Selection of patients with myelodysplastic syndromes from a large electronic medical records database and a study of the use of disease-modifying therapy in the United States

Objectives Treatment patterns for patients with myelodysplastic syndromes (MDS) outside clinical trials are not well described. Our objective was to evaluate treatment patterns and patient characteristics that influence time to disease-modifying therapy in patients with MDS in the USA. Design, participants and outcome measures Patients with MDS treated with erythropoiesis-stimulating agents (ESAs), iron chelation therapy, lenalidomide (LEN) and the hypomethylating agents (HMAs) azacitidine and decitabine, were retrospectively identified in the GE Centricity Electronic Medical Record database between January 2006 and February 2014; LEN and HMAs were defined as 'disease-modifying' therapies. Multivariable Cox regression models were used to ascertain patient characteristics associated with time to disease-modifying therapy. Results Of the 5162 patients with MDS, 35.7%, 40.3% and 4.6% received 1, >= 1 and >= 2 therapies, respectively. ESAs were the first-line (72.5%) and only (64.0%) treatment in the majority of patients who received >= 1 therapy. ESA-only patients were older and had more comorbidities, including isolated anaemia. LEN and HMAs were first-line treatment in 12.4% of patients each; 32.7% received LEN or HMAs at any time. The majority of del(5q) patients (77.6%) received >= 1 therapy, most commonly LEN, compared with 40% of patients without del(5q). A shorter time to disease-modifying therapy was significantly associated with absence of comorbidities, diagnosis after February 2008, lower baseline haemoglobin level, age <80 years and male gender (p<0.002 for all). Conclusions A high proportion of patients diagnosed with MDS in the USA do not receive approved disease-modifying therapies. It is important to improve access to these therapies.