Background Globally, the incidence of prostate cancer is increasing, particularly in low-and middleincome countries. It is the most common cancer among men worldwide, with higher mortality in low and middle-income countries. In Ethiopia, it is the second most common cause of cancer morbidity and mortality among men. Despite a few studies done regarding the disease burden, the evidence is scarce about the survival and prognostic determinants of prostate cancer patients in Ethiopia. Thus, this study assessed the survival and prognostic determinants of patients with prostate cancer. Methods We retrospectively followed patients who were newly diagnosed from 2012 to 2016 at the Oncology Department of Tikur Anbessa Specialized Hospital. We extracted the data from patient charts that were available in the cancer registry using a checklist with the help of oncology nurses. Kaplan-Meier survival analyses with the log-rank test were used to estimate and compare the probability of survival among covariate categories. After checking for assumptions, a multivariable Cox regression analysis was performed to identify prognostic determinants of survival. Results The median survival time was 28 months with an overall 2-, 3-and 5-year survival of 57%, 38.9% and 22%, respectively. The overall survival differs according to the clinical stage (P-value<0.01), presence or absence of distant metastasis (P<0.01) and androgen deprivation therapy (ADT) (P<0.05). Cancer stage at diagnosis (adjusted hazard ratio (AHR) = 0.309,

This article describes data collected retrospectively on a cohort of esophageal, colorectal and prostate cancer patients registered in the patient log book of Tikur Anbessa Specialized Hospital, Ethiopia, from January 1, 2012 to December 31, 2017. The key variables studied include histological characteristics of each type of cancer, clinical and TNM stages, baseline laboratory results (Carcinoembryonic antigen (CEA) for colorectal cancer, Prostate-Specific Antigen (PSA) for prostate cancer, hemoglobin level, etc.), clinical characteristics including sign and symptoms, family history of cancer, diagnostic and treatment modalities a patient received for each type of cancer. The event status (death) was also collected using death certificates (whenever available) and supplemented by telephone interviews with the patient or attendant. Furthermore, lifestyle characteristics of patients including tobacco use, alcohol consumption, khat (‘ Catha edulis ’) chewing, etc. and socioeconomic characteristics including age, sex, region of residence, marital status, and educational level were also collected. The aim that led to conduct the study that generated these data was to describe clinical presentation, histological characteristics, survival pattern, and to identify determinants of mortality among cancer patients in Ethiopia. Thus, independent survival analyzes were performed using Kaplan-Meier estimates and life table analysis. Furthermore, Cox's proportional hazards regression was developed to investigate the survival pattern and determinants of cancer specific mortality among colorectal, esophageal and prostate cancer patients.

No abstract