Acute Leukemia of Ambiguous Lineage - Analysis of Survival Using SEER-Medicare Database

Background: Acute leukemia of ambiguous lineage (ALAL) is a group of leukemias which cannot be classified uniquely into myeloid or lymphoid lineage based on immunophenotyping. It includes the subtypes of acute undifferentiated leukemia and mixed phenotype acute leukemia. Due to its rarity, information on its survival at population level and optimal management strategies is not clear.

Methods: We used the SEER-Medicare database to describe the overall survival (OS) and treatment pattern of elderly patients (age ≥65) with ALAL. ICD-0-3 codes (9801, 9805, 9808, 9809) were used to identify patients from the database with pathologically confirmed ALAL, diagnosed between 1991-2010 with active follow-up. Patients diagnosed only on the basis of autopsy/death certificate and those who survived less than 30 days were excluded. To obtain treatment related information, in addition to the above mentioned criteria, patients included also had continuous part A and part B Medicare coverage and were not enrolled in HMO for 1 year before and anytime after the diagnosis.  Information on patient demographics, chemotherapeutic agents, and OS were obtained. Initial chemotherapy was grouped into AML-like, ALL-like and other regimens based upon the individual drugs that were given, as identified through the claims data. OS analysis was done using Kaplan-Meier method and its determinants were analyzed using Cox proportional hazard regression method with a significant p value < 0.05.

Results: A total of 705 patients were identified who met the study criteria. Median age of the cohort was 80 years. Baseline characteristics included males (51.3%), females (48.7%), whites (88.2%), blacks (6.5%), and other races (5.3%). Median OS according to the age groups were as follows- 4 months for age 65-70, 3 months for age 71-75, 2 months for age 76-80 and 2 months for age > 80 (p < 0.001).  Median OS was 2 months in both males and females (p=0.19) and also in all races (p=0.49). Patients with acute undifferentiated leukemia had a median OS of 2 months and acute biphenotypic leukemia had a median OS of 3 months (p=0.74).  A total of 151 patients received chemotherapy and the details of individual drugs given could be identified in 74 patients. Median OS was 7 months in patients who received chemotherapy and 2 months in those who did not receive chemotherapy (p < 0.001). Median OS did not significantly differ based on the type of therapy given – AML-like (Median OS 9.5 months), ALL-like (Median OS 6 months), and others (Median OS 6 months) (p = 0.70). On multivariate analysis, age less than 80 was associated with lower mortality [Age 65-70 vs age > 80:  HR 0.66, 95% CI 0.52-0.85;  age 71-75 vs age > 80:  HR 0.80, 95% CI 0.65-0.99 ; age 76-80 vs age > 80: HR 0.80, 95% CI 0.66-0.98, p=0.004)]. Similarly, treatment with chemotherapy reduced the hazard for mortality (HR 0.51, 95% CI 0.42-0.62, P < 0.001).

Conclusion: Our study suggests that the overall surival for ALAL in the elderly population remains poor. Although treatment is associated with a significant improvement in survival, only about 21% of patients have received therapy. Optimal choice of chemotherapy for this disease needs to be determined by more prospective studies.