Clinical prediction of opioid use disorder in chronic pain patients: a cohort-retrospective study with a pharmacogenetic approach.

BACKGROUND: Opioids are widely used in chronic non-cancer pain (CNCP) management. However, they remain controversial due to serious risk of causing opioid use disorder (OUD). Our main aim was to develop a predictive model for future clinical translation that include pharmacogenetic markers.

METHODS: An observational study was conducted in 806 pre-screened Spanish CNCP patients, under long-term use of opioids, to compare cases (with OUD, N.=137) with controls (without OUD, N.=669). Mu-opioid receptor 1 (OPRM1, A118G, rs1799971) and catechol-O-methyltransferase (COMT, G472A, rs4680) genetic variants plus cytochrome P450 2D6 (CYP2D6) liver enzyme phenotypes were analyzed. Socio-demographic, clinical and pharmacological outcomes were also registered. A logistic regression model was performed. The model performance and diagnostic accuracy were calculated.

RESULTS: OPRM1-AA genotype and CYP2D6 poor and ultrarapid metabolizers together with three other potential predictors: 1) age; 2) work disability; 3) oral morphine equivalent daily dose (MEDD), were selected with a satisfactory diagnostic accuracy (sensitivity: 0.82 and specificity: 0.85), goodness of fit (P=0.87) and discrimination (0.89). Cases were ten-year younger with lower incomes, more sleep disturbances, benzodiazepines use, and history of substance use disorder in front of controls.

CONCLUSIONS: Functional polymorphisms related to OPRM1 variant and CYP2D6 phenotypes may predict a higher OUD risk. Established risk factors such as young age, elevated MEDD and lower incomes were identified. A predictive model is expected to be implemented in clinical setting among CNCP patients under long-term opioids use.