Abstract
Thrombotic events are regularly observed in patients receiving treatment for Hodgkin Lymphoma (HL). However, sound data on incidence and risk factors are not known. The aim of the present study was thus to provide a comprehensive analysis of thrombotic events after multimodality treatment for HL.
A total of 5,773 patients ≤ 60 years treated within the German Hodgkin Study Group (GHSG) trials HD13-15 between January 2003 and December 2009 were included in this analysis. All reported venous and arterial thrombotic events occurring within 1 year after trial enrollment were evaluated and detailed information on patient characteristics, localizations, time of occurrence and risk factors were collected. We excluded thromboses of superficial veins and thrombophlebitis. Descriptive statistics and logistic regression were used for statistical analysis.
A total of 193 thrombotic events occurred for an incidence of 3.3%; there were 11 events in early-favorable, 27 in early-unfavorable and 155 in advanced stage HL, resulting in incidence rates of 0.7%, 1.3% and 7.3%, respectively. The incidence in advanced stage HL was significantly higher than in early stage HL (p<0.001); 175 events were venous and 18 arterial. The most common venous events consisted of arm vein thrombosis in 49.1% (n=86), DVT in 29.1% (n=51), PE in 13.1% (n=23) and sinus vein thrombosis in 1.7% (n=3) of cases. The most common arterial events were MI in 55.6% (n=10), lower extremities arterial thromboembolism in 22.2% (n=4) and CVI in 16.7% (n=3) of cases. 30.6% (n=59) of events were associated with intravenous catheters, 8.8% (n=17) were likely due to tumor compression and 1.0% (n=2) occurred despite prophylactic anticoagulation.
We found that 2.6% (n=5), 77.2% (n=149) and 20.2% (n=39) of cases occurred before, during and after chemotherapy respectively. In advanced HL patients treated with 8 x BEACOPPesc, 6 x BEACOPPesc or 8 x BEACOPP-14 , the incidence rates were 6.2% (n=44), 5.5% (n=39) and 10.1% (n=72) respectively. The incidence in patients treated with BEACOPP-14 was significantly higher than in patients treated with the other two regimens (p<0.01). Opposed to rather evenly distributed events during chemotherapy with BEACOPPesc, thromboses during treatment with BEACOPP-14 occurred more frequently at the beginning of chemotherapy.
We then analyzed potential risk factors in advanced stage patients using logistic regression analysis, adjusting for treatment in order to identify a high-risk group for developing a thrombotic event. The well-established Khorana score was not associated with a higher risk of thrombosis (odds ratio (OR) per unit [95% confidence interval]: 0.91 [0.76-1.1], p=0.33). Additionally, we screened 21 potential risk factors, including the thrombocyte-to-lymphocyte ratio. Only age (OR per year: 1.02 [1.01-1.03], p=0.01) and smoking (OR: 1.61 [1.07-2.43], p=0.02) emerged as significant risk factors. None of the other following potential risk factors was prognostic: thrombocyte-to-lymphocyte ratio, thrombocytes, leukocytes, lymphocytes, hemoglobin, albumin, WHO activity index, erythropoietin treatment, sex, body mass index, B-symptoms, erythrocyte sedimentation rate, extranodal disease, large mediastinal mass, more than 2 affected areas or Ann Arbor stage (all p≥0.10). Yet, when only including venous events, which are potentially preventable by prophylactic anticoagulation, neither age nor smoking were significant risk factors anymore (p≥0.10).
This study is the largest and most comprehensive analysis of thrombotic events in HL patients to date. Compared to both, early-favorable and early-unfavorable HL, advanced stage patients are at higher risk for thrombotic events. The most widely used Khorana score estimating thrombosis risk in cancer outpatients was not prognostic in the HL population investigated here. This is unsurprising considering the young age of the patient population investigated. Other risk factors were also not prognostic. This data does not imply a need for prophylactic anti-coagulation in outpatients treated for early-stage HL. In advanced-stage HL patients, routine prophylactic anticoagulation is not warranted. However, individual patients with additional risk factors that could not be evaluated such as history of thrombosis or reduced mobility might still benefit from prophylactic treatment.
Disclosures
Engert: Takeda, BMS: Consultancy, Honoraria, Research Funding.
Association between risk factors and retinal nerve fiber layer loss in early stages of diabetic retinopathy