A retrospective study by Ali et al. on cancer surveillance in patients with primary sclerosing cholangitis (PSC) was published in June in Hepatology (DOI:10.1002/hep.29730). The study demonstrated that hepatobiliary malignancy was diagnosed at earlier tumour stages and that the outcome of PSC patients was significantly improved, when cancer surveillance was performed using cross-sectional imaging plus carboanhydrate antigen 19-9 levels. Further studies with standardized surveillance protocols are required to validate these findings.
One of the major clinical challenges in primary sclerosing cholangitis (PSC) is the increased risk for hepatobiliary malignancy. Cholangiocarcinoma (CCA) develops in approximately 10-20 % of patients over time and is regarded as the major cause of death in PSC. In addition, the risk for gallbladder cancer (GBC) and for hepatocellular carcinoma (HCC) is increased. There is an urgent clinical need to diagnose hepatobiliary cancer at early stages in PSC. It is, however, unclear, whether surveillance strategies, particularly for CCA detection, improve patients´ outcome and survival. As a consequence, current clinical guidelines recommend (bi)annual ultrasound for GBC surveillance, but are controversial with respect to CCA surveillance recommendations.
New insights on this issue come from a study by Ali et al., who have recently published their experience on cancer surveillance in PSC in Hepatology in June 2018 (DOI:10.1002/hep.29730). In this study Ali and colleagues reviewed medical reports from 830 newly diagnosed PSC patients from the Mayo Clinic Rochester between 1995-2015. 79 patients developed hepatobiliary malignancy during follow-up (78 % CCA, 21% HCC, 6% GBC, 3% mixed CCA/HCC, 1% HCC plus GBC) and went into the final analysis. Patients were then categorised according to their surveillance status prior to cancer diagnosis, which included (at least) annual imaging with ultrasound (US), computer tomography (CT) or magnetic resonance imaging (MRI) / magnetic resonance cholangiopancreatography (MRCP) and carboanhydrate antigen (CA) 19-9 levels (plus alphafeto protein levels in cirrhotic PSC patients). Surveillance was performed in 51% of patients. Primary study endpoints were the recurrence of hepatobiliary malignancy, hepatobiliary malignancy-related death and all-cause mortality.
The cumulative follow-up in patients who developed hepatobiliary malignancy was 712 and 283 person-years pre- and post-cancer diagnosis, respectively. 62% of patients died after cancer diagnosis. As compared to the no surveillance group, patients with PSC who underwent cancer surveillance had a significantly higher 5-year and 10-year overall survival (68% and 56% vs 20% and 13%). This also applied to hepatobiliary malignancy-related events (recurrence or cancer related death), which were significantly more frequent in the no surveillance group (75% vs 32). The most obvious explanation for the better outcome of the surveillance group was the earlier cancer diagnosis: lymph node-positive hepatobiliary malignancy and hepatobiliary malignancy with intra- and extrahepatic metastasis were significantly more frequent at cancer diagnosis, when patients had not undergone surveillance. As a result, patients in the surveillance group were more likely to receive liver transplantation as a curative treatment (65% vs 23%).
CCA is regarded as the highest burden for PSC patients. The subgroup of patients, who developed CCA, seemed to be too small for significant results. There was, however, a trend towards a lower frequency of extrahepatic CCA recurrence and increased survival of patients with intrahepatic CCA, if surveillance was performed. Interestingly, the overall survival depended on the imaging modality used for cancer surveillance with US showing similar outcomes as compared to MRI/MRCP but better outcomes as compared to CT.
By nature, the study of Ali et al. has several limitations. As a retrospective study, surveillance was determined based on clinicians´ preferences, indicating that there was no real randomization into the surveillance and no surveillance group. In addition, no information was provided on patients who underwent surveillance but did not develop hepatobiliary malignancy. Future studies should evaluate not only the potential benefit of cancer surveillance (ideally prospective and with standardized surveillance protocols) but also adverse events resulting from surveillance strategies (e.g. unnecessary surgical interventions due to false positive results). Still, the study of Ali et al. underlines the central meaning of surveillance strategies in PSC and we recommend a detailed look at this important publication in the field of PSC.
I. Department of Medicine, University Medical Centre Hamburg-Eppendorf, Hamburg, Germany.