Thrombosis as a Treatment Complication in Hodgkin Lymphoma Patients

A Comprehensive Analysis of Three Prospective Randomized German Hodgkin Study Group (GHSG) Trials

S. Borchmann; H. Müller; I. Hude; M. Fuchs; P. Borchmann; A. Engert


Ann Oncol. 2019;30(8):1329-1334. 

In This Article


Patient Population

Patient characteristics are presented in Table 1. Our study cohort was typical of the general HL population. However, we excluded patients >60 years in the HD 14 and HD 15 trial and those with a WHO activity index >3 in all trials included in this study, rendering the study population likely to be at a slightly lower risk of thrombotic events than the general HL population due to trial selection.

Thrombotic Events

In total, we observed 193 thrombotic events for a total incidence of 3.3% in our trials. Of these, 11 events occurred in early-favorable stage (0.7%), 27 in early-unfavorable stage (1.3%) and 155 in advanced stage HL patients (7.3%), the latter incidence being significantly higher than in early-stages (P < 0.001). Of the 193 thrombotic events, 175 were venous thrombotic events (90.7%), 3 newly diagnosed post-thrombotic syndromes due to previously undiagnosed thrombotic events and 15 arterial thrombotic events (7.8%). Focusing on the small group of arterial thrombotic events, we still observed a significantly increased incidence in advanced stage patients (n = 11, 0.5%) compared with early-stage patients (n = 4, 0.1%, P = 0.0057).

Different treatments in early-stages of HL had no detectable effect on the incidence of thrombotic events. We observed 5 thrombotic events in 623 patients treated with ABVD (doxorubicin, bleomycin, vinblastine, dacarbazine) (0.8%), no thrombotic events in 209 patients treated with ABV (doxorubicin, bleomycin, vinblastine) (0.0%), 4 thrombotic events in 620 patients treated with AVD (doxorubicin, vinblastine, dacarbazine) (0.6%) and 2 thrombotic events in 186 patients treated with AV (doxorubicin, vinblastine) (1.1%). In early-unfavorable stage, we observed 10 thrombotic events in 833 patients treated with 4 cycles of ABVD (1.2%) and 15 thrombotic events in 1174 patients treated with 2 cycles of BEACOPPesc (bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, prednisone) +2 cycles of ABVD (1.3%).

As treatment with BEACOPPesc is not recommended for patients older than 60 years, this age group was excluded in the HD14 and HD15 trials investigating early-unfavorable and advanced stage HL. In early-favorable stage, we observed a thrombotic event rate of 1.4% in 212 patients aged ≥60 years. Patients aged 50–60 years had thrombotic event rates of 1.2% (5/431) in early-stages and 10.0% (27/269) in advanced stages.

DVT of the arm (81/175, 46.3%), DVT of the leg (43/175, 24.6%) and PE (10/175, 5.7%) were the most frequent venous events, while myocardial infarction (10/15, 66.7%) and arterial thromboembolism of the leg (3/15, 20%) were the most common arterial events.

We found that 17 thrombotic events were associated with local tumor compression (8.8%), 59 were catheter-associated (30.6%) and 2 occurred despite prophylactic anticoagulation (1.0%).

Timing of Thrombotic Events

Most venous thrombotic events occurred during chemotherapy (138/175, 78.9%), with 38 of 175 (21.7%) venous thrombotic events occurring within the first 30 days of chemotherapy and 71 of 175 (40.6%) within the first 60 days. Fewer events were observed between diagnosis and the start of chemotherapy (5/175, 2.9%) as well as after chemotherapy (32/175, 18.3%). In comparison, 0 of 15, 9 of 15 (60.0%) and 6 of 15 (40.0%) arterial events occurred before, during and after chemotherapy, respectively.

Risk Factors for Thrombotic Events

The results of a logistic regression analysis of potential risk factors in advanced stage HL are depicted in Table 3. The treatment patients received was a major risk factor for venous thrombotic events. The relative frequency for any venous thrombotic events in patients treated with 8×BEACOPP-14 was 9.4% (67/710) compared with 5.7% with 8×BEACOPPesc (40/705) and 5.1% with 6×BEACOPPesc (36/711), the first being statistically significantly higher than with 8×BEACOPPesc (P = 0.0079). In addition to treatment, only age [OR (per year) =1.02, P = 0.01] and smoking (OR =1.61, P = 0.02) were risk factors for any thrombotic events. Of note, a higher Khorana score[12] was not associated with an increased risk for any [OR (per point) =0.91, P = 0.31] and any venous [OR (per point) =0.92, P = 0.41] thrombotic event.

Thrombotic Events in Female Patients Using Oral Contraception

Additionally, we carried out a subgroup analysis of female patients with available data on oral contraception (n = 248). We observed a tendency towards increased risk of venous thrombotic events with 10 of 147 (6.8%) and 4 of 101 (3.9%) thromboses in patients with and without oral contraception use during treatment, respectively (OR =1.77, CI =0.54–5.81). Because of the low event numbers in this subset analysis, we did not perform formal statistical testing in this subgroup.