Efficacy and Safety of Elvitegravir/Cobicistat/Emtricitabine/Tenofovir Alafenamide as Maintenance Treatment in HIV/HBV-coinfected Patients

Yu-Shan Huang, MD; Chien-Yu Cheng, MD; Bo-Huang Liou, MD; Po-Liang Lu, MD, PhD; Shu-Hsing Cheng, MD, PhD; Yuan-Ti Lee, MD; Chun-Eng Liu, MD; Hsin-Yun Sun, MD; Chia-Jui Yang, MD; Hung-Jen Tang, MD; Shih-Ping Lin, MD; Mao-Wang Ho, MD; Sung-Hsi Huang, MD; Hung-Chin Tsai, MD, PhD; Chen-Hsiang Lee, MD; Chien-Ching Hung, MD, PhD


J Acquir Immune Defic Syndr. 2021;86(4):473-481. 

In This Article


Participant Characteristics

During the study period, 274 HIV/HBV-coinfected participants who met the inclusion criteria were enrolled. Among them, 268 and 261 completed the 24 and 48 weeks of follow-up, and 13 (4.7%) withdrew from the study before week 48, with 7 (2.6%) withdrew because of adverse effects (Figure 1). The adverse events leading to discontinuations of E/C/F/TAF were gastrointestinal discomfort in 3 participants, insomnia in 2, depression in one, and headache in one.

Figure 1.

Flow diagram of the study. E/C/F/TAF, coformulated elvitegravir, cobicistat, emtricitabine, and tenofovir alafenamide.

The baseline characteristics of the participants are shown in Table 1. Most were middle-aged men who have sex with men. Before switching to E/C/F/TAF, the median duration of exposure to TDF was 4.0 years (range, 2.4–6.0). Overall, 12.8% of the participants tested HBeAg-positive, and 94.2% had a plasma HBV DNA <20 IU/mL. For participants with detectable HBV DNA (≥20 IU/mL) before E/C/F/TAF, the median HBV DNA was 47 IU/mL (range, 29–108).

Virologic Efficacy

The virologic responses of HBV and HIV to E/C/F/TAF at weeks 24 and 48 are shown in Figure 2, in which 92.7% and 89.8% of the participants achieved plasma HBV DNA <20 IU/mL and 4.7% and 5.1% of participants had HBV DNA ≥20 IU/mL at weeks 24 and 48, respectively (Figure 2A). The median HBV DNA of 14 participants with a detectable HBV DNA at week 48 was 31 IU/mL (range, 28–38). During the follow-up period, 5 participants had >1 log10 increase in plasma HBV DNA from having achieved <20 IU/mL and 2 had persistent detectable HBV DNA, and the highest HBV DNA level detected was 1000 IU/mL. All 7 participants had undetectable HIV RNA at the end of the study. After adjusting for age, baseline CD4 count, duration of TDF exposure, baseline HBsAg level, HBV DNA level, and HBeAg positivity, a detectable HBV DNA (≥20 IU/mL) was significantly correlated with baseline HBeAg positivity (adjusted odds ratio 3.652, 95% confidence interval 1.248 to 10.690, P = 0.018) and duration of previous TDF therapy (per 1-year increase, adjusted odds ratio 0.812, 95% CI: 0.677 to 0.975, P = 0.026) by a generalized estimating equation analysis (see Table 1, Supplemental Digital Content 1, http://links.lww.com/QAI/B580). The proportion of participants who maintained HIV suppression at weeks 24 and 48 was 95.6% and 94.2%, respectively (Figure 2B).

Figure 2.

Virologic responses of (A) HBV (HBV DNA <20 IU/mL) and (B) HIV (HIV RNA <50 copies/mL) to E/C/F/TAF at weeks 24 and 48.

Serologic Response of HBV and Liver Function

At week 48 of E/C/F/TAF, 3 (8.6%) of 35 HBeAg-positive participants had a loss of HBeAg and 2 (5.7%) had HBeAg seroconversion. One HBeAg-negative participant had HBeAg seroreversion at week 48. The median HBsAg level decreased significantly in HBeAg-positive participants from 3.3 log10 IU/mL to 3.2 log10 IU/mL (P < 0.001) but not in HBeAg-negative participants (see Figure 1, Supplemental Digital Content 1, http://links.lww.com/QAI/B580). The loss of HBsAg occurred in 3 (1.1%) participants, and one of them had an equivocal anti-HBs titer at week 48. The median ALT and aspartate aminotransferase levels of the participants decreased significantly, and 26 (53.1%) of 49 participants who had ALT ≥40 IU/mL at baseline achieved ALT normalization.

Evolution of Renal Function and Lipid Profile

The median serum creatinine increased from 0.94 mg/dL to 1.00 mg/dL, and the median eGFR decreased from 98.8 mL/min/1.73m2 to 94.9 mL/min/1.73m2 through 48 weeks (both P < 0.001). The participants experienced significant decreases in urine protein–creatinine ratio, UACR, and urine β2-microglobulin–creatinine ratio (Figure 3A). All fasting lipid levels increased from baseline to week 48, including total cholesterol (TC) to high-density lipoprotein (HDL)-cholesterol (TC/HDL-cholesterol) ratio (Figure 3B). The mean percentage increase from baselines in triglycerides, total cholesterol, LDL-cholesterol, and HDL-cholesterol was 38%, 17%, 22%, and 12%, respectively, at week 48. A modest weight gain was noted among 89 patients with follow-up measurements, with a median weight increase from 69 kg at baseline to 72 kg at week 48 (P < 0.001). No correlation was found between the increases in lipids and weight gain with linear regression. There were no statistically significant changes in fasting blood glucose and HbA1C from baseline to week 48 (see Table 2, Supplemental Digital Content 1, http://links.lww.com/QAI/B580).

Figure 3.

Changes in (A) quantitative proteinuria [median (95% confidence interval)] and (B) lipid profile through week 48. The asterisk (*) indicates that the median level at week 48 is significantly different from baseline (P < 0.05). UPCR, urine protein-tocreatinine ratio; TC, total cholesterol; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol.

Evolution of Bone Mineral Density

Of the 181 (66.1%) participants who underwent BMD assessment, the mean BMD of the spine and hip increased significantly at week 48 (+1.77% and +1.33%, respectively) (Figure 4A). The comparisons between the participants who underwent the BMD assessment and those who did not are shown in Supplemental Digital Content 1 (see Table 3, http://links.lww.com/QAI/B580). The prevalence of osteopenia at the spine and hip decreased from 20.6% and 27.6% to 18.4% and 24.5%, respectively, through week 48. The prevalence of osteoporosis at the spine and hip also decreased from 1.7% and 3.3% at baseline to 0.6% and 1.3%, respectively, at week 48 (Figure 4B). The changes in BMD, T-score, and Z-score are shown in Supplemental Digital Content 1 (see Table 2, http://links.lww.com/QAI/B580).

Figure 4.

Changes in (A) BMD of the spine and hip and (B) the proportion of participants with osteopenia and osteoporosis through week 48.

HDV Coinfection

The seroprevalance of anti-HDV IgG remained stable from baseline (14.6%) to week 48 (13.5%) (see Figure 2A, Supplemental Digital Content 1, http://links.lww.com/QAI/B580). Three participants (1.1%) had anti-HDV seroconversion during the follow-up. Among participants with positive anti-HDV IgG, 25% (10/40) and 18.9% (7/37) had detectable plasma HDV RNA (≥100 copies/mL) at baseline and week 48, respectively. Five participants (1.8%) had persistent HDV viremia during the study period. The median HDV RNA among participants with detectable HDV RNA was 4.7 log10 copies/mL and 4.1 log10 copies/mL at baseline and week 48, respectively (see Figure 2B, Supplemental Digital Content 1, http://links.lww.com/QAI/B580). Seven (41%) of the 17 participants with detectable HDV RNA experienced ALT elevation for greater than 2-fold of the upper limits of normal during the 48 weeks of follow-up.