Skip to main content


Improving the prediction of thromboses in cancer patients

(Vienna, 14 September 2010) Up to 20% of all cancer patients develop a thrombosis in the course of their disease; this thrombosis constitutes an additional burden for the patients and often leads to negative effects on the further progression of the illness. In a study conducted at MedUni Vienna, which was published in the internationally renowned medical journal "Blood", it has now been possible to develop a new risk score that enables the assessment and prediction of the risk of thrombosis in cancer patients. The findings of this study could soon be integrated into therapeutic and/or prophylactic decision-making.

People suffering from cancer run a high risk of developing a venous thrombosis (the development of a blood clot in the deep leg veins or pulmonary embolism) which is potentially life-threatening and frequently a cause of shorter survival and higher mortality. On the one hand the tendency to develop a thrombosis arises from the tumour itself, on the other it depends on various factors that lead to an activation of blood coagulation. From a clinical viewpoint, an effective medication-based thrombosis prophylaxis would be urgently required in some cancer patients to prevent thrombotic complications and their consequences. It is therefore of major interest to identify – back at the diagnosis stage – those cancer patients who are at a particularly high risk of thrombosis and who would therefore benefit particularly from prophylactic blood-thinning therapy.

Accordingly, scientists of MedUni Vienna headed by Univ. Prof. Dr. Ingrid Pabinger-Fasching and Dr. Cihan Ay have set themselves the goal in this prospective study of exploring any risk factors that can predict the emergence of thrombosis in patients with cancer to enable targeted treatment in line with the individual risk profile. The Vienna Cancer and Thrombosis Study (CATS) was initiated by Univ. Prof. Dr. Ingrid Pabinger-Fasching at the Department of Medicine I (Head: Univ. Prof. C. Zielinski) in 2003. In previous work for this study it was possible to identify different laboratory parameters/biomarkers to predict the risk of thrombosis in cancer patients.

One current result of the study was accepted for publication in the internationally renowned medical journal "Blood" and also announced to the US public in a press release.  Now it has been possible to describe an improved risk model that enables the assessment of the risk of thrombosis. In this prospective observational study, 819 patients with cancer were examined between October 2003 and December 2008, of which 7.4% developed thrombosis in the period of observation. (For comparison, the rate of thrombosis in the general population is about 0.001%). The covered tumour entities comprised brain, breast, lung, stomach, colon, pancreas, kidney, prostate and lymph gland tumours.

A currently existing risk prediction model for the emergence of venous thromboses in cancer patients which includes the parameters of tumour entity, body-mass index, number of thrombocytes and leukocytes as well as haemoglobin level (all these are factors that are linked with an increased risk of cancer-associated thromboses) was complemented with the two new predictive biomarkers (laboratory parameters) soluble P-selectin and D-dimer in order to conduct a more precise stratification of tumour patients in high- and low-risk groups. sP-selectin is a cell adhesion molecule that promotes the development of blood clots; D-dimer is an activation marker of the coagulation system.

The significance of the two biomarkers for the prediction of venous thromboses in cancer patients was already described before in the Vienna Cancer and Thrombosis Study. Taking them into consideration in this new risk prediction model clearly improved the precision with the stratification of patients into various risk categories. Some 1/3 (35%) of all patients in the highest risk category of this new risk prediction model developed a thrombosis in the period under observation whereas in patients in the lowest risk category the rate of thrombosis was only 1%.

"Because the risk of thrombosis is not the same in all cancer patients and medication-based blood thinning leads to a high risk of haemorrhagic complications particularly in cancer patients, the classification of cancer patients by their risk of thrombosis is important," explains Univ. Prof. Dr. Ingrid Pabinger-Fasching, Professor of Haemostaseology at MedUni Vienna; she deduces from her study that patients with a high risk of thrombosis could benefit from routine thrombosis prophylaxis whereas the risk of haemorrhage seems to be predominant in patients with a low risk of thrombosis, so these would not be suitable candidates for routine anticoagulation (blood thinning).

"Our extended model reveals that cancer patients with a high risk of thrombosis can be identified even better," says Dr. Cihan Ay, first author of the study and employee of the Division of Haematology and Haemostaseology, and adds: "This model can help physicians in clinical practice make a tailored therapy decision and improve the prevention of leg vein thrombosis and pulmonary embolism by thrombosis prophylaxis, which on the one hand could optimise clinical benefits and also cost effectiveness and, on the other, minimise the risk of haemorrhagic complications by applying blood-thinning prophylactic therapy."

» Publication in BLOOD
Prediction of venous thromboembolism in cancer patients
Cihan Ay, Daniela Dunkler, Christine Marosi, Alexandru-Laurentiu Chiriac, Rainer Vormittag, Ralph Simanek, Peter Quehenberger, Christoph Zielinski and Ingrid Pabinger