Use of Clinical Global Impressions-Severity (CGI-S) to Assess Response to Antidepressant Treatment in Patients with Treatment-Resistant Depression
Joachim Morrens, Maju Mathews, Vanina Popova, Stephane Borentain, Benoît Rive, Beatriz Gonzalez Martin Moro, Carol Jamieson, Qiaoyi Zhang
Abstract
Background: This post-hoc analysis evaluated the agreement between Clinical Global Impressions-Severity (CGI-S) score- and Montgomery-Åsberg Depression Rating Scale (MADRS) total score-based assessment of response in patients with treatment-resistant depression (TRD) treated with esketamine nasal spray plus a newly initiated oral antidepressant (ESK-NS + AD). Methods: Data were analyzed from a phase 3, randomized, double-blind study (TRANSFORM-2) of flexibly dosed esketamine or placebo nasal spray plus a newly initiated oral-AD in adults with moderate-to-severe TRD. Patients with ≥50% reduction in MADRS from baseline at the end of the 4-week acute treatment phase were defined as responders. For the CGI-S-based assessment of response, patients with ≥2 points decrease from baseline or a CGI-S score of ≤3 (mildly depressed to normal) were considered responders. Cohen's kappa coefficient was calculated to assess level of agreement between MADRS and CGI-S-based assessments. Results: At the end of 4-week treatment, the proportion of responders among all study patients (n=201) was similar when assessed using the MADRS (61%) and CGI-S (62%) methods, with substantial agreement (Cohen's kappa=0.76; sensitivity=92%; specificity=84%) between both methods. When restricting analysis to ESK-NS + AD-treated patients (n=101) who had a higher response rate (on MADRS: 69%; on CGI-S: 68%), the agreement remained substantial (Cohen's kappa=0.75; sensitivity=91%; specificity=84%). Conclusion: The CGI-S may be a practical and reliable alternative to the MADRS to assess response to ESK-NS + AD in patients with TRD and can be used in real-world practice to support informed treatment decisions.