Background In an integrated analysis of 2 global Phase 3 studies, we evaluated HCC incidence and risk at 8 years in patients treated with TAF and those treated initially with TDF and then switched to TAF for up to 6 years (y).

Methods In 2 studies, HBeAg-positive (n=859) and -negative (n=439) patients with HBV DNA ≥20,000 IU/mL and ALT >60 U/L (males) or >38 U/L (females) were randomized to TAF or TDF in a double-blind (DB) phase for up to 3 y, followed by open-label (OL) TAF through Y8. HCC was assessed by local standards of care and by hepatic ultrasonography. Three validated models (REACH-B, aMAP, and mPAGE-B) were utilized to assess HCC risk by initial treatment assignment and collectively. Using the REACH-B model, standard incidence ratios (SIRs) for HCC were calculated.

Results Through Y8, HCC was diagnosed in 21/1298 patients (1.6%; TAF 1.4%; TDF→TAF 2.1%; P=0.33)— 11 in DB/10 in OL phase. Eight of 21 HCC cases were in cirrhotic patients. The median time to HCC onset was 729 days (TAF 1291, TDF→TAF 460 days). Advanced age, male, and cirrhosis were more common in HCC vs non-HCC patients (P<0.05). Proportionately, more HCC patients were HBV genotype C (76% vs 47%) and had BL HBV DNA between 6 to ≤ 8 log₁₀IU/mL (57% vs 38%). With treatment over 8y, by REACH-B, HCC incidence was significantly reduced (21 observed vs 74.6 predicted; SIR [95% CI] 0.28; P<0.0001; figure 1). Of patients predicted to be low risk for HCC at BL, nearly all remained low risk at Y8 by aMAP (98%) and mPAGE-B (97%), and substantial proportions estimated to be medium or high risk at BL shifted to lower risk at Y8 (aMAP: 45% and 72%; mPAGE-B: 27% and 51%, respectively).

Conclusions CHB patients treated with TAF alone or switched from TDF to TAF for up to 8 y showed a reduced observed vs predicted risk for HCC development.