Study explores the diverse expression patterns of T cell priming markers across cancers

By Tarun Sai LomteAug 10 2023Reviewed by Sophia Coveney

A recent study published in npj Genomic Medicine explored transcriptomics of T-cell priming markers (TPMs) across different cancers.

Background

Immune interventions by vaccines, cytokines, chimeric antigen receptor (CAR) T cells, and immune checkpoint blockade (ICB) have been studied as cancer treatment approaches. The Food and Drug Administration (FDA) approved the first immune checkpoint inhibitor, ipilimumab, for advanced melanoma in 2011. Programmed death (PD)-1 and PD-ligand (PDL)-1 inhibitors were approved in subsequent years for different cancers.

Despite the success of ICB, only some patients are responsive. Comprehensive immune signatures and response markers have been explored to circumvent this challenge and identify patients who would benefit from ICB. Another potential strategy involves clinical trials centered around cancer immunity stimulation through T cells. However, preliminary findings from clinical trials are less promising, with low response rates of up to 20% in most trials.

About the study

In the present study, researchers evaluated the transcriptomic diversity of TPMs across advanced cancer types. Various types of solid tumors from 514 patients were analyzed for RNA expression of TPMs. The team obtained data on patients’ sex, age, microsatellite instability (MSI), PDL1 status, and tumor mutational burden (TMB). Tumor samples were subject to RNA extraction and sequencing.

RNA expression of 397 genes was measured, focusing on 15 TPMs – cluster of differentiation 27 (CD27), CD28, CD86, CD80, CD40 and its ligand (CD40LG), CD137, granzyme B (GZMB), inducible T cell co-stimulator (ICOS) and its ligand (ICOSLG), interferon-gamma (IFNG), T-box transcription factor 21 (TBX21), OX40 and its ligand (OX40LG), and glucocorticoid-induced TNFR-related (GITR).

The researchers extracted genomic DNA from tumors with > 20% neoplastic nuclei for MSI. The MSI next-generation sequencing (NGS) assay evaluated 29 homopolymer loci. For TMB, DNA was extracted from tumors with > 30% neoplastic nuclei, and DNA libraries were prepared, enriched, and sequenced. PDL1 was measured using three immunohistochemistry assays.

Patients’ baseline characteristics and TPM frequencies were summarized using descriptive statistics. Hierarchical clustering was performed using Ward’s method based on TPM expression patterns. The similarity of each sample’s TPM expression was visualized using principal component analysis, and the silhouette coefficient was estimated.

Findings

The median age of patients was 60.8, and most patients were females. Colorectal cancer was the most common cancer present, followed by pancreatic, breast, ovarian, and stomach cancers. The proportion of patients with low TPM expression ranged between 14.7% for ICOSLG and 50% for IFNG. The proportion of those with intermediate TPM expression ranged between 40.1% for IFNG and 55.6% for OX40.

The proportion of patients with high expression ranged between 9.9% for IFNG and 37.4% for ICOSLG. Distinct TPM expression patterns were identified in 502 patients. The researchers observed variance in TPM expression based on histological types, i.e., colorectal cancer patients more commonly showed high ICOSLG and GZMB expression. Nevertheless, there were no statistically significant differences in TPM expression by cancer type.

The heatmap of expression profiles of TPMs revealed no specific patterns by histological type. Patients with unstable microsatellite status (high MSI), TMB ≥ 10 mutations/mb, or PDL1 ≥ 1% showed significantly elevated expression of IFNG and GZMB than those with stable microsatellite status, TMB < 10 mutations/mb, or PDL1 < 1%, respectively. High PDL1 expression was also associated with significantly increased ICOS, GITR, and CD137 levels.

The optimal number of clusters was three, each with a characteristic TPM expression pattern. Cluster 1 (hot) comprised most TPMs with high expression, cluster 2 (cold) contained most TPMs with low expression, and cluster 3 (mixed) contained the remainder. These clusters were associated with PDL1 status. No significant differences were evident by TMB or cancer type.

Conclusions

In sum, the study revealed unique expression patterns of TPMs in 97.7% of cancer patients. The researchers observed increased expression of IFNG and GZMB in patients with high MSI, TMB, or PDL1 expression. The expression of GITR, ICOS, and CD137 was elevated in patients with high PDL1 expression.

Hierarchical clustering revealed three clusters significantly associated with PDL1 expression but not with the histological type of cancers. The authors posit that individualized patient selection based on TPM immunomic profiles may help to optimize the efficacy of immunotherapeutics targeting T cell priming.