Multicentre external validation of the GES score for predicting HCC risk in Japanese HCV patients who achieved SVR following DAAs

A simple score combining clinical and biochemical parameters (general evaluation score (GES)) has shown value in predicting hepatocellular carcinoma (HCC) risk after hepatitis C virus (HCV) eradication in Egyptian patients with HCV genotype 4. We aimed to apply the GES to predict HCC risk in Japanese HCV patients who achieved sustained virological response (SVR) following direct‐acting antivirals (DAAs). This multicentre retrospective cohort study included 187 HCV patients without a history of HCC treatment who achieved SVR. The GES was calculated using pre‐ and post‐treatment data. The median age of the patients was 66 years; 49% were male, 89% had cirrhosis and 69% had HCV genotype 1. During the mean 36‐month follow‐up, 19 (10.2%) developed HCC. Regarding the pretreatment scores, 75 (40.1%), 58 (31.0%) and 54 (28.9%) patients had low‐, intermediate‐ and high‐risk scores, respectively. The 4‐year cumulative incidence (CumI) was 1.64% in the low‐risk group, 2.82% in the intermediate‐risk group and 6.88% in the high‐risk group (log‐rank P = .029). In patients with cirrhosis, 60 (36.1%), 57 (34.3%) and 49 (29.5%) had low‐, intermediate‐ and high‐risk scores respectively. The 4‐year CumI was 0.98% in the low‐risk group, 2.86% in the intermediate‐risk group and 6.67% in the high‐risk group (log‐rank P = .02). The GES calculated with pretreatment data was more useful than that calculated with post‐treatment data (Harrell's C statistic: 0.670 vs 0.587). This tool incorporates changes over time to estimate variations in HCC risk and could help identify low‐risk patients for whom HCC surveillance can be discontinued.

Shiha et al demonstrated and validated the usefulness of a simple HCC risk score, the general evaluation score (GES), which could identify three patient groups with different levels of HCC risk during follow-up, utilizing simple and readily available predictors and thus can be easily assessed in clinical practice, namely, older age, male sex, low albumin and high AFP levels and the presence of cirrhosis at baseline. 6 In response to the authors' recommendation to evaluate the GES in different patient populations and ethnicities, we attempted to validate the GES in 187 Japanese HCV patients with cirrhosis and advanced liver fibrosis, who achieved a sustained virological response (SVR) following DAA therapy and were recruited from the Fukushima Liver Academic Group (FLAG) and its satellite centres between December 2014 and April 2019. The average duration of follow-up was 35 months after the initiation of DAA therapy.

| Cohorts
Of the 689 patients enrolled, this study included 187 HCV patients with cirrhosis and advanced liver fibrosis, who achieved SVR be-

| Diagnosis of cirrhosis
Patients were diagnosed with cirrhosis (F4) when they fulfilled more than one of the following criteria; • Definite clinical signs and laboratory parameters of cirrhosis (eg splenomegaly, ascites, albumin ≤3.5 g/dL, platelet count ≤100,000/μl);

Lay Summary
The eradication of hepatitis C virus (HCV) with directacting antivirals (DAAs) has been shown to significantly reduce the risk of hepatocellular carcinoma (HCC). Moreover, the predictors of HCC occurrence after DAA therapy in patients without a history of HCC treatment have not been clearly elucidated. We validated the usefulness of a simple HCC risk score, the general evaluation score (GES), which utilizes simple and readily available predictors, namely, older age, male sex, low albumin and high AFP levels and the presence of cirrhosis at baseline, that can be easily assessed in clinical practice. This dynamic tool incorporates changes over time to estimate variations in HCC risk and might help to identify patients with low risk for whom could be followed up at longer intervals.
• Abdominal ultrasonographical signs suggestive of cirrhosis (eg PV dilatation, presence of collaterals, coarse or nodular echo pattern, mild splenomegaly, minimal ascites); • LSM (>13.0 kPa) Patients were classified as having advanced liver fibrosis (F3) by FibroScan (>10 and ≤13 kPa) depending on EASL Recommendations on Treatment of Hepatitis C 2018. 7 Patients were not included in the current study if their liver stiffness measurement was less than 10 kPa.

| Diagnosis of HCC
The diagnosis of HCC was made in accordance with EASL 8 and AASLD 9 guidelines. Multiphase CT or MRI was performed if there were any focal hepatic lesions diagnosed by abdominal ultrasound and/or if AFP was >20 ng/mL. MSCT. The diagnosis of HCC was based on the characteristic arterial enhancement and early washout in the delayed phase.
The macroscopic classification, histological differentiation and architectural pattern were evaluated according to the general guidelines and standards for clinical and pathological studies of primary liver cancer. 10

| GES score
The General Evaluation Score (GES) was then derived using hazard ratios (HRs) of the variables in the multivariable model (Table S1). 6 Patients were then stratified into three groups based on the GES

| Ethical consideration
The study protocol conformed to the ethics guidelines outlined in the Declaration of Helsinki. The study protocol was reviewed, and opt-out consent was approved by the Ethics Committee of Fukushima Medical University (No. 29021). The need to obtain informed consent from the participants was waived by the Ethics Committee of Fukushima Medical University because of the retrospective nature of the study. The study was conducted in accordance with relevant guidelines.

| Analysis according to pretreatment risk
The total number of patients included in the analysis was 187 (21 non- and 10 developed in the high-risk group (10/54, 18.5%) ( Table 2). The 4-year cumulative incidence was 1.64% (95% CI = 0.52-3.95) in the low-risk group, 2.82% (95% CI = 1.03-6.24) in the intermediate-risk group and 6.88% (95% CI = 3.49-12.26) in the high-risk group. Analysis of the cumulative incidence of HCC showed a significant difference among the three risk groups (P = .029, Figure 1A). Harrell's C statistic for this external validation group was 0.6703. incidence of HCC showed a significant difference among the three risk groups (P = .020, Figure 1B). Harrell's C statistic for this external validation group was 0.6960.

| Analysis according to post-treatment risk
The total number of patients included in this analysis was 97 (13 F3 noncirrhotic patients and 84 cirrhotic patients), and the HCC patients    Figure 2A). Harrell's C statistic for this external validation group was 0.5874.

TA B L E 3
Characteristics of studied patients according to the development of HCC and post-treatment risk Figure 2B). Harrell's C statistic for this external validation group was 0.6196.

| D ISCUSS I ON
In this study, we externally validated a simple predictive score for HCC (called the GES) comprising routinely available laboratory parameters (albumin and AFP) plus age and sex. The original GES study has some limitations, as all patients in the discovery and validation cohorts had genotype 4; thus, it remains unclear whether their findings are valid in DAA-treated HCV patients with other genotypes. 6 The GES in our Japanese FLAG cohort (mainly genotypes 1 and 2) successfully stratified the patients with pretreatment scores into three A new assessment tool for hepatic function, the ALBI score, which consists of only albumin and bilirubin measurements, has recently been proposed. 15,16 The ALBI score may be used to evaluate liver function damage and the prognosis of patients with liver cancer. 17,18 Several HCC risk models using the ALBI score have been developed and validated to stratify patients based on their risk of HCC development. 19,20,21 The GES utilizes simple and readily available predictors and thus can be easily employed in clinical practice.
To our knowledge, this is the first study to assess the performance of an HCC risk score (GES) among DAA-treated Japanese patients with HCV genotypes 1 and 2. These findings showed that the GES had excellent performance for HCC risk stratification in DAAtreated HCV patients with other genotypes. However, our study has several limitations, which should be noted. First, the sample size was relatively small. In addition, we used a retrospective design; thus, our results need further confirmation in a prospective study. Moreover, Harrell's C-statistic was only 0.6703, which is fair.
In conclusion, the GES was a useful score for Japanese patients after HCV eradication. Post-SVR patients with high-risk scores should be monitored carefully for the possible development of HCC.

ACK N OWLED G EM ENTS
The authors thank Hiroto Wakabayashi (Department of Gastroenterology, Takeda General Hospital), Haruo Nakayama

CO N FLI C T O F I NTE R E S T
The authors declare no conflicts of interest.

AUTH O R S ' CO NTR I B UTI O N
GS designed the study. KA provided the external validation Cohort.
KA, MF, MH, AT and HO supervised clinical work. NM performed the statistical analyses. GS, NM and RS interpreted the data. All authors drafted the paper, provided input for the manuscript and approved the final version.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author, [K.A], upon reasonable request.