918P - SPP1-related M2 macrophage signatures predict prognosis and immunotherapy response in patients with nasopharyngeal carcinoma

Background

The M2 macrophages in the tumor immune microenvironment (TIME) have been identified as independent prognostic factors in nasopharyngeal carcinoma (NPC). However, the clinical significance of SPP1-related M2 macrophages remains largely unexplored.

Methods

To reconstruct a pseudotime trajectory of macrophage polarization in NPC and analyze the gene expression patterns along the trajectory, we applied publicly available single-cell sequencing data (GSE150825). GeneSwitches was used to sequence the critical regulatory genes during M2 macrophage activation and identify their order. To construct the M2 macrophage-activation genes signature (MAGens), we analyzed the transcriptome data of the Fujian NPC dataset (N=188) and the published dataset (GSE102349, N=88) using 101 combinations of machine learning algorithms that included univariate and correlation analyses.

Results

We utilized single-cell transcriptomics and GeneSwitches analyses to identify SPP1 as the top switching gene in M2 macrophage activation and determine the order of the switch. Analysis of the bulk transcriptome revealed that high expression of SPP1 is associated with poor prognosis and increased infiltration of M2 macrophages, which was validated using immunohistochemistry. We then established a 13-gene MAGens that accurately predicts progression-free survival in patients with nasopharyngeal carcinoma (NPC) (5-year AUC: 0.95) and serves as an independent prognostic factor based on uni- and multivariate analyses. We found that the low MAGens group had a higher level of immune infiltrating cells and immune modulators, indicating an inflamed but relatively immunosuppressive microenvironment that could potentially benefit from immunotherapy. Lastly, we verified the predictive value of MAGens in the Fujian NPC dataset, where MAGens score exhibited a significant positive correlation with known immunotherapy indicators CD8, PD-1, and PD-L1.

Conclusions

As a new predictive biomarker, MAGens offer significantly improved precision in selecting the NPC population that could benefit from immunotherapy. We recommend the immediate validation and application of these unique MAGens.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

National Natural Science Foundation of China; Major Research Projects for Young and Middle-aged Health Researchers of Fujian Province, China; Joint Funds for the Innovation of Science and Technology, Fujian Province; Innovative Medicine Subject of Fujian Provincial Health Commission, China; and High-level Talent Training Program of Fujian Cancer Hospital.

Disclosure

All authors have declared no conflicts of interest.