Proteogenomic analysis of melanoma brain metastases from distinct anatomical sites identifies pathways of metastatic progression

Erin M. Taylor,Alan J. Tackett,Brittany G. Griffin,Murat Gokden,Stephanie D. Byrum,Christopher P. Wardell,Analiz Rodriguez,Issam Makhoul,Jacob L. Edmondson,Sara C. Shalin

doi:10.1186/s40478-020-01029-x

Erin M. Taylor, Alan J. Tackett + Show 8 more

Open Access

https://doi.org/10.1186/s40478-020-01029-x

Copy DOI

Abstract

Melanoma brain metastases (MBM) portend a grim prognosis and can occur in up to 40% of melanoma patients. Genomic characterization of brain metastases has been previously carried out to identify potential mutational drivers. However, to date a comprehensive multi-omics approach has yet to be used to analyze brain metastases. In this case report, we present an unbiased proteogenomics analyses of a patient’s primary skin cancer and three brain metastases from distinct anatomic locations. We performed molecular profiling comprised of a targeted DNA panel and full transcriptome as well as proteomics using mass spectrometry. Phylogeny demonstrated that all MBMs shared a SMARCA4 mutation and deletion of 12q. Proteogenomics identified multiple pathways upregulated in the MBMs compared to the primary tumor. The protein, PIK3CG, was present in many of these pathways and had increased gene expression in metastatic melanoma tissue from the cancer genome atlas data. Proteomics demonstrated PIK3CG levels were significantly increased in all 3 MBMs and this finding was further validated by immunohistochemistry. In summary, this case report highlights the potential role of proteogenomics in identifying pathways involved in metastatic tumor progression. Furthermore, our multi-omics approach can be considered to aid in precision oncology efforts and provide avenues for therapeutic innovation.

Highlights

In various cancer patient cohorts, the development of brain metastases (BM) is present in 24–45% of patients
In summary, we used a multi-omics approach to characterize multiple melanoma brain metastases (MBM) in a treatment naïve patient and were able to identify PIK3CG as a protein potentially correlated to metastatic progression
Our study is novel as an unbiased proteogenomic approach has yet to be applied to MBM samples matched to the primary tumor

Summary

Introduction

In various cancer patient cohorts, the development of brain metastases (BM) is present in 24–45% of patients. Proteins are often the targets of cancer based therapies making proteogenomics integral for the identification of therapeutic targets [16] In this case report, we use a multi-omics approach (i.e. genomics, transcriptomics, and proteomics) to study three anatomically distinct MBM in a treatment naïve patient. Hierarchial clustering of significantly expressed gene transcripts and proteins demonstrated Phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit γ (PIK3CG) was identified in multiple upregulated pathways for the metastatic brain tumors and gene expression was significantly increased in metastatic melanoma samples in comparison to primary melanoma from TCGA data (Fig. 3c).

Conclusion and discussion

Findings

Methods