Individual risk factors for venous thromboembolism (VTE) in cancer patients treated with targeted therapies (TT)

• Published in: Journal of clinical oncology Vol. 27; no. 15_suppl; p. e20549
• Main Authors: Soini, B., Caffo, O., Ferro, A., Ambrosini, G., Murgia, V., Valduga, F., Caldara, A., Eccher, C., Tumulo, S., Galligioni, E.
• Format: Journal Article
• Language: English
• Published: 20.05.2009
• ISSN: 0732-183X; 1527-7755
• DOI: 10.1200/jco.2009.27.15_suppl.e20549

Abstract only e20549 Background: Many anticancer drugs, alone or in combination, can significantly increase the risk of VTE in cancer patients. Vascular complications however, including VTE, have emerged as a relevant toxicity also with targeted therapies (TT) (Elice, Am. J. Hematol.2008). Recently, a simple model for predicting risk of VTE in cancer patients receiving chemotherapy (CT) has been developed, based on a number of clinical and laboratory variables (Khorana, Blood 2008). In this report, we have evaluated the incidence of VTE and assessed the predictive value of the Khorana's model, in cancer patients receiving TT, alone or combined with CT. Methods: From a consecutive series of 278 patients, treated with TT ± CT between January 2006 and August 2008, we have identified for each patient the baseline values, before TT treatment, of the following variables: age, performance status, primary site of cancer, stage, comorbidities, body mass index, hemoglobin, leukocyte and platelet count and use of erythropoiesis stimulating agents. All variables have been correlated with the occurrence of VTE, at the uni- and multivariate analysis. Results: TT treatment consisted of trastuzumab in 103 patients (37%), bevacizumab in 69 (25%), cetuximab in 29 (10%), rituximab in 41 (15%), sunitinib in 16 (6%), erlotinib in 12 (4%) and sorafenib in 8 (3%). In 163 patients (59%) TT were administered concurrently with CT. In a period of 2–360 days (median 125), 14 patients (5%) developed deep vein thrombosis, complicated by pulmonary embolism in 5. None of the variables of Khorana's model has shown any significant value, in predicting VTE risk, as well as any association with different TT (i.e. bevacizumab or sunitinib) or the concomitant administration of CT. Conclusions: In our experience, the Khorana's model does not appear able to predict the risk of VTE, in patiens receiving TT treatments. Other models may be required for identifying high risk patients, who would benefit from a prophylaxis of thrombotic events. No significant financial relationships to disclose.