The gene signature of tertiary lymphoid structures within ovarian cancer predicts the prognosis and immunotherapy benefit

Abstract

Ovarian cancer (OC) has the lowest survival rate among gynecologic malignancies. Ectopic lymphocyte aggregates, namely tertiary lymphoid structures (TLSs), have been reported as positive biomarkers for tumor prognosis. However, the related gene signature of tertiary lymphoid structure in ovarian cancer was less understood. Therefore, this study first exhibited the organizational patterns of tertiary lymphoid structure by H&E staining and immunohistochemistry (IHC), and confirmed the improved survival values of tertiary lymphoid structure and quantified tumor-infiltrating lymphocytes (CD20+ B cells and CD8+ T cells) in ovarian cancer patients. Secondly, we collected the genes involved in tertiary lymphoid structure from databases. By the univariate regression analysis, the tertiary lymphoid structure gene signature (CETP, CCR7, SELL, LAMP3, CCL19, CXCL9, CXCL10, CXCL11, and CXCL13) with prognostic value, characteristically of ovarian cancer, was constructed in the TCGA dataset and validated in the GSE140082 dataset. Thirdly, by performing CIBERSORT and Tumor Immune Dysfunction and Exclusion (TIDE) analysis, we found that the high expression of this gene signature was positively correlated with developed immune infiltration and reduced immune escape. The improved IPS score and application in the IMvigor210 dataset received PD-L1 proved the predictive value of immunotherapy for this gene signature. Furthermore, this signature showed a better correlation between tumor mutation burden and classical checkpoint genes. In conclusion, Tertiary lymphoid structure plays important role in tumor immunity and the gene signature can be evaluated as a biomarker for predicting prognosis and guiding immunotherapy in ovarian cancer.

A pan‐cancer analysis of prognostic significance and immunological role of lysosomal‐associated membrane protein 3

“…This results that the higher the LAMP3 expression, the better prognosis does not conform to conventional thinking. In OV, the high expression group of tertiary lymphoid structures (TLSs) including the LAMP3 gene exhibited a favourable prognosis for OS and this result is consistent with our database analysis 29 . LAMP3 was reported to be sharply related to locoregional control in patients who received radiation therapy 27 .…”

Spatial transcriptomics reveals discrete tumour microenvironments and autocrine loops within ovarian cancer subclones

“…In the case of the two subclones present in patient 5, single-cell resolution spatial transcriptomics with CosMx SMI revealed both were highly infiltrated by CCL5 +T cells, C1QC+ and CXCL9+ macrophages. Notably, these cells formed clusters resembling tertiary lymphoid structures which have been reported as predictive of prognosis in HGSOC 51 (Fig. 4e ).…”

Spatiotemporal architecture of immune cells and cancer-associated fibroblasts in high-grade serous ovarian carcinoma

“…S5D), we found that plasma cells identified were enriched in PDPN-positive ROIs and that the immune infiltrates in PDPN-positive ROIs were enriched in genes associated with plasma cell differentiation and/or recruitment (JCHAIN, CD38, IGHG3, CD79A, SLAMF7, IGKC, PRDM1, and XBP1; fig. S5E), while the immune infiltrates in PDPN-negative ROIs were enriched in genes typically associated with T and B cells in mature TLSs (CXCL13, MS4A1, CD8A, CD3D, CD3E, CCL19, TRBC1, CD2, CXCL9, and CCR7) ( 61 ). As an independent validation method for the expression of different cytokines and chemokines and their receptors in PDPN-positive and TLS-positive areas, we stratified patients with ovarian cancer in The Cancer Genome Atlas (TCGA) into PDPN High TLS Low and PDPN Low TLS High based on the expression of PDPN and the TLS gene signature ( 62 ).…”