Impact of pleural effusion (PE) on treatment adherence, discontinuation, switching, and dose modification in patients with chronic myelogenous leukemia (CML).

Abstract

6616 Background: CML patients may develop PE as an adverse event of some tyrosine kinase inhibitor (TKI) therapies. PE requires medical care and may compromise the course of CML treatment. The objective of this retrospective study is to estimate the impact of PE on the course of TKI therapy in CML patients. Methods: The MarketScan and Ingenix Impact databases (2001-2009) were combined to identify adult CML patients (ICD-9CM code 205.1x) who received ≥1 prescription of any TKI before the index date and had continuous enrollment ≥6 months prior to and after the index date, defined as 30 days before the first PE diagnosis (ICD-9CM code 511.9x) for PE patients and randomly selected among all the eligible calendar dates for PE-free controls. Study cohorts were matched on a 1:1 ratio using propensity score matching and followed for 6 months after the index date. Treatment adherence was measured by the proportion of days covered (PDC) and compared between PE and PE-free patients using generalized linear models where as discontinuation, switching, and dose decrease rates were compared using Cox-proportional hazard models. Treatment discontinuation was defined as a gap of ≥60 days between the last day of supply of a prescription fill and the next prescription fill of a TKI. Switching was defined as the initiation of a TKI other than the index TKI. Dose decrease was measured relative to a patient’s index dose. Results: A total of 186 matched pairs were evaluated. PE patients were significantly less adherent compared to PE-free patients with an average PDC of 0.51 vs. 0.64 (p<.05). PE patients also had significantly higher discontinuation and switching rates compared to PE-free patients and similar rates of dose decrease. Conclusions: Compared to PE-free patients, significantly more PE patients discontinued their TKI therapy or switched to another TKI, and a similar proportion had a dose decrease. Inherent limitations of the data, such as the lack of information on CML phase, preclude drawing conclusions as to whether therapy was discontinued or switched due to PE or not. PE PE-free Hazard ratio P Discontinuation 39% 27% 1.47 0.048* Switch 12% 4% 4.60 0.002* Dose decrease 10% 7% 1.67 0.323