Prospective genotype-guided drug selection and dosing in alcohol withdrawal syndrome complicated by seizures: a randomized clinical trial
Year: 2026
Type: article
Study designs: Randomized Controlled Trial, Clinical Trial
Abstract: OBJECTIVES: To determine whether prospective genotype-guided drug selection and dosing can reduce cumulative withdrawal severity and improve treatment tolerability in alcohol withdrawal syndrome (AWS) complicated by seizures. METHODS: In this single-center, parallel-group randomized clinical trial, 100 inpatients with AWS and seizures were allocated 1:1 to a genotype-guided prescribing strategy or control. All patients were genotyped on admission for CYP2D6*4, CYP2C19*2, CYP2C19*17, CYP3A4*1B, CYP3A5*3, and ABCB1 3435C>T. Intervention-group physicians received an individualized report with recommendations on benzodiazepine and anticonvulsant selection and starting dose; control-group physicians received a format-matched report without individualized recommendations. The primary outcome was area under the CIWA-Ar curve over 0-72 h, assessed by a blinded rater. RESULTS: AUC CIWA-Ar was 774.3 ± 175.9 vs. 903.6 ± 199.5 score-hours (mean difference -129.3; 95 % CI -204.0 to -54.6; p<0.001). Confirmed stabilization by 48 h occurred in 31/50 vs. 20/50 patients (log-rank p=0.020; Benjamini-Hochberg-adjusted p=0.051, not significant after multiplicity correction). UKU side-effect scores were lower on days 3 and 5 (both p<0.001), and excessive sedation occurred in 1/50 vs. 9/50 patients (p=0.016). Total benzodiazepine load was lower with genotype guidance (25.0 vs. 27.5 mg diazepam equivalents; p=0.016). The composite rate of clinically significant adverse events and the rates of delirium and unplanned treatment modification did not differ significantly. CONCLUSIONS: A prospective genotype-informed prescribing strategy was associated with lower cumulative withdrawal severity and less excessive sedation while using a lower benzodiazepine load. Because the intervention also changed anticonvulsant selection and drug concentrations were not measured, these findings support the clinical strategy as a whole rather than any single genotype-to-dose rule. Multicenter, prospectively registered trials with measured drug exposure are warranted.
Language: English
Subfield: Pharmacology
Domain: Life Sciences
Open Access status: hybrid
Estimated APC paid: $2,712
OpenAlex ID: W7233988579
