Association between primary tumor site of colorectal cancer and RAS/BRAF mutational status with venous thromboembolism: A retrospective-prospective cohort study from 3 Croatian oncology centers

Right-sided colorectal cancer (RCRC) is associated with several negative clinicopathological and molecular features and also with worse survival than left-sided colorectal cancer (LCRC). A recent study has shown an association between RCRC and increased incidence of venous thromboembolism (VTE). However, the role of RAS/BRAF mutations considering the risk of VTE in metastatic colorectal cancer (mCRC) remains unclear. The aim of our study was to examine impact of tumor sidedness and RAS/BRAF status regarding VTE occurrence in a cohort of patients with mCRC. We conducted a multicenter, combined retrospective-prospective, observational study, with 234 patients (n=234) and known KRAS/NRAS/BRAF status who were treated with first-line chemotherapy with or without biological therapy for mCRC in three Croatian oncologic centers (General Hospital of Šibenik-Knin County, General Hospital Zadar and General Hospital Dubrovnik) from June 2013 to March 2020. Considering prospective arm of the study 40 patients with informed consent were enrolled with minimum follow-up of 1 year. The primary endpoint was incidence of VTE and its association with location of the primary tumor and RAS/BRAF status. VTE was defined as deep venous thrombosis and/or pulmonary embolism occurred 6 months before or at any time after the diagnosis of mCRC. Numerical data were described by the median and the interquartile range (IQR). Categorical variables were tested by Chi-square test and by Fisher's exact test. The normality of the distribution was tested by the Shapiro-Wilk test. Differences between two independent groups were tested by Mann-Whitney's U test. Logistic regression analysis was used to analyze the independent factors associated with VTE. All P values were two-sided. The level of significance was set at Alpha of 0.05. The median follow-up time was 21 months (IQR13-33). Of 234 patients included in analysis 26% had RCRC and 74% had LCRC. KRAS, NRAS and BRAF mutations were detected in 50%, 3% and 5% of patients, respectively. KRAS/NRAS mutations were more common in RCRC than in LCRC (62% vs 50% p=0.001) and the same finding has been noticed with BRAF mutations (12% vs 2% p =0.001). A total of 56 (24%) patients experienced VTE. The incidence was 43% in RCRC patients and 17.3% in LCRC patients (p < 0.001). According to mutational status, incidence was 48.2% in KRAS/NRAS-mutated patients, 10.7% in BRAF-mutated patients and 41% in KRAS/NRAS/BRAF wilde type patients. The results of logistic regression adjustment for known predictors of VTE including Khorana score, use of bevacizumab or anti-EGFR therapy and chemotherapy confirmed that RCRC was significant predictor of VTE (OR=4.7; 95% CI 1.74 – 12.57). However KRAS/NRAS/BRAF mutations were not associated with VTE (OR =0.86; 95% CI 0.11-6.59). In our study, primary tumor sidedness has been associated with VTE in patients with mCRC. Our results have not confirmed an association between biomarker status (KRAS, NRAS and BRAF) and an increased risk of VTE in mCRC patients.