Research Paper Volume 13, Issue 24 pp 26063—26094

Tertiary lymphoid structure stratifies glioma into three distinct tumor subtypes

Xingwang Zhou1, , Wenyan Li1, , Jie Yang1, , Xiaolan Qi2, , Yimin Chen1, , Hua Yang1, , Liangzhao Chu1, ,

  • 1 Department of Neurosurgery, The Affiliated Hospital of Guizhou Medical University, Guiyang 550004, Guizhou Province, PR China
  • 2 Key Laboratory of Endemic and Ethnic Diseases, Ministry of Education and Key Laboratory of Medical Molecular Biology of Guizhou Province, Guizhou Medical University, Guiyang 550004, Guizhou Province, PR China

Received: September 13, 2021       Accepted: December 11, 2021       Published: December 26, 2021
How to Cite

Copyright: © 2021 Zhou et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


Objective: Tertiary lymphoid structure (TLS), also known as ectopic lymphoid organs, are found in cancer, chronic inflammation, and autoimmune diseases. However, the heterogeneity of TLS in gliomas is unclear. Therefore, it is necessary to identify TLS differences and define TLS subtypes.

Methods: The TLS gene profile of 697 gliomas from The Cancer Genome Atlas (TCGA) was used for consensus clustering to identify robust clusters, and the reproducibility of the stratification method was assessed in Chinese Glioma Genome Atlas (CGGA) cohort1, CGGA_cohort2, and GSE16011. Analyses of clinical characteristics, immune infiltration, and potential biological functions were performed for each subtype.

Results: Three resulting clusters (A, B, and C) were identified based on consensus clustering on the gene expression profile of TLS genes. There was a significant prognostic difference among the clusters, with a shorter survival for C than B and A. In comparison with the A and B subtypes, the C subtype was significantly enriched in primary immunodeficiency, intestinal immune network for lgG production, antigen processing and presentation, natural killer cell-mediated cytotoxicity, complement and coagulation cascades, cytokine-cytokine receptor interaction, leukocyte transendothelial migration, and some immune-related diseases. The levels of 23 immune cell types were higher in the C subtype than in the A and B subtypes. Finally, we developed and validated a riskscore based on TLS subtypes with better performance of prognosis prediction.

Conclusions: This study presents a new stratification method according to the TLS gene profile and highlights TLS heterogeneity in gliomas.


TLS: Tertiary lymphoid structure; TCGA: The Cancer Genome Atlas; CNGA: Chinese Glioma Genome Atlas; Genotype-Tissue Expression; CNAs: copy number alterations; DEGs: Differentially expressed genes; FDR: false discovery rate; FDCs: Follicular dendritic cells; GCs: germinal centers; HEVs: endothelial venules; IRS: immunoreaction score.