We read with interest the article by Liang et al., presenting a systematic review and meta-analysis of cancer risk in primary biliary cirrhosis (PBC) patients. We congratulate the authors for their accuracy and the exhaustive data presented, which clearly demonstrate that PBC is closely associated with a greater risk of hepatocellular carcinoma (HCC), but not of other extrahepatic cancers. We would like, however, to highlight some key points.
PBC is characterized by a long natural history that is influenced by a number of factors that modulate mortality risk, including: clinical symptoms/signs of liver disease, associated autoimmune conditions, comorbidity with other nonautoimmune-associated conditions (i.e., osteoporosis), and response to ursodeoxycholic acid (UDCA).
Table 1 summarizes the 10 most important studies dealing with the incidence of HCC in PBC patients, also reported in Liang et al.'s review article, for a total of 6,040 PBC patients. Five studies report a relationship between HCC and PBC histological stage; all of them clearly indicate that HCC arises in advanced histological stages. This behavior is most commonly observed in all types of liver disease in Western countries, with viral and nonviral etiology, with only very few exceptions.
|Author||Country||Study Period||No. PBC Patients||No. HCC||Histological Stage||Risk Factors|
|Floreani A||Italy||1973–1996||175||4||IV||HCV, cigarette smoking|
|Cavazza Aa||Italy||1973–2007||327||11||III–IV||Male gender|
|Jones D||UK||1975–1995||667||16||IV||Male gender|
|Shibuya A||Japan||1980–1998||396||14||III–IV||Male gender, age at diagnosis, history of blood transfusions|
|Jackson H||UK||1987–2002||930||7||NA||UDCA seems protective|
|Kuiper EMM||Holland||1990–2007||375||9||2 III/7 IV||Lack of response to UDCA|
These data also confirm our previous observation, i.e., that the relative risk for HCC in female patients with stage IV PBC is similar to that of female patients with cirrhosis of different etiologies.
Only a few studies report a statistical analysis of the risk factors associated with HCC development. The most impressive data regard the association with male gender and the lack of response to UDCA (although these data should be considered with caution). In our opinion it is remarkable that UDCA may favorably influence the natural history of PBC in responders. There are no sufficient elements to consider UDCA as a protective agent toward HCC, however.
Screening for HCC with cross-sectional imaging, with or without alpha-fetoprotein at 6-month intervals, should be recommended for PBC patients with advanced stage disease (histological stage IV or Mayo prognostic score >4.1), or evidence of portal hypertension. Due to the lack of an evidence-based association with other extrahepatic liver malignancies, cancer screening in PBC patients should not differ from what is indicated in the general population.
Having said that, it must be also kept in mind that, overall, PBC-related HCC appears to be a low-impact problem, since in a consecutive series of about 3,000 consecutive cases of HCC diagnosed in Italy and registered in the ITA.LI.CA database, only 10 (0.03%) are diagnosed in PBC patients.
ANNAROSA FLOREANI, M.D.
FABIO FARINATI, M.D.
Department of Surgery, Oncology and Gastroenterology, University of Padova, Padova, Italy