Among women with uterine cancer, there is a substantial and persistent thrombotic burden, particularly for those with advanced-stage disease.
Abstract
Using the Surveillance, Epidemiology, and End Results (SEER)-Medicare database, we identified women 65 years of age and older diagnosed with uterine cancer between 2008 and 2017 to evaluate the cumulative incidence of venous thromboembolism (VTE). Patients were followed from 6 months before to up to 10 years after diagnosis. A total of 683 patients with prevalent VTE (within 6 months before cancer diagnosis) were excluded. Among the remaining 25,204 patients with a median age of 72 years, 4,907 (19.5%) had VTE events during the follow-up period. The 10-year cumulative incidence of VTE increased with stage at diagnosis: 23.1% (95% CI 22.1–24.1%) for localized, 38.9% (95% CI 36.8–40.9%) for regional, and 56.5% (95% 51.0–61.7%) for distant disease. Our findings indicate a substantial and persistent thrombotic burden, particularly for patients with advanced-stage uterine cancer.
Uterine cancer is the most common gynecologic cancer in the United States, with 69,120 estimated new cases in 2025.1 Patients with cancer have a fourfold to sevenfold increase in risk for venous thromboembolism (VTE) as a result of tumor-specific procoagulant properties, host inflammatory responses, treatment-related factors, and patient characteristics.2 The true burden of VTE according to stage at diagnosis of uterine cancer is not well characterized. The objective of this study was to estimate the stage-specific, cumulative incidence of VTE in patients 65 years of age and older with uterine cancer.
METHODS
We conducted a retrospective cohort study using the Surveillance, Epidemiology, and End Results (SEER)–Medicare database. We identified women 65 years of age and older diagnosed with uterine cancer from 2008 to 2017. Subjects were followed up for up to 10 years. Venous thromboembolism, including deep vein thrombosis or pulmonary embolism, was identified with the use of billing codes from Medicare claims after diagnosis.3 Patients who had VTE claims within 6 months before cancer diagnosis were excluded. Cumulative incidence of VTE was estimated separately by stage (localized, regional, and distant disease) with 95% CIs. All analyses were performed with SAS 9.4.
RESULTS
Among 25,204 patients at risk, 71.4% had localized disease, 21.2% had regional disease, and 7.5% had distant disease. Most had a low Charlson comorbidity score of 0 or 1 (Table 1). Over the course of 10 years of follow-up, 4,907 patients (19.5%) developed VTE. Risk of VTE diverged early by stage: The cumulative incidence for patients with localized, regional, and distant disease was 6.6% (95% CI, 6.2–7.0%), 14.4% (95% CI, 13.4–15.4%), and 28.5% (95% CI, 26.3–30.7%) at 1 year; 9.3% (95% CI, 8.8–9.7%), 21.0% (95% CI, 19.9–22.2%), and 36.6% (95% CI, 34.1–39.1%) at 2 years; 15.4% (95% CI, 14.8–16.0%), 31.2% (95% CI, 29.7–32.7%), and 51.1% (95% CI, 47.3–54.7%) at 5 years; and 23.1% (95% CI, 22.1–24.1%), 38.9% (95% CI, 36.8–40.9%), and 56.5% (95% CI, 51.0–61.7%) at 10 years (Fig. 1).
Table 1.
Characteristics of Women Aged 65 and Older Who Were Diagnosed With Uterine Cancer From 2008 to 2017
| All | Stage | |||
| Localized | Regional | Distant | ||
| All | 25,204 (100.0) | 17,984 (71.4) | 5,337 (21.2) | 1,883 (7.5) |
| Age (y) | ||||
| 65–69 | 8,815 (35.0) | 6,546 (36.4) | 1,684 (31.6) | 585 (31.1) |
| 70–74 | 7,302 (29.0) | 5,254 (29.2) | 1,484 (27.8) | 564 (30.0) |
| 75–79 | 4,691 (18.6) | 3,219 (17.9) | 1,074 (20.1) | 398 (21.1) |
| 80 and older | 4,396 (17.4) | 2,965 (16.5) | 1,095 (20.5) | 336 (17.8) |
| Year of diagnosis | ||||
| 2008 | 2,436 (9.7) | 1,644 (9.1) | 624 (11.7) | 168 (8.9) |
| 2009 | 2,471 (9.8) | 1,691 (9.4) | 604 (11.3) | 176 (9.3) |
| 2010 | 2,364 (9.4) | 1,689 (9.4) | 489 (9.2) | 186 (9.9) |
| 2011 | 2,325 (9.2) | 1,610 (9.0) | 516 (9.7) | 199 (10.6) |
| 2012 | 2,475 (9.8) | 1,786 (9.9) | 509 (9.5) | 180 (9.6) |
| 2013 | 2,593 (10.3) | 1,838 (10.2) | 552 (10.3) | 203 (10.8) |
| 2014 | 2,543 (10.1) | 1,801 (10.0) | 565 (10.6) | 177 (9.4) |
| 2015 | 2,604 (10.3) | 1,862 (10.4) | 538 (10.1) | 204 (10.8) |
| 2016 | 2,749 (10.9) | 2,107 (11.7) | 462 (8.7) | 180 (9.6) |
| 2017 | 2,644 (10.5) | 1,956 (10.9) | 478 (9.0) | 210 (11.2) |
| Race and ethnicity | ||||
| AI/API, NH | 948 (3.8) | 657 (3.7) | 208 (3.9) | * |
| Black, NH | 2,009 (8.0) | 1,166 (6.5) | 568 (10.6) | 275 (14.6) |
| Hispanic | 1,360 (5.4) | 901 (5.0) | 324 (6.1) | 135 (7.2) |
| White, NH | 20,788 (82.5) | 15,181 (84.4) | 4,224 (79.1) | 1,383 (73.4) |
| Unknown | 99 (0.4) | 79 (0.4) | 13 (0.2) | * |
| Marital status | ||||
| Unmarried | 8,386 (33.3) | 5,860 (32.6) | 1,871 (35.1) | 655 (34.8) |
| Married | 8,265 (32.8) | 6,099 (33.9) | 1,588 (29.8) | 578 (30.7) |
| Unknown | 8,553 (33.9) | 6,025 (33.5) | 1,878 (35.2) | 650 (34.5) |
| Urban/rural | ||||
| Metropolitan | 21,842 (86.7) | 15,533 (86.4) | 4,637 (86.9) | 1,672 (88.8) |
| Urban | 3,009 (11.9) | 2,191 (12.2) | 636 (11.9) | 182 (9.7) |
| Rural | 353 (1.4) | 260 (1.4) | 64 (1.2) | 29 (1.5) |
| SEER region | ||||
| Eastern | 11,811 (46.9) | 8,399 (46.7) | 2,543 (47.6) | 869 (46.1) |
| Midwest | 6,269 (24.9) | 4,490 (25.0) | 1,293 (24.2) | 486 (25.8) |
| West | 7,124 (28.3) | 5,095 (28.3) | 1,501 (28.1) | 528 (28.0) |
| SES | ||||
| Low | 2,853 (11.3) | 1,898 (10.6) | 723 (13.5) | 232 (12.3) |
| Medium-low | 3,487 (13.8) | 2,442 (13.6) | 753 (14.1) | 292 (15.5) |
| Medium | 4,266 (16.9) | 3,069 (17.1) | 869 (16.3) | 328 (17.4) |
| Medium-high | 5,630 (22.3) | 4,047 (22.5) | 1,177 (22.1) | 406 (21.6) |
| High | 7,829 (31.1) | 5,698 (31.7) | 1,591 (29.8) | 540 (28.7) |
| Unknown | 1,139 (4.5) | 830 (4.6) | 224 (4.2) | 85 (4.5) |
| Comorbidity | ||||
| 0 | 10,386 (41.2) | 7,646 (42.5) | 2,096 (39.3) | 644 (34.2) |
| 1 | 7,193 (28.5) | 5,107 (28.4) | 1,530 (28.7) | 556 (29.5) |
| 2 or more | 7,625 (30.3) | 5,231 (29.1) | 1,711 (32.1) | 683 (36.3) |
| Histology | ||||
| Endometrioid | 16,641 (66.0) | 13,498 (75.1) | 2,718 (50.9) | 425 (22.6) |
| Nonendometrioid | 4,830 (19.2) | 2,128 (11.8) | 1,699 (31.8) | 1,003 (53.3) |
| Sarcoma | 531 (2.1) | 327 (1.8) | 87 (1.6) | 117 (6.2) |
| EM NOS | 2,884 (11.4) | 1,840 (10.2) | 772 (14.5) | 272 (14.4) |
| Other | 318 (1.3) | 191 (1.1) | 61 (1.1) | 66 (3.5) |
| Grade | ||||
| Well | 6,927 (27.5) | 6,263 (34.8) | 615 (11.5) | 49 (2.6) |
| Moderate | 5,696 (22.6) | 4,378 (24.3) | 1,175 (22.0) | 143 (7.6) |
| Poor | 7,474 (29.7) | 3,741 (20.8) | 2,516 (47.1) | 1,217 (64.6) |
| Unknown | 5,107 (20.3) | 3,602 (20.0) | 1,031 (19.3) | 474 (25.2) |
AI, American Indian; API, Asian and other Pacific Islanders; NH, non-Hispanic; SEER, Surveillance, Epidemiology, and End Results; SES, socioeconomic status; EM NOS, endometrioid not otherwise specified.
Values are n (%).
The first row was row percentage, and the rest were column percentage.
These groups were combined because of small cell counts.
Fig. 1. Cumulative incidence rate of venous thromboembolism (VTE) by localized, regional and distant stage (P<.001). Shading represents 95% CIs.
Sassine. Venous thromboembolism in Patients with Uterine Cancer. O&G Open 2026.
DISCUSSION
Our data suggest that VTE is common in patients with uterine cancer across all stages of disease, with a markedly higher cumulative incidence observed in patients with distant disease. Although VTE risks peak during the treatment period, later events, particularly in patients with localized disease who are likely cured, may reflect noncancer causes, and our 10-year estimates represent cumulative observed burden rather cancer-attributable risk.
A previous SEER-Medicare study reported a VTE incidence of 7.2% within 24 months of cancer diagnosis among 23,122 patients with uterine cancer,4 with a higher rate among patients with aggressive histologic subtypes. Patients with VTE had a higher rate of stage IV disease. In our cohort, VTE incidence at 24 months was modestly higher than previously reported, consistent with our stage-stratified design. A recent study evaluating only patients with advanced uterine cancer (stage III and IV) receiving neoadjuvant or adjuvant chemotherapy found an overall incidence of 27.4% at 1 year of follow-up.5 This is in concordance with our data that showed a cumulative incidence of 28.5% at 1 year for patients with distant disease. Although other investigators have reported a much lower total incidence of VTE (1.4%) in 26,256 patients with uterine cancer over a follow-up of 47 months,6 it was based on a different population, a cohort of Korean patients with a much younger mean age of 54 years.
The highest rise in the incidence of VTE occurred within the first year after cancer diagnosis among all stage groups. This suggests that the highest risk of VTE occurs at the time of primary treatment and aligns with prior work noting that VTE occurred in 54 of 10,000 women during the first 6 months after primary treatment initiation compared with 136 of 10,000 during follow-up of 24 months.6
We recognize several study limitations. First, by using the Medicare database, our analysis was limited to women 65 years of age or older with unmeasured confounders, including but not limited to body mass index (BMI), VTE history, anticoagulant use, and treatment-related factors, that may contribute to VTE risk. Second, VTE may be undercaptured with claims data. Likewise, there may be misclassification of a small number of patients with a history of VTE as having new, incident events when relying on administrative codes and claims data. Third, we did not perform histology-stratified analyses and acknowledge that VTE rates may differ according to tumor histology and other disease characteristics. Despite these limitations, our study provides important data suggesting a high incidence of VTE in patients with uterine cancer, particularly in those with distant disease at the time of diagnosis.
Footnotes
Financial Disclosure Dr. Wright has received royalties from UpToDate, honoraria from the American College of Obstetricians and Gynecologists, and research support from Merck. Dr. Xu has received honoraria and travel funding from the American Association of Gynecologic Laparoscopists. The other authors did not report any potential conflicts of interest.
Jason D. Wright, Editor-in-Chief of Obstetrics & Gynecology, was not involved in the review or decision to publish this article.
Each author has confirmed compliance with the journal's requirements for authorship.
Peer reviews and author correspondence are available at https://links.lww.com/AOG/E786.
Dr. Wright and Dr. Hou are co–first authors.
REFERENCES
- 1.National Cancer Institute. SEER preliminary cancer incidence rate estimates for diagnosis years 2000 to 2023, 2025. Accessed August 13, 2025. https://seer.cancer.gov/statistics/preliminary-estimates/ [Google Scholar]
- 2.Ay C, Pabinger I, Cohen AT. Cancer-associated venous thromboembolism: burden, mechanisms, and management. Thromb Haemost 2017;117:219–30. doi: 10.1160/TH16-08-0615 [DOI] [PubMed] [Google Scholar]
- 3.Agency for Healthcare Research and Quality. AHRQ patient safety indicator technical documentation, version v2025, Accessed August 13, 2025. https://qualityindicators.ahrq.gov/measures/psi_resources [Google Scholar]
- 4.Rauh-Hain JA, Hariton E, Clemmer J, Clark RM, Hall T, Boruta DM, et al. Incidence and effects on mortality of venous thromboembolism in elderly women with endometrial cancer. Obstet Gynecol 2015;125:1362–70. doi: 10.1097/AOG.0000000000000866 [DOI] [PubMed] [Google Scholar]
- 5.Llanora JD, Uppal S, McLaughlin HD. Incidence and timing of venous thromboembolism in patients with advanced endometrial cancer. Obstet Gynecol 2025;146:293–6. doi: 10.1097/AOG.0000000000005978 [DOI] [PubMed] [Google Scholar]
- 6.Yuk JS, Lee B, Kim K, Kim MH, Seo YS, Hwang SO, et al. Incidence and risk of venous thromboembolism according to primary treatment type in women with endometrial cancer: a population-based study. BMC Cancer 2021;21:1166. doi: 10.1186/s12885-021-08853-x [DOI] [PMC free article] [PubMed] [Google Scholar]