Ejso最新文献

Objectives: Intratumoral tertiary lymphoid structures (iTLSs) are prognostic biomarkers for hepatocellular carcinoma (HCC). This study aimed to develop a machine learning approach based on topologically distinct intratumoral heterogeneity (ITH) scores derived from CT images to predict iTLS status and patient outcomes.

Methods: In this multicenter study, patients from Centers 1 and 2 were divided into training (n = 475) and internal validation (n = 204) cohorts, with an independent cohort (n = 208) used for external validation from Center 3. Two complementary ITH scores were developed: a 2D score integrating local radiomics with global pixel patterns on the largest cross-sectional slice, and a 3D score extending this quantification to the entire tumor volume. A stacking ensemble classifier (2D3D-TD-ITH-Ensemble) incorporating clinicoradiological features and ITH scores was constructed to predict iTLS status. Model performance was compared with clinical and traditional radiomics models. SHapley Additive exPlanations (SHAP) analysis was used for interpretability. Disease-free survival (DFS) was assessed using Kaplan-Meier analysis.

Results: The 2D3D-TD-ITH-Ensemble demonstrated superior diagnostic performance compared to reference models. In the internal validation cohort, the ensemble model achieved an AUC of 0.904, outperforming the radiomics (AUC 0.887) and clinical models (AUC 0.811). Consistent results were observed in the external validation cohort, where the ensemble model yielded an AUC of 0.890, versus 0.864 for the radiomics model and 0.817 for the clinical model. SHAP analysis identified the 3D ITH score as the most influential contributor to model output. Furthermore, HCC patients with the presence of iTLS and lower 3D ITH scores exhibited significantly better DFS (p < 0.05).

Conclusions: The preoperative CT-based 3D ITH score serves as a robust non-invasive biomarker for predicting iTLS status and prognosis in HCC, potentially guiding stratified immunotherapy strategies.

Background: The prognostic impact of vascular resection in intrahepatic cholangiocarcinoma (iCCA) remains uncertain, particularly in the context of advanced disease requiring complex surgery. This study evaluated the association between vascular resection and long-term survival and examined how tumor biology, assessed with tumor burden score (TBS) and CA19-9, influenced outcomes.

Methods: Patients who underwent upfront curative-intent hepatectomy for iCCA were identified from an international multi-institutional database. Multivariable Cox regression assessed overall (OS) and recurrence-free survival (RFS). Among patients undergoing vascular resection, biologic heterogeneity was explored using optimized TBS and CA19-9 cutoffs to define biologic risk subgroups.

Results: Among 1757 patients, 9.7% (n = 171) required major vascular resection. These patients were more likely to have bilateral tumors (29.2% vs. 21.3%), lymph node metastasis (33.9% vs. 24.8%), and T3/T4 tumors (57.3% vs. 27.6%) (all p < 0.05). Vascular resection was not an independent predictor of OS (aHR 1.11, 95% CI 0.82-1.51) or RFS (HR 1.02, 95% CI 0.81-1.27). Within the vascular resection cohort, higher TBS (aHR 1.11, 95% CI 1.03-1.19) and higher CA19-9 (log-transformed; aHR 1.16, 95% CI 1.02-1.32) were independently associated with worse OS. Five-year survival decreased stepwise from favorable to unfavorable biologic profiles. Patients with favorable biology demonstrated survival comparable to individuals undergoing hepatectomy without vascular resection.

Conclusion: Major vascular resection did not independently worsen prognosis after iCCA hepatectomy. Instead, tumor biology-captured by TBS and CA19-9-was strongly associated with long-term outcomes, highlighting the importance of biologic risk stratification when considering vascular resection.

Background: While tumour-free (R0) margins are essential in oesophageal cancer surgery, the ideal length of the proximal resection margin remains unclear. Although longer margins may improve oncological outcomes, they can increase surgical complexity and morbidity. Current evidence is limited and inconsistent, mostly based on small retrospective studies. This population-based cohort study aimed to assess the association between proximal resection margin length and 5-year mortality following oesophagectomy for oesophageal cancer.

Methods: This binational, population-based cohort study included 1830 patients who underwent curatively intended oesophagectomy with tumour-free margins for oesophageal cancer in Sweden or Finland between 2006 and 2020. The main exposure was the length of the proximal resection margin, categorized as 0.1-<2.0 cm, 2.0-<5.0 cm, 5.0-<8.0 cm, and ≥8.0 cm. The outcomes were all-cause and disease-specific 5-year mortality. Multivariable Cox regression provided adjusted hazard ratios (HR) with 95% confidence intervals (CI).

Results: There was a gradual decrease in the risk of 5-year mortality outcomes with increasing proximal resection margin lengths up until the category 5.0 to <8.0 cm, with HR 0.83 (95% CI 0.68-1.01) for all-cause mortality and HR 0.80 (95% CI 0.65-0.98) for disease-specific mortality. A longer resection margin (>8.0 cm) did not further decrease the all-cause 5-year mortality (HR 0.87, 95% CI 0.70-1.09) or disease-specific 5-year mortality (HR 0.85, 95% CI 0.67-1.07).

Conclusions: A proximal resection margin of 5 to 8 cm may be appropriate to optimize the chance of 5-year survival in oesophageal cancer patients who undergo oesophagectomy, but further research is necessary to confirm its applicability.

Background: Resection margin status and lymph node involvement are well-established predictors of recurrence following resection of perihilar cholangiocarcinoma (pCCA). However, even patients with favorable pathology including negative surgical margins (R0) and node-negative disease (N0) may experience recurrence. We sought to develop a clinically relevant tool to risk stratify patients relative to tumor recurrence following an R0N0 resection of pCCA.

Methods: pCCA patients undergoing curative-intent resection with R0 and N0 tumor were identified from an international multi-institutional database. A pathology-based risk score was developed to predict recurrence-free survival (RFS). In addition, genomic profiling was performed in a subset of patients to evaluate the prognostic relevance of genetic alterations.

Results: Among 298 patients with resected R0N0 pCCA, 131 (44.0%) developed disease recurrence. Multivariable analysis identified advanced AJCC T category (T2b or T3/T4), perineural invasion, and poor tumor differentiation as independent predictors of inferior RFS. Based on these factors, a three-variable pathology-based risk score stratified patients into low-, intermediate-, and high-risk groups with corresponding 3-year RFS of 85%, 31%, and 27%, respectively. Both intermediate- and high-risk patients had worse RFS versus low-risk patients (high-risk vs. low-risk: median RFS, 15.0 vs. 92.9 months; intermediate-risk vs. low-risk: median RFS, 23.0 vs. 92.9 months; both p < 0.001). KRAS mutations occurred in 29% of profiled patients, which was associated with reduced RFS (mutant vs. wild-type KRAS: median RFS, 11.0 vs. 24.0 months, p = 0.011).

Conclusions: Recurrence risk among patients with R0N0 pCCA was heterogeneous. The proposed risk score stratified patients into markedly different risk categories relative to recurrence, which may help guide utilization of adjuvant therapy as well as surveillance in the postoperative setting.

Background: Achieving tumor-free margins without unintentional tissue loss is essential in breast-conserving surgery (BCS). Calculated resection ratio (CRR) and tumor eccentricity measuring tumor displacement from the specimen center have been proposed as complementary quality metrics to reoperation and mastectomy rates. The objective of this study was to identify factors influencing CRRs and reoperations in BCS.

Methods: The prospective, multi-center FIBRATIO study included 206 women undergoing unilateral BCS for invasive cancer and/or ductal carcinoma in situ across five Finnish centers. Tumor, specimen, and breast volumes were measured radiologically and histopathologically. CRRs, defined as total resection volume (TRV) divided by optimal resection volume, were calculated both radiologically (CRR^rad) and histologically (CRR^pat). Eccentricity and relative eccentricity (adjusted for tumor size) were also assessed. Associations with clinical and imaging variables, and reoperations were analyzed using multivariable analyses.

Results: Median CRR^rad was 2.3 [interquartile range (IQR) 1.5-3.7] and CRR^pat 2.4 (IQR 1.4-3.7). Relative eccentricity was 1.0 (IQR 0.5-2.0), higher in smaller tumors and correlated with CRRs. Reoperation occurred in 14% of patients and was associated with larger lesion size and lower CRRs. High CRRs were associated with large breast volume, non-dense breasts, and non-palpable tumors. CRRs decreased with increasing tumor size. Tumor spiculation was associated with higher CRR^rad. Statistically significant inter-surgeon and inter-center variability in CRRs and reoperation rates was observed.

Conclusions: BCS is associated with variable and often excessive resection of healthy tissue. Identifying patients at risk for over-resection may improve surgical planning. Incorporating CRR into quality metrics alongside reoperation and mastectomy rates could enhance benchmarking.