关键词: CDK4/6抑制剂, 不良事件, FAERS, 真实世界, 比例失衡法, 药物警戒

Abstract:

Background and purpose: The development and approval of inhibitors of cyclin-dependent kinase 4/6 (CDK4/6) is an essential milestone in treating hormone receptor-positive metastatic breast cancer. The efficacy of these drugs is similar, but the adverse events (AE) are different, directly affecting the physician's choice of drug. There is no systematic study on the safety of CDK4/6 inhibitors in the real world. In this study, we compared the differences in AE of CDK4/6 inhibitors through signal mining in the FDA Adverse Event Reporting System (FAERS) and identified unknown AE signals to provide a reference for the clinical choice of treatment and monitoring AE. Methods: All data in the FAERS database were extracted from the first quarter of 2004 to the first quarter of 2023. After removing duplicates, data were analyzed by the disproportionality method for reports ranking palbociclib, abemaciclib, or ribociclib as the primary suspect. Signals were identified using the reporting odds ratio (ROR) and MHRA methods. Positive signals were required to meet the following criteria: the number of reports ≥3, the lower limit of the 95% confidence interval of the ROR >1, proportional reporting ratios (PRR) >2, and the χ² >4. Results: A total of 85 562 reports of AE associated with CDK4/6 inhibitors were identified. The highest signal intensity of palbociclib was observed in hematologic and lymphatic AE (leukopenia ROR = 20.01). Palbociclib had lower AE signals in the gastrointestinal, hepatic, and renal systems than the other drugs (diarrhea ROR = 1.95, gamma-glutamyltransferase increased ROR = 0.36, blood creatinine increased ROR = 1.01). Abemaciclib had the strongest signal in the gastrointestinal system (diarrhea ROR = 13.54); it also showed a strong AE signal in the hepatic and renal systems (gamma-glutamyltransferase increased ROR = 2.58, blood creatinine increased ROR = 7.74) and a lower AE signal than the other drugs in the blood and lymphatic systems (leukopenia ROR = 5.34). Ribociclib had a lower AE signal intensity in the blood and lymphatic system than palbociclib (leukopenia ROR = 7.55); however, among hepatic AE, ribociclib had the highest signal intensity of increased gamma-glutamyltransferase (ROR = 4.05). In rare severe hepatic systemic AE, abemaciclib had the strongest signal in hepatic failure (ROR = 3.50) and drug-induced liver injury (ROR = 4.68). Erythema multiforme was a newly identified signal in the abemaciclib reports (ROR = 3.06). Conclusion: The safety profile of CDK4/6 inhibitors varies. Analysis of the FAERS database revealed hematologic and lymphatic system toxicities for palbociclib and ribociclib and gastrointestinal and hepatorenal toxicities for abemaciclib. Erythema multiforme was found as a novel severe AE for abemaciclib. Individualized drug selection and monitoring of AE based on the patient’s physiological status and AE are needed during treatment.

Key words: CDK4/6 inhibitors, Adverse events, FAERS, Real world, Disproportionality analysis, Pharmacovigilance