Clinical judgement and geriatric assessment for predicting prognosis and chemotherapy completion in older patients with a hematological malignancy

Prior research has focused on comparing clinical judgment with geriatric assessments in identifying potentially unfit patients or the prognostic value of geriatric impairments. In this study, we set out to compare and combine geriatric impairments with clinical judgment in predicting mortality and non-completion of chemotherapy in older patients with a hematological malignancy. Between March 2004 and August 2014, a multi-dimensional geriatric assessment was performed in consecutive patients aged >= 65 years diagnosed with a hematological malignancy at the Innsbruck university hospital. Associations between geriatric assessment, clinical judgment (derived from initial treatment decision) and outcome (mortality, non-completion) were analyzed. Patient sample consisted of 157 patients, of which 37% was 80 + years of age. Aggressive non-Hodgkin lymphoma (29%), acute myeloid leukemia (27%) and myelodysplastic syndromes (20%) were most common diagnoses. Prevalence of an impaired geriatric assessment (i.e. impairments in >= 2 of eight assessed domains) was 71% and similar for patients receiving standard (71%) or attenuated treatment (72%). Clinical judgment was the strongest predictor of mortality (p = 0.003); addition of geriatric assessment allowed for further stratification of patients with a good (>75% one year survival), intermediate (50-60%) and poor prognosis (<30%). Cumulative burden of geriatric impairments further increased risk of mortality. Clinical judgment was greatly superior to geriatric assessment in identifying patients at risk for non-completion. The combination of clinical judgment and geriatric assessment allows for better prognostic stratification of patients compared to clinical judgment alone.