MRI Prediction of ECT Response: New Systematic Review and Meta-Analysis
Out on PubMed, from investigators in Amsterdam, The Netherlands, is this paper:
Magnetic resonance imaging for individual prediction of treatment response in major depressive disorder: a systematic review and meta-analysis.
Transl Psychiatry. 2021 Mar 15;11(1):168. doi: 10.1038/s41398-021-01286-x.PMID: 33723229
The abstract is copied below:
No tools are currently available to predict whether a patient suffering from major depressive disorder (MDD) will respond to a certain treatment. Machine learning analysis of magnetic resonance imaging (MRI) data has shown potential in predicting response for individual patients, which may enable personalized treatment decisions and increase treatment efficacy. Here, we evaluated the accuracy of MRI-guided response prediction in MDD. We conducted a systematic review and meta-analysis of all studies using MRI to predict single-subject response to antidepressant treatment in patients with MDD. Classification performance was calculated using a bivariate model and expressed as area under the curve, sensitivity, and specificity. In addition, we analyzed differences in classification performance between different interventions and MRI modalities. Meta-analysis of 22 samples including 957 patients showed an overall area under the bivariate summary receiver operating curve of 0.84 (95% CI 0.81-0.87), sensitivity of 77% (95% CI 71-82), and specificity of 79% (95% CI 73-84). Although classification performance was higher for electroconvulsive therapy outcome prediction (n = 285, 80% sensitivity, 83% specificity) than medication outcome prediction (n = 283, 75% sensitivity, 72% specificity), there was no significant difference in classification performance between treatments or MRI modalities. Prediction of treatment response using machine learning analysis of MRI data is promising but should not yet be implemented into clinical practice. Future studies with more generalizable samples and external validation are needed to establish the potential of MRI to realize individualized patient care in MDD.
The pdf is here.
...Our results show that machine learning analysis of MRI data can predict antidepressive treatment success with an AUC of 0.84, 77% sensitivity, and 79% specificity (Fig. 2). Furthermore, we did not find a difference in classification performance between studies using pharmacotherapy and ECT. Although ECT showed somewhat higher sensitivity and specificity, CIs largely overlapped between the two intervention types.
...It would be useful to have a single predictive test for therapy resistant patients, especially to guide decision-making for invasive treatments such as ECT. For example, ECT is associated with cognitive side effects that are preferably avoided in case the treatment is unsuccessful. In addition, ECT is only applied in 1–2% of patients with persistent or severe depression and a biomarker that indicates a high probability of success may reduce the hesitance of its use.
and from the text:
This is yet another pie-in-the-sky-ish attempt to "personalize" psychiatric medicine with a biomarker. The review included a total of 957 patients, 285 of whom received ECT. Machine learning analysis of MRI data for response prediction was pretty good, but not ready for clinical use, as acknowledged by these authors.
As noted before, ECT is so effective for appropriately chosen patients, that such efforts seem like a huge effort for small gains. Saving a patient from unsuccessful ECT is a laudable goal, but denying a seriously ill patient a course of ECT based on a baseline MRI would be very unfortunate...For now, clinicians will continue to prescribe ECT based on clinical symptoms, history, and severity/urgency of illness.
This article will be good reading for dedicated ECT neuroimagers (~20minutes); for others, review of the abstract and above text excerpts should suffice.
(Please see also blog post of March 20, 2021,"Classics in ECT: "Prediction of Clinical Response to ECT," Fink, Abrams and Feldstein, 1973)
Comments
Post a Comment