Affect and self-control outcomes following DLPFC and medial orbitofrontal rTMS in tobacco use disorder: Secondary analyses from a randomized sham-controlled trial

Dear Editor,

Repetitive transcranial magnetic stimulation (rTMS) has emerged as a promising neuromodulatory treatment for tobacco use disorder (TUD). Several studies have demonstrated that excitatory stimulation of the dorsolateral prefrontal cortex (DLPFC) can reduce cigarette craving and consumption by strengthening executive control networks involved in decision-making and behavioral regulation [1–3]. In contrast, addiction neuroscience models emphasize the role of reward valuation circuitry, particularly the orbitofrontal cortex (OFC), which contributes to cue-induced craving and reinforcement learning [4]. These models suggest two potential neuromodulation strategies for treating addiction: enhancing executive control networks or suppressing reward-related circuitry.

In a recent randomized clinical trial from our group, excitatory 10 Hz rTMS targeting the left DLPFC produced significantly greater reductions in cigarettes per day compared with either inhibitory medial orbitofrontal cortex (mOFC) stimulation or sham treatment, and these improvements were associated with increased prefrontal cortical activity [5]. That primary report focused on smoking outcomes and neuroimaging results. The present letter reports secondary analyses from the same trial examining psychological processes that may contribute to treatment response, specifically affect regulation and self-control, which were not included in the primary publication.

Negative affect and impaired self-regulation are core features of addiction and major drivers of relapse in tobacco use disorder [6,7]. Smokers frequently report worsening mood during quit attempts, and deficits in behavioral control can undermine sustained abstinence. Understanding how neuromodulation influences these processes may therefore inform the development of more effective circuit-based interventions for addiction.

We conducted a randomized, sham-controlled study examining rTMS targeting two addiction-relevant neural circuits. Treatment-seeking smokers were randomized to one of three conditions: (1) sham stimulation, (2) excitatory 10 Hz rTMS targeting the left DLPFC, or (3) inhibitory 1 Hz rTMS targeting the left medial orbitofrontal cortex. Stimulation was delivered daily for 15 sessions over three weeks using individualized imaging-guided targeting and electric-field–based dosing procedures.

Participants abstained from smoking for 2 h before each treatment session. Smoking cue videos were presented during stimulation to engage addiction-related neural circuits. The DLPFC protocol delivered 3000 pulses per session at 10 Hz, whereas the mOFC protocol delivered 900 pulses per session at 1 Hz. Sham stimulation followed identical procedures to those for active magnetic stimulation except with a specific sham coil.

Participants completed repeated behavioral assessments throughout the treatment period. In the present analyses, we focused on negative affect and self-control explicitly. Negative affect was assessed using the Positive and Negative Affect Schedule – Short Form (PANAS-SF) [8], and self-control was measured using the Brief Self-Control Scale (BSCS) [9]. Assessments were conducted at baseline and weekly during the three-week rTMS course, with an additional follow-up assessment 4 weeks after the last rTMS.

A total of 46 participants with tobacco use disorder were enrolled in the trial, and 35 completed the treatment protocol and were included in the analyses (sham = 9; DLPFC = 12; mOFC = 14). Baseline demographic and smoking characteristics were comparable across groups, including age, years of smoking, cigarettes per day, and nicotine dependence severity.

Stimulation target significantly influenced affective outcomes (Fig. 1A). Participants receiving inhibitory rTMS targeting the medial orbitofrontal cortex demonstrated significantly greater increases in negative affect (e.g. worsening) than the DLPFC and the sham groups. Mixed-effects model showed a significant group effect (F_(2,124) = 5.81, p = 0.005), post-hoc comparisons: mOFC > DLPFC and sham). Thus, mOFC inhibitory TMS made subjects worse in dealing with negative affect.

Fig. 1.

Changes in negative affect and self-control across the three-week rTMS treatment course by treatment group. (A) Mean change in negative affect (PANAS-Negative) from baseline are shown for the sham, dorsolateral prefrontal cortex (DLPFC), and medial orbitofrontal cortex (mOFC) stimulation groups across assessment time points. Error bars represent standard errors of the mean. In panel A, the mOFC group exhibited significantly greater increases in negative affect at week 3 compared to the other groups (**p < 0.01). A mixed model analysis reveals a significant treatment group effect: F_(2,124) = 5.81, p = 0.005; post hoc: mOFC > DLPFC (p < 0.01) or sham (p < 0.05). (B) mean change in self-control (BSCS) from baseline i are shown for the sham, dorsolateral prefrontal cortex (DLPFC), and medial orbitofrontal cortex (mOFC) stimulation groups across assessment time points. Error bars represent standard errors of the mean. A mixed model analysis reveals a significant treatment group effect: F _{(2, 127)} = 3.36, p = 0.036; post hoc: mOFC > DLPFC (p = 0.014); the Sham group does not differ from either mOFC or DLPFC.

A similar pattern was observed for self-control outcomes. Participants receiving mOFC stimulation showed significantly reduced self-control compared with those receiving excitatory DLPFC stimulation (p = 0.014) (Fig. 1B). However, mOFC self-control did not differ significantly from sham.

Thus, neuromodulation of distinct nodes within addiction-related networks can produce markedly different psychological outcomes. Inhibitory stimulation of the medial orbitofrontal cortex, a region involved in reward valuation and affective processing, increased negative affect and reduced behavioral self-regulation. Although suppressing reward circuits may theoretically reduce craving, these results suggest that such inhibition may simultaneously exacerbate emotional dysregulation and weaken behavioral control.

Conversely, excitatory stimulation of the DLPFC produced more favorable trajectories for self-control. The DLPFC is a central hub within executive control networks responsible for cognitive regulation, inhibitory control, and goal-directed behavior. Strengthening this network may enhance top-down modulation of reward circuitry and support improved emotional regulation.

These findings complement and offer and an explanation of the smoking outcome results reported previously from the same randomized trial [5], which demonstrated that excitatory 10 Hz rTMS targeting the left DLPFC produced greater reductions in cigarettes per day than either inhibitory mOFC stimulation or sham treatment. Taken together, the current results suggest that rTMS may facilitate smoking cessation by strengthening executive control networks rather than suppressing reward circuitry alone. Enhancing prefrontal regulatory systems may shift the balance between executive and reward networks toward improved behavioral control and emotional regulation.

Several limitations should be acknowledged. First, given the modest sample size typical of mechanistic neuromodulation trials, these findings should be considered exploratory. Especially, the BSCS results did not show any difference compared to the sham treatment. Second, longer follow-up periods will be necessary to determine whether these psychological changes predict sustained abstinence.

Despite these limitations, the present findings suggest that enhancing executive control circuitry may produce more favorable emotional and behavioral regulation than suppressing reward-related orbitofrontal circuits during neuromodulation treatments for tobacco use disorder. rTMS targeting distinct nodes within addiction networks may produce substantially different effects on emotional and self-regulatory processes that influence relapse risk. Future studies integrating neuroimaging biomarkers and network-level analyses may further improve the precision and clinical effectiveness of neuromodulatory interventions for addiction.