Vol.:(0123456789) 1 3 Cancer Causes & Control https://doi.org/10.1007/s10552-018-1004-5 BRIEF REPORT Challenges in assessing the real incidence of chronic lymphocytic leukemia: 16 years of epidemiological data from the province of Girona, Spain Marta Solans 1,2,3  · Gemma Osca‑Gelis 3  · Raquel Comas 3  · Josep Maria Roncero 4  · David Gallardo 4  · Rafael Marcos‑Gragera 2,3  · Marc Saez 1,2 Received: 4 September 2017 / Accepted: 27 January 2018 © Springer International Publishing AG, part of Springer Nature 2018 Abstract Determining chronic lymphocytic leukemia (CLL)/small lymphocytic lymphoma (SLL) incidence is challenging for two reasons: cancer registries tend to underreport CLL cases and its diagnostic criteria changed markedly in 2008. No studies have reported incidence rates dealing with both difculties, and thus CLL/SLL burden in Europe is currently uncertain. Herein, we present accurate CLL/SLL incidence in a Spanish region during 1998–2013, using the population-based Girona Cancer Registry (GCR). We detected an 18.2% under-reporting of CLL/SLL cases when combining records from the GCR and additional information sources (i.e., records of fow cytometry laboratories, hospital registries and hematologists’ databases). In addition, age-adjusted rates (using the 2013 European population) changed from 7.57 (95% CI 6.87; 8.30) in 1998–2008 to 6.35 (95% CI 5.51; 7.30) in 2009–2013. Overall, completeness of CLL/SLL data requires accurate diagnosis and reporting of cases. Revision of cancer registry operations to include CLL/SLL-specifc surveillance is likely to ensure that the monitoring of this malignancy is entirely accurate. Keywords Chronic lymphocytic leukemia · Incidence · Under-reporting · 2008 Guidelines · Cancer registry Introduction Chronic lymphocytic leukemia (CLL) is the commonest B-cell neoplasm in Western countries [1]. Following the 2016 WHO recommendations, CLL and small lympho- cytic lymphoma (SLL) are classified together since both share clinical and pathological features. Determining CLL/SLL incidence is critical for unravelling its etiol- ogy—still poorly understood, and for informing health care administrators tasked with resource allocation. How- ever, its epidemiology is challenging for two main rea- sons. First, unlike other cancers, registry data has been shown to underestimate CLL disease incidence [ 2 –4]. Tumor registry reports generally rely on tissue pathol- ogy and inpatient treatment data; however, CLL can be diagnosed by flow cytometry analysis of peripheral blood cells, and the majority of patients are initially untreated. Second, changes in the diagnostic criteria of CLL/SLL have further affected its epidemiological surveillance [5–7]. Initially, the National Cancer Institute sponsored Working Group 1996 (NCI-WG 1996) guidelines classi- fied individuals with CLL-phenotype cells as CLL if the patient’s absolute lymphocyte count was ≥ 5.0 × 10 9 /L. The following 2008 WHO criteria effectively decreased CLL incidence by excluding lymphocytes other than B-cells from the absolute numerical cut-off, subsequently reclassifying patients with low B-cell counts and no clini- cal symptoms as monoclonal B lymphocytosis (MBL), a premalignant condition. Currently, European incidence * Marta Solans martasolans@gmail.com 1 Consortium for Biomedical Research in Epidemiology and Public Health (CIBERESP), Carlos III Institute of Health, Madrid, Spain 2 Research Group on Statistics, Econometrics and Health (GRECS), University of Girona, Campus de Montilivi, Carrer de la Universitat de Girona 10, 17003 Girona, Spain 3 Epidemiology Unit and Girona Cancer Registry, Oncology Coordination Plan, Department of Health, Autonomous Government of Catalonia, Catalan Institute of Oncology, Girona, Spain 4 Haematological Service, University Hospital Dr. Josep Trueta, Catalan Institute of Oncology, Girona, Spain