Machine learning-based integrative analysis identifies CXCL13-driven tertiary lymphoid structures as favorable immune
Jie Jiang1, Jiuhui Xu1, Lu Xie1
1Department of Musculoskeletal Tumor, People's Hospital, Peking University, Beijing, 100044, China.
Cellular Oncology (Dordrecht, Netherlands)
|May 16, 2026
Summary
Tertiary lymphoid structures (TLS) in osteosarcoma correlate with better survival and can predict immunotherapy response. The gene CXCL13 is key to TLS formation and may offer new therapeutic targets for osteosarcoma.
Area of Science:
- Oncology
- Immunology
- Bioinformatics
Background:
- Osteosarcoma is an aggressive bone cancer with poor immunotherapy response.
- Tertiary lymphoid structures (TLS) influence anti-tumor immunity but are understudied in osteosarcoma.
- Understanding TLS in osteosarcoma is crucial for improving treatment outcomes.
Purpose of the Study:
- To characterize TLS in osteosarcoma.
- To develop a prognostic index based on TLS (TLSPI).
- To identify molecular determinants of TLS and their clinical significance.
Main Methods:
- TLS identification via immunohistochemistry and multiplex immunofluorescence.
- Development of a TLSPI using machine learning on transcriptomic data (TARGET, GEO, PKUPH cohorts).
- Analysis of immune infiltration, functional enrichment, and immunotherapy response prediction; validation of CXCL13.
Main Results:
- TLS presence associated with improved osteosarcoma survival (p=0.024).
- TLSPI-related genes involve immune activation and antigen presentation.
- Low-TLSPI group showed increased immune infiltration, immune checkpoints, and predicted immunotherapy sensitivity.
- High CXCL13 expression correlated with better survival and B-cell activation.
Conclusions:
- A TLS-based prognostic index (TLSPI) for osteosarcoma has clinical potential.
- CXCL13 is a key mediator of TLS formation and anti-tumor immunity.
- Findings suggest therapeutic implications for osteosarcoma immunotherapy targeting TLS and CXCL13.
