Abstract 5602: Regulation of immune checkpoint genes revealed by a melanoma tumor cancer genome atlas (TCGA) analysis - potential implications for improving immunotherapy

Abstract

Abstract Introduction: The interface between T lymphocytes and cancer or antigen presenting cells (C/APCs) is multi-faceted and complex. This interface, now designated 'the immunological synapse', comprises of both co-inhibitory and co-stimulatory transmembrane protein pairs ('checkpoint proteins') that all serve to modulate the signal transmitted to the T lymphocyte, leading to either activation, anergy or exhaustion. Immune checkpoint inhibitors have anti-neoplastic activity in a wide range of malignancies, but not all cancers and not all patients with a given cancer respond to the currently available checkpoint inhibitors. Micro-RNAs (miRNAs) are short intracellular RNA molecules known to be master regulators of gene expression. Our aim was to study the associations between a miRNA that was previously implicated in cancer and immune checkpoint genes. Methods: Bioinformatic analyses of the expression of mRNAs and miRNAs from 451 samples was performed using the melanoma TCGA database. Correlation coefficients between the expression of mRNAs or mRNAs/miRNAs were calculated using the Spearman rho method. Survival analysis was performed using the Kaplan-Meier method. Potential 3'UTR binding sites of miRNAs were found using the web-based tool www.targetscan.org. Results: Of 22 mRNAs of checkpoint genes assessed, the expression of 19 was highly positively correlated to each other. These mRNAs code for both co-inhibitory and co-stimulatory proteins at the T cell and the C/APC sides of the immunological synapse, suggesting that there is joint transcriptional regulation on the expression of these checkpoint genes. The expression of one specific miRNA was also significantly positively correlated with the expression of eight of these checkpoint mRNAs (PD1L, PD1L, B7.1, ICOS, BTLA, LAG3, CTLA4, TNFRSF9). The significant positive correlation between this miRNA and B7.1 and PD1L was verified in 18 cell lines in vitro, with rho correlation coefficients of 0.72 and 0.61, respectively (p=0.001 and 0.01, respectively). This possibly indicates a joint transcriptional regulation on this miRNA and the mRNAs. Bioinformatic analyses suggest that this miRNA may potentially target the 3'UTR of 6 of these mRNAs (PD1, PD1L, B7.1, ICOS, BTLA, TNFRSF9). Data from 163 stage III melanoma patients with documented survival data show that a high expression level of this miRNA and a low expression level of either B7.1, B7.2, CD28, ICOS, ICOSL, OX40, OX40L, TNFRSF18 or PD1L was each significantly associated with poor prognosis relative to all other expression groups. Eight of these nine mRNAs are co-stimulatory. Conclusions: It is tempting to speculate that aberrant co-regulation of the miRNA-mRNAs, leading to high levels of the miRNA and low levels of co-stimulatory checkpoint genes is associated with worse outcome, potentially as a result of 'immune evasion' due to decreased co-stimulation at the synapse. Citation Format: Raya Leibowitz-Amit, Jason Roszik, Dror Avni, Elizabeth Grimm. Regulation of immune checkpoint genes revealed by a melanoma tumor cancer genome atlas (TCGA) analysis - potential implications for improving immunotherapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 5602. doi:10.1158/1538-7445.AM2017-5602