Background
Colorectal cancer (CRC) is the third most commonly diagnosed cancer and the second leading cause of cancer death worldwide, with an estimated 1.9 million cancer cases and 935,000 deaths in 2020.1 Previous studies reported a strong association between the occurrence of CRC and a higher level of socioeconomic development due to the high prevalence of obesity, smoking, lack of physical activity, and unhealthy diet in economically developed countries.2–5 As a result, CRC incidence rates are increasing in countries with emerging economies, including many countries in Sub-Saharan Africa (SSA), because people in these countries are increasingly adopting unhealthy behaviors in part due to marketing by the food and beverage industries.6–11 Among the Black population of Zimbabwe, for example, CRC incidence rates increased by 4% per year between 1991 and 2010.7 The magnitude of the increase in CRC incidence in Zimbabwe and other SSA countries is likely to be underestimated because of underdiagnosis and underreporting of new cancer cases in the region.12
However, there are limited data on receipt of guideline-concordant treatment and survival for patients diagnosed with CRC in SSA. The few existing studies are based on the experience of individual clinicians, small case series, and hospital-based registries,13–18 or studies in the private healthcare sector.
None of the previous studies evaluated treatment concordance with NCCN Harmonized Guidelines for SSA, which were created in early 2018 to address the varying resources in the region. In brief, the NCCN Harmonized Guidelines for SSA recommend transabdominal resection or transanal local excision for stage I rectal cancer and neoadjuvant chemotherapy with radiotherapy for stage II and III rectal cancer.19 Patients with colon cancer appropriate for resection should undergo colectomy with en bloc removal of at least 12 lymph nodes, whereas neoadjuvant chemotherapy is recommended for T4b or locally unresectable colon cancer.20 All patients with stage IV CRC should receive some form of (radio-)chemotherapy, with surgical resection depending on tumor characteristics and patient condition. Radiotherapy in a single fraction is an option for surgically unresectable stage IV disease.
In this study, we examined concordance of treatment for CRC with NCCN Harmonized Guidelines based on patients diagnosed from 11 population-based cancer registries in SSA, providing context-specific data to support the guidelines and establish a “preguideline benchmark.”
Methods
Study Design and Patients
This is a multinational retrospective study that includes data from 11 regional population-based cancer registries in SSA: Abidjan (Ivory Coast), Addis Ababa (Ethiopia), Bamako (Mali), Brazzaville (Congo), Bulawayo (Zimbabwe), Cotonou (Benin), Eldoret (Kenya), Kampala (Uganda), Maputo (Mozambique), Nairobi (Kenya), and Namibia.
All of these centers are members of the African Cancer Registry Network (AFCRN; https://afcrn.org/), a regional hub for the International Agency for Research on Cancer’s Global Initiative for Cancer Registry Development in Low- and Middle-Income Countries. Supplemental eTable 1 (available with this article at JNCCN.org) shows populations covered by each registry, selected numbers of patients for the study, and diagnosis years.
At the time of study planning, the AFCRN encompassed 26 population-based cancer registries in SSA; of these, 11 registries consented to participate in the study. From each of the 11 participating registries, we randomly selected between 60 and 100 patients with CRC with the ICD-O-3 topography codes C18, C19, and C20 and date of diagnosis between 2011 and 2015, with the exact year varying by registry. All patients diagnosed in Cotonou, Brazzaville, and Maputo were included in the study because of smaller population coverage or fewer diagnosis years.
The registry database provided demographic and clinical data according to the AFCRN Standard Procedure Manual and ICD-O-3,21,22 whereas additional information such as diagnostic procedures, treatment, and vital status was obtained from medical records, phone calls, and home visits for select patients. Histopathologic, surgical, endoscopic, and clinical information was used to assign the primary site of the tumor as colon or rectum.
To make the report more comprehensive, we created 3 overlapping study subpopulations, as shown in supplemental eFigure 1. First, the “entire cohort” encompassed all 653 patients initially selected, representing a population-based setting for the registry areas after excluding patients who were diagnosed before 2011, aged <15 years, double-counted, or false-positive (misdiagnosis). We used this cohort for descriptive cross-sectional analyses. Second, the “traced therapy cohort” encompassed the 356 patients for whom additional information on treatment, stage at diagnosis, and follow-up for vital status was obtained from medical records in treating facilities. This cohort was used to describe stage-specific guideline concordance of cancer-directed therapy. Third, the “traced survival cohort” encompassed the 308 traced patients with a follow-up of at least 30 days to minimize survivorship bias in the study population.23,24 This cohort was used to determine prognostic factors for overall survival (OS) based on a Cox multiple regression model.
Study Variables
The primary outcome was guideline concordance of care, and the secondary outcome was survival at the time of data collection.
Stage
Based on clinical and pathologic information in hospital records, stages were classified as I, II, III, or IV according to the TNM staging system of the combined AJCC/Union for International Cancer Control. If detailed information in medical records was lacking to classify tumors according to TNM, additional information from imaging, surgical reports, or clinical notes was used to assign stages.25 The adequacy of surgical resection was determined based on the most invasive treatment received; for example, colectomy in a patient who underwent polypectomy followed by colectomy.
Stage II was separated into low risk and high risk according to current international guidelines. Cases whose records did not contain information on lymph node status were categorized as Nx (lymph node status unknown); analogously, cases with missing tumor size information were categorized as Tx. In contrast, for patients with missing information on the tumor’s distant spread, we assumed there were no metastases (thus M0), as recommended by the 8th revision of the TNM classification.26
Guideline Concordance
We developed a stage-dependent simplified therapy evaluation scheme based on the NCCN Harmonized Guidelines for SSA for Colon and Rectal Cancers,19,20 shown in supplemental eTable 2. For patients diagnosed with nonmetastasized disease (stage I–III), receipt of treatment was categorized as optimal treatment, standard of care, and treatments with minor and major deviations, backed by empirical evidence (supplemental eTable 3).
Resources Available
As a rough surrogate for resource availability in the countries included in the study, we used the Human Development Index (HDI) with the predefined categories of low HDI (<0.5) and medium HDI (0.5 to <0.8) from 2014.27
Statistical Analyses
We used SPSS Statistics, version 25 (IBM Corp). OS was estimated using the Kaplan-Meier method, and differences in OS according to prognostic factors were assessed using the log-rank test. Hazard ratios (HRs) were estimated using a Cox regression model. We assessed reversed Kaplan-Meier curves stratified by factors included in the Cox regression model for censoring at random.
Ethics
The study protocol was approved by the AFCRN Review Committee (7.12.2017) and by the Martin-Luther-University Halle-Wittenberg Review Board (votum no. 2019-009). The study was based on deidentified secondary data, which were collected under existing regulations and national laws in the respective registries.
Results
Patient and Tumor Characteristics
Table 1 shows baseline and tumor characteristics of the entire cohort according to metastasis (M1) at the time of diagnosis. Median age at diagnosis was 54 years, with 1 in 5 patients aged <40 years. Among the cohort, 56.2% of patients were registered as having colon cancer, and 43.8% were registered as having rectal cancer. Of those patients with known stage (n=303), almost one-third (31.0%; n=94) presented with metastasis (stage IV), and 54.5% were staged by imagery, mainly by CT scan and ultrasound.
Baseline Patient Cohort Characteristics
Table 2 shows treatment characteristics by stage at diagnosis for traced patients (n=356). Of the 262 patients who were traced and without metastasis (stage I–III, M0) at time of diagnosis and considered potentially curable, 71.8% received some form of surgery, 54.2% received chemotherapy, and 17.2% received radiotherapy. Among the patients who received surgery, lymph node examination was documented for one-third of them. Overall, only 3.1% (n=8) of M0 patients received guideline-concordant therapy according to the simplified treatment evaluation scheme (supplemental eTable 2). Approximately 20.6% (n=54) of the patients received guideline-concordant therapy with minor deviations and 31.7% (n=83) received guideline-concordant therapy with major deviations. Notably, no cancer-directed therapy was documented for 35.1% of the M0 patients, and it was assumed they did not receive any form of cancer treatment.
Treatment Characteristics of Patients With Nonmetastatic (M0) and Metastatic (M1) Disease
Evaluation of Cancer-Directed Therapy by Registry
Figure 1A portrays treatment patterns by registry for all 653 patients included in the study. The proportion of M0 patients who received adequate treatment, including treatment with minor deviations, ranged from 2% in Kampala (Uganda) to 29.7% in Namibia. Figure 1B shows treatment evaluation of all 653 patients from a cross-sectional approach. It should be noted again that only 1% of the total study population received guideline-concordant treatment. Furthermore, 10% of patients did not receive any form of cancer-directed therapy.
Evaluation of concordance with NCCN Harmonized Guidelines (A) separately per center and (B) overall (N=653).
Abbreviation: HDI, human development index.
Citation: Journal of the National Comprehensive Cancer Network 21, 9; 10.6004/jnccn.2023.7041
Evaluation of concordance with NCCN Harmonized Guidelines (A) separately per center and (B) overall (N=653).
Abbreviation: HDI, human development index.
Citation: Journal of the National Comprehensive Cancer Network 21, 9; 10.6004/jnccn.2023.7041
Evaluation of concordance with NCCN Harmonized Guidelines (A) separately per center and (B) overall (N=653).
Abbreviation: HDI, human development index.
Citation: Journal of the National Comprehensive Cancer Network 21, 9; 10.6004/jnccn.2023.7041
Survival
OS in the traced cohort (n=356) at 1, 2, and 3 years was 70.9% (95% CI, 65.5%–76.3%), 55.2% (95% CI, 49%–61.4%), and 45.3% (95% CI, 38.9%–51.7%), respectively (supplemental eFigure 2). OS statistically significantly differed by stage at time of diagnosis (log rank P≤.001).
Two-year OS was 72.6% (95% CI, 58.4%–86.8%) for patients with stage I–II low-risk cancer, 57.9% (95% CI, 45.3%–70.5%) for those with stage II high-risk cancer, 62.2% (95% CI, 49.6%–74.8%) for those with stage III cancer, and 39% (95% CI, 26.6%–51.4%) for those with stage IV cancer (supplemental eFigure 3). In patients without a documented stage, the 2-year OS was 50.5% (95% CI, 33.7%–67.3%). OS also significantly differed by initiation of therapy, as depicted in Figure 2.
Kaplan-Meier curve showing overall survival, with 95% confidence intervals, based on concordance with NCCN Harmonized Guidelines.
Citation: Journal of the National Comprehensive Cancer Network 21, 9; 10.6004/jnccn.2023.7041
Kaplan-Meier curve showing overall survival, with 95% confidence intervals, based on concordance with NCCN Harmonized Guidelines.
Citation: Journal of the National Comprehensive Cancer Network 21, 9; 10.6004/jnccn.2023.7041
Kaplan-Meier curve showing overall survival, with 95% confidence intervals, based on concordance with NCCN Harmonized Guidelines.
Citation: Journal of the National Comprehensive Cancer Network 21, 9; 10.6004/jnccn.2023.7041
Table 3 shows the results of a multivariable Cox regression model for all M0 patients with follow-up of at least 30 days (n=226) after adjusting for age and sex. The overall risk of mortality (HR) was significantly greater in patients diagnosed with advanced-stage disease than in those with early stage-disease; patients with higher ECOG performance status versus lower performance status; patients treated with major deviations from NCCN Harmonized Guidelines or no cancer-directed therapy versus in those treated according to NCCN Harmonized Guidelines or with minor deviations from the guidelines; and patients residing in countries with low HDI versus medium HDI. For example, compared with patients who received standard treatment or standard treatment with minor deviations, the risk of mortality in patients with major treatment deviations and those with no cancer-directed therapy was 1.85 (95% CI, 0.98–3.49) and 3.49 (95% CI, 1.83–6.66) higher, respectively. Likewise, the risk of mortality was 1.67 (95%, 1.07–2.62) higher in patients residing in countries with low HDI versus medium HDI.
Multivariable Cox Regression Analysis in the Traced Cohort With Stage M0 CRC With Follow-Up ≥30 Days
Supplemental eTable 4 presents multivariable Cox regression results for all patients with metastasized disease and follow-up of at least 30 days for vital status (n=81) after accounting for age and sex. Compared with patients who received surgery, the risk of mortality was significantly higher (HR, 2.1; 95% CI, 1.1–4.0) in patients who did not receive surgery or whose receipt of surgery was unknown.
Discussion
Based on 11 population-based cancer registries in SSA, we examined concordance of treatment for CRC with NCCN Harmonized Guidelines and survival in the region and found that (1) OS of patients was poor, (2) fewer than 1 in 20 patients with nonmetastatic disease received treatment that was concordant with NCCN Harmonized Guidelines, and (3) disparity in OS among patients with CRC in countries with low HDI versus medium HDI was quite large.
Our findings of the low OS rate were generally similar to previous findings from mostly hospital-based studies. Agyemang-Yeboah et al13 examined a total of 221 patients recorded from 2009 to 2015 in a teaching hospital in Ghana and found that survival rates at 2 and 5 years were 40% and 16%, respectively. In a study from a referral hospital in Tanzania of 332 patients with CRC, only 9 (3%) were available for follow-up 5 years after diagnosis. Similar to our study, most patients (96.7%) presented late with advanced stages of disease, and a quarter of them were diagnosed with distant metastases.28 Data on cancer survival from SSA are generally scarce. A recent study based on 1,448 patients with CRC from 11 countries in SSA, with some overlap of cases in the present study, found an OS of 72.0% (95% CI, 69.5%–74.4%) at 1 year, 50.4% (95% CI, 47.6%–53.2%) at 3 years, and 43.5% (95% CI, 40.6%–46.3%) at 5 years.29
In contrast to SSA, the contemporary survival rate for CRC is substantially higher in Western countries largely because of broad dissemination of early detection methods and guideline-concordant treatments.30 For example, 1- and 3-year relative survival rates for invasive CRC in the United States are 84.3% (95% CI, 83.8%–84.7%) and 71.8% (95% CI, 71.2%–72.4%), respectively.31
The low survival rate in the present study (in SSA) largely reflects lack of CRC awareness in the general population and among health workers, the absence of opportunistic or organized screening programs,32,33 lack of neoadjuvant and adjuvant chemotherapies, poor quality of surgical treatment,34 delay in receiving care,34 and the high cost of diagnostics and treatment that most patients cannot afford and the subsequent abandonment of treatment in the region.35 Only 3% of patients with stage I–III disease (n=262) received treatment concordant with NCCN Harmonized Guidelines, and 20% received guideline-concordant treatment with minor deviations.
Possible reasons for the low receipt of guideline-concordant treatment remain a matter of speculation and probably include lack of awareness about the benefit of modern medicine; stigma attached to oncologic diseases, especially with colostomy bags or improvised devices playing an important role; reliance on alternative treatment and concern about side effects of drugs28,36; and a less well-developed health care infrastructure (eg, shortage of oncologists, surgeons, and cancer drugs); and the high cost of cancer care. The Lancet Commission on Global Surgery estimates that >95% of the population in central, eastern, and western SSA do not have access to safe, affordable, and timely surgical care.37 Only 19% of all surgeons are located in low-income countries and low-middle-income countries, although these regions account for 48% of the world population.38
Only 17% of all patients received documented cross-sectional imaging by CT or MRI, with radiography or ultrasound often used instead, which may indicate significant understaging. Few patients received radiotherapy in our study, despite its crucial role in treating rectal cancer. Only 63 patients received radiotherapy, which accounted for 17% of all patients with nonmetastatic disease, consistent with the limited availability of radiotherapy equipment on the African continent. In fact, only 5 of the 10 countries studied had any oncology center with radiation capability in 2014. Africa has the least developed radiotherapy services globally, with <1 teletherapy machine per million people on average. Unfortunately, the situation has improved only slightly over the past decade.39,40
As the NCCN Harmonized Guidelines for SSA for CRC were published in 2018, the diagnosis and treatment of patients in our study (2011–2015) occurred when there were no established standards of care for CRC specific to the unique circumstances of SSA. Therefore, our data represent a preguideline benchmark and can be considered as a starting point for further studies examining guideline concordance to demonstrate potential improvement in cancer care since the introduction of the NCCN Harmonized Guidelines for SSA.
Guideline-concordant care with major deviations had a significant impact on survival, although this deviation was based on a lower number of lymph nodes resected for surgical quality (an established prognostic factor),41,42 underlining the major role of radical surgery in resource-limited settings in curative CRC care. Much can be done, even without the newest technologies and most expensive treatment strategies, but access to even basic care is not established in many parts of SSA.43 Our study highlights the importance of diagnosing cancer early. Treatment in earlier stages is not only more effective in terms of morbidity and mortality but also has been shown to be much more cost-effective than treatment in advanced stages.44
In the multivariable Cox regression analysis for the traced cohort, we found that patients from countries with a low HDI had a significantly higher risk of death than patients from countries with a medium HDI. Outcomes depend on available resources that on a national level are often estimated using the HDI. The HDI can function as a rough indicator of critical inequalities of global cancer burden, mortality, and outcomes.32,43,45 In 2015, by adopting sustainable development goals, the United Nations General Assembly recognized low-income countries as being particularly vulnerable to the growing global cancer burden. Our findings underline the urgent need to allocate resources to cancer care.
We acknowledge several limitations inherent in the retrospective design of the study. First, it is likely that treatment effects were overestimated due to selection bias in our study. Assignment of treatment was obviously not randomized (and manifestly impossible in a comparison of adequate and substandard care). Patients who died very shortly after the date of diagnosis did not have a chance to receive any treatment at all. Furthermore, observational time-to-event analyses are prone to immortal time bias (also known as survivor[ship] bias), given that patients who had received treatment had to be alive by that time, or rather had to survive long enough after surgical resection to receive chemotherapy.46 To reduce the overestimation of treatment effects and immortal time bias, we included only patients who survived at least 30 days after the time of diagnosis for the Cox regression analyses. It should be noted that censoring was not associated with age, sex, or stage at diagnosis by the reverse Kaplan-Meier method, and is therefore considered as random for the Cox regression analyses.
Another limitation is that we do not know whether the inadequate number of lymph nodes examined was due to inadequate surgical resection or to insufficient pathologic evaluation. Thus, we risk attributing deviations from guidelines solely to surgery when pathology deficiencies may also contribute.
Data were missing for a substantial proportion of the patients (46% of all patients without additional information). Explanations are manifold and cover inadequate archiving systems (missing at random), treatment documents located abroad (for wealthy patients), or lack of treatment initiation (due to lack of resources). It remains speculative that the patients who were not traceable (despite thorough investigation) may have received inadequate care or even none at all; or, conversely, they were lost to follow-up due to referral to cancer treatment centers abroad. It is reasonable to assume that the opportunity for treatment abroad is reserved for only the most affluent patients due to high costs. We consider the large amount of censoring and the limited diagnostic and therapeutic data not only as limiting factors of our study but also as findings that disclose the concerning situation of CRC care in SSA.
Conclusions
To our knowledge, this is the first study to examine receipt of treatment and survival among patients with CRC in a population-based setting and from multiple centers in SSA. It is also the first study to examine concordance of CRC care with the NCCN Harmonized Guidelines for SSA. Our finding of discrepancies between the recommendations of the NCCN Harmonized Guidelines and actual receipt of therapy, especially for patients with potentially curable diseases, is alarmingly large. The proportion of patients with documented initiation of adequate therapy was <5% in some centers and did not exceed 25% in any of the centers. At the same time, countries with a low HDI showed worse survival rates than countries with a medium HDI (despite an adjustment for factors such as age, disease stage, and ECOG performance status). These data reinforce the importance of concrete efforts to improve the health care infrastructure, from diagnostic services to workforce numbers and drug availability in the region, so that quality of care for patients with CRC and other cancers can be improved.
However, despite the high proportion of missing data, we were able to show that the NCCN Harmonized Guidelines are an effective and meaningful basis for treatment decisions for patients with CRC, even under precarious conditions (as in SSA). The effect is clear: patients who received treatment concordant with NCCN Harmonized Guidelines survived longer. Records of roughly 45% of the patients in the population-based registries could not be traced to a treatment facility. It can be assumed that most of these patients never received adequate and guideline-concordant treatment; therefore, no clinical records were initiated. Our study underlines the major challenges regarding diagnosis and therapy in patients with CRC in the region. For policymakers and institutions in SSA, our study results can be an important basis for targeted and meaningful investments and measures to improve the outcomes and survival of patients with CRC in the region.
Despite intensive efforts and large-scale cancer control programs in SSA, progress in expanding diagnosis and treatment facilities and the oncology workforce after 2015 has been limited. Lack of resources, including in diagnostics and medical oncology, as well as the “brain drain” of oncology specialists due to high clinical workload and job dissatisfaction, continue to pose significant challenges.47 With increasing cancer incidence in low- and middle-income countries, there is an urgent need for significant efforts to improve cancer therapy.
References
- 1.↑
Sung H, Ferlay J, Siegel RL, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2021;71:209–249.
- 2.↑
Bingham SA, Day NE, Luben R, et al. Dietary fibre in food and protection against colorectal cancer in the European Prospective Investigation into Cancer and Nutrition (EPIC): an observational study. Lancet 2003;361:1496–1501.
- 3.↑
Friedenreich C, Norat T, Steindorf K, et al. Physical activity and risk of colon and rectal cancers: the European prospective investigation into cancer and nutrition. Cancer Epidemiol Biomarkers Prev 2006;15:2398–2407.
- 4.↑
Pischon T, Lahmann PH, Boeing H, et al. Body size and risk of colon and rectal cancer in the European Prospective Investigation into Cancer and Nutrition (EPIC). J Natl Cancer Inst 2006;98:920–931.
- 5.↑
Robsahm TE, Aagnes B, Hjartåker A, et al. Body mass index, physical activity, and colorectal cancer by anatomical subsites: a systematic review and meta-analysis of cohort studies. Eur J Cancer Prev 2013;22:492–505.
- 6.↑
Arnold M, Sierra MS, Laversanne M, et al. Global patterns and trends in colorectal cancer incidence and mortality. Gut 2017;66:683–691.
- 7.↑
Chokunonga E, Borok MZ, Chirenje ZM, et al. Trends in the incidence of cancer in the Black population of Harare, Zimbabwe 1991–2010. Int J Cancer 2013;133:721–729.
- 9.↑
Lorenzoni C, Vilajeliu A, Carrilho C, et al. Trends in cancer incidence in Maputo, Mozambique, 1991–2008. PloS One 2015;10:e0130469.
- 10.↑
Somdyala NI, Parkin DM, Sithole N, et al. Trends in cancer incidence in rural Eastern Cape Province; South Africa, 1998–2012. Int J Cancer 2015;136:E470–474.
- 11.↑
Zingoni C, Norris SA, Griffiths PL, et al. Studying a population undergoing nutrition transition: a practical case study of dietary assessment in urban South African adolescents. Ecol Food Nutr 2009;48:178–198.
- 12.↑
Katsidzira L, Gangaidzo I, Thomson S, et al. The shifting epidemiology of colorectal cancer in sub-Saharan Africa. Lancet Gastroenterol Hepatol 2017;2:377–383.
- 13.↑
Agyemang-Yeboah F, Yorke J, Obirikorang C, et al. Patterns and presentations of colorectal cancer at Komfo-Anokye teaching hospital Kumasi, Ghana. Pan Afr Med J 2017;28:121.
- 14.↑
Lambert R, Saito H, Lucas E, et al. Survival from digestive cancer in emerging countries in Asia and Africa. Eur J Gastroenterol Hepatol 2012;24:605–612.
- 15.↑
Rahman GA. Rectal cancer: pattern and outcome of management in University of Ilorin Teaching Hospital, Ilorin, Nigeria. Ann Afr Med 2010;9:164–169.
- 16.↑
Selemane C, Jamisse L, Arroz J, et al. Demographic, clinical and pathological characterization of patients with colorectal and anal cancer followed between 2013 and 2016 at Maputo Central Hospital, Mozambique. Ecancermedicalscience 2021;15:1205.
- 17.↑
Sharma A, Alatise OI, Adisa AO, et al. Treatment of colorectal cancer in Sub-Saharan Africa: results from a prospective Nigerian hospital registry. J Surg Oncol 2020;121:342–349.
- 18.↑
Wismayer R, Kiwanuka J, Wabinga H, et al. Prognostic factors for survival of colorectal adenocarcinoma patients in Uganda. Cancer Manag Res 2022;14:875–893.
- 19.↑
Benson AB III, Venook AP, Al-Hawary MM, et al. NCCN Harmonized Guidelines for Sub-Saharan Africa: Rectal Cancer. Version 2.2018. To view the most recent version, visit https://www.nccn.org
- 20.↑
Benson AB III, Venook AP, Al-Hawary MM, et al. NCCN Harmonized Guidelines for Sub-Saharan Africa: Colon Cancer. Version 2.2018. To view the most recent version, visit https://www.nccn.org
- 21.↑
African Cancer Registry Network. Standard procedure manual. Accessed July 1, 2023. Available at: https://afcrn.org/index.php/resources2/53-standard-procedure-manual/131-sop
- 22.↑
World Health Organization. International Classification of Diseases for Oncology (ICD-O-3). 1st rev, 3rd ed. IARC; 2014.
- 23.↑
Geneletti S, Richardson S, Best N. Adjusting for selection bias in retrospective, case-control studies. Biostatistics 2009;10:17–31.
- 24.↑
Miller DP, Gomberg-Maitland M, Humbert M. Survivor bias and risk assessment. Eur Respir J 2012;40:530–532.
- 25.↑
Piñeros M, Parkin DM, Ward K, et al. Essential TNM: a registry tool to reduce gaps in cancer staging information. Lancet Oncol 2019;20:e103–111.
- 26.↑
Brierley JD, Gospodarowicz MK, Wittekind C, eds. TNM Classification of Malignant Tumours, 8th ed. Wiley Blackwell; 2016.
- 27.↑
United Nations Development Programme. Human development report 2015: work for human development. Accessed July 1, 2023. Available at: http://hdr.undp.org/en/content/human-development-report-2015
- 28.↑
Chalya PL, McHembe MD, Mabula JB, et al. Clinicopathological patterns and challenges of management of colorectal cancer in a resource-limited setting: a Tanzanian experience. World J Surg Oncol 2013;11:88.
- 29.↑
Gullickson C, Goodman M, Joko-Fru YW, et al. Colorectal cancer survival in sub-Saharan Africa by age, stage at diagnosis and Human Development Index: a population-based registry study. Int J Cancer 2021;149:1553–1563.
- 30.↑
Jiang Y, Yuan H, Li Z, et al. Global pattern and trends of colorectal cancer survival: a systematic review of population-based registration data. Cancer Biol Med 2021;19:175–186.
- 31.↑
Surveillance, Epidemiology, and End Results Program. SEER*Stat database: colon and rectum: recent trends in SEER relative survival rates, 2000–2020. Accessed July 1, 2023. Available at: https://seer.cancer.gov/statistics-network/explorer/application.html?site=20
- 33.↑
Irabor DO. Emergence of colorectal cancer in West Africa: accepting the inevitable. Niger Med J 2017;58:87–91.
- 34.↑
Chawla S, Kurani S, Wren SM, et al. Electricity and generator availability in LMIC hospitals: improving access to safe surgery. J Surg Res 2018;223:136–141.
- 35.↑
Sankaranarayanan R, Swaminathan R, Jayant K, et al. An overview of cancer survival in Africa, Asia, the Caribbean and Central America: the case for investment in cancer health services. IARC Sci Publ 2011;162:257–291.
- 36.↑
Yawe KT, Bakari AA, Pindiga UH, et al. Clinicopathological pattern and challenges in the management of colorectal cancer in sub-Saharan Africa. J Chin Clin Med 2007;2:688–695.
- 37.↑
Alkire BC, Raykar NP, Shrime MG, et al. Global access to surgical care: a modelling study. Lancet Glob Health 2015;3:e316–323.
- 38.↑
Holmer H, Lantz A, Kunjumen T, et al. Global distribution of surgeons, anaesthesiologists, and obstetricians. Lancet Glob Health 2015;3:S9–11.
- 39.↑
Abdel-Wahab M, Bourque JM, Pynda Y, et al. Status of radiotherapy resources in Africa: an International Atomic Energy Agency analysis. Lancet Oncol 2013;14:e168–175.
- 40.↑
International Atomic Energy Agency. What is DIrectory of RAdiotherapy Centres (DIRAC)? Accessed April 28, 2023. Available at: https://dirac.iaea.org
- 41.↑
Chang GJ, Rodriguez-Bigas MA, Skibber JM, et al. Lymph node evaluation and survival after curative resection of colon cancer: systematic review. J Natl Cancer Inst 2007;99:433–441.
- 42.↑
Le Voyer TE, Sigurdson ER, Hanlon AL, et al. Colon cancer survival is associated with increasing number of lymph nodes analyzed: a secondary survey of intergroup trial INT-0089. J Clin Oncol 2003;21:2912–2919.
- 43.↑
Farmer P, Frenk J, Knaul FM, et al. Expansion of cancer care and control in countries of low and middle income: a call to action. Lancet 2010;376:1186–1193.
- 44.↑
Ralaidovy AH, Gopalappa C, Ilbawi A, et al. Cost-effective interventions for breast cancer, cervical cancer, and colorectal cancer: new results from WHO-CHOICE. Cost Eff Resource Alloc 2018;16:38.
- 45.↑
Pervaiz R, Faisal F. Cancer incidence and mortality are associated with human development index and health setups in Africa. J Egypt Natl Cancer Inst 2017;29:123–126.
- 47.↑
Vanderpuye V, Hammad N, Martei Y, et al. Cancer care workforce in Africa: perspectives from a global survey. Infect Agent Cancer 2019;14:11.
