AI Can Predict Response to Early Sertraline for Chronic Major Depression

Findings based on neuroimaging and clinical data can speed knowledge of efficacy to one week
AI Can Predict Response to Early Sertraline for Chronic Major Depression
Adobe Stock
Medically Reviewed By:
Meeta Shah, M.D.

WEDNESDAY, Feb. 14, 2024 (HealthDay News) -- Early sertraline treatment response can be predicted using neuroimaging and clinical data in outpatients with chronic major depressive disorder, according to a study published online Feb. 7 in The American Journal of Psychiatry.

Maarten G. Poirot, from University of Amsterdam, and colleagues assessed whether a multimodal machine learning approach could predict early sertraline response in patients with major depressive disorder. The analysis included 229 adult outpatients with unmedicated recurrent or chronic major depressive disorder who underwent magnetic resonance neuroimaging and had clinical data collected before and one week after treatment.

The researchers found that internal cross-validation performance was significantly better than chance in predicting response to sertraline (balanced accuracy [bAcc], 68 percent; area under the receiver operating characteristic curve [AUROC], 0.73). Using external cross-validation on data from placebo nonresponders (bAcc, 62 percent; AUROC, 0.66) and placebo nonresponders who were switched to sertraline (bAcc, 65 percent; AUROC, 0.68) resulted in differences that suggest specificity for sertraline treatment versus placebo treatment.

"This is important news for patients. Normally, it takes six to eight weeks before it is known whether an antidepressant will work," coauthor Liesbeth Reneman, M.D., Ph.D., also from University of Amsterdam, said in a statement. "With this method, we can already prevent two-thirds of the number of 'erroneous' prescriptions of sertraline and thus offer better quality of care for the patient. Because the drug also has side effects."

Several authors disclosed ties to the pharmaceutical or medical technology industries.

Abstract/Full Text (subscription or payment may be required)

Related Stories

No stories found.