Dose Modification In MDS Pts Receiving Lenalidomide As a Means To Improve Patient Persistence

Abstract

BackgroundLenalidomide (LEN) has been studied in transfusion-dependent MDS patients (pts) with IPSS Low/Int-1 risk del5q disease with 55%-70% of pts becoming transfusion-independent (TI) (List, JCO, 2006). A review of Medicare claims showed MDS pts' response rates were higher when receiving >3 cycles of LEN therapy (Zeidan, Cancer 2013, in print). Interestingly, cytopenias and other cytotoxic effects of LEN can challenge physicians and pts to remain on treatment despite the finding that in del5q pts effects may indicate an expected TI Response under study conditions (Sekeres, JCO, 2008). For example, 80% of pts with del5q MDS on the Phase II MDS-003 study required dose modification (DM) due to an adverse event (AE) in achieving excellent response rates, yet less than a third had an AE leading to discontinuation of drug, suggesting that AEs can be managed through DM without discontinuation (Celgene, data on File). AimThis study examined DM and discontinuation in MDS real world pts treated with LEN in the US; results were compared with those seen in the MDS-004 clinical trial. MethodsCommercial and Medicare Advantage MDS pts newly initiating therapy with LEN were identified between 01/07 and 12/12 in a US claims database. Both pharmacy and medical benefits and ≥1 fill for LEN indicating a 10 mg/day dose were required; index date was the first LEN claim date. Pts were 18+ years old and continuously enrolled in the health plan 6 months prior to the index date (baseline period) and ≥6 months after the index date (follow-up). At least 1 claim with MDS diagnosis (ICD-9-CM 238.72-238.75) was required during study period. Pts with evidence of LEN during the baseline period were excluded.Pts with a 60+ day gap in LEN fills were considered to have discontinued LEN. Pts with an initial LEN fill of 10 mg/day dose who were subsequently prescribed 5 mg/day dose during the follow-up (prior to discontinuation) were considered to have had dose reduction. LEN fills prior to discontinuation was captured.The clinical study report summarizing the clinical trial data from MDS-004 was examined to identify the proportion of pts who dose modified and/or discontinued during trial.Real world use of LEN was then modeled using clinical trial results from MDS-004 to identify pts who may have benefitted from DM vs. discontinuation of drug. ResultsTable 1 summarizes use patterns in the MDS-004 clinical trial and real world claims database and estimates impact on real world use of LEN if DM strategy was applied based on MDS-004. In the real world dataset, 168 MDS pts were identified as initiating LEN therapy at 10mg/day. 26% (44 pts) had evidence of DM from 10 mgs to 5 mgs/day in the real world compared to 55% (38 pts) with DM in the MDS-004 study, where LEN dose reductions were required due to AEs. Similarly, 32% (54 pts) discontinued therapy in the real world, compared to 6(9%) pts in the clinical study. Pts who dose reduced in the real world had a higher average persistence of 10.8 LEN fills prior to discontinuation vs. an average of 4.25 LEN fills for pts without of dose reduction.Table 1DM, Discontinuation, and Persistence on Therapy of ‘Real-World' vs. Clinical DataA) MDS-004 Data for del5q pts starting at 10mg QD n (%) n= 69B) Real World Data MDS pts n (%) n=168C) Real World Data modeled after MDS-004 del5q pts n (%) n=168D) B and C DifferentialReceived DM38 (55%)44 (26%)92 (55%)48 (29%) Additional Real World pts who would have received DM in Clinical TrialDiscontinued6 (9%)54 (32%)15 (9%)39 (23%) Additional Real World pts who would have continued on study in Clinical Trial ConclusionsAnalysis of real world practices examined shows a small proportion of pts dose reduced LEN compared to rates in the clinical trial. LEN's overall impact may be unrealized as studies of prescribing patterns from US claims data suggest that many pt treatment courses are shortened or discontinued after a single course. Additonally, pts who DM were associated with longer persistence on LEN compared to those who did not. Further studies are warranted to assess the clinical factors association with more early discontinuation of LEN vs compared to DM, but we suspect this was in part due to expected and common side effects that were managed more commonly under study conditions. Modeling dosing patterns used in clinical trials of LEN suggests that there are likely pts that can stay on drug longer through DMs, which can be done safely and chances for response would be increased under such care. Disclosures:No relevant conflicts of interest to declare.