Comprehensive geriatric assessment predicts azacitidine treatment duration and survival in older patients with myelodysplastic syndromes

Background: Treatment of older patients with myelodysplastic syndrome (MDS) is based on disease biology and performance status. Performance status, however, does not reflect increasing co-morbidities, functional dependence or psychosocial issues in older patients. Patients and Methods: This prospective study evaluated the burden of geriatric related health issues, assessed feasibility of “tailored” Comprehensive Geriatric Assessment (CGA), and compared treatment duration and survival in older patients with MDS and oligoblastic acute myeloid leukemia with and without deficits in CGA domains (n = 98). Results: Although only 27 (28%) patients had an Eastern Cooperative Oncology Group score ≥2, 78% (n = 77) patients had deficits in at least one CGA domain. Deficits were spread across all CGA domains, including dependence for instrumental activity of daily living (iADL; n = 33, 34%). Importantly, patients who were dependent for iADL (3.7 ± 2.6 vs 12.1 ± 7.9; p =.009), had cognitive impairment (3.5 ± 2.1 vs. 10.9 ± 7.9; p =.034) or impaired mobility (3.8 ± 2.5 vs. 13.2 ± 7.6; p =.001) completed significantly less azacitidine cycles as compared to those without these deficits. Cox-proportional regression showed that iADL dependency (hazard ratio 3.37; p =.008) and higher comorbidities (hazard ratio 4.7; p <.001) were associated with poor prognosis independent of disease related factors. Poor survival of iADL dependent patients was seen in both azacitidine (6 vs 19 months; p <.001) and supportive care cohorts (26 vs 48 months; p =.01). Conclusion: CGA detected geriatric related health issues, predicted poor survival and identified patients less likely to continue and benefit from azacitidine. Hence, CGA should be included in the treatment decision algorithm of older patients with MDS.