Health Care Utilization, Risk of Infection and Bleeding Among MDS Patients During Periods of Transfusion Dependency or Independency with or without Active Therapy

Clinical manifestations of MDS include recurrent infections from neutropenia and bleeding from thrombocytopenia. These events often result in increased healthcare resource utilization (ER visits and hospitalizations) and are associated with disease-related deaths. Understanding the interplay between disease activity, transfusion needs, and medical interventions that can initially worsen baseline cytopenias, may help to improve treatment approaches for MDS patients.

This retrospective claims review analyzed the occurrence of infection and significant bleeding (defined as GI, intracranial, and hospitalized bleeds as well as bleeding deaths) related to pre-defined follow-up periods which captured the clinical activity of MDS based on transfusion dependency vs. independency and active therapy Rx (defined as use of lenalidomide, 5-azacitidine, or decitabine).

Claims data from a US national commercial health insurer were retrospectively reviewed to assess the impact of RBC transfusion dependency (TD) on bleeding and infection events. Pts ≥ 18 yrs with ≥ 1 claim for MDS (ICD-9-CM codes 238.72–238.75) between 01 Jan 07 and 31 Dec 09 were assessed using the 1^st MDS date as the index date. Continuous enrollment in a commercial or Medicare Advantage plan with a medical and pharmacy benefit for 6 mos before the index date (baseline period) and for a variable period after the index date was required. Pt follow-up was divided according to 4 unique treatment periods: transfusion independent (TI) periods without active therapy termed “watch and wait” (A), TI periods on active therapy (B), (TD) periods on no active therapy (C), TD on active therapy (D). Health care utilization focused on infectious complications and significant bleeding events and was captured within each follow-up period. TD was defined as ≥ to 2 RBC transfusions in 8 weeks and TI as < 2 transfusions in 8 weeks. The number (%) of periods with events is reported. Health care utilization was also calculated as “incidence per person-year” to account for the variable length of the pre-defined periods and to compare results across the follow-up period cohorts.

A total of 3, 886 pts with MDS with a mean age of 67.1 years (SD=14.9) represented 4, 007 total follow-up periods. Each patient could account for multiple periods. The follow-up periods (A, B, C, D) appear to predict expected disease activity based on the results of incidence per person-year in each period. For example – the TI periods (A and B) had lower incidence per person-year compared to the TD periods (C and D). Furthermore, within the TI periods, those requiring medical treatment (B) had more events compared to the watch and wait periods (A). Interestingly, the TD periods were not differentiated based on medical intervention and the TI periods with therapy (B) had lower events than either TD periods.

This retrospective analysis helps to define the health care utilization patterns of MDS pts based on transfusion requirements and treatment interventions. Periods of TD are associated with higher numbers of medical events compared to TI periods. These data also suggest that periods of TI on therapy had fewer events compared to either TD period and may allay concerns that active therapy could worsen baseline cytopenias. In fact, active therapy plays a role in lowering supportive care requirements and resources utilization for many pts and may also reduce the incidence of MDS-related medical events. Results demonstrate that active therapy should be considered in all eligible TD MDS pts.

Smith:Celgene: Consultancy; Genzyme: Consultancy; Incyte: Consultancy; Infinity: Consultancy; Merck-Serono: Research Funding; Synta: Research Funding; Celator: Research Funding; Calistoga: Research Funding; BMS: Research Funding; Novartis: Research Funding. Mahmoud:Celgene: Employment. Khan:Celgene: Employment.