Comparison of drug–drug interaction checking databases for interactions involving BCR-ABL tyrosine kinase inhibitors
Ayşe Günay, Eren Demirpolat, Mükerrem Betül Yerer Aycan, Ali ÜnalBackground and Aims: BCR-ABL tyrosine kinase inhibitors (TKIs) are used for the treatment of chronic myeloid leukemia and are commonly involved in clinically significant drug–drug interactions (DDIs). In this study, we aimed to evaluate the consensus of DDI checking databases for interactions involving BCR-ABL TKIs.
Methods: We checked DDIs of 100 drugs with six BCR-ABL TKIs—dasatinib, imatinib, nilotinib, ponatinib, bosutinib, and asciminib—in two subscription-based databases (UpToDate and Micromedex) and two open-access databases (Drugs.com and Medscape). Databases were compared in terms of severity ratings, literature support ratings, and general interaction mechanism definitions using Fleiss’ and Cohen’s kappa statistics.
Results: A total of 410 interactions were found. Nilotinib was the most interacted TKI, with 88 interactions. Drugs.com detected the highest number of interactions (n = 355). The overall agreement levels of databases for the severity ratings and general mechanisms were calculated as 0.13 (p = 0) and 0.28 (p = 0), respectively. The Micromedex- UpToDate pair showed the highest agreement level in terms of severity ratings and general mechanism definitions, with kappa values of 0.23 and 0.45, respectively.
Conclusion: The differences among databases for DDIs involving BCR-ABL TKIs were statistically significant. Therefore, healthcare practitioners should check DDIs in multiple databases.