Resting fMRI-guided TMS evokes subgenual anterior cingulate response in depression

Abstract

Background: Depression alleviation following treatment with repetitive transcranial magnetic stimulation (rTMS) tends to be more effective when TMS is targeted to cortical areas with high (negative) resting state functional connectivity (rsFC) with the subgenual anterior cingulate cortex (sgACC). However, the relationship between sgACC-cortex rsFC and the TMS-evoked response in the sgACC is still being explored and has not yet been established in depressed patients.

Objectives: In this study, we investigated the relationship between sgACC-cortical (site of stimulation) rsFC and induced evoked responses in the sgACC in healthy controls and depressed patients.

Methods: For each participant (N = 115, 34 depressed patients), a peak rsFC cortical 'hotspot' for the sgACC and control targets were identified at baseline. Single pulses of TMS interleaved with fMRI readouts were administered to these targets to evoke downstream fMRI blood-oxygen-level-dependent (BOLD) responses in the sgACC. Generalized estimating equations were used to investigate the association between TMS-evoked BOLD responses in the sgACC and rsFC between the stimulation site and the sgACC.

Results: Stimulations over cortical sites with high rsFC to the sgACC were effective in modulating activity in the sgACC in both healthy controls and depressed patients. Moreover, we found that in depressed patients, sgACC rsFC at the site of stimulation was associated with the induced evoked response amplitude in the sgACC: stronger positive rsFC values leading to stronger evoked responses in the sgACC.

Conclusions: rsFC-based targeting is a viable strategy to causally modulate the sgACC. Assuming an anti-depressive mechanism working through modulation of the sgACC, the field's exclusive focus on sites anticorrelated with the sgACC for treating depression should be broadened to explore positively-connected sites.

Keywords: Depression; Functional connectivity; TMS; fMRI; sgACC.

Fig. 1.

Experimental protocol. Representation of the primary stages of the experimental protocol. In the top right corner, sites of stimulation across participants are overlaid on the MNI brain template. Red spheres represent Individualized sgACC targets (rsFC hotspot to the sgACC) and blue spheres represent individualized control targets (anatomical landmark or rsFC hotspot to a different seed). The sagittal section drawing has been modified from Patrick J. Lynch “Brain human sagittal section”. (Patrick J. Lynch, medical illustrator; C. Carl Jaffe, MD, cardiologist. http://patricklynch.net/ https://creativecommons.org/licenses/by/2.5/)

Fig. 2.

Average evoked response after stimulation over the individualized sgACC targets. Sagittal view of the average signal change map following stimulation of the sgACC rsFC-based targets in healthy controls and depressed patients overlaid on the MNI brain template. The sgACC region of interest is shown in red.

Fig. 3.

Evoked response as a function of function of connectivity. Marginal distributions and relationship between rsFC to the sgACC at the site of stimulation and ER in the sgACC for healthy controls (N = 81) and MDD patients (N = 34). The red and blue lines represent the within-group relationships estimated from the unadjusted GEE model with rsFC by group (MDD and HC) interaction.

Fig. 4.

Relationship between behavioral measures and evoked response. Relationship between sgACC ER and (A) baseline Positive affect, (B) Negative affect, and (C) DASS21 depression subscores in MDD patients. The black lines represent the slope of a linear regression fit to the data.