenz

AZT

Zidovudine

TAF

Tenofovir Alafenamide

ADV

Adefovir

NRTIs

Nucleoside Reverse Transcriptase Inhibitors

FTC

Emtricitabine

RCTs

Randomized Controlled Trials

d4T

Stavudine

SUCRA

Surface Under the Cumulative Ranking curve

Author contributions

Zhang Xinru conducted literature search. Xie Zhuohua, Deng Yanting and Deng Jiasheng evaluated the study articles and made decisions on inclusion and exclusion of the articles. Zhang Xinru, Chen Jieyi, Fang Yibin performed statistical analyses. All authors (Zhang Xinru, Chen Jieyi, Fang Yibin, Xie Zhuohua, Deng Yanting, Deng Jiasheng and Zhou Zhipin) were involved in the manuscript development and its revision. All authors contributed to the critical revision of the manuscript.

Funding

This work was supported by grants of the National Natural Scientific Foundation of China (No.81760751), Guangxi Provincial Natural Scientific Foundation (No. 2021GXNSFAA075020).

Data availability

The data of this study can be obtained from the corresponding author according to reasonable requirements.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.