Prognostic Value of Cytogenetics in Patients with Acute Myeloid Leukemia Aged More Than 60 Years at Diagnosis: Identification of Subgroups Taking Benefit from Intensive Chemotherapy. Results from a Large Single Center Retrospective Study.

Abstract

Cytogenetic analysis performed at diagnosis is known as the most important independent prognostic factor in adult acute myeloid leukemia (AML) treated with intensive chemotherapy. This major predictor of outcome has been mostly identified in younger patients who are categorized into three risk groups: favorable, intermediate and adverse. However, the prognostic value of chromosome abnormalities in elderly patients with AML aged 60 years or older has been evaluated in much fewer studies and overall prognosis remains very poor in this population with a very small number of long-term survivors using current standard chemotherapy. To further investigate the prognostic relevance of cytogenetic abnormalities in older patients, we reviewed 376 patients with de novo or secondary AML aged 60 years or more and available cytogenetic data at diagnosis. APL were excluded and no patient received allogeneic bone marrow transplantation. Of 376 patients, 280 (74%) received induction chemotherapy based on association with cytarabine (ara-C) and anthracyclins according to GOELAMS SA2, SA3, SA4 and SA2002 prospective trials between 1988 and 2005. Patients who were not treated with induction regimen received best supportive care or low dose ara-C. Overall complete remission rate was 42%, induction failure 38% and induction death 20%. With a median follow-up of 11.5 years for survivors, overall survival (OS) was 4.5% at 5 years (95% confidence interval [CI], 3.8%–5.8%) and disease free survival (DFS) was 4.3% at 5 years (95% CI, 3.7%–5.9%). Using a proportional hazards model with normal cytogenetics as reference group, 3 prognostic subgroups were identified for OS: low-risk (t(8;21) and inv(16)) in 20 of 280 patients (7%), standard-risk (+8, +11, −13/del(13q), t(11q23), +22, all in no complex karyotypes) in 175 of 280 patients (63%) and high-risk (−5/del(5q), −7/del(7q), −17/del(17p), −20/del(20q), +21, rare aberrations, complex≥3 and complex≥5) in 85 of 280 patients (30%). In multivariate analyses using a Cox model with backward selection procedure, high-risk cytogenetics (p<0.001), age>65 years (p=0.01), multilineage dysplasia on marrow at diagnosis (p=0.008) and white blood cell count (WBC) (p=0.03) appeared as independent prognostic factors on OS and DFS. Secondary AML did not appear as prognostic factor in multivariate analysis (p=0.18) but in a log-rank test, AML secondary to hematologic malignancy did significantly worse than AML secondary to solid tumors in both OS and DFS (p=0.01 and p=0.04, respectively). OS distribution between treated and untreated patients was compared using a log-rank test with homogeneous subgroups based on cytogenetic risk group and age. Three-years OS was significantly better in patients <70 years with standard-risk cytogenetics (p=0.008 if age<65years and p=0.02 if age <70 years). For high-risk cytogenetic group, 3-years OS was improved only in patients <65 years with WBC<30G/L (p=0.03). These results confirm the independent prognostic value of pretreatment karyotype, age >65 years, WBC and associated multilineage dysplasia in elderly patients with AML. Overall prognosis remains very poor and intensive chemotherapy does not improve OS in patients aged >70 years and high-risk cytogenetic groups.