Genomic analysis of metastatic cutaneous squamous cell carcinoma

Abstract

Purpose: A rare 5% of cutaneous squamous cell carcinomas (cSCC) metastasize, lack FDA-approved therapies, and carry a poor prognosis. Our aim was to identify recurrent genomic alterations in this little-studied population of metastatic cSCCs.

Experimental design: We performed targeted sequencing of 504 cancer-associated genes on lymph node metastases in 29 patients with cSCC and identified mutations and somatic copy-number alterations associated with metastatic cSCC. We determined significantly mutated, deleted, and amplified genes and associated genomic alterations with clinical variables.

Results: The cSCC genome is heterogeneous with widely varying numbers of genomic alterations and does not appear to be associated with human papillomavirus. We found previously identified recurrently altered genes (TP53, CDKN2A, NOTCH1/2) but also a wide spectrum of oncogenic mutations affecting RAS/RTK/PI3K, squamous differentiation, cell cycle, and chromatin remodeling pathway genes. Specific mutations in known oncogenic drivers and pathways were correlated with inferior patient outcomes. Our results suggest potential therapeutic targets in metastatic cSCC, including PIK3CA, FGFR3, BRAF, and EGFR, similar to those reported in SCCs of the lung and head and neck, suggesting that clinical trials could be developed to accrue patients with SCCs from multiple sites of origin.

Conclusions: We have genomically characterized a rare cohort of 29 metastatic cSCCs and identified a diverse array of oncogenic alterations that can guide future studies of this disease.

Conflict of interest statement

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed by the other authors.

©2015 American Association for Cancer Research.

Figures

Figure 1. Integrated view of selected recurrently altered genes

1A - Genomic overview of sequencing and variant calling. The top three plots show shows the distribution of CNA types across the samples, distribution of SNV types across the samples, and the average coverage of tumor samples and their matched normal samples where available. 1B - Heatmap representation of selected recurrently altered genes. CNAs are colored in red for high-level amplification events and green for homozygous deletion events. For simplicity, low-level CNA events are not shown. SNVs are colored by type in purple, beige, or blue, and also labeled: I for insertion or deletion (indel), S for missense, C for COSMIC, * for truncating, and O for other types of nonsynonymous mutation (splice site, non-stop). Significantly mutated genes as determined by Mutsig CV are those whose q-scores pass threshold of 0.1 (or –logQ-value greater than 1) on the left-hand plot. The genes are listed in order of decreasing number of alterations across the samples, as shown on the right-hand plot.

Figure 2. Overview of copy number alterations

GISTIC plot showing the most recurrently gained (red) and lost (blue) loci in metastatic cSCCs. Peaks are considered significant if they pass the q-value threshold of 0.25. The majority of peaks contain only one gene, as we determined CNAs using 504 cancer-associated genes.

Figure 3. Recurrently altered pathways in metastatic cSCC

Pathway diagrams depicting the percentage of samples with alterations in 3A - RAS/RAF/MEK/ERK and PI3K/AKT signaling 3B - cell cycle, and 3C - squamous cell differentiation Alterations are classified as activating (high-level amplification or known activating mutation colored red), inactivating (homozygous loss or truncating mutation colored blue), or potentially cancer associated (COSMIC mutation colored white). For each pathway, we show integrated heatmaps (similar to Fig 1B) to show the detailed alteration pattern of each gene; however, we now also include light red and light green to represent low-level CNAs. Note that each heatmap is sorted independently across the samples, to best illustrate the pattern of mutations, such as mutual exclusivity or concurrence.

Figure 4. Alterations in RTK/RAS/PI3K pathway and chromatin remodeling genes associated with progression-free survival (PFS) in metastatic cSCC

Kaplan-Meier survival curves of metastatic cSCC patients comparing patients with or without mutation in A) RTK/RAS/PI3K pathway, B) chromatin remodeling genes, or C) both.