Characteristics and Predictors of Venous Thromboembolism Among Lymphoma Patients Undergoing Chemotherapy: A Cohort Study in China

Evaluation of the incidence, predictors, risk assessment scores and outcomes of thromboembolism in a cohort of Egyptian NHL patients - Real World Experience

Non-Hodgkin's Lymphoma (NHL) is the most common subtype of lymphoma. The incidence of venous thromboembolism (VTE) in aggressive NHL was estimated recently to be 11%. Several risk assessment scores and factors are available to help identify cancer patients at risk for developing VTE. Patients with a pathologically confirmed diagnosis of NHL were identified at the Oncology Center of Mansoura University. The study included 777 patients: 719 with DLBCL-NOS, 26 with Anaplastic-B-cell, and 32 with T-cell-rich-NHL. Data were retrospectively collected from electronic medical records, including clinical, radiological, and laboratory information related to VTE and NHL. The median age at NHL diagnosis was 53 years, (range: 18-98). There was a male predominance, 51.4% of the cases. At initial lymphoma diagnosis, VTE was identified in 46 (5.9%) patients, and 61 (7.9%) patients experienced VTE while undergoing chemotherapy. According to logistic regression analysis, a PS (performance status) ≥ 2, bulky lesions, and mediastinal masses were significant predictors of VTE at presentation, with P-values of 0.022, 0.002, and < 0.001, respectively. Meanwhile, NHL patients who developed VTE during chemotherapy had significantly poorer PS, higher absolute neutrophilic counts (ANC), neutrophil/lymphocyte ratio (NLR), platelet-lymphocyte ratio (PLR) and lactate dehydrogenase (LDH) levels than lymphoma patients without VTE, with P-values of 0.003,  0.034, 0.049, 0.01 and 0.007, respectively, as determined by multivariate analysis. The ROC curve identified the cut-off values of 4.875 × 109/L for ANC, 2.985 for NLR, 144.85 for PLR, and 417.5 U/L for LDH as potential markers for predicting VTE in NHL patients. Patients with a PS ≥ 2 and values exceeding these cut-offs for ANC, NLR, and PLR experienced significantly higher incidences of VTE than other groups, with P-values of 0.003, < 0.001, < 0.001, and < 0.001, respectively. At the end of the follow-up, the overall survival was significantly shortened by VTE occurring during chemotherapy, hypoalbuminemia,  intermediate-high and high international prognostic index (IPI) scores (intermediate-high and high), responses other than CR and relapse, all with P-values < 0.05. ECOG PS and Inflammatory markers such as NLR, PLR, and neutrophilic count could serve as predictors of the development of thrombotic events in patients with NHL-DLBCL. Additionally, the occurrence of VTE during chemotherapy is an independent poor prognostic marker for overall survival (OS).

Incidence of venous thromboembolism and predictive ability of age-adjusted international prognostic index for prediction of venous thromboembolism in Asian patients with diffuse large B-cell lymphoma

Diffuse large B-cell lymphoma (DLBCL) is one of the malignancies at high risk for the development of venous thromboembolism (VTE). We aimed to evaluate the incidence of VTE and the predictive ability of the age-adjusted international prognostic index (aaIPI) for the prediction of VTE among DLBCL patients. This was a retrospective cohort study including adult patients with newly diagnosed DLBCL. Differences in VTE occurrence within one year after diagnosis of DLBCL were estimated across aaIPI groups using the Kaplan-Meier model, Cox's model, and Gray's model with deaths regarded as competing events. Five hundred and ninety-one newly diagnosed DLBCL patients with a median age of 58 (range 16-93) years were included in this study. At a median follow-up time of 365 (range 2-365) days, VTE events were objectively diagnosed in 32 patients, giving a one-year cumulative incidence of VTE of 5.4% (95% confidence interval [CI], 3.7-7.6). Patients with aaIPI ≥ 2 had a significantly higher risk of VTE than patients with aaIPI < 2 (hazard ratio, 3.5; 95% CI, 1.6-7.8; p = 0.001 based on Cox's model and sub-distribution hazard ratio, 3.0; 95% CI, 1.3-6.7; p = 0.007 using Gray's model). The C-statistic of aaIPI was 0.65 (95% CI, 0.58-0.72). We demonstrated that the incidence of VTE in Asian DLBCL patients was not uncommon. The aaIPI was effective in determining the risk of VTE in DLBCL patients, even when including death as a competing event. aaIPI may be helpful in identifying patients at higher risk of VTE in DLBCL patients.

Development and validation of a nomogram model for predicting the risk of venous thromboembolism in lymphoma patients undergoing chemotherapy: a prospective cohort study conducted in China

BACKGROUND

The mechanism of Venous thromboembolism (VTE) is complicated and difficult to prevent due to factors such as bone marrow invasion, therapy, and immune-mediated effects. This study aims to establish a nomogram model for predicting the risk of thrombosis in lymphoma patients undergoing chemotherapy, which has been increasing over the past 30 years.

METHODS

The data of lymphoma patients from the Affiliated Cancer Hospital of Chongqing University in China between 2018 and 2020 were analyzed. This included age, sex, body mass index, ECOG score, histological type, Ann Arbour Stage, white blood cells count, haemoglobin level, platelet count, D-dimer level, and chemotherapy cycle. Univariate and multivariate cox analysis was used to determine the risk factors for VTE. Characteristic variables were selected to construct a nomogram model which was then evaluated using ROC curve and calibration.

RESULTS

Age, sex, PLT, D-dimer and chemotherapy cycle were considered as independent influencing factors of VTE. The mean (standard deviation) of the C index, AUC and Royston D statistics of 1000 cross-validations of the Nomogram model were 0.78 (0.01), 0.81 (0.01) and 1.61(0.07), respectively. It indicates a good calibration degree and applicability value as shown by the calibration curve. The DCA curve showed a rough threshold range of 0.05-0.60 with a good model.

CONCLUSIONS

We have established and validated a nomogram model for predicting the risk of thrombosis in lymphoma patients. This model can assess the risk of thrombosis in each individual patient, enabling the identification of high-risk groups and targeted preventive treatment.