Thrombin Generation and D-Dimer Significantly Predict for Early Disease Progression and Mortality in Patients with Gastrointestinal Cancer

Background: Abnormalities of laboratory coagulation tests are common in cancer patients, underlying a subclinical hypercoagulable state. Due to the reciprocal interaction between coagulation and cancer, the biomarkers of hemostatic system activation are under evaluation as a tool for predicting cancer outcomes, disease progression, and mortality. Aim: In this analysis of a prospective cohort of patients with newly diagnosed metastatic gastro-intestinal (GI) cancer, we wanted to evaluate whether the pre-chemotherapy abnormalities of hemostatic biomarkers may predict for early-disease progression (E-DP), i.e. within 6 months from cancer diagnosis, and for 1-year overall survival (1-year OS). Methods: The study cohort included 304 newly diagnosed metastatic GI cancer patients, candidate to chemotherapy associated or not with immunotherapy, enrolled from March 2012 to March 2019 in the HYPERCAN study (ClinicalTrials.gov, ID# NCT02622815), an ongoing Italian prospective, multicenter, observational study; 191 healthy subjects acted as a control group. At diagnosis, before starting any curative chemotherapy, plasma samples were collected and tested for the following hypercoagulation biomarkers: D-dimer and fibrinogen by an automated coagulometer analyzer (ACL TOP500, Werfen Group), and prothrombin fragment 1+2 (F1+2) by ELISA (Siemens). In addition, thrombin generation (TG) potential was evaluated by Calibrated Automated Thrombogram (CAT) assay at 5 pM tissue factor and endogenous thrombin potential (TG ETP) and TG peak were analyzed. Clinical data [i.e. age, sex, BMI, ECOG Performance Status (ECOG-PS), relevant comorbidies] and the hemochromocytometric parameters were recorded at enrollment, whereas E-DP was clinically monitored every 3 chemotherapy cycles and during follow-up. Results: A cohort of 304 (205M/99F) metastatic GI cancer patients (206 colorectal and 98 gastric cancers) with a median age of 66 years (min-max: 29-85) was available for analysis. At enrollment, patients presented with a hypercoagulable state, as shown by significantly higher (p