Persistent Racial Disparities in Breast Cancer Incidence by Stage and Subtype in a High‐Burden State Authors: Rifath Ara Alam Barsha, Jasmine Miller‐Kleinhenz Published: 2026 ABSTRACT ABSTRACTIntroductionBreast cancer incidence varies by race, stage at diagnosis, and molecular subtype. However, these patterns are often obscured in overall estimates, particularly in high‐burden settings. This study examined incidence trends overall and by race, stage at diagnosis, and molecular subtype among women in Mississippi.MethodsWe used population‐based data from the Mississippi Cancer Registry, including Black and White women aged ≥ 18 years diagnosed with primary breast cancer from 2010 to 2019. Age‐adjusted incidence rates (IRs) per 100,000 population were standardized to the 2000 United States standard population. Trends were assessed using Joinpoint regression estimating average annual percent changes (AAPCs) and annual percent changes (APCs). Multivariable Poisson regression estimated incidence rate ratios (IRRs) with 95% confidence intervals (CIs).ResultsAmong 21,130 women (64.4% White; 35.6% Black), the overall age‐adjusted IR was 166.0 per 100,000 with no significant change over time (AAPC = 0.48%; 95% CI: −0.27 to 1.24). Black women had higher incidence than White women (178.0 vs. 159.1). Among White women, incidence increased modestly (AAPC = 0.60%; 95% CI: 0.06–1.13). Trends among Black women were not significant overall (AAPC = 0.36%; 95% CI: −0.25 to 1.04) but increased from 2014 to 2019 (APC = 1.95%; 95% CI: 0.77–4.26) after an initial decline. Localized‐stage incidence increased in both groups, while Black women had higher regional‐ and distant‐stage incidence. In adjusted models, Black women had higher incidence of HER2‐overexpressing (IRR = 1.55; 95% CI: 1.15–2.10) and triple‐negative cancer (IRR = 2.35; 95% CI: 1.74–3.17) than White women.ConclusionOverall incidence remained stable, but substantial heterogeneity across subgroups underscores the need for stratified surveillance and equity‐focused cancer control strategies in high‐burden settings.Keywords: breast cancer incidence, cancer surveillance, disparities, joinpoint regression, molecular subtype, stage at diagnosis Introduction Breast cancer is the most commonly diagnosed cancer among women in the United States (US), except for skin cancer, and remains a major public health concern [1]. It represents nearly one‐third of all new cancer diagnoses among women each year [1] and has increased by an average of 0.6% per year between 2013 and 2022 [2]. Molecular classification is essential for understanding breast cancer heterogeneity, prognosis, and treatment response. Breast cancer is categorized into four main subtypes based on hormone receptor (HR) and HER2 (Human Epidermal Growth Factor Receptor 2) status: HR‐positive/HER2‐negative, HR‐positive/HER2‐positive, HR‐negative/HER2‐positive, and HR‐negative/HER2‐negative [3]. These subtypes differ in clinical aggressiveness, availability of targeted therapies, and responsiveness to treatment [4, 5], with HR‐positive/HER2‐negative being the most common [6]. Stage at diagnosis also remains a critical determinant of breast cancer outcomes, with early‐stage cancers associated with better survival than regional or distant stage [2, 7]. Additionally, prior research has documented disparities in both stage and molecular subtype across racial groups [8, 9], reflecting complex interactions between tumor biology and socio‐structural determinants of health. Investigating incidence patterns by stage and molecular subtype is therefore critical for understanding breast cancer epidemiology and informing early detection and prevention strategies. Reliance on national or aggregate estimates may obscure meaningful heterogeneity across populations and geographic settings, particularly when opposing trends occur across subpopulations. Mississippi represents a high‐burden setting in which such patterns may be especially pronounced. Among women in Mississippi, breast cancer has the highest incidence of any cancer type [10] and consistently ranks among the highest in breast cancer mortality in the US [11]. Moreover, unlike national patterns, breast cancer incidence is higher among Black women than White women in Mississippi [12]. These differences highlight the need to examine incidence trends by race alongside tumor characteristics such as stage at diagnosis and molecular subtype. To our knowledge, no prior study has comprehensively examined long‐term breast cancer incidence trends in Mississippi across these dimensions using population‐based data from the Mississippi Cancer Registry (MCR), highlighting the need for approaches that move beyond overall trends to better characterize heterogeneity. Therefore, to address existing gaps in knowledge, this study examined breast cancer incidence trends overall and by race, stage at diagnosis, and molecular subtype among women in Mississippi from 2010 to 2019. Given Mississippi's demographic, socioeconomic, and health profile, such an investigation is critical for identifying populations at greatest risk and guiding state‐specific cancer control strategies. As a high‐burden setting, Mississippi may also offer insights relevant to other settings and inform more targeted, equity‐focused interventions. Materials and Methods Data Source We analyzed data from the Mississippi Cancer Registry. This statewide population‐based registry has collected information on all newly diagnosed cancer cases in Mississippi since 1996 [13]. The MCR has rigorous quality control processes and holds gold certification from the North American Association of Central Cancer Registries (NAACCR), ensuring the reliability and validity of its information [13]. Data for this study included all female breast cancer cases diagnosed between 2010 and 2019. Ethical Considerations The study used de‐identified, population‐based registry data provided under a Data Use Agreement with the MCR. Ethical approval for secondary data analysis was obtained from the University of Mississippi Medical Center Institutional Review Board (IRB) (IRB # UMMC‐IRB‐2025‐6). The requirement for informed consent was waived because the study used de‐identified registry data. Study Population The study included Black and White women aged 18 years and older diagnosed with primary breast cancer between 2010 and 2019. The study period was limited to 2010–2019 to avoid potential disruptions in breast cancer incidence patterns associated with the COVID‐19 pandemic, which substantially affected cancer screening and diagnosis beginning in 2020 [14, 15]. Men were excluded due to their lower overall national incidence (< 1%) [16] and because this study's focus is on breast cancer incidence among women. The final analytic sample consisted of 21,130 women. Study Measures Outcome Measure The primary outcome of the study was the age‐adjusted incidence rate (per 100,000 women) of breast cancer. Covariates of Interest Race: The analytic sample was restricted to women identified as Black or White. Women from other racial groups were excluded due to small case counts and limited statistical stability for rate estimation and trend analysis. Race was categorized as Black and White. Molecular Subtype: We defined molecular subtype based on the joint expression of HR and HER2 status. Tumors were grouped into the following molecular subtypes: luminal A (HR+/HER2−), luminal B (HR+/HER2+), HER2‐overexpressing (HR−/HER2+), triple‐negative breast cancer (TNBC; HR−/HER2−), and Unknown subtype. The unknown category included cases in which receptor status was unknown, could not be determined, or in which testing was ordered but results were not reported. Cases with ER‐negative/PR‐positive receptor patterns were excluded because this receptor combination is uncommon and may reflect testing variability or potential misclassification and cannot be reliably classified into standard molecular subtype categories. Stage at diagnosis: Stage at diagnosis was defined as in situ, localized, regional, distant, and unknown, based on standardized registry definitions. The unknown category included cases with unknown stage or unstaged breast cancer. Statistical Analyses Annual age‐adjusted incidence rates per 100,000 women were calculated using the direct standardization method, standardized to the 2000 US standard population [17]. Age‐adjusted rates were computed overall and stratified by race, stage at diagnosis, and molecular subtype. Incidence rate ratios (IRRs) with corresponding 95% confidence intervals (CIs) were estimated using three multivariable Poisson regression models. Model 1 adjusted for race, age, and year of diagnosis. Model 2 additionally included stage at diagnosis and a race and stage interaction term to assess racial differences by stage at diagnosis. Model 3 replaced stage at diagnosis with molecular subtype and included race and subtype interaction to evaluate whether racial differences varied by molecular subtype. Incidence rate calculations and regression analysis were performed in R version 4.4.3 (R Foundation for Statistical Computing). Temporal trends in incidence rates were evaluated using the National Cancer Institute's (NCI) Joinpoint Regression Program, Version 5.4.0.0 [18]. Joinpoint regression fits log‐linear models to annual age‐adjusted rates and identifies statistically significant changes in trend over time. Based on the number of years of available data, models were estimated, allowing a maximum of 1 joinpoint per model according to NCI Joinpoint guidelines. The final model was selected using the Weighted Bayesian Information Criterion (Weighted BIC) approach. For each model, the Annual Percent Change (APC) and corresponding 95% CIs were estimated. The average annual percent changes (AAPCs) were also calculated to summarize overall trends from 2010 to 2019. Confidence intervals were calculated using the empirical quantile method. APC and AAPC estimates were considered statistically significant if the two‐sided 95% CI did not include zero (α = 0.05). Separate Joinpoint models were fitted for race, stage at diagnosis, and molecular subtype to evaluate subgroup‐specific temporal trends. Results Descriptive Characteristics A total of 21,130 women aged 18 years and older were diagnosed with primary breast cancer in Mississippi between 2010 and 2019 (Table 1). Among these, 64.4% were White women and 35.6% were Black women. About half of the cases were diagnosed at the localized stage (49.5%), followed by regional (26.4%), in situ (16.3%), and distant‐stage cases (6.3%). Luminal A subtype was the most common subtype (53.1%), followed by TNBC (13.2%), luminal B (8.8%) and HER2‐overexpressing (4.1%). TABLE 1Trends in breast cancer incidence among women in Mississippi, 2010–2019.CharacteristicsTotal no. (%)Age‐adjusted incidence rates (per 100,000)Average annual percent change (AAPC) (95% CI)Overall21,130 (100)166.00.48 (−0.27, 1.24)RaceWhite13,608 (64.4)159.10.60* (0.06, 1.13)Black7522 (35.6)178.00.36 (−0.25, 1.04)Stage at diagnosisIn situ3449 (16.3)26.9−0.02 (−2.21, 2.17)Localized10,469 (49.5)81.41.60* (0.45, 2.76)Regional5577 (26.4)44.9−1.04 (−2.71, 0.59)Distant1331 (6.3)10.3−0.41 (−1.67, 1.82)Unknown304 (1.4)2.44−1.31 (−6.40, 3.93)Molecular subtypeLuminal A11,218 (53.1)87.11.12 (−0.18, 2.53)Luminal B1863 (8.8)15.10.10 (−4.48, 4.81)HER2‐overexpressing874 (4.1)6.95−2.11 (−4.89, 0.58)Triple negative2785 (13.2)22.71.00 (−0.53, 2.52)Unknown4390 (20.8)34.1−0.91 (−3.12, 1.34)Abbreviation: CI, confidence interval.*Indicates that the AAPC is statistically significant; 95% CI does not include 0 (2‐sided p < 0.05). Trends in breast cancer incidence among women in Mississippi, 2010–2019. Abbreviation: CI, confidence interval. Indicates that the AAPC is statistically significant; 95% CI does not include 0 (2‐sided p < 0.05). Overall Breast Cancer Incidence Rates and Trends The overall incidence was 166.0 per 100,000 women, with no statistically significant change over time (AAPC = 0.48%; 95% CI = −0.27 to 1.24) (Table 1). Black women had a higher age‐adjusted incidence rate (178.0 per 100,000) compared with White women (159.1 per 100,000). Over the study period, incidence among White women increased slightly but significantly (AAPC = 0.60%; 95% CI = 0.06–1.13; p < 0.05), whereas the trend among Black women was not statistically significant (AAPC = 0.36%; 95% CI = −0.25–1.04). Joinpoint analyses of race‐specific trends are shown in Figure 1. Among Black women, incidence decreased from 2010 to 2014 (APC = −1.59%; 95% CI = −3.75 to 0.02) and then increased significantly from 2014 to 2019 (APC = 1.95%; 95% CI = 0.77–4.26; p < 0.05). Among White women, incidence showed no significant change from 2010 to 2016 (APC = 0.34%; 95% CI = −1.19 to 1.89) or from 2016 to 2019 (APC = 1.28%; 95% CI = −3.21 to 5.97). Trends in age‐adjusted breast cancer incidence rates by race, Mississippi, 2010–2019. (A) Age‐adjusted breast cancer incidence trends among Black women; (B) age‐adjusted breast cancer incidence trends among White women. APC, annual percent change. * indicates that the APC is statistically significant; 95% CI does not include 0 (two‐sided p < 0.05). Regarding stage at diagnosis, the highest incidence rate was observed for localized disease (81.4 per 100,000), followed by regional disease (44.9 per 100,000) and in situ tumors (26.9 per 100,000) (Table 1). Distant‐stage cancers had an incidence rate of 10.3 per 100,000. Over the study period, localized‐stage incidence increased significantly (AAPC = 1.60%; 95% CI = 0.45–2.76; p < 0.05), while no significant temporal changes were observed for in situ, regional‐stage, and distant‐stage cancers. By molecular subtype, luminal A tumors had the highest incidence rate (87.1 per 100,000), followed by TNBC (22.7 per 100,000), luminal B (15.1 per 100,000), and HER2‐overexpressing (6.95 per 100,000) tumors (Table 1). Temporal trends for any of the subtypes were not statistically significant. Stage‐ and Subtype‐Specific Trends by Race Stage at Diagnosis Among White women, the highest incidence was observed for localized‐stage disease (83.9 per 100,000), followed by regional‐stage (39.9 per 100,000), in situ (24.9 per 100,000), and distant‐stage (8.06 per 100,000) (Table 2). Over the study period, localized‐stage incidence increased significantly (APC = 1.75%; 95% CI = 0.56–2.92; p < 0.05), whereas trends for other stages were not statistically significant. TABLE 2Trends in breast cancer incidence among women in Mississippi by race, stage at diagnosis, and molecular subtype, 2010–2019.CharacteristicsTotal no. (%)Age‐adjusted incidence rates (per 100,000)Annual percent change (APC) (95% CI)Stage at diagnosisWhiteIn situ2115 (15.5)24.92010–2019: −1.01 (−2.46, 0.41)Localized7306 (53.7)83.92010–2019: 1.75* (0.56, 2.92)Regional3294 (24.2)39.92010–2019: −0.65 (−1.99, 0.65)Distant701 (5.2)8.062010–2019: −1.49 (−6.27, 3.40)Unknown192 (1.4)2.292010–2019: 4.10 (−4.45, 13.33)BlackIn situ1334 (17.7)31.02010–2019: 1.34 (−0.24, 2.90)Localized3163 (42.1)74.62010–2019: 1.80* (0.34, 3.24)Regional2283 (30.4)54.4 2010–2013: −6.92* (−11.81, −0.30) 2013–2019: −0.51 (−3.73, 7.42) Distant630 (8.4)14.92010–2019: 1.53 (−4.35, 7.58)Unknown112 (1.5)2.912010–2019: −10.54* (−15.78, −5.17)Molecular subtypeWhiteLuminal A7797 (57.3)89.52010–2019: 1.49* (0.17, 2.82)Luminal B1196 (8.8)14.82010–2019: 1.09 (−4.13, 6.53)HER2‐overexpressing520 (3.8)6.172010–2019: −3.05 (−7.42, 1.34)Triple negative1372 (10.1)16.92010–2019: −0.39 (−2.16, 1.35)Unknown2723 (20.0)31.72010–2019: −0.84 (−3.90, 2.29)BlackLuminal A3421 (45.5)81.5 2010–2014: −3.44 (−8.50, 0.39) 2014–2019: 4.43* (1.41, 10.40) Luminal B667 (8.9)15.62010–2019: −1.14 (−5.24, 2.87)HER2‐overexpressing354 (4.7)8.312010–2019: −2.26 (−6.67, 2.30)Triple negative1413 (18.8)33.22010–2019: 2.19 (−0.05, 4.47)Unknown1667 (22.2)39.32010–2019: −1.23 (−3.39, 0.92)Abbreviation: CI, confidence interval.*Indicates that the APC is statistically significant; 95% CI does not include 0 (2‐sided p < 0.05). Trends in breast cancer incidence among women in Mississippi by race, stage at diagnosis, and molecular subtype, 2010–2019. 2010–2013: −6.92* (−11.81, −0.30) 2013–2019: −0.51 (−3.73, 7.42) 2010–2014: −3.44 (−8.50, 0.39) 2014–2019: 4.43* (1.41, 10.40) Abbreviation: CI, confidence interval. Indicates that the APC is statistically significant; 95% CI does not include 0 (2‐sided p < 0.05). Among Black women, localized‐stage incidence was 74.6 per 100,000, followed by regional‐stage (54.4 per 100,000) and in situ tumors (31.0 per 100,000) (Table 2). Distant‐stage cancers occurred at a higher rate (14.9 per 100,000) than among White women. Localized‐stage incidence increased significantly among Black women (APC = 1.80%; 95% CI = 0.34–3.24; p < 0.05). Regional‐stage incidence declined significantly between 2010 and 2013 (APC = −6.92%; 95% CI = −11.81 to 0.30; p < 0.05), followed by no significant change thereafter. Molecular Subtype Among White women, luminal A subtype had the highest incidence (89.5 per 100,000) and increased significantly over time (APC = 1.49%; 95% CI = 0.17–2.82; p < 0.05) (Table 2). Incidence rates for TNBC (16.9 per 100,000), luminal B (14.8 per 100,000), and HER2‐overexpressing subtypes (6.17 per 100,000) were lower, and none showed statistically significant trends. Among Black women, luminal A subtype also had the highest incidence rate (81.5 per 100,000) (Table 2). Trends showed a decline from 2010 to 2014, followed by a significant increase from 2014 to 2019 (APC = 4.43%; 95% CI = 1.41–10.40; p < 0.05). Triple‐negative breast cancer had a higher incidence rate among Black women (33.2 per 100,000) and showed a nonsignificant increasing trend (APC = 2.19%; 95% CI = −0.05 to 4.47). Race‐specific incidence and trends by stage and subtype are presented in Figures S1–S6. Multivariable Poisson Regression Models of Breast Cancer Incidence In Model 1, Black women had a higher incidence rate compared with White women; however, the association was not statistically significant (IRR = 1.13; 95% CI = 0.97–1.30) (Table 3). Age was strongly associated with breast cancer incidence, with rates increasing substantially across age groups relative to women aged 18–29 years. The highest incidence was observed among women aged 65–74 years (IRR = 86.30; 95% CI = 64.91–114.74; p < 0.005). TABLE 3Incidence rate ratios for breast cancer incidence by race, stage at diagnosis, and molecular subtype among women in Mississippi, 2010–2019.CharacteristicsModel 1Model 2Model 3IRR (95% CI)IRR (95% CI)IRR (95% CI)RaceWhiteRef.Ref.Ref.Black1.13 (0.97, 1.30)1.28 (0.92, 1.80)0.89 (0.72, 1.10)Age (years)18–29Ref.Ref.Ref.30–345.64*** (4.31, 7.37)5.64*** (4.39, 7.23)5.64*** (4.32, 7.35)35–4419.82*** (15.48, 25.38)19.82*** (15.78, 24.90)19.82*** (15.58, 25.22)45–5442.24*** (32.76, 54.47)42.24*** (33.41, 53.41)42.24*** (33.07, 53.96)55–6462.03*** (47.79, 80.51)62.03*** (48.76, 78.91)62.03*** (48.31, 79.64)65–7486.30*** (64.91, 114.74)86.30*** (66.41, 112.16)86.30*** (65.95, 112.93)75–8471.83*** (53.43, 96.55)71.83*** (54.75, 94.24)71.83*** (54.40, 94.84)85+47.86*** (35.87, 63.86)47.86*** (36.67, 62.47)47.86*** (36.56, 62.65)Year of diagnosis1.00 (0.97, 1.03)1.00 (0.98, 1.03)Stage at diagnosisIn situNARef.NALocalizedNA3.45*** (2.59, 4.60)NARegionalNA1.56** (1.19, 2.04)NADistantNA0.33*** (0.26, 0.43)NAUnknownNA0.09*** (0.07, 0.12)NAMolecular subtypeLuminal ANANARef.Luminal BNANA0.15*** (0.12, 0.19)HER2‐overexpressingNANA0.07*** (0.05, 0.08)Triple negativeNANA0.18*** (0.14, 0.22)UnknownNANA0.35*** (0.28, 0.44)Race × Stage at diagnosisBlack × In situNARef.NABlack × LocalizedNA0.67 (0.46, 1.02)NABlack × RegionalNA1.10 (0.75, 1.62)NABlack × DistantNA1.42 (0.98, 2.08)NABlack × UnknownNA0.92 (0.60, 1.44)NARace × Molecular subtypeBlack × Luminal ANANARef.Black × Luminal BNANA1.27 (0.95, 1.70)Black × HER2‐overexpressingNANA1.55** (1.15, 2.10)Black × Triple negativeNANA2.35*** (1.74, 3.17)Black × UnknownNANA1.40 (1.00, 1.95) Note: Model 1 adjusted for race, age group, and year of diagnosis. Model 2 additionally included stage at diagnosis and race × stage interaction terms. Model 3 replaced Stage and included molecular subtype and race × subtype interaction terms.Abbreviations: CI, confidence interval; IRR, incidence rate ratios.*** p < 0.001.** p < 0.01. Incidence rate ratios for breast cancer incidence by race, stage at diagnosis, and molecular subtype among women in Mississippi, 2010–2019. Note: Model 1 adjusted for race, age group, and year of diagnosis. Model 2 additionally included stage at diagnosis and race × stage interaction terms. Model 3 replaced Stage and included molecular subtype and race × subtype interaction terms. Abbreviations: CI, confidence interval; IRR, incidence rate ratios. p < 0.001. p < 0.01. After further adjustment for stage at diagnosis, Model 2 showed similar findings with Black women having a higher incidence, although the association remained nonsignificant (IRR = 1.28; 95% CI = 0.92–1.80) (Table 3). Compared with in situ tumors, incidence rates were significantly higher for localized‐stage (IRR = 3.45; 95% CI = 2.59–4.60; p < 0.005) and regional‐stage cancers (IRR = 1.56; 95% CI = 1.19–2.04; p < 0.01), whereas distant‐stage cancers had lower incidence rates (IRR = 0.33; 95% CI = 0.26–0.43; p < 0.005). Race‐by‐stage interaction terms were not statistically significant. When molecular subtype was added in Model 3, the direction of the association changed, with incidence among Black women lower than White women, and again the association was nonsignificant (IRR = 0.89; 95% CI = 0.72–1.10) (Table 3). Compared with luminal A tumors, incidence rates were significantly lower for luminal B (IRR = 0.15; 95% CI = 0.12–0.19; p < 0.005), HER2‐overexpressing (IRR = 0.07; 95% CI = 0.05–0.08; p < 0.005), and TNBC (IRR = 0.18; 95% CI = 0.14–0.22; p < 0.005) subtypes. The race‐by‐subtype interaction indicated that Black women had a higher incidence of HER2‐overexpressing (IRR = 1.55; 95% CI = 1.15–2.10; p < 0.01) and TNBC (IRR = 2.35; 95% CI = 1.74–3.17; p < 0.001) compared with White women. Discussion In this population‐based study, although the overall breast cancer incidence in Mississippi did not change significantly over the study period, important heterogeneity emerged when trends were examined by race, stage at diagnosis, and molecular subtype. Black women experienced higher age‐adjusted incidence rates than White women. However, the corresponding adjusted incidence rate ratio did not reach statistical significance, indicating that the overall racial difference in breast cancer incidence was modest after adjustment for age and year of diagnosis. This highlights the importance of interpreting descriptive incidence rates alongside adjusted analyses that account for differences in age and year of diagnosis. Black women also bore a disproportionate burden of aggressive subtypes, including TNBC and HER2‐overexpressing tumors. In addition, the incidence of localized‐stage disease increased over time, while disparities in later‐stage diagnosis persisted. Together, these findings demonstrate that stable overall incidence trends can obscure meaningful differences across population subgroups. The overall incidence pattern observed in this study is generally consistent with national trends, where breast cancer incidence slowly increased from the mid‐2000s, largely driven by an increased diagnosis of localized stage disease [19]. However, while national data indicate slightly lower breast cancer incidence among Black women than White women [8], we observed a higher incidence among Black women in Mississippi. This reversal underscores the importance of examining state‐level data, particularly in regions with elevated breast cancer mortality. This finding suggests that local contextual factors, such as socioeconomic conditions and access to healthcare, may influence breast cancer risk and detection. In addition, we observed a decline in breast cancer incidence among Black women between 2010 and 2014, followed by a significant increase from 2014 to 2019. Although the reasons for this pattern cannot be determined from the present study, it may reflect temporal changes in breast cancer screening, healthcare utilization, diagnostic practices, or other population‐level factors [20, 21]. Further research is needed to better understand the factors underlying these temporal trends. Stage‐specific trends provide additional insight into differences in breast cancer incidence. The increased incidence of localized‐stage cancers observed in this study may reflect improvements in early detection, such as greater use of mammography or increased awareness [19, 22]. Although these factors were beyond the scope of the present analysis, they represent plausible explanations supported by previous research [19, 22]. However, Black women continued to experience a higher incidence of regional‐ and distant‐stage cancers than White women, indicating persistent disparities in timely diagnosis. These findings are consistent with prior studies demonstrating that Black women are more likely to be diagnosed at later stages of breast cancer [23, 24, 25, 26]. These patterns suggest that gains in early detection may not be equitably distributed across populations, reinforcing the need for targeted interventions to improve timely diagnosis among underserved groups. In addition, structural barriers, including limited access to screening and health services [27, 28], transportation challenges [29, 30], and delays in follow‐up after abnormal findings [31, 32] may contribute to these disparities. Although these factors were not directly assessed in the present study, they provide plausible contextual explanations for the observed patterns. In Mississippi, where many residents live in medically underserved areas and access to primary care is limited [33], these barriers may be particularly relevant. Consistent with prior research [3, 34, 35], luminal A was the most common subtype among diagnosed breast cancer cases in Mississippi. However, important racial differences were observed for molecular subtype. Black women had a higher incidence of TNBC and HER2‐overexpressing breast cancer compared with White women. These subtypes are associated with more aggressive disease, poorer prognosis, and fewer treatment options [3, 34]. Previous studies have suggested that differences in tumor biology [36, 37], metabolic factors [38, 39], reproductive patterns [40, 41], and social determinants of health [23, 42, 43] may contribute to these disparities. The higher burden of aggressive subtypes may help explain the persistently higher breast cancer mortality rates in Mississippi. This study highlights how population‐level stability in cancer incidence trends may conceal important disparities in disease burden. These findings underscore the importance of examining disaggregated cancer surveillance data to better identify high‐risk populations and assess progress in cancer control. Although this study was conducted in Mississippi, a high‐burden state with unique demographic and healthcare characteristics, the analytic approach used here may be informative for other population‐based cancer registries seeking to identify subgroup‐specific disparities. However, the specific incidence patterns observed should not be assumed to apply directly to other states or populations. Routine application of a similar stratified surveillance approach may strengthen cancer surveillance by improving the detection of disparities, monitoring progress, and informing targeted cancer control strategies. Future research should extend this framework by incorporating multilevel measures of geographic and socioeconomic context, including rurality and area‐level deprivation, to better understand the structural drivers of these patterns. The persistence of a higher burden of aggressive forms of breast cancer among Black women suggests the need to further examine how structural and biological factors jointly shape outcomes in high‐mortality settings. Moreover, longitudinal studies linking incidence patterns with treatment access and survival outcomes are needed to understand the long‐term impacts of these disparities. Implications Our study's findings have important implications for cancer control efforts in Mississippi and similar high‐burden settings. The higher incidence of aggressive subtypes among Black women underscores the need for more targeted strategies to improve early detection and access to high‐quality care. That may include enhanced risk stratification, culturally tailored outreach, and improved access to treatments. Continued surveillance of subtype‐specific trends is critical to monitor progress and guide resource allocation. Although increases in localized‐stage diagnoses suggest progress in early detection, persistent disparities in advanced‐stage incidence indicate that barriers to timely diagnosis remain. Although insurance coverage was not evaluated in this study, previous research suggests that Medicaid expansion may improve access to cancer screening and timely care [44, 45]. In Mississippi, the absence of Medicaid expansion may therefore represent one contextual consideration for future cancer control efforts. Expanding screening access, strengthening community‐based outreach, and improving care navigation may help reduce disparities in breast cancer outcomes. Strengths and Limitations This study has several strengths. It used a large, high‐quality, population‐based cancer registry data that meet NAACR gold certificate standards. In addition, the extended study period allowed for a robust assessment of temporal trends. Moreover, stratified analyses by race, stage at diagnosis, and molecular subtype provided a more detailed understanding of heterogeneity in breast cancer incidence. Additionally, to the best of our knowledge, our study is the first study to comprehensively assess the temporal trends by race, stage at diagnosis, and molecular subtype in Mississippi, the state with persistently the highest mortality. This study also has limitations. First, analyses were restricted to Black and White women because of small sample sizes in other racial groups, which may limit generalizability. Second, potential misclassification of tumor characteristics, including stage at diagnosis and molecular subtype, may have occurred due to incomplete or inaccurate reporting. Approximately 20% of tumors were classified as having an unknown molecular subtype because of incomplete receptor information. Although incomplete receptor data may have influenced subtype‐specific incidence estimates, the proportion of tumors with unknown molecular subtype remained relatively stable over the study period, declining only modestly from 24.0% in 2010 to 19.6% in 2019. This suggests that changes in reporting completeness were unlikely to have substantially influenced the observed temporal trends. Third, the registry lacks detailed individual‐level information on socioeconomic and risk factors such as reproductive history and comorbidities that could be associated with the observed trends. The registry also lacks information on screening utilization, healthcare access, and other structural factors that may influence breast cancer diagnosis. Consequently, these factors could not be directly examined in this study. Finally, because this is an observational study using registry data, causal inferences cannot be made. Conclusions Our study's findings highlight important differences in breast cancer incidence patterns that are obscured by overall trends. Stable population trends should not be interpreted as reassuring, as they may conceal worsening disparities in breast cancer outcomes. These results underscore the need for stratified cancer surveillance and sustainable, equity‐focused strategies to reduce disparities and improve breast cancer outcomes in Mississippi and potentially other high‐burden settings with similar demographic and healthcare characteristics. Author Contributions Jasmine Miller‐Kleinhenz: conceptualization, funding acquisition, methodology, writing – review and editing, supervision. Rifath Ara Alam Barsha: methodology, writing – original draft, conceptualization, writing – review and editing, formal analysis, data curation. Funding This study was supported by grant R00CA267674 (Dr. Jasmine Miller‐Kleinhenz) from the National Cancer Institute. The funding agency was not involved in the study design; the collection, analysis, and interpretation of the data; the preparation, review, or approval of the manuscript; or the decision to submit for publication. Conflicts of Interest The authors declare no conflicts of interest. Supporting information Figure S1: Age‐adjusted breast cancer incidence rates by race and stage at diagnosis. Figure S2: Age‐adjusted breast cancer incidence rates by race and molecular subtype. Figure S3: (A–E) Joinpoint analysis of breast cancer incidence for Black women in Mississippi by stage at diagnosis, 2010–2019. Figure S4: (A–E) Joinpoint analysis of breast cancer incidence for White women in Mississippi by stage at diagnosis, 2010–2019. Figure S5: (A–E) Joinpoint analysis of breast cancer incidence for Black women in Mississippi by molecular subtype, 2010–2019. Figure S6: (A–E) Joinpoint analysis of breast cancer incidence for White women in Mississippi by molecular subtype, 2010–2019. Acknowledgments The authors acknowledge the Mississippi Cancer Registry for providing the data used in this study. Data Availability Statement The data used in this study are not publicly available and cannot be shared due to data use agreements with the Mississippi Cancer Registry. Researchers may request access to the data through the registry. References Associated Data Supplementary Materials Figure S1: Age‐adjusted breast cancer incidence rates by race and stage at diagnosis. Figure S2: Age‐adjusted breast cancer incidence rates by race and molecular subtype. Figure S3: (A–E) Joinpoint analysis of breast cancer incidence for Black women in Mississippi by stage at diagnosis, 2010–2019. Figure S4: (A–E) Joinpoint analysis of breast cancer incidence for White women in Mississippi by stage at diagnosis, 2010–2019. Figure S5: (A–E) Joinpoint analysis of breast cancer incidence for Black women in Mississippi by molecular subtype, 2010–2019. Figure S6: (A–E) Joinpoint analysis of breast cancer incidence for White women in Mississippi by molecular subtype, 2010–2019. Data Availability Statement The data used in this study are not publicly available and cannot be shared due to data use agreements with the Mississippi Cancer Registry. Researchers may request access to the data through the registry.