A case of ruptured hepatocellular carcinoma ( HCC ) in resolving non‐alcoholic steatohepatitis associated with type 2 diabetes: I s early detection of diabetes‐related HCC feasible?

At present, one of the most critical problems for patients with type 2 diabetes mellitus (T2DM) is digestive organ cancers, which are now considered a major cause of death. The risk of hepatocellular carcinoma (HCC), in particular, is approximately fourfold greater in diabetic compared with non-diabetic individuals. Non-alcoholic fatty liver disease (NAFLD), including non-alcoholic steatohepatitis (NASH), appears to play a significant role in T2DM-related hepatocarcinogenesis, but the detailed underlying pathological mechanism remains to be clarified. One would assume that individuals with T2DM and NAFLD should at least be closely followed-up as high-risk patients. But, is this sufficient as a strategy in the care of T2DM patients? Recently, we encountered a case of a ruptured HCC associated with T2DM. Herein we report on this case to alert physicians and pathologists and to instigate discussion about current and future strategies for T2DMrelated HCC. A 52-year-old man was urgently admitted to Takatsuki General Hospital because of intraabdominal hemorrhage. Imaging detected a 50-mm tumor in the right lobe of the liver (Fig. 1a) and a clinical diagnosis of a ruptured HCC was made. Embolization was performed immediately and controlled the bleeding successfully. In the past 2 years, the patient had been treated for T2DM and hypertension with metformin, linagliptin, and amlodipine. His body mass index was 24 kg/m and waist circumference was 84 cm. He was not a habitual drinker. Serum markers of hepatitis viruses and autoantibodies were negative, liver function tests had never been abnormal, and he had no history of chronic liver disease or tumor (Fig. 1b). Three weeks after admission, the liver tumor was surgically removed (Fig. 1c). Histological examination revealed complete coagulative necrosis (Fig. 1d), but the tumor could still be recognized as HCC (Fig. 1e). The background liver was non-cirrhotic and had steatosis in <5% of hepatocytes (Fig. 1f ), minimal acinar inflammation, and rare typical ballooned hepatocytes (Fig. 1h) containing Mallory–Denk bodies (MDB) (Fig. 1i). There was only Zone 3 sinusoidal fibrosis (Fig. 1g). Based on the currently accepted histological criteria for NAFLD and NASH, a diagnosis of steatohepatitis could not be made because of the absence of “significant” steatosis. However, the additional presence of hepatocyte ballooning with MDB and typical Zone 3 sinusoidal fibrosis supported our final clinicopathological diagnosis of HCC associated with “resolving NASH”. The antidiabetic drugs may have alleviated steatosis and disease activity, and the overall histological picture in the background liver represented NASH resolution. The presence of fat droplets in 5% of hepatocytes is currently approved as the threshold for the histological diagnosis of pathological fat accumulation in liver tissue, and is possibly the lowest level detectable with modern imaging devices. However, strictly adhering to this criterion for diagnosing steatosis entails the risk of missing the diagnosis of NAFLD or NASH and related complications. It is important to recognize that diabetes-related “NAFLD/NASH”, especially in treated patients, may present with a non-pathologic level (<5%) of steatosis and that HCC can develop asymptomatically, like a silent killer, in non-cirrhotic liver. Eventually, all diabetic patients need close follow-up with regular imaging examinations to possibly detect T2DM-related HCCs at an early stage.