Normal component of TMS-induced electric field is correlated with depressive symptom relief in treatment-resistant depression

Abstract

Repetitive transcranial magnetic stimulation (rTMS) is a non-invasive brain stimulation technique that modulates brain activity by producing an electric field in the targeted brain region [[1]Klomjai W. Katz R. Lackmy-Vallée A. Basic principles of transcranial magnetic stimulation (TMS) and repetitive TMS (rTMS).Ann Phys Rehabil Med. 2015; 58: 208-213https://doi.org/10.1016/j.rehab.2015.05.005Google Scholar]. Theta burst stimulation (TBS), a patterned form of rTMS, is an FDA approved therapy for treatment resistant depression (TRD). However, reported effect sizes and response rates are heterogeneous [[2]Blumberger D.M. et al.Effectiveness of theta burst versus high-frequency repetitive transcranial magnetic stimulation in patients with depression (THREE-D): a randomised non-inferiority trial.Lancet. 2018; 391: 1683-1692https://doi.org/10.1016/S0140-6736(18)30295-2Google Scholar]. Recent research suggests that the diversity of the induced electric fields (EF) in the brain due to individual differences in head morphology may partly explain the response variation to TMS [[3]Aberra A.S. et al.Simulation of transcranial magnetic stimulation in head model with morphologically-realistic cortical neurons.Brain Stimul. 2020; 13: 175-189https://doi.org/10.1016/j.brs.2019.10.002Google Scholar]. Yet, predicting the clinical response based on EF calculations continues to be a challenge, as we still lack a clear understanding of the key EF component responsible for the observed treatment response. To inform this research gap, Fox et al. used positron emission tomography (PET) to examine TMS-induced cortical activation and found that “TMS active regions” had a higher EF normal component (E⊥) in the inward direction (i.e., entering the cortex perpendicular to its surface) [[4]Fox P.T. et al.Column-based model of electric field excitation of cerebral cortex.Hum Brain Mapp. 2004; 22: 1-14https://doi.org/10.1002/hbm.20006Google Scholar]. Conversely, recent computational simulation modeling research reported that the tangential component (E||) and the overall magnitude of the EF (||E||) were most relevant to TMS-induced effects [[5]Weise K. et al.A novel approach to localize cortical TMS effects.Neuroimage. 2020; 209116486https://doi.org/10.1016/j.neuroimage.2019.116486Google Scholar]. The current study aimed to systematically investigate the relationship between TMS-induced EF characteristics and antidepressant effects in patients with TRD. We hypothesized that the reduction of depressive symptoms after three weeks of stimulation treatment would be proportional to the magnitude of one or more of the proposed EF components in the stimulated brain region. The analysis was based on a subsample of patients from a longitudinal, clinical trial published previously [[6]Stöhrmann P. et al.Effects of bilateral sequential theta-burst stimulation on functional connectivity in treatment-resistant depression: first results.medRxiv. 2022; (Advance online publication p. 2022.02.16.22271078)https://doi.org/10.1101/2022.02.16.22271078Google Scholar]. Data of twelve TRD patients were included. Patients underwent 3 weeks of daily sequential bilateral TBS. Standard intermittent TBS (iTBS; 20 trains, 600 pulses) was utilized to stimulate the left dorsolateral prefrontal cortex (DLPFC) at Montreal Neurological Institute (MNI) coordinates x = −38, y = 44, z = 26, whereas continuous TBS (cTBS; 600 continuous pulses) was used to stimulate the right DLPFC at MNI coordinates x = 38, y = 44, z = 26 in individual subject space, and at an intensity of 120% resting motor threshold (RMT). Targets were based on a previous study that indicated largest antidepressant effects and strongest negative functional connectivity with the subgenual cingulate cortex compared with other DLPFC locations [[7]Fox M.D. et al.Efficacy of transcranial magnetic stimulation targets for depression is related to intrinsic functional connectivity with the subgenual cingulate.Biol Psychiatr. 2012; 72: 595-603https://doi.org/10.1016/j.biopsych.2012.04.028Google Scholar]. MRI scans and clinical assessments were performed at baseline and after the last TBS session (for details, see Ref. [[6]Stöhrmann P. et al.Effects of bilateral sequential theta-burst stimulation on functional connectivity in treatment-resistant depression: first results.medRxiv. 2022; (Advance online publication p. 2022.02.16.22271078)https://doi.org/10.1101/2022.02.16.22271078Google Scholar]). TBS induced EF were simulated in anatomically realistic, volume conductor head models by the finite element method (FEM) using SimNIBS v3.2 [[8]Saturnino G.B. Madsen K.H. Thielscher A. Electric field simulations for transcranial brain stimulation using FEM: an efficient implementation and error analysis.J Neural Eng. 2019; 16066032https://doi.org/10.1088/1741-2552/ab41baGoogle Scholar]. Individual T1-and T2-weighted structural MRI data were used for head model reconstruction. Segmentation was inspected to prevent tissue boundary establishment errors. A coil model of MagVenture MC-B70 was used for computations, which displays similar penetration depth and focality to the coil used for stimulation (MagVenture Cool-B70) [[9]Drakaki M. et al.Database of 25 validated coil models for electric field simulations for TMS.Brain Stimul. 2022; 15: 697-706https://doi.org/10.1016/j.brs.2022.04.017Google Scholar]. The TMS coil was positioned over the patient's head according to the location extracted from the neuro-navigation system (LOCALITE® TMS Navigator Germany) with the handle pointing towards the vertex. The stimulation intensity was determined using the relative dI/dt (current rate-of-change) value obtained directly from the TMS stimulator (MagPro X100, Magventure, Tonica Elektronik A/S, Denmark). Left and right stimulation regions in the DLPFC at MNI coordinates (±38, 44, 26) were chosen as regions of interest (ROIs) and defined as spheres (r = 5mm) in head models (Fig. 1a). EF characteristics were analyzed on the ROI level. ||E||, E|| and E⊥ were calculated as the sum of the respective average values within the left and right ROI to simulate the real stimulation “dose” that patients received. The magnitude of E⊥ component (|E⊥|) was defined as the sum of the average absolute value of both left and right E⊥ inside an ROI. In an exploratory post-hoc analysis, EF characteristics were also investigated separately for the left and right ROI. Generalized liner mixed model (GLMM, repeated measure: time; main effect: time, ||E||, E||, E⊥ and |E⊥|; interaction effect: time by each characteristics of EF; dependent variable: IDS-C score) and post-hoc pairwise comparisons were used to analyze the relationship between EF characteristics and clinical improvement. Following the GLMM, we performed Pearson's correlations to confirm associations between antidepressant treatment effects and EF characteristics. Post-hoc analyses were corrected for multiple comparisons using the Bonferroni procedure. The level of statistical significance was set at P < 0.05. GLMM indicated a significant interaction effect between time and |E⊥|sum on IDS-C scores (F = 8.911, P = 0.01), but not for other EF characteristics. Post-hoc pairwise comparison showed significantly lower estimated IDS-C scores after treatment compared to baseline scores (mean ± SD = −15.5 ± 2.327; t14 = −6.662, P < 0.001, corrected) (see supplementary file for further details). Pearson correlation analyses confirmed a negative correlation between change in IDS-C scores after treatment and the |E⊥|sum component (r = −0.752, P = 0.005, corrected P = 0.0125, Fig. 1b). No other EF characteristics showed significant associations with IDS-C scores (Fig. 1c–e). When investigating each ROI separately, we observed a negative correlation between |E⊥|right and change of IDS-C scores (r = −0.781, P = 0.003, Fig. S1a) but not for the left ROI (r = −0.458, P = 0.135 Fig. S1b). To our knowledge, this is the first study using a computational simulation modeling approach to investigate the association between TMS-induced EF characteristics and clinical improvement in TRD patients. Our findings indicate that the |E⊥| component rather than ||E|| or E|| in the DLPFC is related to symptom improvement. This is in line with a previous study indicating a correlation between the E⊥component and TMS induced effects [[10]Opitz A. et al.Physiological observations validate finite element models for estimating subject-specific electric field distributions induced by transcranial magnetic stimulation of the human motor cortex.Neuroimage. 2013; 81: 253-264https://doi.org/10.1016/j.neuroimage.2013.04.067Google Scholar]. However, the exact mechanism of how TMS stimulates the brain is still a matter of debate. On the one hand, research suggests that not only E⊥ but also the ||E|| and E|| substantially contribute to the neuronal depolarization [[3]Aberra A.S. et al.Simulation of transcranial magnetic stimulation in head model with morphologically-realistic cortical neurons.Brain Stimul. 2020; 13: 175-189https://doi.org/10.1016/j.brs.2019.10.002Google Scholar,[5]Weise K. et al.A novel approach to localize cortical TMS effects.Neuroimage. 2020; 209116486https://doi.org/10.1016/j.neuroimage.2019.116486Google Scholar] On the other hand, the cortical column cosine model claims that the stimulation effect is proportional to the cosine of the angle between the E_field and the normal of the cortical surface [[4]Fox P.T. et al.Column-based model of electric field excitation of cerebral cortex.Hum Brain Mapp. 2004; 22: 1-14https://doi.org/10.1002/hbm.20006Google Scholar]. The latter explanation seems to be the most parsimonious explanation for our data, with recent findings bolstering the rationale for this theoretical model. Aberra et al. [[3]Aberra A.S. et al.Simulation of transcranial magnetic stimulation in head model with morphologically-realistic cortical neurons.Brain Stimul. 2020; 13: 175-189https://doi.org/10.1016/j.brs.2019.10.002Google Scholar] reported that TMS initiates action potentials at the axon terminal, with layer 5 pyramidal cells (the main integration unit of the cortical column) demonstrating the lowest depolarization thresholds and preferentially activated by the E⊥ component. More studies are needed to resolve these controversial results, and to clarify the clinical significance of EF characteristics for left and right DLPFC in bilateral stimulation protocols. Lastly, we observed a significant correlation only for the right but not for left DLPFC when examining each hemisphere separately, an observation that requires replication. In conclusion, our results suggest that the magnitude of E⊥ component of the TMS-induced EF is associated with the antidepressant response to bilateral sequential TBS in patients with TRD. Future studies using a tailored TMS treatment approach may consider maximizing the |E⊥| component in order to enhance clinical effects. However, our results are exploratory and must be interpreted with caution due to the limited sample size. Bella B.B. Zhang: Methodology, Software, Data Curation, Formal analysis, Writing-Original Draft, Writing-Review & Editing, Visualization; Peter Stöhrmann: Methodology, Software, Data Curation, Formal analysis, Writing-Review & Editing; Godber M Godbersen: Data Curation, Stimulation Treatment, Writing-Review & Editing; Jakob Unterholzner: Data Curation, Stimulation Treatment, Writing-Review & Editing, Siegfried Kasper: Resources, Methodology, Writing-Review & Editing; Rupert Lanzenberger: Conceptualization, Methodology, Writing-Review & Editing, Supervision, Project administration; Georg S. Kranz: Conceptualization, Methodology, Writing-Original Draft, Writing-Review & Editing, Supervision, Project administration. The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: In the past three years S. Kasper has received grant/research support from Lundbeck; he has served as a consultant or on advisory boards for Angelini, Biogen, Esai, Janssen, IQVIA, Lundbeck, Mylan, Recordati, Sage and Schwabe; and he has served on speaker bureaus for Abbott, Angelini, Aspen Farmaceutica S.A., Biogen, Janssen, Lundbeck, Recordati, Sage, Sanofi, Schwabe, Servier, Sun Pharma and Vifor. Without any relevance to this work, R. Lanzenberger declares that he received travel grants and/or conference speaker honoraria within the last three years from Bruker BioSpin MR and Heel, and has served as a consultant for Ono Pharmaceutical. He received investigator-initiated research funding from Siemens Healthcare regarding clinical research using PET/MR. He is a shareholder of the start-up company BM Health GmbH since 2019. G.S. Kranz declares that he received conference speaker honorarium from Roche, AOP Orphan and Pfizer. J Unterholzner received travel grants from MEDICE Pharma GmbH & Co. KG. The other authors do not report any conflict of interest. This research was supported by the Austrian Science Fund ( FWF ) [grant number KLI 551, PI: S. Kasper] and the General Research Fund ( GRF ) under the University Grants Committee ( UGC ) of the Hong Kong Special Administrative Region [grant numbers 15100120 and 25100219, PI: G.S. Kranz]. B.B.B. Zhang is funded by the Ernst Mach Grant, a scholarship of the OeAD, Austria's Agency for Education and Internationalisation. We would like to thank Richard Frey, Gregor Gryglewski, Marius Hienert, Marie Spies, Christoph Kraus, Alexander Kautzky, Arkadiusz Komorowski, Paul Michenthaler, Pia Baldinger-Melich for clinical support, and Sebastian Ganger, Andreas Hahn and Murray Reed for technical support of the main study. We would further like to thank all participants for their time and effort in participating. The following is the Supplementary data to this article: Download .docx (.14 MB) Help with docx files Multimedia component 1

Topics & Concepts