https://orcid.org/0000-0002-3981-658X
Key Points
There are age-, sex- and race-related disparities in receipt of HMAs among MDS patients favoring younger (65-74 years old), White males.
High HMA treatment discontinuation rates and incomplete cycles in practice show significant divergence from recommended clinical guidelines.
Abstract
Compared to data from clinical trials, US population-level data show decreased effectiveness of hypomethylating agents (HMAs) in patients with myelodysplastic syndromes (MDS). We sought to identify factors associated with patterns of HMA use. In this retrospective cohort study, we identified 49,514 individuals ≥65 years of age with incident MDS during the years 2012-2013 using the 2011-2014 Medicare claims dataset. We collected data on demographics, clinical characteristics, disease severity, and area-level socioeconomic measures. Multivariable logistic regression analysis was used to evaluate factors associated with receipt of HMA and duration of HMA therapy. A total of 7,935 patients (16.1%) received HMAs. In adjusted analyses, the oldest age cohort (>85 years) had lower odds of receiving HMAs compared to their younger counterparts (65-74 years) [adjusted odds ratio (aOR): 0.41 (0.38-0.44)]. Females and Black patients had significantly lower odds than males and Whites to receive HMA [(aOR: 0.81, (0.77-0.86) for females; aOR: 0.70 (0.62-0.8) for Blacks]. In HMA recipients, factors associated with lower odds of receiving >4 cycles of HMAs included patients treated with decitabine [(aOR: 0.7, (0.62-0.78)], having 2-3 cytopenias [(aOR: 0.69, (0.61-0.78)], being nursing home residents [aOR = 0.64; 95% CI: 0.46-0.90), and having high frailty [(aOR: 0.50 (95% CI: 0.34-0.75)]. We identified age-, sex- and race-related disparities in receipt of HMAs, favoring younger, White males. The duration of therapy in HMA-treated patients in routine clinical practice showed wide divergence from recommended clinical guidelines.
Author notes
Data Sharing Statement
The data (Medicare claims files) underlying this article were provided by Centers for Medicare and Medicaid Services (CMS) under the data user agreement (RSCH-2018-52214). As per the data user agreement, the data can only be shared between the signatories of the agreement. This data therefore is not publicly accessible as it is not permitted to those who are not signatories to the DUA as per CMS requirement.
