Translational cancer research最新文献

Background: Metastatic pancreatic ductal adenocarcinoma (mPDAC) has a poor prognosis, with a significant number of patients experiencing early death. Identifying these high-risk patients at diagnosis is critical for personalizing treatment intensity, facilitating timely palliative care discussions, and improving clinical trial stratification. Therefore, this study aimed to develop and validate a machine learning (ML)-based algorithm to estimate the probability of early death in patients with mPDAC.

Methods: We recruited a total of 14,820 patients diagnosed with mPDAC from the Surveillance, Epidemiology, and End Results (SEER) databases. Key exclusion criteria were missing data on survival time or essential variables. The cohort was randomly split into a training set (70%) and an internal test set (30%). For external validation, we retrospectively enrolled patients with mPDAC from a Chinese medical center (2017-2019), representing a distinct geographic and healthcare population. The primary outcome was early death, defined as all-cause mortality within three months of diagnosis. Baseline clinical predictors included demographic, tumor, and treatment characteristics. Four ML models were constructed based on clinical and pathological features. The effectiveness of these models was assessed through various metrics such as the area under the curve (AUC), calibration plots, and decision curve analysis (DCA). The optimal model was selected based on 10-fold cross-validation and its generalizability was internally and externally validated. Additionally, Shapley values for relevant features were calculated using the SHapley Additive exPlanations (SHAP) method.

Results: The extreme gradient boosting classifier (XGBoost) model demonstrated the best performance (AUC =0.757). Crucially, it maintained strong generalizability in the independent external Chinese cohort (AUC =0.780), demonstrating robust cross-population applicability. According to the feature importance ranking plot generated, chemotherapy stood out as the most crucial feature, followed by age, and marital status.

Conclusions: We developed and validated an interpretable ML model that accurately predicts the risk of early death in mPDAC patients. The model's robust performance across US and Chinese populations underscores its broad clinical utility. This tool can assist clinicians in identifying high-risk individuals at diagnosis, thereby informing personalized treatment strategies, prioritizing palliative care, and optimizing resource allocation in diverse healthcare settings.

Background: Malignant bone tumors are rare and highly heterogeneous tumors with poor clinical prognosis and numerous challenges in treatment. Traditional prognostic models may lead to biased assessment of tumor-specific mortality risk due to failure to account for competing risk events such as non-tumor causes of death. The objective of this study was to develop and validate a competing risk model for cancer-specific survival (CSS) in patients diagnosed with malignant bone tumors, and to improve the accuracy of prognostic prediction.

Methods: A total of 3,508 patients with osteosarcoma, chondrosarcoma, and Ewing sarcoma from the Surveillance, Epidemiology, and End Results (SEER) database between 2000 and 2022 were included, and divided into a training set (2,455 cases) and a validation set (1,053 cases) at a ratio of 7:3. Univariate and multivariate Cox regression analyses were used to screen independent risk factors for cancer-specific mortality (CSM), construct a competing risk model, and draw a nomogram. The model performance was evaluated using the consistency index (C-index), area under the receiver operating characteristic curve (AUC), calibration curve, and decision curve analysis (DCA), and compared with the TNM (tumor, node, metastasis) staging system.

Results: Age, gender, primary tumor site, tumor size, clinical stage, surgery, TNMstage, pathological type (Ewing sarcoma), radiotherapy, and chemotherapy were identified as independent risk factors for CSM. The C-index of the model was 0.77 [95% confidence interval (CI): 0.75-0.79] in the training set and 0.79 (95% CI: 0.77-0.82) in the validation set, with AUC >0.8 in both. The calibration curve showed a high degree of agreement between predicted and actual survival rates. DCA results indicated that the clinical net benefit of this model was significantly better than the TNM staging system. Risk stratification showed that the 5-year CSM rate in the high-risk group (65%) was significantly higher than that in the low-risk group (22%, P<0.001).

Conclusions: The competing risk model constructed in this study can accurately predict the CSS probability of patients with malignant bone tumors, with better performance than traditional staging systems, providing a new tool for the development of individualized treatment plans and the identification of high-risk patients.

Background: Breast cancer (BC) is a major health concern among women. SanHuang decoction (SHD), a Chinese herbal formula, has exhibited the potential for clinical application in patients with BC. This study aimed to clarify the mechanism of SHD's action in the treatment of BC based on macrophage function and nuclear factor kappa B (NF-κB) pathway.

Methods: M1 polarized macrophages were cocultured with MDA-MB-231 BC cells, and treated with 10 mg/mL of SHD for 72 hours. Scratch and transwell assays were used to observe the inhibitory effect of SHD, and the levels of interleukin-6 (IL-6), tumor necrosis factor alpha (TNF-α), IL-1β, cleaved caspase 3, B-cell lymphoma 2 (BCL-2), BCL-2-associated X (BAX), P65, phosphorylated P65 (p-P65), inhibitor of kappa B (IκB), and phosphorylated IκB (p-IκB) were detected via Western blotting. Forty patients with BC were randomly divided into a control group [neoadjuvant chemotherapy (NACT)] and observation group (NACT + SHD). After 9 weeks of treatment, magnetic resonance imaging scans were performed to evaluate the tumor size, and immunofluorescence staining was performed to investigate the effect of SHD on the regulation of M1 macrophages in tumor tissue.

Results: SHD inhibited the migration ability of MDA-MB-231 cells, and the effect was significantly enhanced after coculture with M1/THP-1. Western blot results showed that SHD could significantly increase the expression of TNF-α, IL-1β, IL-6, BAX, cleaved caspase 3, p-P65, and p-IκB while decreasing that of BCL-2. In vivo studies showed that SHD could reduce the tumor size and increase the expression of M1 macrophages.

Conclusions: SHD could enhance the expression of M1/THP-1 cytokines, promote the inflammatory response in the tumor microenvironment, and thus inhibit the proliferation of BC and regulate its pathological process, likely through activating NF-κB signaling pathway.

Background: Tumor habitat analysis holds significant application potential in oncology, yet systematic bibliometric studies to characterize its research landscape remain limited. This study aims to comprehensively assess the current status, hotspots, and trends in this field using rigorous bibliometric methods, providing a theoretical framework for future research.

Methods: English publications on medical imaging applications in tumor habitat analysis indexed in Web of Science Core Collection (WOSCC) and PubMed (inception to April 2025) were retrieved. VOSviewer and CiteSpace were used to visualize and analyze country/region contributions, authors, journals, references, and keyword co-occurrences.

Results: A final set of 127 studies was included, revealing a rapid acceleration in research; annual output grew from a single article in 2014 to a peak of 36 in 2024, with 30 articles already published by April 2025. China was the most productive country, while the United States anchored the densest international collaboration network. Key contributors included the University of Ulsan, the journal Frontiers in Oncology, and the most prolific author, Young-Hoon Kim, while Zhou, Mu was the most-cited author. Co-citation analysis revealed two dominant intellectual clusters: one centered on the tumor microenvironment and imaging biology (led by Gillies and Gatenby), and another focused on radiomics and machine learning for prognosis. Keyword analysis confirmed a clear trend toward integrating artificial intelligence (AI), multi-parametric magnetic resonance imaging (MRI), and multi-omics, with a growing emphasis on reproducibility.

Conclusions: Tumor habitat analysis research is growing rapidly, but methodological standardization and data integration remain critical challenges. Addressing these gaps will enhance result comparability and advance translational oncology.

Background: Breast cancer (BC) is the most common malignancy in women and remains a major cause of cancer-related death. Growing observational and multi-omics evidence suggests that dysregulated lipid metabolism and alterations in circulating lipidomes are involved in BC development, yet their causal relationships and underlying mechanisms remain unclear. This study aimed to elucidate the causal effect of lipidomes on BC, as well as its possible mechanism of action, and to investigate the mediating effect of tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL) on the risk of BC.

Methods: Two-sample Mendelian randomization (MR) was used to analyze genome-wide association study (GWAS) data on 179 lipidomes-related single nucleotide polymorphisms (SNPs) and BC (15,680 cases, 167,189 controls) to identify potential mediators. MR-Egger regression, Cochran's Q, MR Pleiotropy RESidual Sum and Outlier (MR-PRESSO), and leave-one-out analysis ensured result robustness. Molecular docking was performed using AutoDock to model TRAIL-death receptor 4 (DR4) interactions with lipidomes. Foldseek homology search analyzed structural similarities, and molecular dynamics simulations identified key binding residues and interaction stability.

Results: This modelling results revealed that in the 179 lipidomes, a notable positive causal relationship was found between phosphatidylcholine (16:0_16:0) and BC risk [P=0.02, odds ratio (OR) =1.0764, 95% confidence interval (CI): 1.0124-1.1443], as well as a causal association between phosphatidylcholine (16:0_16:0) and TRAIL (P=0.02, OR =0.9219, 95% CI: 0.8588-0.9896), along with a negative causal association between TRAIL and BC (P=0.02, OR =0.9504, 95% CI: 0.9110-0.9916). Sensitivity analysis revealed no significant heterogeneity. According to structural homology searches, the molecular recognition between TRAIL and DR4 exhibited evolutionary structural conservation. The binding of phosphatidylcholine significantly weakened the hydrogen bonds, van der Waals interactions, and binding free energy between TRAIL and DR4, thereby disrupting the TRAIL-mediated apoptotic signaling pathway in BC cells.

Conclusions: MR implicates PC (16:0_16:0) in BC risk with TRAIL as a putative mediator, nominating PC (16:0_16:0) for biomarker development and the lipid-TRAIL pathway as a potential therapeutic avenue. Translation will require replication and prospective validation, especially beyond European ancestry.

Background: Colorectal cancer (CRC) is a malignant disease that poses a significant threat to human health; however, early diagnostic and treatment strategies for it remain limited. Immune evasion is a critical factor contributing to treatment failure in CRC. Various cell subtypes within the tumor microenvironment (TME) play essential roles in this process. However, there is currently a lack of a systematic and novel classification of immune evasion-related cell subtypes and an analysis of their dynamic interaction networks within the CRC TME. This study aims to explore a novel classification of immune evasion-related subtypes in CRC, elucidate their underlying mechanisms, and assess their value for immunotherapy and prognosis.

Methods: This study investigates immune escape-related gene expression profiles utilizing single-cell RNA sequencing (scRNA-seq), which were subsequently validated through multiple immunohistochemistry (mIHC) techniques. Non-negative matrix factorization (NMF) clustering was employed to identify novel subtypes associated with immune escape. Additionally, CellChat and pseudotime analysis were utilized to explore intercellular interactions and differentiation pathways. Kyoto Encyclopedia of Genes and Genomes (KEGG), Single-Cell Regulatory Network Inference and Clustering (SCENIC), and immune checkpoint analyses were conducted to elucidate the functional characteristics of these novel subtypes. Furthermore, Cox proportional hazards regression analysis and Kaplan-Meier survival analysis were performed to assess the response to immunotherapy and prognosis.

Results: The expression profile of immune escape-related genes in the TME of CRC was initially plotted. This analysis identified 11 distinct types of immune escape-related cells in the TME, and confirmed that TGF-β⁺JAK1⁺Calretinin⁺ could serve as a candidate cell marker for the immunosuppressive state of the TME. Furthermore, novel subtypes of cancer-associated fibroblasts (CAFs), CD8⁺ T cells, macrophages, and B cells were identified. These subtypes exhibit unique gene expression profiles and functional characteristics associated with CRC immune escape. The cell interaction network, formed by these subtypes and other cells within the TME, facilitates CRC immune escape by reshaping the immunosuppressive microenvironment. Additionally, CFLAR⁺B_cells-C3, CALR⁺CD8⁺T_cells-C2, and TAP1⁺Mac-C2 may serve as potential biomarkers for predicting responses to immunotherapy in CRC patients. In contrast, HEXIM1⁺CAF-C1 may act as an independent risk factor for poor prognosis in CRC.

Conclusions: Our findings enhance understanding of immune escape mechanisms in CRC, show how novel subtypes affect prognosis, and offer insights for new diagnostic and treatment strategies.

Background: Pancreatic neuroendocrine tumors (pNETs) present diagnostic and therapeutic challenges because of their rarity and heterogeneity. This study aimed to identify key genes involved in the development of pNETs and potentially effective prognostic biomarkers.

Methods: We analyzed three datasets from the Gene Expression Omnibus, which included 135 pNET and 13 normal tissues, to identify secretogranin II (SCG2) as a key gene in pNETs. Enrichment analysis revealed that SCG2 expression was negatively correlated with inflammatory responses and interferon signaling. Immune infiltration analysis showed that high SCG2 expression was associated with lower Stromal, Immune, and ESTIMATE scores and a shift in immune cell composition, including reduced γδ T cells and M1 macrophages, and increased M2 macrophages. Potential therapeutic molecules, including telomerase inhibitors and caspase activators, were identified using the Connectivity Map. In our clinical validation of 130 patients with pNETs from The Fourth Hospital of Hebei Medical University, quantitative reverse transcription polymerase chain reaction (qRT-PCR) and immunohistochemistry (IHC) confirmed higher SCG2 expression in pNET tissues.

Results: SCG2 was significantly upregulated in pNETs. High SCG2 expression was associated with significantly lower Stromal, Immune, and ESTIMATE Score. Significant reductions in γδ T cells and macrophages M1, and a significant increase in macrophages M2 were observed, accompanied by a significant decline in various immune chemokines in the high SCG2 expression group. The expression of SCG2 in pNETs was significantly higher, as verified by qRT-PCR and IHC in clinical cohorts. Furthermore, high SCG2 expression was associated with shorter overall survival [hazard ratio (HR) =1.68; 95% confidence interval (CI): 1.432-7.408; P=0.005] and disease-free survival (HR =4.997; 95% CI: 1.288-19.386; P=0.02), establishing it as an independent prognostic factor for pNETs.

Conclusions: These findings indicate that SCG2 is a potential prognostic marker and therapeutic target for pNETs. It may be involved in disease progression by modulating the tumor immune microenvironment, suggesting a possible role in prognostic evaluation and clinical management.

Background: Alectinib is a second-generation tyrosine kinase inhibitor (TKI) that selectively targets anaplastic lymphoma kinase (ALK) rearrangements and is recommended as first-line therapy for patients with advanced ALK-positive non-small cell lung cancer (NSCLC). Pivotal clinical trials have demonstrated its superior efficacy and favorable safety profile compared with earlier ALK inhibitors and chemotherapy. However, long-term real-world outcomes remain insufficiently characterized, particularly in patients harboring concurrent ALK alterations and additional rare genetic variants, whose clinical relevance is often unclear.

Case description: We report a case of a 41-year-old female diagnosed with stage IV lung adenocarcinoma (LUAD) following routine imaging. Comprehensive diagnostic evaluation, including positron emission tomography/computed tomography (PET-CT), cervical lymph node biopsy, and targeted next-generation sequencing, revealed an EML4-ALK fusion (variant 1) together with a concurrent RET p.R820H mutation. The patient initiated first-line treatment with alectinib at a daily dose of 1,200 mg. A partial response was achieved within two months of therapy, and disease control was sustained throughout long-term follow-up. Remarkably, after more than 62 months of continuous alectinib treatment, the patient remained progression-free, with no evidence of disease relapse, distant metastasis, or treatment-related adverse events. The identified RET p.R820H alteration is currently classified as a variant of uncertain significance, and its functional or clinical impact has not been established.

Conclusions: This case demonstrates an exceptionally durable response to first-line alectinib in an ALK-positive LUAD patient with a concurrent rare RET variant. It underscores the long-term efficacy and tolerability of alectinib and highlights the importance of comprehensive genomic profiling in guiding personalized targeted therapy for genetically complex NSCLC.