Predicting Treatment Success for Hepatitis C Using IL28B Genotype
Author Information
Author(s): Thomas R. O'Brien, James E. Everhart, Timothy R. Morgan, Anna S. Lok, Raymond T. Chung, Yongwu Shao, Mitchell L. Shiffman, Myhanh Dotrang, John J. Sninsky, Herbert L. Bonkovsky, Ruth M. Pfeiffer, HALT-C Trial Group
Primary Institution: National Institutes of Health, Bethesda, Maryland, USA
Hypothesis
Can a model based on IL28B genotype and clinical variables predict the probability of sustained virological response (SVR) in chronic hepatitis C patients?
Conclusion
A clinical prediction model based on IL28B genotype and clinical variables can yield useful individualized predictions of treatment success for chronic hepatitis C.
Supporting Evidence
- IL28B rs12979860-CC genotype was the strongest predictor of SVR with an adjusted odds ratio of 7.56.
- The model achieved an area under the curve (AUC) of 78.5%, indicating good predictive ability.
- Subjects with a predicted probability of SVR <10% had an observed SVR rate of 3.8%.
Takeaway
Doctors can use a patient's genes to better guess if a treatment for hepatitis C will work, helping to avoid unnecessary treatments.
Methodology
The study used step-wise logistic regression and leave-one-out cross-validation to develop a predictive model for SVR based on IL28B genotype and clinical variables.
Potential Biases
Potential bias due to the exclusion of non-European American subjects and those with other liver diseases.
Limitations
The model was limited to European American patients infected with HCV genotype 1 and may not be generalizable to other populations.
Participant Demographics
The study included 646 European American patients with chronic hepatitis C, predominantly male (75.4%) with a median age of 49 years.
Statistical Information
P-Value
p<0.0001
Confidence Interval
95% CI, 3.20–17.87
Statistical Significance
p<0.001
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website