Psychiatry · Medical Literature
[Correspondence] Cognitive safety after magnetic seizure therapy: from group differences to patient-level risk
Abstract
Daniel L Blumberger and colleagues1 report an important trial comparing magnetic seizure therapy (MST) with right unilateral ultra-brief (RUL-UB) electroconvulsive therapy (ECT) for major depressive disorder. Their results suggest that MST preserved much of the antidepressant efficacy of RUL-UB ECT while producing fewer cognitive adverse effects.1 The lower rate of autobiographical memory worsening after MST is clinically relevant, given that memory concerns remain a major reason for patients declining ECT.
Clinical relevance
Does the favorable group-level cognitive safety profile of magnetic seizure therapy (MST) observed in the CREST-MST trial adequately characterize patient-level risk of cognitive worsening when compared with right unilateral ultra-brief pulse-width (RUL-UB) electroconvulsive therapy (ECT) in adults with major depressive disorder? __SNAPSHOT__ {"studyDesign":"Correspondence (letter to the editor) commenting on the CREST-MST trial (Blumberger et al., Lancet Psychiatry 2026;13:376–386), which was itself a randomized, double-blind, parallel-group, non-inferiority trial. This summary pertains to the correspondence piece by Stubberud (Lancet Psychiatry, June 18, 2026), not the primary trial.","setting":"Academic/journal commentary; the primary trial was conducted at three academic centers in Canada (CAMH, Toronto) and the USA (University of Texas Southwestern Medical Center, Dallas; UC San Diego). Author affiliation: Department of Psychology, University of Oslo, Norway.","sampleSize":"Not applicable for this correspondence. The primary CREST-MST trial enrolled 236 analyzed participants (nearly 300 enrolled) across the three sites.","population":"Adults aged ≥18 years with non-psychotic major depressive disorder (MDD), baseline HRSD-24 score ≥21, deemed appropriate for convulsive therapy. Correspondence raises questions about which subpopulations within this group may carry disproportionate patient-level cognitive risk.","intervention":"Magnetic seizure therapy (MST), delivered using a twin coil in a midline frontal position, up to 21 treatment sessions.","comparator":"Right unilateral ultra-brief pulse-width electroconvulsive therapy (RUL-UB ECT), up to 21 treatment sessions.","primaryOutcome":"Co-primary outcomes in the referenced trial: (1) remission of depression on the Hamilton Rating Scale for Depression-24 item (HRSD-24); (2) worsening of autobiographical memory on the Autobiographical Memory Test (AMT). The correspondence specifically interrogates whether group-level AMT differences adequately reflect individual patient-level cognitive risk.","followUp":"Not reported in this correspondence. The primary CREST-MST trial was conducted between 2018 and 2024; specific post-treatment follow-up duration for cognitive outcomes is not reported in available abstracts.","fundingSource":"Not reported for this correspondence. The primary CREST-MST trial was funded by the U.S. National Institute of Mental Health (NIMH), with in-kind equipment support from MagVenture (manufacturer of the MST device used in the trial)."}
Key findings
This is a correspondence piece responding to the CREST-MST trial. The primary trial (Blumberger et al.) demonstrated that MST was non-inferior to RUL-UB ECT for depression remission as measured by the HRSD-24, with approximately 48% of patients in both treatment arms showing meaningful clinical response. MST also showed a more favorable cognitive safety profile, with substantially fewer patients experiencing clinically meaningful worsening of autobiographical memory on the AMT. Adverse events in the MST arm were predominantly mild. Stubberud's correspondence raises a methodologically important concern: that group-level cognitive outcome differences, while statistically and clinically meaningful at the population level, may obscure heterogeneity in patient-level cognitive risk. Specifically, Stubberud argues that some individual patients receiving MST may still experience significant cognitive worsening, and that aggregate group comparisons are insufficient to identify or predict who those patients are. Precise effect sizes, confidence intervals, p-values, and NNT/NNH figures for the correspondence's specific arguments are not reported (only abstract available); these values for the primary trial are reported in Lancet Psychiatry 2026;13(5):376–386.
Practice implications
The CREST-MST trial establishes that MST is non-inferior to RUL-UB ECT for depression remission and is associated with a significantly more favorable autobiographical memory safety profile at the group level—a finding of genuine clinical importance given that memory concerns are a leading reason patients decline convulsive therapy. Stubberud's correspondence adds an important methodological caveat: favorable mean group outcomes do not preclude clinically meaningful cognitive deterioration in individual patients receiving MST. Clinicians should not interpret the MST cognitive safety advantage as an absence of individual-level risk. Patients who experienced cognitive worsening despite being in the MST group are not adequately characterized by the group comparison alone. This is particularly relevant for patients with pre-existing cognitive vulnerability (e.g., older adults, those with neurocognitive comorbidities), who were not specifically sub-analyzed in available abstracts. MST is not yet FDA-approved, and regulatory clearance, clinician training infrastructure, and implementation frameworks remain to be established before widespread clinical deployment. Practice should not change based on this correspondence alone; however, clinicians following the CREST-MST primary trial should incorporate individual cognitive risk assessment into pre-treatment counseling for patients considering MST. Only the abstract of the correspondence is available; the full analytical argument and any supporting data analysis by Stubberud cannot be fully evaluated. __LIMITATIONS__ 1. Publication type: This is a correspondence/letter, the lowest tier of evidence, without primary data, formal statistical analysis, or peer-reviewed methodology section. 2. Abstract-only access: Full text of the Stubberud correspondence was not available; clinical interpretation is limited accordingly. 3. The primary CREST-MST trial was conducted exclusively at three academic centers in North America, limiting generalizability to community and international settings. 4. In-kind equipment support from MagVenture (the MST device manufacturer) for the primary trial introduces a potential conflict of interest that may have influenced equipment performance or reporting, though government (NIMH) primary funding partially mitigates this concern. 5. Long-term cognitive outcomes beyond the acute treatment period are not reported in available abstracts, leaving durability of the cognitive safety advantage unestablished. 6. The exclusion of patients with substance dependence, major medical comorbidities, and other psychiatric diagnoses limits representativeness of real-world treatment-resistant depression populations. 7. MST currently lacks regulatory approval in most jurisdictions, making the external validity of efficacy and safety data contingent on future implementation contexts. 8. The correspondence does not provide independent quantitative analysis of patient-level cognitive trajectories; its argument is conceptual and methodological rather than empirical. __BOTTOM_LINE_CATEGORY__ potentially_practice_informing __BOTTOM_LINE__ The CREST-MST trial provides the first large-scale randomized evidence that MST is non-inferior to RUL-UB ECT for depression remission and carries a meaningfully better autobiographical memory safety profile at the group level, with approximately 48% clinical response in both arms. Stubberud's correspondence appropriately cautions that group-level cognitive advantages may mask individual patients who still experience significant cognitive worsening with MST, and clinicians should conduct individualized cognitive risk assessment rather than assume universal cognitive safety. MST is not yet regulatory-approved, and confirmation of long-term cognitive and efficacy outcomes is needed before it can be adopted as a routine clinical alternative to ECT.