In elderly patients with AML, which patients should be considered fit or unfit for standard induction therapy?

The outcomes for acute myeloid leukemia (AML) patients older than 65 years of age are universally poor. Adverse chromosomal abnormalities, multidrug resistance, and antecedent BM failure are common in older patients with this diagnosis. Many elderly patients are frail and few are fit candidates for remission induction, with treatment-related mortality rising steeply in those older than 70 years of age. Nonetheless, achieving complete remission (CR) of leukemia is a requisite end point for prolonged survival, and data from large population registries have validated the use of intensive induction approaches over less-intensive therapies in patients up to the age of 80 years.¹  When faced with the conundrum of whether to use standard therapy in older patients, the prevailing wisdom has been to take a risk-adapted approach balancing the risk of induction death and the probability of achieving remission of leukemia. For patients not fit for intensive treatment, intermediate-intensity options are available, and trials of the hypomethylating agents, decitabine and azacytidine, in de novo elderly AML patients have been performed recently (successfully in myelodysplastic syndrome and AML with dysplasia and low BM blasts).^2,3  To provide clinicians with more objective guidance when considering intensive therapy for older patients, we performed a brief evidence-based review and present the results herein.

We carried out a comprehensive literature search of the PubMed database between January 1950 and May 2012. We used a combination of truncated and nontruncated MeSH terms and key words and built the search initially as follows: (“Leukemia, Myeloid, Acute” OR “Acute Myelocytic Leukemia(s)” OR “Acute Myeloid Leukemia”) AND (“aged” OR “elderly”) AND (“drug therapy, combination” OR “combination chemotherapy” OR “polychemotherapy”). This strategy provided 3124 hits. We then filtered for age > 65 years and articles written in English only and narrowed the search to 1544 citations. Adding the MeSH term “decision making” and the key words “hematopoietic stem cell transplantation comorbidity index” narrowed the search to 49 citations. Reference review added 4 additional manuscripts, resulting in a total of 52 articles, which were reviewed individually for relevance and content. After exclusion of review articles, practice guidelines, and therapeutic studies not dealing specifically with the issue of patient selection, we were left with 16 citations. All of these reports were retrospective and focused on identification of disease factors, host factors, or integrated assessments of both that might allow for outcome prediction in elderly AML patients treated intensively. To identify objective assessments of outcome, we picked 6 reports that compiled prognostic variables into comorbidity scoring systems.^4,6–10 

The Hematopoietic Stem Cell Transplantation Comorbidity Index (HCT-CI) is a validated model that predicts for outcome in AML patients undergoing stem cell transplantation and assigns patients a numeric risk score based on 17 weighted clinical parameters. This model was applied retrospectively to a large cohort of older adults receiving standard AML induction (3 days of idarubicin with 7 days of cytarabine). In this analysis, low-risk patients achieved CR rates similar to those of younger patients (64%), with few toxic induction deaths observed (3%).⁴  The HCT-CI was independently validated in a second, recent study.⁵  Kantarjian et al assigned risk categories based on age, karyotype, Eastern Cooperative Oncology Group (ECOG) performance status, abnormal organ function, treatment outside a laminar flow room, and duration of an antecedent disorder to older patients with AML and advanced myelodysplastic syndrome receiving intensive chemotherapy.⁶  Of the patients in the favorable-risk category, half were alive at 1 year and CR rates were in excess of 60%. Ten percent of these patients died during induction. In a multivariate analysis of AML patients older than 70 years who received cytarabine-based intensive chemotherapy, investigators at the MD Anderson Cancer Center (Houston, TX) demonstrated the following independent prognostic factors for 8-week mortality: age ≥ 80 years, complex karyotypes (≥ 3 abnormalities), poor ECOG performance status, and elevated serum creatinine > 1.3 mg/dL. Patients with none (28%), 1 (40%), 2 (23%), or ≥ 3 factors (9%) had estimated 8-week mortality rates of 16%, 31%, 55%, and 71%, respectively.⁷  The Spanish cooperative group (CETLAM) explored 3 variables (leukocyte count > 100 × 10⁹/L, serum creatinine > 1.2 mg/dL, and age > 50 years) in a large cohort of AML patients enrolled in multicenter studies to identify low-risk (no risk factors), intermediate-risk (1 risk factor), and high-risk (2 or 3 risk factors) groups. Six-week induction death rates of 6%, 19%, and 32%, respectively, were reported for each group.⁸  In another publication, the French cooperative group (ALFA) analyzed more than 400 elderly AML patients treated in a clinical trial and developed a “go, no go” decision index. These investigators proposed that patients who had unfavorable karyotypes and at least 2 of the following features: age > 75 years, ECOG performance status > 2, and WBC counts exceeding 100 × 10⁹/L, should not be treated with standard induction therapy.⁹  Finally, retrospective data from the German cooperative group (German Acute Myeloid Leukaemia Cooperative Group) identified 7 variables predicting for complete response or early death in older AML patients treated intensively: body temperature, age, de novo (versus secondary) leukemia, hemoglobin, platelet count, fibrinogen, and serum lactate dehydrogenase. Using these variables, they designed an online risk assessment score to assist with clinical decision making.¹⁰ 

The results of our analysis suggest that comorbidity scoring integrating patient- and disease-specific features is the best way to define patients fit and unfit for induction with intensive AML therapy. Decisions about fitness for treatment should not be made on the basis of age alone, but should arise from a comprehensive geriatric assessment of the patient. We recognize the growing importance of quality of life data in the clinical decision making process, however, there are no data showing how to use this information in an objective way. In conclusion, for the patient presented in the vignette, for whom no dysplastic features nor antecedent history of cytopenias are evident, it is reasonable to wait for confirmation of an intermediate- or good-risk karyotype and to proceed with standard induction therapy based on the use of clinical decision-making tools to justify this decision (grade 2C).

Conflict-of-interest disclosure: The authors declare no competing financial interests. Off-label drug use: None disclosed.

Ronan Swords MD, FRCPI, FRCPath, Sylvester Comprehensive Cancer Center, Division of Hematology/Oncology, Miller School of Medicine, 1475 NW 12th Ave, Miami, FL 33136; Phone: 305-243-3084; Fax: 305-243-6361; e-mail: rswords@med.miami.edu.