Obesity and recurrence‐free survival in patients with hepatocellular carcinoma after achieving sustained virological response to interferon therapy for chronic hepatitis C

Abstract Aim Some patients who achieve a sustained virological response (SVR) to interferon (IFN) treatment for chronic hepatitis C prior to hepatic resection for hepatocellular carcinoma (HCC) experience postoperative recurrence. This study investigated the relationship between obesity and postoperative HCC recurrence in SVR patients. Methods Fifty‐nine patients who had achieved SVR before hepatic resection were evaluated. Patients had a solitary tumor ≤5 cm in diameter or ≤3 lesions each ≤3 cm in size with no macroscopic vascular invasion (Milan criteria). Patient characteristics potentially associated with recurrence risk were investigated. Results Three‐, 5‐, and 7‐year recurrence‐free survival after surgery were 65%, 44%, and 41%, respectively. Univariate analysis showed that obesity (P < .01), hypertension (P = .038), and non‐anatomical resection (P = .022) were significantly associated with a lower recurrence‐free survival rate. In a multivariate analysis, obesity (hazard ratio, 2.8; 95% confidence interval [CI] 1.3‐6.1; P < .01) and non‐anatomical resection (hazard ratio, 2.7; 95% CI 1.1‐6.2; P = .025) were independently associated with postoperative recurrence. Three‐, 5‐, and 7‐year overall survival rates after surgery were 100%, 80%, and 64% in obese patients and 100%, 92%, and 82% in non‐obese patients, respectively (P = .014). However, other variables showed no significant difference in the overall survival rate. Conclusions Obesity and non‐anatomical resection were independent risk factors for HCC recurrence after hepatic resection and successful IFN therapy. Obesity is an important clinical problem to consider to improve postoperative outcomes in such patients.


| INTRODUCTION
Hepatocellular carcinoma (HCC) is the sixth most common cancer worldwide, with an increasing incidence in Western countries, 1 and chronic hepatitis C virus (HCV) infection is one of the causes. Interferon (IFN) can reduce hepatic inflammation and fibrosis and reduce the risk of HCC recurrence after resection of HCV-related HCC, especially in patients with a sustained virological response (SVR). 2 However, the annual incidence of HCC recurrence after HCV eradication is 0.3%-2.4%. 3,4 The recognized risk factors for HCC development after SVR include advanced age, hepatic fibrosis, and alcohol abuse. 4,5 However, the risk factors associated with HCC recurrence after hepatic resection in SVR patients are unclear.
Obesity is a serious public health problem worldwide, and its incidence is increasing in Japan along with the Westernization of lifestyles. Obesity is closely associated with the development of metabolic syndrome, which includes hypertension, cardiovascular disease, and diabetes mellitus, and is independently associated with the risk of developing several malignancies, including breast, renal, and pancreatic cancer. 6 Obesity has also been reported to be independently associated with the risk of hepatocarcinogenesis 6,7 resulting from the accumulation of fat in the liver and consequent hepatocellular injury. 6,8,9 Obesity may influence HCC recurrence, but its involvement in postoperative recurrence in SVR patients with hepatic resection for HCC has not been investigated.
In the present study, we estimated the impact of obesity on HCC recurrence after hepatic resection following successful IFN therapy for HCV infection.

| Patients
A total of 68 patients who were seronegative for hepatitis B surface antigen and seropositive for anti-HCV antibodies but had an SVR and a history of curative hepatic resection for HCC between January 1994 and December 2015 at the Department of Hepato-Biliary-Pancreatic Surgery, Osaka City University Hospital, were eligible. SVR was defined as a serum HCV-RNA under the detection sensitivity limit at 6 months after the termination of IFN therapy.

| Metabolic syndrome-related factors
The metabolic syndrome-related factors assessed were obesity, diabetes mellitus, dyslipidemia, and a history of hypertension. Obesity was defined as a body mass index (BMI) ≥25 kg/m 2 following the criterion for adult Asian populations. 10 Diabetes mellitus was defined as a fasting serum glucose ≥126 mg/dL or the use of antidiabetes medications. 11 Dyslipidemia was defined as serum triglycerides ≥150 mg/dL or the use of indicated medications. 12 A history of hypertension was determined by interview. To evaluate the change in BMI after surgery, body weight measured at 3, 6, or 12 months after surgery was carried out.

| Surgical treatment
Indication for surgery was based on an algorithm consisting of the presence/absence of ascites, serum total bilirubin level, and results of the indocyanine green (ICG) retention test. 13 For anatomical resection, we divided the hepatic parenchyma along the demarcation line appearing after the occlusion of the portal vein and hepatic artery in hemihepatectomy and sectoriectomy, or after the injection of ICG into the portal vein conflicting the tumor-bearing area in segmentectomy. In non-anatomical resection, we divided the hepatic parenchyma along the line to maintain a surgical margin of 5 mm.

| Detection of recurrence
Patients were evaluated by serum alpha-fetoprotein, ultrasonography, computed tomography, or magnetic resonance imaging 1 month after surgery and every 3 months thereafter. If recurrence of HCC was suspected, selective hepatic angiography and an ultrasoundguided biopsy were used to establish a definitive diagnosis, as required. Sites of recurrence were categorized into near the resection site (NR), which was the residual part of the tumor-bearing third-order portal branches and the part adjacent to the liver transection surface, or distant from the resection site (DR).

| Histology
Histological evaluation followed the guidelines of the Liver Cancer Study Group of Japan. 14 Grade (severity of active hepatitis) and stage (degree of hepatic fibrosis) of non-malignant hepatic tissue were scored using a histological activity index 15,16 including periportal necrosis with or without bridging necrosis, intralobular degeneration with focal necrosis, portal inflammation, and fibrosis.

| Statistical analyses
Differences in clinicopathological findings were tested with the Mann-Whitney U-test and Fisher's exact probability method. Kaplan-Meier method was used to calculate recurrence-free survival rates, and between-group differences in the survival were evaluated by log-rank test. Cox's proportional hazard model was used for multivariate analysis.
Variables potentially associated with recurrence were selected based on previous study results or on our own clinical experience 5,8 and included age (<65 or ≥65 years); gender; history of alcohol abuse Influence of metabolic syndrome-related variables and clinical and pathological characteristics on recurrence-free survival was analyzed. P-values were derived from two-tailed tests, and the threshold for statistical significance was set at P < .05.

| Patient characteristics
Clinicopathological characteristics of the 59 patients are shown in Table S1. Mean age was 66 years (range, 48-76 years), 48 were men and 11 women, 31% were obese, 29% had diabetes mellitus, 20% had dyslipidemia, and 44% had hypertension. Twenty-seven patients had hepatic steatosis scores of 0, 17 had scores of 1, 11 had scores of 2, and four had scores of 3.
The 3-, 5-, and 7-year overall survival rates after surgery were 100%, 89%, and 77%, respectively. A total of 11 patients died during the study period, nine as a result of cancer-related causes and two from other causes. Among obese patients, all five patients died of cancer-related causes. The 3-, 5-, and 7-year overall survival rates after surgery were 100%, 80%, and 64% in obese patients and 100%, 92%, and 82% in non-obese patients, respectively (P = .014). Other variables showed no significant difference in the overall survival rate (Table S2).

| Characteristics of obese and non-obese patients
Obesity was an independent risk factor for recurrence in this patient series ( Table 3). Comparison of the clinicopathological characteristics of obese and non-obese patients showed that age, gender ratio, history of alcohol abuse, presence of diabetes mellitus, tumor characteristics, and type of surgery in obese and non-obese patients were not significantly different. However, the prevalence of hepatic steatosis was significantly higher in obese patients (72%) than in non-obese patients (56%) (P = .023). Furthermore, we evaluated the change in BMI after surgery in obese and non-obese patients. Of 59 patients, 50 patients had body weight measurements taken at 3, 6, or 12 months after surgery without recurrence. In obese patients

| DISCUSSION
Multivariate analysis confirmed obesity and non-anatomical resection as independent risk factors for postoperative HCC recurrence after successful IFN therapy, as the risk of recurrence was increased 2.8and 2.7-fold, respectively, in patients with these factors. To our knowledge, this is the first report of obesity as an independent risk factor for postoperative recurrence of HCC in patients with an SVR.
An SVR might be expected to reduce the background hepatocarcinogenic potential in patients with a history of chronic hepatitis C and HCC surgery, 2 as the tumor recurrence rate was 49%, and the recurrence-free survival was 41% at 7 years after hepatic resection. These results indicate that the risk of HCC recurrence persists for a long time in patients who achieve an SVR.
Obesity may promote hepatocarcinogenesis. 6,7 In a large, popula- young adulthood was significantly associated with an increased risk of HCC. 18 Ohki et al reported that a BMI >25 was an independent risk factor for HCC development in chronic hepatitis C patients, with a relative risk of 1.86. 8 In this study, obese patients had a significantly increased risk of HCC recurrence. The cause of this increased risk is not clear, but adipokine-related hepatocarcinogenesis might be involved. Obesity has been reported to affect hepatocarcinogenesis through adipose tissue remodeling with the secretion of proinflammatory adipokines and accumulation of ectopic lipids, resulting in liver lipotoxicity. 9,19 Level of serum adiponectin is inversely correlated with BMI, and a low serum adiponectin has been associated with an increase in the risk of various cancers. 8,9 Obesity may also promote hepatocarcinogenesis through hepatic steatosis, which can cause hepatic inflammation by a mechanism similar to that of nonalcoholic steatohepatitis with pathological findings of lobular inflammation and hepatocellular injury (ballooning). 20 However, in this study, hepatic steatosis, lobular inflammation, and ballooning were not independently associated with HCC recurrence. One possible reason for this discrepancy is the persistent influence of HCV infection. HCV infection can cause hepatic steatosis, and this steatosis persists even after HCV eradication. 3 Another possible reason is the influence of alcohol abuse. Alcohol abuse is known to cause hepatic steatosis and steatohepatitis. These factors may have promoted hepatic steatosis, lobular inflammation, and ballooning in this study.
Hepatocellular carcinoma recurrence after curative treatment includes intrahepatic metastasis of original tumors and newly developed tumors (multicentric recurrence). 21 Time to recurrence after surgery (early recurrence within 2 years of surgery or late recurrence at 2 or more years after surgery) is commonly used to distinguish the type of recurrence. 22 In the present study, recurrence-free survival showed that the difference in the recurrence rate between obese and non-obese patients increased, even in the early postsurgical period. However, on examining the patterns of recurrence between obese and non-obese patients, multiple recurrences near the resection site were similarly observed between the two groups, whereas solitary recurrence distant from the resection site tended to be more frequently observed in obese patients than in non-obese patients. Theoretically, recurrence derived from residual micrometastases can develop near the resection site. Furthermore, multiple recurrences near the resection site are considered likely to be a result of intrahepatic metastasis. In contrast, solitary intrahepatic recurrence distant from the resection site may be considered a pattern of multicentric recurrence. 22 Therefore, multicentric recurrence may be associated with an increased risk of HCC recurrence in obese patients.
Non-anatomical resection was also an independent risk factor for HCC recurrence. Anatomical resection along the portal tributaries is considered an effective way to remove occult intrahepatic HCC metastases. 23,24 In the present study, the patients who underwent anatomical resection had no multiple recurrences near the resection site. Furthermore, solitary recurrence distant from the resection site was more frequently observed in the patients who underwent anatomical resection than in those who underwent non-anatomical resection. Therefore, recurrence derived from micrometastases of the original tumors might be markedly reduced by anatomical resection. We previously reported that anatomical resection might prevent HCC recurrence in patients achieving an SVR and a biochemical response (defined as normal ALT activity) for at least 1 year. 25 Similarly, anatomical resection might be recommended for patients with HCC following HCV eradication.
In the present study, several common risk factors for HCC recurrence showed no significant difference between obese and non-obese patients, including liver-related factors 3,4 (albumin, ALT, T-Bil, and liver cirrhosis) and tumor-related factors 23 27 In the present study, only obesity was identified as a significant risk factor for death. However, given the small number of deceased patients, the accumulation of more patients and a further investigation will be necessary to confirm the prognostic factors for overall survival.
Improvement of obesity theoretically can be related to less recurrence or better overall survival. However, in the present study, only four patients showed improvement of obesity. Therefore, further investigation is necessary to clarify the impact of improvement of obesity on recurrence or death. have not been determined. 28,29 This study may be useful as a reference for further investigations of HCC recurrence after achieving SVR induced by DAA.
In conclusion, obesity and non-anatomical resection were independent risk factors for HCC recurrence after liver resection following successful IFN therapy for chronic hepatitis C. Obesity is an important clinical problem to consider in order to improve postoperative outcomes in such patients.