Primary colon cancer site as a predictor of bevacizumab effectiveness in metastatic colon cancer patients.

Abstract

574 Background: Bevacizumab has been approved by the FDA for first-line treatment in combination with chemotherapy for metastatic colon cancer. Although bevacizumab extends progression free survival and overall survival for many metastatic colon cancer patients, primary cancer site has not been examined as a predictor of its effectiveness. Methods: Data regarding bevacizumab treatment for colorectal cancer were abstracted from the medical record of 175 patients treated at a large multi-specialty clinic. Data were collected regarding demographics, body mass index, weight loss, lifestyle variables (smoking status, drinking habit), history of colon polyps, family history of cancer, primary site of tumor in the colon, tumor stage, lymph node involvement, metastasis, tumor grade of metastasis, cancer cell type, symptoms related to the cancer, and other comorbidities (cardiovascular disease, diabetes mellitus, Crohn's disease, diverticulitis, and irritable bowel syndrome).Tumor response was measured by radiologic and biochemical markers. Univariate and multivariate analyses were performed using Fisher's exact test, Wilconox rank sum test, and unconditional logistic regression modeling. Results: Univariate analysis demonstrated that positive tumor response was positively correlated with blood in stool (p=0.0301), advanced lymph node stage (p=0.048), and adenocarcinoma cancer cell type (p=0.0387), but negatively correlated with chemotherapy treatment of the primary tumor (p=0.0387) and primary site of the colon cancer (p=0.0328). Multivariate stepwise logistic regression analysis demonstrated that patients with primary tumor located in the transverse or descending/sigmoid colon were more likely (odds ratio=5.88, 95% confidence interval: 1.72–20.00; p=0.0046) to have a positive tumor response than patients whose primary tumor site was cecal/ascending colon. Conclusions: Primary site of metastatic colon cancer may influence the effectiveness of bevacizumab therapy. Future randomized, controlled trials are needed to examine the primary site as a predictor of tumor response in metastatic colon cancer patients using bevacizumab therapy. No significant financial relationships to disclose.