Does co-administration of paroxetine change oxycodone analgesia: An interaction study in chronic pain patients.

Oxycodone is a strong opioid and it is increasingly used in the management of acute and chronic pain. The pharmacodynamic effects of oxycodone are mainly mediated by the μ-opioid receptor. However, its affinity for the μ-opioid receptor is significantly lower compared with that of morphine and it has been suggested that active metabolites may play a role in oxycodone analgesia. Oxycodone is mainly metabolized by hepatic cytochrome (CYP) enzymes 2D6 and 3A4. Oxycodone is metabolized to oxymorphone, a potent μ-opioid receptor agonist by CYP2D6. However, CYP3A4 is quantitatively a more important metabolic pathway. Chronic pain patients often use multiple medications. Therefore it is important to understand how blocking or inducing these metabolic pathways may affect oxycodone induced analgesia. The aim of this study was to find out whether blocking CYP2D6 would decrease oxycodone induced analgesia in chronic pain patients. The effects of the antidepressant paroxetine, a potent inhibitor of CYP2D6, on the analgesic effects and pharmacokinetics of oral oxycodone were studied in 20 chronic pain patients using a randomized, double-blind, placebo-controlled cross-over study design. Pain intensity and rescue analgesics were recorded daily, and the pharmacokinetics and pharmacodynamics of oxycodone were studied on the 7th day of concomitant paroxetine (20 mg/day) or placebo administration. The patients were genotyped for CYP2D6, 3A4, 3A5 and ABCB1. Paroxetine had significant effects on the metabolism of oxycodone but it had no statistically significant effect on oxycodone analgesia or use of morphine for rescue analgesia. Paroxetine increased the dose-adjusted mean AUC0-12h of oxycodone by 19% (-23 to 113%; P = 0.003), and that of noroxycodone by 100% (5-280%; P < 0.0001) but decreased the AUC0-12 h of oxymorphone by 67% (-100 to -22%; P < 0.0001) and that of noroxymorphone by 68% (-100 to -16%; P < 0.0001). Adverse effects were also recorded in a pain diary for both 7-day periods (placebo/paroxetine). The most common adverse effects were drowsiness and nausea/vomiting. One patient out of four reported dizziness and headache during paroxetine co-administration, whereas no patient reported these during placebo administration (P = 0.0471) indicating that these adverse effects were due to paroxetine. No statistically significant associations of the CYP2D6 or CYP3A4/5 genotype of the patients and the pharmacokinetics of oxycodone or its metabolites, extent of paroxetine-oxycodone interaction, or analgesic effects were observed probably due to the limited number of patients studied. The results of this study strongly suggest that CYP2D6 inhibition does not significantly change oxycodone analgesia in chronic pain patients and that the analgesic activity of oxycodone is mainly due to the parent compound and that metabolites, e.g. oxymorphone, play an insignificant role. The clinical implication of these results is that induction of the metabolism of oxycodone may lead to inadequate analgesia while increased drug effects can be expected after addition of potent CYP3A4/5 inhibitors particularly if combined with CYP2D6 inhibitors or when administered to poor metabolizers of CYP2D6.