Background: Primary mediastinal B-cell lymphoma (PMBCL) is a rare lymphoma subtype that typically affects young individuals, with a predominance in women (2:1). Although venous thromboembolism (VTE) is a well recognized complication in lymphoma patients, its incidence and associated risk factors in PMBCL are not well established. This study aimed to describe the clinical characteristics and incidence of VTE in this patient population. Moreover, since we hypothesize that the mediastinal tumor bulk can be a risk factor due to venous compression, we compared these findings to patients with bulky mediastinal Hodgkin Lymphoma (HL).

Methods: We retrospectively analyzed patients diagnosed with PMBCL between 2015 and 2025 at two referral hospitals in Brazil. Of the 42 identified cases, 30 had complete medical records and were included in the analysis. Evaluated variables included sex, age, occurrence and type of VTE, timing of diagnosis, site of the thrombotic event, concomitant risk factors (immobility, recent surgery, contraceptive use, pregnancy, prior VTE, history of thrombophilia, obesity), tumor mass diameter, presence of superior vena cava syndrome (SVCS), IPI score, extranodal involvement, and treatment modality. Patients were categorized into a VTE or non-VTE (NVTE) subgroups. Statistical analyses were performed in Python using Fisher's exact test for categorical variables and the Mann-Whitney U test for numerical variables, with a significance level of 0.05. As a comparator, we analysed the incidence of VTE in patients with bulky mediastinal HL.

Results: The mean follow-up was 37.7 months. Patients were predominantly female (60%), with a mean age of 36.4 ± 8.9 years (range 17–51). At diagnosis, the mean tumor mass diameter was 11.5 ± 3.3 cm (range 6–20), with 20 out of 30 patients presenting with masses >10 cm. SVCS was observed in 30% (9/30) of cases; 29/30 had IPI scores between 0–3, and 53.3% (16/30) had extranodal involvement. Initial treatment with R-DA-EPOCH was administered in 29 patients (96.7%), and 10% (3/30) experienced refractory or relapsed disease. The incidence rate (IR) of VTE was 40% (12/30), with 5 events occurring shortly before or at diagnosis and 4 within the first 5 weeks of treatment. Among the 12 VTE events, 7 were deep vein thrombosis (DVT), 3 catheter-related thrombosis (CRT), 1 pulmonary embolism (PE), and 1 superficial venous thrombosis (SVT). The most common DVT sites were the internal jugular and subclavian veins (4/7 cases), while the others involved deep upper limb veins. Among device-associated VTE cases, two were related to totally implanted catheters, and one to a peripherally inserted central catheter (PICC). Concomitant risk factors were found in 4/12 patients: 3 patients had obesity and one was pregnant at the time of the event. No patients died due to VTE. The only death in the cohort occurred 14 months after diagnosis, due to pulmonary infection. When comparing subgroups, tumor mass size and the presence of SVCS appeared clinically relevant but not statistically significant (p = 0.15 and p = 0.102, respectively). In contrast, for the bulky mediastinal HL group, we found a mean tumor mass of 12.06 ± 3.68 cm and a VTE IR of 11.7%. The 40% VTE IR observed in PMBCL patients was significantly higher than rates reported in the literature for lymphomas in general. Interestingly, no SVCS was observed in the HL group, compared to 30% in the PMBL group.

Conclusions: PMBCL is a lymphoma subtype closely linked to VTE and often underrepresented in the development of risk scores. Interestingly, there was a significant difference in VTE incidence between PMBCL and bulky mediastinal HL, suggesting that tumor mass diameter alone does not account for the occurrence of VTE. We hypothesize that the use of high-dose steroids within the R-DA-EPOCH regimen may contribute to this complication. Our findings support the need for stricter evaluation of prophylactic anticoagulation in this group, especially during the initial months of treatment. Caution is also advised when selecting intravascular devices, preferably avoiding those with higher thrombotic risk, such as PICC, due to the high rate of upper-limb thrombosis in our cohort.

This content is only available as a PDF.
2025
Sign in via your Institution