Abstract Topic Esophageal Cancer: Molecular Biology/Pathology Background With the advancement of personalized medicine, multi-target drug development has garnered significant attention, particularly for complex diseases such as cancer. This study aims to identify potential dual-target inhibitors against Epidermal Growth Factor Receptor (EGFR) and Phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha (PIK3CA), two proteins whose aberrant activation is closely associated with tumorigenesis and progression in various cancers. Methods We collected IC50 values of active compounds for EGFR and PIK3CA from the BindingDB database, which were then standardized to pIC50 values using RDKit. A total of 2048 Extended-Connectivity Fingerprints (ECFPs) were calculated to serve as molecular descriptors. Various machine learning models, including Support Vector Machine (SVM), Decision Tree, Random Forest, Gradient Boosting, K-Nearest Neighbors, and LightGBM, were developed. The optimal model parameters were determined using ten-fold cross-validation and grid search, and model performance was assessed by Mean Absolute Error (MAE), Mean Squared Error (MSE), and the R-squared (R2) value. Results The SVM model demonstrated the best performance and was selected to predict activities for both EGFR and PIK3CA. Conclusion The natural product compounds CNP0456830 and CNP0467494 exhibited the lowest binding free energies for both EGFR and PIK3CA, identifying them as the most promising dual-target inhibitors. This study offers a new direction and a potential therapeutic strategy for personalized drug design in cancer treatment.

BackgroundPARP(Poly(ADP-ribose) polymerase) inhibitors are established as effective treatment for ovarian cancer; however, there have been no systematic studies of PARP inhibitor use in older patients with ovarian cancer.MethodsThis study was conducted at Peking University People’s Hospital. Archival data from patients with advanced age (≥ 65 years) with ovarian cancer between March 2018 and September 2023, who received olaparib or niraparib as maintenance therapy following either first-line platinum-based chemotherapy or platinum-sensitive recurrence (PSR) chemotherapy and achieved complete (CR) or partial response (PR). The two clinical scenarios (first-line maintenance and PSR maintenance) were analyzed separately.ResultsOf 56 included patients, 20 (35.7%) and 34 (60.7%) received olaparib and niraparib, respectively, while 2 (3.6%) switched from olaparib to niraparib due to intolerance. The mean age was 70.14 ± 4.63 years. Median progression-free survival (mPFS) was 24 months in 35 (62.5%) patients receiving PARP inhibitors as first-line maintenance therapy(1L-group). In the 1L group, CA125 level, whether CR was reached, BRCA mutation status, and R0 at initial surgery before PARP inhibitor therapy were associated with PFS in univariate analysis, and the latter three factors were independently associated with PFS on multivariate analysis. In the PSR group, among patients with PSR ≥12 months (n=13), mPFS was not reached; among those with PSR 6–12 months (n=7), mPFS was 9.6 months. Anemia (22.7%) was the most frequent grade 3–4 adverse event in the olaparib group, whereas thrombocytopenia (11.1%) was more common in the niraparib group. Both groups of patients experienced dose reduction, interruption, and discontinuation within the first 6 months. Niraparib had a lower termination rate than olaparib, and long-term PARP inhibitor use was generally well-tolerated.ConclusionsPARP inhibitor (olaparib and niraparib) maintenance therapy yields favorable clinical outcomes for elderly ovarian cancer patients. Three clinical factors, attainment R0(no residual disease) after initial surgery, BRCA mutation, and CA-125 ≤10 U/mL before PARP inhibitor treatment, were predictive of longer PFS in elderly ovarian cancer patients undergoing first-line maintenance therapy with PARP inhibitors.

1049 Background: T-Bren is a HER2-directed antibody-drug conjugate (ADC) consisting of an anti-HER2 monoclonal antibody linked to a potent topoisomerase I inhibitor (Ed-04). In phase I study, T-Bren demonstrated encouraging antitumor activity with a manageable safety profile in patients (pts) with HER2-positive unresectable LA/M breast cancer (BC) who had failed standard treatments. Results of safety and efficacy from a phase II study assessing T-Bren monotherapy or in combination with pertuzumab in treatment-naïve HER2-positive unresectable LA/M BC are presented. Methods: Treatment-naïve pts with HER2-positive (IHC3+, or IHC2+/ISH+) unresectable LA/M BC were treated with T-Bren monotherapy at 4.4mg/kg Q3W (cohort A) or T-Bren 4.4mg/kg Q3W in combination with pertuzumab (cohort B). Primary endpoints were ORR and RP2D for combination treatment. Results: As of Nov 30, 2025, a total of 83 pts were enrolled and treated in cohort A (n = 43) and cohort B (n = 40). All pts were included in the analysis (see table below). The median follow-up was 10.6 months. In cohort A, ORR was 93.0%, confirmed ORR (cORR) was 86.0%. In cohort B, ORR and cORR were 87.5%. Median PFS have not reached in either cohort. The landmark PFS rate at 12-months was 79.1% in cohort A and 90.8% in cohort B. The most common grade 3 and above hematologic TRAEs in both cohorts were neutropenia (51.8%), anemia (37.3%), leukopenia (36.1%), and thrombocytopenia (26.5%); the most frequent grade 3 and above non-hematologic TRAEs were weight decreased (7.2%), hypokalemia (6.0%), and nausea (6.0%). Grade 3 and above TRAEs were able to be effectively managed with standard supportive measure including dose reductions, as demonstrated by the TRAE leading to discontinuation rate of 4.8%. No treatment related death or ILD was reported. Conclusions: T-Bren as monotherapy or in combination with pertuzumab has demonstrated a promising antitumor activity and a manageable safety profile in pts with treatment-naïve HER2-positive unresectable LA/M BC. Phase III study assessing T-Bren in treatment-naïve HER2-positive unresectable LA/M BC is in preparation. Clinical trial information: NCT06445400 . Cohort A: T-Bren 4.4 mg/kg D1Q3W (N=43) Cohort B: T-Bren 4.4 mg/kg D1Q3W+ pertuzumab D1Q3W (N=40) Total (N=83) ORR, % (95% CI) 93.0 (80.9, 98.5) 87.5 (73.2, 95.8) 90.4 (81.9, 95.7) cORR, % (95% CI) 86.0 (72.1, 94.7) 87.5 (73.2, 95.8) 86.7 (77.5, 93.2) DCR, % (95% CI) 100 (91.8, 100) 97.5 (86.8, 99.9) 98.8 (93.5, 100) 12-mo PFS rate, % (95% CI) 79.1 (32.9, 95.2) 90.8 (74.1, 97.0) 89.9 (77.8, 95.6) 12-mo OS rate, % (95% CI) 100 (100, 100) 95.0 (81.5, 98.7) 97.4 (89.9, 99.3)

Abstract Topic Esophageal Cancer: Molecular Biology/Pathology Background To use bioinformatics methods to evaluate the prognostic value of Programmed Cell Death Related Genes (PCDRGs) in esophageal carcinoma (EC), and to explore the development and immune regulatory mechanisms of EC from multiple perspectives. Methods Using TCGA, GSE53622 data sets, and downloaded key regulatory genes of 18 PCD patterns, combined with 10 different machine learning methods to develop a prediction model, named this model ‘Characteristics of Cell Deaths’ (CDS). Seven prognosis-related genes were screened out by the model. The correlationbetween these seven genes and EC was analyzed. Results The PCDRGs prognostic model developed using the StepCox[both] + RSF method performed the best. CDS showed significant and powerful performance in predicting EC clinical outcomes and was able to serve as an independent risk factor in TCGA and GEO datasets. Conclusion This study successfully developed a novel EC PCDRGs model, which could predict the prognosis and drug treatment sensitivity of EC patients in the future based on further validation.

Abstract Topic Esophageal Cancer: Surgical Treatment of Esophageal Cancer – technique Background Robot-assisted minimally invasive esophagectomy (RAMIE) is increasingly adopted, yet comparative patient-reported outcome (PRO) data against conventional minimally invasive esophagectomy (MIE) remains scarce. We aimed to compare perioperative PROs, including quality of life, nutritional status, cough-related quality of life, and anxiety, between RAMIE and MIE for esophageal squamous cell carcinoma (ESCC). Methods We enrolled 214 patients with ESCC who underwent RAMIE (n = 47) or conventional MIE (n = 167) from an ongoing prospective cohort study. PROs were assessed using the EQ-5D-5L, PG-SGA, Leicester Cough Questionnaire (LCQ), and Self-rating Anxiety Scale (SAS). A 1:1 propensity score-matched (PSM) analysis was performed to balance baseline covariates. Nutritional trajectory was tracked from admission through discharge using serial PG-SGA assessments. Results Before PSM, the RAMIE group had significantly higher preoperative PG-SGA scores at admission (5.18 ± 3.32 vs 4.09 ± 3.01, P = 0.015) and pre-surgery (5.44 ± 3.69 vs 3.78 ± 2.88, P = 0.006), indicating worse baseline nutritional status. However, PG-SGA scores at discharge were comparable (4.56 ± 2.74 vs 4.42 ± 2.28, P = 0.931), demonstrating nutritional convergence. No significant differences were observed in EQ-5D-5L utility index (0.92 vs 0.93, P = 0.934), LCQ total score (121.6 vs 122.6, P = 0.751), or SAS standard score (45.4 vs 45.4, P = 0.551). The RAMIE group had a higher anastomotic leak rate (14.9% vs 4.8%, P = 0.025) but no Clavien-Dindo grade ≥III complications. Conclusion RAMIE and conventional MIE achieved comparable perioperative PROs in ESCC patients. Despite worse baseline nutritional status in the RAMIE group, nutritional recovery by discharge was equivalent. These findings support RAMIE as a viable alternative with comparable patient-centered outcomes, though the higher anastomotic leak rate warrants further investigation.

Abstract Topic Esophageal Cancer: Surgical Treatment of Esophageal Cancer – early outcomes and complications Background Postoperative complications following oesophagectomy occur in 25–60% of patients, yet traditional patient-reported outcome (PRO) assessment relies on standardised questionnaires with limited reliability, particularly among elderly patients with low health literacy. We aimed to develop and externally validate a multimodal artificial intelligence (AI) system that automates PRO assessment from natural patient conversations and provides early warning of major postoperative complications. Methods We conducted a prospective observational study at a tertiary cancer center, enrolling consecutive patients undergoing esophagectomy.The development cohort (Centre 1, n=196) and the temporally independent external validation cohort (Centre 2, n=103) were recruited sequentially. Using the validated Patient Symptom Assessment for Oesophageal Cancer (PSA-ESO) instrument, we collected trimodal recordings (video, audio, text) at up to 25 timepoints per patient, yielding 6,813 evaluable assessments. We fine-tuned the Qwen2.5-Omni-7B multimodal large language model with LoRA adaptation for two tasks: automated PRO symptom grading (Task A) and Temporal Transformer-based early warning of Clavien-Dindo grade II or higher complications (Task B). In a prospective implementation substudy (n=62), we evaluated the clinical impact of real-time alerts on time-to-intervention. Results In the external validation cohort, the trimodal PRO assessment achieved a weighted kappa of 0.801 (95% CI 0.73–0.87) and ICC of 0.858 against expert consensus, significantly outperforming audio-text bimodal (kappa 0.754, p<0.0001) and text-only (kappa 0.689, p<0.0001) configurations. The early warning model achieved an AUROC of 0.873 (95% CI 0.82–0.93) with a mean detection lead time of 28.7 hours before clinical diagnosis. The system detected 76.1% of symptom under-reporting cases. In the implementation substudy, real-time alerts reduced median time-to-intervention from 14.2 hours to 6.8 hours (p=0.003) and were associated with shorter ICU stays (3.1 vs 5.4 days, p=0.028). Conclusion An end-to-end multimodal AI system can accurately automate PRO assessment from natural patient conversations and provide clinically meaningful early warning for postoperative complications, with external validation confirming generalisability across cohorts. This conversation-based approach represents a paradigm shift from questionnaire-based PRO evaluation in surgical oncology, with particular relevance for populations with limited health literacy.

Abstract Topic Esophageal Cancer: Adjuvant and Neo-Adjuvant Therapies Background The survival benefit of neoadjuvant therapy for esophageal squamous cell carcinoma (ESCC) remains controversial, as landmark trials were conducted predominantly in adenocarcinoma populations. This study evaluated the association between neoadjuvant therapy and long-term survival in a large international multicenter ESCC cohort using propensity score methods and multivariable Cox regression. Methods We retrospectively analyzed 2,449 patients with ESCC who underwent curative esophagectomy at five institutions across three countries (Japan, China, and South Korea) between 2008 and 2022. Patients receiving neoadjuvant therapy (n=941) were compared with a surgery-first cohort (n=880) after excluding 41 cases with inconsistent survival data. Propensity score matching (PSM) 1:1 was performed using nearest-neighbor matching (caliper=0.1 SD of logit propensity score) with exact matching on center, adjusting for sex, age, clinical T and N stage. Inverse probability of treatment weighting (IPTW) with trimming (1st–99th percentile) served as a sensitivity analysis. Center-stratified univariate and multivariable Cox proportional hazards regression were applied. Primary endpoints were overall survival (OS) and disease-free survival (DFS). Results In analysis, neoadjuvant therapy appeared protective for OS (HR 0.60, 95% CI 0.48–0.75, p<0.001) while DFS was worse (HR 1.57, 95% CI 1.34–1.84, p<0.001). This discordant finding reversed after center-stratified Cox regression, revealing substantial confounding by center. Univariate analysis showed neoadjuvant therapy was associated with worse OS (HR 2.22, 95% CI 1.75–2.83, p<0.001) and DFS (HR 2.31, 95% CI 1.89–2.82, p<0.001). Multivariable analysis confirmed neoadjuvant therapy as an independent predictor of worse OS (HR 1.48, 95% CI 1.11–1.96, p=0.007) and DFS (HR 1.39, 95% CI 1.09–1.77, p=0.007). PSM yielded 255 pairs; after matching, neoadjuvant therapy was associated with worse OS (HR 1.56, 95% CI 1.05–2.30, p=0.027) and DFS (HR 1.60, 95% CI 1.18–2.19, p=0.003). IPTW sensitivity analysis yielded consistent results (OS: HR 2.21, 95% CI 1.74–2.82; DFS: HR 2.30, 95% CI 1.89–2.81; both p<0.001). Conclusion In this international multicenter analysis of 2,449 ESCC patients, neoadjuvant therapy was not associated with improved survival after esophagectomy and remained an independent predictor of worse OS and DFS across univariate, multivariable, PSM, and IPTW analyses. These findings underscore the critical importance of histology-specific evidence and support the urgent need for contemporary randomized controlled trials designed specifically for ESCC populations.

Abstract Topic Esophageal Cancer: Surgical Treatment of Esophageal Cancer Background Preoperative risk stratification for esophagectomy complications relies on clinical prediction models; however, their discriminative performance in multi-institutional settings remains poorly defined. We hypothesized that standard preoperative variables would demonstrate limited predictive validity across heterogeneous surgical cohorts. Methods We analyzed 2,490 patients undergoing esophagectomy across four institutions in Asia (total n=2,490; individual center n range 75–1,012). Four outcomes were studied: recurrent laryngeal nerve palsy (RLNP), anastomotic leak (AL), pulmonary complications (PC), and vocal cord palsy (VCP). Logistic regression models with bootstrap-validated odds ratios (1,000 iterations) were evaluated by 5-fold cross-validated AUC. SHAP (SHapley Additive exPlanations) via Gradient Boosting Machines quantified variable importance. Decision curve analysis (DCA) assessed net clinical benefit across threshold probabilities 2–70%. Association between tumor location and each complication was assessed using chi-squared tests. Results All prediction models demonstrated poor-to-fair discrimination: RLNP AUC 0.533, AL AUC 0.586, PC AUC 0.676, and VCP AUC 0.556. Tumor location was the only statistically significant categorical predictor of RLNP—upper/cervical location was associated with higher RLNP incidence compared to middle thoracic tumors (39.1% vs. 28.0%; OR 1.22, 95%CI 1.05–1.42; p=0.009). No significant association was observed between tumor location and AL, PC, or VCP. DCA demonstrated negligible clinical net benefit for RLNP and AL models; only the PC model provided modest benefit (max net benefit gain +0.057) at threshold probabilities of 5–20%. SHAP analysis identified FEV1%, PNI score, and BMI as the highest-importance variables for RLNP prediction, with tumor location ranking sixth—indicating that location contributes a statistically real but clinically modest signal. Conclusion Standard preoperative variables are insufficient for individualized risk stratification of RLNP, anastomotic leak, or vocal cord palsy after esophagectomy. Statistical significance (p=0.009 for location–RLNP association) does not translate to clinically meaningful predictive power (AUC 0.533). Tumor location should be incorporated into RLNP preoperative counseling. Improved prediction will require prospective integration of real-time intraoperative data. Pulmonary complication risk approaches clinically actionable prediction (AUC 0.676) and may guide respiratory prehabilitation targeting.

Ovarian endometrioma (OEM) remains a challenging manifestation of endometriosis, particularly in women with infertility, recurrent disease, or concern about ovarian reserve. Ultrasound-guided sclerotherapy has emerged as a minimally invasive treatment option in selected patients, but its appropriate use depends heavily on imaging-based decision-making. This review examines the role of ultrasound across the care pathway of sclerotherapy for OEM, with emphasis on diagnosis, risk stratification, patient selection, procedural planning, treatment guidance, and post-treatment surveillance. English-language literature up to May 2026 was searched in PubMed, Embase, Web of Science, and Scopus using terms related to OEM, endometriosis, transvaginal ultrasound, ultrasound-guided aspiration, sclerotherapy, ethanol sclerotherapy, recurrence, ovarian reserve, infertility, and assisted reproduction. Current evidence indicates that transvaginal ultrasound is central not only to confirming the diagnosis of OEM, but also to excluding mimickers and malignant red flags before intervention. In women considered for sclerotherapy, ultrasound provides key information on lesion morphology, disease extent, ovarian mobility, pelvic adhesions, accessibility, and puncture safety. Ultrasound-guided sclerotherapy appears to be a feasible minimally invasive option in selected women, particularly when preservation of ovarian reserve is prioritized. However, the available literature remains heterogeneous with regard to patient selection, procedural protocols, sclerosant exposure, definitions of response and recurrence, and follow-up intervals. Ultrasound may therefore play an important role not only in procedural guidance, but also in decision-making throughout the care pathway of OEM sclerotherapy. A standardized ultrasound-based framework may improve candidate selection, procedural consistency, and post-treatment surveillance, and may help better define the role of sclerotherapy in clinical practice.

Abstract Topic Esophageal Cancer: Surgical Treatment of Esophageal Cancer – long term outcomes Background Esophageal cancer remains a significant global health burden. While surgical and oncologic factors influencing survival after esophagectomy are well studied, the potential impact of surgical timing and medical staff burnout on long-term outcomes remains underexplored. This study investigates whether the month of esophagectomy and esophageal surgery personnel burnout affect survival in esophageal squamous cell carcinoma (ESCC). Methods A retrospective cohort of ESCC patients who underwent esophagectomy between January 2010 and December 2017 was analyzed from a single-center esophageal cancer case management database. Patients were stratified into twelve monthly subgroups based on the month of surgery. Survival outcomes were assessed using Kaplan–Meier analysis, Cox proportional hazards regression, and restricted mean survival time (RMST). Propensity score matching (PSM) was employed to adjust for confounding variables including age, sex, tumor stage, and neoadjuvant therapy. Additionally, a cross-sectional survey was conducted from 2024 to 2025 to evaluate burnout levels among 409 esophageal surgery professionals using the Maslach Burnout Inventory–Human Services Survey (MBI-HSS), with monthly stratification to examine the temporal relationship between staff burnout patterns and surgical outcomes. Results Among 2,758 ESCC patients who underwent esophagectomy, those operated in February had the poorest survival, with a median survival time (MST) of 34.17 months compared to 45.23 months for non-February surgeries (HR: 0.833, 95% CI: 0.677–1.025, P=0.084). The 3- and 5-year overall survival rates were lower for February (47% and 36%) versus other months (56% and 45%). Conversely, April and June surgeries yielded superior outcomes (MST: 61.44 and 63.60 months, respectively). Burnout survey data (n=409) revealed a striking inverse pattern: February showed mild burnout (EE: 16.69, DP: 5.63), coinciding with the Chinese New Year holiday period, whereas non-February months demonstrated severe burnout (EE: 30.95, DP: 13.51). This paradoxical finding suggests that lower burnout alone does not predict better outcomes; rather, February’s reduced surgical volume and case complexity may be contributory factors. Conclusion Surgical timing, particularly esophagectomy performed in February, is associated with inferior long-term survival in ESCC patients. The concurrent low burnout levels during this period suggest that reduced surgical volume and altered case selection during the holiday season, rather than staff fatigue, may underlie this disparity. These findings highlight the importance of considering temporal factors in esophagectomy scheduling and warrant further multicenter validation.

Abstract Topic Esophageal Cancer: Surgical Treatment of Esophageal Cancer – early outcomes and complications Background Postoperative complications following esophagectomy are frequent and clinically heterogeneous. Serum C-reactive protein (CRP) is a widely available inflammatory marker, yet its longitudinal trajectory after esophagectomy and its relationship with complication severity have not been systematically characterized. This study aimed to identify distinct postoperative CRP trajectory patterns and evaluate their association with Clavien-Dindo graded complication outcomes. Methods Among 809 consecutive esophagectomy patients at a single center, 131 patients (16.2%) had complete serial CRP measurements on postoperative days (POD) 1, 3, 5, and 7. K-means clustering was applied to identify distinct CRP trajectory groups. Associations between trajectory groups and complications (Clavien-Dindo grading), anastomotic leak, and ICU length of stay were evaluated using chi-square tests, Kruskal-Wallis tests, and multivariable logistic regression. Receiver operating characteristic (ROC) analysis assessed the predictive utility of individual CRP timepoints and derived indicators. Results Three distinct CRP trajectories were identified: Low-Stable (n=86, 65.6%), Moderate-Declining (n=20, 15.3%), and High-Persistent (n=25, 19.1%). Complication rates increased significantly across trajectories: severe complications (CD≥III) occurred in 15.1%, 25.0%, and 56.0% of patients in each group, respectively (p<0.001). The High-Persistent group showed a 5.71-fold increased odds of severe complications on multivariable analysis (OR 5.71, 95%CI 1.99–16.40, p=0.001). CRP on day 7 demonstrated the highest individual discriminative value (AUC=0.671), while the CRP D3/D1 ratio (AUC=0.632) offered the earliest actionable prediction at POD 3. Conclusion Postoperative CRP trajectory patterns are significantly associated with complication severity after esophagectomy. The High-Persistent trajectory identifies a high-risk subgroup, and the D3/D1 ratio enables early risk stratification by POD 3. Prospective studies with larger cohorts are warranted to validate these findings.

Abstract Topic Esophageal Cancer: Other Background Precise segmentation of esophageal squamous cell carcinoma (ESCC) lesions on CT imaging is essential for treatment planning. While the Segment Anything Model for 3D medical images (SAM-Med3D) shows promise, its performance in delineating tumor boundaries remains suboptimal. We propose a prototype-contrast boundary optimization strategy to enhance segmentation accuracy. Methods In this retrospective multicenter study, CT volumetric data from 310 ESCC patients who underwent esophagectomy were collected from two centers (Center A: n=161, July 2018–February 2023; Center B: n=149). Data from Center A were used for training and internal validation, while Center B served as the external testing cohort. We evaluated SAM-Med3D for ESCC lesion segmentation and compared it with state-of-the-art methods including nnU-Net, SegVol, and MedSAM. To address SAM-Med3D’s limitation in boundary delineation, we introduced a prototype-contrast boundary pixel clustering optimization. This approach employs contrastive learning to refine boundary pixel classification by pulling features toward correct class prototypes while pushing them away from incorrect ones. Performance was assessed using Dice Similarity Coefficient (DSC) and 95th percentile Hausdorff Distance (HD95). Results The optimized SAM-Med3D achieved a DSC of 76.23% on internal validation and 73.57% on external testing, representing a 3.11% improvement over the original SAM-Med3D (internal DSC: 73.12%; external DSC: 70.46%). Our method outperformed all baseline approaches: nnU-Net (internal/external DSC: 72.45%/69.18%), SegVol (70.83%/67.52%), and MedSAM (71.26%/68.34%). The optimized model also achieved the lowest HD95 values (internal: 6.48 mm; external: 7.81 mm), indicating superior boundary delineation accuracy. The performance advantage was consistent across both centers, demonstrating robust generalizability. Conclusion The prototype-contrast boundary pixel clustering optimization effectively enhances SAM-Med3D’s capability for ESCC lesion segmentation, achieving superior performance over nnU-Net, SegVol, and MedSAM across multicenter datasets. This approach addresses the critical challenge of accurate tumor boundary delineation and demonstrates strong generalizability, providing a valuable tool for preoperative assessment and surgical planning in ESCC.

3001 Background: BL-M14D1 is a novel DLL3-directed ADC with a potent topoisomerase I inhibitor Ed-04. Results for safety and efficacy from a phase I study on BL-M14D1 in patients with locally advanced or metastatic (LA/M) SCLC and NEC are presented. Methods: Patients who had LA/M solid tumors were enrolled in dose escalation and dose expansion phase and treated with BL-M14D1 at 0.66, 2.0, 3.5, 4.0, 4.5, 5.0, and 6.0mg/kg on Day 1 every 3 weeks (D1 Q3W). SCLC and NEC patients enrolled in dose expansion phase were treated at 4.0 and 4.5mg/kg D1 Q3W. Results: As of Nov 30, 2025, a total of 127 patients were enrolled, including 87 SCLC and 40 NEC. Median prior line of therapy was 2 (range 1-5). Treatment-related AEs (TRAEs) rate was 98.4% and grade≥3 TRAEs rate was 74.0%. The most frequent grade≥3 TRAEs were hematologic, including thrombocytopenia (54.3%), neutropenia (53.5%), leukopenia (46.5%) and anemia (32.3%), which were able to be effectively managed with standard supportive measures including dose reductions, as demonstrated by a low rate (1.6%) of TRAEs leading to discontinuation. One grade 3 ILD and one treatment-related death was reported. Efficacy data at 4.0 and 4.5mg/kg was presented below. BL-M14D1 has demonstrated a promising efficacy with an ORR/cORR of 71.1%/57.8% and mPFS of 7.2 months in heavily pre-treated SCLC. Furthermore, BL-M14D1 has shown encouraging activity in other gastrointestinal and gynecological NECs. Conclusions: BL-M14D1 has demonstrated promising antitumor activity with a tolerable safety profile at dose levels of 4.0mg/kg and 4.5mg/kg D1 Q3W in heavily pretreated patients with SCLC and NEC, supporting further clinical development of BL-M14D1 in SCLC and other less common NECs. Phase III studies assessing BL-M14D1 in SCLC and NEC are in preparation. Clinical trial information: NCT06505824 . SCLC Total * (N=83) SCLC 4.0mg/kg (N=34) SCLC 4.5mg/kg (N=37) NEC † Total * (N=22) NEC † 4.0mg/kg (N=3) NEC † 4.5mg/kg (N=19) Median prior LoT (range) 2 (1-5) 2 (1-3) 2 (1-5) 2 (1-3) 2 (1-2) 2 (1-3) ORR, % (95% CI) 71.1 (60.1, 80.5) 73.5 (55.6, 87.1) 70.3 (53.0, 84.1) 40.9 (20.7, 63.6) 33.3 (0.8, 90.6) 42.1 (20.3, 66.5) cORR, % (95% CI) 57.8 (46.5, 68.6) 61.8 (43.6, 77.8) 56.8 (39.5, 72.9) 27.3 (10.7, 50.2) 33.3 (0.8, 90.6) 26.3 (9.1, 51.2) DCR, % (95% CI) 94.0 (86.5, 98.0) 97.1 (84.7, 99.9) 91.9 (78.1, 98.3) 90.9 (70.8, 98.9) 100 (29.2, 100) 89.5 (66.9, 98.7) mFU for PFS (mo) (95% CI) 6.8 (5.5, 7.2) 7.7 (5.6, 8.2) 4.4 (4.1, 5.5) 3.0 (1.6, 4.3) NR (1.8, NR) 3.0 (1.4, 3.9) mPFS (mo) (95% CI) 7.2 (5.6, 12.7) 7.2 (5.6, NR) 6.9 (4.2, NR) 5.3 (3.9, NR) 4.6 (3.9, 5.3) NR (2.6, NR) † Sites of NEC included cervix, esophagus, stomach and others. * Efficacy analysis included pts at 4.0 and 4.5 mg/kg who received BL-M14D1 and had at least one post-baseline scan.

LBA1003 Background: Triple-negative breast cancer (TNBC) remains a therapeutic challenge due to its aggressive biology and limited treatment options beyond first-line immunotherapy plus chemotherapy. Patients who progress after taxane-containing regimens face poor outcomes and have a high unmet need. Izalontamab brengitecan (iza-bren) is a first-in-class bispecific antibody–drug conjugate (ADC) targeting EGFRxHER3 with a potent topoisomerase I inhibitor payload. Here we report the results of a pre-planned interim analysis from a phase III study of iza-bren in patients with TNBC. Methods: This randomized, multicenter, open-label, phase III study enrolled patients with unresectable locally advanced or metastatic TNBC who had disease progression after 1–2 prior lines of systemic therapy for advanced disease—including prior taxanes. Patients were randomized 1:1 to receive iza-bren (2.5 mg/kg D1D8 Q3W) or treatment of physician’s choice (TPC) (eribulin, capecitabine, gemcitabine, or vinorelbine). Randomization was stratified by prior lines of therapy (1 vs. 2), prior anti-PD-(L)1 (yes vs. no), and HER2 IHC (0 vs. 1+/2+ ISH-). The dual-primary endpoints were progression-free survival (PFS) by blinded independent central review (BICR) and overall survival (OS). Results: A total of 418 patients were randomized (iza-bren/TPC: 207/211) and 412 were treated (iza-bren/TPC: 207/205). As of the data cutoff (Jan 13, 2026), median follow-up was 11.0 months. The median PFS by BICR was 8.5 months with iza-bren and 3.1 months with TPC (HR, 0.29 [95% CI, 0.22-0.38]; stratified log-rank P<0.0001). The median OS was 15.9 months with iza-bren and 12.5 months with TPC (HR, 0.60 [95% CI, 0.42-0.85]; stratified log-rank P=0.0019). The confirmed objective response rate (cORR) assessed by BICR was 51.7% with iza-bren and 20.5% with TPC (odds ratio, 4.28 [95% CI, 2.75-6.66]). Most common grade ≥3 TEAEs (iza-bren vs TPC) were neutrophil count decreased (58.0% vs 46.8%), white blood cell count decreased (56.0% vs 33.2%), platelet count decreased (52.7% vs 2.4%), and anemia (46.9% vs 3.9%). All-grade ILD was reported in 3 (1.4%) and 0 patients in the iza-bren and TPC arms, respectively. Treatment discontinuation due to TEAEs occurred in 4 (1.9%) patients in the iza-bren arm and 1 (0.5%) patient in the TPC arm. No new safety signals were observed. Conclusions: This pre-planned interim analysis met both dual-primary endpoints of PFS by BICR and OS. Iza-bren demonstrated a statistically significant and clinically meaningful improvement in both PFS and OS compared with chemotherapy, with a manageable safety profile in heavily pre-treated TNBC patients, supporting iza-bren as a new standard of care in this population. Clinical trial information: NCT06382142 .

3030 Background: BL-M05D1 is a novel Claudin18.2-directed ADC with a potent topoisomerase I inhibitor Ed-04. Results for safety and efficacy data from a phase I study on BL-M05D1 in patients (pts) with locally advanced or metastatic (LA/M) Claudin18.2-expressing solid tumors are presented. Methods: Pts who had LA/M solid tumors enrolled in dose escalation and dose expansion phase, and treated with BL-M05D1 at 0.66, 2.0, 3.0, 4.0 or 5.0mg/kg on Day 1 every 3 weeks (D1 Q3W). Gastric cancer/gastroesophageal junction cancer (GC/GEJ), biliary tract cancer (BTC), and pancreatic cancer (PanC) patients with Claudin18.2-expressing enrolled in dose expansion phase were treated at 2.0, 3.0 or 4.0mg/kg D1 Q3W. Results: As of Nov 30, 2025, a total of 245 pts were enrolled (87 GC/GEJ, 53 BTC, 104 PanC) and 1 other). The most frequent grade≥3 TRAEs were all hematological, including thrombocytopenia (33.9%), neutropenia (32.7%), leukopenia (31.8%) and anemia (14.3%); non-hematologic TRAEs were relatively low. One pt at 5.0mg/kg group experienced grade 3 febrile neutropenia as a DLT. No treatment-related deaths or ILD were reported. In PanC, BL-M05D1 has demonstrated a promising efficacy with an ORR of 35.7% and mPFS of 5.6 mo at 4.0mg/kg in CLDN18.2-positive pts; in the second line pts, ORR was 50.0% with a mPFS of 5.7 mo. Furthermore, BL-M05D1 has shown promising efficacies in CLDN18.2-positive GC and BTC with an ORR of 39.2% and 37.9%, and mPFS of 5.4 mo and 5.9 mo, respectively, at 4.0mg/kg. Efficacy results are summarized below. Conclusions: BL-M05D1 has demonstrated a promising antitumor activity with a tolerable safety profile in pts with pretreated Claudin18.2-positive PanC, GC, and BTC. 4.0mg/kg D1 Q3W was chosen as RP2D. The data support further development of BL-M05D1 in Claudin18.2-positive PanC, GC, and BTC. Phase III studies are in preparation. Clinical trial information: NCT06349811 . PanC-Total * PanC CLDN 18.2 + # 4.0 mg/kg PanC CLDN 18.2 + # 1 Prior Chemo in 4.0mg/kg GC-Total * GC CLDN 18.2 + # 4.0 mg/kg GC CLDN 18.2 + # 1 Prior Chemo in 4.0 mg/kg BTC-Total * BTC CLDN 18.2 + # 4.0 mg/kg BTC CLDN 18.2 + # 1 Prior Chemo in 4.0 mg/kg N = 81 N = 28 N = 16 N = 79 N = 51 N = 29 N = 40 N = 29 N = 20 Median prior LoT (range) 2 (1-4) 2 (1- 3) / 2 (1-4) 1 (1-3) / 1 (1-3) 1 (1-3) / ORR, % (95% CI) 17.3 (9.8-27.3) 35.7 (18.6-55.9) 50.0 (24.7-75.3) 35.4 (25.0-47.0) 39.2 (25.8-53.9) 44.8 (26.4-64.3) 37.5 (22.7-54.2) 37.9 (20.7-57.7) 45.0 (23.1-68.5) DCR, % (95% CI) 74.1 (63.1-83.2) 89.3 (71.8-97.7) 100 (79.4-100) 88.6 (79.5-94.7) 88.2 (76.1-95.6) 86.2 (68.3-96.1) 92.5 (79.6-98.4) 93.1 (77.2-99.2) 90.0 (68.3-98.8) mFU for PFS (mo) 2.8 5.5 4.2 4.4 4.4 4.3 2.8 2.7 2.7 mPFS (mo) 4.0 5.6 5.7 4.4 5.4 7.7 6.9 5.9 5.9 * Efficacy analysis included all pts who received at least one dose of BL-M05D1 and with at least one post baseline scan. # PanC CLDN 18.2 +: IHC 2+/3+ ≥ 50%; GC CLDN 18.2 +: IHC 2+/3+ ≥ 40%; BTC CLDN 18.2 +: IHC1+/2+/3+≥30%.

6019 Background: Izalontamab is a bispecific antibody targeting EGFR and HER3. Here we present safety and efficacy results from a phase II study on izalontamab in combination with paclitaxel/docetaxel in patients with R/M HNSCC. Methods: Patients with R/M HNSCC who had progressed on prior anti-PD-(L)1 therapy, with or without platinum-based chemotherapy, and received no more than two prior lines of treatment were enrolled. Izalontamab was administered at 12mg/kg QW in combination with paclitaxel (80mg/m 2 IV QW) (for patients without prior exposure to paclitaxel) or docetaxel (35mg/m 2 IV D1D8D15 Q4W) (for patients with prior exposure to paclitaxel). Primary endpoint was investigator-assessed ORR. Secondary endpoints included PFS, OS, DCR, and DoR. Results: As of Nov 7, 2025, a total of 45 patients (44 HNSCC and 1 sinnonasal) were enrolled, of whom 35 (77.8%) received one prior line of therapy and 10 (22.2%) received two prior lines of therapy. HNSCC patients who received at least one dose of study drug were included in the analysis. Median follow-up was 15.0 months. Overall, ORR was 43.2%, confirmed ORR was 31.8%, median PFS was 4.0 months, and median OS was 10.0 months. In patients treated with izalontamab combo with paclitaxel (n = 34), ORR was 52.9%, confirmed ORR was 41.2%, median PFS was 5.4 months, and median OS was 11.2 months. Grade≥3 TRAEs occurred in 57.8% of patients. The most common grade≥3 TRAE was leukopenia (20.0%), followed by neutropenia (15.6%), rash (13.3%), and anemia (8.9%). One (2.1%) patient discontinued study treatment due to TRAE. No treatment-related death was observed. Conclusions: Izalontamab in combination with paclitaxel showed encouraging antitumor activity and a manageable safety profile in patients with R/M HNSCC who were previously treated with anti-PD-(L)1 therapy. Clinical trial information: NCT05054439 . Izalontamab+Paclitaxel(N=34) Izalontamab+ Docetaxel(N=10) Total(N=44) ORR, % (95% CI) 52.9 (35.1, 70.2) 10.0 (0.3, 44.5) 43.2 (28.3, 59.0) cORR, % (95% CI) 41.2 (24.6, 59.3) 0 31.8 (18.6, 47.6) DCR, % (95% CI) 73.5 (55.6, 87.1) 30.0 (6.7, 65.2) 63.6 (47.8, 77.6) Median DOR (mo) (95% CI) 5.0 (3.7, 11.0) NR (NR, NR) 5.0 (3.7, 11.0) Median PFS (mo) (95% CI) 5.4 (3.9, 6.6) 1.5 (0.4, 3.6) 4.0 (2.8, 5.5) Median OS (mo) (95% CI) 11.2 (6.6, 19.3) 6.6 (0.4, 13.8) 10.0 (6.6, 13.8)

Abstract Topic Esophageal Cancer: Surgical Treatment of Esophageal Cancer – early outcomes and complications Background Postoperative complications after esophagectomy remain a major source of morbidity, yet current surveillance relies predominantly on clinician-initiated assessments and structured database fields that systematically underdetect complications. We aimed to develop and internally validate a multimodal early warning system (MEWS-Eso) integrating patient-reported outcomes (PROs), laboratory trajectories, and large language model (LLM)-extracted clinical text features to enable real-time, automated complication surveillance. Methods This prospective cohort study enrolled 323 consecutive patients undergoing esophagectomy for esophageal cancer at a single tertiary center (2019–2024). PROs were collected using the MD Anderson Symptom Inventory (MDASI) at 10 timepoints from preoperative baseline through 6 months postoperatively (completion rate 91.3%). Laboratory values (complete blood count, C-reactive protein, procalcitonin, albumin) were extracted at 7 perioperative timepoints. Free-text clinical notes (nursing records, physician progress notes, operative reports; median 3,247 words/patient for POD0–POD7) were processed using GPT-4o for structured information extraction and semantic embedding generation. NLP was independently applied to identify complications missed by structured fields. Six models of incrementally increasing modality were compared using 5-fold cross-validation with 100 bootstrap iterations: M1 (PRO-only), M2 (PRO + clinical baseline), M3 (PRO + laboratory), M4 (PRO + text embeddings), M5 (PRO + laboratory + text), and M6 (full multimodal). The primary endpoint was postoperative complications within 90 days under both standard and NLP-augmented outcome definitions. Results NLP text mining identified 72 complications missed by structured fields, most notably anastomotic stricture (43 NLP-detected vs. 3 structured-field recorded; p < 0.001). Under the NLP-augmented definition, complication prevalence increased from 52.3% to 61.0%. AUROC improved progressively with modality addition: M1 (PRO-only) 0.658 [95% CI 0.601–0.715], M3 (PRO + laboratory) 0.761 [0.712–0.810], M5 (PRO + laboratory + text) 0.824 [0.779–0.869], and M6 (full multimodal) 0.847 [0.805–0.889]. The CRP trajectory (POD1–POD5 slope) was the single strongest laboratory predictor (OR 2.41, 95% CI 1.78–3.26). LLM-extracted text features contributed an incremental AUROC gain of +0.063 beyond PRO + laboratory. At the RED-alert threshold (probability ≥0.70), MEWS-Eso achieved sensitivity 72.8%, specificity 83.5%, and positive predictive value 78.3%, with a median early warning lead time of 4.2 days. Removing the PRO module caused the largest performance drop (ΔAUROC = −0.089), followed by laboratory (−0.074) and text (−0.063). Conclusion A multimodal early warning system integrating PROs, laboratory trajectories, and LLM-processed clinical text substantially outperforms single-modality approaches for detecting postoperative complications after esophagectomy. PROs provide the most irreplaceable modality, while NLP-based text mining both corrects systematic outcome misclassification and contributes independent predictive signals. This framework demonstrates the feasibility of automated, patient-centered multimodal surveillance in surgical oncology.

Abstract Topic Esophageal Cancer: New Diagnostic Modalities (Including PET, PET-CT, New Tracers) Background Accurate ultrasound evaluation of cervical lymph nodes (LNs) is essential for staging and treatment planning in esophageal cancer, particularly for decisions regarding three-field lymphadenectomy and neoadjuvant strategies. However, cervical LN assessment in routine ultrasound practice typically follows the AAO-HNS/ASHNS neck level system, which was not designed for esophageal cancer and may not correspond well to oncologic nodal stations defined by the JES or AJCC staging frameworks. This mismatch may limit the clinical interpretability and imaging value of ultrasound reports. Methods We conducted a nationwide, cross-sectional survey among thoracic surgeons and ultrasound physicians from tertiary hospitals across China to evaluate current ultrasound-based cervical LN partitioning practices, familiarity with different nodal classification systems, and the perceived impact of zoning discrepancies on clinical decision-making. The questionnaire also incorporated representative CT and ultrasound images to assess interobserver consistency in imaging-based nodal station identification. Results A total of 312 valid responses were collected from 132 tertiary institutions. All ultrasound physicians (100%) reported using the AAO-HNS system in routine practice, whereas only 2.2% incorporated JES terminology. More than half of surgeons (56.3%) frequently needed to convert ultrasound-reported neck levels into JES/AJCC nodal stations, and 63.8% indicated that such discrepancies influenced preoperative assessment. Both surgeons (91.3%) and ultrasound physicians (87.0%) strongly supported the integration of JES/AJCC-compatible labeling and structured visualization into ultrasound reports. Interobserver agreement for imaging-based nodal station identification was moderate (κ = 0.53), indicating notable variability in anatomical interpretation. Conclusion Current ultrasound-based cervical LN partitioning provides incomplete alignment with esophageal cancer–specific oncologic zoning, resulting in reporting ambiguity and reduced imaging utility in multidisciplinary care. There is strong consensus across imaging and surgical specialties supporting the optimization of ultrasound LN reporting through JES/AJCC-compatible mapping and structured visualization, highlighting an important opportunity to enhance the clinical value of cervical LN ultrasound in esophageal cancer.

Abstract Topic Esophageal Cancer: Surgical Treatment of Esophageal Cancer – early outcomes and complications Background Persistent moderate-to-severe symptoms after minimally invasive esophagectomy (MIE) remain poorly characterized. While patient-reported outcomes (PROs) are increasingly recognized as important endpoints in surgical oncology, the prevalence, risk factors, optimal intervention timing, and prognostic impact of persistent postoperative symptom burden in esophageal squamous cell carcinoma (ESCC) patients have not been systematically evaluated. Understanding these factors is essential for developing evidence-based, timely symptom management strategies to improve postoperative recovery and quality of life. Methods We retrospectively analyzed 345 ESCC patients who underwent MIE at a single tertiary center. Symptoms were assessed using the MDASI, EORTC QLQ-C30, and QLQ-OES18 at baseline through 6 months postoperatively. Persistent moderate-to-severe symptoms at postoperative month 3 were defined by composite criteria (meeting any one): MDASI ≥3 core symptoms scoring ≥4; QLQ-C30 ≥2 functional domains below 66.7; or OES18 ≥2 symptom domains above 33.3. Independent risk factors were identified by multivariable logistic regression. Symptom trajectory analysis identified the critical divergence timepoint. Kaplan–Meier analysis, log-rank tests, multivariable Cox regression, and propensity score matching (1:1, 81 pairs) evaluated associations between symptom persistence and overall survival (OS) and disease-free survival (DFS). Results At postoperative month 3, 89 patients (25.8%) met criteria for persistent moderate-to-severe symptoms. Independent risk factors were anastomotic leak (OR=7.92, P<0.001), adjuvant therapy (OR=3.05, P<0.001), and smoking history (OR=2.58, P=0.029); male sex was protective (OR=0.34, P=0.013). Predictive model AUC was 0.734. Symptom trajectories diverged at postoperative day 14. Five-year OS (63.5% vs 66.1%, HR=1.16, P=0.478) and DFS (72.9% vs 73.8%, HR=0.87, P=0.554) showed no significant differences. After propensity score matching (81 pairs), OS remained non-significant (HR=0.94, P=0.796), while DFS favored the persistent group (HR=0.55, P=0.029), potentially mediated by higher adjuvant therapy exposure. Conclusion Approximately one-quarter of post-MIE ESCC patients experience persistent moderate-to-severe symptoms at 3 months, primarily driven by anastomotic leak and adjuvant therapy. Symptom persistence does not adversely affect overall survival; its association with improved DFS likely reflects the tumor-control benefit of adjuvant therapy. Postoperative day 14 represents the critical symptom trajectory divergence point, providing an optimal window for initiating targeted PRO-based intervention.

Abstract Topic Esophageal Cancer: Other Background Burnout among esophageal surgery personnel has received limited attention despite the specialty’s demanding nature. The post-COVID era has further intensified occupational stress globally. This study aimed to evaluate the international prevalence, severity, and risk factors of burnout among esophageal surgery personnel using the Maslach Burnout Inventory–Human Services Survey (MBI-HSS). Methods A multicenter, anonymous, cross-sectional online survey was conducted from April 2024 to January 2025 across 59 countries spanning six continents. The MBI-HSS (22 items, 0–6 Likert scale) assessed three burnout dimensions: emotional exhaustion (EE, 9 items, 0–54), depersonalization (DP, 5 items, 0–30), and personal accomplishment (PA, 8 items, 0–48). High burnout was defined as EE>26, DP>12, or PA<32. Respondent location was determined by IP geolocation. Demographic variables included occupation, age, gender, and professional title. Independent-samples t-tests, one-way ANOVA, and Pearson correlations were used for subgroup and inter-dimensional analyses. Results A total of 724 respondents from 59 countries completed the survey (644 surgeons [88.9%] and 80 nurses [11.1%]; 527 from China, 197 international). Nearly half reported high-level burnout: 47.9% had high EE (mean 26.70, SD 15.54), 39.1% had high DP (mean 11.38, SD 9.43), and 48.3% had reduced PA (mean 31.41, SD 10.80). EE and DP were strongly correlated (r=0.872, p<0.001), both negatively correlated with PA (r=−0.355, r=−0.402). International respondents exhibited markedly higher burnout than Chinese counterparts (EE 38.2 vs 22.4; DP 18.9 vs 8.6). Surgeons reported higher EE (27.06 vs 23.77) and DP (11.67 vs 9.01) than nurses. Younger personnel (21–30 yr) experienced the highest burnout (EE 29.0–31.0, DP 12.6–14.3), while those aged 51–60 had markedly lower levels (EE 11.3, DP 1.9, PA 38.8). Conclusion This first large-scale international burnout survey among esophageal surgery personnel reveals that nearly half experience high-level burnout in the post-COVID era. International respondents demonstrate substantially higher burnout than Chinese counterparts. Surgeons and younger professionals are disproportionately affected. These findings underscore the urgent global need for targeted interventions including workload optimization, early-career mentorship, and systematic well-being programs in esophageal surgery departments worldwide.