Clinical presentation, treatment patterns, and outcomes of colorectal cancer patients at Tikur Anbessa Specialized Hospital in Addis Ababa, Ethiopia: A prospective cohort study

Abstract Background Colorectal cancer (CRC) is the third most common cause of cancer death in both genders worldwide. Aims This study aimed to evaluate the outcomes and prognostic factors of CRC patients at Tikur Anbessa Specialized Hospital in Ethiopia. Methods and Results A prospective cohort study was conducted on 209 patients from January 2020 to September 2022. Kaplan–Meier curves and bivariate and multivariate Cox regression analyses were used to analyze overall and progression‐free survival, with a significance value of P < .05. Results showed an overall mortality rate was 67.46% (95% confidence interval [CI]: 61.0–74.0), while the 1‐year overall survival (OS) rate was 63.16% (95% CI: 56.23–69.29), with a median follow‐up duration of 20 months. The median OS and progression‐free survival times were 17 and 11 months, respectively. Age above 40 years (hazard ratio [HR] = 1.53, 1.02–2.29, p < .040), lower educational level (high school and below) (HR = 2.20, 1.24–3.90, p < .007), poor performance status (HR = 1.60, 1.03–2.48, p < .035), Hgb ≤12.5 g/dL (HR = 1.55, 1.03–2.08, p < .035), T‐4 disease (HR = 6.05, 2.28–16.02, p < .000), and metastases at diagnosis (HR = 8.53, 3.77–19.25, p < .000) were all associated with poorer survival. Conclusion These findings suggest that poor survival of CRC patients in Ethiopia is largely due to advanced stage of the disease and lack of timely treatment, and highlight the urgent need for improved access to cancer treatment in the region.

worldwide, more than HIV/AIDS, TB, and malaria combined. 3 In 2020, there were 10 million cancer deaths and 19.3 million new cancer cases. 4 The burden of cancer is expected to increase to 27.5 million new cases and 16.2 million cancer deaths by 2040, with 65% of cases occurring in low-and middle-income countries (LMIC). 5 In both genders combined, colorectal cancer (CRC) is the third most frequent cancer worldwide, with an estimated 1.5 million new cases and it has the second-highest mortality rate, 576 858 deaths in 2020. In males, it is the third most common in terms of incidence and mortality; however, in females, it is the third most common in terms of incidence and second in terms of mortality. 4 According to a major Austrian study, men are twice as likely as women to develop CRC. 6 LMICs account for 70% of all CRC-related deaths. 7 In Sub-Saharan Africa (SSA), the crude incidence rate is 4.04 per 100 000, with a male to female ratio of 1.2:1. 8 The incidence of CRC in Ethiopia is rising, with an estimated rate of 8.5 and 6.3 per 100 000 for men and women, respectively. 4 In Addis Ababa, CRC is the most common cancer among men, accounting for 12.4% of all cancers, and the fourth most common cancer among women, accounting for 5.4% of all cancers. 9 CRC also accounts for 12% of cancer cases at the oncology ward at Tikur Anbessa Specialized Hospital (TASH), and 7.7% of all cancer cases in Addis Ababa. 10,11 The incidence of CRC is increasing among young people of all races, particularly African Americans under the age of 50, who often present with more advanced tumors. 12 In the United States, CRC in the young population is typically more aggressive and advanced at diagnosis. The most notable increase in incidence between the ages of 40 and 44. 13 A study in Ghana found that the average age of people with the first CRC diagnosis was 54 ± 16.8 years, and the most frequently reported symptoms were weight loss (44.80%), rectal bleeding (39.82%), and abdominal pain (38.91%). 14 Additionally, a 5-year prospective study in Nigeria revealed that the majority of CRC patients at first diagnosis were between the ages of 51 and 60, and that 31% of patients were 40 years or younger. 15 In CRC, the primary treatment options are surgery, chemotherapy (CT), targeted agents, and radiotherapy (RT), but the choice largely depends on the site of the tumor, stage at presentation, individual patient factors, and increasingly, its molecular subtype. 16 Over time, systemic treatment for CRC has evolved from 5-fluorouracil (5 FU) to combination regimens involving 5-FU, oxaliplatin, irinotecan, or both, as well as the introduction of targeted agents for those with metastatic settings. 16,17 CRC is a heterogenous and complex disease that exhibits varying behaviors depending on its unique molecular pathophysiology and various molecular biomarkers can be prognostic or predictive. 18 Recent studies have focused on CRC molecular classification in order to develop molecularly defined subgroups for prognostic and treatment purposes.
The consensus molecular subtype (CMS) is a transcriptome-based classification of CRC that is often utilized and includes 4 subtypes: CMS1 for microsatellite instability, CMS2 for chromosomal instability, CMS3 for metabolic dysregulation, and CMS4 for mesenchymal activation. CMS4 has the worst prognosis, while CMS1 has a favorable prognosis. 19,20 Recent studies have also focused on molecular classification in order to develop molecularly defined subgroups for treatment purpose. 17 Anti-epidermal growth factor receptor (anti-EGFR) monoclonal antibodies, such as cetuximab and panitumumab, are used to treat metastatic cancer that expresses EGFR. Tumors with RAS and BRAF mutations, on the other hand, are not sensitive to these anti-EGFR agents. 21,22 Similarly, 5-FU-based traditional CT are ineffective against tumors with a high mutation burden and microsatellite instability; instead, immunotherapies such as pembrolizumab are more effective. 23,24 The use of RT also changed the course of CRC, particularly for rectal cancer. 25 Colorectal cancer mortality varies between countries based on human development index and racial characteristics, which is linked to the stage of disease at presentation, patient health-seeking behavior, and treatment accessibility. In high-income countries, mortality is decreasing, while in LMIC, it is increasing. SSA has the highest CRC mortality to incidence ratio in the world. 7 Studies have shown that recurrence-free survival and the stage of the disease are the most important predictors of CRC survival. In China, the three-and five-year survival rates were 74% and 68%, respectively, 26 while in Europe, the 1, 3, and 5-year survival rates were 90%, 70%, and 63%, respectively. 27 Despite advances in personalized cancer treatment in developed countries, cancer management in developing nations remains limited due to the lack of access to surgical care, availability of CT and RT, limited availability of healthcare professionals, and associated high costs. 17,28 This resulted in most patients in developing countries to continue having poor clinical outcomes and survival. 29 Only 27 of the 43 SSA nations have structured cancer registration systems; data quality varies, and national coverage is limited. 7 Significant numbers of CRC patients do not receive standard recommended therapies in most low-income countries, including Ethiopia. This is attributable to health-seeking behavior and, more crucially, the scarcity of integrated care. 30 In Ghana, a retrospective hospital-based study showed that the median OS of CRC patients over a five-year period was 15 months, with an OS rate of 16.0%. 29 In Nigeria, between 2013 and 2017; 71% of patients diagnosed with CRC underwent the necessary surgical procedure and 50.5% received the recommended CT. The median OS for those with stage III and IV CRC was 24 and 10.5 months, respectively. 31 However, in Ethiopia; the treatment outcomes, including, progression free survival (PFS), recurrence, and prognostic factors associated with OS of CRC patients were not studied. To address this gap in knowledge, we conducted a study at TASH in Addis Ababa, Ethiopia, to assess the treatment patterns and outcomes, as well as factors associated with OS of CRC patients. This is a pioneer prospective study of its kind that provides an understanding of the treatment patterns and outcomes of CRC patients, which will give insight into the gaps in the current treatment interventions and inform future strategies for quality comprehensive cancer care in the country. It will have important implications for the development of a multi-level assignment in the prevention, diagnosis, management, and follow-up of CRC patients and serve as a solid foundation for future investigators.

| Study setting and participants
This study was conducted in the adult oncology department of TASH, Addis Ababa University, which houses the only RT center in Ethiopia.
It is the only hospital that provides comprehensive cancer care, including surgery, CT, RT, and palliative care services.

| Study design
Institutional-based unmatched prospective cohort study was conducted to assess clinical presentation, treatment pattern, and outcome of CRC patients at TASH. Participants had clinical assessments, which included confirmatory diagnostic tests, appropriate primary and metastatic staging. Treatment details were documented, and patient outcomes were monitored. Patients' PFS and OS were determined through clinical follow-up and telephone interviews.

| Source population
All CRC patients seen at TASH oncology center during the study period.

| Study population
All CRC patients who started treatment at TASH oncology center from January 1, 2020 to September 10, 2022.  x-ray (CXR) for pulmonary metastatic evaluation; if CXR is suspicious that should be confirmed with chest computerized tomography scan.
All eligible patients starting from January 1, 2020, were prospectively included in the study until August 10, 2021. Patients with an ambiguous diagnosis, inadequate baseline CBC, LFT, RFT laboratory workup, inadequate diagnostic and metastatic staging workup, and those who did not consent to be part of the study were excluded from this study.

| Sample size
This study included all eligible patients with a diagnosis of CRC. A total of 415 CRC patients were observed at the oncology center's outpatient department, with 209 meeting the criteria for inclusion and forming the final sample size.

| Study variables
The study variables consisted of dependent variables, such as treatment outcome (alive, death, disease progression), and independent variables, such as sociodemographic factors (age, gender, marital status, educational status, habits, monthly income, comorbidities), clinical and pathologic characteristics, including, performance status, hemoglobin level, histology, stage at presentation, and types of treatments received (surgery, CT, and RT).

| Operational definitions
We operationalized the stages of CRC at presentation as stages I and

| Data collection tools and procedures
Three oncology resident physicians underwent two-day training for the purpose of data collection and extraction techniques. The principal investigator and two supervisors monitored the process, while customized patient-centered interview questionnaires and data extraction questionnaires were adopted and used to collect information from patients via phone or in-person interviews. Pretesting was done on 10 patients prior to the study and appropriate modifications were made based on the results. Each eligible patient's chart was given a unique code number before data collection. Information such as socio-demographic status, economic status, stage at presentation, and location of disease were collected. The patients were followed every 3 months for treatment details, PFS, and mortality rate.

| Data analysis
After checking completeness, data entered into Epidata version 4.60 and then transferred to Statistical Package for Social Sciences (SPSS) T A B L E 3 Cox regression hazard model for risk factor associated with overall survival among colorectal cancer patient attending Tikur Anbessa Specialized Hospital from January 1, 2020 to September 10, 2022, Ethiopia.

| Stage at presentation
More than three-fourths (79.4%) of the patients were found to have advanced disease (stages III/IV). Of the 104 patients who underwent surgery, 56 (53.8%) were not staged preoperatively (Figure 1).

| Treatment delivered
The treatment status of 209 patients was evaluated, of which

| Factors affecting overall survival
Age over 40 years increases the risk of death by 53% when compared to younger age groups (HR = 1.53, 95% CI, 1.02-2.29, P < .040). Poor performance status (ECOG-II & III) was associated with a 60% greater likelihood of dying than good functional status (HR = 1.60, 95% CI, 1.03-2.48, P < .035). Hemoglobin (Hgb) levels less than 12.5 g/dL were associated with a 1.5-fold higher risk of death (HR = 1.55, 1.03-2.08, P < .035) than those without anemia. Tumor depth of invasion was also an independent predictor of survival, with T-4 disease associated with a 6-fold increased risk of mortality compared to T1 and T2 disease (HR = 6.05, 95% CI, 2.28-16.02, P < .001). Metastases at diagnosis was associated with an eight-fold increased risk of death compared to early-stage cancer (stage I/II, HR = 8.53, 95% CI, 3.77-19.25, P < .001). Inability to deliver recommended surgery and CT were also poor prognostic factors (See Table 3, Figure 3A-F).

| DISCUSSION
The findings of this study are of great importance and novelty as they provide insight into the treatment type and outcomes of CRC patients in Ethiopia. The results highlight the need for improved access to cancer treatment in the region, given the poor survival rate due to advanced stages of the disease and a lack of timely treatment. Additionally, the study identified prognostic factors for survival, which are important for developing strategies to improve CRC outcomes in Ethiopia.
In this study, 67.46% of patients died in a median follow-up duration of 20 months, with a 1-year OS rate of 63.13% (95% CI, 56.23%-69.29%). This is similar to a retrospective study in Ghana, where the 1-year survival rate was 64%. 29 Comparatively to other regions, SSA has a low rate of CRC survival, where a meta-analysis found the 1-year survival is 74%, and even lower in lower income countries. 32 Our finding is far lower than many other studies performed in different parts of the world, such as a study from Brazil, where the 5-year OS was 63.5% and 60.6% for colonic and rectal malignancies, respectively, 33 and a Chinese study where the 3-and 5-year OS was 74% and 68%, respectively. 26 In a study in Vietnam, the 1, 3, and 5 OS survival showed 84.7%, 56%, and 45%, respectively, 34 which are greater than 1-year survival of our patients and this demonstrates the need of addressing the situation, which necessitates the involvement of all stakeholders.
The MS time observed in our patient is 17 months, and 11 months for stage IV disease, which is comparable to a previous study in Ghana, MS which is 15 months, where lack of treatment delivery was the main reason for poor survival. 29 However, for stage -III patients MS we found was 19 months, which is lower than a report from Nigeria, MS for stage-III is 24 months, due to curative treatment delivery being relatively better in such locally advanced cases. 31 In comparison to the sur- There are several factors that could affect patient survival seen in many studies. In our assessment, a significant number of patients, 67 (32.06%), were under the age of 40 at the time of diagnosis. We found that the probability of death increased with age greater than 40 years (HR = 1.53, 95% CI, 1.02-2.29, p < .040). This is consistent with a study from Canada, which found that age younger than 45 years was an independent predictor of better survival. 38 However, early-onset (age < 50) CRC has a poor prognosis due to its advanced stage at diagnosis and poorly differentiated histology. 39 Colonoscopy screening at age 45 and flexible sigmoidoscopy screening at age 40 is beneficial in identifying more patients with early-onset CRC in individuals who are at average risk. 40 The difference in prognosis between the two groups could be attributed to the aggressive nature of tumor biology in the younger age group, whereas multiple comorbidities and frailty in the older age group resulted in a barrier to initiating recommended cancer therapy.
Despite advances in cancer treatment, socioeconomic inequities continue to play a significant impact on all types of cancer survival.
Socioeconomically disadvantaged groups are at a higher risk of death. 41 In our investigation, the MS time for patients with a college or higher

| Strength of the study
The strength of this study lies in its prospective nature and the fact that it was conducted at the largest and only comprehensive oncology center in Ethiopia.

| Limitations of the study
The COVID-19 pandemic presented a limitation to the study as

ACKNOWLEDGMENTS
We wish to thank Miss. Muluwork, Mr. Nemie, and Mrs. Ayalnesh, who showed profound support during this research.

FUNDING INFORMATION
The Addis Ababa University School of Public Health received funding for this work through an American Cancer Society-supported colorectal cancer thematic research project award. However, they are not involved in the methodology, data collecting, analysis, or manuscript writing.