Deep rTMS of the insula and prefrontal cortex in smokers with schizophrenia: Proof-of-concept study

Abstract

Patients with schizophrenia have a high prevalence of cigarette smoking and respond poorly to conventional treatments, highlighting the need for new therapies. We conducted a mechanistic, proof-of-concept study using bilateral deep repetitive transcranial magnetic stimulation (dTMS) of insular and prefrontal cortices at high frequency, using the specialized H4 coil. Feasibility of dTMS was tested for disruption of tobacco self-administration, insula target engagement, and insula circuit modulation, all of which were a priori outcomes of interest. Twenty patients completed the study, consisting of weekday dTMS sessions (randomization to active dTMS or sham; double-blind; 10 patients per group), a laboratory tobacco self-administration paradigm (pre/post assessments), and multimodal imaging (three MRI total sessions). Results showed that participants assigned to active dTMS were slower to initiate smoking their first cigarette compared with sham, consistent with smoking disruption. The imaging analyses did not reveal significant Time × Group interactions, but effects were in the anticipated directions. In arterial spin labeling analyses testing for target engagement, an overall decrease in insula blood flow, measured during a post-treatment MRI versus baseline, was numerically more pronounced in the active dTMS group than sham. In fMRI analyses, resting-state connectivity between the insula and default mode network showed a numerically greater change from baseline in the active dTMS group than sham, consistent with a functional change to insula circuits. Exploratory analyses further suggested a therapeutic effect of dTMS on symptoms of psychosis. These initial observations pave the way for future confirmatory studies of dTMS in smoking patients with schizophrenia.

Conflict of interest statement

The authors declare no competing interests.

© 2022. The Author(s).

Figures

Fig. 1. Study participants.

a Flow diagram for the trial. ASL = arterial spin labeling. b Demographics of patients who completed the study. For b, numbers are frequencies or M ± SD. dTMS = deep repetitive transcranial magnetic stimulation, PANSS = Positive and Negative Syndrome Scale. For additional description of the Fagerström scale, motivation to stop smoking, and cigarettes smoked per day (the latter assessed with a Timeline Follow-Back calendar), see the Supplement. Socioeconomic status was measured with the Hollingshead Index. No statistical tests showed significant differences between the treatment groups.

Fig. 2. Overview of study procedures.

The study included pre- and post-treatment smoking self-administration sessions (cigarette graphic), which were preceded by overnight inpatient stay to monitor smoking abstinence (bed graphic); weekday deep repetitive transcranial magnetic stimulation (dTMS) treatment for 15 weekdays over 3 weeks (Visits 2-16; TMS helmet graphic), and three functional MRI scans (MRI scanner graphic), during which ASL was acquired (once at baseline, once after the first treatment, and once after the final treatment). MRI was acquired before the smoking self-administration on Days 1 and 17, and after the first dTMS session on Day 2. Resting-state functional connectivity was also acquired during the same MRI sessions, but only during the first and last scan (Visits 1 and 17). Clinical assessment with the PANSS was acquired at four time points (clipboard graphic), approximately once per week. Participants made a quit attempt approximately by Visit 8 (red prohibition graphic). Screening/consenting could occur up to a week before primary study procedures began.

Fig. 3. Effects of dTMS on behavior.

a, b In a laboratory task of tobacco choice, participants had the chance to smoke or earn money for abstaining. a There was a significant interaction on the latency to smoke a first cigarette, where patients in the active dTMS group took longer to smoke after treatment than before it (significant effect of Visit in this group indicated by asterisk). b While there was a similar direction of effects regarding the total number of cigarettes smoked during the task, no main effects or interactions were significant (ns = not significant). In addition to these smoking variables, c dTMS improved severity of psychosis, as measured by the Positive and Negative Syndrome Scale (PANSS). Uniquely in the active dTMS group, there was a linear decrease in positive symptoms over the course of the study (significant linear contrast indicated by asterisk).

Fig. 4. Effects of dTMS on insula-centric neural functioning.

a Regions of interest (ROIs) for the insula (teal) and the default mode network (DMN) (red). b Both groups showed a decrease in insula cerebral blood flow (CBF) during Visit 2, which occurred directly after the first dTMS treatment. Exploratory analyses suggested that this decrease during Visit 2 was significant in the active dTMS group (indicated by asterisk), but not the sham group, consistent with a more robust numerical difference between baseline and first treatment in those receiving the active stimulation. c The active dTMS group showed numerically increased (less negative) insula-DMN connectivity from Visit 1 to Visit 17 (p < 0.15, indicated by †), which was not evident in the sham group. The y-axis depicts the z-transformed partial r, between the bilateral insula and the DMN (with the insula as the seed).