EHA 2016

Decoding the risk of thromboembolic events in lymphoma patients

Prof Darko Antic - Clinic For Hematology, Clinical Center Serbia, Belgrade, Serbia

We investigated thromboembolic events in lymphoma patients and we made a thrombosis lymphoma score with predicted factors about thromboembolic events in patients with chronic lymphoproliferative disorders.

Could you give us some background information about these events?

Thromboembolic events are a serious complication in cancer patients. Up to 12% of patients with cancer have thromboembolic events during their course of disease. Post mortem analysis confirmed that almost half of patients with cancer have thromboembolic disease. Also patients with thromboembolic events and cancer have an eightfold higher risk for death than patients without thromboembolic events. Because of that it is a very important thing for the treatment of cancer patients. When we talk about thromboembolic events in lymphoma patients we need to know that there is no adequate prognostic score for predicting thromboembolic events in patients with lymphoma. Also we need to know the incidence of thromboembolic events in patients with haematological malignancy is very different. For example, patients with acute leukaemia have an incidence of up to 12% but when we talk about lymphoma patients the incidence is very different, from 2-60%. What are the reasons for the very wide range of incidence? Namely we believe that that wide range is caused by different patient characteristics, different types of lymphomas, different treatment related factors and different follow-up types. Factors that predispose development of thromboembolic events are tumour by itself, also infections, different treatment and some acquired thrombophilic states. Mainly in a former patient, he has a thromboembolic event, he needs to receive anticoagulant therapy. Also his therapy for primary disease needs to be modified or interrupted. Also patients on anticoagulant therapy have a risk for bleeding, it’s a very important thing, and because of that the way of treatment of that patient needs to be changed.

So that’s what the score will be assessing?

We analysed 1,820 patients with chronic lymphoproliferative disorders. We analysed the arterial and venous events during the period from the time of diagnosis to the finish of treatment and three months after that. We analysed patients from our institution, from the Clinical Center of Serbia, it is an internal study. We confirmed that 5% of our patients had thromboembolic events, 99 patients of 1,820 patients in the whole group. Patients with high grade lymphomas, with aggressive lymphomas had significantly greater risk for the development of thromboembolic events than patients with other types of lymphomas. Where is the problem? When we talk about thromboprophylaxis we need to know next. We have a problem with hospitalised patients and we have a problem with outpatient settings. There are a few guidelines about that, namely ASCO has guidelines about hospitalised patients and ASCO recommends using thromboprophylaxis in patients with active cancer who are admitted to hospital with active disease or reduced mobility. ESMO criteria are more restricted but ACCP recommends using the Padua prediction score. When we talk about outpatient settings only ASCO has guidelines. Doctors need to use the Khorana score. The Khorana score is based on site of cancer, pre-chemotherapy blood count and body mass index but there is no specific prognostic score for patients with haematological malignancy, for patients with lymphoma. Because of that we analysed all parameters from the Padua score, all parameters from the Khorana score as well as specific parameters from lymphoma patients as well as mediastinum involvement, extra-nodal disease, development of neutropenia during treatment and after that we made our thrombosis lymphoma score based on eight risk factors. That thrombosis lymphoma score divides our group of patients into three groups – low risk, intermediate risk and high risk patients, namely. Patients with high risk in our group had thromboembolic events in 70% of cases and our prognostic score has very good positive predictive value, negative predictive value, specificity and sensitivity.

What does this score mean for patients and for providers?

For patients it means that patients in the high risk group need to receive thromboprophylaxis during treatment. They need to receive treatment but we need to take into account their status of cancer, their status of lymphoma, their status of bleeding risk and treatment type. An important question is what does it mean for the doctor? Many doctors support the claim that thromboprophylaxis is underused in patients with cancer. For example a low number of patients with cancer received thromboprophylaxis compared with patients, for example other hospitalised patients, for example, I don’t know, stroke, acute myocardial infarction, severe lung disease. Because of that our score is very important for doctors and those doctors need to think about the risk of thromboembolic events in haematological malignancy. Why? Because patients with haematological malignancy very often have a low number of platelets. Because of that, doctors don’t think about potential thromboembolic events but it is not true and every patient with a haematological malignancy needs to be classified and if a patient is a high risk patient they need to receive thromboprophylaxis. It’s a very important thing for doctors.

So I suppose a conclusion to that would be how would you like to see the ThroLy score be brought out to wider clinical use?

Our score has a limitation, we are aware about that. It is a score from a single institution experience and it is a score without external validation. In a further period we will make external validation of our score and we will try to publish our results. After that I think that our score will be used from doctors worldwide.