Background:

 

Colorectal cancer (CRC) is the third most common tumor disease worldwide. A typical complication in cancer is the occurrence of thromboembolic events significantly worsening patient prognosis. Vice versa, patients having been treated for venous thrombosis exhibit a four-fold higher risk of developing cancer in the consecutive years. While this mutual interaction between thrombosis and cancer is well known, current treatment regimens do not address both diseases simultaneously. Thus, the aim of the present study is to identify possible effects of the direct oral FXa inhibitor Apixaban as well as of the protease-activated receptor for FXa (PAR2) on colorectal cancer cell growth in vivo.

 

Methods:

 

Male and female C57Bl/6J (WT) and PAR2 knockout (PAR2 ko) mice with C57Bl/6 background were s.c. injected with murine colon carcinoma cells (MC38) in the right flank. Tumor growth was observed over 21 days. Apixaban was applied orally by gavage. Treatment groups were: I. control (vehicle treatment), II. 5 mg/ body weight and III. 50 mg/ body weight Apixaban daily. At the end of the treatment period organs were removed for molecular analyses. Liver and tumor tissues were also examined histologically. All animal experiments were approved by the responsible animal welfare committee. The addition, direct effect of FXa and PAR2 on growth and migration was determined in culture human and murine cancer cells in vitro. 

 

Results:

 

PAR2-deficiency resulted in a reduced tumor volume compared to WT animals. This effect also evident by a significantly reduced tumor weight in PAR2 ko mice after excision of the tumor at the end of the study. In comparison, oral treatment with Apixaban (5 mg and 50 mg) did not markedly affect tumor development in vivo. In addition, a significantly lower rate of complications was seen in PAR2 ko mice during the experiment. Interestingly, the spleen was larger in relation to body weight in PAR2 ko animals in all three experimental groups. We observed no obvious histological differences within the tumor tissue between the treatment groups or animal strains. Further molecular analyses are currently ongoing. In comparison, FXa did directly stimulate cancer growth and migration in both cultured human and murine cells in vitro. These effects were mimicked by peptide-stimulated activation of PAR2 but not after activation the classical thrombin receptor PAR1. In addition, we observed a significantly increased activation of mitogenic signaling pathways such as p44/42, p38 and AKT pointing towards an elevated sensitization of CRC cell to coagulation factor signaling by FXa.

 

 

Conclusion:

While deficiency of the FXa receptor PAR2 resulted in reduced growth of CRC cells in vivo, oral treatment with Apixaban did not affect tumor size or weight. In vitro, FXa did stimulate CRC growth and migration indicating a potential direct effect of FXa on CRC growth. Whether insufficient drug concentration or its bioavailability at the site of tumor growth may account for the observed lack of effect in vivo is to date unclear. Pharmacological inhibition of PAR2, on the other hand, could represent a future therapeutic strategy in the treatment of CRC.