INTRODUCTION: Thromboembolism (TE) in lymphoma patients is gathering substantial attention due to its impact on morbidity and mortality of those patients. The association between lymphoma and increased risk for TE development, especially venous thromboembolism (VTE), has lately been well established through numerous publications. Thrombosis Lymphoma (ThroLy) score has been initially developed as a simple risk assessment model for the risk of TE development in lymphoma patients. It has been both internally and externally validated in several studies, which dominantly included patients with non-Hodgkin lymphoma (NHL). Therefore, aim of our study is to analyse and validate ThroLy score in an extensive cohort of Hodgkin lymphoma (HL) patients.
METHODS: A total of 5509 newly diagnosed HL patients, from the German Hodgkin Study Group (GHSG) HD13-15 trials, were included in this study. Data has been obtained for all venous and arterial TE events in HL patients from time of diagnosis to 3 months after the last cycle of therapy. TE was diagnosed objectively based on radiographic studies (duplex venous ultrasound, contrast-enhanced thoracic computed tomography scan, magnetic resonance imaging (MRI) - for central nervous system (CNS) thrombosis, or angiograms (for arterial thrombosis), clinical examination and laboratory evaluation. Based on ThroLy score, patients were divided in three risk categories: low (score 0-1), intermediate (score 2-3) and high risk (score >3). Patients with intermediate and high-risk score were classified at risk. The validation was conducted through Chi-square test, ROC analysis and logistic regression.
RESULTS: The mean patients' age was 35.9 years (range, 18-75 years); 55.7% were males. The majority of patients had limited or intermediate stage of disease: Ann Arbor stage I 10.6%, and stage II 57.5%. 190 (3.4%) patients developed thromboembolic events, 173 patients with VTE (3.14%), and 17 with arterial TE (0.31%), respectively. Chi-square test showed statistically significant association between TE and ThroLy score, both in three risk groups (chi-square = 18.236, p≤0.001) and two risk groups: low and at risk (chi-square = 18.029, p≤0.001). The sensitivity, specificity, negative and positive predictive value were 49%, 65%, 95%, and 97%, respectively. Binary logistic regression of ThroLy score showed statistically significant performance in prediction of TE events in HL patients, with satisfactory validity indicators (Ombinus Test chi-square = 11.668, p=0.001; AIC=44.956, BIC = 97.869). Diagnostic accuracy of ThroLy score was calculated via ROC curve (area under curve (AUC)=0.57).
CONCLUSION: ThroLy score demonstrated its capability of risk prediction for TE events in HL patients. The limited statistical performance of the ThroLy score requires further research towards possible score enhancement.
Engert:AstraZeneca: Honoraria; MSD Sharp & Dohme: Honoraria; Bristol Myers Squibb: Honoraria, Research Funding; Affimed Therapeutics: Research Funding; Sandoz: Honoraria; Takeda: Honoraria, Research Funding.
Asterisk with author names denotes non-ASH members.