Effectiveness and Renal Safety of Tenofovir Alafenamide Fumarate Among Chronic Hepatitis B Patients

Real-World Study

Mina S. Farag; Scott Fung; Edward Tam; Karen Doucette; Alexander Wong; Alnoor Ramji; Brian Conway; Curtis Cooper; Keith Tsoi; Philip Wong; Giada Sebastiani; Mayur Brahmania; Sarah Haylock-Jacobs; Carla S. Coffin; Bettina E. Hansen; Harry L.A. Janssen


J Viral Hepat. 2021;28(6):942-950. 

In This Article


Patient Population and Virologic Response

A total of 176 patients met the inclusion criteria. At baseline, 143 patients were NA-experienced, and 33 patients were NA naïve. Baseline characteristics are shown in Table 1. The mean age was 52 years(SD 13), 73% were male, 80% of Asian ethnicity, mean liver stiffness using FibroScan was 7.8(6.4), and 18% were diagnosed with cirrhosis. Most NA-experienced patients(88%) had been pre-treated with TDF for at least 12 months before switching to TAF.

To assess the effectiveness of TAF treatment on viral suppression, HBV DNA and HBsAg were investigated before and after switching to TAF. The majority of NA-naïve patients achieved undetectable HBV DNA after one year of TAF treatment(75%). Additionally, HBsAg showed a significant decline after TAF initiation(−0.03[−0.04 - −0.01]log10 IU/mL/month, p<0.001).

Among NA-experienced patients at the time of the switch, 72% had undetectable HBV DNA, and after two years of TAF treatment, 96% had achieved an undetectable HBV DNA(Figure 1). Among the NA-experienced HBeAg-positive patients, two achieved HBeAg loss and three achieved HBeAg seroconversion after starting TAF.

Since the majority of the NA-experienced patients were treated with TDF before switching to TAF(88%), we focused our renal analysis on TDF-treated patients. There were not enough patients to evaluate the effect of switching from other NA therapies.

Figure 1.

HBV DNA Detectability Among NA-Experienced and Naïve Patients. Kaplan–Meier curve showing the proportion of patients with HBV DNA below detection limit for NA-experienced and NA-naïve patients

Renal Outcomes Among Patients Switched From TDF to TAF

During the TDF treatment period, eGFR declined significantly(−0.18[−0.31 - −0.05]mL/min/month, p= 0.008). However, this trend significantly reversed(p=0.02) leading to stable eGFR after switching to TAF(0.00[−0.06 – 0.06]mL/min/month, p=1.0)(Figure 2A). This stabilization of eGFR was sustained through the third year of follow-up(week 0: 74.9, Week 52:75.0, Week 104:75.1, Week 156:75.2 mL/min). In addition, serum phosphate showed improvement after starting TAF treatment(TDF:0.03[−0.01 – 0.08]log10 IU/mL, p=0.2; TAF:1.0[−0.04 – 0.2]log10 IU/mL, p=0.2).

Figure 2.

eGFR Changes. A, General linear model curve demonstrating eGFR changes among NA-experienced patients before and after starting TAF. B, General linear model curve showing eGFR changes according to CKD stage among TAF-treated patients who switched from TDF. C, General linear model showing eGFR changes among NA-experienced patients with or without renal comorbidity before and after starting TAF

Renal Outcomes According to Chronic Kidney Disease Stage at Baseline

Patients receiving TDF were categorized at baseline (TAF switch) according to different stages of CKD based on their eGFR [stage 1(≥90), 2(60–89), 3a(45–59), ≥3b(< 45)](Figure 2B). At baseline, 29% had eGFR >90, 36% had eGFR 60–89, 20% had eGFR 45–59, and 15% had eGFR <45. After switching to TAF, kidney function deterioration as observed during TDF was halted (Stage 2:TDF:(−0.25[−0.52 – 0.02]mL/min/month, p=0.07, TAF:(+0.02[−0.09 – 0.13]mL/min/month, p=0.75, difference in trend (p=0.09); (Stage 3a:TDF:-0.13[−0.45 – 0.18]mL/min/month, p=0.4, TAF: +0.05 [−0.10 – 0.20]mL/min/month, p=0.5), difference in trend (p=0.34); (≥Stage 3b: TDF:-0.36[−0.68 - −0.04]mL/min/month, p=0.03; TAF:+0.02[−0.15 – 0.19]mL/min/month, p=0.8), difference in trend(p=0.04)).

To validate the current recommendation of the EASL guidelines[5]s which recommend switching to TAF in those patients with eGFR<60 mL/min, we also compared the influence of TAF vs TDF on kidney function among patients with eGFR below 60 mL/min(n=59) or above 60 mL/min(n=71) at the time of TAF switch. Out of 59 with eGFR <60 mL/min, 31(53%) patients had eGFR deterioration during TDF that was reversed to eGFR increase after one year of TAF treatment(p=0.009) vs ten patients who continued to show eGFR decline after switching to TAF. Among the 71 patients with eGFR >60 mL/min at the time of TAF switch, 25(35%) had eGFR increase during TAF after it had deteriorated during TDF treatment(p<0.001) vs ten patients who continued to show eGFR decrease after switching to TAF.

Renal Outcomes Among Patients With Renal Comorbidity

Patients who were treated with TDF prior to switching to TAF with renal comorbidity(n=39), were compared to those without(n=121)(Figure 2C). Patients without renal comorbidity experienced stable eGFR after switching to TAF compared to a decline in eGFR while on TDF(TDF:-0.16[−0.32 – 0.00]mL/min/month, p=0.05); TAF:+0.03[−0.04 – 0.09]mL/min/month, p=0.4); difference in trend(p=0.048). In contrast, patients with renal comorbidity had a profound deterioration in their kidney function during the TDF treatment period which continued after switching to TAF, albeit to a much lesser extent(TDF:-0.27[−0.25 - −0.03]mL/min/month, p=0.03); TAF:-0.11[−0.23 – 0.02]mL/min/month, p=0.1); difference in trend(p=0.25). The continued minimal eGFR decline after switch to TAF was thus likely due to the underlying renal comorbidity.

Renal Outcomes According to TDF Dose

We compared eGFR changes after starting TAF according to the TDF dose given before switching: full-dose TDF(n=116) or reduced-dose TDF(n=9)(Figure 3). Patients who had been switched to reduced-dose TDF due to a prior eGFR decline had a lower eGFR at baseline compared to full-dose TDF patients(full-dose TDF: 78.1 mL/min vs reduced-dose: 66.6 mL/min). One year of full-dose TDF treatment was associated with significant eGFR decline(−2.80[−1.03 - −4.58]mL/min, p=0.002). This trend that was reversed after switching to TAF with eGFR showing no further decline. For those receiving reduced-dose TDF, there was a trend towards a continued yet more limited eGFR decline following the switch to TAF.

Figure 3.

eGFR Changes according to TDF dose before switching to TAF. General linear model of eGFR changes among NA-experienced patients receiving full dose or reduced-dose TDF

Other Clinical Outcomes

Patients treated with TDF before switching to TAF were categorized, according to their ALT levels at baseline based on the current AASLD guidelines(Figure 4). Among NA-experienced patients with abnormal ALT at baseline, ALT declined significantly after switching to TAF compared to TDF(TDF:+0.002[−0.001 – 0.005]log10 ULN U/L/month, p=0.2); TAF:-0.005 [−0.006 – −0.004]log10 ULN U/L/month, p<0.001). Among the NA-naïve group, patients had a significant ALT decline after starting TAF(−0.006[−0.008 - −0.004]log10 ULN U/L/month, p<0.001; Figure S1).

Figure 4.

ALT Changes in NA-Experienced Patients according to ALT levels at Baseline. General linear model of ALT changes among NA-experienced patients with or without abnormal ALT at baseline

Liver stiffness was measured using FibroScan. Among NA-experienced patients, liver stiffness significantly decreased after starting TAF, compared with TDF treatment period(median [IQR] (−0.3 [−1.1 – 0.0]kPa, p=0.05). During the TAF treatment period, total Cholesterol and LDL Cholesterol showed a mean change of −0.03[−0.09 – 0.03]mmol/L, p=0.3 and −0.04[−0.10 – 0.02]mmol/L, p=0.2, respectively. Throughout this study, none of the patients developed hepatic decompensation, hepatocellular carcinoma or underwent liver transplantation.