Tumour-Associated Tissue Factor (TF)-mRNA Is A Precursor for Rapid TF-Microvesicle Release and A Potential Predictive Marker for the Risk of Pulmonary Embolism (PE) in Gastrointestinal Cancer Patients

Author(s): Camille Ettelaie, Naima E. Benelhaj, Sophie Featherby, Farzana Haque, Miriam J Johnson, Justin Cooke, Anthony Maraveyas

Predicting which cancer patients are at risk of thrombosis remains a key challenge to effective thromboprophylaxis. This study was based on the hypothesis that the rapid release of TF-containing MV occurs in cancer cells that possess high levels of TF mRNA, permitting the transient but amplified TF-protein production in response to cellular activation. To gather preliminary clinical evidence for this hypothesis the correlation between the levels of tumour-associated TF mRNA and incidence of Pulmonary Embolism (PE) in Gastrointestinal cancer patients (GI) was assessed using stringently-selected patient cohorts. Furthermore, the rapid TF-MV release was assessed in three cell lines with high TF mRNA in which protein-translation or mRNA-transcription were inhibited separately. On applying the exclusion criteria, the study accrued 9 clinical samples with incidental PE (colonic n=5; gastroesophageal n=4) which were type, gender and stage of cancer matched one-to-one with patients without PE (9+9 samples). Total-RNA was extracted from the samples using a FFPE-RNA extraction kit and TF mRNA was quantified using a quantitative real-time PCR procedure along with a standard curve prepared using in vitro-transcribed TF mRNA. Relative amounts of PAR2 mRNA were also determined. Analysis of absolute amounts of tumour-associated TF mRNA showed significant increase in patients who developed PE (mean=26.931±15.371 pg/100ng-total RNA; median=5.340 pg/100ng-total RNA; range=0.4-131.43 pg/100ng-total RNA) compared to those who didn’t (mean=0.098±0.023 pg/100ng-total RNA; median=0.110 pg/100ng-total RNA; range=0-0.19 pg/100ng-total RNA). Receiver Operating Characteristic (ROC) analysis returned an area under the curve of 1. In contrast, no significant difference in PAR2 mRNA was recorded. To provide an explanation for these findings, inhibition of protein-translation using cycloheximide prevented the incorporation of TF but not the MV release. However, blocking of