Histological characteristics, survival pattern and prognostic determinants among colorectal cancer patients in Ethiopia: A retrospective cohort study

Abstract

Background: The incidence of colorectal cancer (CRC) and associated mortality are rising in low- and middle-income countries. In Ethiopia, colorectal cancer is among the leading causes of cancer morbidity and mortality in both sexes. Although some studies provided estimations on the national burden and regional distribution, the histological characteristics, survival pattern and determinants among colorectal cancer patients are not well-documented.

Aim: This study aimed to describe the histological characteristics, to determine the patterns of survival, and identify factors that determine mortality rate among CRC patients in Ethiopia.

Methods: A retrospective cohort study was conducted among CRC patients registered at cancer treatment center of Tikur Anbessa Specialized Hospital, from January 2012 to December 2016. Data were extracted from a total of 161 patient medical records using a pretested abstraction form and supplemented by phone calls with the patients/caregivers. To determine colorectal cancer specific survival overtime, we performed a Kaplan-Meier survival analysis and significance of variation in survival across covariates and treatment categories was tested using log-rank test. A multivariable Cox proportional-hazards model was performed to identify determinants of survival after diagnosis with colorectal cancer.

Results: Overall, the median survival time was 21 months [95%CI: 16-35], with two-, three- and five-year CRC-specific survival rates of 46.8%, 39.5% and 28.7% respectively. In the multivariable Cox regression model, the rate of death due to CRC is significantly higher for patients with elevated baseline carcinoembryonic antigen (CEA) level (Adjusted Hazard Ratio (AHR) = 2.31, 95%CI: 1.27-4.19), stage IV at diagnosis (AHR = 2.66, 95%CI: 1.44-4.91), and mucinous or signet-ring cell carcinoma histology type (AHR = 4.92, 95%CI: 1.75-13.80). Moreover, patients who underwent surgery showed a better survival than those who did not (AHR = 0.35, 95%CI: 0.14-0.88).

Conclusion: In Ethiopia, patients diagnosed with CRC showed a low rate of cancer-specific survival. Histology type, stage of cancer and CEA level at diagnosis, and the type of treatment a patient received significantly determine mortality rate. Hence, cancer screening programs could help to detect the disease at an earlier stage and to initiate available treatments timely so as to extend the lifespan of CRC patients.

Keywords: Cohort; Colorectal cancer; Determinants; Ethiopia; Histological characteristics; Survival.

Figure 1

Kaplan-Meier survival curve of the overall survival for colorectal cancer patients in Tikur Anbessa Specialized Hospital, Addis Ababa, 2012 to 2016. The curve shows the median survival is 21 months [95%CI: 16–35].

Figure 2

Kaplan-Meier survival curve of the overall survival by TNM stage of cancer at diagnosis among colorectal cancer patients in Tikur Anbessa Specialized Hospital, Addis Ababa, 2012 to 2016. The log-rank test showed that the difference in mortality rate according to TNM stage is statistically significant (p < 0.001).

Figure 3

Kaplan-Meier survival curve of the overall survival by clinical stage of cancer at diagnosis among colorectal cancer patients in Tikur Anbessa Specialized Hospital, Addis Ababa, 2012 to 2016. The log-rank test showed the difference in mortality according to clinical stage at diagnosis is statistically significant (p < 0.001).

Figure 4

Kaplan-Meier survival of the difference in overall survival by baseline CEA level among colorectal cancer patients in Tikur Anbessa Specialized Hospital, Addis Ababa, 2012 to 2016. The log-rank test indicated a significant difference in survival according to baseline CEA level (p < 0.001). CEA: carcinoembryonic antigen.

Figure 5

Kaplan-Meier survival curve of the overall survival based on surgery among colorectal cancer patients in Tikur Anbessa Specialized Hospital, Addis Ababa, 2012 to 2016. The log-rank test indicated those who underwent surgery showed a better overall survival than those who did not (p = 0.025).