Megan Brooks

November 09, 2009

November 9, 2009 (Boston, Massachusetts) — In contrast to a study published in 2008, a new study suggests that hepatitis B virus (HBV) infection does not increase the risk for pancreatic cancer.

The study was reported here at The Liver Meeting 2009, the 60th Annual Meeting of the American Association for the Study of Liver Diseases.

"We were very much intrigued by a report from M.D. Anderson that prior type B hepatitis, not current, was somehow associated with an increased risk of pancreatic cancer," study investigator Stuart C. Gordon, MD, of the Division of Gastroenterology and Hepatology at Henry Ford Health Hospital, Detroit, Michigan, told Medscape Gastroenterology. "We did a retrospective cohort study — the M.D. Anderson study was a matched control study — and we were unable to validate their findings."

Using administrative data from an integrated healthcare delivery system, Dr. Gordon and colleagues identified all patients who underwent HBV testing between 1995 and 2008. Individuals younger than 18 years, those with fewer than 2 clinical encounters, or those who were followed up for fewer than 30 days were excluded.

They divided patients into an HBV-negative cohort (negative hepatitis B surface antigen [HBsAg] and antibody to hepatitis B core antigen [anti-HBc], n = 28,719), previous exposure cohort (antibody to hepatitis B surface antigen [anti-HBs] and anti-HBc positive or anti-HBc positive alone, n = 5141), and active infection cohort (HBsAg and anti-HBc positive, hepatitis Be antigen positive, or HBV DNA positive, n = 404).

No differences were found between cohorts with regard to mean age (67.9 vs 68.0 vs 62.0 years) or diabetes (21.9% vs 23.3% vs 21.8%). The HBV-negative cohort was more likely to be female (50.0% vs 38.9% vs 33.2%; < .001), white (61.2% vs 39.3% vs 41.3%; < .001), and HIV negative (98.4% vs 90.7% vs 85.4%; < .001).

On univariate analysis, prior HBV infection was associated with pancreatic cancer, with a hazard ratio of 1.827 (95% confidence interval, 1.154 - 2.891; = .156). However, in a multivariate model that included race, sex, and age, "Prior type B hepatitis infection fell out as an important predictor; it was no longer clinically significant," Dr. Gordon told Medscape Gastroenterology. The hazard ratio on multivariate analysis was 1.411 (95% confidence interval, 0.877 - 2.271; = .156).

"Age and diabetes trumped prior type B hepatitis when looking at it in multivariate analysis," Dr. Gordon noted.

"Additional studies including histopathologic investigation of pancreatic cancer tissue for HBV DNA should help clarify any potential relationship between HBV and pancreatic cancer," the authors conclude in a meeting poster.

Robert G. Gish, MD, hepatologist and medical director of the Liver Disease Management and Transplant Program at the California Pacific Medical Center, San Francisco, who was not involved in the study, told Medscape Gastroenterology, "This is a reassuring retrospective data set that concludes that chronic hepatitis B viral infection does not increase pancreatic cancer risk. We look forward to a prospective study that can be performed in parallel to a liver cancer surveillance program, to confirm these very important findings."

The study was funded by Henry Ford Health System, Detroit, Michigan. Dr. Gordon and Dr. Gish have disclosed no relevant financial relationships.

The Liver Meeting 2009: American Association for the Study of Liver Diseases (AASLD) 60th Annual Meeting: Poster 1486. Presented November 2, 2009.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.