The impact of diabetes and glucose-lowering therapies on hepatocellular carcinoma incidence and overall survival.

Abstract

Purpose: The incidence of hepatocellular carcinoma (HCC) in the United Kingdom has increased 60% in the past 10 years. The epidemics of obesity and type 2 diabetes are contributing factors. In this article, we examine the impact of diabetes and glucose-lowering treatments on HCC incidence and overall survival (OS).

Methods: Data from 1064 patients diagnosed with chronic liver disease (CLD) (n = 340) or HCC (n = 724) were collected from 2007 to 2012. Patients with HCC were followed up prospectively. Univariate and multivariate logistic regression determined HCC risk factors. Kaplan-Meier curves were used to examine survival and Cox proportional hazards analysis estimated hazard ratios (HRs) for death according to use of glucose-lowering therapies.

Findings: Diabetes prevalence was 39.6% and 10.6% within the HCC and CLD cohorts, respectively. The odds ratio for having HCC in patients with diabetes was 5.55 (P < 0.001). Univariate analysis found an increased association of HCC with age, sex, cirrhosis, hemochromatosis, alcohol abuse, diabetes, and Child's Pugh score. In multivariate analysis age, sex, cirrhosis, Child's Pugh score, diabetes status, and insulin use retained significance. Diabetes status did not significantly affect OS in HCC; however, in people with diabetes and HCC, metformin treatment was associated with improved OS (mean survival, 31 vs 24 months; P =0.016; HR for death = 0.75; P = 0.032).

Implications: Diabetes is significantly associated with HCC in the United Kingdom. Metformin treatment is associated with improved OS after HCC diagnosis. Treatment of diabetes should be appropriately reviewed in high-risk populations, with specific consideration of the potential hepatoprotective effects of metformin in HCC.