National Burden and Trend of Cancer in Ethiopia, 2010–2019: a systemic analysis for Global burden of disease study

Over the last two decades, we have tracked the national burden of cancer and its trends in Ethiopia, providing estimates of incidence, death, and disability adjusted life years. In Ethiopia, there were an estimated 53,560 (95% UI 52,480–55,540) new incident cases, 39,480 deaths (95% UI 32,640–46,440), and 1.42 million (95% UI 1.16–1.68) DALYs of cancer 2019. Cancer incidence, death, and DALYs counts increased by 32% (95% UI 11–55%), 29% (95% UI 12–44%), and 19% (95% UI − 2 to 44%) between 2010 to 2019, respectively, while age-standardised incidence, death, and DALYs rates increased by 5% (95% UI − 7 to 18%), 2% (95% UI − 9 to 14%), and − 2% (95% UI − 15 to 12%) respectively. In 2019, the leading incidence cases were leukemia, cervical cancer, breast cancer, colon and rectum cancer, and stomach cancer, while leukemia, breast cancer, cervical cancer, and stomach cancer were the most common killer cancers in Ethiopia. According to the findings of this study, tobacco-related cancers such as pancreatic, kidney, and lung cancer have increased in Ethiopian females over the last decade, while genitourinary cancer has increased in Ethiopian males. Another significant finding was that infection-related cancers, such as stomach cancer and Hodgkin lymphoma, have been rapidly declining over the last decade.

Stomach cancer, leukemia, larynx cancer, nasopharynx cancer, esophageal cancer, Hodgkin's lymphoma, gallbladder and biliary tract cancer, and mesothelioma have decreased age-standardised incidence rate from 2010 to 2019, ranging from − 1 to − 14%, and other malignant cancers have seen an increment in age-standardised incidence rate, ranging from 1 to 30%.

Discussion
Between 2010 and 2019, the absolute number of cancer incidence cases, mortality, and DALYs increased significantly in Ethiopia. However, the age-standardised rate of cancer incidence, death, and DALYs shows erratic trends. From 2010 to 2019, we found that the trends in the age-standardised cancer incidence rate were fairly stable. Similarly, neighboring countries such as Djibouti, Eritrea, Kenya, Somalia, South Sudan, and Sudan have seen similar trends. However, there were contrast trend in high income countries and global 4 . Cancer is becoming more prevalent around the world, particularly in low and middle-income countries. According to WHO projections, low and middle income (LMIC) countries will bear two-thirds of the cancer burden in 2040 3 .The main reasons for the rapid rise in cancer in low and middle income countries are population growth, aging, sociodemographic, and epidemiological transitions (LMICs) 1 .
The change in the incidence of cancer cases in Ethiopia, on the other hand, was primarily driven by population growth and aging. The stable age-standardised cancer incidence rate suggests that epidemiological and sociodemographic transitions play a minor role in cancer pathogenesis in Ethiopia. In 2019, behavioral risks, metabolic, occupational exposure, and air pollutions were attributed to approximately 20% (17-26%) of cancer in Ethiopia; however, changes in overall risk factors were less than 10% between 2010 and 2019.From 2010 to 2019, the age-standardised rate of cancer death in Ethiopia increased. The findings of this study stand in stark contrast to the age-standardised cancer rates in high-income countries 12 and global trends 4 .
The increased age-standardised cancer death rate calls into question national policy in terms of progress in treatment and management, primary prevention, and secondary prevention modality implementation. Cancer is  www.nature.com/scientificreports/ responsible for one out of every six deaths worldwide, according to a WHO report 3 . Many global initiatives have been launched to address the cancer burden. However, global, regional, and national efforts for cancer prevention in low and middle-income countries remain insufficient and inequitable 4 . In high-income countries, a strong health-care system, a large human resource capacity, and effective primary and secondary prevention methods are responsible for lower mortality 12 whereas the main causes of increased cancer related mortality are a lack of workforce capacity, poor cancer care infrastructure, a lack of cancer centers for diagnosis and treatment, a lack of financial security, and a lack of universal health coverage 3,13 . Despite an increase in the age-standardised death rate of overall cancer, some cancer types had decreased age-standardised death rates, such as thyroid cancer, gall bladder and biliary tract cancer, cervical cancer, Nasopharynx cancer, esophageal cancer, leukemia, larynx cancer, stomach cancer, and Hodgkin lymphoma, which ranged from − 1 to − 18% over the last one decades. The main agent for lowering mortality rates will be progress in human resource capacity building, adoption of diagnostic imaging and pathological laboratories, early detection and treatment, surgical advancement, and adaptation of an effective cervical cancer screen. Most infection-related cancers, such as cervical, stomach, nasopharynx, and Hodgkin lymphoma, have steadily declined in Ethiopia over the last one decade. Cancer trends and outcomes are disproportionately high in low and middle-income countries 1,3,4 . Low health care budgets 14 , overburdened health-resources with communicable diseases, child and maternal health, low universal health coverage 15 , and an increased burden of cancer all contribute to significant universal healthcare disparities and inequity in low and middle-income countries. According to current evidence, primary and secondary prevention strategies could prevent more than half of all cancers 16 . A screening program based on guidelines has shown a reduction in cancer-related mortality in cervical, breast, prostate, and colorectal cancer 16 . Screening has been primarily responsible for lower rates of death and disability-adjusted life years (DALYs) for cervical cancer in Ethiopia. Currently, evidence-based modification of primary risk factors such as smoking (lung, kidney, pancreatic, and larynx cancer), H.pylori (stomach cancer), reduced alcohol consumption (liver cancer), and salted and western diets (colorectal and other GI malignancy) have aided in cancer prevention 12,17 . Global organizations advocate for and implement National Cancer Control Plans (NCCP) to address the cancer burden in low-income countries 18

Limitation
Although GBD studies provide qualitative and compressive evidence for policymakers, researchers, and planners, the quality and quantity of data sources available for estimation is dependent on the quality and quantity of data sources available for estimation. Cancer mortality is primarily estimated using the cancer registry, vital registration, and, to a lesser extent, other data sources. Ethiopia has only one population-based cancer registry, which only covers 3-5 percent of the total population.

Conclusion
Overall cancer related mortality and incidence rates increased in Ethiopia. Disparities in cancer prevention, care, and control are the primary causes of these trends. Researchers, health care professionals, and policymakers must work together to develop screening guidelines and protocols, improve cancer care infrastructure, capacity building, surgical and chemoradiotheray policy, and maximize primary cancer prevention, secondary cancer prevention, early diagnosis, treatment, and rehabilitative care to reduce cancer-related mortality and disability.  www.nature.com/scientificreports/

Methods
We extracted data from the GBD 2019 results too (http:// ghdx. healt hdata. org/ gbd-resul ts-tool). The method and data sources are described in detail in GBD 2019 publications and previous GBD publications 11,20,21 . The Guideline for Accurate and Transparent Health Estimated Reporting (GATHER) statement was used to create GBD 2019. In summary, the 2019 Global Burden of Disease, Injury, and Risk Factors Study reported national estimates of cancer incidence, mortality, and DALYs from 1990 to 2019. Estimates for GBD 2019 were analyzed and evidence for 363 non-fatal diseases, 302 deaths, and 87 risk factors were reported in 204 countries and 21 regions 11 . To calculate disease incidence, mortality, and DALYs, the GBD study collects data from vital registration, verbal autopsy, cancer registry, sample vital registry, censes, demographic and health surveys, published and unpublished health data, and other sources. GBD produced sound and up-to-date evidence of trends at the global, regional, and national levels as a result of the shift in the global agenda and increased focus on noncommunicable disease and injury. GBD studies used three main standardised modeling tools to process data, model, and generate each estimation of disease by age, location, sex, and year-Cause of Death Ensemble (CODEm), DisMod-MR, and Spatiotemporal Gaussian Process Regression (ST-GPR). Cancer registry incidence data were used to calculate the mortality rate. The first model was MIR, which is based on a cancer registry and includes both mortality and incidence. MIR is a liner-step mixed-effects model that includes a logit link function, HAQ, age, and gender as covariates. Spatiotemporal Gaussian process regression was used to smooth and adjust the final model. The final model CODEm was built using observed mortality data and MIR model estimated mortality. To estimate cancer incidence, the final cancer specific mortality estimates are divided by MRI. DisMod-MR is a Bayesian meta-regression tool that uses all available data to estimate the incidence and prevalence of each disease over time. Years lived with disability (YLDs) are calculated by dividing 10-year cancer prevalence into four categories: (1) diagnosis/treatment, (2) remission, (3) metastasis/dissemination, and (4) terminal phase. Years of life lost (YLLs) are calculated by multiplying the estimated number of deaths by age by the age's standard life expectancy. The sum of YLDs and YLLs yields disability-adjusted life-years (DALYs). For age standardised rates and all rates reported per 100,000, the GBD world population is used. All estimates have 95 percent confidence intervals (UI). The GBD2019 publications contain detailed descriptions of methodology, modeling, and data sources. The GBD2019 publications contain detailed descriptions of methodology, modeling, and data sources 11,20,21 . We focused on the national burden of cancer in Ethiopia, estimating the burden in terms of incidence, DALYs, and mortality for 29 cancer categories.