N W Bailey1, K E Hoy2, N C Rogasch3, R H Thomson2, S McQueen2, D Elliot2, C M Sullivan2, B D Fulcher3, Z J Daskalakis4, P B Fitzgerald5. 1. Monash Alfred Psychiatry Research Centre, Monash University Central Clinical School, Commercial Rd, Melbourne, Victoria, Australia. Electronic address: neil.bailey@monash.edu. 2. Monash Alfred Psychiatry Research Centre, Monash University Central Clinical School, Commercial Rd, Melbourne, Victoria, Australia. 3. Brain and Mental Health Laboratory, Monash Institute of Cognitive and Clinical Neurosciences, Monash University, Clayton, 3168 VIC, Australia. 4. Temerty Centre for Therapeutic Brain Intervention, Centre for Addiction and Mental Health, Department of Psychiatry, University of Toronto, Toronto, ON, Canada. 5. Monash Alfred Psychiatry Research Centre, Monash University Central Clinical School, Commercial Rd, Melbourne, Victoria, Australia; Epworth Healthcare, The Epworth Clinic, Camberwell, 3004, Victoria, Australia.
Abstract
BACKGROUND: Repetitive transcranial magnetic stimulation (rTMS) is an effective treatment for depression, but only some individuals respond. Predicting response could reduce patient and clinical burden. Neural activity related to working memory (WM) has been related to mood improvements, so may represent a biomarker for response prediction. PRIMARY HYPOTHESES: We expected higher theta and alpha activity in responders compared to non-responders to rTMS. METHODS: Fifty patients with treatment resistant depression and twenty controls performed a WM task while electroencephalography (EEG) was recorded. Patients underwent 5-8 weeks of rTMS treatment, repeating the EEG at week 1 (W1). Of the 39 participants with valid WM-related EEG data from baseline and W1, 10 were responders. Comparisons between responders and non-responders were made at baseline and W1 for measures of theta (4-8 Hz), upper alpha (10-12.5 Hz), and gamma (30-45 Hz) power, connectivity, and theta-gamma coupling. The control group's measures were compared to the depression group's baseline measures separately. RESULTS: Responders showed higher levels of WM-related fronto-midline theta power and theta connectivity compared to non-responders at baseline and W1. Responder's fronto-midline theta power and connectivity was similar to controls. Responders also showed an increase in gamma connectivity from baseline to W1, with a concurrent improvement in mood and WM reaction times. An unbiased combination of all measures provided mean sensitivity of 0.90 at predicting responders and specificity of 0.92 in a predictive machine learning algorithm. CONCLUSION: Baseline and W1 fronto-midline theta power and theta connectivity show good potential for predicting response to rTMS treatment for depression.
BACKGROUND: Repetitive transcranial magnetic stimulation (rTMS) is an effective treatment for depression, but only some individuals respond. Predicting response could reduce patient and clinical burden. Neural activity related to working memory (WM) has been related to mood improvements, so may represent a biomarker for response prediction. PRIMARY HYPOTHESES: We expected higher theta and alpha activity in responders compared to non-responders to rTMS. METHODS: Fifty patients with treatment resistant depression and twenty controls performed a WM task while electroencephalography (EEG) was recorded. Patients underwent 5-8 weeks of rTMS treatment, repeating the EEG at week 1 (W1). Of the 39 participants with valid WM-related EEG data from baseline and W1, 10 were responders. Comparisons between responders and non-responders were made at baseline and W1 for measures of theta (4-8 Hz), upper alpha (10-12.5 Hz), and gamma (30-45 Hz) power, connectivity, and theta-gamma coupling. The control group's measures were compared to the depression group's baseline measures separately. RESULTS: Responders showed higher levels of WM-related fronto-midline theta power and theta connectivity compared to non-responders at baseline and W1. Responder's fronto-midline theta power and connectivity was similar to controls. Responders also showed an increase in gamma connectivity from baseline to W1, with a concurrent improvement in mood and WM reaction times. An unbiased combination of all measures provided mean sensitivity of 0.90 at predicting responders and specificity of 0.92 in a predictive machine learning algorithm. CONCLUSION: Baseline and W1 fronto-midline theta power and theta connectivity show good potential for predicting response to rTMS treatment for depression.
Authors: Amin Zandvakili; Noah S Philip; Stephanie R Jones; Audrey R Tyrka; Benjamin D Greenberg; Linda L Carpenter Journal: J Affect Disord Date: 2019-03-30 Impact factor: 4.839
Authors: Robin F H Cash; Luca Cocchi; Rodney Anderson; Anton Rogachov; Aaron Kucyi; Alexander J Barnett; Andrew Zalesky; Paul B Fitzgerald Journal: Hum Brain Mapp Date: 2019-07-22 Impact factor: 5.038
Authors: Juliana Corlier; Andrew Wilson; Aimee M Hunter; Nikita Vince-Cruz; David Krantz; Jennifer Levitt; Michael J Minzenberg; Nathaniel Ginder; Ian A Cook; Andrew F Leuchter Journal: Cereb Cortex Date: 2019-12-17 Impact factor: 5.357