Global burden of female breast cancer and its association with socioeconomic development status, 1990–2044

Abstract Background Breast cancer is a widespread disease in women worldwide. Aim We aimed to explore the global epidemiological trends of female breast cancer (FBC) between 1990 and 2044. Methods and Results Disease burden, population, and socio‐demographic index (SDI) data were obtained from the Global Health Data Exchange (GHDx) database. We analyzed temporal trends, age differences, risk factors, and geographic patterns of FBC disease burden globally and explored the association between age‐standardized incidence rate (ASIR) of FBC and SDI. Bayesian age‐period‐cohort model was also performed to predict the changes in FBC incidence worldwide from 2020 to 2044. First, the global ASIR of FBC increased by 14.31% from 1990 to 2019 (95% Uncertainty Interval 4.75% to 23.98%). The death rate presented a falling trend. Second, alcohol use is the most‐highlighted risk factor for FBC in some high‐income regions such as Europe. A high fasting plasma glucose levels is the most prominent risk factor for FBC in Latin America and Africa. Third, the ASIR of the FBC increases with the SDI. Fourth, the incidence is expected to increase faster among women aged 35–60 years and fastest for those aged 50–54 years from 2020 to 2044. Countries with a high incidence of FBC that is expected to increase significantly include Barbados, Burkina Faso, Senegal, Monaco, Lebanon, Togo, and Uganda. Conclusion The disease burden of FBC varies worldwide; the findings suggest attaching importance to the control of middle and low‐middle SDI regions. Public health as well as cancer prevention experts should pay more attention to regions and populations at an increased risk of developing FBC, focusing on their prevention and rehabilitation while conducting further epidemiological studies to investigate the risk factors of their increase.


| BACKGROUND
The incidence of breast cancer is increasing worldwide, with 4.4 million cases predicted by 2070. 1 Women constitute the majority of breast cancer patients, and female breast cancer (FBC) remains widespread among women in most regions. 2 FBC accounts for roughly 24.5% of all cancer cases and for 15.5% of cancer deaths in women, leading in most countries in terms of incidence and death rates in 2020. 3 There are considerable variations in the morbidity, death, and survival rates of FBC between regions. 4 Age-standardized incidence rate (ASIR) ranged from 112.3/100000 in Belgium to 35.8/100000 in Iran, while age-standardized death rate (ASDR) ranged from a high of 41.0 per 100 000 in Fiji to a low of 6.4 per 100 000 in Korea. 5 Therefore, FBC is a public health issue that warrants attention.
FBC risk factors are multifaceted and include both uncontrollable (age, 6 family history, 7 race 8 ) and controllable (BMI, 6 smoking, 6 alcohol consumption, 9 diabetes, 10 timing of first delivery, 11 breastfeeding, 6 organic solvent occupational exposure, 12 and electromagnetic occupational exposure 13 ) factors. Meanwhile, COVID-19 had a significant impact on FBC screening, breast surgery, and genetic counseling, 14 with a general decrease in the number of patients accessing prevention, screening, diagnosis, and treatment during the outbreak.
Local management of FBC, adjuvant systemic therapy, and treatment of patients with advanced disease have evolved in recent years 15 toward identifying more conservative ways to treat the cancer and provide the best quality of life for patients. 16 However, no valid vaccine has yet been produced to protect against FBC. 17 Furthermore, extended and more active FBC therapies have raised the prevalence of long-term survival. 18 Therefore, it is necessary to accurately predict the risk of FBC and develop individualized strategies to identify better ways to prevent the disease.
Most previous studies were based on all breast cancer patients, 19 and the prediction is localized. 20 This study focuses on FBC, with the prediction covering all countries. We analyzed the temporal trends, age differences, risk factors, and geographic patterns of FBC disease burden worldwide and examined the association between ASIR and socio-demographic index (SDI). In addition, we made age-specific and country-specific projections of changes in FBC incidence worldwide from 2020 to 2044 to better identify populations at increased risk of FBC. On this basis, trends worthy of public health and cancer prevention experts' close attention were highlighted.

| Data analysis
We described the trend of an interval or a whole period using the average annual percentage change (AAPC) and its 95% Uncertainty Interval (UI). This metric is derived from the Joinpoint regression analysis, which has been widely used to analyze cancer mortality and incidence data. 25 In addition, we determined the annual percentage change in each identified trend of FBC rates using the calendar year as a regression variable. The AAPC throughout the considered period was also calculated. Based on a Poisson regression model, the positions of joinpoints and regression coefficients were estimated, while the optimal number of joinpoints was selected by means of a permutation test. Each P-value was calculated using the Monte Carlo methods, and the overall asymptotic significance level was maintained through a Bonferroni correction. 26 A P-value of less than 0.05 was considered statistically significant. If the lower UI of AAPC is above 0, it reveals an uphill tendency of the indicator, and if the upper UI of AAPC is below 0, it indicates a downward trend of the indicator. Additionally, if the confidence interval contains 0, it indicates that the trend of change is not statistically significant. 27 We obtained 30 years of data for 22 regions, with a total of 660 sets, based on which we explored the relationship between the ASIR and SDI of FBC. As this relationship is difficult to transform into a linear model with a simple function, a polynomial regression was performed. The general form of the polynomial is: The purpose of the polynomial fit is to find a set of p 0 , p 1 , …p n , so that the fitted equations match the actual sample data as closely as possible. The significance level was set at P < .05. The adjusted R-squared was chosen to judge the fit of the model, which represents the proportion of the variance of y by the fitted values. For a series of true values (y i ) and fitted It takes a value between 0 and 1, with a value closer to 1 representing a better fit. 28 We also examined the temporal trends in ASIR for the five SDI regions during 1990-2019.
We performed Bayesian age-period-cohort (APC) analyses for incidence prediction, which shows better coverage and precision than other prediction methods as it involves no parametric assumptions. 29  Europe, DALYs of FBC were mainly attributed to drinking. In Africa and the Americas, DALYs of FBC were mainly attributed to high fasting plasma glucose. In Oceania and Southeast Asia, DALYs of FBC were mainly attributed to high body mass index. Figure 4 shows the ASDR for the different risk factors leading to FBC in each GBD region, presenting similar characteristics to attributable DALYs. Figure 5 illustrates the relationship between the ASIR and SDI of the FBC. We selected an order of 2 for the polynomial fitting process (ρ < 0.05), and the adjusted R-squared value was 0.6322, which was a good fit. 3.4 | Age-specific and country-specific projections of female breast cancer incidence in 2020-2044       36 and chronic pain. 37 In the meantime, some patients may undergo mastectomy, which may lead to concerns about body image and sexuality. 38 Therefore, psychological disorders are also a common problem for patients with FBC after surgery. 39 In this context, we must focus on the rehabilitation of FBC survivors, 40  contributing to some of the differences in death rates across regions.

| Relationship between the incidence of female breast cancer and SDI
As for age structure, the worldwide disease burden of FBC increases with age, which is in accordance with the available studies. 43 Meanwhile, we predicted the incidence of FBC showed a noticeable rising trend between 35 and 60 years, with the fastest rising rate in 50-54 years. Most randomized clinical trials and FBC screening guidelines recommend a uniform protocol, suggesting that all women should begin screening at 50 years old. 44 However, when considered in conjunction with the predicted results of the risk of incidence by age group, it is necessary to adopt a risk-appropriate age of screening onset and optimal time interval 45 to ensure equity and validity of breast cancer screening claims. 46 This finding suggests that focus should be placed on the detection and control of FBC in middle-aged and older women.
In terms of risk factors, alcohol use 47 and high fasting plasma glucose 48 are the remarkable influential risk factors for FBC, which is in agreement with previous studies. 49 Especially in Europe, the disease burden from alcohol use is even worse. As European countries have a traditional wine-oriented diet, 50 their total per capita alcohol consumption is higher than that of most developing countries, 51 with the existence of alcohol dependence and abuse. 52 The results of most epidemiological as well as experimental animal studies, 47 suggest that alcohol intake can lead to the development of FBC through different mechanisms. In some African regions, the disease burden from high fasting plasma glucose is even worse. High fasting plasma glucose will likely develop into diabetes, which increases the risk of breast cancer.
Diabetes is a major challenge faced by many African health systems. 53 Studies 54 have shown that over half of the cases of diabetes in Africa are undiagnosed. The results of the projection study 55  hormones, which further increases the risk of breast cancer. 48,56,57 From the relationship between ASIR and SDI, the results of our data corroborate that the ASIR of FBC increases with SDI, which is consistent with previous studies. 58 The SDI 59  The association of fatalism with screening behavior may lead to a higher incidence in some racial/ethnic minority populations. [68][69][70] The fatalism believed by some African-American populations affects their cancer screening behavior and may reduce the use of screening mammograms by women, 71

ACKNOWLEDGMENTS
We highly appreciate the works by the Global Burden of Disease Study 2019 collaborators.

FUNDING INFORMATION
This research was funded by Guangdong Provincial Natural Science Funds (grant number 2022A1515011871).