The Validity of Biomarkers in Detecting Alcohol Dependence
Because of concerns about denial in the clinical history, the search continues for biomarkers for chronic alcoholism. To help identify valid biomarkers, researchers examined the performances of carbohydrate-deficient transferrin (%CDT),* gamma-glutamyltransferase (GGT), and mean corpuscular volume (MCV) in 101 current drinkers with alcohol dependence, 117 patients with liver disease (e.g., hepatitis, nonspecific elevated GGT, and nonalcoholic cirrhosis), and 115 social drinkers.
- The percent of patients with dependence who had a positive %CDT (sensitivity) was 73%. The sensitivities of GGT and MCV were 71% and 64%, respectively.**
- Sensitivity increased to 83% when %CDT was combined (through a mathematical equation) with GGT and to 88% when MCV was added to the combination.
- The percent of patients without dependence who had a negative test of %CDT (specificity) was 94% in those with elevated GGT, 88% in those with hepatitis, and 70% in those with nonalcoholic cirrhosis. Specificities of GGT and MCV in these patients were much lower.
The %CDT assay had better sensitivity in patients with alcohol dependence and better specificity in patients with various forms of liver disease than did the GGT and MCV. The combination of %CDT with GGT and MCV increased sensitivity to 88% without loss of specificity. Before widespread adoption of these biomarkers occurs, further study is needed to understand their costs and real-world performance alone and in combination with a skilled clinical interview.Peter D. Friedmann, MD, MPH
**The %CDT cut-off was 3% (set for 95% specificity in social drinkers), the GGT cut-off was 33 U/L for women and 56 U/L for men, and the MCV cut-off was 95 fL.
Hock B, Schwarz M, Domke I, et al. Validity of carbohydrate-deficient transferrin (%CDT), γ-glutamyltransferase (γ-GT) and mean corpuscular erythrocyte volume (MCV) as biomarkers for chronic alcohol abuse: a study in patients with alcohol dependence and liver disorders of non-alcoholic and alcoholic origin. Addiction. 2005;100(10):1477–1486.