Patients with autoimmune hepatitis had a higher risk of developing extrahepatic malignancies compared with matched controls in a large US-based cohort study.
“While autoimmune hepatitis (AIH) is known to increase liver cancer risk, its association with cancers outside the liver remains poorly defined, particularly within the US population,” study author Jake MacDonald, MD, MEd, told GI & Hepatology News ahead of Digestive Disease Week® (DDW) 2026, where the work was presented. “With over 31,000 propensity-matched AIH patients, this is the largest US study to address that question, and the findings have meaningful implications for how we monitor and counsel these patients.”
Using data from TriNetX, a federated network of 67 US healthcare systems, Dr. MacDonald, an internal medicine resident at Emory University, Atlanta, and colleagues conducted a retrospective cohort study of adults aged 18 to 75 years with autoimmune hepatitis (AIH). They excluded patients with prior malignancy or coexisting liver diseases, including hepatitis B, hepatitis C, primary biliary cholangitis, and primary sclerosing cholangitis, as well as those with HIV or prior liver transplantation. The AIH cohort was matched 1:1 with controls without AIH using propensity scores based on age, sex, cirrhosis, nicotine dependence, and heavy alcohol use.
After matching, the study included 31,837 patients with AIH, with a mean follow-up of 4.2 years compared with 4.7 years in controls. The overall incidence of extrahepatic malignancy was 6% in patients with AIH vs 5% in controls, reflecting a 23% higher risk of developing cancer.
Several types of cancer occurred more often in patients with AIH. Stomach cancer was observed in 0.16% of patients with AIH vs 0.05% of controls, representing more than 3.5 times the risk of controls. "A remarkable finding given its low baseline incidence in the US," Dr. MacDonald noted.
Gallbladder or biliary tract cancers occurred in 0.09% vs 0.05%, or about twice the risk. Hematologic malignancies, including myeloma and non-Hodgkin lymphoma (NHL), were reported in 0.98% vs 0.56%, corresponding to a 90% higher risk.
Other increased risks included pancreatic cancer (0.33% vs 0.21%; a 68% higher risk), lung cancer (0.93% vs 0.61%; a 66% higher risk), and skin cancer (1.0% vs 0.8%; a 38% higher risk).
In contrast, there were no meaningful differences in several common cancers. Colon cancer occurred in 0.56% of patients with AIH vs 0.50% of controls. Thyroid cancer rates were 0.44% vs 0.49%, and breast cancer rates were 1.96% vs 2.33%, with no statistically significant differences.
“The non-uniform distribution of cancer risk across tumor types suggests specific, potentially targetable biological mechanisms that merit further study,” Dr. MacDonald said.
So far, he continued, there are no guidelines specifically focused on monitoring extrahepatic cancers in patients with AIH. “Our findings argue for a more proactive approach, ensuring patients are current with age-appropriate screening and maintaining a low threshold for investigating new symptoms or laboratory abnormalities that may signal an underlying malignancy,” Dr. MacDonald said. “Formal surveillance guidelines will require prospective studies, but the data support clinical vigilance.”
At the meeting, session moderator Cynthia Levy, MD, Associate Director of the Schiff Center for Liver Diseases at the University of Miami, described the study as “well done and the level of research that we need to have, so I appreciate this.”
Limitations include the retrospective design and reliance on administrative coding, which may introduce misclassification bias. The database lacks detailed information on immunosuppressive therapy, disease severity, and duration of AIH, all of which may influence cancer risk. Additionally, residual confounding cannot be excluded despite matching. Follow-up duration, while several years, may not fully capture long-term cancer development.
According to Dr. MacDonald, propensity matching enabled the researchers to account for major cancer risk factors such as cirrhosis, smoking, and alcohol use, so that the observed differences in risk are more likely due to AIH and its treatment. “The key unanswered question is why — whether specific immunosuppressive medications, disease severity, or duration of illness drive these risks,” he said. “Answering that question will be essential to developing rational, evidence-based surveillance strategies for this population.”
Dr. MacDonald and his coauthors reported having no relevant disclosures.
DDW is AGA’s annual meeting, jointly sponsored by AGA, AASLD, ASGE, and SSAT. Learn more at ddw.org.