Patient risk model can help target costly HCV treatment to those with most urgent need
ANN ARBOR, Mich. -- A team of researchers at the University of Michigan Health System has developed a risk prediction model that helps identify which hepatitis C patients have the most urgent need for new anti-viral drugs.
The model, described in the June issue of Hepatology, uses routine lab values and machine-learning methods to help doctors predict the health outlook of patients diagnosed with hepatitis C.
Using a dataset from a previous National Institutes of Health study the Hepatitis C Antiviral Long-term Treatment Against Cirrhosis (HALT-C) trial, the U-M team used clinical data such as age, body mass index and virus type and routine lab measurements to estimate patients’ risk of progression of liver disease. “Offering immediate treatment to patients identified as high risk for poor health outcomes would allow these patients to benefit from highly effective treatments as other patients continue to be monitored and their risk assessment updated at each clinic visit,” says lead study author Monica Tincopa, M.D., MSc., a fellow in gastroenterology at the University of Michigan Health System.
The strength of the new model includes incorporation of many more lab values than most traditional models can handle. Plus, machine learning methods help analyze how lab values change over time, including the slope and acceleration of lab values such as platelet count, hepatic panel and AST to platelet ratio index (APRI), lab markers of liver injury and liver health.
Among the patients predicted as low-risk, only 6 percent will have cirrhosis (liver scarring) complications in the next year, compared to 56 percent in the high-risk group, according to the U-M study model.
“Ideally we would treat all patients. Until logistic and financial barriers are solved, clinicians and policy makers are faced with trying to target these therapies to patients with the most urgent need,” Tincopa says. “The model allows us to identify these patients with greater accuracy.”
The risk prediction tool can be added to an existing electronic medical record as a health care decision guide for doctors. It can establish how often patients come in for doctor visits or have monitoring tests. The groundwork is being laid to make care accessible and affordable, including drug cost discounts for certain health care programs, and increased competition among drug companies that could potentially drive down prices.
Additional authors: The paper's senior author is Akbar K. Waljee, M.D., an assistant professor of gastroenterology at the U-M Medical School and member of the U-M Institute for Healthcare Policy and Innovation. Co-authors are Anna S.F. Lok, M.D., Peter D.R. Higgins, M.D., Yiwei Zhang, of the University of Michigan Division of Gastroenterology and Ji Zhu, Ph.D., professor of statistics at the U-M College of Literature Science and Arts.
Reference: “Improvement of predictive models of risk of disease progression in chronic hepatitis by incorporating longitudinal data,” Hepatology, official journal of the American Association for the Study of Liver Disease, Vol. 61, Issue No. 6, June 2015.
Funding: Researchers were supported by the National Institutes of Health and a Veterans Health Services Research and Development career development award. Learn more about the Hepatology Program at the University of Michigan.
About U-M’s Division of Gastroenterology: The U-M is one of the largest gastroenterology practices in the country and is a leader in the prevention, diagnosis, and treatment of diseases of the gastrointestinal tract and liver. Our 50-plus physicians are experts in the diagnosis and treatment of all diseases of the gastrointestinal system, from simple to complex, including those of the esophagus, stomach, small intestine, colon, rectum, liver, gallbladder, pancreas and biliary tract.