Objective: The risk of venous thromboembolism has been reported to increase when receiving bevacizumab. Many cancer patients are reported to have elevated D-dimer levels. It is not clear what D-dimer level might indicate an increased risk of venous thromboembolism in the colorectal cancer patients treated with bevacizumab-containing chemotherapy.
Methods: The D-dimer levels and any event concurrent with an elevated D-dimer level were evaluated in patients receiving bevacizumab. The D-dimer cut-off level was determined using the receiver-operating characteristic analysis. The selection criteria were as follows: histologically proven metastatic and unresectable colorectal adenocarcinoma; no prior chemotherapy containing bevacizumab; D-dimer test performed repetitively on the baseline and during bevacizumab administration; no venous thromboembolism identified at the baseline; and enhanced computed tomographic scan performed every 2 months.
Results: Sixty-nine patients were included. The chemotherapy regimens with bevacizumab included the regimen of 5-fluorouracil, leucovorin and oxaliplatin (FOLFOX), the regimen of 5-fluorouracil, leucovorin and irinotecan (FOLFIRI), and leucovorin-modulated 5-fluorouracil. The median baseline D-dimer level was 1.2 µg/ml. The appropriate D-dimer cut-off level was 3 µg/ml with the negative predictive value of 98% and relative risk of 6.9. Twenty-one of 69 patients showed elevated D-dimer levels of >3 µg/ml, with 11 patients for unknown reasons, 6 with tumor progression, 3 with venous thromboembolism and 1 with sepsis. In the remaining 48 patients whose D-dimer levels were ≤3 µg/ml, only one patient developed a venous thromboembolism.
Conclusions: A D-dimer cut-off level of 3 µg/ml might be a useful indicator level to exclude venous thromboembolism or show an increased risk for venous thromboembolism in colorectal cancer patients treated with bevacizumab-containing chemotherapy.