Cancer stem cells (CSCs) — known to be resistant to genotoxic radiation and chemotherapy — are fundamental to therapy failure and cancer relapse. Here, we reveal that glioma CSCs are hypersensitive to radiation, but a temporal DNA repair mechanism converts the intrinsic sensitivity to genomic instability and treatment resistance. Transcriptome analysis identifies DNA-dependent protein kinase (DNA-PK) as a predominant DNA repair enzyme in CSCs. Notably, DNA-PK activity is suppressed after irradiation when ROS induce the dissociation of DNA-PKcs with Ku70/80, resulting in delayed DNA repair and radiosensitivity; subsequently, after ROS clearance, the accumulated DNA damage and robust activation of DNA-PK induce genomic instability, facilitated by Rad50-mediated cell-cycle arrest, leading to enhanced malignancy, CSC overgrowth, and radioresistance. Finally, we show a requisite in vivo role for DNA-PK in CSC-mediated radioresistance and glioma progression. These findings identify a time-sensitive mechanism controlling CSC resistance to DNA-damaging treatments and suggest DNA-PK/Rad50 as promising targets for CSC eradication.

Tumor-associated macrophages (TAMs) are pivotal immune cells within the tumor microenvironment (TME), exhibiting dual roles across various cancer types. Depending on the context, TAMs can either suppress tumor progression and weaken drug sensitivity or facilitate tumor growth and drive therapeutic resistance. This study explores whether targeting TAMs can suppress the progression of head and neck squamous cell carcinoma (HNSCC) and improve the efficacy of chemotherapy. Bioinformatics analyses were performed to evaluate TAMs infiltration levels in HNSCC tumor tissues and examine their associations with patients’ clinicopathological characteristics and prognosis. Flow cytometry was utilized to measure the expression of key macrophage markers and assess apoptosis following treatment with colony stimulating factor 1 receptor (CSF1R) inhibitors (BLZ945, PLX3397). Additionally, immunohistochemistry was employed to detect CD68 and CD8 expression. In vivo, the antitumor efficacy of CSF1R inhibitors was tested in mouse HNSCC tumor model, both as monotherapy and in combination with cisplatin, to evaluate potential synergistic effects. Bioinformatic analysis identified TAMs as the predominant infiltrating immune cells in the TME of HNSCC, with significantly higher infiltration levels in tumor tissues compared to adjacent non-tumor tissues. High TAMs infiltration was associated with poorer overall survival (OS), disease-free survival (DFS), human papillomavirus (HPV) infection status, and advanced disease stages. The TAMs-related genes prediction model demonstrated high prognostic accuracy. CSF1R is primarily expressed in TAMs, where high CSF1R expression may suppress antigen binding and activation. In vitro experiments showed that CSF1R inhibitors induce TAMs apoptosis, enhance their phagocytic activity, and reduce CD206 expression and IL-10 secretion, thereby diminishing their immunosuppressive function. In vivo experiments revealed that while CSF1R inhibitors alone had limited efficacy in suppressing tumor growth, their combination with cisplatin significantly enhanced therapeutic efficacy, as evidenced by increased CD8+ T cells infiltration within the TME. Targeting TAMs via CSF1R inhibition enhances the therapeutic efficacy of cisplatin in HNSCC. These findings suggest that CSF1R inhibitors hold promise as a component of combination therapy for HNSCC.

Background: A growing number of clinical studies have confirmed that mRNA vaccines are effective in the treatment of malignant tumors; however, their efficacy in head and neck squamous cell carcinoma (HNSCC) has not been determined. This study aimed to identify the potential antigens of HNSCC for mRNA vaccine development and further distinguish the immune subtypes of HNSCC to select suitable patients for vaccination. Methods: We obtained gene expression profiles and corresponding clinical information of HNSCC from Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA). We visualized the genetic alterations of potential antitumor antigens using cBioPortal and obtained the immune gene set from Immport. The correlation between the expression of the identified antigens and the infiltration of antigen-presenting cells was visualized by Tumor Immune Estimation Resource (TIMER). We evaluated the potential biological functions of different samples and described the immune landscape. Results: Increased expression of three potential tumor antigens, CCR4, TMCO1, and SPACA4, associated with superior prognoses and infiltration of antigen-presenting cells, was identified in HNSCC. Three immune subtypes (C1–C3) with different molecular, cellular, and clinical characteristics were defined. Patients with C3 tumor had a better prognosis, representing an immune “cold” phenotype, which may be more suitable for mRNA vaccination. In addition, different immune characteristics were observed among the three immune subtypes, including markers of immune cells, mutation burden, expression of immune checkpoints, and immune modulators. Finally, the immune landscape of HNSCC showed a high degree of heterogeneity between individual patients. Conclusion: CCR4, TMCO1, and SPACA4 may be potential antigens for developing mRNA vaccines against HNSCC, especially for patients with C3 tumor. This study could provide a theoretical basis for the development of an mRNA vaccine against HNSCC.

Objective: Poorly differentiated thyroid cancer (PDTC) is a rare, heterogeneous carcinoma from follicular cells, characterized by poor differentiation, aggressive spread, and poor prognosis. Currently, there is no specific staging system for PDTC. This study aimed to develop a new TNM staging system tailored to PDTC for improved disease management.Methods: A new TNM staging system was designed and internally validated using data from the US SEER database (2004-2016) on PDTC cases. External validation was performed using data from four major institutions in China. Prognostic factors influencing cancer-specific survival (CSS) were identified through Cox regression analyses. Patients were stratified into subgroups based on adjusted hazard ratios (AHRs), weighted by these prognostic factors. The new system classified patients into five stages with distinct 5-year CSS outcomes.Results: The study analyzed 1,201 PDTC cases from SEER and 85 cases from China. Among the 876 patients in the training cohort, the new TNM staging system showed superior discrimination compared to the 8th edition of the AJCC TNM system. The 5-year CSS rates for the new stages I, II, III, IVA, and IVB were 96.3%, 88.4%, 69.4%, 43.3%, and 22.3%, respectively. The new system outperformed the 8th edition in predicting CSS, as shown by time-dependent ROC curves, C-index, and calibration plots. Both internal and external validation confirmed its predictive abilities.Conclusion: The current AJCC staging system inadequately predicts PDTC prognosis. The new TNM staging system developed in this study offers improved stratification and prognosis prediction, potentially guiding more effective clinical management for PDTC.

Postoperative papillary thyroid cancer (PTC) patients often have enlarged cervical lymph nodes due to inflammation or hyperplasia, which complicates the assessment of recurrence or metastasis. This study aimed to explore the diagnostic capabilities of computed tomography (CT) imaging and radiomic analysis to distinguish the recurrence of cervical lymph nodes in patients with PTC postoperatively. A retrospective analysis of 194 PTC patients who underwent total thyroidectomy was conducted, with 98 cases of cervical lymph node recurrence and 96 cases without recurrence. Using 3D Slicer software, Regions of Interest (ROI) were delineated on enhanced venous phase CT images, analyzing 302 positive and 391 negative lymph nodes. These nodes were randomly divided into training and validation sets in a 3:2 ratio. Python was used to extract radiomic features from the ROIs and to develop radiomic models. Univariate and multivariate analyses identified statistically significant risk factors for cervical lymph node recurrence from clinical data, which, when combined with radiomic scores, formed a nomogram to predict recurrence risk. The diagnostic efficacy and clinical utility of the models were assessed using ROC curves, calibration curves, and Decision Curve Analysis (DCA). This study analyzed 693 lymph nodes (302 positive and 391 negative) and identified 35 significant radiomic features through dimensionality reduction and selection. The three machine learning models, including the Lasso regression, Support Vector Machine (SVM), and RF radiomics models, showed.