We previously reported that the sensitivity and specificity of anti-integrin αvβ6 antibodies for diagnosing PSC are 89.1% and 96.7%, respectively [6]. However, patients with PSC were recruited from only two institutions. The present study included all cases from the Japanese nationwide PSC registry study. We obtained results similar to those of the previous study, confirming that anti-integrin αvβ6 autoantibodies are useful biomarkers for PSC.

There are no specific biomarkers for PSC, and its diagnosis mainly depends on cholangiography. Thus, it is sometimes difficult to distinguish PSC from other hepatobiliary diseases such as CCA, IgG4-SC, AIH, or PBC [16,17,18,19]. Regarding autoantibodies, the perinuclear anti-neutrophil cytoplasmic antibody is reported to be positive in 26–94% of PSC, 22–88% of AIH, and 0–33% of PBC cases [20]. Anti-glycoprotein 2 antibody is reported to be positive in 30.8–52.2% of patients with PSC [21] and 36% of patients with CCA [22]. Radiological images of PSC also overlap with those of other biliary diseases such as CCA or IgG4-SC [23]. Although current European and American guidelines advocate magnetic resonance cholangiopancreatography as the preferred method over endoscopic retrograde cholangiopancreatography (ERCP) [17, 18, 24], some cases require ERCP to detect small or subtle biliary lesions [25, 26]. However, ERCP occasionally leads to severe complications such as fatal pancreatitis [27]. Anti-integrin αvβ6 autoantibodies could noninvasively contribute to diagnosing PSC accurately.

The genetic predisposition for PSC is shared with that of IBD [28,29,30], and 60–80% of patients with PSC have IBD in Northern Europe and the United States [3]. In the present study, 80.9% of patients with PSC had concomitant IBD, and 64.7% of patients with PSC had UC. The antibody concentrations in PSC patients with UC were significantly higher than those in PSC patients without IBD. These data suggest that the presence of the antibody in patients with PSC merely reflects accompanying UC. However, we found that more than half of the patients without IBD also had this antibody, and its concentrations in PSC patients without IBD were significantly higher than those in the controls. Notably, although two PSC patients without IBD who initially tested negative for anti-integrin αvβ6 antibodies at enrollment tested positive at the 1-year follow-up, they did not develop IBD during the follow-up period. Thus, the presence of this antibody indicates that it is associated with PSC as well as that of UC.

Regarding the negative correlation between autoantibody concentrations and age in patients with PSC, we speculate that accompanying IBD might have confounded the result. In this PSC registry, younger patients had IBD, especially UC. As indicated in the subgroup analysis in Fig. 4a, PSC patients with UC had higher antibody concentrations than PSC patients without IBD. Multivariate analysis revealed that the age of patients did not influence the antibody concentration whereas the presence of UC did.

Integrin αvβ6 is reportedly expressed in the colonic epithelium and bile duct epithelium, and IgGs from patients with PSC and UC exhibit inhibitory effects on the binding between integrin αvβ6 and fibronectin in vitro [4, 6]. Weil et al. reported a case of ITGB6 homozygous germline mutation, which showed lethal cholestatic liver injuries and bloody diarrhea, clinical features of PSC and UC, respectively [31]. Taken together, anti-integrin αvβ6 autoantibodies may play an important role in the clinical manifestation of PSC as well as that of UC.

In this study, we collaborated with Medical and Biological Laboratories Co., Ltd. to establish the Anti-Integrin αvβ6 ELISA Kit, which enables easier detection of anti-integrin αvβ6 antibodies. This ELISA kit includes antigen pre-coated strip plates, rendering it ready-to-use and eliminating the need for electrolyte addition to the buffer [4]. Furthermore, standard material facilitated the standardization of anti-integrin αvβ6 autoantibody titers. Notably, the results of this kit and those of the in-house method showed a significant correlation, and there was no significant difference between the two methods in the ROC curves for the diagnosis of PSC.

This study had some limitations. Although all patients were recruited from the Japanese nationwide PSC registry, the number of patients with PSC was limited. In addition, in this PSC registry, the ages of the patients were younger and the IBD prevalence was higher than those previously reported in a nationwide survey in Japan [32], potentially limiting the generalizability of our findings. Thus, large-scale studies involving other ethnicities are warranted. Moreover, only a few patients in the control groups underwent colonoscopy. Although no abdominal pain or diarrhea was reported in patients in any of the control groups, the possibility of undiagnosed IBD cannot be ruled out. Furthermore, whether anti-integrin αvβ6 autoantibodies have pathogenic roles in PSC manifestation or a secondary phenomenon was not elucidated. Nevertheless, our findings further support evidence that anti-integrin αvβ6 autoantibodies are a useful biomarker for diagnosing PSC.

In conclusion, this study confirmed that anti-integrin αvβ6 autoantibodies had high specificity and sensitivity for diagnosing PSC in the Japanese nationwide registry with a standardized ELISA kit. These autoantibodies are useful as non-invasive diagnostic biomarkers for PSC.