Persistently detectable serum HBV DNA and pgRNA is associated with subsequent HCC development in chronic hepatitis B patients receiving chronic NRTI treatment

Abstract

Background & Aims: Although antiviral therapy reduces the risk of hepatocellular carcinoma (HCC) in chronic hepatitis B (CHB) patients, HCC risk is not eliminated in those achieving undetectable serum hepatitis B virus (HBV) DNA by standard assays. We aimed to assess whether HBV DNA and pgRNA detected by highly sensitive assays are associated with HCC development in a retrospective case-control study. Methods: A total of 124 patients [62 HCC cases and 62 non-HCC controls (matched with age, gender and treatment duration) on ≥3 years of entecavir (ETV) with undetectable HBV DNA by Cobas Taqman assay version 2.0 (Roche Diagnostics; lower limit of detection [LLOD] 20 IU/mL) were included. Stored serum from 3 (-3 years), 2 (-2 years) and 1 (-1 year) prior to HCC diagnosis or last follow-up (LFU) were assessed using highly sensitive HBV DNA and pgRNA assays. Total viral nucleic acid was extracted and subjected to HBV DNA and total nucleic acid (HBV DNA + pgRNA) measurements. Detection of HBV DNA was performed using a semi-quantitative PCR assay with LLOD of 10 IU/mL. Quantitation of pgRNA was performed using a RT-qPCR assay with and lower limit of quantification (LLOQ) of 100 copies/mL with DNA standard or 20 IU/mL traceable to 1 IU/mL of WHO HBV standard (cpm). Serum HBV pgRNA level was derived from the difference between the total HBV nucleic acid levels and HBV DNA levels with LLOD of 10 IU/mL. Results: Among the 124 patients (59.6% male, median age 60.9 years old, duration of ETV 50.6 months), 117, 105 and 78 had retrievable samples -3 years, -2 years and -1 year prior to HCC or LFU, respectively. For HCC patients, only 27.4%, 43.4% and 33.3% had undetectable HBV DNA and only 9.7%, 9.4% and 10% had undetectable pgRNA by the highly sensitive assays at time -3, -2 and -1 year, respectively. For non-HCC patients, only 34.5%, 41.2% and 60.4% had undetectable HBV DNA and only 9.1%, 19.6% and 25% had undetectable pgRNA by the highly sensitive assays at time -3, -2 and -1 year, respectively (Figure 1). The median HBV DNA ([90.6 (IQR 0 – 90.6) vs. 0 (IQR 0 – 90.6) IU/mL, p=0.002] and pgRNA [(153 (IQR 9 – 4871) vs. 9 (IQR 2.25 – 9) IU/mL, p=0.015) was significantly higher in patients with HCC compared to those without at -1 year. Among the 71 patients with samples from all 3 time points, a significantly higher proportion of patients with HCC had quantifiable serum HBV pgRNA in ≥2 out of 3 samples compared to patients without HCC (60% vs 24%, respectively; p=0.002). Detectable HBV pgRNA at -2 years and -1 year was associated with subsequent HCC development (log rank P<.05), while detectable HBV DNA at -3 years showed trend for higher risk of HCC (log rank P=0.066). Conclusions: More than half of CHB patients on ETV with HBV DNA < LLOQ by a standard assay had persistent HBV DNA and pgRNA detected by highly sensitive assays which was associated with subsequent HCC development. More profound viral suppression is needed for further reduction of HCC risk in CHB patients.