Skip to main content
NCBI home page
JNCI Journal of the National Cancer Institute logoLink to JNCI Journal of the National Cancer Institute
. 2025 Jun 20;117(9):1891–1903. doi: 10.1093/jnci/djaf135

Trends in uterine cancer incidence and mortality: insights from a natural history model

William D Hazelton 1, Matthew Prest 2, Ling Chen 3, Kevin Rouse 4, Elena B Elkin 5, Jennifer S Ferris 6, Xiao Xu 7, Nina A Bickell 8, Chung Yin Kong 9, Stephanie Blank 10, Eric J Feuer 11, Goli Samimi 12, Brandy M Heckman-Stoddard 13, Tracy M Layne 14, Jason D Wright 15, Evan R Myers 16, Laura J Havrilesky 17,✉
PMCID: PMC12415960  NIHMSID: NIHMS2152498  PMID: 40509873

Abstract

Background

Uterine cancer incidence and mortality are increasing, with concomitant disparities in outcomes between racial groups. Natural history modeling can evaluate risk factors, predict future trends, and simulate approaches to reducing mortality and disparities.

Methods

We designed a natural history model of uterine cancer using a multistage clonal expansion design. The model is informed by National Health and Nutrition Examination Survey, National Health Examination Survey, age, time period, birth cohort, and birth certificate data on reproductive histories and body mass index (BMI). We fit and calibrated the model to Surveillance, Epidemiology, and End Results data by race and ethnicity as well as histologic subgroup. We projected future incidence and estimated the degree of contribution of BMI, reproductive history, and competing hysterectomy to excess uterine cancer incidence.

Results

The model accurately replicated Surveillance, Epidemiology, and End Results incidence for endometrioid, nonendometrioid, and sarcoma subgroups for non-Hispanic Black and non-Hispanic White patients. For endometrioid, nonendometrioid, and sarcomas, BMI-attributable risks are greater for non-Hispanic White than for non-Hispanic Black patients; reproductive history–attributable risks are greater for non-Hispanic Black patients. Between 2018 and 2050, endometrioid incidence is projected to rise by 64.9% in non-Hispanic Black individuals and17.5% in non-Hispanic White individuals; the projected rise for the nonendometrioid subgroup is 41.4% in non-Hispanic Black individuals and 22.5% in non-Hispanic White individuals; the sarcoma incidence projected increase is 36% in non-Hispanic Black individuals and 29.2% in non-Hispanic White individuals.

Conclusions

Uterine cancer risk is substantially explained by reproductive history and BMI, with differences observed between non-Hispanic Black and non-Hispanic White individuals and future projections indicating perpetuation of disparities. Lower rates of hysterectomy and rising obesity rates will likely contribute to continued increases in uterine cancer incidence.

Introduction

With an average annual percentage change of 0.8% for incidence and 1.9% for mortality from 2015 to 2019, uterine cancer has the fastest-growing burden among cancers in women in the United States.1 This burden is disproportionately distributed: Between 1999 and 2016, mortality increased 29% for Black women compared with 18% for White women.2 This discrepancy is due to greater increases in incidence (46% [average annual percentage change = 2.4%] among Black women vs 9% [average annual percentage change = 0.5%] among White women2) and poorer survival (62% 5-year survival for Black women vs 84% for White women3).

At least part of this increase is attributable to rising rates of obesity4,5 and subsequent increases in exposure to unopposed estrogen through increased peripheral estrogen synthesis in adipocytes5; a 5 kg/m2 increase in body mass index (BMI) is associated with a relative risk of endometrial cancer of 1.59 (95% confidence interval [CI] = 1.50 to 1.68).6 Aspects of reproductive history associated with unopposed estrogen exposure, such as younger age at menarche, younger age at last pregnancy, and older age at menopause, are associated with increased risk, whereas reproductive history factors counteracting estrogen, such as oral contraceptive use and multiple pregnancies, are protective.7-10 Racial and ethnic differences in these factors, such as higher BMI11,12 and a lower average age at menarche13 among Black women, may account for observed differences in cancer incidence, although there may be complex correlations or interactions between obesity and reproductive history (eg, higher parity is not as protective for Black women as it is for White women14).

Simulation modeling can improve our understanding of the future impact of currently observed trends and inform prioritization and design of studies of emerging prevention, screening, and treatment strategies. The Duke University Uterine Cancer Model (DU-CAM) is a comprehensive model developed as part of the Cancer Intervention and Surveillance Modeling Network (CISNET) Uterus Modeling Group to simulate uterine cancer natural history and estimate future trends in incidence, mortality, and the potential impact of various population-level strategies to reduce these burdens.

Methods

Model description

DU-CAM is a multistage clonal expansion model that simulates the natural history of uterine cancer while accounting for age, race, trends in reproductive history, BMI, and prior hysterectomy. DU-CAM relates individual reproductive and BMI histories to the cellular processes underlying the development of uterine cancer. Using reproductive history and corresponding BMI measurements from the National Health and Nutrition Examination Survey (NHANES) and National Health Examination Survey, we estimated their effects as risk factors for uterine cancer incidence and mortality, incorporating hysterectomy as a competing risk.15 Use of relevant publicly available databases was approved by the Duke Health System Institutional Review Board (No. Pro00110594).

Biologically based model

Because the underlying biological processes in oncogenesis that occur before cancer diagnosis are unobservable, all mathematical models of cancer involve imputation of progression rates by fitting to observed age-specific incidence. Although Markov state-transition models typically impute progression of the physical tumor between stages, clonal expansion models explicitly simulate processes of cell division, mutation, and promotion.16-20 DU-CAM combines a 2-stage clonal expansion component of premalignant growth with a single-stage clonal expansion component of malignant growth, metastasis, and American Joint Committee on Cancer (AJCC) stage detection (Figure 1), allowing estimation of the time-dependent probability of uterine cancer incidence at each age. We briefly outline the model here; details are provided in Supplementary Methods.

Figure 1.

Figure 1.

Open in a new tab

DU-CAM model schematic. The DU-CAM model combines a stochastic 2-stage clonal expansion model of premalignant cell initiation, clonal expansion, malignant transformation, and malignant lag time, with a stochastic single-stage clonal expansion model that begins with a single malignant cell that may undergo malignant clonal expansion, possible metastatic spread, and incidence. Incident cancers are staged according to AJCC stages I, II, III, and IV based on single-stage clonal expansion model stochastic threshold probabilities for malignant clone size and metastatic status. The premalignant 2-stage clonal expansion model is linked to the malignant single-stage clonal expansion model by calibrating the single-stage clonal expansion model mean time to cancer incidence to equal the estimated malignant lag time from the calibrated 2-stage clonal expansion model. The 2-stage clonal expansion premalignant model was first calibrated to cancer incidence by race, ethnicity, and histology using thousands of women’s BMI and reproductive histories from NHANES. Multiple dose-response models were tested and compared for BMI and reproductive history, potentially influencing initiation, promotion, or malignant transformation or the malignant lag time. Reproductive history age intervals were defined as intervals between birth, menarche, first pregnancy, first birth, last pregnancies (combined), last birth, menopause, hysterectomy, and end of follow-up or cancer incidence. Likelihood-based model comparisons identified best-fitting models by race, ethnicity, and histology as BMI influencing malignant transformation and reproductive history influencing premalignant promotion. Mathematically, both the 2-stage clonal expansion and single-stage clonal expansion models use probability-generating functions that represent the evolving probability distributions for numbers of premalignant and malignant cells, respectively, as they evolve throughout a woman’s lifetime, given her BMI and reproductive history. AJCC = American Joint Committee on Cancer; BMI = body mass index; DU-CAM = Duke University Uterine Cancer Model; NHANES = National Health and Nutrition Examination Survey.

Likelihood-based gradient search methods were used to optimize 2-stage clonal expansion premalignant growth parameters (mutation, cell division, death, promotion, and transition from premalignant to malignant cell) and single-stage clonal expansion malignant growth parameters (malignant cell division, death, metastasis, and clone size).21 Flexible dose-response relationships were assumed for BMI and reproductive history (total number of live births and ages at menarche, first and last birth, and menopause17,19-21) on each 2-stage clonal expansion parameter; hysterectomy eliminated uterine cancer risk. Single-stage clonal expansion malignant growth parameters were calibrated to age-, race-, and histology-specific AJCC stage distributions in Surveillance, Epidemiology, and End Results (SEER) data.22,23

Population-based data

Race and ethnicity were categorized collectively as non-Hispanic Black, non-Hispanic White, and Hispanic. By consensus, the CISNET Uterus Working Group prioritized initial model development on only non-Hispanic White and non-Hispanic Black because of the particularly high burden of disease among non-Hispanic Black women. We modeled 3 parallel histology-based disease cohorts (Supplementary Methods, Table S1): (1) endometrioid (the most common subtype with the best prognosis24), (2) nonendometrioid high-risk carcinomas (more common in non-Hispanic Black women and with a less favorable prognosis)24, and (3) leiomyosarcomas (sarcoma) (worst prognosis25).

We harmonized demographic, BMI, and reproductive history variable data by race and ethnicity (non-Hispanic Black and non-Hispanic White) for NHANES-III (1991) and all NHANES surveys between 2000 and 2020.15 Using individual reproductive history and corresponding BMI measurements, we estimated age-specific and year-specific BMI percentile distributions to simulate BMI trajectories across reproductive history events. This approach inherently incorporates complex correlations among race, age, BMI, and reproductive history without the need to explicitly estimate those correlations.26 We used birth certificate data from the National Center for Health Statistics to adjust for age-, race-, year-, and birth order–specific weight retention.27,28

We used maximum likelihood estimation methods to optimize fits to age-specific incidence and mortality data by race and ethnicity (non-Hispanic White and non-Hispanic Black), AJCC stage, and histologic type (endometrioid, nonendometrioid, and sarcoma), from 18 SEER registries spanning the years 2000-2018.3 Likelihood optimization was performed using the Bhat program in R, version 4.2.2, software (R Foundation for Statistical Computing).21 A Poisson likelihood was evaluated for SEER population and incidence data given the DU-CAM model hazard (or risk) at each age and year that was calculated as the mean hazard across all corresponding NHANES histories.

For each histology and race and ethnicity category, multiple candidate models were optimized and compared, assuming that an individual’s BMI and reproductive history have a dose-response influence on premalignant initiation, clonal growth, and malignant transformation, or time from first malignant cell until cancer incidence. Best-fitting models for each histology and race and ethnicity category (based on maximum likelihood estimation, adjusted for the number of parameters) were designated “full models.”29 We separately optimized nested “submodels,” where either BMI alone or reproductive history alone was assumed to modify any of the multistage model parameters.

Modeling endometrial cancer incidence

Cross-sectional age-specific and birth cohort–specific US endometrial cancer incidence derived from SEER data were mapped to US cross-sectional reproductive history and BMI profiles, based on birth cohort, age, and race and ethnicity.

Uterine cancer mortality component

We used reproductive history and BMI histories from NHANES to calibrate to uterine cancer mortality using a separate multistage clonal expansion model (Figure S1), including a calendar year trend to adjust for changes in clinical care, other-cause mortality, and other unknown factors between 2000 and 2020.

Statistical analysis

Calibration

Calibration targets were SEER-18 uterine cancer incidence (endometrioid, nonendometrioid, sarcoma), AJCC stage distributions, and uterine cancer mortality (all histologies) (Table S1). DU-CAM calibration was performed by single year of age and year of incidence for each histologic category (endometrioid, nonendometrioid, and sarcoma) by diagnosis year and race and ethnicity (non-Hispanic Black and non-Hispanic White). Endometrial intraepithelial neoplasia, a precursor to endometrioid cancers, was modeled as a secondary result of fitting the DU-CAM model to endometrioid cancer incidence by adjusting lag time used in the 2-stage clonal expansion model to approximate the observed age-specific peak in endometrial intraepithelial neoplasia incidence around the ages of 50-54 years.30

Projected incidence

Future incidence rates to 2050 by race and ethnicity were generated by projecting forward reproductive history and BMI trends observed in NHANES between 2000 and 2020 while keeping age-related “background” parameters constant.

Attributable risks for BMI and reproductive history on incidence and mortality

Beginning with a full model of mortality or incidence by histology, we set dose-response parameters to zero for BMI, then for reproductive history, then for both together to estimate a background effect not associated with BMI or reproductive history. With zero dose response for BMI, the remaining age-specific hazard represents contributions from background and reproductive history; we performed a similar exercise using zero dose response for reproductive history. We calculated attributable risks to BMI, reproductive history, and background as the relative contributions of those components to the total age-specific hazard (Supplementary Methods).

Impact of hysterectomies

To evaluate the potential impact of hysterectomy on incidence, we performed a counterfactual comparison of incidences with uterine cancer risk set to zero after hysterectomy (base case) to a scenario where women who had undergone hysterectomy were still “at risk” for uterine cancer.

Results

Model selection

Using maximum likelihood estimation methods, we compared more than 20 dose-response models for each histology category for non-Hispanic Black and non-Hispanic White women, resulting in a full model that almost uniformly provided the likelihood-based best-fit model for endometrioid, nonendometrioid, and sarcoma among non-Hispanic Black and non-Hispanic White women. The best full models for endometrioid, nonendometrioid, and sarcoma all include a linear dose response for BMI that affects malignant transformation (1 parameter) and a piece-wise constant effect of premalignant promotion during 4 reproductive history intervals (4 parameters): (1) between menarche and first birth (or 10 years following menarche for nulliparous women), (2) during pregnancies, (3) between first and subsequent pregnancies or until menopause, and (4) after menopause. For endometrioid cancers in non-Hispanic Black women, BMI additionally affected the initiation rate (1 additional parameter).

DU-CAM fits to incidence and mortality

Table 1 displays age-adjusted SEER incidence and corresponding DU-CAM estimates for endometrioid cancers, nonendometrioid cancers, and sarcomas, with stage distributions and uterine cancer mortality for the years 2000, 2010, and 2018, as well as DU-CAM–predicted incidence and mortality for the years 2030, 2040, and 2050. Between 2018 and 2050, endometrioid cancer incidence is projected to rise by 64.9% in non-Hispanic Black women and 17.5% in non-Hispanic White women, the nonendometrioid cancer projected rise is 41.4% in non-Hispanic Black women and 22.5% in non-Hispanic White women, and the sarcoma incidence projected increase is 36% in non-Hispanic Black women and 29.2% in non-Hispanic White women. The non-Hispanic Black to non-Hispanic White incidence ratio for nonendometrioid cancers is projected to increase from 1.46 in 2000 to 2.66 by 2050. Figure 2 displays incidence projections by year in each race and ethnicity and histology cohort. By 2050, overall uterine cancer mortality is projected to rise by 27.1% in non-Hispanic Black women and 13.1% in non-Hispanic White women.

Table 1.

DU-CAM model calibration to SEER uterine cancer incidence, by histology and AJCC stage, and uterine cancer mortality.

Likelihood-based DU-CAM model calibration to SEER
 DU-CAM model projected BMI and reproductive history trends
Outcome 2000
 2010
 2018
 2030
 2040
 2050
Non-Hispanic Black Non-Hispanic White Non-Hispanic Black Non-Hispanic White Non-Hispanic Black Non-Hispanic White Non-Hispanic Black Non-Hispanic White Non-Hispanic Black Non-Hispanic White Non-Hispanic Black Non-Hispanic White
Endometrioid
SEER endometrioid incidencea (95% CI) 23.7 (19.5 to 27.1) 43.7 (41.6 to 45.7) 30.8 (27.6 to 33.2) 48.3 (46.8 to 49.8) 34.6 (31.9 to 36.7) 49.3 (47.8 to 50.9) — — — — — —
DU-CAM endometrioid incidencea (95% CI) 28.8 (23.5 to 33.5) 45.3 (42.8 to 47.6) 28.6 (24.4 to 32.3) 46.3 (44.4 to 48.4) 35.3 (31.4 to 38.9) 53.0 (50.9 to 55.2) 44.9 (41.3 to 48.3) 55.6 (53.7 to 57.7) 50.1 (46.3 to 53.8) 58.2 (55.9 to 60.2) 58.2 (51.9 to 63.4) 62.3 (59.4 to 65.2)
SEER AJCC stage I:II:III:IV, % 81:5:11:3 79:8:8:6 73:8:13:6 85:3:7:5 83:3:6:9 83:3:10:3 — — — — — —
DU-CAM AJCC stage I:II:III:IV, % 77:5:12:6 79:7:8:6 77:5:11:7 84:4:8:4 81:4:8:7 84:4:9:3 82:4:8:6 85:3:9:3 83:3:8:6 86:3:8:3 83:3:8:6 86:3:8:3
Nonendometrioid
SEER nonendometrioid incidencea (95% CI) 12.4 (10.0 to 17.0) 8.5 (7.0 to 10.5) 14.7 (13.2 to 17.7) 7.7 (6.7 to 8.9) 22.3 (21.2 to 25.0) 8.5 (7.8 to 9.5) — — — — — —
DU-CAM nonendometrioid incidencea (95% CI) 16.3 (13.3 to 21.3) 8.0 (6.5 to 10.0) 16.6 (14.5 to 20.1) 7.6 (6.6 to 8.9) 20.5 (18.8 to 23.5) 8.9 (8.1 to 9.9) 24.3 (22.8 to 27.2) 9.8 (9.0 to 10.8) 25.6 (4.1 to 28.5) 10.2 (9.4 to 11.1) 29.0 (27.2 to 32.1) 10.9 (10.0 to 12.0)
SEER AJCC stage I:II:III:IV, % 66:5:11:18 72:7:6:14 46:2:22:30 57:6:22:15 50:1:18:31 58:2:12:28 — — — — — —
DU-CAM AJCC stage I:II:III:IV, % 60:5:11:24 65:7:10:18 50:2:19:29 60:5:15:20 50:1:19:30 59:3:15:23 50:1:18:31 60:2:13:25 51:2:17:30 60:2:13:25 52:2:16:30 60:2:13:25
Sarcoma
SEER sarcoma incidencea (95% CI) 2.0 (1.6 to 3.0) 2.2 (1.9 to 2.7) 2.1 (2.0 to 2.3) 2.2 (2.0 to 2.3) 2.3 (2.1 to 2.5) 2.3 (2.1 to 2.5) — — — — — —
DU-CAM sarcoma incidencea (95% CI) 1.9 (0.9 to 2.9) 2.1 (1.7 to 2.6) 2.1 (1.1 to 3.1) 2.2 (2.0 to 2.3) 2.5 (1.5 to 3.5) 2.4 (2.3 to 2.7) 2.7 (1.7 to 3.7) 2.6 (2.4 to 2.8) 3.0 (2.0 to 4.0) 2.8 (2.6 to 3.0) 3.3 (2.3 to 4.3) 3.1 (2.9 to 3.4)
SEER AJCC stage I:II:III:IV, % 53:11:14:21 71:2:10:17 61:7:3:29 65:12:6:18 54:8:12:26 62:4:8:27 — — — — — —
DU-CAM AJCC stage I:II:III:IV, % 53:9:15:23 69:4:9:18 56:9:7:28 65:8:8:19 55:9:10:26 63:6:8:23 55:8:11:26 65:5:8:24 56:8:12:24 64:5:7:24 56:8:12:24 64:5:7:24
Uterine cancer mortality (all)
SEER uterine cancer mortalitya (95% CI) 15.3 (13.3 to 17.3) 7.5 (7.0 to 8.0) 16.1 (14.3 to 18.0) 8.4 (7.9 to 8.9) 19.4 (17.5 to 21.3) 9.3 (8.8 to 9.8) — — — — — —
DU-CAM uterine cancer mortality (95% CI) 15.3 (12.9 to 17.8) 7.4 (6.8 to 7.9) 17.5 (15.2 to 19.7) 8.7 (8.2 to 9.2) 19.2 (17.0 to 22.4) 9.1 (8.6 to 9.6) 21.3 (19.0 to 23.5) 9.5 (9.0 to 10.1) 22.8 (20.6 to 25.1) 9.9 (9.4 to 10.5) 24.4 (21.9 to 27.0) 10.3 (9.7 to 10.9)
Open in a new tab

Abbreviations: AJCC = American Joint Committee on Cancer; BMI = body mass index; DU-CAM = Duke University Uterine Cancer Model; SEER = Surveillance, Epidemiology, and End Results.

a

Incidence per 100 000 population, age adjusted using the Centers for Disease Control and Prevention Master List: 2000 projected population aged 40-84 y. Values not provided for future years.

Figure 2.

Figure 2.

Open in a new tab

DU-CAM incidence projections, by year, in each race and ethnicity and histology cohort. Circles represent annual SEER incidence, and triangles represent 10-year binned SEER incidence. Solid lines represent calibration to SEER data, and dashed lines represent future projections based on continuation of BMI and reproductive health trends observed in NHANES data between 2000 and 2020. BMI = body mass index; DU-CAM = Duke University Uterine Cancer Model; NHANES = National Health and Nutrition Examination Survey; SEER = Surveillance, Epidemiology, and End Results.

Best-fitting full model results, by histology

Table S2 presents best-fitting models by race and ethnicity as well as histology.

Endometrioid

Models included a linear BMI dose response on malignant transformation among non-Hispanic Black and non-Hispanic White women and on initiation among non-Hispanic Black women. With non-Hispanic Black and non-Hispanic White premenarche premalignant promotion rates as a reference, (1) promotion decreased between menarche and first birth (or for 10 years following menarche for nulliparous women), (2) was low during pregnancies, (3) was markedly elevated between pregnancies and until menopause, and (4) was still elevated but decreased after menopause. Promotion rates for non-Hispanic Black women during intervals 3 and 4 were higher for non-Hispanic Black women than for non-Hispanic White women.

Nonendometrioid

Models included a linear BMI dose response on malignant transformation. Promotion rates for non-Hispanic Black and non-Hispanic White women were elevated during interval 3 and decreased in 4, similar to endometrioid cancers. Promotion was elevated, however, during interval 1 for non-Hispanic Black women and during interval 2 for non-Hispanic White women.

Sarcoma

Models included a linear BMI dose response on malignant transformation similar to endometrioid and nonendometrioid cancers. Promotion was elevated during reproductive history intervals 1, 2, 3, and 4 for non-Hispanic White women and intervals 1 and 3 for non-Hispanic Black women.

Contributions of BMI, reproductive history, and hysterectomy

Table S3 displays NHANES-derived trends in obesity and reproductive history factors, including hysterectomy, between 2000 and 2020. Figures 3 through 5 display age-specific estimates of the contributions of BMI, reproductive history, and background risk factors to endometrial intraepithelial neoplasia and cancer incidence in the years 2000, 2010, and 2018 as well as age-specific risk in the counterfactual absence of hysterectomy; Figures S2 through S4 display projections. Background represents age-related incidence (and contributions from unknown factors) that would continue following menarche in the absence of dose response from BMI and reproductive history. Table 2 displays age-adjusted attributable risk estimates between 2000 and 2050 for BMI and reproductive history as a percentage of incidence for the endometrioid, nonendometrioid, and sarcomas subgroups and counterfactual increases in risk in the absence of hysterectomy.

Figure 3.

Figure 3.

Open in a new tab

Age-specific estimates of the contribution of BMI, reproductive history, and background to endometrioid cancer and endometrial intraepithelial neoplasia risk among non-Hispanic White and non-Hispanic Black women in the years 2000, 2010, and 2018. Circles represent annual SEER incidence. Dotted curves represent age-specific risk in the (counterfactual) absence of hysterectomy. BMI = body mass index; SEER = Surveillance, Epidemiology, and End Results.

Table 2.

DU-CAM model estimates of attributable risk associated with BMI, reproductive history, and incidence increases without hysterectomy.

Attributable risks for DU-CAM model with SEER calibration, %
 Attributable risks for DU-CAM model with projected BMI and reproductive history, %a
Outcome 2000
 2010
 2018
 2030
 2040
 2050
Non-Hispanic Black Non-Hispanic White Non-Hispanic Black Non-Hispanic White Non-Hispanic Black Non-Hispanic White Non-Hispanic Black Non-Hispanic White Non-Hispanic Black Non-Hispanic White Non-Hispanic Black Non-Hispanic White
Endometrioid
BMI-attributable risk (95% CI) 37 (7 to 62) 48 (42 to 55) 37 (13 to 55) 51 (47 to 56) 39 (22 to 52) 51 (47 to 55) 45 (32 to 56) 51 (47 to 55) 49 (36 to 59) 52 (49 to 56) 51 (40 to 61) 54 (50 to 57)
Reproductive history–attributable risk (95% CI) 46 (16 to 71) 14 (7 to 20) 46 (22 to 64) 13 (9 to 18) 45 (29 to 59) 13 (9 to 17) 41 (28 to 52) 14 (11 to 18) 38 (26 to 48) 14 (10 to 17) 36 (25 to 46) 13 (9 to 16)
Risk increase without hysterectomy (95% CI) 96 (0 to 177) 52 (31 to 71) 97 (28 to 150) 49 (35 to 64) 67 (20 to 110) 36 (21 to 51) 58 (22 to 90) 35 (21 to 49) 54 (20 to 86) 32 (17 to 48) 44 (0 to 99) 25 (0 to 54)
Nonendometrioid
BMI-attributable risk (95% CI) 17 (0 to 67) 45 (17 to 82) 17 (0 to 53) 48 (29 to 73) 18 (7 to 43) 51 (38 to 67) 20 (11 to 41) 51 (40 to 66) 21 (13 to 41) 53 (42 to 67) 22 (14 to 41) 54 (44 to 67)
Reproductive history–attributable risk (95% CI) 80 (50 to 131) 30 (1 to 67) 80 (61 to 117) 27 (7 to 51) 80 (67 to 106) 25 (13 to 41) 77 68 to 99) 25 (14 to 40) 76 (67 to 97) 24 (14 to 38) 75 (66 to 94) 23 (13 to 36)
Risk increase without hysterectomy 116 (55 to 273) 78 (48 to 129) 119 (90 to 194) 75 (60 to 97) 85 (69 to 132) 51 (43 to 64) 68 (58 to 103) 50 (43 to 61) 60 (51 to 94) 48 (41 to 59) 45 (32 to 81) 39 (30 to 52)
Sarcoma
BMI-attributable risk (95% CI) 63 (0 to 100) 43 (4 to 99) 62 (0 to 100) 45 (30 to 64) 61 (0 to 100) 46 (33 to 66) 61 (0 to 100) 50 (37 to 69) 61 (0 to 100) 51 (39 to 68) 61 (0 to 100) 51 (40 to 67)
Reproductive history–attributable risk (95% CI) 34 (0 to 100) 48 (9 to 105) 36 (0 to 100) 46 (30 to 65) 36 (0 to 100) 45 (31 to 66) 37 (0 to 100) 42 (29 to 60) 37 (0 to 100) 41 (29 to 58) 37 (0 to 100) 41 (30 to 57)
Risk increase without hysterectomy (95% CI) 76 (22 to 128) 58 (52 to 72) 71 (21 to 119) 52 (51 to 54) 47 (6 to 86) 36 (35 to 39) 43 (5 to 79) 37 (36 to 39) 33 (0 to 66) 32 (31 to 35) 21 (0 to 53) 23 (21 to 26)
Uterine cancer mortality (all)
BMI-attributable risk (95% CI) 9 (0 to 34) 17 (7 to 26) 9 (0 to 28) 16 (8 to 24) 10 (0 to 28) 18 (10 to 26) 11 (0 to 28) 20 (12 to 28) 12 (0 to 28) 21 (14 to 29) 13 (0 to 28) 22 (15 to 30)
Reproductive history–attributable risk (95% CI) 82 (56 to 100) 10 (0 to 20) 82 (62 to 100) 7 (0 to 16) 81 (63 to 100) 5 (0 to 14) 80 (63 to 97) 4 (0 to 12) 80 (63 to 96) 4 (0 to 11) 79 (63 to 95) 4 (0 to 11)
Open in a new tab

Abbreviations: BMI = body mass index; DU-CAM = Duke University Uterine Cancer Model; SEER = Surveillance, Epidemiology, and End Results.

a

Reproductive history– and BMI-attributable risks and counterfactual risk increases without hysterectomy are estimated from the DU-CAM model and are age adjusted using the Centers for Disease Control and Prevention Master List: 2000 projected population aged 40-84 y.

Hysterectomy

Within NHANES, hysterectomy rates decreased between 2000 and 2018, with higher premenopausal rates among non-Hispanic Black women than non-Hispanic White women. The counterfactual impact on age-adjusted incidence of endometrioid, nonendometrioid, and sarcoma, assuming no hysterectomy, is shown in Figures 3, 4, and 5 and in Table 2, projected to 2050. The absence of hysterectomy increases all age-adjusted rates, with a greater impact in non-Hispanic Black women: In 2018, the age-adjusted endometrioid rate was 67% (95% CI = 20% to 110%) higher in non-Hispanic Black women and 36% (95% CI = 21% to 51%) higher in non-Hispanic White women, the age-adjusted nonendometrioid rate was 85% (95% CI = 69% to 132%) higher in non-Hispanic Black women and 51% (95% CI = 43% to 64%) higher in non-Hispanic White women, and the age-adjusted sarcoma rate was 47% (95% CI = 6% to 86%) higher in non-Hispanic Black women and 36% (95% CI = 35% to 39%) higher in non-Hispanic White women.

Figure 4.

Figure 4.

Open in a new tab

Age-specific estimates of the contribution of BMI, reproductive history, and background to nonendometrioid cancer risk among non-Hispanic White and non-Hispanic Black women in the years 2000, 2010, and 2018. Circles represent annual SEER incidence. Dotted curves represent age-specific risk in the (counterfactual) absence of hysterectomy. BMI = body mass index; SEER = Surveillance, Epidemiology, and End Results.

Figure 5.

Figure 5.

Open in a new tab

Age-specific estimates of the contribution of BMI, reproductive history, and background to uterine sarcoma risk among non-Hispanic White and non-Hispanic Black women in the years 2000, 2010, and 2018. Triangles represent 10-year binned SEER incidence. Dotted curves represent age-specific risk in the (counterfactual) absence of hysterectomy. BMI = body mass index; SEER = Surveillance, Epidemiology, and End Results.

Body mass index

Figures 3 through 5 illustrate that BMI contributes more than reproductive history to the age-specific cancer risk at earlier ages, with decreasing relative impact at older ages. The BMI contribution to endometrioid and nonendometrioid cancers is greater for non-Hispanic White women than for non-Hispanic Black women. Also, BMI appeared to contribute to increasing endometrioid, nonendometrioid, and sarcoma incidence between 2000 and 2018, with Table 2 showing projected age-adjusted risks rising until 2050.

Reproductive history

The impact of reproductive history on age-specific endometrioid, nonendometrioid, and sarcoma incidence begins later than for BMI and contributes a higher fraction of risk at older ages compared with BMI. Reproductive history contributes relatively more to non-Hispanic Black women’s risk than to non-Hispanic White women’s risk of endometrioid and nonendometrioid cancers, and nonendometrioid incidence rates are also higher in non-Hispanic Black women than in their non-Hispanic White counterparts (Figure 4).

Discussion

DU-CAM and the CISNET Uterus Working Group parallel natural history models represent the first large-scale collaborative project to comprehensively model uterine cancer. The DU-CAM approach of merging models of specific cellular events with observable postdiagnosis events while capturing correlations between risk factors provided accurate fits to SEER incidence calibration data by race and ethnicity as well as histologic subtypes while approximating observed stage distributions and overall mortality.

By modeling an interface between national population-based health surveys and population-based national cancer incidence data, we were able to estimate substantial attributable uterine cancer risk to BMI and reproductive history events as well as the impact of hysterectomies.

DU-CAM predicts continued and substantial increases in the incidence of endometrioid, nonendometrioid, and uterine sarcoma through 2050, with non-Hispanic Black women seeing greater increases than non-Hispanic White women for every histologic type. Our modeling approach provides insight into potential mechanisms and cancer-control strategies:

  • The strong association between BMI and uterine cancer risk broadly suggests that interventions aimed at reducing the prevalence of obesity could substantially reduce future cancer incidence.

  • We identified distinct and complementary etiological mechanisms for BMI and reproductive history that remain consistent across the endometrioid, nonendometrioid, and sarcoma subgroups and between non-Hispanic White and non-Hispanic Black women. In all full models, BMI has a linear dose response on malignant transformation, and reproductive history influences premalignant promotion. We identified an additional increase in endometrioid premalignant initiation for non-Hispanic Black women that may suggest effects from other early-life exposures.

  • Low estimated promotion rates between menarche and first birth for endometrioid cancers could be due to a protective effect with the onset of menstrual cycling where periodic shedding of tissue could eliminate a fraction of small clones within the endometrial epithelium or to protective effects of some contraceptive methods,31,32 which should be considered when counseling about contraceptive method choices.

  • Increasing BMI would lead to cancers arising in smaller clones and at earlier ages. Increasing specific reproductive history factors, such as earlier menarche and nulliparity, would cause faster growth of premalignant clones, leading to higher cancer incidence at earlier ages. As shown in Figure 2, both contribute to a predicted increase in early onset for all histologic types as BMI and reproductive history risk factors increase. Population trends in BMI and individual reproductive history factors are consistent with increased uterine cancer incidence (Table S3). Based on the literature, the relationship among BMI, reproductive history, and cancer risk is complex, with slight differences between non-Hispanic Black and non-Hispanic White women. In non-Hispanic White women, uterine cancer risk decreases with each successive live birth, but the same effect is weaker in non-Hispanic Black women14; this finding may be partially attributable to greater weight retention after pregnancy in non-Hispanic Black women.33 Interventions to optimize weight gain during pregnancy34 and postpregnancy weight retention35 may have additional benefits in reducing uterine cancer risk, particularly among younger non-Hispanic Black women. Because of the complex associations between these individual components, estimating the individual attributable risk for each factor is not currently possible; further insights may be gained through analyses of additional datasets that include exposures and uterine cancer outcomes.

  • Our modeling estimates a higher incidence of each cancer subtype in the absence of hysterectomy that may be attributable to more than a simple reduction in a competing risk (Figures 3-5). Non-Hispanic Black women historically have higher rates of premenopausal hysterectomy, especially for conditions associated with abnormal uterine bleeding.36-38 Because women with these conditions may be at increased risk of uterine cancer39,40 or delayed diagnosis,41 declining hysterectomy rates may exacerbate existing disparities. The DU-CAM model can potentially use emerging data on these conditions’ effects on the endometrial cellular environment42 and cancer risk. Although methodologically challenging, stronger evidence of associations between these conditions and future risk of uterine cancer would be an important consideration for informed decision making about treatment options43 as well as a factor in estimating prior probability when evaluating screening or diagnostic strategies.

Study limitations

Because NHANES does not include sufficient individuals to accurately estimate cancer outcomes and SEER does not include reproductive or BMI histories, available data do not provide a direct connection between individual BMI and reproductive history and cancer incidence/outcomes, a common issue in modeling population-level effects of risk factors and subsequent cancer incidence.29 Data from large cohorts that allow direct linkage between individual exposures and cancer outcomes, as has been done with lung cancer,44 would strengthen these analyses, and next steps include validation in relevant datasets with uterine cancer outcomes.14,45,46 In addition, changes in data-collection standards and sample size limit the ability of datasets such as NHANES to generate parameter estimates for other racial and ethnic groups.

Conclusion

DU-CAM is a novel, comprehensive natural history model that accounts for the joint effects of age, birth cohort, reproductive history, BMI, hysterectomy, and race and ethnicity on the incidence and outcomes of 3 major subclassifications of uterine cancer. Our modeling emphasizes the substantial impact of BMI and reproductive history events on initiating uterine cancers and the potential for further rises in incidence and deepening disparities related to the ongoing upward trajectory of BMI and reduction in hysterectomy rates. Subsequent collaborative modeling work with the other Uterine Incubator investigators will help generate greater insight into potential strategies for reducing these disparities.

Supplementary Material

djaf135_Supplementary_Data
djaf135_supplementary_data.docx (1.1MB, docx)

Acknowledgments

The authors wish to acknowledge the assistance of Amelia Scott and Reena Vattakalam.

Contributor Information

William D Hazelton, Division of Public Health Sciences, Fred Hutchinson Cancer Center, Seattle, WA, United States.

Matthew Prest, Columbia University Irving Cancer Research Center, New York, NY, United States.

Ling Chen, Columbia University Irving Cancer Research Center, New York, NY, United States.

Kevin Rouse, Columbia University Irving Cancer Research Center, New York, NY, United States.

Elena B Elkin, Columbia University Irving Cancer Research Center, New York, NY, United States.

Jennifer S Ferris, Columbia University Irving Cancer Research Center, New York, NY, United States.

Xiao Xu, Columbia University Irving Cancer Research Center, New York, NY, United States.

Nina A Bickell, Tisch Cancer Center, Icahn School of Medicine at Mount Sinai, New York, NY, United States.

Chung Yin Kong, Tisch Cancer Center, Icahn School of Medicine at Mount Sinai, New York, NY, United States.

Stephanie Blank, Tisch Cancer Center, Icahn School of Medicine at Mount Sinai, New York, NY, United States.

Eric J Feuer, National Cancer Institute, Bethesda, MD, United States.

Goli Samimi, National Cancer Institute, Bethesda, MD, United States.

Brandy M Heckman-Stoddard, National Cancer Institute, Bethesda, MD, United States.

Tracy M Layne, Tisch Cancer Center, Icahn School of Medicine at Mount Sinai, New York, NY, United States.

Jason D Wright, Columbia University Irving Cancer Research Center, New York, NY, United States.

Evan R Myers, Department of Obstetrics and Gynecology, Duke University School of Medicine, Duke Cancer Institute, Durham, NC, United States.

Laura J Havrilesky, Department of Obstetrics and Gynecology, Duke University School of Medicine, Duke Cancer Institute, Durham, NC, United States.

Author contributions

William D. Hazelton (Conceptualization, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing—original draft, Writing—review & editing), Matthew Prest (Conceptualization, Data curation, Writing—review & editing), Ling Chen (Conceptualization, Data curation, Writing—review & editing), Kevin Rouse (Data curation, Writing—review & editing), Elena B. Elkin (Conceptualization, Writing—review & editing), Jennifer S. Ferris (Conceptualization, Writing—review & editing), Xiao Xu (Conceptualization, Data curation, Writing—review & editing), Nina A. Bickell (Conceptualization, Writing—review & editing), Chung Yin Kong (Conceptualization, Writing—review & editing), Stephanie Blank (Conceptualization, Writing—review & editing), Eric J. Feuer (Conceptualization, Writing—review & editing), Goli Samimi (Conceptualization, Writing—review & editing), Brandy M. Heckman-Stoddard (Conceptualization, Writing—review & editing), Tracy M. Layne (Conceptualization, Writing—review & editing), Jason D. Wright (Conceptualization, Funding acquisition, Supervision, Writing—review & editing), Evan R. Myers (Conceptualization, Formal analysis, Funding acquisition, Methodology, Supervision, Writing—original draft, Writing—review & editing), and Laura J. Havrilesky (Conceptualization, Funding acquisition, Investigation, Methodology, Supervision, Writing—original draft, Writing—review & editing).

Supplementary material

Supplementary material is available at JNCI: Journal of the National Cancer Institute online.

Funding

This research was funded by the Cancer Intervention and Surveillance Modeling Network Uterine Incubator grant No. U01CA265739 from the National Cancer Institute. The funding agency had no role in the study design, interpretation of results, or writing of the manuscript. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Conflicts of interest

All authors disclose funding of the current study by the Cancer Intervention and Surveillance Modeling Network Uterine Incubator U01 grant No. 5U01CA265739. The following authors have no additional disclosures: M.P., C.Y.K., J.F., E.E., G.S., K.R., L.C., T.L., B.H.S., B.H., N.B., and L.H. E.F. discloses a consultancy with Information Management Services Inc to provide support for National Cancer Institute work. E.M. reports consulting fees with Merck (human papillomavirus vaccination), Moderna (cytomegalovirus vaccination) and Hologic (cervical cancer screening). J.W. discloses research grants from Merck as well as royalties/licenses with the American College of Obstetricians and Gynecologists and UpToDate. X.X. discloses honoraria from the American Association of Gynecologic Laparoscopist for service on a practice guideline committee.

Data availability

Data on the harmonized NHANES BMI and reproductive history dose, model dose response calibrations, and model parameters are available at https://doi.org/10.5281/zenodo.15392555; any additional questions about data should be directed to the corresponding author.

References

  • 1. National Cancer Institute. Annual report to the nation 2020: overall cancer statistics. https://seer.cancer.gov/report_to_nation/statistics.html#:∼:text=Rates%20of%20New%20Cancer,-Adults&text=However%2C%20incidence%20rates%20increased%20for%208%20of%20the%2018%20most, and%20oral%20cavity%20and%20pharynx.
  • 2. Henley SJ, Miller JW, Dowling NF, et al.  Uterine cancer incidence and mortality—United States, 1999-2016. MMWR Morb Mortal Wkly Rep. 2018;67:1333-1338. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • *3. SEERStat Database: Incidence—SEER Research Data, 18 Registries, Nov 2020 Sub (2020-2018) Counties, National Cancer Institute, DCCPS, Surveillance Research Program, released April 2024, based on the November 2020 submission.
  • 4. Hales CM, Carroll MD, Fryar CD, et al.  Prevalence of obesity among adults and youth: United States, 2015-2016. NCHS Data Brief. 2017;288:1-8. https://www.ncbi.nlm.nih.gov/pubmed/29155689 [PubMed] [Google Scholar]
  • 5. Onstad MA, Schmandt RE, Lu KH.  Addressing the role of obesity in endometrial cancer risk, prevention, and treatment. J Clin Oncol. 2016;34:4225-4230. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Renehan AG, Tyson M, Egger M, et al.  Body-mass index and incidence of cancer: a systematic review and meta-analysis of prospective observational studies. Lancet. 2008;371:569-578. [DOI] [PubMed] [Google Scholar]
  • 7. Wu QJ, Li YY, Tu C, et al.  Parity and endometrial cancer risk: a meta-analysis of epidemiological studies. Sci Rep. 2015;5:14243. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Setiawan VW, Pike MC, Karageorgi S, et al. ; Australian National Endometrial Cancer Study Group. Age at last birth in relation to risk of endometrial cancer: pooled analysis in the epidemiology of endometrial cancer consortium. Am J Epidemiol. 2012;176:269-278. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Hemminki K, Bermejo JL, Granstrom C.  Endometrial cancer: population attributable risks from reproductive, familial and socioeconomic factors. Eur J Cancer. 2005;41:2155-2159. [DOI] [PubMed] [Google Scholar]
  • 10. Peeri NC, Bertrand KA, Na R, et al.  Understanding risk factors for endometrial cancer in young women. J Natl Cancer Inst. 2025;117:76-88. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Ogden CL, Fryar CD, Martin CB, et al.  Trends in obesity prevalence by race and Hispanic origin-1999-2000 to 2017-2018. JAMA. 2020;324:1208-1210. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Francoeur AA, Liao CI, Chang J, et al.  Associated trends in obesity and endometrioid endometrial cancer in the United States. Obstet Gynecol. 2025;145:e107-e116. [DOI] [PubMed] [Google Scholar]
  • 13. Chumlea WC, Schubert CM, Roche AF, et al.  Age at menarche and racial comparisons in US girls. Pediatrics. 2003;111:110-113. [DOI] [PubMed] [Google Scholar]
  • 14. Cote ML, Alhajj T, Ruterbusch JJ, et al.  Risk factors for endometrial cancer in black and white women: a pooled analysis from the Epidemiology of Endometrial Cancer Consortium (E2C2). Cancer Causes Control. 2015;26:287-296. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. National Health and Nutrition Examination Survey Data. Centers for Disease Control and Prevention (CDC). National Center for Health Statistics (NCHS). U.S. Department of Health and Human Services, Centers for Disease Control and Prevention; 2000-2020. [Google Scholar]
  • 16. Heidenreich WF, Luebeck EG, Moolgavkar SH.  Some properties of the hazard function of the two-mutation clonal expansion model. Risk Analysis. 1997;17:391-399. [DOI] [PubMed] [Google Scholar]
  • 17. Moolgavkar SH, Knudson AG. Jr., Mutation and cancer: a model for human carcinogenesis. J Natl Cancer Inst. 1981;66:1037-1052. [DOI] [PubMed] [Google Scholar]
  • 18. Moolgavkar SH, Dewanji A, Venzon DJ.  A stochastic two-stage model for cancer risk assessment. I. The hazard function and the probability of tumor. Risk Anal. 1988;8:383-392. [DOI] [PubMed] [Google Scholar]
  • 19. Moolgavkar SH, Luebeck G.  Two-event model for carcinogenesis: biological, mathematical, and statistical considerations. Risk Anal. 1990;10:323-341. [DOI] [PubMed] [Google Scholar]
  • 20. Luebeck EG, Curtius K, Jeon J, et al.  Impact of tumor progression on cancer incidence curves. Cancer Res. 2013;73:1086-1096. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Luebeck GMR, Bhat: General likelihood exploration; 2022.
  • 22. Hazelton WD, Goodman G, Rom WN, et al.  Longitudinal multistage model for lung cancer incidence, mortality, and CT detected indolent and aggressive cancers. Math Biosci. 2012;240:20-34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. de Koning HJ, Meza R, Plevritis SK, et al.  Benefits and harms of computed tomography lung cancer screening strategies: a comparative modeling study for the U.S. Preventive Services Task Force. Ann Intern Med. 2014;160:311-320. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Feinberg J, Albright B, Black J, et al.  Ten-year comparison study of type 1 and 2 endometrial cancers: risk factors and outcomes. Gynecol Obstet Invest. 2019;84:290-297. [DOI] [PubMed] [Google Scholar]
  • 25. D'Angelo E, Prat J.  Uterine sarcomas: A review. Gynecol Oncol. 2010;116:131-139. [DOI] [PubMed] [Google Scholar]
  • 26. Briggs AH, Claxton K, Sculpher MJ, Decision Modelling for Health Economic Evaluation. Oxford University Press; 2006. [Google Scholar]
  • 27. Nehring I, Schmoll S, Beyerlein A, et al.  Gestational weight gain and long-term postpartum weight retention: a meta-analysis. Am J Clin Nutr. 2011;94:1225-1231. [DOI] [PubMed] [Google Scholar]
  • 28. Centers for Disease Control and Prevention NCfHS. National Vital Statistics System, Natality. CDC WONDER Online Database; 2005-2023. [Google Scholar]
  • 29. Hazelton WD, Curtius K, Inadomi JM, et al.  The role of gastroesophageal reflux and other factors during progression to esophageal adenocarcinoma. Cancer Epidemiol Biomarkers Prev. 2015;24:1012-1023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Reed SD, Newton KM, Clinton WL, et al.  Incidence of endometrial hyperplasia. Am J Obstet Gynecol. 2009;200:678.e1-678.e6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Gierisch JM, Myers ER.  Oral contraceptive use and risk of cancer—response. Cancer Epidemiol Biomarkers Prev. 2014;23:677-678. [DOI] [PubMed] [Google Scholar]
  • 32. Minalt N, Caldwell A, Yedlicka GM, et al.  Association between intrauterine device use and endometrial, cervical, and ovarian cancer: an expert review. Am J Obstet Gynecol. 2023;229:93-100. [DOI] [PubMed] [Google Scholar]
  • 33. Kent-Marvick J, Cloyes KG, Meek P, et al.  Racial and ethnic disparities in postpartum weight retention: a narrative review mapping the literature to the National Institute on Minority Health and Health Disparities Research Framework. Womens Health (Lond). 2023;19:17455057231166822. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Cantor AG, Jungbauer RM, McDonagh M, et al.  Counseling and behavioral interventions for healthy weight and weight gain in pregnancy: evidence report and systematic review for the US preventive services task force. JAMA. 2021;325:2094-2109. [DOI] [PubMed] [Google Scholar]
  • 35. Amorim Adegboye AR, Linne YM.  Diet or exercise, or both, for weight reduction in women after childbirth. Cochrane Database Syst Rev. 2013;2013:CD005627. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Gartner DR, Delamater PL, Hummer RA, et al.  Integrating surveillance data to estimate race/ethnicity-specific hysterectomy inequalities among reproductive-aged women: who's at risk?  Epidemiology. 2020;31:385-392. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Harvey SV, Pfeiffer RM, Landy R, et al.  Trends and predictors of hysterectomy prevalence among women in the United States. Am J Obstet Gynecol. 2022;227:611.e1-611 e12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Simms KT, Yuill S, Killen J, et al.  Historical and projected hysterectomy rates in the USA: implications for future observed cervical cancer rates and evaluating prevention interventions. Gynecol Oncol. 2020;158:710-718. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Wise LA, Sponholtz TR, Rosenberg L, et al.  History of uterine leiomyoma and risk of endometrial cancer in black women. Cancer Causes Control. 2016;27:545-552. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Kho PF, Mortlock S, Rogers PAW, et al. ; Endometrial Cancer Association C. Genetic analyses of gynecological disease identify genetic relationships between uterine fibroids and endometrial cancer, and a novel endometrial cancer genetic risk region at the WNT4 1p36.12 locus. Hum Genet. 2021;140:1353-1365. [DOI] [PubMed] [Google Scholar]
  • 41. Doll KM, Nguyen A, Alson JG.  A conceptual model of vulnerability to care delay among women at risk for endometrial cancer. Gynecol Oncol. 2022;164:318-324. [DOI] [PubMed] [Google Scholar]
  • 42. Navarro A, Bariani MV, Yang Q, et al.  Understanding the impact of uterine fibroids on human endometrium function. Front Cell Dev Biol. 2021;9:633180. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Aarts JWM, Thompson R, Alam SS, et al.  Encounter decision aids to facilitate shared decision-making with women experiencing heavy menstrual bleeding or symptomatic uterine fibroids: a before-after study. Patient Educ Couns. 2021;104:2259-2265. [DOI] [PubMed] [Google Scholar]
  • 44. Meza R, Hazelton WD, Colditz GA, et al.  Analysis of lung cancer incidence in the Nurses' Health and the Health Professionals' Follow-Up Studies using a multistage carcinogenesis model. Cancer Causes Control. 2008;19:317-328. [DOI] [PubMed] [Google Scholar]
  • 45. The Million Women Study. Oxford Population Health CEU. Accessed May 1, 2025. https://www.ceu.ox.ac.uk/research/the-million-women-study
  • 46. Sister Study. National Institute of Environmental Health Sciences. Accessed May 1, 2025. https://www.niehs.nih.gov/research/atniehs/labs/epi/studies/sister

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

djaf135_Supplementary_Data
djaf135_supplementary_data.docx (1.1MB, docx)

Data Availability Statement

Data on the harmonized NHANES BMI and reproductive history dose, model dose response calibrations, and model parameters are available at https://doi.org/10.5281/zenodo.15392555; any additional questions about data should be directed to the corresponding author.

Articles from JNCI Journal of the National Cancer Institute are provided here courtesy of Oxford University Press

RESOURCES