RESUMO
OBJECTIVE: Prognostic factors in colorectal cancer have lesser been evaluated in developing countries. This study aims to determine overall survival and prognostic factors for metastatic colorectal cancer patients who were non-operable and received chemotherapy. METHODS: The study retrospectively investigated 67 inoperable metastatic colorectal cancer patients at Square Hospital, Bangladesh. The primary endpoint was overall survival, and the secondary endpoints were prognostic association with factors. Survival probabilities were calculated by non-parametric Kaplan-Meier method and compared by log-rank test. Univariate and multivariable Cox proportional hazard models were implemented to assess the prognostic association. RESULTS: Median survival of the entire cohort was 14 months (95% confidence interval: 11-25). In multivariable analysis, two prognostic factors were independently associated with survival: Karnofsky performance status and carcinoembryonic antigen. Patients with Karnofsky performance status <70 had significant higher risk of death than those with Karnofsky performance status ≥70 (adjusted hazard ratio 4.25, 95% confidence interval: 2.15-8.39). Higher risk of death was found to be associated with higher carcinoembryonic antigen: adjusted hazard ratio was 1.72 (95% confidence interval: 0.81-3.68) and 2.96 (95% confidence interval: 1.25-7.01) for patients with carcinoembryonic antigen 10-100 and >100 ng/ml, respectively, while comparing with carcinoembryonic antigen <10 ng/ml. The presence of peritoneal metastasis and grade-III tumour significantly worsened the survival in univariate analysis (hazard ratio 2.46, 95% confidence interval: 1.32-4.57 and hazard ratio 1.74, 95% confidence interval: 1.01-3.03, respectively) but not in multivariable analysis (adjusted hazard ratio 1.92, 95% confidence interval: 0.88-4.18 and adjusted hazard ratio 1.25, 95% confidence interval: 0.66-2.36, respectively). CONCLUSION: The study reported survival of stage IV colorectal cancer patients undergo chemotherapy and identified that Karnofsky performance status and carcinoembryonic antigen are the poor prognostic factors to this cohort adjusting for other factors.
