¹Taipei City Psychiatric Center, Taipei City Hospital, Department of Addiction Sciences, Taipei, Taiwan, ²School of Medicine, Taipei Medical University, Department of Psychiatry, Taipei, Taiwan, ³National Health Research Institutes, Center for Neuropsychiatric Research, Miaoli, Taiwan

Background: Neurofilament light chain (NfL), a marker of neuroaxonal injury, has recently attracted increasing attention in alcohol use disorder (AUD). Alcohol withdrawal syndrome (AWS) is a core feature of AUD and contributes to its progressive course, with delirium tremens (DT) representing its most severe manifestation. Aldehyde dehydrogenase 2 (ALDH2) is the primary enzyme responsible for metabolizing acetaldehyde, a toxic byproduct of alcohol metabolism. A functional ALDH2 single nucleotide polymorphism (rs671) is associated with reduced enzymatic activity and increased acetaldehyde-related neurotoxicity. The mu-Opiod Receptor (OPRM1) A118G polymorphism (rs1799971) has been linked to naltrexone treatment response; however, its influence on longer-term (12-week) naltrexone outcomes remains unclear.

Aims & Objectives: We aimed to characterize changes in NfL levels during alcohol withdrawal and to evaluate their potential utility in predicting DT. We further examined the impact of the ALDH2 rs671 polymorphism on interindividual variability in NfL levels and assessed the association between the OPRM1 rs1799971 polymorphism and response to 12 weeks of naltrexone treatment.

Method: Blood NfL was measured in patients with alcohol dependence (AD) and healthy controls, alongside genotyping for rs671 and rs1799971. NfL, craving, and psychological symptoms were assessed at baseline and after 1–2 weeks of detoxification. Patients were categorized by prospective DT occurrence (DT: n = 25; non-DT: n = 199) and followed during 12-week open-label naltrexone treatment.

Results: Baseline NfL was higher in AUD patients before withdrawal than controls (264.2 ± 261.8 vs. 72.1 ± 35.6 pg/mL, p < 0.001) and declined after alcohol withdrawal, correlating with improvements in craving, depression, and anxiety (p < 0.001). GA carriers of rs671, associated with lower ALDH2 activity, had higher NfL at baseline and one and two weeks of post-withdrawal period. Baseline NfL predicted DT occurrence in the following 2 two weeks, with a 27.2 pg/mL cut-off yielding 80% sensitivity and 72.4% specificity (AUC 0.84, OR 9.43; p < 0.001), with an odds ratio of 9.43 (95% CI 3.26–27.32; p < 0.001). Last, G-allele carriers of rs1799971 exhibited significant NfL reductions over 12 weeks of naltrexone, while non-carriers remained stable.

Discussion & Conclusions: NfL levels are elevated in AD and decrease with early abstinence. ALDH2 rs671 and OPRM1 rs1799971 polymorphisms may modulate neuroaxonal injury, recovery, and naltrexone treatment response. Moreover, NfL may predict DT occurrence, highlighting neuroaxonal pathology in its pathophysiology. These findings highlight the potential of integrating pharmacogenetics with objective biomarker (NfL) to characterize DT phenotype and recovery in AUD.