Abstract
Background: Venous thromboembolism (VTE) is a frequent yet potentially preventable complication that increases morbidity and mortality among cancer patients. Chronic Lymphocytic Leukemia (CLL) is a heterogeneous hematological cancer. Patient specific factors (age, gender, performance status) and CLL specific markers (e.g. IgHV mutational status, Binet stage, and beta 2-microglobulin (B2M) levels) influence the clinical course of CLL. A considerable proportion of CLL patients are diagnosed with additional cancer after the CLL diagnosis. The risk of VTE in CLL is higher than the background population throughout the clinical course of CLL in contrast to solid tumors where VTE typically develops within the first few months after cancer diagnosis.
Aim: To investigate the risk of VTE in CLL patients according to patient specific factors, CLL specific markers, CLL treatment and additional cancer after the CLL diagnosis.
Methods: The Danish National CLL Registry provided prospectively collected clinical data from all CLL patients diagnosed in Denmark since 2008. These data were linked at person level by the unique civil registration number to the Danish National Patient Registry for information on VTE and additional cancers, to the Danish National Prescription Database for information on prescribed and reimbursed anticoagulation treatment and to the Danish Civil Registration System for information on vital status and emigration. Study entry was date of CLL diagnosis; subjects were followed until VTE, death, emigration or administrative censoring. Cumulative incidence of VTE according to Binet stage, B2M levels, IgHV mutational status and additional cancers diagnosed later than CLL were calculated. Additional cancer after the CLL diagnosis was defined as a time-varying exposure. Subjects contributed to the number being at risk of VTE in the "no additional cancer" group until the possible date of additional cancer at which they shifted to the "additional cancer after CLL" group. Death and additional cancer were treated as competing risks by use of the Aalen-Johansen estimator. In Cox proportional hazard models, we estimated the hazard ratios (HR) of VTE; WHO-performance score, previous VTE and sex were included in the models as categorical variables, age was included as a continuous variable and anticoagulation treatment and additional cancer were included as time-varying covariates in the appropriate models.
Results: From 2008 through 2015, 3609 subjects were diagnosed with CLL, 60.1% were males and the median age at CLL diagnosis was 70.4 years. Median follow-up was 2.6 years. During the study period, 12.7% (461) were diagnosed with an additional cancer and 19.9% (721) died. Anticoagulation treatment was given to 6.3% (227) of the CLL patients in the study period. The majority of the CLL patients had mutated IgHV, and B2M levels below 4mg/L (Table 1). A VTE was registered in 92 CLL patients during the study period. The cumulative incidence of VTE was highest in CLL patients exposed to additional cancer followed by CLL patients with B2M levels above 4 mg/L and unmutated IgHV. (Figure 1). CLL patients with additional cancer after the CLL diagnosis had a 3-4 fold higher risk of VTE compared with CLL patients without additional cancer. Previous VTE increased the risk of VTE after the CLL diagnosis by 5 fold. This was to some extent because a larger proportion of CLL patients with previous VTE were exposed to an additional cancer after the CLL diagnosis (23.9%) compared with the total study population (12.8%). The HR of VTE in case of previous VTE however, remained high after adjustment for additional cancer cancer and other risk factors (Table 1). Unmutated IgHV and B2M level >4 mg/L increased the risk of VTE, these associations persisted after adjustment for additional cancer (Table 1). CLL treatment tended to increase the risk of VTE, patient related factors and CLL specific markers did not affect the association markedly.
Conclusion: The risk of VTE was especially high in CLL patients with additional cancer. Previous VTE was also a substantial (expected) risk factor whereas high age and CLL specific markers (B2M level > 4 mg/L, unmutated IgHV) were significant but less strong risk factors of VTE.
No relevant conflicts of interest to declare.
