Background: Acute myeloid leukemia (AML) is a common hematologic neoplasm with high incidence and mortality in the elderly. Our aims were to explore risk factors for early mortality in elderly AML patients and develop a new prognostic score.
Methods: We enrolled newly diagnosed AML patients age 60 and above at Taipei Veterans General Hospital between July 2008 and May 2017. The primary endpoint was early mortality, defined as death within two months after AML diagnosis. A multivariate Cox proportional hazards model was used to build a risk-scoring system incorporating significant risk factors for AML.
Results: The final cohort included 277 elderly AML patients. The median age was 74 (range 60-96), and 61.7% were male. The two-month mortality rate was 29.9%. Age ≥ 80 (adjusted HR 1.88), myocardial infarction (adjusted HR 1.87), ECOG ≥ 2 (adjusted HR 2.10), complex karyotype (adjusted HR 3.21), bone marrow blasts ≥ 70% (adjusted HR 1.88), WBC ≥ 100 × 109 /L (adjusted HR 3.31), and estimated glomerular filtration rate (eGFR) < 45 (adjusted HR 2.60) were identified as independent predictors for early mortality in the multivariate analysis. A simplified score incorporating the seven factors was developed with good predictive ability measured by Harrell's C statistic [0.72 (95% CI 0.66-0.78)].
Conclusions: We identified seven potential risk factors for early mortality and built up a new prognostic score for elderly AML patients. The new score may help clinicians stratify patients and initiate appropriate management. Further validation of our findings on other cohorts is warranted.
Keywords: acute myeloid leukemia; early mortality; elderly; epidemiology; prognostic models.
© 2020 The Authors. Cancer Medicine published by John Wiley & Sons Ltd.