Genomic and tumour microenvironmental biomarkers of immune checkpoint inhibitor response in advanced Taiwanese melanoma

Abstract

Objective: Genomic biomarkers predicting immune checkpoint inhibitor (ICI) treatment outcomes for Asian metastatic melanoma have been rarely reported. This study presents data on next-generation sequencing (NGS) and tumour microenvironment biomarkers in 33 cases.

Methods: Thirty-three patients with advanced melanoma, who underwent ICI treatment at the Chang Gung Memorial Hospital in Taiwan, were recruited. The study evaluated clinical outcomes, including response rate, disease control rate, progression-free survival (PFS) rate and overall survival (OS) rate. Archived tissue samples from 33 cases were subjected to NGS by ACTOnco, and ACTTME was employed in 25 cases.

Results: The most prevalent driver mutations were BRAF mutations (24.2%), followed by NRAS (15.2%), KIT (12.1%), KRAS (9.1%) and NF1 (9.1%) mutations. Acral/mucosal melanomas exhibited distinct mutation patterns compared to non-acral melanomas. Tumour mutational burden estimated using ACTOnco was not associated with ICI efficacy. Notably, genetic alterations in the p53 pathway (CDKNA2 loss, MDM2 gain/amplification and TP53 mutation) accounted for 36.4% and were significantly associated with unfavourable PFS (median PFS 2.7 months vs. 3.9 months, P = 0.0394). Moreover, 26 genes were identified as differentially expressed genes that were upregulated in patients with clinical benefits compared to those without benefits. Four genes, GZMH, GZMK, AIM2 and CTLA4, were found to be associated with both PFS and OS.

Conclusion: Genetic alterations in the p53 pathway may be critical in Asian patients with melanoma undergoing ICI treatment. Further investigation is required to explore this mechanism and validate these findings.

Keywords: ICIs; genomic and TME biomarkers; melanoma; p53.

Figure 1

Flowchart, progression‐free survival (PFS) and overall survival (OS) in this study. (a) Flowchart for sample selection. (b) Progression‐free survival (PFS) and (c) overall survival (OS) of 33 patients with melanoma undergoing immune checkpoint inhibitor (ICI) treatment and gene testing, using ACTOnco, a gene panel of 440 single‐nucleotide variations (SNV) and copy number variations (CNV). The median PFS and OS were 3.3 and 11.1 months, respectively.

Figure 2

Summary of genomic alterations, clinicopathological features and clinical response to immuno‐oncology (IO) in 33 patients with advanced melanoma. (a) SNV and small insertions and deletions; and (b) CNV of important oncogenic drivers detected in each sample. Data are grouped by melanoma subtypes (acral/mucosal, non‐acral cutaneous and unknown primary melanoma). Hotspot gene alterations are coloured according to mutation type. Copy number changes: homozygous deletion (dark blue), amplification (red) and gain (pink). The frequencies of mutations detected in the whole cohort are shown on the left. Stars indicate driver mutations.

Figure 3

The association between tumour mutational burden (TMB) and treatment outcomes. (a) No significant difference in TMB was noted between responders and non‐responders (P = 0.0663). Responders were classified as tumours with responses. (b) No significant difference in TMB was noted between patients with and without clinical benefit (P = 0.7215). Clinical benefit indicates a tumour with a response and stable disease. No significant difference was noted in (c) PFS and (d) OS between patients with melanoma harbouring higher or lower than median TMB (1.9 mutations per Mb).

Figure 4

The association between genetic alterations in the CDKN2A/MDM2/TP53 axis and treatment outcomes. (a) The genetic landscape of the p53 pathway (CDKN2A/MDM2/TP53 axis). (b) The PFS and (c) OS between patients with melanoma harbouring p53 pathway alterations (CDKN2A loss, MDM2 gain, TP53 mutation). (b) The melanomas harbouring p53 pathway alterations had worse PFS than those without p53 pathway alterations (median PFS 2.7 months vs. 3.9 months, P = 0.0394).

Figure 5

Gene expression in patients with or without clinical benefit. (a) Heatmap of immune‐associated genes and clinical benefit among patients undergoing immunotherapy. (b) A volcano plot showing the significance of immune association genes among patients with or without clinical benefit. (c) PFS and (d) OS of patients with melanoma based on the four‐gene signature.