Potential drug interactions in patients with a history of cancer.

e19651 Background: Cancer patients (CP) are frequently exposed to polypharmacy, which may increase their risk for drug interactions. Our study aims were to 1) determine and compare the prevalence and clinical significance of potential drug interactions (PDI) in CP vs. non-cancer patients (NCP) and 2) identify clinical risk factors associated with PDI in CP. METHODS Self-reported prescription medication data from 4,975 patients were extracted from the U.S. population-based National Health and Nutrition Examination Survey and screened for PDI using specialized drug interaction software (iFacts, version 13.10). The clinical significance of each PDI was graded on a 5-point scale (1=most significant and 5=least significant). Clinical significance was judged based on the degree of severity of drug interaction, the rate of onset of side effects (immediate vs. delayed), and the level of scientific evidence for the drug interaction. Summary statistics and logistic regression models were used to assess the impact of cancer history on the risk and significance of PDI. RESULTS After screening, 302 CP and 907 NCP were eligible for analyses: mean ages were 62 (SD=1.14) and 60 (SD=0.92) years; 44% and 41% were men; 88% and 77% were White, respectively. CP indicated using a mean of 4.5 (SD=0.22) different classes of medications while NCP reported using 3.8 (SD=0.09) different drugs. In terms of PDI, 40% and 43% of CP and NCP, respectively, had interactions. Among them, 12% were rated as level 1 in significance in both patient groups. In multivariate analysis, CP were significantly less likely to be at risk for any type of PDI (OR 0.71, 95% CI 0.52-0.97, p=0.034) when compared to NCP. For level 1 PDI, CP remained less at risk than NCP (OR 0.65, 95% CI 0.46-0.92, p=0.017). Additional factors that were associated with any PDI or level 1 PDI in CP were advanced age and low household income. Medications that were most commonly prescribed among CP with level 1 PDI included atorvastatin (5.2%) and digoxin (4.3%). CONCLUSIONS Although CP appear to be less susceptible to PDI than NCP, the prevalence of medication interactions among CP remains suboptimal. Specific subgroups of CP may be particularly prone to PDI, underscoring the importance of increased vigilance.