Decision Support Tool Personalizes Antidepressant Treatment in MDD

Key Takeaways

• PETRUSHKA was associated with lower all-cause antidepressant discontinuation at 8 weeks than usual care.
• Fewer discontinuations due to adverse events were reported, and depressive and anxiety scores were lower at 24 weeks.
• The authors said that the lack of a double-blind design and substantial missing data limited the validity of the findings.

In adults with major depressive disorder, 17% of participants assigned to PETRUSHKA discontinued antidepressant treatment by 8 weeks, compared with 27% in usual care, in a randomized clinical trial published in JAMA. The web-based decision-support system personalized antidepressant selection, and symptom scores also favored PETRUSHKA by 24 weeks, although substantial missing data limited confidence in those later findings.

PETRUSHKA was evaluated in a multicenter international randomized clinical trial at 47 sites in Brazil, Canada, and the UK, published online March 4, 2026. Among adults aged 18 to 74 years with major depressive disorder, 540 participants were randomized 1:1, with 271 assigned to PETRUSHKA and 269 to usual care. Of 520 eligible participants, 493 entered the primary analysis; median age was 35 years, 58% were female, and mean baseline PHQ-9 and GAD-7 scores were 16.6 and 11.5. PETRUSHKA combined clinical and demographic predictors with patient preferences, and the primary endpoint was discontinuation due to any cause at 8 weeks.

For the primary endpoint, 17% (41/241) in the PETRUSHKA group and 27% (69/252) in usual care discontinued treatment for any cause at 8 weeks. The adjusted relative risk was 0.62, with a 95% CI of 0.44-0.88 and P=.007. Discontinuation due to adverse events occurred in 9% (22/241) and 16% (39/252), respectively, with an adjusted relative risk of 0.59, 95% CI 0.36-0.97, and P=.04. More participants in the PETRUSHKA group remained on treatment through 8 weeks.

At 24 weeks, mean PHQ-9 scores were 7.1 (SD 5.4) with PETRUSHKA and 9.2 (SD 6.5) with usual care (n=129 in each group). The adjusted between-group mean difference was -1.92, with a 95% CI of -3.06 to -0.78 and P<.001. Mean GAD-7 scores were 4.6 (SD 4.1) and 5.8 (SD 4.9) (n=133 and n=126, respectively), with an adjusted between-group mean difference of -1.39, 95% CI -2.26 to -0.52, and P=.002. The first participant was screened on November 29, 2022, and the last follow-up occurred on January 15, 2025.

The authors said the lack of a double-blind design and a large amount of missing data limited validity. Those limitations affected interpretation of both treatment persistence and symptom findings during follow-up. Overall, PETRUSHKA was associated with more participants remaining on antidepressant treatment at 8 weeks and with lower depressive and anxiety scores at 24 weeks.