The global, regional, and national burden of early, middle, and late-onset esophageal cancer and trends from 1990 to 2021: a systematic analysis of the Global Burden of Disease Study 2021

Introduction

Esophageal cancer represents a significant global health challenge, ranking seventh in cancer-related mortality worldwide with more than 511,000 new cases and 445,000 deaths estimated in 2022 (1). Histologically, esophageal cancer is classified into esophageal adenocarcinoma (EAC) and esophageal squamous cell carcinoma (ESCC), each characterized by unique molecular characteristics and population screening criteria (2). It is typically a multifactorial disease of the elderly, with its incidence rising significantly with age. Moreover, the burden of esophageal cancer varies significantly across regions and populations, reflecting disparities in underlying risk factors, socioeconomic development, and the geographic distribution of subtypes (3).

Although the overall incidence and mortality of esophageal cancer have gradually decreased globally attributed to early screening and multimodal therapies, our understanding of the differences in disease burden across various age groups remains incompletely understood and controversial (4). Evidence indicated that early-onset esophageal cancer (EOEC), defined as occurring in individuals under 50 years, appeared to be associated with more tumor stages and poorer survival outcomes (5). Notably, recent research has increasingly focused on the risk factors and prognosis of EOEC (6,7), as significant disparities in epidemiology and clinicopathological characteristics are evident across age groups (8). Current screening guidelines recommend endoscopic screening for esophageal cancer in individuals over 50 years old, but the most appropriate age threshold to achieve the best detection rates and screening effectiveness remains insufficiently understood (9). Recent data from the Surveillance, Epidemiology, and End Results (SEER) suggested that 10.3% of esophageal cancers were diagnosed in individuals younger than 55 years of age, with 63.5% of new cases occurring among individuals older than 65 years of age (10). Furthermore, earlier epidemiological studies have explored the disease burden of esophageal cancer, utilizing data from public research databases or national registries (11,12). Nevertheless, most research has primarily addressed regional or national burdens, with a notable emphasis on Asian countries, particularly China (12,13). Hence, studying the age-stratified epidemiological patterns of esophageal cancer is crucial for a deeper understanding of the disease.

The Global Burden of Disease (GBD) study provided an invaluable framework for comprehensively assessing the burden of esophageal cancer through a range of estimated metrics, serving as a reliable tool for understanding the disease, identifying risk factors, and proposing potential solutions (14). In this study, we aimed to systematically evaluate the temporal trends in incidence, mortality, and disability-adjusted life-years (DALYs) of early, middle, and late-onset esophageal cancer at the global, regional, and national levels from 1990 to 2021, perform frontier and decomposition analysis, quantify cross-country inequalities, and forecast the changes to 2035. Our analyses were stratified by age of onset, gender, and sociodemographic index (SDI), providing insights to inform more effective targeted prevention and treatment strategies. We present this article in accordance with the GATHER reporting checklist (15) (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-782/rc).

Methods

Data collection and definitions

The GBD 2021 (https://ghdx.healthdata.org/gbd-2021) systematically assessed the health burden of 371 diseases and injuries along with 88 risk factors for 204 countries and 21 GBD regions (Table S1), utilizing the latest epidemiological data and refined standardized methodologies (16). In this study, we extracted these estimates and their 95% uncertainty intervals (UIs) for incidence, mortality, and DALYs relating to esophageal cancer. The SDI is a composite indicator that quantifies the development levels of different countries and regions, based on fertility rate, education level, and income per capita (17). In this study, countries and regions were grouped into five quintiles based on their SDI values: low, low-middle, middle, high-middle, and high. In addition, we further examined the gender-specific distribution of the disease burden. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

In the GBD 2021, esophageal cancer was defined according to the 10th revision of the International Classification of Diseases (ICD-10; codes C15–C15.9, D00.1, and D13.0). Our study categorized esophageal cancer cases into three distinct groups based on age at onset: early-onset (EOEC, 15–49 years), middle-onset (MOEC, 50–69 years), and late-onset (LOEC, ≥70 years) esophageal cancer, aligning with prior research (8,18).

Statistical analyses

The age-standardized rates (ASRs), based on the global standard population (Table S2), were calculated to ensure the comparability of age-standardized incidence rate (ASIR), mortality rate (ASMR), and DALYs rate (ASDR). All rates were reported per 100,000 population.

Joinpoint regression model was employed to evaluate temporal trends of disease burden, using Joinpoint Trend Analysis Software developed by the National Cancer Institute of the United States (19). Annual percent change (APC) was applied to delineate the local trends across different time intervals and average annual percent change (AAPC) was calculated to demonstrate the overall trend from 1990 to 2021, along with their 95% confidence intervals (CIs). In addition, considering that different research methods to explore the temporal trends of ASRs could introduce potential biases into results, we conducted sensitivity analyses by comparing the results of estimated annual percentage change (EAPC) with that of AAPC. Furthermore, we explored the relationship between SDI and burden trends of esophageal cancer at the regional and national level. Frontier analysis was employed to investigate the association between ASDR and sociodemographic development. The effective difference, defined as the gap between a country’s ASDR and its frontier, reflects the potential health gains that remain unrealized at the corresponding SDI level (20). Slope index of inequality (SII) and concentration index, reflecting absolute and relative health inequalities respectively, were used to quantify the cross-country inequalities of esophageal cancer burden (21). Decomposition analysis was performed to estimate the contribution of three factors to changes in incidence, mortality, and DALYs for esophageal cancer from 1990 to 2021 (i.e., aging, population growth, and epidemiological changes). We also analyzed the contribution of risk factors to esophageal cancer burden. In GBD 2021, four major risk factors contributed to DALYs of esophageal cancer: smoking, high alcohol use, diet low in vegetables, and chewing tobacco.

Bayesian age-period-cohort (BAPC) model integrated nested Laplace approximations, a logarithmic linear Poisson model, was used in our study to forecast future epidemiological trajectories (22). The BAPC model assumed the multiplicative impact between age, period, and cohort and second-order random walk priors were employed to adjust for overdispersion. Despite challenges such as wide prediction intervals in certain cases, it provides well-calibrated and sensible forecasts, outperforming alternative approaches like linear power models (22). To validate the accuracy and stability of prediction results, a sensitivity analysis was conducted using the autoregressive integrated moving average (ARIMA) model (23). Given its ability to incorporate age, period, and cohort effects, the BAPC model offers a more biologically plausible framework than ARIMA, which mainly relies on historical patterns; therefore, we prioritized BAPC projections, using ARIMA as a supplementary sensitivity analysis. We constructed these models based on esophageal cancer data from 1990 to 2021 and forecasted ASR trends from 2022 to 2035.

All statistical analyses and visualizations were performed using R software (version 4.2.1). Two-sided P<0.05 was defined with statistical significance.

Results

Global trends of EOEC, MOEC, and LOEC

In EOEC, the global ASIR declined from 1.95 per 100,000 (95% UI: 1.73 to 2.19) in 1990 to 1.08 (95% UI: 0.96 to 1.22) in 2021, with an AAPC of −1.9 (95% CI: −2.05 to −1.74, P<0.001) (Table 1 and Table S3). Similarly, the global age-standardized mortality and DALYs rates both decreased over the study period. With respect to gender, the ASRs declined more significantly in males than females for ASIR (AAPC: −1.93 vs. −1.73), ASMR (AAPC: −2.36 vs. −2.06), and ASDR (AAPC: −2.36 vs. −2.02) (Table 1 and Figure 1A-1C). As for MOEC, the ASIR also decreased from 29.61 per 100,000 (95% UI: 26.06 to 32.93) in 1990 to 19.55 (95% UI: 17.06 to 22.40) in 2021 globally, with an AAPC of −1.34 (95% CI: −1.44 to −1.24, P<0.001) (Table 1 and Table S3). The ASMR and ASDR followed a similar downward trend from 1990 to 2021. Unlike EOEC, the ASRs of MOEC declined more significantly in females than males over the past 30 years (Table 1 and Figure 1D-1F). In contrast to EOEC and MOEC, the incidence rates of LOEC also decreased from 54.93 per 100,000 (95% UI: 48.32 to 60.19) in 1990 to 51.19 (95% UI: 44.19 to 58.00) in 2021, with an AAPC of −0.24 (95% CI: −0.38 to −0.10, P<0.001) (Table 1 and Table S3). The trends of mortality and DALYs were similar to that of the incidence rates (Figure 1G-1I).

Figure 1 Global trends in incidence, mortality and DALYs of EOEC (A-C), MOEC (D-F) and LOEC (G-I) from 1990 to 2021. ASDR, age-standardized disability-adjusted life year rate; ASIR, age-standardized incidence rate; ASMR, age-standardized mortality rate; DALYs, disability-adjusted life years; EOEC, early-onset esophageal cancer; LOEC, late-onset esophageal cancer; MOEC, middle-onset esophageal cancer.

Joinpoint regression analysis revealed that the ASRs of EOEC (Figure 2A-2C) and MOEC (Figure 2D-2F) followed a similar downward trend from 1990 to 2021. Of note, the global ASIR of LOEC increased from 1990 to 2004, followed by a continuous decline since 2004, with a slight increase from 2014 to 2021 (APC =0.34, P<0.05). Detailed information about ASMR and ASDR of LOEC was also provided (Figure 2G-2I and Table 1). When stratified by gender, a significant increase was observed for ASIR of LOEC in males from 2016 to 2021 (APC: 0.80), and the 2015–2021 period also exhibited a slight increase in ASMR and ASDR but were not statistically significant (Figure S1). However, in females, the ASMR and ASDR of LOEC both showed a significant decline from 2014 to 2021 (APC: −0.26 and APC: −0.23, respectively) (Figure S2).

Figure 2 Joinpoint regression analysis of global incidence, mortality and burden for EOEC (A-C), MOEC (D-F) and LOEC (G-I) from 1990 to 2021. *, with significance, P value <0.05. APC, annual percentage change; ASDR, age-standardized disability-adjusted life year rate; ASIR, age-standardized incidence rate; ASMR, age-standardized mortality rate; EOEC, early-onset esophageal cancer; LOEC, late-onset esophageal cancer; MOEC, middle-onset esophageal cancer.

Trends of EOEC, MOEC, and LOEC by regions and nations

The burden of esophageal cancer exhibited significant variation across regions and countries. According to the GBD region classification, the ASIR for EOEC was highest in Southern Sub-Saharan Africa at 2.39 (95% UI: 2.03 to 2.78), followed by East Asia at 2.30 (95% UI: 1.81 to 2.90) (Table 1). In contrast, the lowest ASIR was observed in Andean Latin America, recorded at 0.20 (95% UI: 0.15 to 0.26). Western Sub-Saharan Africa showed a slight increasing trend in the ASIR during the period from 1990 to 2021, with an AAPC of 1.39 (95% CI: 1.22 to 1.56, P<0.001). Similarly, the ASMR and ASDR exhibited a comparable downward trend across most GBD regions from 1990 to 2021, except for Western Sub-Saharan Africa. For MOEC, the regions with the highest ASIR were East Asia at 40.35, Southern Sub-Saharan Africa at 35.91, and Eastern Sub-Saharan Africa at 35.37. Andean Latin America exhibited the lowest ASRs, followed by Central Latin America. In terms of LOEC, the age-standardized incidence, mortality, and DALYs rates were found to be highest in East Asia and lowest in Central Latin America. Notably, Central Asia experienced the most significant downward trend from 1990 to 2021 in ASIR (AAPC: −3.11; 95% CI: −3.94 to −2.27; P<0.001), ASMR (AAPC: −3.13; 95% CI: −3.99 to −2.28; P<0.001), and ASDR (AAPC: −3.13; 95% CI: −3.93 to −2.33; P<0.001).

Stratified by countries and territories, Chad (AAPC: 2.57; 95% CI: 2.15 to 2.99, P<0.001), Lesotho (AAPC: 2.42; 95% CI: 2.08 to 2.75, P<0.001), and Liberia (AAPC: 2.39; 95% CI: 2.06 to 2.72, P<0.001) exhibited the most significant rise in ASIR for EOEC from 1990 to 2021 (Figure 3A,3B and Table S4). Notably, two of the three countries (Chad and Liberia) are located in Western Sub-Saharan Africa, aligning with our regional analysis. Malawi had the highest ASMR at 5.04 (95% UI: 3.38 to 7.39) and ASDR at 236.52 (95% UI: 158.03 to 348.20) for EOEC in 2021 (Figures S3,S4 and Tables S5,S6). In MOEC, the ASIR varied from approximately 1.81 to 85.87. Among all countries, Malawi, Eswatini, and Lesotho exhibited the highest ASIR (Figure 3C,3D and Table S4). The most significant decrease in the ASMR of MOEC was observed in Uzbekistan (AAPC: −4.83; 95% CI: −5.92 to −3.71, P<0.001) (Figure S3 and Table S5). Kuwait, Tunisia, and Algeria, which are all located in North Africa and Middle East, had the lowest ASDR of MOEC (Figure S4 and Table S6). As for LOEC, similarly, Malawi leaded in three ASRs in 2021, with the highest ASIR at 166.24 (95% UI: 120.31 to 221.99), ASMR at 197.64 (95% UI: 142.66 to 263.99), and ASDR at 3,157.13 (95% UI: 2,288 to 4,203.9) (Figure 3, FigureS3E,S3F, Figure S4E-S4F, and Tables S4-S6). Mongolia ranked second in ASIR at 147.92 (95% UI: 111.78 to 188.65), while China ranked third at 125.96 (95% UI: 101.01 to 152.79). Besides, we also listed the countries and territories with top 5 significantly increased EAPC of ASRs from 1990 to 2021 (Table S7), which was used to assess the robustness of the results with joinpoint regression method. The results indicated that the temporal trends were mostly consistent. The distribution of ASIR, ASMR, and ASDR stratified by gender among 204 countries and territories was detailed in Figures S5-S10.

Figure 3 Among 204 countries and territories, the ASIR of EOEC (A), MOEC (C) and LOEC (E) in 2021, and the EAPC of ASIR for EOEC (B), MOEC (D) and LOEC (F) from 1990 to 2021. ASIR, age-standardized incidence rate; EAPC, estimated annual percent changes; EOEC, early-onset esophageal cancer; LOEC, late-onset esophageal cancer; MOEC, middle-onset esophageal cancer.

Disease burden of EOEC, MOEC, and LOEC based on SDI

In EOEC, the high-middle SDI regions reported the highest ASIR at 1.51 (95% UI: 1.20 to 1.90), ASMR at 1.05 (95% UI: 0.84 to 1.31), and ASDR at 49.00 (95% UI: 39.32 to 61.52) in 2021 (Table 1). Middle SDI regions exhibited the most significant decline from 1990 to 2021. In regard to MOEC, the ASIR was highest in high-middle SDI regions, recorded at 26.25 (95% UI: 21.11 to 33.08) and lowest in low-middle SDI regions, recorded at 11.44 (95% UI: 10.23 to 13.1) (Table 1). The most notable reduction in ASIR was observed in middle SDI regions, with an AAPC of −2.21 (95% CI: −2.32 to −2.11, P<0.001). The temporal trends were similar in ASMR and ASDR, with regions across different SDI levels all showing a decreasing trend between 1990 and 2021. As for LOEC, the middle SDI regions reported the highest ASRs, while low-middle SDI regions had the lowest for the three measures. The ASIR in high SDI regions increased from 34.85 (95% UI: 32.16 to 36.46) in 1990 to 38.53 (95% UI: 34.19 to 41.05) in 2021, with an AAPC of 0.3 (95% CI: 0.12 to 0.49, P=0.001) (Table 1 and Table S3). In contrast, most other SDI regions demonstrated overall downward trends. At the regional level, from 1990 to 2021, ASRs exhibited a declining trend with increasing SDI for EOEC and MOEC in most regions; however, this relationship reversed for LOEC in regions with a relatively high SDI (>0.8) (Figure S11). A similar relationship was observed for both males and females (Figures S12,S13). At the national level in 2021, higher SDI were generally associated with lower ASRs of esophageal cancer; however, the rates tended to reach the lowest point at an SDI of around 0.80 and increased with a further rise in SDI, particularly for LOEC (Figure S14). Similar patterns were observed in males and females (Figures S15,S16).

To analyze the potential improvement space of the ASDR for esophageal cancer considering country’s development status, frontier analysis was applied with data from 1990 to 2021 (Figure 4, Figure S17 and Table S8). For EOEC and MOEC, the countries and territories furthest from the frontier line were the same, including Malawi, Eswatini, Lesotho, Zimbabwe, and Zambia. Notably, these countries are all located in Sub-Saharan Africa, which indicates that there is considerable potential to alleviate the burden of esophageal cancer. In regard to LOEC, the countries and territories furthest from the frontier line included Malawi, Mongolia, Zimbabwe, China, and Zambia. It should be noted that the effective differences varied between males and females (Tables S9,S10).

Figure 4 Frontier analysis based on SDI and ASDR of EOEC (A-C), MOEC (D-F) and LOEC (G-I) in 2021. The frontier is depicted as a solid black line, with countries and territories represented by dots. The top 15 countries showing the largest effective difference (i.e., the greatest gap in ASDR from the frontier) are labeled in black. Examples of frontier countries with low SDI (<0.5) and small effective differences are labeled in blue, while examples of countries and territories with high SDI (>0.85) but relatively large effective differences for their development level are labeled in red. Red dots indicate an increase in ASDR from 1990 to 2021, while blue dots indicate a decrease in ASDR during the same period. In cases where a top 15 countries also meet the criteria for red or blue dots, color-coding by ASDR trend takes priority, so fewer than 15 black-labeled countries may appear. ASDR, age-standardized disability-adjusted life year rate; EOEC, early-onset esophageal cancer; LOEC, late-onset esophageal cancer; MOEC, middle-onset esophageal cancer; SDI, socio-demographic index.

To quantify the SDI-related cross-country inequalities of DALYs, SII and concentration index were calculated. Generally, the regression curves had negative slopes, indicating a higher disease burden was disproportionately shouldered by economically underdeveloped regions (Figure 5, Figure S18). From 1990 to 2021, SII values for both sexes combined and for males alone exhibited a worsening inequality for EOEC, MOEC, and LOEC in countries with lower SDI, while females experienced improving inequality (Table S11). Additionally, health inequality concentration curves for DALYs were developed to measure the relative SDI-related inequalities (Figure S19). The concentration index showed a decreasing trend for both sexes combined from 1990 to 2021, suggesting an increasing concentration of DALYs in low-income countries and territories (Table S12).

Figure 5 Health inequality regression curves for the DALYs of EOEC (A-C), MOEC (D-F) and LOEC (G-I) worldwide, in 1990 and 2021. The SII values, i.e., the slopes of the regression curves, are labeled next to the corresponding curves. DALYs, disability-adjusted life-years; EOEC, early-onset esophageal cancer; LOEC, late-onset esophageal cancer; MOEC, middle-onset esophageal cancer; SDI, socio-demographic index; SII, slope index of inequality.

Decomposition analysis

Decomposition analysis was performed at the global level, across 5 SDI quintiles and 21 GBD regions, to investigate the relative contributions of aging, population, and epidemiologic changes in incidence, mortality, and DALYs of esophageal cancer (Figure 6 and Table S13). Globally, aging and population contributed 198.72% and 339.53%, respectively, to the reduction of DALYs in EOEC. The contribution of population growth to overall incidence of MOEC was most pronounced across all SDI regions while aging had the least effect. Population growth contributed most to the increase of incidence in LOEC, especially in middle-SDI regions, at 142.01%. The effects of three population-level determinants on disease burden differed across GBD regions and genders (Figure S20,S21 and Tables S14,S15).

Figure 6 Changes in incidence, mortality, and DALYs for EOEC (A-C), MOEC (D-F), and LOEC (G-I) attributed to aging, population growth, and epidemiological changes at the global, SDI quintile, and regional levels from 1990 to 2021. Black dots represent the cumulative contribution of all three factors to the observed changes. A positive value for any component reflects its contribution to an increase in esophageal cancer incidence, mortality, and DALYs, while a negative value indicates a reduction in these measures. DALYs, disability-adjusted life-years; EOEC, early-onset esophageal cancer; LOEC, late-onset esophageal cancer; MOEC, middle-onset esophageal cancer; SDI, socio-demographic index.

Proportion of DALYs attributable to risk factors

We identified the percentage contributions of four major risk factors to DALYs of esophageal cancer in 1990 and 2021: smoking, high alcohol use, diet low in vegetables, and chewing tobacco (Figure 7). Globally, smoking was the most important contributor to DALYs both in 1990 and 2021 among three groups. The region’s socio-demographic progression could also affect the contributions of risk factors to DALYs.

Figure 7 Percentage contributions of major risk factors to DALYs of EOEC, MOEC and LOEC at the global, SDI quintile, and regional levels in 1990 and 2021. DALYs, disability-adjusted life-years; EOEC, early-onset esophageal cancer; LOEC, late-onset esophageal cancer; MOEC, middle-onset esophageal cancer; SDI, socio-demographic index.

In the low SDI regions, diet low in vegetables was identified as the primary risk factor in 2021 among EOEC (24.3%), MOEC (25%), and LOEC (25.1%). While in the high SDI regions, high alcohol use was the main risk factor for EOEC (26.5%), and smoking was the main risk factor for MOEC (41.7%) and LOEC (42.5%). Moreover, the percentage contributions of risk factors varied by different GBD regions. Compared with that in 1990, the percentages of DALYs due to smoking showed a downward trend in most GBD regions. For example, in Tropical Latin America, the percentage of smoking was decreased by 17.4% in EOEC, 20.0% in MOEC, and 20.0% in LOEC, respectively. The percentage contributions of risk factors differed between males and females (Figures S22,S23). For example, smoking contributed more to esophageal cancer-related DALYs in males, while diet low in vegetables contributed more to females. The gender differences in these key risk factors could promote more effective prevention strategies to reduce the disease burden of esophageal cancer.

Future forecasts of global burden of esophageal cancer

The BAPC model was used to predict the disease burden of esophageal cancer from 2022 to 2035 (Figure 8). The ASIR of EOEC was projected to increase globally for both sexes combined, rising from approximately 1.08 per 100,000 persons in 2021 to about 1.18 by 2035. The ASMR and ASDR of MOEC are projected to decrease for both sexes combined (Table S16). Furthermore, the ASRs of LOEC in males were projected to show an upward trend while that in females are expected to decrease, so the gaps between males and females appeared to expand over time (Figures S24,S25). Besides, we also used the ARIMA model to validate the predictions (Figures S26-S28 and Table S17). While the results aligned with the BAPC model for most measures, certain differences remained. For example, the ASIR of EOEC was projected to increase with the BAPC model while it was expected to decrease with the ARIMA model.

Figure 8 BAPC model predicted trends of ASIR, ASMR and ASDR for EOEC (A-C), MOEC (D-F), and LOEC (G-I): observed (1990–2021) and predicted rates (2022–2035). The blue region in the figures shows the upper and lower limits of the 95% UI. ASDR, age-standardized disability-adjusted life year rate; ASIR, age-standardized incidence rate; ASMR, age-standardized mortality rate; BAPC, Bayesian age-period-cohort; EOEC, early-onset esophageal cancer; LOEC, late-onset esophageal cancer; MOEC, middle-onset esophageal cancer; UI, uncertainty interval.

Discussion

Globally, esophageal cancer represents a major public health challenge, contributing to both disability and mortality. Our comprehensive analysis demonstrated the great changes in the disease burden associated with early, middle, and late-onset esophageal cancer from 1990 to 2021 at the global, regional, and national levels, highlighting the importance of understanding this disease across different age groups. Compensated for the limited and outdated results of prior research (1,24), our study also investigated risk factors, SDI-related burden, and gender-specific estimates.

Globally, the ASRs for EOEC and MOEC have exhibited an overall downward trend. However, the joinpoint regression analysis revealed that the global ASIR of LOEC increased from 2014 to 2021. Esophageal cancer often cannot be discovered in its early stages due to lack of symptoms or atypical presentations. However, with advancements in medical technology, the widespread implementation of early screening and diagnosis, particularly the development of endoscopic techniques, has significantly enhanced the ability to diagnose esophageal cancer in early stage (25). Patients that previously went undetected may now be identified and recorded, leading to an apparent increase in incidence rates. Although the ESCC has historically been the dominant histological subtype worldwide, the incidence of EAC has risen substantially in recent years, especially in high-income regions such as Europe and North America, reflecting a notable shift in the global epidemiological profile of esophageal cancer (4,26,27). Representatively, with changes in dietary patterns, as well as increases in smoking rates and obesity prevalence, the incidence of gastroesophageal reflux disease (GERD) has risen in many regions, particularly among the elderly (28). In 2019, the peak for ASIR of GERD was observed at age 75–79 years for males and 70–74 years for females (29). GERD can lead to esophagitis and Barrett’s esophagus, ultimately increasing the risk of EAC (28). Nevertheless, due to the paucity of studies that concurrently stratify trends by histological subtype and age group, it remains unclear whether the observed increase in LOEC incidence is partially driven by a concomitant rise in EAC. This limitation emphasizes the need for more granular epidemiological research to better understand the drivers of these patterns.

At regional and national levels, the highest ASRs of early, middle, and late-onset esophageal cancer were mainly observed in Sub-Saharan Africa and East Asia, which may be associated with the so-called African esophageal cancer corridor (30) and Asian esophageal cancer belt (31). In Africa, alcohol consumption, hot beverages, dental fluorosis, and oral health appear to be associated with the high disease burden (32,33). Moreover, the ESCC, typically occurring in the upper and middle parts of the esophagus, comprises the majority of cases in this region (4). Compared to adenocarcinoma, it is more likely to be affected by environmental and lifestyle factors (34). As for LOEC, Malawi, which is located in Eastern Sub-Saharan Africa, exhibited the highest ASRs in 2021. Consistent with other research, this elevated risk could be attributed to nutritional, lifestyle, and infectious factors (35). Similarly, Asia continued to face the most severe disease burden due to its high population density and socio-economic factors. In recent decades, the epidemiology of esophageal cancer in the Asia belt has undergone changes, with shift towards adenocarcinoma, which is closely associated with obesity (36). Accordingly, preventive strategies can be implemented to encourage individuals to adopt healthier lifestyle habits such as taking physical exercise, reducing alcohol consumption, getting enough fruits and vegetables (37). Considering the great differences in pathophysiology, risk factors, survival and epidemiological trends between EAC and ESCC, future research should emphasize outcomes associated with histological subtypes (4).

Considering sociodemographic development, higher disease burden was found in lower SDI countries with the cross-country inequality analysis. This suggested that despite significant advancements in the medical management for esophageal cancer, these developments appeared to be predominantly concentrated in high-income, developed countries (38). The incidence of esophageal cancer was higher in developing countries, commonly due to alcohol and tobacco consumption, poor nutrition, diet low in fruits and vegetables and exposure to environmental risk factors (39). In addition, because of the inadequate healthcare resources in these countries, treatment options primarily rely on surgery, thereby exacerbating the disease burden. Frontier analysis also demonstrated the potential improvement space for each country and territory. Guinea, a country with low SDI located in Western Sub-Saharan Africa, showed relatively small effective difference for three age groups, which meant excellent performance in reducing the disease burden (Figure 4). With limited resources, Guinea have made significant achievements in managing the disease burden of esophageal cancer, highlighting the effectiveness of its policies and clinical practice (30). Conversely, some countries and territories with relatively high SDI such as United Kingdom, Ireland and Monaco performed poorly, showing large effective differences for their development level. Current research on esophageal cancer were conducted mainly in high-income countries. In the future, low-income countries should be supported and encouraged to enhance data collection and analysis to improve global statistical outcomes (16). It is noteworthy that the COVID-19 pandemic had an impact on the epidemiological trends of esophageal cancer. During the pandemic, the limited healthcare resources, patients’ concerns about infection and the preventive measures implemented by hospitals resulted in delays in the diagnosis and treatment of esophageal cancer, particularly the surgical resection (40). Patients presented with more advanced stages and transition to palliative care, which subsequently impacted their prognosis (41). The multidisciplinary management of esophageal cancer was disproportionally affected in low-income and middle-income countries (42). The COVID-19 pandemic likely disrupted diagnosis and care for esophageal cancer, potentially leading to underestimation of incidence and mortality (43). This limitation should be considered when interpreting the most recent data. However, as the healthcare systems gradually return to pre-pandemic management model, the disease burden of esophageal cancer is expected to undergo changes in the coming years.

The results of this study also showed the gender differences in the burden of esophageal cancer. For instance, the ASMR and ASDR of the LOEC in males exhibited a consistent upward trend from 2016 to 2021 while those in females both showed a significant decline from 2014 to 2021. These results were consistent with the previous studies, which investigated the protective effects of estrogen against the esophageal cancer risk in women (44,45). The expression of estrogen receptor beta in patients with esophageal cancer is lower than healthy controls, which may indicate new therapeutic avenues in the future (46). Besides, due to the economic and socio-cultural differences, women generally exhibit a stronger awareness of health issues compared to men (47). They typically pay more attention to their health status and regular check-ups, which may lead them to seek medical care earlier when symptoms associated with esophageal cancer arise. The projected ASRs from 2022 to 2035 indicated the gap between males and females appears to expand over time. From the perspective of lifestyle, this could be attributed to different exposure to risk factors (Figures S22,S23). The prevalence of smoking and alcohol drinking is higher in men, which could increase the risk of esophageal squamous-cell carcinoma (48). Future efforts to promote lifestyle changes should also take gender differences into account and implement targeted strategies to achieve greater benefits.

Furthermore, the ASIR of EOEC and LOEC were projected to increase globally for both sexes combined from 2022 to 2035 with the BAPC model. The prediction results aligned with previous studies. A population-based cohort study involving 59,584 patients conducted in The Netherlands has indicated an increased incidence of EAC among adults under the age of 50 years, while the incidence of ESCC in this age group has shown a declining trend (49). Another study based on SEER database in the US has found that the incidence of early-onset EAC (younger than 50 years) is increasing, along with more advanced stages and poorer prognosis (50). Therefore, there is an urgent need for strategic planning to enhance the capacity of the healthcare workforce to effectively address the projected increases. We should prioritize early diagnosis and treatment of esophageal cancer, as well as primary prevention, in order to mitigate the overall disease burden. Early detection of esophageal cancer, particularly precancerous lesions including Barrett’s esophagus and squamous dysplasia, primarily relies on endoscopy (25). Recently, artificial intelligence systems have also been used to assist in endoscopic diagnosis, which could significantly improve the detection rate of early-stage ESCC and demonstrate good safety (51). Besides, the results also indicated that the risk factors varied in different age groups (Figure 7). Environmental and behavioral risk factors were closely associated with increased ESCC incidence risk among young individuals in Tanzania (6). A pooled analysis also revealed that gastroesophageal reflux disease and obesity were stronger risk factors for early-onset EAC compared to older age groups (52). Thus, future attention should be focused on individualized prevention strategies designed for different age groups.

There are several limitations in this study. Firstly, the estimates provided by the GBD 2021 could be affected by the accuracy and completeness of data sources across different countries and regions. Particularly in some low-income countries, the epidemiological data may be inadequate, potentially resulting in underestimation of the true burden. Secondly, the GBD data did not contain information about the histological subtypes; therefore, our findings only represent the general overview of the disease burden. Given the predominant subtype varied in countries and regions, the results should be interpreted with caution. This gap also underscores the need for more detailed, subtype-specific research to better understand these trends. Thirdly, the methodology and techniques of GBD relied on various rigorous assumptions and modeling approaches. But the variations in esophageal cancer management, including diagnosis and reporting, across different regions may still introduce biases that could affect the certainty of the estimates.

Conclusions

In summary, we provided updated estimates of the global burden of early, middle, and late-onset esophageal cancer, which is beneficial for developing practical policy and implementing effective interventions. The findings revealed a general decline in ASRs for EOEC and MOEC between 1990 and 2021, contrasted by a recent upward trend in the disease burden of LOEC. Further high-quality epidemiological studies and surveillance programs are needed to understand these trends and guide targeted interventions.

Acknowledgments

We are extremely grateful for the comprehensive and systematic work by the Global Burden of Disease study 2021 collaborators.

Footnote

Reporting Checklist: The authors have completed the GATHER reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-782/rc

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-782/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-782/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2024;74:229-63. [Crossref] [PubMed]
2. Deboever N, Jones CM, Yamashita K, et al. Advances in diagnosis and management of cancer of the esophagus. BMJ 2024;385:e074962. [Crossref] [PubMed]
3. Yang H, Wang F, Hallemeier CL, et al. Oesophageal cancer. Lancet 2024;404:1991-2005. [Crossref] [PubMed]
4. Morgan E, Soerjomataram I, Rumgay H, et al. The Global Landscape of Esophageal Squamous Cell Carcinoma and Esophageal Adenocarcinoma Incidence and Mortality in 2020 and Projections to 2040: New Estimates From GLOBOCAN 2020. Gastroenterology 2022;163:649-658.e2. [Crossref] [PubMed]
5. Sawas T, Manrique GC, Iyer PG, et al. Young Adults With Esophageal Adenocarcinoma Present With More Advanced Stage Tumors and Have Shorter Survival Times. Clin Gastroenterol Hepatol 2019;17:1756-62. [Crossref] [PubMed]
6. Buckle GC, Mmbaga EJ, Paciorek A, et al. Risk Factors Associated With Early-Onset Esophageal Cancer in Tanzania. JCO Glob Oncol 2022;8:e2100256. [Crossref] [PubMed]
7. Shi M, Tang JW, Cao ZR. Nomograms for predicting survival in early-onset esophageal cancer. Expert Rev Gastroenterol Hepatol 2021;15:437-46. [Crossref] [PubMed]
8. Kolb JM, Han S, Scott FI, et al. Early-Onset Esophageal Adenocarcinoma Presents With Advanced-Stage Disease But Has Improved Survival Compared With Older Individuals. Gastroenterology 2020;159:2238-2240.e4. [Crossref] [PubMed]
9. Chen R, Dou L, Zhou J, et al. Optimal starting age of endoscopic screening for esophageal cancer in China: A multicenter prospective cohort study. Cancer Med 2023;12:9988-98. [Crossref] [PubMed]
10. Hankey BF, Ries LA, Edwards BK. The surveillance, epidemiology, and end results program: a national resource. Cancer Epidemiol Biomarkers Prev 1999;8:1117-21.
11. Teng Y, Xia C, Cao M, et al. Esophageal cancer global burden profiles, trends, and contributors. Cancer Biol Med 2024;21:656-66. [Crossref] [PubMed]
12. Tang X, Wang P, Huang S, et al. Trend of gastrointestinal and liver diseases in China: Results of the Global Burden of Disease Study, 2019. Chin Med J (Engl) 2024;137:2358-68. [Crossref] [PubMed]
13. Wu G, Wu Q, Xu J, et al. Mortality burden and future projections of major risk factors for esophageal cancer in China from 1990 to 2019. Gen Thorac Cardiovasc Surg 2024;72:192-201. [Crossref] [PubMed]
14. Global burden and strength of evidence for 88 risk factors in 204 countries and 811 subnational locations, 1990-2021: a systematic analysis for the Global Burden of Disease Study 2021. Lancet 2024;403:2162-203. [Crossref] [PubMed]
15. Stevens GA, Alkema L, Black RE, et al. Guidelines for Accurate and Transparent Health Estimates Reporting: the GATHER statement. Lancet 2016;388:e19-23. [Crossref] [PubMed]
16. Global incidence, prevalence, years lived with disability (YLDs), disability-adjusted life-years (DALYs), and healthy life expectancy (HALE) for 371 diseases and injuries in 204 countries and territories and 811 subnational locations, 1990-2021: a systematic analysis for the Global Burden of Disease Study 2021. Lancet 2024;403:2133-61. [Crossref] [PubMed]
17. Global Burden of Disease Collaborative Network. Global Burden of Disease Study 2021 (GBD 2021) Socio-Demographic Index (SDI) 1950–2021. Seattle, United States of America: Institute for Health Metrics and Evaluation (IHME); 2024. Available online: https://ghdx.healthdata.org/record/global-burden-disease-study-2021-gbd-2021-socio-demographic-index-sdi-1950%E2%80%932021
18. Ugai T, Sasamoto N, Lee HY, et al. Is early-onset cancer an emerging global epidemic? Current evidence and future implications. Nat Rev Clin Oncol 2022;19:656-73. [Crossref] [PubMed]
19. Kim HJ, Fay MP, Feuer EJ, et al. Permutation tests for joinpoint regression with applications to cancer rates. Stat Med 2000;19:335-51. [Crossref] [PubMed]
20. Pan H, Zhao Z, Deng Y, et al. The global, regional, and national early-onset colorectal cancer burden and trends from 1990 to 2019: results from the Global Burden of Disease Study 2019. BMC Public Health 2022;22:1896. [Crossref] [PubMed]
21. Jiang CY, Han K, Yang F, et al. Global, regional, and national prevalence of hearing loss from 1990 to 2019: A trend and health inequality analyses based on the Global Burden of Disease Study 2019. Ageing Res Rev 2023;92:102124. [Crossref] [PubMed]
22. Riebler A, Held L. Projecting the future burden of cancer: Bayesian age-period-cohort analysis with integrated nested Laplace approximations. Biom J 2017;59:531-49. [Crossref] [PubMed]
23. Wagner B, Cleland K. Using autoregressive integrated moving average models for time series analysis of observational data. BMJ 2023;383:2739. [Crossref] [PubMed]
24. Jiang Y, Lin Y, Wen Y, et al. Global trends in the burden of esophageal cancer, 1990-2019: results from the Global Burden of Disease Study 2019. J Thorac Dis 2023;15:348-64. [Crossref] [PubMed]
25. Yang H, Hu B. Recent advances in early esophageal cancer: diagnosis and treatment based on endoscopy. Postgrad Med 2021;133:665-73. [Crossref] [PubMed]
26. Chen R, Zheng R, Zhang S, et al. Patterns and trends in esophageal cancer incidence and mortality in China: An analysis based on cancer registry data. J Natl Cancer Cent 2023;3:21-7. [Crossref] [PubMed]
27. Li J, Sun K, Zhu Q, et al. Temporal Trends and Sex Differences in the Incidence of Esophageal Squamous Cell Carcinoma and Adenocarcinoma from CI5 VIII-XII Data - Global, 1993-2017. China CDC Wkly 2025;7:491-6. [Crossref] [PubMed]
28. Maret-Ouda J, Markar SR, Lagergren J. Gastroesophageal Reflux Disease: A Review. JAMA 2020;324:2536-47. [Crossref] [PubMed]
29. Zhang D, Liu S, Li Z, et al. Global, regional and national burden of gastroesophageal reflux disease, 1990-2019: update from the GBD 2019 study. Ann Med 2022;54:1372-84. [Crossref] [PubMed]
30. Van Loon K, Mwachiro MM, Abnet CCThe African Esophageal Cancer Consortium, et al. A Call to Action. J Glob Oncol 2018;4:1-9. [Crossref] [PubMed]
31. Jin W, Huang K, Ding Z, et al. Global, regional, and national burden of esophageal cancer: a systematic analysis of the Global Burden of Disease Study 2021. Biomark Res 2025;13:3. [Crossref] [PubMed]
32. Menya D, Maina SK, Kibosia C, et al. Dental fluorosis and oral health in the African Esophageal Cancer Corridor: Findings from the Kenya ESCCAPE case-control study and a pan-African perspective. Int J Cancer 2019;145:99-109. [Crossref] [PubMed]
33. Middleton DRS, Mmbaga BT, Menya D, et al. Alcohol consumption and oesophageal squamous cell cancer risk in east Africa: findings from the large multicentre ESCCAPE case-control study in Kenya, Tanzania, and Malawi. Lancet Glob Health 2022;10:e236-45. [Crossref] [PubMed]
34. van den Brandt PA. The impact of a healthy lifestyle on the risk of esophageal and gastric cancer subtypes. Eur J Epidemiol 2022;37:931-45. [Crossref] [PubMed]
35. Geßner AL, Borkowetz A, Wilhelm TJ, et al. Risk factors for esophageal cancer in a high-incidence area of Malawi. Cancer Causes Control 2021;32:1347-54. [Crossref] [PubMed]
36. Grille VJ, Campbell S, Gibbs JF, et al. Esophageal cancer: the rise of adenocarcinoma over squamous cell carcinoma in the Asian belt. J Gastrointest Oncol 2021;12:S339-49. [Crossref] [PubMed]
37. Kubo Y, Kitagawa Y, Miyazaki T, et al. The potential for reducing alcohol consumption to prevent esophageal cancer morbidity in Asian heavy drinkers: a systematic review and meta-analysis. Esophagus 2022;19:39-46. [Crossref] [PubMed]
38. Mwachiro M, White R. Management of Esophageal Cancer Treatment in Resource-Limited Settings. Thorac Surg Clin 2022;32:397-404. [Crossref] [PubMed]
39. Hull R, Mbele M, Makhafola T, et al. A multinational review: Oesophageal cancer in low to middle-income countries. Oncol Lett 2020;20:42. [Crossref] [PubMed]
40. Doeve BH, Bakx JAC, Siersema PD, et al. The impact of the COVID-19 pandemic on the diagnosis, stage, and treatment of esophagogastric cancer. J Gastroenterol 2023;58:965-77. [Crossref] [PubMed]
41. Baxter MA, Khan KS, Gall LS, et al. Diagnosis, treatment, and outcome of patients with oesophagogastric cancer during the COVID-19 pandemic: national study. Br J Surg 2023;110:456-61. [Crossref] [PubMed]
42. Fialka NM, El-Andari R, Jogiat U, et al. Impact of the COVID-19 pandemic on esophageal cancer resource allocation: a systematic review. J Thorac Dis 2024;16:1576-89. [Crossref] [PubMed]
43. Global age-sex-specific mortality, life expectancy, and population estimates in 204 countries and territories and 811 subnational locations, 1950-2021, and the impact of the COVID-19 pandemic: a comprehensive demographic analysis for the Global Burden of Disease Study 2021. Lancet 2024;403:1989-2056. [Crossref] [PubMed]
44. Wang BJ, Zhang B, Yan SS, et al. Hormonal and reproductive factors and risk of esophageal cancer in women: a meta-analysis. Dis Esophagus 2016;29:448-54. [Crossref] [PubMed]
45. Brusselaers N, Maret-Ouda J, Konings P, et al. Menopausal hormone therapy and the risk of esophageal and gastric cancer. Int J Cancer 2017;140:1693-9. [Crossref] [PubMed]
46. Peri S, Niv Y. Estrogen receptor beta (ERβ) in esophageal cancer - a systematic review and meta-analysis. Scand J Gastroenterol 2024;59:1178-83. [Crossref] [PubMed]
47. Ketor CE, McDonnell J, Benneh CK, et al. Determinants of health-seeking behaviour in rural district of Jasikan, Ghana: a cross-sectional study. BMC Public Health 2024;24:2972. [Crossref] [PubMed]
48. Henick BS, Taylor AM, Nakagawa H, et al. Squamous cell cancers of the aero-upper digestive tract: A unified perspective on biology, genetics, and therapy. Cancer Cell 2025;43:178-94. [Crossref] [PubMed]
49. Al-Kaabi A, Baranov NS, van der Post RS, et al. Age-specific incidence, treatment, and survival trends in esophageal cancer: a Dutch population-based cohort study. Acta Oncol 2022;61:545-52. [Crossref] [PubMed]
50. Codipilly DC, Sawas T, Dhaliwal L, et al. Epidemiology and Outcomes of Young-Onset Esophageal Adenocarcinoma: An Analysis from a Population-Based Database. Cancer Epidemiol Biomarkers Prev 2021;30:142-9. [Crossref] [PubMed]
51. Yuan XL, Liu W, Lin YX, et al. Effect of an artificial intelligence-assisted system on endoscopic diagnosis of superficial oesophageal squamous cell carcinoma and precancerous lesions: a multicentre, tandem, double-blind, randomised controlled trial. Lancet Gastroenterol Hepatol 2024;9:34-44. [Crossref] [PubMed]
52. Saha B, Verma A, Dierkhising R, et al. Barrett’s Esophagus Risk Factors in Patients Without Gastroesophageal Reflux: A Large Population-Based Study. Clin Gastroenterol Hepatol 2024;22:2150-2152.e2. [Crossref] [PubMed]