Abstract
Objective
To evaluate the association between immunohistochemically (IHC) defined molecular subtypes and substantial lymphovascular space invasion (LVSI) in early-stage endometrial cancer (EC) according to the International Federation of Gynecology and Obstetrics (FIGO) 2023 classification.
Methods
This study included 575 patients who underwent primary surgery for FIGO stage I–II EC between January 2023 and April 2025. LVSI was assessed using World Health Organization criteria and classified as no/focal (Group 1, n=520) or substantial (Group 2, n=55). Molecular subtypes were determined using a surrogate IHC-based approach (no specific molecular profile, mismatch repair-deficient [MMRd], p53-abnormal [p53abn]); POLE mutation analysis was performed in a subset of patients. Clinicopathological variables were compared between groups, and independent predictors of substantial LVSI were identified using multivariable logistic regression.
Results
Substantial LVSI was observed in 9.6% of patients (n=55). MMRd (18.5% vs. 34.5%) and p53abn (11% vs. 20%) subtypes were more frequent in the substantial LVSI group (p=0.001). Deep myometrial invasion (≥50%) (21.9% vs. 70.9%, p<0.001), cervical stromal invasion (4.8% vs. 16.4%, p=0.003), and adjuvant therapy (34.6% vs. 92.7%, p<0.001) were also associated with substantial LVSI. Multivariable analysis identified MMRd subtype (adjusted odds ratio [aOR]=2.916; 95% confidence interval [CI]=1.466–5.801; p=0.002) and ≥50% myometrial invasion (aOR=7.935; 95% CI=4.173–15.091; p<0.001) as independent predictors.
Conclusion
In early-stage EC, substantial LVSI is associated with the MMRd molecular subtype and deep myometrial invasion. Understanding the molecular tumor biology underlying LVSI may facilitate more individualized risk stratification and treatment strategies.
Keywords: DNA Mismatch Repair, Endometrial Neoplasms, Immunohistochemistry, Neoplasm Invasiveness, Neoplasm Staging
Synopsis
This study evaluated molecular subtypes and substantial lymphovascular space invasion (LVSI) in 575 patients with early-stage endometrial cancer. Mismatch repair-deficient tumors were independently associated with substantial LVSI after adjustment for clinicopathological factors.
INTRODUCTION
The diagnostic approach, prognostic assessment, and treatment strategies for endometrial cancer (EC) have undergone substantial changes over the past decade. One of the most significant advances has been the translation of the 4 molecular subtypes identified by The Cancer Genome Atlas (TCGA) into clinical practice through immunohistochemistry-based surrogate methods, particularly via the ProMisE algorithm [1,2]. This molecular classification was incorporated into the World Health Organization (WHO) tumor classification in 2020, subsequently integrated into the 2023 International Federation of Gynecology and Obstetrics (FIGO) staging system for surgical and prognostic assessment, and most recently established as a core component of risk stratification in the 2025 European Society of Gynecological Oncology/European Society for Radiotherapy and Oncology/European Society of Pathology guidelines [3,4,5].
According to the WHO classification, EC is divided into 4 molecular subtypes: ultramutated POLE-mutated tumors, mismatch repair-deficient (MMRd) tumors, p53-abnormal (p53abn) tumors, and tumors with no specific molecular profile (NSMP). While POLE mutations require DNA sequencing analysis, mismatch repair (MMR) protein expression and p53 status can be reliably assessed using immunohistochemical methods, allowing molecular classification to be widely implemented in routine clinical practice [3].
LVSI has long been recognized as an adverse prognostic factor in EC. The PORTEC trials clearly demonstrated that the presence of extensive, or substantial, LVSI is a strong and independent predictor of pelvic regional recurrence, distant metastasis, and overall survival [6]. Consequently, LVSI has become an integral component of contemporary risk stratification systems, including the current guidelines [4,5]. However, despite the increasing prognostic relevance of molecular subtypes, the biological basis of LVSI and its association with specific molecular profiles remain incompletely understood.
Therefore, the present study aims to evaluate the relationship between substantial LVSI and immunohistochemically (IHC) assessed molecular subtypes in patients with early-stage EC staged according to the FIGO 2023 criteria. Elucidating this association may contribute to a better understanding of the biological background of LVSI and further refine the clinical value of molecular classification in risk stratification.
MATERIALS AND METHODS
1. Study population
Following approval by the Ankara Etlik City Hospital Ethics Committee (approval No. AEŞH-BADEK-2025-0080), a total of 575 patients with early-stage EC (FIGO 2023 stage I–II) who underwent surgery at the gynecologic oncology clinic from January 2023 to April 2025 were included in this study. Eligible patients had received primary treatment consisting of total hysterectomy and bilateral salpingo-oophorectomy, had LVSI evaluated according to a 3-tier classification in the pathology report, and had IHC assessment of p53 and MMR proteins (MLH1, PMS2, MSH2, and MSH6).
Patients with FIGO stage III–IV disease, those who received neoadjuvant therapy, cases with incomplete pathological evaluation, unclear reporting of LVSI or IHC results, or unavailable medical records were excluded. Written informed consent was obtained from all patients.
2. Surgical procedure
All patients underwent surgical staging, including total hysterectomy and bilateral salpingo-oophorectomy. Nodal assessment was performed using either sentinel lymph node (SLN) mapping or intraoperative frozen-section assessment. In cases evaluated with frozen section, the indication for lymph node dissection was determined according to the Mayo criteria [7]. The surgical approach was determined according to preoperative clinical assessment and imaging findings and included open (laparotomy) or minimally invasive techniques (laparoscopic, robotic, or vaginal/vNOTES).
3. Pathological evaluation
All pathological specimens were reviewed by 2 experienced gynecologic pathologists. LVSI was graded according to the WHO 3-tier system as follows: absence of LVSI, defined as no vascular invasion identified on any slide; focal LVSI, defined as involvement of 1–4 vessels on at least one slide; and substantial LVSI, defined as involvement of ≥5 vessels on at least one slide [3]. Based on LVSI status, patients were categorized into 2 groups: Group 1, comprising patients with no/focal LVSI (n=520), and Group 2, comprising patients with substantial LVSI (n=55).
All cases were staged according to the FIGO 2023 staging system. Stage I was defined as disease confined to the uterine corpus. Stage II was defined as cervical stromal invasion without extrauterine extension, the presence of substantial LVSI, or aggressive histological types with myometrial invasion. According to the FIGO 2023 classification, stage II was further subdivided into stage IIA (cervical stromal invasion of non-aggressive histological types), stage IIB (substantial LVSI of non-aggressive histological types), and stage IIC (aggressive histological types with any myometrial involvement) [4].
4. Molecular classification
Immunohistochemical markers have been routinely evaluated at our institution since January 2023. Loss of expression of one or more MMR proteins (MLH1, MSH2, MSH6, or PMS2) was classified as MMRd. Diffuse strong nuclear overexpression or complete absence (null pattern) of p53 staining was defined as p53abn.
POLE mutation analysis was not routinely performed due to its cost and technical complexity and was therefore available only in a limited subset of patients. Testing was conducted in selected cases, particularly in FIGO 2023 stage II patients, where additional molecular information was considered relevant for adjuvant treatment decision-making, and in patients who consented to further testing. As POLE status was not available for all patients, molecular subtyping was based on immunohistochemical assessment of MMR and p53 and does not represent a complete TCGA-style classification. Tumors with intact MMR protein expression and wild-type p53 staining were classified as NSMP in this study [3].
Decisions regarding adjuvant treatment were made by a multidisciplinary tumor board based on pathological findings.
5. Clinicopathological variables
Clinical variables including age, body mass index (BMI), parity, serum cancer antigen 125 level, and tumor diameter were recorded. Molecular subtype, histological type (endometrioid vs. non-endometrioid), depth of myometrial invasion (<50% vs. ≥50%), presence of cervical stromal invasion (negative vs. positive), performance of lymph node assessment, surgical approach (laparotomy vs. minimally invasive), and receipt of adjuvant therapy were compared according to LVSI status.
FIGO 2023 stage was evaluated descriptively and was not formally compared between LVSI groups because substantial LVSI is incorporated into the FIGO 2023 stage II definition.
6. Statistical analysis
Statistical analyses were performed using IBM SPSS Statistics for Mac, version 22.0 (IBM Corp., Armonk, NY, USA). Continuous variables were summarized using median (range), minimum, and maximum values, while categorical variables were presented as frequencies and percentages. Comparisons between groups were conducted using the χ2 test or Fisher’s exact test, as appropriate.
Variables with a p-value <0.10 in univariable analysis, as well as clinically relevant factors, were considered for inclusion in the multivariable logistic regression analysis. Given the limited number of patients with substantial LVSI (n=55), the number of variables included in the final model was restricted to preserve model stability and reduce the risk of overfitting. Priority was given to molecular and pathological variables directly related to the predefined study hypothesis, including molecular subtype, histological type, depth of myometrial invasion, and cervical stromal invasion. Although age and tumor diameter differed between groups in the univariable analysis, they were not included in the final model to preserve model stability and reduce the risk of overfitting.
FIGO stage was not included in the multivariable model, because substantial LVSI is incorporated into the definition of stage II in the FIGO 2023 staging system. Therefore, including FIGO stage in the model would represent a definition-driven circular association. Results were reported as odds ratios (ORs) with 95% confidence intervals (CIs), and a p-value <0.05 was considered statistically significant.
RESULTS
The median age of the study cohort was 60 years (31–85), with a median parity of 3 (0–11), median BMI of 32 kg/m2 (19–64), and median tumor size of 33 mm (2–180). LVSI was absent in 85.6% of patients (n=492), while focal LVSI was observed in 4.9% (n=28) and substantial LVSI in 9.6% (n=55). According to the FIGO 2023 classification, 75% of patients (n=429) had stage I disease and 25% (n=146) had stage II disease (Table 1).
Table 1. Patient characteristics.
| Variables | ||
|---|---|---|
| No. of patients | 575 | |
| Age (yr) | 60 (31–85) | |
| Parity | 3 (0–11) | |
| BMI (kg/m2) | 32 (19–64) | |
| CA-125 | 13 (3–4,290) | |
| Tumor size (mm) | 33 (2–180) | |
| LVSI | ||
| No | 492 (85.6) | |
| Focal | 28 (4.9) | |
| Substantial | 55 (9.6) | |
| FIGO 2023 | ||
| Stage I | 429 (75) | |
| Stage II | 146 (25) | |
Data are expressed as median (minimum–maximum) or number (%).
BMI, body mass index; CA-125, cancer antigen 125; FIGO, International Federation of Gynecology and Obstetrics; LVSI, lymphovascular space invasion.
When molecular subtype distributions were compared, 0.8% of patients in Group 1 had POLE-mutated tumors (n=4), 69.8% had NSMP tumors (n=363), 18.5% had MMRd tumors (n=96), and 11% had p53abn tumors (n=57). In Group 2, no POLE-mutated tumors were identified; the proportions of NSMP, MMRd, and p53abn tumors were 45.5% (n=25), 34.5% (n=19), and 20% (n=11), respectively (p=0.001). Endometrioid histology was observed in 87.7% of patients in Group 1 (n=456) and 81.8% of patients in Group 2 (n=45) (p=0.216). POLE mutation analysis was available in 40 patients, and POLE mutations were identified in 4 patients (10%). The distribution of molecular subtypes according to LVSI status among POLE-tested patients is presented descriptively in Table S1.
Deep myometrial invasion (≥50%) was observed in 21.9% of patients in Group 1 (n=114) and in 70.9% of patients in Group 2 (n=39) (p<0.001). Cervical stromal invasion was detected in 4.8% of patients in Group 1 (n=25) and 16.4% of patients in Group 2 (n=9) (p=0.003). Regarding surgical approach, laparotomy was performed in 21.9% of patients in Group 1 and 40% of patients in Group 2, whereas minimally invasive surgery was used in 78.1% and 60% of patients, respectively (p=0.003). The rate of adjuvant treatment was significantly higher in Group 2 than in Group 1 (92.7% vs. 34.6%, p<0.001). Lymph node assessment was performed in 75.2% of patients in Group 1 (n=391) and 92.7% of patients in Group 2 (n=51) (p=0.003). Among patients who underwent nodal evaluation, SLN mapping was performed in 41.2% (161/391) and 31.4% (16/51), respectively (Table 2). The distribution of FIGO 2023 stage according to LVSI status is presented descriptively in Table S2, without statistical comparison.
Table 2. Histopathological and clinical characteristics according to the degree of LVSI.
| Characteristics | Group 1 (no/focal LVSI; n=520) | Group 2 (substantial LVSI; n=55) | p | |
|---|---|---|---|---|
| Age (yr) | 59 (31–85) | 63 (44–79) | 0.006 | |
| CA-125 | 12.8 (3–4,290) | 15 (3–4,290) | 0.509 | |
| Tumor size (mm) | 30 (2–180) | 50 (14–130) | <0.001 | |
| Molecular subtypes | 0.001 | |||
| POLEmut | 4 (0.8) | 0 | ||
| NSMP | 363 (69.8) | 25 (45.5) | ||
| MMRd | 96 (18.5) | 19 (34.5) | ||
| p53abn | 57 (11) | 11 (20) | ||
| Histological type | 0.216 | |||
| Endometrioid | 456 (87.7) | 45 (81.8) | ||
| Non-endometrioid | 64 (12.3) | 10 (18.2) | ||
| Myometrial invasion | <0.001 | |||
| <50% | 406 (78.1) | 16 (29.1) | ||
| ≥50% | 114 (21.9) | 39 (70.9) | ||
| Cervical stromal invasion | 0.003 | |||
| Negative | 495 (95.2) | 46 (83.6) | ||
| Positive | 25 (4.8) | 9 (16.4) | ||
| Lymph node assessment | 0.003 | |||
| No | 129 (24.8) | 4 (7.3) | ||
| Yes | 391 (75.2) | 51 (92.7) | ||
| Surgical approach | 0.003 | |||
| Laparotomy | 114 (21.9) | 22 (40) | ||
| MIS | 406 (78.1) | 33 (60) | ||
| Adjuvant treatment | <0.001 | |||
| No | 340 (65.4) | 4 (7.3) | ||
| Yes | 180 (34.6) | 51 (92.7) | ||
Data are expressed as median (minimum–maximum) or number (%). The p≤0.05 indicates statistically significant difference. Bold values indicate statistical significance (p<0.05). FIGO 2023 stage was not statistically compared between LVSI groups because substantial LVSI is incorporated into the FIGO 2023 stage II definition.
CA-125, cancer antigen 125; FIGO, International Federation of Gynecology and Obstetrics; LVSI, lymphovascular space invasion; MIS, minimally invasive surgery; MMRd, mismatch repair-deficient; NSMP, no specific molecular profile; p53abn, p53-abnormal.
In multivariable logistic regression analysis, MMRd tumors were independently associated with substantial LVSI compared with the NSMP subtype (adjusted odds ratio [aOR]=2.92, 95% CI=1.46–5.80, p=0.002). Deep myometrial invasion (≥50%) was also identified as an independent predictor of substantial LVSI (aOR=7.94; 95% CI=4.17–15.09; p<0.001). Although p53abn tumors showed a trend toward higher rates of substantial LVSI, this association did not reach statistical significance (aOR=2.79; 95% CI=0.99–7.85; p=0.052). Histological subtype and cervical stromal invasion were not independently associated with substantial LVSI (p=0.668 and p=0.204, respectively) (Table 3).
Table 3. Independent predictors of substantial LVSI: multivariable logistic regression analysis.
| Variables | aOR | 95% CI | p-value |
|---|---|---|---|
| MMRd (vs. NSMP) | 2.916 | 1.466–5.801 | 0.002 |
| p53abn (vs. NSMP) | 2.787 | 0.989–7.850 | 0.052 |
| Non-endometrioid histology (vs. endometrioid) | 1.253 | 0.447–3.507 | 0.668 |
| Myometrial invasion ≥50% (vs. <50%) | 7.935 | 4.173–15.091 | <0.001 |
| Cervical stromal invasion (vs. negative) | 1.815 | 0.723–4.553 | 0.204 |
Bold values indicate statistical significance (p<0.05).
aOR, adjusted odds ratio; CI, confidence interval; LVSI, lymphovascular space invasion; MMRd, mismatch repair-deficient; NSMP, no specific molecular profile; p53abn, p53-abnormal.
DISCUSSION
In this study, we found that MMRd is independently associated with the presence of substantial LVSI in early-stage EC, highlighting an important molecular determinant underlying lymphovascular dissemination in this disease. Although deep myometrial invasion was also identified as an independent predictor of substantial LVSI, the consistent association observed in MMRd tumors underscores the relevance of integrating molecular markers into the evaluation of LVSI pathogenesis.
MMR deficiency represents one of the key molecular subtypes influencing both biological behavior and prognostic outcomes in EC. Previous studies have shown that MMRd tumors are characterized by a pronounced inflammatory tumor microenvironment, increased tumor-infiltrating lymphocytes, and enhanced immune activation; these biological features may facilitate closer interactions between tumor cells and stromal or vascular structures, thereby predisposing tumors to the development of LVSI [8,9,10].
Nevertheless, studies systematically evaluating the relationship between molecular subgroups and LVSI remain limited. Siegenthaler et al. [11] reported a higher prevalence of MMRd tumors among patients with extensive LVSI, although the highest proportion of substantial LVSI was observed in the p53abn subgroup. In their analysis of the prognostic impact of LVSI across molecular subtypes, the presence of LVSI was associated with recurrence in the MMRd, NSMP, and p53abn groups. Similarly, in a study by Loukovaara et al. [12] focusing on early-stage endometrioid-type EC, the prognostic significance of LVSI varied according to molecular subtype: both focal and substantial LVSI were associated with survival outcomes in the MMRd subgroup, whereas only substantial LVSI was prognostic in the NSMP subgroup, and no degree of LVSI demonstrated prognostic relevance in p53abn tumors. Notably, the highest rate of substantial LVSI was observed in the MMRd group (15.1%). In line with these findings, although substantial LVSI rates were comparable between the MMRd and p53abn subtypes in our cohort, only the MMRd subtype remained independently associated with substantial LVSI in multivariable analysis. These results suggest that the biological characteristics of MMRd tumors may contribute to LVSI development and that evaluation of the MMRd subtype may provide additional prognostic value in risk stratification, even among patients with early-stage disease.
Ongoing debate persists regarding the definition and interpretation of LVSI in EC. While the WHO defines substantial LVSI as involvement of at least 5 vessels, Peters et al. [13] proposed a threshold of 4 or more involved vessels [3]. Similarly, several retrospective studies, including those by Bhatnagar et al. [14], have considered involvement of 3 or more vessels as clinically significant LVSI [15]. These discrepancies in definition represent a major source of heterogeneity across studies. Another important challenge is the substantial interobserver variability in LVSI assessment. In a study conducted among gynecologic pathologists, complete agreement regarding LVSI extent was achieved in only 14% of cases, highlighting the subjective nature of LVSI evaluation [16]. Consequently, reported rates of focal and substantial LVSI vary widely in the literature, ranging from approximately 5% to as high as 15% in different series [11,12,14,17]. In our cohort, the frequencies of focal LVSI (4.9%, n=28) and substantial LVSI (9.6%, n=55) were consistent with previously reported data.
It should also be emphasized that the distribution of FIGO 2023 stage according to LVSI status was interpreted descriptively in this study. Under the FIGO 2023 staging system, substantial LVSI itself is incorporated into the definition of stage II disease. Therefore, the high frequency of FIGO 2023 stage II disease in the substantial LVSI group represents a definition-driven circular relationship rather than an independent clinicopathological association. For this reason, FIGO stage was not statistically compared between LVSI groups and was not included in the multivariable model.
The strengths of this study include a large cohort staged according to the FIGO 2023 classification, standardized application of an IHC-based molecular panel across all cases, and 3-tier LVSI assessment performed by 2 experienced gynecologic pathologists in accordance with WHO criteria. These features enabled a robust comparison of molecular and histopathological parameters. However, several limitations of this study should be acknowledged. First, the relatively small number of patients in the substantial LVSI group may have limited the statistical power of the multivariable analysis. In addition, the events-per-variable ratio was in the marginal range, which may increase the risk of model instability and overestimation of ORs. In particular, although the estimated effect size for p53abn tumors was comparable to that observed for MMRd tumors, statistical significance was not reached and the CI remained wide. Post hoc power analysis suggested adequate power for the MMRd comparison (power=0.86), whereas power was more limited for the p53abn comparison (power=0.68). These findings suggest that the borderline result for p53abn may reflect limited sample size and insufficient statistical power rather than the absence of a true association.
In addition, POLE mutation analysis was not available for all patients and therefore could not be incorporated into the multivariable model. As a result, some POLE-mutated tumors may have been misclassified as NSMP, particularly among tumors with intact MMR expression and wild-type p53 staining. This potential contamination of the NSMP reference group may have influenced the estimated effect sizes of molecular subgroups, including a possible overestimation of the relative effect of MMRd. Although the distribution of molecular subtypes according to LVSI status among POLE-tested cases was evaluated descriptively, this subgroup was small and consisted of selected cases; therefore, a robust sensitivity analysis was not feasible, and the findings from this subgroup should be interpreted with caution. Furthermore, POLE mutation testing using next-generation sequencing is a costly technique and is not routinely implemented in many centers worldwide, and the prevalence of POLE-mutated EC is generally low [18,19,20]. LVSI assessment is subject to interobserver variability, and formal agreement analysis was not performed in this study. Although all cases were reviewed by experienced gynecologic pathologists and consensus was reached, the absence of quantitative interobserver agreement data may be considered a limitation of the study. Finally, survival analysis could not be performed due to the relatively short follow-up period, precluding evaluation of the prognostic impact of molecular profiles in patients with LVSI.
In conclusion, our findings suggest that substantial LVSI in early-stage EC is associated with the MMRd subtype and deep myometrial invasion (≥50%). Substantial LVSI may reflect a more aggressive tumor biology that is influenced not only by conventional histopathological factors but also by underlying molecular characteristics. Incorporating LVSI assessment with immunohistochemistry-based molecular classification may enhance risk stratification and support more tailored adjuvant treatment decisions in early-stage EC. Nevertheless, given the heterogeneity of existing data and the limited number of available studies, larger prospective investigations with comprehensive molecular profiling are warranted to better elucidate the biological mechanisms underlying LVSI.
ACKNOWLEDGEMENTS
The authors would like to thank all individuals who contributed to the conduct of this study.
Footnotes
Conflict of Interest: No potential conflict of interest relevant to this article was reported.
- Conceptualization: Ö.H.N., K.V.
- Data curation: Ö.H.N., B.Ş.K., Ö.N., E.G., K.O.K., Ş.D.F., H.C.
- Formal analysis: Ö.H.N., B.Ş.K., Ş.D.F., K.V.
- Investigation: Ö.H.N., Ş.D.F., H.C.
- Methodology: Ö.H.N., K.V.
- Resources: B.Ş.K.
- Supervision: K.V.
- Writing - original draft: Ö.H.N.
- Writing - review & editing: Ö.H.N.
SUPPLEMENTARY MATERIALS
Distribution of molecular subtypes according to LVSI status among patients with available POLE mutation analysis (n=40)
Distribution of FIGO 2023 stage according to LVSI status
References
- 1.Cancer Genome Atlas Research Network. Kandoth C, Schultz N, Cherniack AD, Akbani R, Liu Y, et al. Integrated genomic characterization of endometrial carcinoma. Nature. 2013;497:67–73. doi: 10.1038/nature12113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Kommoss S, McConechy MK, Kommoss F, Leung S, Bunz A, Magrill J, et al. Final validation of the ProMisE molecular classifier for endometrial carcinoma in a large population-based case series. Ann Oncol. 2018;29:1180–1188. doi: 10.1093/annonc/mdy058. [DOI] [PubMed] [Google Scholar]
- 3.Höhn AK, Brambs CE, Hiller GGR, May D, Schmoeckel E, Horn LC. 2020 WHO classification of female genital tumors. Geburtshilfe Frauenheilkd. 2021;81:1145–1153. doi: 10.1055/a-1545-4279. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Berek JS, Matias-Guiu X, Creutzberg C, Fotopoulou C, Gaffney D, Kehoe S, et al. FIGO staging of endometrial cancer: 2023. Int J Gynaecol Obstet. 2023;162:383–394. doi: 10.1002/ijgo.14923. [DOI] [PubMed] [Google Scholar]
- 5.Concin N, Matias-Guiu X, Cibula D, Colombo N, Creutzberg CL, Ledermann J, et al. ESGO-ESTRO-ESP guidelines for the management of patients with endometrial carcinoma: update 2025. Lancet Oncol. 2025;26:e423–e435. doi: 10.1016/S1470-2045(25)00167-6. [DOI] [PubMed] [Google Scholar]
- 6.Bosse T, Peters EEM, Creutzberg CL, Jürgenliemk-Schulz IM, Jobsen JJ, Mens JWM, et al. Substantial lymph-vascular space invasion (LVSI) is a significant risk factor for recurrence in endometrial cancer--a pooled analysis of PORTEC 1 and 2 trials. Eur J Cancer. 2015;51:1742–1750. doi: 10.1016/j.ejca.2015.05.015. [DOI] [PubMed] [Google Scholar]
- 7.Mariani A, Dowdy SC, Cliby WA, Gostout BS, Jones MB, Wilson TO, et al. Prospective assessment of lymphatic dissemination in endometrial cancer: a paradigm shift in surgical staging. Gynecol Oncol. 2008;109:11–18. doi: 10.1016/j.ygyno.2008.01.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Ramchander NC, Ryan NAJ, Walker TDJ, Harries L, Bolton J, Bosse T, et al. Distinct immunological landscapes characterize inherited and sporadic mismatch repair deficient endometrial cancer. Front Immunol. 2020;10:3023. doi: 10.3389/fimmu.2019.03023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Guo J, Tang B, Fu J, Zhu X, Xie W, Wang N, et al. High-plex spatial transcriptomic profiling reveals distinct immune components and the HLA class I/DNMT3A/CD8 modulatory axis in mismatch repair-deficient endometrial cancer. Cell Oncol (Dordr) 2024;47:573–585. doi: 10.1007/s13402-023-00885-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Kir G, Olgun ZC, Soylemez T, Aydin A, Demircan B, Kaya IA, et al. PD-L1 expression in mismatch repair-deficient endometrial carcinoma and tumor-associated immune cells: differences between MLH1 methylated and nonmethylated subgroups. Int J Gynecol Pathol. 2021;40:575–586. doi: 10.1097/PGP.0000000000000750. [DOI] [PubMed] [Google Scholar]
- 11.Siegenthaler F, Epstein E, Büchi CA, Gmür A, Saner FACM, Rau TT, et al. Prognostic value of lymphovascular space invasion according to the molecular subgroups in endometrial cancer. Int J Gynecol Cancer. 2023;33:1702–1707. doi: 10.1136/ijgc-2023-004606. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Loukovaara M, Pasanen A, Bützow R. Molecular subgroup-specific prognostic value of semiquantitative lymphovascular space invasion in early-stage endometrioid endometrial cancer. Gynecol Oncol. 2025;197:96–101. doi: 10.1016/j.ygyno.2025.04.589. [DOI] [PubMed] [Google Scholar]
- 13.Peters EEM, León-Castillo A, Smit VTHB, Boennelycke M, Hogdall E, Hogdall C, et al. Defining substantial lymphovascular space invasion in endometrial cancer. Int J Gynecol Pathol. 2022;41:220–226. doi: 10.1097/PGP.0000000000000806. [DOI] [PubMed] [Google Scholar]
- 14.Bhatnagar AR, Ghanem AI, Alkamachi B, Allo G, Lin CH, Hijaz M, et al. The prognostic impact of substantial lymphovascular space invasion in women with node negative FIGO stage I uterine carcinoma. Gynecol Oncol. 2024;188:44–51. doi: 10.1016/j.ygyno.2024.06.009. [DOI] [PubMed] [Google Scholar]
- 15.Pifer PM, Jaishankar S, Bhargava R, Schad MD, Keller A, Musunuru HB, et al. Is substantial lymphovascular space invasion prognostic in patients with pathologically lymph node-negative endometrial cancer? Int J Radiat Oncol Biol Phys. 2023;117:148–153. doi: 10.1016/j.ijrobp.2023.02.053. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Peters EEM, Bartosch C, McCluggage WG, Genestie C, Lax SF, Nout R, et al. Reproducibility of lymphovascular space invasion (LVSI) assessment in endometrial cancer. Histopathology. 2019;75:128–136. doi: 10.1111/his.13871. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Öncü HN, Ege G, Öztürk N, Köksal OK, Körpe B, Hanedan C, et al. The relationship between lymph node metastasis and immunohistochemical molecular subtypes in endometrial cancer: a cohort study of 339 patients. J Obstet Gynaecol Res. 2025;51:e70065. doi: 10.1111/jog.70065. [DOI] [PubMed] [Google Scholar]
- 18.Lim JWJ, Dzyubak O, Moshkovich M, Maganti M, Han K, Hodgson A, et al. Impact of FIGO 2023 staging criteria on stage migration and survival outcomes in early-stage endometrial cancer: a retrospective cohort study. Gynecol Oncol. 2025;202:93–101. doi: 10.1016/j.ygyno.2025.09.015. [DOI] [PubMed] [Google Scholar]
- 19.Zheng S, Donnelly ED, Strauss JB. Race, prevalence of POLE and POLD1 alterations, and survival among patients with endometrial cancer. JAMA Netw Open. 2024;7:e2351906. doi: 10.1001/jamanetworkopen.2023.51906. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Capasso I, Perrone E, Loverro M, Giannarelli D, Nero C, Tortorella L, et al. Focal lymphovascular space invasion: friend or foe? A large retrospective analysis on stage I endometrioid endometrial carcinomas. Eur J Cancer. 2025;228:115736. doi: 10.1016/j.ejca.2025.115736. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Distribution of molecular subtypes according to LVSI status among patients with available POLE mutation analysis (n=40)
Distribution of FIGO 2023 stage according to LVSI status