Background/Objectives: Lymphovascular space invasion (LVSI) is an established prognostic factor in endometrial cancer, yet current risk stratification relies on a binary focal/substantial classification. Whether further substratification into a “prominent” LVSI category (≥10 involved vessels) provides independent prognostic value beyond the FIGO 2023 staging framework remains unclear. We aimed to investigate the prognostic significance and clinical outcomes regarding the extent of lymphovascular space invasion (LVSI) in patients with LVSI-positive endometrioid endometrial cancer (EEC). Methods: We retrospectively identified 94 patients with International Federation of Gynecology and Obstetrics (FIGO) 2009 stage I–IV EEC who underwent surgical staging. LVSI-positive cases were classified as focal (1–4 involved vessels), substantial (5–9 involved vessels), or prominent (≥10 involved vessels). Results: Of the 94 LVSI-positive patients, 58 (61.7%) had focal, 11 (11.7%) had substantial, and 25 (26.6%) had prominent LVSI. Five-year overall survival (OS) rates were 95.3%, 72.7%, and 53.0% for focal, substantial, and prominent LVSI, respectively (p < 0.001), with corresponding progression-free survival rates of 87.6%, 71.6%, and 34.3% (p < 0.001). On multivariable analysis, prominent LVSI emerged as an independent predictor of both mortality (HR 14.31, p < 0.001) and disease progression (HR 4.02, p = 0.009), retaining significance after re-assessment for FIGO 2023 stage. Tumor size > 4 cm and p53-abnormal status were also independently associated with OS. Conclusions: Prominent LVSI was independently associated with aggressive disease and poor survival in EEC; recognizing its extent as a prognostic parameter may improve both clinical management and future risk classification frameworks. As these findings are derived from a single-institution, LVSI-positive cohort with an empirically defined threshold, external validation is warranted before clinical adoption.

Objective In patients with Non-Small Cell Lung Cancer (NSCLC), spread through air spaces (STAS) describes tumor cells within alveolar spaces beyond the main tumor margin and is considered an invasive pattern associated with recurrence. This study aimed to preoperatively predict STAS using metabolic parameters obtained from F-18 FDG PET/CT imaging. Methods This retrospective single-center study analyzed 108 patients who underwent surgical resection between 2019 and 2024. Histological assessment determined the presence and subtype of STAS as limited or extended. PET/CT parameters included SUVmax, MTV, and TLG, measured by both a fixed threshold (42% of SUVmax) and patient-specific adaptive thresholds. Peritumoral rim values were calculated volumetrically. Variables associated with STAS in univariate analysis were tested with ROC analysis to determine diagnostic performance, and survival outcomes were evaluated using the Kaplan–Meier method. Results STAS was detected in 56.5% (n = 61) of patients, with the extended subtype observed in 67.2% (n = 41) of these cases. SUVmax was significantly higher in STAS-positive tumors and demonstrated modest discriminative ability, with an AUC of 0.619. A cut-off value of 5.3 achieved very high sensitivity (0.97) but only moderate specificity (0.64). None of the volumetric parameters, including MTV and TLG, differentiated between limited and extended STAS. Survival was numerically worse in patients with STAS, and extended STAS tended to have shorter median overall survival compared with limited STAS, though these differences did not reach statistical significance. Conclusion It was concluded that STAS, which plays a crucial role in the management of lung cancer treatment and prognosis prediction, can be predicted by SUVmax and may serve as a useful non-invasive parameter in the preoperative evaluation. However, other volumetric parameters showed no significant predictive value, and the high sensitivity but moderate specificity of SUVmax limit its standalone clinical utility. Future studies combining metabolic, radiomic, and morphological features may yield more accurate tools for preoperative STAS detection and tailored surgical planning.