Skip to content
DraconiticResearch
Search Sources API
203,234 studies31 sourcesUpdated 30 min ago
Peer-reviewedCC BY-NC-NDTranslational cancer research

Development and external validation of an 18-gene tertiary lymphoid structure-related prognostic signature in gastric cancer.

  1. 1Department of Gastrointestinal Surgery, The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China.
  2. 2Department of Thyroid and Breast Surgery, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China.

Abstract

Tertiary lymphoid structures (TLS) contribute to local antitumor immunity, but TLS-related transcriptional signals do not directly measure histological TLS abundance or maturation. This study aimed to develop and externally evaluate a biologically constrained TLS-related gene (TLSRG) signature for prognostic stratification in gastric cancer and characterize its associated immune-stromal programmes. Transcriptomic and clinical data from 350 patients in The Cancer Genome Atlas stomach adenocarcinoma (TCGA-STAD) cohort and 431 patients in GSE84437 were used for development and external evaluation, respectively. Least absolute shrinkage and selection operator (LASSO)-Cox regression was applied to a predefined 39-gene panel. Overall survival (OS) was assessed using Kaplan-Meier analysis, Cox regression, the concordance index (C-index), and time-specific receiver operating characteristic (ROC) analysis. Additional analyses evaluated TLS-related cell-state and functional transcriptional surrogates, model-gene-overlap-excluded scores, cross-cohort association networks, 18-gene co-expression, and area under the curve (AUC) uncertainty using 2,000 patient-level bootstrap resamples. A combined clinical model underwent internal cross-validation using the precomputed risk score. The 18-gene signature separated survival in TCGA-STAD [high vs. low risk: hazard ratio (HR) =3.108; 95% confidence interval (CI): 2.178-4.436] and GSE84437 (HR =1.410; 95% CI: 1.071-1.856). The score remained associated with survival after age, sex, and pathological stage or tumor-node-metastasis (TNM) adjustment in separate TCGA-STAD models. External 1-, 3-, and 5-year AUCs were 0.526 (95% CI: 0.443-0.607), 0.597 (0.540-0.654), and 0.606 (0.548-0.661), indicating limited discrimination, particularly at 1 year. Risk was consistently associated with epithelial-mesenchymal transition (EMT)/stromal and angiogenesis scores and with CXCL12 expression after excluding CXCR4 from the two-gene axis score. Cell-state and immune-function contrasts were cohort-dependent rather than uniformly reduced in high-risk tumors. The 18-gene network retained 102 consensus co-expression edges; CXCL13, ICOS, PDCD1, CD79A, and CCR7 had the highest connectivity, whereas CXCR4 had the largest model coefficient. The combined model had an internal C-index of 0.704 vs. 0.617 for clinical variables alone. The TLSRG signature was associated with survival and reproducible stromal/angiogenic transcriptional features, although external discrimination was modest. Prospective spatial and functional studies are needed to determine its relationship with TLS maturation and clinical utility.

Related

  1. An interpretable machine-learning signature predicts early triple-negative breast cancer recurrence but adds no prognostic value beyond clinicopathological factors in external validation
    Haocheng Nan, Rong Di, Xiaorong Kou +1
    Scientific ReportsPeer-reviewedCC BY-NC-ND 4.0Full textOncologyCrossref
  2. Development and external validation of an XGBoost-based prognostic model for small cell lung cancer with brain metastases
    Pei Jiao, Chen Chen, Koudong Zhang +2
    Translational cancer researchPeer-reviewedCC BY-NC-NDFull textGeneral medicineEurope PMC
  3. A Pathological Image-Based Pathomics Signature Combined with Clinicopathological Features for Predicting Lymph Node Metastasis in Esophageal Squamous Cell Carcinoma: A Development and External Validation Study
    Kaiming Peng, Jieming Lu, Peipei Zhang +5
    CancersPeer-reviewedCC BY 4.0Full textOncologyCrossrefEurope PMC
  4. Development and external validation of a psychological stress-related gene signature for overall survival prediction in bladder cancer.
    Pengwei Luo, Jiajun Xie, Rong Kang +5
    Translational andrology and urologyPeer-reviewedCC BY-NC-NDFull textSurgery & anaesthesiaEurope PMC
  5. Baseline 25-hydroxyvitamin D and progression-free survival in de novo metastatic hormone-sensitive prostate cancer: prognostic model development and external validation.
    Galip Can Uyar, Kadriye Başkurt, Orhun Akdoğan +8
    The oncologistPeer-reviewedCC BYFull textGeneral medicineEurope PMC
BibTeX
@article{han2026development,
  title = {{Development and external validation of an 18-gene tertiary lymphoid structure-related prognostic signature in gastric cancer.}},
  author = {Xiaying Han and Qimiao Chen and Zhixing Hao and Jing Chen and Jia Qi and Jian Chen},
  journal = {Translational cancer research},
  year = {2026},
  month = sep,
  doi = {10.21037/tcr-2026-2004},
  url = {https://europepmc.org/article/MED/42997890},
  note = {CC BY-NC-ND},
}
RIS
TY  - JOUR
TI  - Development and external validation of an 18-gene tertiary lymphoid structure-related prognostic signature in gastric cancer.
AU  - Xiaying Han
AU  - Qimiao Chen
AU  - Zhixing Hao
AU  - Jing Chen
AU  - Jia Qi
AU  - Jian Chen
PY  - 2026
DA  - 2026/09/28
JO  - Translational cancer research
DO  - 10.21037/tcr-2026-2004
UR  - https://europepmc.org/article/MED/42997890
AB  - Tertiary lymphoid structures (TLS) contribute to local antitumor immunity, but TLS-related transcriptional signals do not directly measure histological TLS abundance or maturation. This study aimed to develop and externally evaluate a biologically constrained TLS-related gene (TLSRG) signature for prognostic stratification in gastric cancer and characterize its associated immune-stromal programmes. Transcriptomic and clinical data from 350 patients in The Cancer Genome Atlas stomach adenocarcinoma (TCGA-STAD) cohort and 431 patients in GSE84437 were used for development and external evaluation, respectively. Least absolute shrinkage and selection operator (LASSO)-Cox regression was applied to a predefined 39-gene panel. Overall survival (OS) was assessed using Kaplan-Meier analysis, Cox regression, the concordance index (C-index), and time-specific receiver operating characteristic (ROC) analysis. Additional analyses evaluated TLS-related cell-state and functional transcriptional surrogates, model-gene-overlap-excluded scores, cross-cohort association networks, 18-gene co-expression, and area under the curve (AUC) uncertainty using 2,000 patient-level bootstrap resamples. A combined clinical model underwent internal cross-validation using the precomputed risk score. The 18-gene signature separated survival in TCGA-STAD [high vs. low risk: hazard ratio (HR) =3.108; 95% confidence interval (CI): 2.178-4.436] and GSE84437 (HR =1.410; 95% CI: 1.071-1.856). The score remained associated with survival after age, sex, and pathological stage or tumor-node-metastasis (TNM) adjustment in separate TCGA-STAD models. External 1-, 3-, and 5-year AUCs were 0.526 (95% CI: 0.443-0.607), 0.597 (0.540-0.654), and 0.606 (0.548-0.661), indicating limited discrimination, particularly at 1 year. Risk was consistently associated with epithelial-mesenchymal transition (EMT)/stromal and angiogenesis scores and with CXCL12 expression after excluding CXCR4 from the two-gene axis score. Cell-state and immune-function contrasts were cohort-dependent rather than uniformly reduced in high-risk tumors. The 18-gene network retained 102 consensus co-expression edges; CXCL13, ICOS, PDCD1, CD79A, and CCR7 had the highest connectivity, whereas CXCR4 had the largest model coefficient. The combined model had an internal C-index of 0.704 vs. 0.617 for clinical variables alone. The TLSRG signature was associated with survival and reproducible stromal/angiogenic transcriptional features, although external discrimination was modest. Prospective spatial and functional studies are needed to determine its relationship with TLS maturation and clinical utility.
SN  - 2218-676X
SN  - 2219-6803
ER  -