Prior hepatitis B virus infection as a co-factor of chronic hepatitis C patient survival after resection of hepatocellular carcinoma

From 2004 to 2016, HBsAg-negative patients who underwent curative liver resection for HCC in Nihon University Itabashi Hospital were included in this study. Each participant provided written, informed consent, and this study was approved by the institutional review board of Nihon University. All patients were closely observed during each outpatient visit post-surgery. PBI patients were defined as those who were negative for HBsAg, but positive for HBcAb [22, 23]. HCV patients were defined as those positive for HCV antibody and HCV-RNA detected. Clinical characteristics and outcomes were compared between patients with and without PBI.

Patients who underwent initial liver resection for HCC, but were positive for HBsAg were excluded from the study. Patients who were negative for HBsAg and HBcAb but positive for HBsAb, and those who were positive for HCV antibody but negative for HCV-RNA were also excluded. In addition, patients whose HBsAb or HBcAb status had not been determined were also excluded.

Indicators for liver resection were determined by assessing the liver functional reserve according to Clinical Practice Guidelines for Hepatocellular Carcinoma in Japan [24]. Briefly, liver resection was contraindicated in patients who had refractory ascites, hepatic encephalopathy, or both [25]. Patients with up to three lesions were candidates for liver resection.

In order to assess the existence of esophageal varices, gastrointestinal endoscopy was performed preoperatively for all candidates considered eligible for liver resection. Patients with F3 varices (largest size) or F2 varices (enlarged tortuous) or blue varices positive for red color signs were treated prophylactically using esophageal variceal ligation [26].

Liver resection was performed on all patients according to criteria based on liver function, as described previously [27]. Briefly, transection of the liver was performed under ultrasonographic guidance using the clamp-crushing method with the inflow-blood-occlusion technique. Curative resection was defined as the complete removal of recognizable HCC diagnosed preoperatively. Postoperative complications were stratified according to the Clavien-Dindo classification [28], which defines morbidities as complications with a score of ≥ IIIa. Complications specific to liver resection were defined as previously described [29].

All patients were followed for postoperative recurrence, as described previously [30]. Briefly, tumor marker levels were measured, and imaging studies, including computed tomography and ultrasonography, were performed every 3 months on all patients. Recurrence was diagnosed by dynamic computed tomography and/or gadolinium-ethoxybenzyl-diethylenetriamine pentaacetic acid-enhanced magnetic resonance imaging. The date of recurrence was defined as the date of examination when the recurrence of HCC was detected. In patients with recurrent HCC, the recurrence-free period was defined as the time from the date of surgery to the date of recurrence. Recurrent HCC was managed aggressively by repeated liver resection, transcatheter arterial chemoembolization, radiofrequency ablation, and chemotherapy according to the HCC status and liver function at the time of recurrence. Liver function was estimated 6 months after operation based on Child-Pugh classification.

Data collected from each group were statistically analyzed with Fisher’s exact test and the Wilcoxon rank-sum test. Survival curves were generated using the Kaplan–Meier method and compared by the log-rank test. Prognostic factors for overall survival were identified with the Cox proportional hazards regression model. A p value of less than 0.10 was set as the cut-off value for elimination. The following 25 variables considered as potential confounders were examined; age (≥ 75 years vs < 75 years), sex, PBI, alcohol abuse, diabetes mellitus, esophageal varices, AST (≥ 80 U/L vs < 80 U/L), ALT (≥ 80 U/L vs < 80 U/L), platelet count (≥ 10 × 10⁴ μL vs < 10 × 10⁴ μL), creatinine (≥ 1.2 mg/dL vs < 1.2 mg/dL, Child-Pugh classification (A vs B), indocyanine green clearance rate at 15 min (≥ 15% vs < 15%), serum alpha-fetoprotein level (≥ 100 ng/mL vs < 100 ng/mL), serum des-gamma-carboxy prothrombin level (≥ 100 ng/mL vs < 100 ng/mL), operation time (≥ 300 min vs < 300 min), clamp time (≥ 75 min vs < 75 min), bleeding (≥ 300 mL vs < 300 mL), transfusion, resection (anatomic vs non-anatomic), multiple tumors, tumor size (≥ 3.0 cm vs < 3.0 cm), differentiation grade (well-differentiated vs moderately differentiated vs poorly differentiated), tumor thrombus of the portal vein and hepatic vein, tumor exposure at surgery, and liver cirrhosis. In all analyses, a p value < 0.05 was considered to be statistically significant.

Of the 1053 total patients that underwent curative liver resection for HCC during the time period of the study, 872 (82.8%) were negative for HBsAg (Fig. 1). After excluding patients based on the criteria described above, 428 (40.6%) patients, including 165 (18.6%) patients with PBI, were positive for HCV antibody and HCV-RNA; and 317 (30.1%) patients, including 104 (9.8%) patients with PBI were negative for HCV antibody. The demographics and background information for the HCV-positive patients are listed in Table 1. The frequency of alcohol abuse also trended higher in the PBI patients compared to the non-PBI patients (23.6% vs. 16.7%, respectively), but statistical significance was not detected (p =  0.080). As seen in Table 2, there was no significant or trending variable between the PBI and non-PBI patients in the group of HCC patients negative for HCV antibody (non-B non-C).

Among the patients positive for HCV antibody and HCV-RNA, the amount of bleeding during surgery for PBI patients (median 316 mL; range, 5–3887) was significantly greater compared to that of non-PBI patients (240 mL; 10–4530; p = 0.029) (Table 1). In comparison, there was no significant variable between the PBI and non-PBI patients negative for HCV (Table 2). Complication rates and histological diagnosis were also not significantly different between the two PBI groups in both HCV-positive and HCV-negative patients.

After a median follow-up of 3.6 years (range, 0.2–14.2), a total of 290 patients (67.7%) experienced HCC recurrence; 277 patients (95.5%) in the remnant liver, eight patients (2.7%) in distant sites such as the lung, bone, and lymph nodes, and five patients (1.7%) with both intra- and extra-hepatic recurrences. The treatments for recurrent HCC were not significantly different between the two groups (Table 3). During the follow-up period, 105 (63.6%) patients with PBI and 190 (72.2%) patients without PBI were diagnosed as Child-Pugh classification A (p = 0.068).

In the patients with HCV-positive, the median overall survival of PBI patients was 4.7 years (95% confidence interval [CI], 3.9–5.9), which was significantly shorter compared with that of non-PBI patients (6.6 years, 5.3–9.8; p = 0.015) (Fig. 2). On the other hand, there was no significant difference in the median recurrence-free survival between the two groups (1.8 years, [95% CI, 1.4–2.0] vs 2.0 years, 1.7–2.3; p = 0.205). The overall survival and recurrence-free survival rates at 5 years were 45.9 and 17.0% in PBI patients, respectively, and 62.0 and 22.1% in non-PBI patients, respectively. On multivariate analysis, independent factors for overall survival were PBI (hazard ratio 1.38 [95% CI, 1.02–1.87]; p = 0.033), multiple tumors (p = 0.007), tumor size (p = 0.002), and liver cirrhosis (p <  0.001) (Table 4).

After a median follow-up of 3.0 years (range, 0.2–12.8 years) for the non-B non-C patients in the study, a total of 192 patients (60.5%) experienced HCC recurrence; 160 patients (83.3%) in the remnant liver, 22 patients (11.4%) in distant sites, and 10 patients (5.2%) with both intra- and extra-hepatic recurrences.

For the 104 patients with PBI and the 213 patients without PBI in this group, the median overall survival was 6.5 years (95% CI, 4.8–7.1) and 7.5 years (5.5–NA; p = 0.932), respectively. The recurrence-free survivals were 2.4 years (95% CI, 1.5–3.3) and 2.2 years (1.7–2.7; p = 0.983), respectively (Fig. 3). The 5-year overall survival rates were 61.2 and 59.9%, and 5-year recurrence-free survival rates were 25.3 and 27.7% in the two groups, respectively.