Evaluation of drug-drug interactions in cancer patients treated at a university hospital in North Cyprus using two interaction databases

Background: Drug interactions in oncology are of clinical importance owing to the inherent use of multiple medications in cancer treatment which predisposes patients to drug-related problems. Aim: This study aimed to compare two electronic databases based on the frequency, mechanism and severity of drug-drug interactions (DDIs) in cancer treatment at Near East University Hospital. Method: A retrospective observational study of hospitalized cancer patients who had received more than one chemotherapy and/or supportive-care drugs from April 2017 to April 2019. Lexi-interact tool by Lexicomp and Drugs.com databases were used to check (DDIs and all detected interactions were categorized based on the severity-level and mechanism of interaction. Results: A total of 681 prescriptions were evaluated and the median medication per patient was 4 (IQR 3-6). Drugs.com identified potential DDIs in 129 (84.9%) patients while Lexicomp identified potential DDIs in 113 (74.3%) patients. Drugs.com reported DDIs of 394 pairs while Lexicomp reported DDIs of 313 pairs. More than 50% of the potential DDIs were classified as pharmacodynamic interactions in both databases. There were varied reports of severity of potential DDIs, but the test of agreement using kappa index was 0.592 (95% CI: 0.502-0.682, P = 0.0001) and this was interpreted as a moderate agreement between the two databases. Conclusion: Lexicomp documented more detailed information relevant to clinical practice. However, Drugs.com with more sensitivity, detected more potential DDIs. Therefore, we suggest the use of at least two drug databases for quality screening, especially for patients predisposed to polypharmacy.

Automated summary

This study compared two electronic databases on drug-drug interactions in cancer treatment. The results showed a moderate agreement between the databases, with Drugs.com being more sensitive and Lexicomp providing more detailed clinical information. Using at least two drug databases is recommended for quality screening, especially for patients predisposed to polypharmacy.