Deep sequencing of multiple regions of glial tumors reveals spatial heterogeneity for mutations in clinically relevant genes

Abstract

Background: The extent of intratumoral mutational heterogeneity remains unclear in gliomas, the most common primary brain tumors, especially with respect to point mutation. To address this, we applied single molecule molecular inversion probes targeting 33 cancer genes to assay both point mutations and gene amplifications within spatially distinct regions of 14 glial tumors.

Results: We find evidence of regional mutational heterogeneity in multiple tumors, including mutations in TP53 and RB1 in an anaplastic oligodendroglioma and amplifications in PDGFRA and KIT in two glioblastomas (GBMs). Immunohistochemistry confirms heterogeneity of TP53 mutation and PDGFRA amplification. In all, 3 out of 14 glial tumors surveyed have evidence for heterogeneity for clinically relevant mutations.

Conclusions: Our results underscore the need to sample multiple regions in GBM and other glial tumors when devising personalized treatments based on genomic information, and furthermore demonstrate the importance of measuring both point mutation and copy number alteration while investigating genetic heterogeneity within cancer samples.

Figures

Figure 1

Experimental approach. (A) Each tumor was divided into three to five regions to assay intratumoral heterogeneity. Each individual region was subdivided into four pieces for use in next generation sequencing (NGS), histology, cell culture and xenotransplantation. (B) Molecular inversion probe method. Oligonucleotide probes were previously designed against 33 cancer genes [6]. MIPs have a common backbone sequence, molecular tag sequence as well as targeting arms homologous to regions flanking targets of interest. After polymerase extension and ligation, targeted sequence is captured within a circular molecule. Captured sequences are amplified in a barcoding-PCR reaction and multiple samples are pooled and sequenced on the same lane. After tag-correction (not shown), reads corresponding to each tumor region are mapped to the human reference sequence to be used to identify copy number amplifications and point mutations specific to one region or another. Additional details are provided in Figure S1 in Additional file 1. (C) Example of comparisons: MIP captures of regions C and D can detect both TP53 point mutation heterogeneity and EGFR amplification heterogeneity within a tumor. Tumors with mutational heterogeneity were required to share either a point mutation or copy number alteration (in this case mutation of PTEN) across all regions to ensure that differences in observed mutation were not due to varying levels of tumor cellularity.

Figure 2

Summary of heterogeneity observed across all samples. (A) Protein-altering mutations detected across all tumor regions. Genes mutated twice in the same tumor region are not identified here but can be found within a table of all mutations (Table S3 in Additional file 1). (B) High level gene amplifications detected by smMIP assay. Copy number was estimated by comparing all tumor samples against 12_X, a universal control from BI12 (see Figure S2 in Additional file 1 for analysis using patient matched controls). ‘Amplification’ indicates genes with coverage three-fold higher than median coverage across a sample. ‘High Amplification’ indicates genes with coverage six-fold higher than median coverage across a sample. Region X refers to brain tissue grossly uninvolved by tumor. Our approach would miss any low-level gene amplifications within these tumors.

Figure 3

Intratumoral heterogeneity ofTP53andRB1determined from smMIP sequencing. Tumor BI09 was sectioned into five regions (A to E). Brain tissue grossly uninvolved by tumor was used as a control (X). Each region was assayed for mutations in 33 genes, including TP53 and RB1. This plot shows the allele balance of TP53 and RB1 mutations within each tumor region. Regions A and B have a high allele fraction mutation in TP53, while regions D and E have a high allele fraction mutation in RB1. Sanger results validated TP53 and RB1 mutations in each region and also revealed that all regions shared a R132H mutation in IDH1 (Figure S7 in Additional file 1).

Figure 4

Heterogeneity ofPDGFRAamplification in BI05. (A) Copy number estimates based on smMIP probe data. PDGFRA amplification (labeled) occurs in regions A and B with no amplification in regions C, D or E. (B) Results from Taqman qPCR targeting both PDGFRA and EGFR performed in quadruplicate. PDGFRA amplification occurs in regions A and B (between four- and eight-fold amplification) with no significant amplification in regions C, D and E. EGFR amplification occurs in all regions of BI05, consistent with MIP sequencing results. Heterogeneity of PDGFRA amplification was also confirmed through immunohistochemistry of regions A and E (Figure S9 in Additional file 1). Error bars represent the mean +/- one standard deviation from quadruplicate values.

Figure 5

Heterogeneity ofPDGFRAamplification in BI06. (A) Copy number estimates based on smMIP probe data. PDGFRA amplification (labeled) occurs in region A with only mild amplification in region B and no clear detectable amplification in regions C, D or E. (B) Results from Taqman qPCR targeting PDGFRA performed in quadruplicate. Region X refers to a region of brain tissue grossly uninvolved by tumor. PDGFRA amplification occurs in region A (approximately four-fold amplification) with only mild amplification in regions B, C, D and E. Error bars represent the mean +/- one standard deviation from quadruplicate values.