Chronic eosinophilic leukemia, not otherwise specified, (CEL-NOS) is a rare myeloproliferative neoplasm defined by an autonomous clonal expansion of eosinophil precursors. CEL-NOS has long been considered an aggressive disease with a poor prognosis and high risk of transformation to blast phase. This impression was reinforced by a case series of 10 patients with CEL-NOS in which median overall survival was 22 months and in which one half of patients transformed to acute leukemia at a median of 20 months from time of diagnosis (Helbig et al, American Journal of Hematology, 2012). However, this has not been confirmed in any population-based or large cohort studies.

We sought to evaluate whether this poor prognosis is consistent across a larger patient database. We identified 487 patients with CEL-NOS in the Surveillance Epidemiology and End Result-18 database. Patients were predominantly male (60.5%) with a median age of 57 at time of diagnosis. Overall survival at 2 years from diagnosis for the cohort was 92.7% with a disease-specific survival of 97.2%. At 5 years overall survival remained relatively high (88.7%) as did disease-specific survival (95.6%). Median follow-up was 56.5 months (range 0-191 months). Median survival for patients who died of CEL was 5 months (range 0-46 months). This raises the possibility of a dichotomous patient population wherein there is one group of patients with a good prognosis, and a second group with aggressive disease and significantly poorer prognosis. Of note, the majority of the patients who died from CEL-NOS (21/23) died within two years of diagnosis.

Only three of the patients studied developed acute leukemia, two with acute myeloid leukemia and one acute lymphoblastic leukemia; all three died of their second malignancy within 28-60 months. An additional 11 patients developed lymphoma, 2 Hodgkin lymphoma and 9 non-Hodgkin lymphoma (NHL), including peripheral T-cell lymphoma (3/9), diffuse large B-cell lymphoma (3/9), angioimmunoblastic T-cell lymphoma (2/9), and Burkitt's lymphoma (1/9).

This data suggests that the current impression of CEL-NOS as a rare and aggressive disease with a poor prognosis may be more nuanced than previously appreciated. The limitations of SEER data are significant and include concerns regarding both accuracy of diagnosis and collection bias. However, the incongruence between current disease perception and the large-scale data available suggest that further study is both warranted and necessary. Next steps would include the creation of a prospective national registry of patients with CEL-NOS. This would assist both in giving appropriate disease-counseling for patients and in better assessing which patients might benefit from treatment with chemotherapy.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution