We read with great interest the review by Rudnik and Perlmutter1 on the mechanisms of carcinogenesis in PiZZ α1-antitrypsin deficiency (A1ATD). We would like to endorse their hypothesis and previous animal models2 by describing the development of hepatocellular carcinoma (HCC) in a 12-year-old girl with end-stage liver disease related to PiZZ A1ATD.
The diagnosis of PiZZ A1ATD was made at the age of 3.5 years following investigations into the child's hepatosplenomegaly. She was noted to have a history of unexplained prolonged neonatal jaundice and mild developmental delay. Her clinical course was complicated by an episode of hematemesis at age 4 that required variceal sclerotherapy. At age 11, she suffered a blunt abdominal trauma. A computed tomography scan suggested a splenic infarction and significant splenic varices, but no intrahepatic lesions. The child recovered clinically under conservative management. Eight months later, a follow-up ultrasound scan detected a subcapsular nodule (diameter 2.6 cm) in segment V of the right lobe. A biphasic computed tomography scan indicated arterialization of the nodule and the presence of small para-aortic and mesenteric nodes. Her serum α-fetoprotein level was 4 (normal value <7 IU/L), her carcino-embryonic antigen level was less than 2 (nv <5 μg/L), and her Ca-199 level was 11 (nv <37 kU/L). The child was scheduled for liver transplantation, and a hepatectomy 5 months later confirmed a single subcapsular nodule measuring 0.8 × 2.5 × 1.5 cm located in the right lobe of the liver, which exhibited cirrhosis. Six months after the operation, the child is well with normal graft function and no signs of recurrence.
The tumor was composed of two-cell-thick liver cell plates surrounded by diffusely capillarized CD34-positive sinusoids. Lesional hepatocytes showed minimal cytological atypias and no mitotic activity. The proportion of proliferating hepatocytes was less than 1% using Ki67 (MIB-1) immunostaining (Fig. 1A). Only about a third of the lesional hepatocytes contained diastase-resistant periodic acid Schiff (dPAS)-positive fine intra-cytoplasmic granules, confirmed to be α1-antitrypsin via immunohistochemistry (Fig. 1B). Hepatocytes with α1-antitrypsin deposits were more present at the periphery of the tumor and in particular at the interface with fibrous tissue (Fig. 1C). The background liver showed periseptal hepatocytes containing abundant intracytoplasmic dPAS-positive deposits of variable size, ranging from fine granular material to globules up to 20 μm in diameter (Fig. 1D).
The extent of deposition of dPAS-positive material is an age-related phenomenon in PiZ A1ATD3 that is also observed in older individuals who carry only one PiZZ allele.4, 5 The diminished presence of dPAS-positive deposits in the HCC cells of our patient is compatible with the hypothesis that globule-containing periportal hepatocytes could generate a proliferative stimulus, leading to a higher turnover of globule-devoid perivenular hepatocytes and, ultimately, neoplastic change. Whether the abdominal trauma suffered 8 months earlier played a role in the development of HCC in our patient remains speculative. The minimal atypia, very low proliferative rate, slow clinical progression, and incipient accumulation of the dPAS-positive material in the lesional hepatocytes indicate that hepatocarcinogenesis in metabolic conditions, as recently postulated,1 may follow a different, less aggressive clinical course.