Non-alcoholic fatty liver disease is an independent predictor of recurrence after resection of colorectal liver metastases

Abstract

Background: Obesity and metabolic syndrome have been associated with increased risk for recurrence after hepatic resection of colorectal liver metastases (CRLM). Although previous studies of hepatic fibrosis and fatty liver disease suggest a protective effect against intrahepatic recurrence (IHR), the impact of non-alcoholic fatty liver disease (NAFLD) on recurrence patterns has not been well-defined. Methods: Histologic evaluation of hepatic parenchymal disease was quantified using the validated NAFLD activity score (NAS) including sinusoidal dilatation, hepatic steatosis, and fibrosis for 365 CRLM patients undergoing hepatic resection between April 2003 and March 2007. These were compared to clinicopathologic factors, intraoperative findings including surgeon evaluation of gross hepatic appearance (normal, fatty, or fibrotic) and outcomes. Fisher’s exact test was used to examine the association between substantial NAFLD, defined as NAS ≥ 3, and other clinical characteristics with gross hepatic appearance. Recurrence-free survival (RFS) and overall survival (OS) were estimated using Kaplan-Meier methods. Cumulative incidence of recurrence patterns (any intrahepatic recurrence (IHR), extrahepatic recurrence only (EHR), and death without recurrence (DWR), were estimated using competing risks methods and compared using Fine and Gray regression. Results: Intraoperative hepatic appearance was not associated with NAFLD (p=0.16), body mass index (BMI) > 25 (p = 0.23), diabetes (p = 0.56), or chemotherapy (5-FU, FOLFIRI, FOLFOX, FOLFIRINOX, p = 0.73, 0.10, 0.43, 0.52, respectively). Median follow-up was 54.1 months (range:1–178). Five-year RFS was 29% (95%CI = 25–34%). Five-year cumulative incidence for IHR, EHR, and DWR were 35%, 26%, and 9% respectively. After controlling for known clinical confounders, NAFLD remained significantly associated with the risk of IHR (HR = 1.83, 95%CI = 1.08-3.10, p = 0.02, Figure 1). NAFLD (HR = 0.19 95%CI=0.05-0.78, p = 0.02), tumor size >5cm (HR = 1.87, 95%CI = 1.15-3.05, p = 0.01), and hepatic artery infusion chemotherapy (HR = 1.59, 95%CI = 1.05–2.41, p = 0.03) were independently associated with risk of EHR. NAFLD was not associated with OS (HR = 0.91, 95%CI = 0.59–1.40, p = 0.665), however, tumor size > 5cm (HR = 1.73, 95%CI = 1.28–2.34, p 200 (HR = 2.28, 95%CI = 1.35–3.86, p 3 (HR = 2.10, 95%CI = 1.56–2.83, p < 0.01) were associated with OS on univariate analysis. Conclusion: Intraoperative hepatic appearance during hepatectomy for CRLM is not associated with the histologic presence of NAFLD. Although NAFLD was not associated with OS, it was an independent predictor of time to IHR. NAFLD may create a tumorigenic milieu supportive of IHR. The presence of NAFLD offers prognostic information for risk of timing and location of recurrent disease.