Article Structure
This article is organized into the following sections: Highlights; Clinical background and unmet need; Study design and methods; Key findings; Clinical interpretation; Expert commentary and context with current evidence; Limitations and generalizability; Practice implications; Conclusion; Funding and trial registration; References.
Highlights
More than half of patients with atypical endometrial hyperplasia (AEH) had concurrent endometrial carcinoma (EC) at hysterectomy in this single-institution cohort.
Hypertension was independently associated with markedly higher odds of occult EC, while adenomyosis and endometrial polyps were associated with lower odds.
Among patients with EC, 39% met Mayo criteria on final pathology, translating to an overall 21% risk of features associated with nodal assessment at hysterectomy.
Lack of antecedent medical care was linked to a greater likelihood of meeting Mayo criteria, underscoring the role of structural inequity in cancer risk stratification.
Clinical Background and Unmet Need
AEH, now often categorized within the framework of endometrial intraepithelial neoplasia, is a recognized precursor to endometrioid endometrial carcinoma. The central clinical challenge is that a substantial proportion of patients diagnosed with AEH on preoperative sampling already harbor invasive carcinoma at definitive surgery. This uncertainty influences counseling, surgical planning, referral to gynecologic oncology, and decisions about whether lymph node assessment should be considered at the time of hysterectomy.
Most published series on concurrent EC in AEH derive from populations that are not enriched for racial and ethnic minorities, and relatively few studies have focused on how access to care and comorbid conditions shape occult cancer risk. That gap matters. Endometrial cancer outcomes in the United States are marked by persistent racial disparities, and these are not fully explained by tumor biology alone. Delays in diagnosis, fragmented access to gynecologic care, and differences in preoperative workup may all influence stage at presentation and the complexity of surgical management.
The present study by Soiffer and colleagues addresses this clinically important space by examining predictors of concurrent EC and predictors of final pathology meeting Mayo criteria among patients with AEH treated at a single urban center serving a predominantly minority population.
Study Design and Methods
This was a retrospective single-institution study of patients with AEH who underwent definitive treatment with hysterectomy between 2014 and 2022. The investigators analyzed demographic, clinical, imaging, and pathologic characteristics to identify factors associated with two related outcomes: first, the presence of concurrent EC at hysterectomy; and second, among those with EC, the likelihood of meeting Mayo criteria on final histopathologic review.
The cohort included 123 patients. The population was racially and ethnically diverse and minority-enriched: 45% Hispanic, 22% Black, 20% non-Black/non-Hispanic other, and 13% undisclosed. The abstract emphasizes that this is an underrepresented urban population, which is a key strength because it broadens the evidence base beyond more homogeneous datasets.
Regression modeling was used to identify associations. Univariate analyses were performed, followed by multivariate models to determine independent predictors. While the abstract does not provide the full list of candidate variables tested, reported predictors included hypertension, adenomyosis, endometrial polyps, antecedent medical care, and endometrial echogenicity of at least 2 cm.
The second endpoint used Mayo criteria, a pathologic framework traditionally applied to estimate nodal metastasis risk in endometrial cancer and to guide selective lymphadenectomy. In broad terms, Mayo low-risk criteria incorporate endometrioid histology, grade 1 or 2 disease, less than 50% myometrial invasion, and tumor size 2 cm or less; patients not meeting these low-risk criteria may be considered at higher risk for nodal involvement. In this study, the authors examined how often occult cancers arising from an AEH diagnosis ultimately crossed that threshold.
Key Findings
Concurrent endometrial carcinoma was common
Of 123 patients with AEH who underwent hysterectomy, 56% were found to have EC on final surgical pathology. This is clinically striking. Historically, many series have reported concurrent carcinoma in roughly 30% to 50% of patients with AEH or endometrial intraepithelial neoplasia, depending on sampling method, pathology criteria, and patient selection. A 56% prevalence places this cohort at the upper end of that range and supports the authors’ conclusion that occult cancer burden may be higher than previously appreciated in this setting.
Hypertension increased the odds of occult cancer
In multivariate analysis, hypertension was independently associated with higher odds of concurrent EC, with an odds ratio of 5.52 and a 95% confidence interval of 1.56 to 19.62. Although obesity and diabetes are often emphasized in endometrial neoplasia risk, this finding suggests that hypertension may serve as an accessible clinical marker of a higher-risk metabolic phenotype in patients with AEH. The confidence interval is wide, reflecting the modest sample size, but the effect estimate is large enough to merit attention.
Adenomyosis and endometrial polyps were associated with lower odds of cancer
Two intrauterine pathologies were independently associated with reduced odds of concurrent EC: adenomyosis, with an odds ratio of 0.26 (95% confidence interval 0.07 to 0.91), and endometrial polyps, with an odds ratio of 0.25 (95% confidence interval 0.08 to 0.81). These are among the more novel observations in the study.
The adenomyosis finding is especially interesting because it raises the possibility that coexisting benign uterine pathology may correlate with different symptom patterns, earlier evaluation, or distinct microenvironments affecting tumor development or detection. Likewise, a polyp-associated diagnosis of AEH may in some cases reflect more localized disease captured earlier during evaluation for abnormal uterine bleeding. These hypotheses remain speculative, but they offer plausible biologic and clinical explanations.
A meaningful subset met Mayo criteria
Among 62 evaluable EC cases, 24 patients, or 39%, met Mayo criteria on final histopathologic review. The authors translate this into an overall 21% likelihood in the full AEH cohort. This is not a trivial figure. It means that about 1 in 5 patients initially diagnosed with AEH may ultimately have cancer features relevant to nodal risk assessment and potentially to intraoperative surgical decision-making.
For gynecologic surgeons, that finding has direct practical implications. If the preoperative diagnosis is AEH, but the probability of finding EC with higher-risk features is substantial, surgical planning may need to account for the possibility of staging procedures, frozen-section limitations, referral patterns, or the availability of sentinel lymph node mapping.
Access to care influenced risk stratification
Lack of antecedent medical care was associated with an increased likelihood of meeting Mayo criteria, with a reported p value of 0.047. This is one of the most important translational findings in the study. It suggests that social and structural determinants of health are not simply background variables; they may materially influence the severity of occult malignancy at the time definitive treatment is finally delivered.
Although the abstract does not quantify the exact operational definition of antecedent medical care, the signal is clear: patients disconnected from routine care pathways may present later, be worked up less completely, or experience delays that allow occult malignancy to progress. This is particularly relevant in urban safety-net settings.
Endometrial echogenicity at least 2 cm predicted higher-risk occult cancer
In multivariate analysis, endometrial echogenicity of at least 2 cm was associated with markedly increased odds of meeting Mayo criteria, with an odds ratio of 13.22 and a 95% confidence interval of 1.62 to 108.17. The confidence interval is very wide, again indicating a limited sample and probable event sparsity, but the direction of effect is clinically sensible. A larger endometrial lesion burden on imaging would be expected to correlate with larger tumor size or more advanced pathology.
By contrast, adenomyosis remained negatively associated with concurrent cancer meeting Mayo criteria, with an odds ratio of 0.22 (95% confidence interval 0.05 to 0.98), reinforcing the possibility that this coexisting condition identifies a subset with less aggressive or less advanced occult disease.
Clinical Interpretation
The practical message from this study is that AEH should not be viewed as a uniformly low-complexity preinvasive diagnosis, especially in populations facing healthcare inequities. In this cohort, the probability of invasive cancer was high, and the probability of pathologic features potentially relevant to nodal assessment was also clinically meaningful.
Several bedside implications follow. First, preoperative counseling should explicitly address the likelihood of finding EC at hysterectomy. Second, referral to a surgeon or center capable of intraoperative risk assessment and nodal mapping may be reasonable for selected patients, particularly those with risk factors such as hypertension, limited prior healthcare engagement, or substantial endometrial thickening or echogenicity on ultrasound. Third, the absence of racial or ethnic differences within this single-center cohort should not be misread as evidence that disparities are absent; rather, the study points toward access-related variables as an important layer of risk that may cut across racial and ethnic groups within a shared safety-net environment.
The adenomyosis and polyp findings are hypothesis-generating rather than practice-changing. They should not currently be used to downgrade concern or to avoid appropriate oncologic planning. However, they may help guide future work on phenotype refinement in AEH.
Expert Commentary and Context With Current Evidence
Current guideline statements already recognize AEH or endometrial intraepithelial neoplasia as a lesion with substantial concurrent cancer risk. The American College of Obstetricians and Gynecologists Clinical Consensus on management of endometrial intraepithelial neoplasia or atypical endometrial hyperplasia emphasizes hysterectomy as definitive treatment for patients who do not desire fertility preservation and notes the frequency of occult carcinoma in hysterectomy specimens. Likewise, the Society of Gynecologic Oncology has highlighted the need for careful preoperative counseling and individualized operative planning.
The present study aligns with that broad framework but adds two clinically relevant nuances. First, the reported 56% occult cancer rate is higher than many prior estimates, suggesting that risk may be amplified in some real-world populations. Second, the study links care access with the likelihood of meeting criteria associated with nodal risk, pushing the discussion beyond pathology and into health systems design.
From a biologic standpoint, the hypertension association may reflect the broader metabolic milieu driving estrogen-dependent endometrial carcinogenesis. Chronic hyperinsulinemia, inflammation, adipokine imbalance, and vascular dysfunction often cluster, and hypertension may function as a surrogate for that burden. Whether it is independently causal or simply a strong clinical marker cannot be resolved from this design.
The finding that no racial or ethnic differences were detected should be interpreted with caution. The sample size may not support precise subgroup comparisons, and racial categories alone are blunt instruments for studying disparities. In many settings, insurance status, longitudinal care access, neighborhood deprivation, language concordance, and time-to-procedure may be more proximal determinants of cancer severity than race or ethnicity per se.
Limitations and Generalizability
This study has the usual limitations of retrospective single-center analyses. Selection bias is possible because only patients treated definitively with hysterectomy were included; patients managed conservatively, including those pursuing fertility-sparing therapy, were not represented. Pathology interpretation may also vary across time and reviewers, although this is an issue shared by much of the AEH literature.
The sample size is modest, particularly for the analysis of Mayo criteria among the subset with EC. This is evident in the wide confidence intervals for several multivariable estimates, especially the ultrasound variable. Such findings should be considered exploratory until replicated.
Some potentially important covariates are not available in the abstract, including body mass index distribution, diabetes status in multivariable models, method of endometrial sampling, use of office biopsy versus dilation and curettage, time from biopsy to surgery, and sentinel node mapping practices. These factors could affect both occult cancer prevalence and the applicability of Mayo-style risk assessment.
Even so, the study’s single-center nature is also a strength. It reflects the lived clinical reality of a specific urban population that is often underrepresented in the literature. For clinicians serving similar communities, these data may be more relevant than estimates derived from demographically different referral cohorts.
Practice Implications
For clinicians evaluating a patient with AEH, this study supports a more deliberate preoperative assessment when certain features are present. Hypertension, limited prior access to care, and substantial endometrial echogenicity may justify heightened concern for invasive disease and for pathology that could alter intraoperative management.
In practical terms, institutions may consider pathways that ensure patients with AEH are triaged to surgeons comfortable with staging decisions or sentinel lymph node mapping when indicated. Preoperative imaging review and careful ultrasound reporting may also be more valuable than often assumed. At a systems level, improving continuity of gynecologic care and reducing delays from abnormal bleeding evaluation to definitive treatment may decrease the proportion of patients who ultimately present with more consequential occult cancer.
At the same time, clinicians should resist overextending the data. The results do not prove that every patient with AEH requires nodal assessment, nor do they establish adenomyosis or polyps as protective in a causal sense. The appropriate response is better risk stratification, not reflex escalation for all.
Conclusion
This single-institution study adds important evidence to the management of AEH in a minority-enriched urban population. The prevalence of concurrent EC was high at 56%, and approximately 21% of the overall cohort had occult cancer meeting Mayo-criteria thresholds relevant to nodal risk assessment. Hypertension emerged as an independent predictor of concurrent cancer, while adenomyosis and endometrial polyps were associated with lower odds. Lack of antecedent medical care increased the likelihood of higher-risk occult cancer features, highlighting the clinical consequences of inequity.
The study’s central contribution is not only its quantification of risk, but its reminder that preinvasive gynecologic pathology is embedded in real-world healthcare access. Future multicenter studies should validate these predictors, test modern sentinel node-based strategies in AEH populations with high occult cancer rates, and more rigorously integrate social determinants of health into preoperative risk models.
Funding and ClinicalTrials.gov
Funding information was not provided in the abstract. No ClinicalTrials.gov registration is applicable or reported for this retrospective single-institution study.
References
Soiffer JL, Suresh D, Jalai C, Xu Y, Fnu S, Hussein NM, Lanjewar SD, Lin J, Scanlon LR, Lin KY, Nevadunsky NS, Isani SS. Risk of endometrial carcinoma in patients with atypical endometrial hyperplasia in a population of predominantly racial and ethnic minorities, a single institution study. Gynecologic Oncology. 2026-05-30;210:77-83. PMID: 42217257.
ACOG Clinical Consensus No. 5. Management of Endometrial Intraepithelial Neoplasia or Atypical Endometrial Hyperplasia. Obstetrics and Gynecology. 2023;142(3):e80-e92.
Trimble CL, Kauderer J, Zaino R, et al. Concurrent endometrial carcinoma in women with a biopsy diagnosis of atypical endometrial hyperplasia: a Gynecologic Oncology Group study. Cancer. 2006;106(4):812-819.
Abu-Rustum NR, Yashar CM, Bradley K, et al. NCCN Clinical Practice Guidelines in Oncology: Uterine Neoplasms. National Comprehensive Cancer Network. Current guideline version available online.
Mariani A, Dowdy SC, Keeney GL, et al. Prospective assessment of lymphatic dissemination in endometrial cancer: a paradigm shift in surgical staging. Gynecologic Oncology. 2008;109(1):11-18.