This study aimed to demonstrate tumor immune microenvironment (TIME) subtypes as prognostic factors for local recurrence-free survival (LRFS) in esophageal squamous cell carcinoma (ESCC) patients treated with definitive chemoradiotherapy (dCRT) and to develop an interpretable, TIME-guided subregional radiomics model for predicting LRFS. 986 ESCC patients from three centers were enrolled. For a subset with available biopsies, multiplex immunofluorescence was used to assess CD3, CD8, and PD-L1 expression, defining prognostic TIME subtypes. Pre-treatment contrast-enhanced CT (CECT) images were used for tumor segmentation, intratumoral subregion partitioning and radiomic feature extraction. Features associated with TIME subtypes were selected to build a subregional radiomic signature (Subrad signature) for LRFS prediction using an advanced machine learning framework. This signature was integrated with clinical factors to construct a Clinic-Subregional radiomic model (CliSubrad-M). A “PD-L1-low/TILs-high” TIME subtype was identified as an independent favorable prognostic factor for LRFS (HR = 0.34, P < 0.001). The Subrad signature, comprising seven TIME-related features from various subregions, demonstrated robust predictive performance (C-indexes: 0.631–0.637). The integrated CliSubrad-M (built with Random Survival Forest) achieved the best performance, with C-indexes of 0.702 (training), 0.679 (internal testing), 0.653 (external testing1), and 0.625 (external testing2), significantly outperforming the clinical model (all P < 0.05). The model effectively stratified patients into distinct risk groups with significant differences in LRFS. SHAP analysis confirmed the signature’s strong contribution and enhanced model interpretability. The developed interpretable subregional radiomics model, integrating TIME biology with clinical factors, provides a non-invasive tool for predicting LRFS in ESCC received dCRT.

The prognostic value of skeletal muscle mass and muscle quality in patients with locally advanced esophageal squamous cell carcinoma (LA-ESCC) undergoing neoadjuvant chemoradiotherapy (NCRT) remains unclear. We investigated the association of low skeletal muscle index (SMI) and low skeletal muscle radiodensity (SMD, reflecting myosteatosis) with long-term survival. We conducted a multicenter retrospective study including consecutive LA-ESCC patients treated between 2014 and 2023 at three tertiary hospitals. SMI and SMD were measured at the third lumbar vertebral level on preoperative CT images. Optimal cutoff values were determined using time-dependent receiver operating characteristic curve analysis. Overall survival (OS) was analyzed using Kaplan–Meier curves and Cox proportional hazards models. Sex-specific analyses were conducted to explore potential effect modification. A total of 420 patients were included, with a median follow-up of 39.1 months, the 5-year OS rate was 60

BackgroundWith the widespread application of neoadjuvant immunochemotherapy (nICT) in locally advanced resectable esophageal squamous cell carcinoma (ESCC), determining the optimal postoperative adjuvant treatment strategy has become a critical clinical issue. This study aimed to evaluate the efficacy of different adjuvant treatment patterns and identify beneficial populations.MethodsThis was a multicenter real-world study involving 612 ESCC patients from six national centers who underwent R0 resection after nICT. Patients were divided into three groups based on postoperative treatment patterns: Clinical Observation (CO group, n=337), Postoperative Adjuvant Chemotherapy (POCT group, n=80), and Postoperative Adjuvant Chemoimmunotherapy (POICT group, n=195). Study endpoints included overall survival (OS), disease-free survival (DFS), and recurrence patterns.ResultsMultivariate analysis of the whole cohort showed that both POCT and POICT reduced the risk of death (HR = 0.51, P = 0.007; HR = 0.49, P<0.001) and disease progression (HR = 0.48, P = 0.003; HR = 0.48, P<0.001) compared with CO. Subgroup analysis indicated that among patients without pathological complete response (non-pCR), both POCT and POICT significantly improved OS and DFS (P<0.05), with no statistical difference between the two groups. In pCR patients, only POICT showed an OS benefit (P = 0.014). Patients with postoperative pathological stage III benefited in both OS and DFS (P<0.05) from POCT and POICT. After propensity score matching (PSM), the 3-year OS (74.7% vs 79.8%) and DFS (63.1% vs 67.2%) rates between the POCT and POICT groups showed no significant difference (P>0.05). Postoperative adjuvant therapy was an independent factor affecting OS in patients with more advanced yT, yN, and yTNM stages, and an independent prognostic factor for DFS in patients with yN2 stage and advanced yTNM stages. Failure pattern analysis revealed no statistically significant differences in the cumulative incidence of locoregional recurrence or distant metastasis between the POCT and POICT groups (χ²=0.279, 2.828; P = 0.597, 0.093).ConclusionFor ESCC patients receiving nICT, postoperative adjuvant treatment can provide survival benefits, particularly more pronounced in non-pCR patients and those with advanced postoperative pathological stages. Adding immunotherapy to chemotherapy did not significantly further improve survival outcomes in the overall PSM-matched population; however, in pCR patients, POICT showed OS benefit while POCT did not, suggesting potential differential efficacy according to pathological response.

Background: Reliably predicting brain metastasis (BM) and brain metastasis‑free survival (BMFS) in patients with primary small cell carcinoma of the esophagus (PSCCE) remains a clinical challenge. Objectives: To develop and externally validate predictive nomograms for assessing BM probability and BMFS in patients with PSCCE. Design: A retrospective model development and external validation study. Methods: Using a training cohort, we constructed two separate nomograms. The first was developed using univariable and multivariable logistic regression to predict the probability of BM. The second was built using a Fine‑Gray competing-risk model to estimate BMFS, treating death without BM as a competing event. The performance of both nomograms was evaluated using the area under the receiver operating characteristic (ROC) curve/time-dependent ROC curve, calibration plots, and decision curve analysis (DCA). External validation was subsequently performed using an independent validation cohort. Results: The study sample included 492 patients in the training cohort (mean age at diagnosis 61.97 ± 8.86 years; 133 (27%) female) and 344 patients in the external validation cohort (mean age at diagnosis 62.97 ± 8.14 years; 108 (31%) female). Age, N stage, and M stage emerged as independent predictors and were included in a nomogram to estimate BM risk. Adding lesion length and initial treatments (radiotherapy, chemotherapy, and surgery) to these three factors allowed the model to predict BMFS. Both nomograms showed strong predictive performance in the training cohort, with area under the ROC curve values of 0.76 and 0.843, respectively, supported by calibration plots and DCA. These results were confirmed in the external validation cohort. The logistic regression‑based nomogram accurately predicted BM probability, while the Fine–Gray model provided reliable estimates of BMFS. Conclusion: We successfully developed and externally validated two complementary nomograms that reliably predict BM probability and BMFS in patients with PSCCE. These tools may aid in risk stratification and inform personalized surveillance strategies.

Background:Definitive chemoradiotherapy is the standard treatment for inoperable locally advanced esophageal squamous cell carcinoma (ESCC). However, the prognosis for these patients remains poor. This study aimed to evaluate iparomlimab and tuvonralimab (QL1706), a novel PD-1/CTLA-4 dual inhibitor, combined with definitive chemoradiotherapy in patients with unresectable stage III-IVA ESCC. Methods:This single-arm, open-label phase 2 trial was conducted at a single center in China between August 2022 and September 2023. 39 patients with unresectable stage III-IVA ESCC were included. QL1706 is composed of iparomlimab (anti-PD-1 IgG4) and tuvonralimab (anti-CTLA-4 IgG1) in a fixed 2:1 ratio. Patients received radiotherapy (50.4 Gy/28 in fractions on 5 days per week), concurrent chemotherapy (paclitaxel 135 mg/m2 d1+ cisplatin 25 mg/m2 d1-3, q3w, 2 cycles), and QL1706 (5 mg/kg q3w for up to 1 year [total of 18 cycles]). The primary endpoint was progression-free survival (PFS). This study is registered with ClinicalTrials.gov (NCT05490719). Findings:20 patients (51.3%) completed the full cycles of QL-1706. Reasons for the premature cessation of QL1706 were disease progression (n = 7), COVID-19 infection (n = 2), pneumonia (n = 2), allergic reaction (n = 2), patient refusal (n = 4), rash (n = 1), and esophageal hemorrhage (n = 1). With a median follow-up of 21.1 months, the median progression-free survival (PFS) was 14.8 (95% CI: 11.2-NA) months. The median overall survival (OS) was immature. The 1-year PFS and OS rates were 58.6% (95% CI: 44.9-76.4) and 84.6% (95% CI: 74.0-96.7). The objective response rate and median duration of response were 84.6% (95% CI: 69.5-94.1) and 12.7 months (95% CI: 8.9-NA). Exploratory biomarker analyses identified several potential predictive biomarkers: 1) immunochemistry staining revealed that PD-L1 combined positive score ≥1 correlated with prolonged PFS (HR 0.37, p = 0.036); 2) Whole-exome sequencing detected high tumor mutation burden associated with better PFS (HR 0.24, p = 0.0066), while the mutated group (TNRC18/CAMSAP3/CARMIL2/ZFHX4 gene alterations) correlated with poor PFS (HR 6.40, p = 0.0002) and OS (HR 6.45, p = 0.0020); and 3) Olink plasma proteomics identified baseline levels of TWEAK and FASLG were both positively associated with PFS and OS (p < 0.05 for all). Grade ≥3 adverse events occurred in 89.7% (35/39) of the patients, predominantly lymphopenia (31/39, 79.5%). Interpretation:QL1706 combined with chemoradiotherapy demonstrated potential antitumor activity and manageable toxicity, supporting further investigation. Funding:National Natural Science Foundation of China, State Key Laboratory of Druggability Evaluation and Systematic Translational Medicine, Tianjin Key Medical Discipline Construction Project, and Tianjin Key Medical Discipline (Specialty) Construction Project.

Evidence for adjuvant immunotherapy after neoadjuvant chemoimmunotherapy (NCIT) remains limited. This study aims to assess survival benefits of adjuvant immunotherapy in esophageal squamous cell carcinoma (ESCC) patients treated with NCIT. This multicenter retrospective study analyzed 398 esophageal cancer patients who underwent NCIT followed by R0 resection (2019–2023). Postoperatively, 147 received immunotherapy (ICIs group) and 251 did not (non-ICIs group). After propensity score matching (PSM), 143 patients were included in each group. The primary endpoints were 2-year overall survival (OS) and disease-free (DFS) rates. Before matching, adjuvant immunotherapy showed no significant OS (hazard ratio [HR] 0.68, p = 0.14) or DFS (HR = 0.75, p = 0.161) benefit versus non-ICIs in ESCC patients. Post-PSM, however, ICIs demonstrated significantly improved 2-year DFS (HR = 0.57, p = 0.013) and OS (HR = 0.47, p = 0.008). Subgroup analyses revealed that ≤65 years, male, lower location, cT3-4, cN1, ypT3-4, ypN2-3, ypIIIB-IVA stage, non-pCR, and non-MPR were significantly associated with survival benefits from adjuvant immunotherapy. Both before and after matching, pCR patients did not benefit from immunotherapy, whereas non-pCR patients exhibited significantly improved DFS (HR = 0.54, p = 0.007) and OS (HR = 0.48, p = 0.01) after matching. Multivariate Cox analysis showed that adjuvant chemoimmunotherapy was the only regimen independently improving both DFS and OS compared to no adjuvant therapy, while immunotherapy alone did not demonstrate significant survival improvement. Adjuvant immunotherapy enhanced survival in NCIT-treated ESCC patients overall, although no survival benefit was observed with adjuvant immunotherapy in pCR patients. Whereas non-pCR patients may benefit from adjuvant immunotherapy, with chemoimmunotherapy may be the optimal strategy. These findings are exploratory. Further validation in longer follow-up time, larger cohorts or prospective studies are needed.

Background Amino acid metabolism plays a crucial role in shaping tumor-immune crosstalk in non-small cell lung cancer (NSCLC). However, the key cellular mediators that translate metabolic states into intercellular signals remain poorly defined.Methods We integrated single-cell RNA-seq with spatial transcriptomics to map immunometabolic architecture. Transcriptional co-variation was decomposed into amino-acid metabolic programs using Non-negative Matrix Factorization (NMF); spatial deconvolution localized programs and cell types in tissue. Myeloid populations were subclustered to resolve macrophage states. Functional assays tested LAP3 overexpression (OE-LAP3) in A549/PC9 cells (qRT-PCR, Western blot, CCK-8, colony formation, wound-healing, Transwell) and a nude-mouse subcutaneous mouse model.Results Integrative single-cell and spatial transcriptomic analyses revealed that tumor epithelial and myeloid cells dominate the NSCLC microenvironment and exhibit lineage-specific activation of amino acid metabolic programs. Notably, LAP3 was selectively enriched in both tumor epithelium and a distinct macrophage subset. Spatial mapping localized this LAP3-high macrophage state to epithelial-myeloid interfaces, where it functions as a signaling hub, actively secreting chemokines, cytokines, adhesion molecules, and extracellular matrix (ECM) components. To test whether LAP3 plays a causal role in tumor behavior, we established stable LAP3-overexpressing A549 and PC9 cell lines, confirming robust upregulation at both mRNA and protein levels. Functionally, LAP3 overexpression significantly suppressed proliferation-evident in CCK-8 time-course and colony formation assays-and impaired motility and invasiveness, as shown by delayed wound healing and reduced cell migration/invasion in Transwell assays. Most importantly, these effects translated in vivo: LAP3-overexpressing xenografts formed markedly smaller tumors in nude mice.Conclusions LAP3 appears to functionally link amino acid catabolism to immune communication in NSCLC, defining an epithelial-macrophage immunometabolic niche where metabolic activity may shape the immune contexture. Its overexpression is associated with attenuated malignant phenotypes and heightened immune engagement, suggesting a potential dual role in restraining tumor aggressiveness and fostering an immune-responsive microenvironment. While these findings support LAP3 as a candidate biomarker for patient stratification and provide a rationale for combining metabolic modulation with immunotherapy, further mechanistic and clinical validation remains necessary.

Background:To compare the efficacy and safety of neoadjuvant therapy (NT) followed by robot-assisted Ivor-Lewis esophagectomy vs. upfront surgery (S) in patients with locally advanced adenocarcinoma of the esophagogastric junction (AEG). Methods:We retrospectively analyzed 156 patients with locally advanced AEG (cT2-4a/N0-2) who underwent robot-assisted Ivor-Lewis surgery with intrathoracic anastomosis between 2017 and 2024. Baseline characteristics, pathological responses, and surgical outcomes were compared between the NT and S groups. Survival data were evaluated using Kaplan-Meier analysis. Results:Of the 156 patients, 41 (26.3%) received NT. The NT group achieved a pathological complete response (pCR) rate of 24.4% (vs. 0% in the S group) and higher node-negative rates (pN0: 56.1% vs. 32.2%, P = 0.009), with significant downstaging in pT stage (T0-2: 41.5% vs. 17.4%, P < 0.001), TNM stage (I-II: 53.7% vs. 23.5%, P < 0.001), and lymphovascular invasion (7.3% vs. 27.0%, P = 0.008). Surgical outcomes were comparable between groups, including conversion rates (9.8% vs. 10.4%, P = 1.000), anastomotic leaks (2.4% vs. 4.3%, P = 1.000), and pneumonia (7.3% vs. 10.4%, P = 0.761), despite more frequent use of overlap anastomosis in the NT group (68.3% vs. 34.8%, P = 0.001). After a median follow-up of 21 months (n = 129), the NT group showed superior 2-year overall survival (OS) (87.5% vs. 71.0%, P = 0.042), with pCR patients achieving 100% 2-year OS. Conclusion:NT significantly improves pathological responses and survival in advanced AEG without increasing surgical risks. The robotic-assisted approach enables the safe implementation of this multimodal treatment strategy. Visual Abstract:http://links.lww.com/MS9/B274.

Purpose:The comparative efficacy of neoadjuvant chemoimmunotherapy (NCIT) vs neoadjuvant chemoradiotherapy (NCRT) for locally advanced esophageal squamous cell carcinoma (LA-ESCC) remains controversial. Methods and materials:This multicenter retrospective cohort study included patients with LA-ESCC who received NCIT or NCRT followed by esophagectomy across 7 Chinese medical centers between January 2012 and January 2024. The primary outcomes were disease-free survival (DFS) and overall survival (OS). Propensity score matching (PSM) was utilized to balance baseline covariates. Results:Among 2535 enrolled patients, 1414 received NCIT and 1121 received NCRT. After 1:1 PSM, 1258 patients (629 per group) were evaluated. With a median follow-up of 32.7 months, no significant differences were observed between the NCIT and NCRT groups in DFS (hazard ratio [HR], 1.14; 95% CI, 0.94-1.38; P = .20) or OS (HR, 0.98; 95% CI, 0.77-1.25; P = .89). The 2-year DFS rates were 68.7% (NCIT) vs 72.0% (NCRT), and 2-year OS rates were 81.3% vs 83.7%, respectively. NCIT showed a trend toward improved distant metastasis-free survival (DMFS) (HR, 0.87; 95% CI, 0.69-1.11; P = .27), whereas NCRT was associated with a trend toward improved locoregional recurrence-free survival (LRFS) (HR, 1.22; 95% CI, 0.96-1.56; P = .11). Conclusions:NCIT and NCRT demonstrated comparable DFS and OS. These findings suggest that both modalities are valid neoadjuvant strategies, possessing differing strengths in local vs systemic tumor control. These observations await confirmation in prospective trials.

Accurate segmentation of the gross tumor volume from computed tomography images is a core step in the development of precise radiotherapy planning for esophageal cancer, which directly affects treatment efficacy and normal tissue protection. Recently, foundation models represented by the segment anything model (SAM) have been increasingly applied in medical image segmentation, and several studies have extended them to tumor target segmentation with promising progress. However, existing SAM-based segmentation methods rely on manual prompts, leading to significant limitations in esophageal cancer. Manual localization of tumor prompt points is difficult, and inappropriate prompts easily cause segmentation errors, increasing the risk of damage to organs-at-risk during radiotherapy. To address this, this study proposes a segmentation method based on SAM that streamlines the inference process by using a coarse-to-fine prompt generation strategy. Inspired by the stepwise refinement of clinical target volume delineation, the core design lies in a coarse-to-fine prompt generation strategy. Specifically, coarse segmentation results generated by nnUNet are first converted into initial prompts to provide global anatomical priors for SAM. Furthermore, a dynamic iterative prompt update mechanism is introduced. During inference, the iterative correction prompts are derived solely from the model's own prediction uncertainty, forming a closed-loop refinement that gradually improves segmentation accuracy. Experimental results based on multi-center datasets show that this method, which iteratively updates prompts based on its own uncertainty, achieves better segmentation performance than existing single-pass methods on both internal and external validation sets. The framework is highly consistent with the logic of clinical target volume delineation and can provide a reliable efficient scheme for the formulation of precise radiotherapy plans for esophageal cancer.

Objective.Existing automatic segmentation methods for esophageal cancer radiotherapy gross tumor volume rely solely on imaging features. Even when coarse masks are introduced as spatial priors, they still lack the pathological semantic guidance essential for clinical decision-making, resulting in segmentation outcomes that are misaligned with clinical practice.Approach.We propose a mask guidance and text prompting segmentation framework (MGTP-Seg) built upon the UNETR backbone. The mask guidance branch incorporates coarse segmentation masks generated by nnU-Net as spatial priors. The text prompting branch encodes clinical information including TNM stage and tumor location via a large language model, and aligns textual semantics with visual features through prompt tuning and a bidirectional attention mechanism. This design achieves end-to-end synergy among imaging details, spatial priors, and high-level clinical knowledge.Main Results.Multi-center experimental results demonstrate that MGTP-Seg achieves a Dice coefficient of 79.86% on the internal validation set, significantly outperforming 3D-UNet, HRNet, SAM-Med3D, and other mainstream methods. Ablation studies confirm the necessity of both branches. The model maintains stable performance on the external validation set (Dice: 78.76%), and text manipulation experiments reveal that the model dynamically adjusts segmentation in response to clinical descriptions, providing interpretability.Significance.MGTP-Seg not only offers an accurate, interpretable, and clinically aligned automatic delineation solution for esophageal cancer radiotherapy, but also establishes a novel methodological paradigm and empirical foundation for integrating spatial priors with semantic knowledge in medical image analysis.

Chemoimmunotherapy has become the standard first-line treatment for advanced non-small cell lung cancer (NSCLC). Deciphering the T-cell subset responsible for chemoimmunotherapy and easily tested conveniently is critical in predicting the treatment outcomes. Based on peripheral blood collected from patients enrolled from a phase 2 clinical study (ClinicalTrials.gov NCT04836728), we performed multi-color flow cytometry and unsupervised analysis to explore correlations with therapeutic outcomes. We integrated single-cell RNA and T-cell receptor (TCR) sequencing in 36 samples, including peripheral blood, tumors and non-tumor tissues, from 8 NSCLC patients to interpret the correlation, which was further verified using blood samples, orthotopic and subcutaneous lung cancer mouse model. The baseline CD28−KLRG1+CD57+ and on-treatment CD28−KLRG1+ CD8+ T cells in peripheral blood were independent factors which indicated improved treatment outcomes in advanced NSCLC patients receiving first-line chemoimmunotherapy. While being in a late-differentiated T-cell status, these cells were clonally expanded and reinvigorated during chemoimmunotherapy, serving as a peripheral T-cell pool for supplying potential tumor-reactive T cells in tumors, and reversely differentiating into less-differentiated subsets. The zinc-metallothionein pathway regulated the CD28−KLRG1+ CD8+ T-cell subset. Zinc supplementation combined with chemoimmunotherapy improved both local and systemic antitumor immune responses in mouse model. Circulating CD28−KLRG1+ CD8+ T cells are valuable and convenient biomarkers for first-line chemoimmunotherapy in advanced NSCLC and provide insight into how late-differentiated or senescent T cells engage in the antitumor immunity when immunotherapy is added to conventional therapies.

LBA4082 Background: The optimal neoadjuvant strategy for resectable locally advanced esophageal squamous cell carcinoma (RLA-ESCC) remains unclear. SCIENCE is a randomized phase III trial comparing neoadjuvant chemotherapy (nCT) plus sintilimab (S), neoadjuvant chemoradiotherapy (nCRT) plus S, and nCRT in RLA-ESCC, aiming to determine whether immunotherapy-based neoadjuvant therapy can improve efficacy without increasing toxicity or compromising surgical feasibility, and to explore ctDNA as an early molecular marker of treatment response. Methods: Pts with histologically confirmed, resectable thoracic ESCC (cT1N2–3M0 or cT2–4aN0–3M0 per AJCC/UICC 8th) were randomized 1:1:1 to nCT + S (A), nCRT + S (B), or nCRT (C), followed by surgery 6–8 weeks after neoadjuvant therapy. Co-primary endpoints were pCR and EFS. Longitudinal ctDNA was assessed (tumor-informed, personalized assay) at baseline, on-treatment, and preoperatively. pCR and EFS were compared using chi-square and log-rank tests, respectively, with overall two-sided α=0.05 controlled by allocating 0.02 to pCR and 0.03 to EFS. Interim analysis for pCR was prespecified. Results: At the time of data cutoff (Jan 19, 2026), 307 pts had completed neoadjuvant therapy and undergone surgery and were evaluable for pCR (Group A/B/C: 100/103/104). Most pts were male (88.3%) and had clinical stage III disease (81.4%). R0 resection was achieved in 303 pts (99 (99.0%) Group A; 102 (99.0%) Group B; 102 (98.1%) Group C). pCR rates were 18.0% in Group A, 57.3% in Group B, and 49.0% in Group C. Compared with Group A, pCR was significantly higher in Group B (OR 6.1, 95% CI 3.3–11.9; p<0.0001) and Group C (OR 4.4, 95% CI 2.3–8.5; p<0.0001). The safety profile was tolerable across three arms. Among 92 ctDNA-evaluable pts, detectability decreased from 96.7% at baseline to 48.9% preoperatively. Preoperative ctDNA positivity was significantly higher in Group A (78.6%) than in Group B (40.0%) and Group C (32.4%) (p<0.001). Importantly, on-treatment ctDNA status after cycle 1 was already associated with pathological response (p=0.0043). Pts with ctDNA clearance had higher pCR rates than those with persistent ctDNA (62.2% vs 9.1%, p<0.001). ctDNA clearance rates increased stepwise across pathological response categories: pCR (84.8%), MPR (48.1%), and non-responders (12.5%). Conclusion: Neoadjuvant CRT + S significantly improved pCR versus CT + S in RLA-ESCC. Preoperative ctDNA clearance status closely tracked pathological response, supporting ctDNA as a promising early molecular marker of treatment effect. EFS will be reported with longer follow-up. Clinical trial information: NCT05244798 . nCT + S nCRT + S nCRT pCR (95% CI) 18% (11, 26.9) 57.3% (47.2, 67) 49% (39.1, 59) P - value (vs. Group A) <0.0001 <0.0001 P - value (vs. Group B) 0.2942

OBJECTIVE:We aimed to develop clinical decision support tools for patients with treatment-naïve advanced esophageal squamous cell carcinoma (ESCC). METHODS:Patients who received first-line immunochemotherapy with or without radiotherapy between 2018 and 2023 across 9 centers were included in the current study. Candidate predictors were routine clinical variables such as patient demographics, tumor characteristics, and treatment details. Feature selection was conducted using the least absolute shrinkage and selection operator (LASSO), Boruta, and stepwise feature selection method. The models were fitted using 4 machine learning algorithms and a Cox proportional hazards model was set as a benchmark, comparing model predictive performance by computing the time-dependent area under the receiver operator curve (tAUC). Concordance index (C-index), Brier score, calibration plots, precision-recall (PR) curves, decision curve analysis (DCA), and held-out testing and external validation were utilized for evaluating model robustness, transportability, and clinical utility. SHapley Additive exPlanations (SHAP) analysis was used to rank feature importance and explain the models. RESULTS:A total of 1,048 patients were included, comprising the training/test (n = 828) and validation cohorts (n = 220). The random survival forest model that was constructed following feature selection using Boruta [Boruta-random survival forest (RSF)] showed the highest predictive performance. The tAUCs for 6-, 12-, and 18-month OS were 0.886, 0.775, and 0.772, respectively, in the training set, which significantly outperformed all other models (P-adjust < 0.05). This superiority persisted in the external validation cohort (all tAUCs > 0.750). The Boruta-RSF model exhibited excellent calibration and the highest clinical net benefit on DCA. A publicly accessible web calculator (https://escc.shinyapps.io/ESCC/) was implemented, enabling individualized overall survival prediction under varying treatment scenarios. CONCLUSIONS:The Boruta-RSF model, leveraging routinely available clinical variables with the corresponding web calculator, facilitates personalized prognostic assessment and optimization of treatment strategies.

ABSTRACTBackground: For patients with unresectable locally advanced esophageal squamous cell carcinoma (ESCC), definitive chemoradiotherapy (CRT) represents the standard treatment regimen. However, patient response remain heterogenous and overall outcomes are suboptimal, highlighting the need for improved therapeutic approaches. Recent studies suggest that combining immune checkpoint inhibitors with radiotherapy may offer preliminary benefit in locally advanced ESCC, yet the optimal sequencing of immunotherapy and CRT remains unclear. Therefore, this study retrospectively assessed the efficacy and safety of induction immunochemotherapy followed by radiotherapy (ICRT) using multicenter, large-sample patient data. Methods: This retrospective study collected clinical data of patients with unresectable locally advanced ESCC who received ICRT in four Chinese cancer centers between January 2018 and December 2024, and included a comparison cohort of patients treated with RT-based therapy. After balancing baseline characteristics using propensity score matching, KaplanMeier (KM) analysis and Cox regression analysis were performed to evaluate progressionfree survival (PFS) and overall survival (OS). Results: A total of 1352 patients with unresectable locally advanced ESCC were included, of whom 767 received ICRT and 585 received CRT. With a median follow-up time of 32.1 months (IQR: 30.2–36.4) for the entire cohort, the ICRT group showed better progression-free survival (median 23.8 vs. 15.6 months; P < 0.001) and overall survival (median 35.9 vs. 26.1 months; P = 0.013) compared with the CRT group after propensity score matching. Multivariate Cox regression analysis identified ECOG performance status, clinical T stage, clinical N stage and ICRT as independent prognostic factors. The survival benefits were most pronounced in patients achieving favorable response (CR + PR) after IC compared to those with SD + PD (median PFS: 26.9 vs. 20.1 months; P < 0.001; median OS: 41.6 vs. 24.9 months; P < 0.001). The most frequently reported grade ≥3 treatment-related adverse events were leukopenia (ICRT 10.7% vs. CRT 12.1%; P = 0.556), neutropenia (5.5% vs. 8.8%; P = 0.051), thrombocytopenia (2.0% vs. 6.4%; P < 0.001), and radiation esophagitis (1.6% vs. 4.3%; P = 0.015), all of which were clinically manageable. Conclusions: This multicenter respectively study provides the first evidence that ICRT conferred a survival benefit with an acceptable safety profile in patients with unresectable locally advanced ESCC. These findings warrant validation in prospective clinical trials.

While radical chemoradiotherpy (CRT) is the standard treatment for patients with unresectable locally advanced esophageal squamous cell carcinoma (LA-ESCC), the prognosis remains extremely poor. Although combining CRT with immunotherapy benefits a subset of patients with advanced disease, recurrence and metastasis after treatment are still common. The overall response rate (ORR) of induction chemoimmunotherapy (ICI) for advanced esophageal cancer is approximately 40

BACKGROUND:The prognostic impact of excessive muscle loss (EML) during neoadjuvant chemoradiotherapy (NCRT) in locally advanced esophageal squamous cell carcinoma (LA-ESCC) remains unclear. METHODS:This multicenter retrospective study analyzed consecutive patients with LA-ESCC undergoing NCRT followed by esophagectomy (2014-2023). Skeletal muscle index was measured longitudinally on computed tomography at the L3 vertebra at three time points: pre-NCRT (T1), post-NCRT (T2), and pre-surgery (T3). EML was evaluated across two intervals: stage A (T1 to T2) and stage B (T2 to T3). Survival outcomes were evaluated using Cox proportional hazards and Kaplan-Meier analyses, with a two-sided P < 0.05 threshold for significance. RESULTS:Among 421 eligible patients (median follow-up 39.1 months), the 5-year overall survival (OS) and progression-free survival (PFS) rates were 60.0% and 53.9%, respectively. EML incidence was 26.6% in stage A and 37.1% in stage B. Multivariate analysis confirmed EML in both stages as independent predictors of poorer OS (stage A: hazard ratio [HR] 1.78; 95% confidence interval [CI] 1.18-2.69; P = 0.006; stage B: HR 1.83; 95% CI 1.23-2.72; P = 0.003). Stage B EML was also independently associated with worse PFS (HR 1.75; 95% CI 1.24-2.46; P = 0.001). Among patients without stage A EML, stable muscle mass in stage B predicted improved OS (HR 0.49; 95% CI 0.38-0.81; P = 0.005). No such survival benefit was observed if EML had already occurred in stage A (P > 0.05). CONCLUSIONS:EML during either NCRT or the subsequent preoperative period independently predicts worse survival in LA-ESCC. Nutritional intervention post-NCRT may improve outcomes, particularly for patients who maintain muscle mass during initial therapy.

Tumor-associated tertiary lymphoid structures (TA-TLS) represent ectopic lymphoid neogenesis within chronically inflamed tumors and are increasingly linked to response to immune checkpoint blockade (ICB) and other anticancer therapies. However, TLS are not uniformly beneficial: their immune outputs vary with maturation, spatial niche (intratumoral, invasive margin, or peritumoral), and the balance between effector programs and suppressive modules. This heterogeneity, together with inconsistent definitions and measurement approaches, has limited cross-cohort comparability and slowed translation of TLS into reliable biomarkers and therapeutic targets. In this review, we synthesize key pathways governing TLS induction, formation, and maturation, emphasizing vascular and stromal scaffold requirements that enable lymphocyte recruitment and compartmentalization. We summarize and reconcile existing TLS classification frameworks across tumor types, and integrate mechanistic interactions among T cells, B cells, dendritic cells, myeloid populations, and stromal elements that shape TLS “quality” and immunological function. Importantly, we highlight therapy-induced TLS remodeling as a dynamic process influenced by chemotherapy, radiotherapy dose/fractionation and sequencing, anti-angiogenic strategies, and ICB itself—helping explain divergent clinical associations and suggesting time-ordered opportunities for intervention. Finally, we discuss multimodal approaches for standardized TLS assessment and outline emerging strategies to safely engineer high-quality TLS to enhance antitumor immunity and rationalize combination regimens.

BACKGROUND:Cytidine triphosphate synthase 1 (CTPS1), a rate-limiting enzyme in nucleotide biosynthesis, is frequently upregulated in cancers. However, its functional role and therapeutic potential in lung adenocarcinoma (LUAD) remain poorly defined. METHODS:We integrated proteomic analyses, tissue microarrays, and functional assays to evaluate CTPS1 expression and clinical significance in LUAD. Genetic knockdown and pharmacological inhibition with the selective inhibitor STP938 assessed malignant phenotypes. The therapeutic efficacy of STP938, alone or combined with osimertinib, was evaluated in xenograft models. Integrated proteomic and transcriptomic analyses explored underlying mechanisms. RESULTS:CTPS1 was significantly upregulated in LUAD and associated with lymph node metastasis and poor overall survival (OS), serving as an independent prognostic factor. CTPS1 promoted proliferation, migration, and invasion while suppressing apoptosis. STP938 inhibited tumour growth in vitro and in vivo, and enhanced the efficacy of osimertinib in EGFR-mutant models. Multi-omics analyses identified MX1-associated interferon signalling as a downstream pathway, with MX1 expression positively correlating with CTPS1 and poor prognosis. CONCLUSIONS:CTPS1 drives LUAD progression and is linked to MX1-associated interferon signalling. Targeting CTPS1 with STP938, alone or in combination with osimertinib, represents a promising therapeutic strategy and supports CTPS1 as a targetable vulnerability in LUAD.