To develop and validate a longitudinal deep-learning based survival prediction model for esophageal squamous cell carcinoma (ESCC) patients who received definitive concurrent chemoradiotherapy (CCRT). A total of 257 ESCC patients from two centers were recruited. Among them, 205 patients were in the training cohort and 52 in the external testing cohort. The CrossFormer algorithm was utilized to extract features from pre- and post- treatment CECT scans. We constructed clinical, Delta-radiomics and deep-learning models. Models were evaluated by the C-index and integrated Brier score (iBS). Prognostic stratification was performed based on risk scores, and model interpretability was evaluated using Grad-CAM and SurvSHAP(t). The Fusion model demonstrated superior predictive performance, achieving a C-index of 0.768 (95% CI: 0.731-0.804) in the training cohort and 0.734 (95% CI:0.665-0.803) in the testing cohort. The Fusion model also showed the best calibration (Training cohort: iBS=0.096; Testing cohort: iBS=0.151). Patients were stratified into high-risk and low-risk groups based on risk scores, with significant differences in overall survival (OS) between the groups (P < 0.001). We developed a model integrating longitudinal CECT scans to predict OS in ESCC patients undergoing CCRT. The results highlight the importance of capturing tumor changes during treatment for accurate prognostic stratification. The model shows its potential for guiding personalized treatment strategies in clinical practice.

The occurrence and development of lung cancer (LC) involve complex interactions between various cell types in the tumor microenvironment (TME). Understanding the spatial distribution and interaction mechanisms of these cells may be the key to overcoming LC. The advancements of single-cell and spatial transcriptome techniques have promoted our understanding of cellular neighborhoods (CNs) and their functions in the pathogenesis of LC. In this review, we focus on the impact of different etiologies on LC CNs and the current research status of CNs in LC. This review may provide new insights into the molecular mechanisms of LC pathogenesis, develop more refined classification principles for LC diagnosis, and offer new perspectives for LC treatment.

Transarterial chemoembolization (TACE) is the standard treatment for intermediate unresectable hepatocellular carcinoma (HCC); however, reliable prognostic markers are still lacking. Sarcopenia has been proposed as a negative prognostic factor in HCC, but its impact on TACE outcomes remains unclear. We retrospectively analyzed 48 HCC patients treated with TACE or transarterial embolization (TAE) at our institution (between 2013 and 2020). Sarcopenia was assessed on computed tomography (CT) or magnetic resonance imaging (MRI) scans at baseline, one month, and six months after treatment according to RECIST criteria. At six months, 27 patients (61.4%) achieved complete or partial response, while 17 (38.6%) experienced stable or progressive disease; four patients were excluded due to missing follow-up data. Sarcopenia was more frequent among responders, increasing from 13.5% at baseline to 22.2% in 6 months, while it was initially absent in non-responders. Conversely, non-responders showed a later increase in sarcopenia (0% at baseline to 29.6% at 6 months), suggesting that late sarcopenia might reflect treatment-related metabolic changes. Overall, the prevalence of sarcopenia increased during follow-up. New-onset sarcopenia was more frequent in non-responders and was associated with lower serum creatinine levels , suggesting a possible link between treatment-related muscle loss and poor therapeutic response. Kaplan-Meier showed that smoking status was associated (p = 0.01) with poorer response at 6 months (), while sarcopenia and low creatinine levels showed borderline associations (p = 0.095). In this exploratory study, baseline CT-defined sarcopenia was not significantly associated with short-term response to TACE. However, treatment-related sarcopenia with low creatinine levels may reflect frailty during follow-up, and poorer therapeutic response. Given the small sample size and limited number of sarcopenic patients, these findings should be considered hypothesis-generating and require validation in larger prospective studies.

MutS Homolog 3 (MSH3), part of the MutSβ DNA mismatch repair complex with MSH2, can reversibly translocate from the nucleus to cytosol via IL-6 signaling, abrogating nuclear MutSβ function and is associated with metastasis and poor patient survival. A polymorphism consisting of deletion of 27-bp proximate to the nuclear localization signal (NLS) (Δ27bpMSH3) allows MSH3 cytosolic retention with IL-6 or oxidative stress. Here, we examined for IL-6-induced post-translational modifications associated with MSH3 cytosolic translocation. We utilized MSH3-genotyped colon cancer cell lines after IL-6 treatment to assess post-translational modification of MSH3 via Western blots (WB). We modified sequences within the MSH3-NLS-EGFP reporter construct to assess MSH3 localization via immunofluorescent microscopy and WB after nuclear-cytosolic fractionation. Immunoprecipitation (IP) followed by WB was used to study post-IL-6-induced interactions with MSH3. MSH3 and Δ27bpMSH3 increased serine phosphorylation after 2 hours followed by tyrosine phosphorylation 18 hours post IL-6 treatment, with Δ27bpMSH3 showing more robust phosphorylation than MSH3 likely due to increased cytosolic translocation. MSH3 cytosolic localization was enhanced by acetylation of lysine residues within MSH3's NLS, specifically at residues K99, K100 and K103. With the observed acetylation control for MSH3 cytosolic localization, IP experiments demonstrate binding of cytosolic-located histone deacetylase 6 (HDAC6) to acetylated Δ27bpMSH3. Polymorphic MSH3 undergoes serine/tyrosine phosphorylation and NLS acetylation upon IL-6 signaling for its nuclear-cytosolic shift and binds HDAC6 in the cytosol which may contribute to anticipated deacetylation and MSH3 protein stability when separated from MSH2. These modifications might be targeted to regulate MSH3's intracellular localization.

Angiogenesis is a pivotal process for tumor progression and metastasis in non-small cell lung cancer (NSCLC). However, the molecular mechanisms by which WNT1-inducible signaling pathway protein 3 (WISP-3) contributes to NSCLC angiogenesis remain poorly defined. This study investigated the role of WISP-3 in regulating pro-angiogenic signaling in lung adenocarcinoma (LUAD) cells. Conditioned medium from H1299 and A549 cells treated with recombinant WISP-3 (0-100 ng/mL) significantly and dose-dependently enhanced the tube formation of human umbilical vein endothelial cells (HUVECs). WISP-3 selectively upregulated platelet-derived growth factor A (PDGF-A) expression at both mRNA and protein levels in NSCLC cell lines, while other angiogenic factors remained unaffected. Notably, knockdown of PDGF-A using siRNA markedly abolished WISP-3-induced HUVEC tube formation, confirming PDGF-A as a critical mediator in this process. Mechanistically, WISP-3 rapidly triggered the phosphorylation of p38 and JNK signaling pathways. These activations led to the phosphorylation of the transcription factor c-Jun, which in turn promoted PDGF-A gene expression. Pharmacological inhibition of p38 (Adezmapimod), JNK (SP600125), or c-Jun (T-5224) effectively suppressed WISP-3-induced c-Jun activation, PDGF-A expression, and subsequent angiogenesis. Collectively, our findings identify a novel WISP-3/p38-JNK/c-Jun/PDGF-A signaling axis that drives vascular remodeling in NSCLC. Targeting WISP-3 or its downstream effectors may represent a promising therapeutic strategy for anti-angiogenic treatment in lung cancer.

IntroductionThe ypT staging system has limited prognostic value after neoadjuvant therapy, as it primarily reflects only tumor characteristics alone. This study proposes a novel staging system that integrates circumferential resection margin (CRM) status with the ypT category to enhance prognostic accuracy following neoadjuvant chemoradiotherapy (nCRT) for rectal cancer.MethodsWe analyzed data from 4,308 rectal adenocarcinoma patients treated with nCRT followed by surgery, using the Taiwan Cancer Registry and National Health Insurance Research Database (2011-2021). CRM involvement was defined as a margin ≤1 mm. Overall survival was assessed using multivariable Cox regression, and prognostic performance of the proposed CRM-integrated ypT staging system was compared with the American Joint Committee on Cancer (AJCC) TNM system using Harrell's c-statistic.ResultsCRM involvement (≤1 mm) was significantly associated with worse 5-year survival (adjusted odds ratio, 0.44; 95% CI, 0.31-0.61). Due to the low rate of CRM positivity in ypT0-2 patients, a modified ypT classification was established: new ypT3 (ypT3 and CRM-), new ypT4A (ypT4A and CRM-), new ypT4B (ypT3 and CRM+ or ypT4B and CRM-), and new ypT4C (ypT4A and CRM+ or ypT4B and CRM+). This system demonstrated better prognostic discrimination than the current AJCC classification (Harrell's c-statistic: 0.756 vs. 0.752, P = 0.034).ConclusionsIncorporating CRM into the ypT stage offers survival stratification and may guide more individualized postoperative treatment strategies for rectal cancer patients after nCRT.

IntroductionAdult-type granulosa cell tumors (AGCTs) are rare ovarian neoplasms with a low overall incidence of recurrence, and also data on secondary recurrence and survival after relapse remain limited. This study aimed to identify factors associated with secondary recurrence and survival after recurrence in patients with recurrent AGCTs.MethodsThis multicenter retrospective study included 52 patients with recurrent AGCTs identified among 484 patients treated between 2000 and 2023. Clinical characteristics, treatment modalities, and outcomes were analyzed, with a particular focus on factors associated with secondary recurrence and survival after first recurrence. Recurrence-free survival and overall survival after first recurrence (OS-FR) were evaluated using Kaplan-Meier analysis.ResultsThe mean follow-up duration was 99.2 ± 61.5 months. Secondary recurrence occurred in 17 patients (32.7%). A serum CA-125 level >35 U/mL at the time of first recurrence was significantly associated with an increased risk of secondary recurrence (p=0.01). Factors significantly associated with improved OS-FR included a CA-125 level ≤35 U/mL at initial diagnosis and at first recurrence, absence of residual disease following surgery for the first recurrence, and administration of salvage chemotherapy (all p<0.05). In subgroup analysis, salvage chemotherapy was associated with improved OS-FR in patients with residual disease or those who did not undergo surgery (p < 0.01), but not in patients who achieved complete cytoreduction (p = 0.67).ConclusionsSecondary recurrence remains a significant clinical challenge in AGCTs. Serum CA-125 levels, surgical outcomes at first recurrence, and the use of salvage chemotherapy may help management strategies in recurrent disease.

BackgroundPapillary thyroid carcinoma (PTC), the most common thyroid malignancy, shows marked clinical heterogeneity despite generally favorable outcomes. Lysosome-dependent cell death (LDCD), a form of programmed death triggered by lysosomal membrane permeabilization, has emerged as a potential cancer therapy target, but its role in PTC remains unclear.MethodsTranscriptomic data from public cohorts were analyzed to identify LDCD-related genes (LDCDRG) associated with PTC prognosis. Cox analysis and LASSO regression analyses were performed to construct a prognostic model. Immune landscape, drug sensitivity, and single-cell expression profiles were examined. Functional experiments were conducted in vitro to verify the biological effects of the key gene LMTK3 on PTC cell proliferation, viability, and invasion.ResultsNineteen LDCDRG were differentially expressed between normal and tumor tissues, defining three molecular subtypes with distinct immune and prognostic profiles. A six-LDCDRG signature (LMTK3, MCM5, NXF1, TUBB4B, LIMCH1 and APH1B) effectively stratified patients into high- and low-risk groups with significantly different survival outcomes and acceptable predictive performance. High-risk patients showed reduced immune infiltration and lower predicted immunotherapy-related immune activity. LMTK3, the highest-risk gene, was highly expressed in PTC cells, and its knockdown suppressed proliferation and invasion in vitro.ConclusionsThe established six-LDCDRG signature provides an exploratory tool for risk stratification and survival prediction, while LMTK3 emerges as potential target worthy of further investigation. These findings deepen our understanding of lysosome-dependent cell death in thyroid carcinogenesis and may provide insights into the development of personalized management strategies and novel treatment approaches for high-risk PTC patients.

Glioblastoma multiforme (GBM), the most aggressive primary brain tumor, is characterized by high recurrence, metabolic plasticity, and complex tumor microenvironmental interactions. The human chitinase and chitinase-like protein family includes five members (CHI3L1, CHI3L2, CHIA, CHID1, and CHIT1) that share conserved chitinase-related domains but exhibit diverse biological functions in immune regulation and tissue remodeling. While chitinase-like proteins are recognized as mesenchymal-associated markers, however, the role of CHID1 in GBM remains largely unexplored. An integrative multi-omics strategy combining TCGA-GBM and CGGA transcriptomic datasets, single-cell RNA sequencing, and enrichment analyses (GSEA, GO, KEGG, and MetaCore) were used to investigate CHID1 expression patterns and associated transcriptional programs. Pharmacogenomic correlations and molecular docking were used to explore potential drug-response associations. CHID1 showed higher expression in GBM compared to the normal brain and was associated with poor overall survival. A single-cell analysis showed tumor-associated expression patterns of CHID1 across malignant samples. Pathway enrichment analyses identified transcriptional programs related to oxidative phosphorylation, redox-related processes, DNA repair, and cell cycle pathways. Collectively, this study provides a comprehensive multi-cohort and multi-modal characterization of CHID1 expression in GBM, integrating bulk transcriptomics, single-cell RNA sequencing, and tissue-level validation. The findings establish CHID1 as a GBM-associated transcriptional marker linked to metabolic and redox-related programs and provide a systematic resource for future investigations into chitinase family-related biology in GBM.