Integrating proteomics to map immune landscape and protein biomarkers for risk stratification in childhood Langerhans cell histiocytosis
- 1Department of Hematology and Oncology, Guangzhou Women and Children’s Medical Center, Guangzhou Medical University
- 2Department of Clinical Laboratory Medicine, Guangdong Provincial People’s Hospital (Guangdong Academy of Medical Sciences), Southern Medical University
- 3Department of Ophthalmology, The Third Affiliated Hospital of Chongqing Medical University
Abstract
Background: Langerhans cell histiocytosis (LCH) is a rare pediatric disease with marked clinical heterogeneity. Current risk stratification relies on the extent of organ involvement and risk-organ status, yet reliable protein biomarkers for distinguishing LCH subtypes are lacking.
Methods: We performed proteomic analysis on 56 formalin‐fixed paraffin‐embedded (FFPE) pediatric LCH tissue samples and 10 normal skin controls, quantifying 6, 914 proteins. Machine-learning approaches were integrated with weighted correlation network analysis for exploratory identification of candidate protein biomarkers and internal classification analysis. Candidate markers were validated by immunohistochemistry. The CIBERSORT deconvolution algorithm was employed to estimate the immune cell infiltration.
Results: Proteomic profiling clearly distinguished LCH from controls. Functional protein modules and candidate protein biomarkers were identified across different LCH subtypes. Significant differences in immune cell infiltration were also observed, revealing enhanced innate immune activation and cytotoxic T cell responses in multisystem LCH compared with single-system LCH, as well as in LCH with risk-organ involvement versus those without. In exploratory internal classification analyses, the SVM model yielded an accuracy, AUC, and F1 score of 1.000 (95% CI: 1.000–1.000) for distinguishing multisystem from single-system LCH. For risk-organ involvement, the KNN model yielded an accuracy of 0.800 (95% CI: 0.500–1.000) and an AUC of 0.880 (95% CI: 0.600–1.000).
Conclusion: To our knowledge, this is the first large-scale FFPE-based proteomic study in pediatric LCH to reveal subtype and risk specific protein signatures and associated immune cell profiles changes. These findings provide novel biological insights into LCH heterogeneity and identify candidate protein biomarkers associated with disease extent and risk-organ involvement. Further validation in larger independent cohorts is required before their predictive or clinical utility can be established.
Related
- Extensive vulvar and perianal langerhans cell histiocytosis with clinical improvement during low-dose thalidomide and topical therapy: a case report
- Adult-onset Langerhans cell histiocytosis in an elderly female with cutaneous and nodal involvement: A case report
- Langerhans Cell Histiocytosis of the Vulva: A Pediatric Case.
- No detectable infectious agents in Langerhans cell histiocytosis with lung involvement
- Fine‐Needle Aspirate Cytology of Feline Pulmonary Langerhans Cell Histiocytosis in Two Cats
BibTeX
@article{liu2026integrating,
title = {{Integrating proteomics to map immune landscape and protein biomarkers for risk stratification in childhood Langerhans cell histiocytosis}},
author = {Jinqi Liu and Huizhuang Shan and Jiaguo Yuan and Weina Zhang and Qingqing Zheng and Haiyan Liu and Pengfei Wang and Huiling Lin and Caiyun Kuang and Meifen Huang and Kaoxian Guo and Jiayi Wang and Hua Jiang and Xinhua Xiao},
journal = {Frontiers in Immunology},
year = {2026},
month = oct,
doi = {10.3389/fimmu.2026.1909960},
url = {https://www.frontiersin.org/articles/10.3389/fimmu.2026.1909960/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Integrating proteomics to map immune landscape and protein biomarkers for risk stratification in childhood Langerhans cell histiocytosis
AU - Jinqi Liu
AU - Huizhuang Shan
AU - Jiaguo Yuan
AU - Weina Zhang
AU - Qingqing Zheng
AU - Haiyan Liu
AU - Pengfei Wang
AU - Huiling Lin
AU - Caiyun Kuang
AU - Meifen Huang
AU - Kaoxian Guo
AU - Jiayi Wang
AU - Hua Jiang
AU - Xinhua Xiao
PY - 2026
DA - 2026/10/21
JO - Frontiers in Immunology
DO - 10.3389/fimmu.2026.1909960
UR - https://www.frontiersin.org/articles/10.3389/fimmu.2026.1909960/full
AB - Background: Langerhans cell histiocytosis (LCH) is a rare pediatric disease with marked clinical heterogeneity. Current risk stratification relies on the extent of organ involvement and risk-organ status, yet reliable protein biomarkers for distinguishing LCH subtypes are lacking. Methods: We performed proteomic analysis on 56 formalin‐fixed paraffin‐embedded (FFPE) pediatric LCH tissue samples and 10 normal skin controls, quantifying 6, 914 proteins. Machine-learning approaches were integrated with weighted correlation network analysis for exploratory identification of candidate protein biomarkers and internal classification analysis. Candidate markers were validated by immunohistochemistry. The CIBERSORT deconvolution algorithm was employed to estimate the immune cell infiltration. Results: Proteomic profiling clearly distinguished LCH from controls. Functional protein modules and candidate protein biomarkers were identified across different LCH subtypes. Significant differences in immune cell infiltration were also observed, revealing enhanced innate immune activation and cytotoxic T cell responses in multisystem LCH compared with single-system LCH, as well as in LCH with risk-organ involvement versus those without. In exploratory internal classification analyses, the SVM model yielded an accuracy, AUC, and F1 score of 1.000 (95% CI: 1.000–1.000) for distinguishing multisystem from single-system LCH. For risk-organ involvement, the KNN model yielded an accuracy of 0.800 (95% CI: 0.500–1.000) and an AUC of 0.880 (95% CI: 0.600–1.000). Conclusion: To our knowledge, this is the first large-scale FFPE-based proteomic study in pediatric LCH to reveal subtype and risk specific protein signatures and associated immune cell profiles changes. These findings provide novel biological insights into LCH heterogeneity and identify candidate protein biomarkers associated with disease extent and risk-organ involvement. Further validation in larger independent cohorts is required before their predictive or clinical utility can be established.
SN - 1664-3224
ER -