Pathogenic variants common in adult-onset cholestasis patients

The first oral presentation by Inbal Houri (Tel Aviv Medical Center, Israel) reported a genetic yield of pathogenic or possibly pathogenic variants in 35% of patients, while a poster by Filippo Gabrielli (University of Modena, Italy) and co-authors reported alterations in 19% of patients.

Houri and co-workers screened 203 patients (56% women, mean age 44 years) with unexplained cholestasis for 77 genes between 2016 and 2025. Overall, 51% of the patients (55% women, mean age 42 years) had one or more genetic variants of uncertain significance that may be associated with their cholestasis and 35% had a possible or confirmed pathogenic variant.

Houri explained that the majority (59.0%) of patients underwent screening after primary biliary cholangitis (PBC) or primary sclerosing cholangitis (PSC) had been ruled out, while 21.0% were screened after a diagnosis of sclerosing cholangitis without inflammatory bowel disease (IBD). A further 0.7% were screened because of a personal history of intrahepatic cholangitis of pregnancy (ICP) and 12.6% because of a significant family history of disease.

The 72 patients with at least one possible pathogenic variant included 32 patients with variants of the progressive familial intrahepatic cholestasis (PFIC) type 3 gene ABCB4, including nine carriers of two variants, four had variants of the PFIC type 1 gene ATP8B1, and one had a variant of the PFIC type 2 gene ABCB11. Three patients had variant forms of NOTCH, which is linked to Alagille syndrome.

The panel also identified variants in other genes linked to liver disease in systemic disorders, including the cystic fibrosis gene CTFR (n=25) and the renal disease genes PKHD1 (n=9) and HNF1B (n=7).

Of note, just 25% of patients with pathogenic variants reported having a relevant family history of cholestatic disease in their medical records, Houri told delegates.

Most of the patients with possible or confirmed pathogenic variants had “clear cholestatic presentation” with elevated gamma-glutamyl transferase, the presenter said, but she noted that there was a wide variation in the levels of laboratory markers, indicative of a “heterogenous population.”

Magnetic resonance imaging findings showed sclerosing cholangitis in 33.3% of the patients with a possible or confirmed pathogenic variant, a normal biliary tree in 31.9%, and nonspecific biliary tree findings in 27.8%.

Houri highlighted that 62% of the patients with pathogenic variants tested positive for antinuclear and smooth muscle autoantibodies and one was positive for antimitochondrial antibodies (AMA) without a clear PBC presentation; she advised delegates to be “wary not to call everything autoimmune just because of a positive AMA,” as that the exact contribution of these autoantibodies to each patient’s cholestasis is unknown.

Overall, 36% of the patients with possible pathogenic variants underwent cholecystectomy and 18% underwent endoscopic retrograde cholangiopancreatography; 16% had intrahepatic stones. The majority (67%) of patients were treated with ursodeoxycholic acid, 10% with fibrates, and 12% with other antipruritic therapy. Ileal bile acid transporter inhibitors were not available to this patient cohort.

Houri said that although the study was from a single center in Toronto and retrospective, it included a large, heterogeneous population with a “high” genetic test yield.

She encouraged clinicians to think “beyond the radiology” and consider genetic testing for patients with PSC, especially for younger individuals, those without IBD, and in the presence of other suspicious findings.

In addition, she questioned whether other cholestasis populations might benefit, such as those with severe ICP, early cholangiocarcinoma, early stone disease, or AMA-negative PBC.  

But acknowledging that “interpretation of genetic results is still unclear in many cases” and that there is a risk of overtesting, with possible implications for family planning and health insurance, for example, she concluded: “Therapeutic implications need to be investigated and clarified.”

The second presentation described 233 patients (46.4% men, average age at presentation 35.6 years) with adult-onset cryptogenic cholestasis, who attended four tertiary outpatient centers in Italy between 2017 and 2023 and were screened for up to 36 genes linked to cholestasis.

Overall, 19.3% of the patients had a pathogenic or likely pathogenic variant, and this yield increased to 45.5% when including those with variants of uncertain significance, Vitale et al reported.

The most common variants of pathogenic or likely pathogenic significance occurred in ABCB4 (n=27), TJP2 (n=18), SLC01B3 (n=10), ABCB11 (n=9), MYO5B (n=8), ABCC2 (n=7), VPS33B (n=6), and ATP8B1 (n=5).

The 45 patients carrying pathogenic or likely pathogenic variants had comparable findings to the 188 patients without such variants for liver stiffness, itch, use of choleretic drugs, and most blood test findings, but variant carriers did have significantly higher levels of bile acids (25.1 vs 13.7 µmol/L).

Patients with pathogenic or likely pathogenic variants were also significantly less likely to be male (33.3 vs 49.5%), and significantly more likely to be younger at testing (41.5 vs 46.8 years), and to have a history of ICP (29.7 vs 7.8%), juvenile cholelithiasis (39.5 vs 19.3%), or low phospholipid-associated cholelithiasis (11.6 vs 1.7%). Pathogenic or likely pathogenic variants were also more common among patients with a family history of PFIC risk factors (44.7 vs 24.6%), the researchers said.

Multivariate analysis confirmed that carriage of a pathogenic or likely pathogenic variant was significantly associated with ICP and juvenile cholelithiasis.

“Conducting natural history studies is essential for comprehending the possible progression of both young and old adult patients who have single or compound heterozygous [pathogenic/likely pathogenic] mutations, as well as the importance of [variants of unknown significance] found in many individuals,” the authors concluded.