Measurement of Thrombin Generation Is a Positive Predictive Biomarker of Venous Thromboembolism (VTE) in Metastatic Cancer Patients Enrolled in the Hypercan Study

Introduction The HYPERCAN study is an ongoing prospective Italian multicenter trial (Thromb.Res. 2014, Suppl 2, 182), designed to evaluate the role of laboratory hypercoagulation screening to predict early diagnosis (in healthy subjects), or prognosis and response to therapy in patients with either limited or metastatic cancers. Four cancer types are included, i.e. non-small cell lung [NSCL], gastric, colorectal, and breast cancers. The occurrence of clinically manifest VTE events, confirmed by objective diagnostic tests, is also recorded. Patients are followed up for 5 years or death.

Aim In a group of patients with metastatic cancer enrolled in the HYPERCAN program, we wanted to evaluate the role of thrombin generation assay (TG), fibrinogen, and D-dimer levels in predicting the occurrence of VTE in the follow up.

Methods As of June 2015, overall 831 patients with metastatic cancer have been enrolled. According to protocol, blood samples from these patients are collected at enrollment (baseline), after 3 and 6 chemotherapy cycles, and at end of treatment, or earlier if cancer disease progression. We measured the levels of TG, fibrinogen, and D-dimer in the baseline citrated plasma samples from the first 281 patients enrolled into the study (158 M/123 F; median age 64 years, range 32-84; NSCL = 56.2%, gastric = 11.1%, colorectal = 14.6%, breast = 18.1%). TG was measured by the Calibrated automated thrombogram (CAT assay, STAGO, France) at 5pM TF and results expressed as endogenous thrombin potential (ETP), fibrinogen and D-dimer were measured by commercial assays (Q.F.A. Thrombin; D-dimer HS; Werfen, Italy). Cut-off values were established by ROC curves; Kaplan Meier analysis was performed to define the VTE risk.

Results Overall the patient baseline ETP values as well as the fibrinogen and D-Dimer levels were significantly greater than those of a control group of healthy subjects (p<0.0001). Among the cancer types, ETP values were highest in patients with NSCL cancer, and lowest in those with gastric cancer (1899±517 nM*min vs 1622±550 nM*min; p=0.024). D-dimer and fibrinogen levels were greatest in gastric (801±186 ng/ml) and NSCL (484±190 mg/dl) cancers, respectively, and were both lowest in breast cancer patients. After a median follow up of 473 days, overall 37 VTE events were recorded in 36 patients: 16 pulmonary embolism (PE), 16 deep vein thrombosis (DVT), 3 superficial vein thrombosis, and 1 PE + DVT. Of these events, 69% occurred in NSCL, 17% in colon, 8.5% in gastric, and 5.5% in breast cancer patients. Median time to VTE was 5 months from enrollment, >80% of VTE developed during chemotherapy. Baseline ETP levels were significantly higher in patients with VTE than in patients without VTE (2020±618 nM*min vs 1799±467 nM*min; p=0.017). Univariate analysis (Kaplan-Meier) showed that patients with ETP>1750 nM*min had about 3-fold higher risk of developing VTE than those with ETP<1750 nM*min (HR:2.841, 95% CI 1.42-5.69 p=0.002). ETP predictive value remained significant by multivariate analysis after correction for age, gender, and tumor site (HR: 2.341, 95% CI 1.15-4.75, p=0.019). Differently, the baseline levels of D-dimer and fibrinogen did not significantly predict for VTE.

Conclusions These results reveal that ETP is a valuable marker in predicting VTE in metastatic cancer patients, therefore it can help to optimize the identification of the high risk subjects, which remains an important challenge and may improve the design of interventional studies of efficacy and safety of primary thromboprophylaxis in cancer patients during chemotherapy.

Project funded by AIRC “5xMILLE” n. 12237 grant from the “Italian Association for Cancer Research (AIRC)”.