The GBD 2019 dataset incorporated epidemiologic metrics for 369 pathological conditions and 87 risk factors (available at http://ghdx.healthdata.org/gbd-results-tool). These estimates encompassed 204 geopolitical entities and 21 subregions from 1990 to 2019 [11]. The SDI was estimated for each country according to education levels, birth rate, and per capita income. Using quintile stratification based on SDI metrics, global territories were categorized into five tiers: high, high-middle, middle, low-middle, and low-SDI regions. The requirements for ethical approval and informed consent were waived due to the public availability of the GBD 2019 database and the absence of identifiable information in the analyses.

AYA breast cancer is commonly defined as a diagnosis of breast cancer occurring between the ages of 15 and 39 years [12]. The data comprised information on prevalence, incidence, death, disability-adjusted life years (DALYs), and corresponding age-standardized rate (ASR), which were examined according to region, country, age, and year groups. DALYs were calculated by combining two components: years of life lost (YLLs) and years lived with disability (YLDs). YLDs were estimated as the product of disease prevalence and the associated disability weight for each health outcome, while YLLs were quantified by measuring deaths against a reference longevity standard. ASRs are widely used to enhance the accuracy of disease-burden comparisons as they consider age variations.

The truncated ASRs and corresponding uncertainty intervals (UIs) were calculated against the world standard population distribution (World Health Organization (WHO) 2000–2025) [13]. The ASR for AYA breast cancer was computed through multiplicative weighting of age-stratified incidence (aᵢ) against corresponding demographic strata weights (wᵢ) from the reference population [14]:

$$ASR=\frac{{\sum }_{i=1}^A{a}_i{\mathrm{w}}_i}{{\sum }_{i=1}^A{\mathrm{w}}_i}\times 100,000$$

Temporal patterns were assessed by deriving the estimated annual percentage change (EAPC) and the corresponding 95% confidence interval (CI) using linear regression modeling. Two formulas were used:

$$\ln (ASR)=\alpha +\beta X+\epsilon$$ $$EAPC=100\times (\exp (\beta )-1)$$

where X is the calendar year (time variable), $\beta$ is the estimated slope coefficient, and $\epsilon$ is an error term.

Increasing ASR trends were inferred from EAPC values greater than zero, which were supported by positive lower bounds of the CIs. Negative values for EAPCs and upper limits of the CIs suggested a downward trend. Pearson’s correlation analysis was carried out to quantify the linear association between ASRs and SDI. All statistics and visualization were performed using R version 4.1.2 (Vienna, Austria, https://www.r-project.org). Several R packages were used, including tidyverse for data manipulation, ggplot2 and patchwork for visualization, epitools for epidemiological analysis, and ggmaps for geospatial mapping. Statistical significance was defined by p-values below 0.05.

The age-standardized prevalence rate (ASPR) of breast cancer among AYAs was 94.8 per 100,000 population in 2019. Similarly, the age-standardized incidence rate (ASIR) was documented as 11.29 per 100,000 population (Table 1, Supplementary Table 1). The EAPCs of ASPR and ASIR were 0.67 (95% CI: 0.59 to 0.75) and 0.60 (95% CI: 0.50 to 0.69), respectively, indicating an increasing trajectory over the last 30 years (Table 1, Supplementary Table 1, and Fig. 1). Notably, however, the age-standardized mortality rate (ASMR) declined from 3.02 (95% UI: 2.78–3.26) in 1990 to 2.86 (95% UI: 2.58–3.17) in 2019, and the EAPC was –0.41 (95% CI: –0.53 to –0.30) (Table 2, Fig. 1). The age-standardized DALY rate (ASDR) was 165.2 (95% UI: 148.54–182.26) in 2019, and the EAPC was –0.34 (95% CI: –0.46 to –0.23), indicating a declining trend (Supplementary Table 2, Fig. 1).

Fig. 1.

The EAPC of age-standardized prevalence, incidence, death and DALYs rates for breast cancer in adolescents and young adults. DALYs, disability-adjusted life years.

High-SDI regions had the highest ASPR (152.13, 95% UI: 133.95–171.47) and ASIR (17.04, 95% UI: 15.01–19.25) for AYA breast cancer in 2019 (Table 1, Supplementary Table 1). The most rapid increase in ASPR and ASIR occurred in the middle-SDI regions (EAPC: 1.77 for ASPR; EAPC: 1.57 for ASIR) (Fig. 1). In the high-SDI regions, both ASIR and ASPR decreased gradually over time, which is supported by negative values of EAPC (–0.14 for ASPR; –0.19 for ASIR). Nevertheless, the ASMR (3.45, 95% UI: 2.86–4.13) and ASDR (197.24, 95% UI: 163.26–236.30) were highest in the low-middle SDI regions (Table 2, Supplementary Table 2). The EAPCs of ASMR and ASDR were positive values in the low-middle and low SDI regions, indicating a rising trend over time (Fig. 1). The results indicated a declining trend in ASMR and ASDR across the high, high-middle, and middle-SDI regions.

With respect to the GBD region, Oceania, Australasia, and Western Europe suffered the highest ASPR and ASIR (Table 1, Supplementary Table 1). The ASPR and ASIR were lowest in Eastern Sub-Saharan Africa, Andean Latin America, and Western Sub-Saharan Africa. An overall upward trajectory in ASPR and ASIR was evident globally except in five regions: Western Europe, high-income North America, Central Asia, Australasia, and Southern Sub-Saharan Africa (Fig. 1). The GBD region with the highest ASMR and ASDR was identified as Oceania (ASMR: 11.36, 95% UI, 7.73–16.22; ASDR: 637.79, 95% UI, 433.84–911.35) (Table 2, Supplementary Table 2). Andean Latin America, high-income Asia Pacific, and East Asia exhibited the lowest ASMR and ASDR for AYA breast cancer. Conversely, the most substantial increases in ASMR and ASDR were observed in North Africa and the Middle East, South Asia, and East Asia (Fig. 1).

The burden of AYA breast cancer exhibited considerable national-level disparities across the 204 countries. Pakistan, India, China, the United States of America, and Indonesia were recorded as the five foremost countries with the highest prevalence and incidence in 2019 (Supplementary Tables 3,4). Higher ASIR and ASPR were observed in Oceania (Solomon Islands, Cook Islands, Nauru, Palau, and Niue) and Western Europe (Monaco, San Marino, Netherlands, United Kingdom, France, and Portugal) (Supplementary Fig. 1A, Supplementary Fig. 1B). Saint Kitts and Nevis, Myanmar, and Kyrgyzstan had the lowest EAPC values for both ASPR and ASIR, which were all less than –2.0 (Supplementary Tables 3,4, and Supplementary Fig. 2A,B). The highest EAPCs for ASPR and ASIR were observed in North Africa and the Middle East (Saudi Arabia, Turkey, Yemen, Oman, and Lebanon), as well as Southern Sub-Saharan Africa (Botswana and Namibia).

The highest DALYs and death rates were documented in India, China, Indonesia, and Pakistan (Supplementary Tables 5,6). Pakistan, Indonesia, and some Oceania countries (Solomon Islands, Kiribati, Nauru, Fiji, Papua New Guinea, and Marshall Islands) had relatively high ASMR and ASDR for AYA breast cancer (Supplementary Fig. 1C,D). Out of the 204 countries, Solomon Island, Zimbabwe, Lesotho, Jamaica, and Yemen demonstrated the largest growth in ASMR and ASDR (Supplementary Tables 5,6, and Supplementary Fig. 2C,D). The EAPCs for these countries all exceeded 1.90.

The risk factors influencing DALYs and deaths varied across different GBD regions (Fig. 2). The burden of DALYs attributable to tobacco was disproportionately concentrated in East Asia (10.82%), high-income Asia Pacific regions (12.02%), and Southern Sub-Saharan Africa (12.89%). North Africa and the Middle East, Western Europe, and Eastern Sub-Saharan Africa displayed the greatest percentages of the DALYs attributed to alcohol use. Elevated fasting glucose accounted for the highest proportion of DALYs in Southern Sub-Saharan Africa. Low physical activity was the primary driver of DALYs in East Asia and Oceania. Both ASDR and ASMR of AYA breast cancer related to increased fasting glucose showed an increasing trend in all SDI regions (Fig. 3A–C, Supplementary Fig. 3A–C, and Supplementary Tables 7,8). In addition, there was an upward trend for ASDR attributed to all risk factors in low-SDI regions.

Fig. 2.

The proportion of AYA breast cancer deaths and DALYs attributable to risk factors among the GBD regions in 2019. DALY, disability adjusted life-year; FPG, fasting plasma glucose; LPA, low physical activity.

Fig. 3.

The age-standardized DALYs rate of AYA breast cancer attributable to risk factors across distinct SDI regions in 1990 and 2019 with corresponding EAPC. (A) The age-standardized DALYs rate in 1990; (B) age-standardized DALYs rate in 2019; (C) EAPC in 1990–2019.

The global prevalence, incidence, death rate, and DALY rate exhibited upward trends across the five subgroups of age (Supplementary Fig. 4). The prevalence and incidence rate increased gradually with the increase of SDI in the age subgroups of 25–29, 30–34, and 35–39 years. In contrast, for the age subgroups of 15–19 and 20–24 years, the highest incidence and prevalence rates of AYA breast cancer were identified in regions with low-middle SDI. The patterns of death and DALY rates among various SDI regions were similar among the five age subgroups. The low-middle and low SDI regions exhibited the greatest rates of both DALYs and deaths.

The trends of the four ASRs across SDI regions are presented in Supplementary Fig. 5. Higher SDI levels were significantly associated with increased ASIR and ASPR (Supplementary Fig. 6A,B). The results indicated a positive correlation between SDI and ASMR in the GBD regions where the SDI was less than 0.4 (Supplementary Fig. 6C). The ASDR pattern related to SDI was similar to that of ASMR (Supplementary Fig. 6D). An inverse relationship was found between EAPC of ASRs and SDI (r: –0.40, p = 0.06 for ASMR; r: –0. 46, p 0.05 for ASDR; r: –0.81, p 0.01 for ASPR; r: –0.83, p 0.01 for ASIR) (Supplementary Fig. 7).