Subtypes of cocaine use disorder and their neurobehavioral profiles.
Leyla R Brucar, Gunner Drossel, Eduardo A Garza-Villarreal, Anna Zilverstand
Abstract
Open AccessCocaine Use Disorder (CUD) afflicts over 1.4 million individuals in the United States alone. However, current treatments are largely ineffective, with up to 85% returning to use long-term. We posit that addressing functional heterogeneity underlying CUD is crucial for improving treatment success. Based on prior work, we hypothesized heterogeneity in function across three domains: approach-related behavior, executive function, and negative emotionality. We analyzed data from 109 participants (ages 22-39, 15% female, N = 61 CUD) from the SUDMEX CONN dataset. To test if distinct 'mechanism-based' subtypes exist within CUD, we conducted a Latent Profile Analysis using phenotypic data as input. The discovered subtypes were characterized regarding their use characteristics, psychiatric diagnostics, and resting-state functional connectivity (rsFC). Three distinct subtypes emerged (p < 0.05, partial eta squared: 0.19-0.51): a 'Relief Type' with high negative emotionality (N = 22); a 'Cognitive Type' with lower executive function (N = 15); and an 'Undefined Type' with no apparent impairments (N = 24). Consistent with this profile, the Relief Type had more comorbid psychiatric diagnoses. Furthermore, each subtype demonstrated a unique neurobiological profile underlying their current cocaine use (pFDR < 0.05): the Relief Type showed aberrant rsFC in Limbic/Memory and Salience networks; the Cognitive Type in Frontoparietal, higher visual, Motor Planning, Salience, and Parietal Association networks; and the 'Undefined Type', in Motor Planning, Ventral Frontoparietal, Salience, and Default-Mode networks. Importantly, despite these distinct profiles, CUD severity was equivalent between all three subtypes. Results highlight significant heterogeneity in the mechanisms underlying CUD, emphasizing the importance of mechanism-based subtyping to inform the development of targeted treatments.