Objective: Scientists have concluded that genetic profiles cannot predict a large percentage of variation in response to citalopram, a common antidepressant. Using the same data, we examined if a different conclusion can be arrived at when the results are personalized to fit specific patients.
Methods: We used data available through the Sequenced Treatment Alternatives to Relieve Depression database. We created three boosted Classification and Regression Trees to identify 16 subgroups of patients, among whom anticipation of positive or negative response to citalopram was significantly different from 0.5 (P≤0.1).
Results: In a 10-fold cross-validation, this ensemble of trees made no predictions in 33% of cases. In the remaining 67% of cases, it accurately classified response to citalopram in 78% of cases.
Conclusion: For the majority of the patients, genetic markers can be used to guide selection of citalopram. The rules identified in this study can help personalize prescription of antidepressants.
aDepartment of Health Systems Administration
bDepartment of Psychiatry, Georgetown University
cMental Health Service Line, Washington VA Medical Center, Northwest, Washington, District of Columbia
dDepartment of Psychiatry, Virginia Commonwealth University School of Medicine, Richmond, Virginia
eDepartment of Psychiatry, Keck School of Medicine of the University of Southern California, Los Angeles, California, USA
fESSEC Business School, Avenue Bernard Hirsch, Cergy-Pontoise Cedex, France
Correspondence to Farrokh Alemi, PhD, Department of Health Systems Administration, Georgetown University, 3700 Reservoir Road, Washington, DC 20007, USA Tel: +1 703 283 3100; fax: +1 202 784 3127; e-mail: email@example.com
Received September 22, 2010
Accepted March 6, 2011