2346P - Spatial and quantified molecular characterization of high endothelial venule predictive of immunotherapy response

Background

High endothelial venules (HEVs) are important structures that mediate lymphocyte entry into tumors. However, it remains difficult to identify HEVs in tumor at the transcriptome level. Therefore, defining a molecular signature of HEVs is of great significance for characterizing HEVs in tumor, and identifying potential patients who could benefit from immunotherapy.

Methods

We collected scRNA-seq data of tumor tissues of 11 tumor types, and integrated 19923 endothelial cells. Genes that were specifically highly expressed in post-capillary venule subpopulation and most related to the function of HEVs were identified as the HEV signature. Bulk-seq data of tumor blood vessel and spatial transcriptome sequencing data of tumor tissue were used to verify the specificity of the signature to identify HEVs. Immunohistochemistry and multiplex immunofluorescence were used to characterize the tumor HEVs and immune microenvironment. Cohorts of tumor patients who received immunotherapy were used to explore the relationship between the HEV signature and patients' response to immunotherapy.

Results

Seven genes including ACKR1, SELE, SELP, IL33, MEOX1, HAPLN3, and CPXM2 were screened out as the HEV signature. HEV and non-HEV endothelial cells can be well distinguished according to the expression level of HEV signature. Spatially, regions with higher expression of HEV signature were associated with the aggregates of lymphocytes. MEOX1 was chosen as a representative of these genes. Using immunohistochemistry of MEOX1 and multiplex immunofluorescence of the tumor microenvironment, we confirmed its co-localization with MECA79 antigen and found that lymphocyte-rich niches formed around MEOX⁺ blood vessels. In addition, higher expression of HEV signature was associated with better response to immunotherapy and longer survival in cohorts obtained from other studies as well as our in-house cohort from Nanfang Hospital.

Conclusions

A 7-gene HEV signature we identified, especially MEOX1, can be used as a new molecular marker to identify HEV in tumors, and its expression can be used to locate and quantify HEV in tumors, and help identify potential patients who could benefit from immunotherapy.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.