ABSTRACTLong‐acting naltrexone is a first‐line pharmacotherapy for alcohol use disorder (AUD), but clinical response is heterogeneous and the underlying biology remains incompletely understood. In a randomised, double‐blind, placebo‐controlled trial of naltrexone implants (n = 70), the prespecified primary endpoint—percentage of heavy‐drinking days (PHDD) over 24 weeks—favoured naltrexone (median 1.49% vs. 13.39%; p = 0.042). In a prespecified Week‐12 mechanistic lipidomics substudy, untargeted liquid chromatography–mass spectrometry (LC–MS) profiling was performed in a responder‐enriched subset comprising naltrexone responders (RN, n = 18), placebo‐treated participants (PL, n = 10) and age‐ and BMI‐matched healthy male controls (HC, n = 10). We derived membrane‐related indices reflecting two prespecified axes of phospholipid remodelling: headgroup balance (PC/PE) and acyl‐chain composition (arachidonic acid [AA] and n‐3 polyunsaturated fatty acids within phospholipid pools). RN showed higher PC/PE and coordinated acyl‐chain shifts versus PL, with species‐level changes preferentially moving toward the healthy‐control direction. Exploratory analyses in naltrexone nonresponders (NR, n = 10) revealed partial Lands‐cycle‐related shifts but lacked the broader dual‐axis configuration observed in RN. In RN + PL (n = 28), higher n‐3 in PE and lower AA in PE at Week 12 were prospectively associated with greater subsequent heavy‐drinking burden. These findings support a dual‐axis membrane remodelling phenotype associated with naltrexone response and prospectively linked to heavy‐drinking burden in AUD, providing a biologically grounded framework for future mechanistic and biomarker studies.

Comorbid substance use disorders (SUDs) are common but difficult to study due to the complex, interacting and overlapping mechanisms through which they affect brain networks. Many datasets collected to investigate a specific SUD include participants with comorbid SUDs. Although most studies treat comorbid SUDs as covariates of no interest, these covariates also contain untapped information. This is particularly relevant as cannabis use disorder (CanUD) has become increasingly prevalent and comorbid with other SUDs that have been more thoroughly studied. While pharmacotherapies have been established for multiple SUDs, none have been approved for CanUD, although naltrexone (NTX) has been associated with reduced use. Here, we conducted a retrospective secondary analysis of functional magnetic resonance imaging (fMRI) data from individuals with primary opioid use disorder (OUD-only, N = 25 pre-NTX, N = 20 on-NTX) with comorbid CanUD (N = 10), alcohol use disorder (AUD, N = 6) or cocaine use disorder (CocUD, N = 7). All participants underwent imaging prior to receiving a therapeutic dose of long-acting intramuscular NTX (Vivitrol), an approved treatment for OUD and AUD but not for CocUD, and again 2 weeks postadministration. At baseline, OUD individuals with comorbid CanUD, AUD or CocUD exhibited distinct functional connectivity (FC) alterations compared to those with OUD-only. These differences were greater in younger participants and primarily involved the default mode network. Following NTX administration, FC differences between the comorbid CanUD and OUD-only groups globally diminished. A similar FC response to NTX was observed in the comorbid AUD group, whereas little change in FC was observed in comorbid CocUD. These findings, combined with prior evidence that NTX reduces cannabis use by dampening reward, suggest NTX may hold promise as a treatment for CanUD.

Studies have used repetitive transcranial magnetic stimulation (rTMS) to stimulate the left dorsolateral prefrontal cortex (DLPFC) in patients with tobacco use disorder (TUD); consequently, decreased craving for smoking was observed. However, the neural mechanism underlying this process remains unclear. Functional MRI data were collected from 59 valid TUD participants (31 rTMS and 28 shams) when they performed a Go/No-go task before and after a continuous 5-day treatment (rTMS on the left DLPFC). Three approaches of data analyses were performed: event-related data analyses to identify the brain regions that are associated with rTMS treatment; functional connectivity (FC) analyses among the left DLPFC (the stimulate region) and other survived brain regions after group comparison; FC between the left DLPFC and all brain regions to find couplings among the brain regions. rTMS decreased the craving for smoking in patients with TUD. Comparing with the sham group, the rTMS group showed enhanced brain responses in the bilateral ACC, bilateral caudate and left thalamus to the No-go smoking cues. Further, FC analyses among the survived brain regions showed that rTMS enhanced the FC in DLPFC-caudate and caudate-left thalamus pathways. Notably, enhanced FC between the DLPFC and bilateral basal ganglia thalamus was observed. The current study demonstrated the effectiveness of rTMS in the treatment of TUD, which is associated with enhanced brain responses that are responsible for executive control and reward processing; it enhanced top-down control by reshaping the prefrontal-striatal pathways.

Millions of people worldwide are affected by betel quid dependence (BQD), a prevalent psychoactive substance-use disorder. To evaluate cerebral glymphatic dysfunction in BQD, this work used the coupling between global blood-oxygen-level-dependent (gBOLD) and cerebrospinal fluid (CSF) signals collected via resting-state functional MRI (rs-fMRI). Sixty-six participants (29 BQD individuals and 37 healthy controls) underwent rs-fMRI scanning. Cross-correlation analysis between cortical BOLD and CSF time series was carried out to compute the gBOLD-CSF coupling intensity. Group comparisons were performed via independent sample t-tests, and correlations with clinical features were assessed using exploratory Pearson's analysis. Bonferroni correction was applied to the five correlation analyses (BQDS, duration, daily consumption, HAMA-14 and HAMD-24). The corrected significance threshold was set at p < 0.01 (0.05/5). BQD individuals exhibited significantly reduced gBOLD-CSF coupling (t = -2.42, p = 0.019), and weaker coupling correlated with longer BQD duration (r = 0.4313, uncorrected p = 0.0195). After Bonferroni correction for multiple comparisons, the correlation between duration and gBOLD-CSF coupling (original p = 0.0195) did not reach the adjusted threshold. Chronic betel quid exposure may impair glymphatic function, reflecting disrupted brain-CSF interactions. The gBOLD-CSF coupling metric could be used as a noninvasive imaging biomarker to evaluate neurotoxic effects and glymphatic dysfunction in substance-use disorders.

Relapse is a defining feature of substance use disorders, yet relapse risk varies widely across individuals. This variability has been linked to individual differences in incentive salience attribution to drug-related cues. We used fMRI to identify cortical and subcortical circuits underlying individual differences in incentive salience attribution among people with tobacco use disorder. Fifty daily cigarette users viewed pleasant, unpleasant, cigarette-related and neutral images during task-based fMRI and completed a resting-state scan. We extracted per-condition BOLD activity from a functionally defined extended visual region (Emotional > Neutral) and applied k-means clustering (k = 2) to derive neuroaffective profiles. Whole-brain voxelwise tests assessed Group × Content (cigarette vs. pleasant) effects. We compared groups' resting-state connectivity at the network level and for subcortical-cortical pairs. Individuals were clustered based on neuroaffective reactivity to drug-related and pleasant stimuli: The C > P profile showed larger BOLD responses to cigarette cues than to pleasant stimuli, whereas the P > C profile showed the opposite pattern. Whole-brain analyses showed significant Group × Content interactions that mirrored these profiles across left dorsolateral and medial prefrontal cortices and amygdala-hippocampal regions. At rest, individuals in the C > P profile showed weaker within-network connectivity of the frontoparietal control network (FPCN) and, in exploratory analyses, reduced FPCN-nucleus accumbens coupling. By replicating and extending prior psychophysiological work, this study identifies neuroaffective profiles that reflect the variation in incentive salience attribution central to neurobehavioral models of relapse. These findings provide convergent neurobiological markers that characterize distinct motivational pathways in substance use disorders.

Plasma neurofilament light chain (NFL), glial fibrillary acidic protein (GFAP), tau protein and ubiquitin carboxy-terminal hydrolase L1 (UCHL1) are candidate biomarkers of alcohol withdrawal (AW)-associated brain toxicity, as they are biomarkers of axonal, neuronal or glial injury. The aim of this study was to investigate the changes of these biomarkers during AW in patients with severe alcohol use disorder (AUD). Plasma NFL, GFAP, tau and UCHL1 levels were measured, with SIMOA, at three times: on Day 1 (T1), on Day 3 or 4 (T2) and on Day 13, 14 or 15 (T3) of AW. They were analysed with a linear mixed model adjusted for age, sex and body mass index. Changes in these levels according to AW symptom severity were evaluated. Twenty-four inpatients with severe AUD were included: 20 men (83.3%), aged 47.4 years [±11.3], with symptoms requiring a median equivalent-diazepam dose of 0.81 mg/kg at T1. A significant increase was observed for NFL level from T1 to T2 (β = 0.349, p = 0.035), but not for GFAP, tau or UCHL1 levels. In AW symptom severity analyses, a significant positive association was found with equivalent-diazepam dose required × T1-T2 time interaction factor for NFL (β = 0.161, p = 0.028) and for GFAP (β = 0.400, p = 9.9 × 10-4). This longitudinal study provided preliminary indication that brain injury could occur within the first days of AW, especially in patients with severe pharmacological dependence. Plasma NFL and GFAP are promising biomarkers of AW-related brain pathology and should be investigated as biomarkers of therapeutic response to test innovative drug strategies for preventing this toxicity. Trial Registration: Clinical Trials: NCT05216705.

Alcohol dependence (AD) is characterized by a high relapse rate. Virtual reality (VR) technology can provide immersive cue exposure therapy (VR-CET) and aversion therapy (VR-AT). This study aimed to evaluate the efficacy of VR-CET, VR-AT and their combination on craving, emotional and sleep states, attentional bias and relapse rate in patients with AD. In this single-centre randomized controlled trial, male inpatients with AD were randomly assigned to one of four groups: control, VR-CET, VR-AT or combined VR-CET + AT (target n = 25 per group; 80 completed, 20 per group). The interventions spanned 15 days with eight sessions (VR-CET + AT ~20 min/session; others ~10 min). Assessments were conducted before and after treatment using the Visual Analogue Scale (VAS), the Pennsylvania Alcohol Craving Scale (PACS), the Hamilton Depression Scale (HAMD), the Hamilton Anxiety Scale (HAMA) and the Pittsburgh Sleep Quality Index (PSQI). Eye-tracking and grasping indices in VR environments were used to assess attentional bias (e.g., alcohol-cue fixation time ratio). Relapse was evaluated by telephone at 4 and 12 weeks post-treatment. Statistical analyses used Shapiro-Wilk tests and ANOVA/Kruskal-Wallis tests with appropriate post hoc comparisons (α = 0.05). All groups showed significant pre- to post-treatment improvements in PACS, HAMD, HAMA, PSQI and VAS scores (all p < 0.001). Between-group comparisons at post-treatment revealed significant differences in alcohol-cue fixation time ratio (p < 0.05), with the VR-CET + AT group showing a lower fixation time ratio than the control group. VAS scores also differed among groups (p < 0.05), with the control group showing higher values than the VR-CET + AT group. Changes in alcohol-cue fixation time ratio from pre- to post-treatment were significantly greater in the VR-CET + AT group than in the control group. Relapse rates at 4 and 12 weeks (47/80 reached by telephone follow-up) did not significantly differ among groups (both p > 0.05). Combining VR-CET with VR-AT reduced craving (VAS) and attentional bias (alcohol-cue fixation time ratio) beyond standard care, whereas all groups improved on clinical scales. Larger and longer trials are warranted to further clarify relapse outcomes. Trial Registration: Chinese Clinical Trial Registry, ChiCTR2500110026. Registered 29 September 2025 (retrospectively registered).

Alcohol use disorder (AUD) is considered a chronic disorder with a highly variable course. Understanding this variability is crucial for identifying factors associated with persistence versus spontaneous remission. We analysed data from N = 462 individuals with AUD in an observational longitudinal cohort study to identify factors associated with spontaneous remission. All participants met Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) criteria for AUD at study entry and were reassessed after 1 year, in which they were classified as being in spontaneous remission (n = 107), falling below the ≥ 2-criteria threshold for AUD (n = 87) or continuing to meet AUD criteria (n = 296). Groups were compared on socio-demographic, clinical and substance use variables at baseline and after 1 year. Additionally, we used machine learning models to identify baseline characteristics predicting a persistent course of AUD. Between-group comparisons revealed that individuals who experienced spontaneous remission reported significantly lower AUD severity (F(2,459) = 25.17, p < 0.001), lower levels of alcohol intake (F(2,459) = 8.31, p = 0.013) and fewer drinking days (F(2,459) = 11.91, p < 0.001) at baseline and after 1 year. Machine learning analysis demonstrated moderate classification performance (AUC = 0.679), with the Alcohol Use Disorders Identification Test (AUDIT) sum score being the most informative predictor for group classification. Our findings indicate significant differences between spontaneously remitted and non-remitted individuals on key alcohol-related variables, supporting the clinical validity of remission as defined by the National Institute on Alcohol Abuse and Alcoholism (NIAAA). In addition, baseline characteristics may help identify individuals at risk of persistent AUD, enabling earlier identification of those who may benefit from specialized treatment. TRIAL REGISTRATION: DRKS number: DRKS00020580.