Chronic lymphocytic leukemia incidence is substantially higher than estimated from tumor registry data

Abstract

Background: Although chronic lymphocytic leukemia (CLL) often is described as the most common leukemia in the U.S. and Western Europe, to the authors' knowledge the true incidence of CLL in the U.S. is unknown. CLL incidence is estimated from tumor registry reports based on tissue pathology and cancer treatment data. Tumor registry data may underestimate the incidence of CLL substantially because CLL can be diagnosed by flow cytometric analysis of peripheral blood cells, and the majority of patients do not require treatment at the time of diagnosis.

Methods: To test the hypothesis that CLL has a higher incidence than estimated from tumor registry data, the authors compared the actual and reported incidence of CLL for a 10-year interval at the Central Arkansas Veterans Healthcare System (CAVHS). The accuracy of surveillance methods for new diagnoses of CLL was confirmed by reviewing the lymphocyte counts in 45,009 CAVHS patients over a 4-year period.

Results: The tumor registry correctly reported 58 of 93 patients with CLL (62.4%) who were diagnosed between January 1, 1990 and December 31, 1999. The tumor registry correctly reported 100% of patients with CLL diagnosed between 1990-1991 but reported only 34.5% of patients with CLL diagnosed between 1998-1999.

Conclusions: The incidence of CLL in the CAVHS was 37.6% higher than estimated from tumor registry data due to an increase in the use of peripheral blood immunophenotype as the only diagnostic test for CLL over the time period of the study. These data suggest that the true incidence of CLL may be substantially higher than estimated from tumor registry data.