Potentially Clinically Relevant Drug–Drug Interactions in Oncology Patients Receiving Chronic Opioid Therapy: Prevalence and Associated Factors

Background: Patients with malignant diseases receiving chronic opioid therapy are particularly susceptible to drug–drug interactions because of extensive polypharmacy, multimodal anticancer treatment, supportive care, and comorbidities. This study aimed to determine the prevalence of potentially clinically relevant drug–drug interactions and to identify factors associated with their occurrence. Methods: This exploratory observational cross-sectional pilot study included 49 adult oncology patients receiving chronic opioid therapy. Complete medication regimens were screened using the Medscape Drug Interaction Checker and Lexicomp. Lexicomp category D and X interactions were analysed descriptively. Because category D interactions were nearly universal and category X interactions were rare, the presence of at least one Medscape “Serious—Use Alternative” interaction was used pragmatically as the binary outcome for regression modelling. Potential predictors were assessed using univariable binary logistic regression and a parsimonious adjusted model with covariates selected using a clinically informed approach. Because of quasi-complete separation, Firth’s penalized-likelihood logistic regression was used as the primary adjusted analysis. Results: Lexicomp category D interactions were identified in 48 of 49 patients (98.0%), whereas category X interactions were present in 2 patients (4.1%). Medscape serious interactions were detected in 33 patients (67.3%). In the Firth-adjusted model, female sex was associated with lower odds of a Medscape serious interaction (adjusted OR 0.12, 95% CI 0.02–0.52), while cardiovascular disease was associated with higher odds (adjusted OR 5.16, 95% CI 1.27–26.31). Stage IV disease showed a positive but statistically non-significant association, and total medication count was not independently associated with the outcome in sensitivity analysis. Because of the small pilot sample and the resulting wide confidence intervals, these findings should be interpreted as exploratory and hypothesis-generating. Conclusions: Oncology patients receiving chronic opioid therapy had a high burden of potentially clinically relevant drug–drug interactions, predominantly Lexicomp category D interactions, warranting consideration of therapy modification and individualized monitoring rather than absolute avoidance. Regular medication review, use of complementary interaction databases, and individualized clinical assessment may improve pharmacotherapy safety in this vulnerable population.