Toward Precision Medicine for Smoking Cessation: Developing a Neuroimaging-Based Classification Algorithm to Identify...

Area of Science:

• Neuroscience
• Addiction Research
• Psychophysiology

Background:

• Neuroimaging advances understanding of addiction's neurobiological mechanisms.
• Identifying distinct brain reactivity profiles in smokers can inform personalized treatment strategies.
• Smokers with greater electrophysiological responses to cigarette cues (C > P) are more prone to relapse than those with greater responses to pleasant stimuli (P > C).

Purpose of the Study:

• Develop a classification algorithm to differentiate smokers based on P > C or C > P neuroaffective profiles.
• Validate the algorithm's predictive accuracy for smoking abstinence in an independent cohort.

Main Methods:

• Discriminant function analysis applied to event-related potentials from 180 smokers viewing emotional images.
• Algorithm classification outcomes assessed against smoking abstinence in a separate cohort of 177 smokers.

Main Results:

• The algorithm classified 111 smokers as P > C and 66 as C > P in the validation dataset.
• Smokers classified as P > C showed a higher abstinence rate (11%) compared to C > P (4.5%), nearly 2.5 times greater.
• Overall 12-month abstinence was low (8.5%), and the difference between groups was not statistically significant, requiring further validation.

Conclusions:

• Psychophysiological techniques show potential for advancing nicotine addiction research and clinical applications.
• A neuroimaging-based classification algorithm may aid in developing precision medicine for substance use disorders.
• Larger sample sizes are essential to reliably confirm the predictive ability of the algorithm in smoking cessation.