Chronic lymphocytic leukemia (CLL) is a heterogeneous disease developing from an early-stage monoclonal B-lymphocytic phase. Data regarding development of prediagnostic lymphocytosis are sparse, and differences in the rate of prediagnostic lymphocytic progression between patients may further add prognostic information for follow-up and treatment decisions in CLL.


We aimed to investigate the development of lymphocytosis prior to diagnosis in a population-based cohort of patients with CLL. Furthermore, we wanted to assess the prognostic information in this prediagnostic progression.


All patients diagnosed with CLL in the Greater Copenhagen area between 2008 and 2016 were included in the analysis. Prediagnostic blood results were obtained from the Copenhagen Primary Care Laboratory (CopLab) Database encompassing all blood test results requested by Copenhagen general practitioners between 2000 and 2016. Data collection was approved by the National Board of Health and the Danish Data Protection Agency.

A piecewise model was employed to identify the onset of prediagnostic lymphocyte increase for each patient. We then fitted a linear mixed model for the lymphocyte count, using only observations taken between the estimated time of lymphocyte increase and time of diagnosis. We next assessed the prognostic significance of the subject-specific progression rate using Cox regression models. Models were adjusted for age, calendar year of diagnosis and the CLL-IPI variables.


A total of 710 CLL patients were identified. Patients had a median follow-up in the CopLab database of 8 years (IQR 5-11) and a median number of observations of 7 (IQR 3-13) prior to diagnosis. The median age at time of diagnosis was 69 years (interquartile range (IQR) 62-77), 57% were male. 86% Binet A, 14% Binet B or C. 25% unmutated IGHV, 5% had del(17p), and 11% had elevated levels of Beta-2-Microglobulin. From time of diagnosis, patients were followed for a median of 3.4 years (IQR 1.8-5.8). Of the 710 patients, 158 died and 170 received treatment for CLL.

The estimated onset of lymphocyte progression was 4.98 years (95% CI [5.79, 4.09]) before the diagnosis. The prediagnostic lymphocyte counts and the piecewise model are shown in Figure 1. The median yearly progression in the lymphocyte count was 31% (IQR [23%, 39%]). Adjusted for CLL-IPI, gender and calendar year of diagnosis, the prediagnostic and patient-specific lymphocyte progression was an independent predictor for progression free survival (hazard ratio (HR) 17.99; 95 % CI [5.76, 56.21]; p<0.001). Figure 2 illustrates three subject specific progression rates(2A) and the predicted progression-free survival for these patients (2B). However, the patient-specific lymphocyte progression did not reach statistical significance as an independent predictor for overall survival (HR 2.16; 95% CI [0.55, 8.54]; p<0.27).


We demonstrate that the lymphocyte progression prior to diagnosis is detectable 5 years before diagnosis. The patient-specific progression of prediagnostic lymphocytosis is significantly associated with progression-free survival independent of CLL-IPI variables and gender. Thus, if these findings can be validated in an independent cohort, it not only enlightens our understanding of the natural history of CLL, but also adds a novel prognostic parameter for individualization of follow-up and treatment in CLL.


Niemann: Roche: Consultancy, Other: Travel grant; Janssen: Consultancy, Membership on an entity's Board of Directors or advisory committees, Other: Travel grant; Novartis: Other: Travel grant; Gilead: Consultancy, Membership on an entity's Board of Directors or advisory committees, Other: Travel grant; Novo Nordisk Foundation: Research Funding; The Danish Cancer Society: Research Funding; AbbVie: Consultancy, Membership on an entity's Board of Directors or advisory committees, Other: Travel grant, Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.