Clinically Actionable Insights into Initial and Matched Recurrent Glioblastomas to Inform Novel Treatment Approaches.

Glioblastoma is the most common primary adult brain tumour, and despite optimal treatment, the median survival is 12-15 months. Patients with matched recurrent glioblastomas were investigated to try to find actionable mutations. Tumours were profiled using a validated DNA-based gene panel. Copy number variations (CNVs) and single nucleotide variants (SNVs) were examined, and potentially pathogenic variants and clinically actionable mutations were identified. The results revealed that glioblastomas were IDH-wildtype (IDH WT; n = 38) and IDH-mutant (IDH MUT; n = 3). SNVs in TSC2, MSH6, TP53, CREBBP, and IDH1 were variants of unknown significance (VUS) that were predicted to be pathogenic in both subtypes. IDH WT tumours had SNVs that impacted RTK/Ras/PI(3)K, p53, WNT, SHH, NOTCH, Rb, and G-protein pathways. Many tumours had BRCA1/2 (18%) variants, including confirmed somatic mutations in haemangioblastoma. IDH WT recurrent tumours had fewer pathways impacted (RTK/Ras/PI(3)K, p53, WNT, and G-protein) and CNV gains (BRCA2, GNAS, and EGFR) and losses (TERT and SMARCA4). IDH MUT tumours had SNVs that impacted RTK/Ras/PI(3)K, p53, and WNT pathways. VUS in KLK1 was possibly pathogenic in IDH MUT. Recurrent tumours also had fewer pathways (p53, WNT, and G-protein) impacted by genetic alterations. Public datasets (TCGA and GDC) confirmed the clinical significance of findings in both subtypes. Overall in this cohort, potentially actionable variation was most often identified in EGFR, PTEN, BRCA1/2, and ATM. This study underlines the need for detailed molecular profiling to identify individual GBM patients who may be eligible for novel treatment approaches. This information is also crucial for patient recruitment to clinical trials.

1. Introduction

Gliomas are the largest group of intrinsic brain tumours with age adjusted incidence rates ranging from 4.67 to 5.73 per 100,000, causing more years of life lost compared with other cancers [1, 2]. Glioblastoma (GBM) is the most malignant glioma and is classified molecularly as IDH-wildtype and IDH-mutant GBM [3-10]. During gliomagenesis, an array of genetic alterations may cause the dysregulation of cell growth signalling and cell cycle pathways [6, 11–15]. In particular, mutations in RTKs (receptor tyrosine kinases) and/or loss of PTEN (phosphatase and tensin homolog) alter the PI3K (phospinositide 3-kinase)/AKT cell growth pathway [11]. Further mutations in CDKN2A or CDK4 (cyclin-dependent kinase) lead to uncontrolled progression of the cell cycle, as do mutations in TP53 [16]. Neural stem cells in the subventricular zone may harbour recurrent driver somatic mutations that are shared with the tumour bulk (e.g., P53, PTEN, EGFR, and TERT) [17]. Telomerase (reactivation or reexpression) can occur in IDH wildtype and mutant GBMs driven either by telomerase reverse transcriptase (TERT) promoter mutations or other mechanisms [8, 18]. The current standard-of-care for glioblastomas remains as maximal safe surgical resection with concurrent radiotherapy and temozolomide (TMZ) chemotherapy (Stupp protocol) [19, 20]. Personalised therapies remain promising although trials have been unsuccessful to date [21-23]. For example, dysregulated PI3K and RTKs (EGFR, MET, PDGFR, FGFR, and BRAF) genes have been targeted with various small molecules, antibodies, and inhibitors [24-29]. To date, entry to clinical trials for GBM has not been based on a detailed molecular analysis of an individual patient's tumour using high throughput sequencing (HTS). HTS-based molecular diagnostics can aid the detection of genetic alterations, information required for personalised medicine [30, 31]. Herein, initial and matched recurrent glioblastomas were examined using HTS with a validated DNA-based diagnostic panel. Potentially pathogenic variants and clinically actionable mutations were identified in different GBM subtypes. Findings were validated using TCGA-GBM and GDC datasets.

2. Materials and Methods

2.1. Clinical Specimens

Ethical approval was given by Brain Tumour Bank South West and Brain UK (Ref: 14/010). All patients had been treated using the Stupp protocol [19]. A total of 72 formalin-fixed paraffin-embedded (FFPE) samples from 54 patients were identified (2009–2014). Only FFPE slides with >30% tumour cells available for macrodissection were selected. Samples lacking cellularity or excessively necrotic were excluded. Following quality control, 67 samples for 46 patients and 19 with matched recurrent samples available were identified. Of these, a total of 49 samples were successfully sequenced for 41 patients (21 males; 20 females; mean age 55 years, range 16–78 years; see Tables 1 and ). Matched initial and recurrent tissue samples were analysed for 8 patients (2 males; 6 females). Recurrent tumours all occurred locally to the initial tumour. Anonymised patient cases in the GBM cohort were numbered 1–11, 16–41, and 43–46, and “a” and “b” indicated initial and recurrent tumour samples, respectively ().

Table 1

Summary of the clinical data for patients genomically profiled in this study (n = 41). Patients with IDH-wildtype and IDH-mutant glioblastoma tumours were identified from the BRASH clinical database between 2009 and 2014.

Characteristic	IDH-wildtype	IDH-mutant
Number of patients	38	3
Age
Mean	54	42
Median (range)	52 (16–78)	50 (19–58)
Gender
Male	19 (50%)	2 (66%)
Female	20 (50%)	1 (33%)
Survival range (months)	2–48	5–12
Tumour location
Temporal lobe	8 (21%)	2 (66%)
Frontal lobe	15 (39%)
Parietal lobe	4 (11%)
Occipital	4 (11%)
More than one lobe	5 (13%)	1 (33%)
Multifocal	1 (3%)
No data	1 (3%)
Tumour recurrence
Initial	38	3
Recurrent	7	1

2.2. HTS Neuro-Oncology Gene Panel

A published HTS DNA-based panel that uses targeted enrichment to examine exonic, selected intronic and promoter regions of 130 clinically relevant neuro-oncology genes was utilised (see ) [30]. The diagnostic panel has been optimised for use either with fresh-frozen or FFPE tissue. Validation studies of the HTS panel analysing ∼200 single nucleotide variants (SNVs), gene fusions, and copy number variants (CNVs) showed 98% concordance with single marker tests [30]. Using the HTS panel, genetic alterations in tumours were characterized, and TERT promoter and IDH1/2 status confirmed.

2.3. DNA Extraction, HTS Library Preparation, Sequencing, and Analysis

Slides were deparaffinised and rehydrated using xylene and ethanol and left to dry. Tissue sections were then microdissected and placed into 180 uL ATL buffer. DNA was extracted from tissue sections (10 × 10 μm) according to manufacturer's instructions using the QIAamp DNA FFPE Tissue Kit (Qiagen, Manchester, UK). Following assessment of DNA quality and quantity, libraries were prepared using 200 ng of genomic DNA with an optical density 260/280 ratio between 1.8 and 2.0. Libraries were constructed using the SureSelectXT Target Enrichment System for Illumina Paired-End Multiplexed Sequencing Library protocol (Agilent). PCR master mixes were prepared using the SureSelectXT Library Prep Kit ILM following manufacturer's guidelines. In accordance with Illumina guidelines, libraries with a concentration of 4 nM were diluted to 20 pM, denatured, and sequenced on a NextSeq 500 (Illumina). HTS data were analysed following the pipeline described by Sahm et al. [30]. In brief, raw reads were demultiplexed, converted to fastq, quality checked, and manually trimmed when necessary. Paired-end reads were aligned to the human genome (version GRch37; hg19), and duplicate sequences were removed.

2.4. CNV Analysis in the GBM Cohort

CNVs were investigated using a coverage analysis. The ratio of on- and off-target reads, coverage per target region, and mean coverage per sample were estimated using the R package TEQC [32]. Measures provided an estimate of read depth, as the number of reconstructed strands across a region of interest, and this was utilised for CNV estimation of genes. Data normalisation and CNV comparison to a reference control were made using the R package seqCNA [33]. This method has previously been validated with 100% concordance for 47 GBM cases using 450 k data [30]. Potential CNV gain or loss is indicated by deviations from a proportional read depth of 50%, considered a normal gene copy number.

2.5. SNV Analysis in the GBM Cohort

Variant calling followed a modified pipeline, as described by Sahm et al. [30]. In brief, variants were called using SAMtools mpileup [34]. Variant calls were then filtered by (a) read depth ≥ 40, (b) genotype quality ≥ 99, (c) minimum allele frequency set at 10, and (d) at least 10% read coverage from each strand using the R package VariantAnnotation [35]. TERT promoter position calls were not filtered due to their low detection rate because of difficulties with their amplification as a GC-rich region [30]. Nonsynonymous filtered variants were annotated with the most up to date information including dbSNP and COSMIC identifiers using the online tool wANNOVAR [36]. Matched normal tissue was unavailable for comparison for the identification of germline mutations. Thus, to try to discern pathogenic from benign variants, the frequency of a variant in the general population was used as a key criterion in their clinical interpretation to try to exclude germline mutations. SNVs were filtered to those with a frequency of ≥0.01 in the 1,000 Genomes database and ≥0.05 in the Genome Aggregation Database (gnomAD), previously known as the Exome Aggregation Consortium database. gnomAD warehouses whole genome sequences from 15,496 unrelated individuals [37]. As the ethnicity of patients in the GBM cohort was unknown, SNV frequencies were compared to overall frequencies (rather than regional) of both databases. Filtered SNVs impacting genes were categorised into biological pathways using GeneCards [38]. SNVs occurring in the potentially clinically actionable genes: EGFR, PTEN, CDKN2A, RB1, TP53, ATM, ATR, MSH6, PDGFRA, PIK3CA, PIK3R1, SMO, PTCH1, BRCA1, BRCA2, and BRAF, were quantified in the initial and matched recurrent tumours. Further filtering was applied to SNV results to try to identify variants of unknown significance (VUS) that are possibly pathogenic and underpin gliomagenesis. VUS considered to be possibly pathogenic, were those that had no frequency recorded in the 1,000 Genomes database, and were predicted to be damaging by both LJB SIFT and FATHMM-MKL software [39]. All genomic positions listed for SNVs identified by this study are from the human genome version GRch37.

2.6. VUS and CNV Analysis in the TCGA-GBM and GDC Datasets

VUS identified as possibly pathogenic mutations in the GBM cohort were further investigated for supporting evidence of their clinical significance using TCGA-GBM and GDC datasets. Frequencies of cases with mutations in genes were investigated in the GDC data portal. Abundance of mutations and copy number alterations within the TCGA-GBM dataset was visualised as an oncoprint plot generated using GlioVis, a data visualisation tool for brain tumour datasets [40].

2.7. Survival Analyses of IDH-Wildtype Glioblastomas

A Cox proportional hazard regression analysis was implemented to determine the relationship between the total number of SNVs (median split) and overall survival. MGMT methylated and unmethylated GBMs were investigated separately. Survival analyses and plotting of results as Kaplan–Meier graphs were carried out using R software [41]. Of the 41 patients, univariate survival analysis was carried out on the 33 IDH-wildtype patients only. Omitted patients included the three IDHMUT patients and a further five patients lacking survival information.

3. Results

3.1. Overview of Genomic Profiling of Glioblastoma Tumours and IDH Status

In all, 49 samples from 41 patients including 8 matched samples were genomically profiled (Tables 1 and ). Results could not be obtained for 5 initial and 13 recurrent samples from 11 patients, giving a sequencing failure rate of ∼22%. SNVs were not identified in 5 samples (9%). Recurrent tumour samples were necrotic with low cellularity, which probably impacted DNA quality and sequencing success. Majority of tumours were IDH-wildtype (38/41; 93%) with the exception of three cases (8, 35, and 39) that were IDH-mutant (). Cases 8, 35, and 39 had a C to T mutation located at the IDH1 diagnostic hotspot R132 (Chr2: 209113112; GRCh37). Only one other case (6a) had an IDH1 mutation located at Chr2: 209108284 (GRCh37). This mutation was 4,828 bp upstream of the diagnostic hotspot (R132); hence, case 6a was considered IDH-wildtype. One case had an IDH2 mutation (Chr15: 90627553); however, this did not coincide with known somatic mutations located at 15q26.1 codons R140 (Chr15: 90631934) and R172 (Chr15: 90631837). TERT mutations were observed in IDH wildtype initial (Chr5: 1254594; Chr5: 1294166) and recurrent tumours (Chr5: 1,254,594); however, none coincided with known somatic mutations in promoter regions at the C228 (Chr5: 1,295,228) and C250 loci (Chr5: 1,295,250; hg19).

3.2. SNVs Detected in Initial and Recurrent IDHWT Glioblastomas

A total of 134 nonsynonymous and three stop-gain SNVs were detected from initial (n = 125; ) and recurrent IDHWT tumours (n = 12; ). Including IDH1/2 mutations, SNVs affected 52 genes across nine biological pathways during the different phases of gliomagenesis (Figures 1 and 2; Tables 2 and 3). Majority of initial tumours had SNVs in a gene in the RTK/Ras/PI(3)K pathways (79%; 30/38) followed by the p53 DNA damage repair pathway (61%; 23/38). Two stop-gain SNVs were identified from the p53 genes MSH2 (Chr2: 47705428; rs63751155) and TP53 (Chr17: 7579315; COSM326717; COSM3388232; COSM326718; COSM3388233; COSM326716) in initial tumours; both variants were predicted to be pathogenic by FATHMM-MKL (). A large proportion of initial IDHWT tumours had SNVs in the p53 pathway genes BRCA1 (18%; 7/38) and BRCA2 (18%; 7/38; Table 4). Six BRCA1 variants were detected including a confirmed somatic mutation in adenocarcinoma (COSM6612515; Chr17: 41244952) [42]. Six BRCA2 variants were detected including confirmed somatic mutations in haemangioblastoma (COSM3753648, Chr13: 32914236; COSM5019704, Chr13: 32953549) [43]. Over half of initial IDHWT tumours had an SNV in a WNT signalling pathway gene (58%; 22/38). Multiple variants (n) were detected for the WNT genes KMT2D/MLL2 (7), CREBBP (4), DICER1 (3), APC (3), TERT (2), and KLF4 (2). IDHWT tumours also showed variation in SHH (16%; 6/38) and NOTCH (8%; 3/38) pathways. A small proportion of initial tumours had SNVs in the G-protein gene, GNAS (5%; 2/38), IDH1/2 (5%; 2/38), and the Rb-specific cell-cycle regulation genes CDK6 and RB1 (5%; 2/38). The RB1 variant was a stop-gain SNV (Chr13: 48953735), but it was not pathogenic. Among IDHWT tumours, 40 SNVs in 21 genes were VUS that were predicted to be functionally damaging (Tables 3 and ). Potentially pathogenic VUS impacted IDH1 and genes in the p53 (ATM, BRCA1, CHEK2, MSH6, PPM1D, and TP53), RTK/Ras/PI(3)K (BRAF, DAXX, EGFR, FGFR2, JAK2, MYB, PIK3CA, PIK3R1, TSC2, and PTEN), SHH (PTCH1 and SMO), and WNT pathways (CREBBP). Two-thirds of initial IDHWT tumours (63%; 24/38) harboured potentially actionable variation most frequently in PTEN (29%; 11/38), followed by BRCA1 (18%; 7/38), BRCA2 (18%; 7/38), TP53 (18%; 7/38), EGFR (16%; 6/38), ATM (16%; 6/38), and ATR (8%; 3/38; see Table 4). Recurrent IDHWT tumours had SNVs in genes in the RTK/Ras/PI(3)K (43%; 3/7), WNT signalling (57%; 4/7), and p53 pathways (29%) in the genes BRCA1 (14%; 1/7) and BRCA2 (14%; 1/7) and GNAS (14%; 1/7). IDHWT recurrent tumours were not mutated in NOTCH, SHH, Rb, or IDH genes (Figure 2 and ). In the matched initial tumour, 16 genes showed variation, four of which were also mutated in the recurrent tumour. An additional three SNVs were recorded only in the recurrent tumour in CSF1R, ATM, and BRCA1. Possibly pathogenic VUS were identified in PTEN in recurrent IDHWT tumours. Almost half of recurrent IDHWT tumours (43%; 3/7) harboured at least one potentially actionable variation in the genes EGFR (14%; 1/7), PTEN (14%; 1/7), BRCA1 (14%; 1/7), BRCA2 (14%; 1/7), and ATM (14%; 1/7; Figure 2 and Table 4).

Figure 1

Summary of the genes identified with SNVs in IDH (n = 38) and IDH diffuse tumours (n = 3; cases 8a, 35a, and 39a). Genes are arranged hierarchically within their pathways for the RTK/Ras/PI(3)K (red), IDH (yellow), NOTCH, SHH, and WNT signalling (variations of green), p53 (blue), Rb (purple), and G-proteins (dark blue) pathways. Numbers across the top axis denote the patient identifier.

Figure 2

Summary of the genes identified with SNVs in matched initial and recurrent IDH (n = 7) and IDH diffuse tumours (n = 1; case 8). Genes are arranged hierarchically within their pathways for the RTK/Ras/PI(3)K (red), IDH (yellow), NOTCH, SHH, and WNT signalling (variations of green), p53 (blue), Rb (purple), and G-proteins (dark blue) pathways. Numbers across the top axis denote the patient identifier; “a” and “b” indicate initial and recurrent tumours, respectively.

Table 2

Summary of the number and proportion of IDH-wildtype and IDH-mutant glioblastoma patients with SNVs in genes in the RTK/Ras/PI(3)K, p53 DNA damage repair, WNT signalling, SHH, NOTCH, Rb, and G-protein pathways.

Pathway	IDH-wildtype				IDH-mutant
	Initial		Recurrent		Initial		Recurrent
	%	N	%	N	%	N	%	N
RTK/Ras/PI(3)K	79	30/38	43	3/7	66	2/3	0	0/1
p53 DNA damage repair	61	23/38	29	2/7	100	3/3	100	1/1
WNT signalling	58	22/38	57	4/7	33	1/3	100	1/1
SHH	16	6/38	0	0/7	0	0/3	0	0/1
NOTCH	8	3/38	0	0/7	0	0/3	0	0/1
Rb	5	2/38	0	0/7	0	0/3	0	0/1
G-protein	5	2/38	14	1/7	0	0/3	100	1/1

Table 3

Comparison of genes with SNVs identified in IDH-wildtype and IDH-mutant initial and recurrent tumours in the GBM cohort with those outlined by Barthel et al. [8], described for the five phases of gliomagenesis.

Gliomagenesis phases	Pathway	Common tumour genetic alterations (Barthel et al.)	IDH wildtype				IDH-mutant
			Barthel et al.	GB-initial	GB-recurrent	GB-potentially pathogenic VUS	Barthel et al.	GB-initial	GB-recurrent	GB-potentially pathogenic VUS	Diagnostic panel (Y/N)
I: initial growth	IDH		—	IDH1	—	Y	IDH1	IDH1	—	Y	Y
	IDH		—	IDH2	—		IDH2	—	—		Y
	Rb		CDK6	CDK6	—		—	—	—		Y
	RTK/Ras/PI(3)K		EGFR	EGFR	EGFR	Y	—	—	—		Y
	RTK/Ras/PI(3)K		MET	MET	—		—	—	—		Y
	RTK/Ras/PI(3)K		PDGFRA	PDGFRA	—		PDGFRA	—	—		Y
	RTK/Ras/PI(3)K		PIK3CA	PIK3CA	—	Y	—	—	—		Y
	RTK/Ras/PI(3)K		PIK3R1	PIK3R1	—	Y	—	—	—		Y
	RTK/Ras/PI(3)K		PTEN	PTEN	PTEN	Y	—	—	—		Y
	WNT		TERT	TERT	TERT		—	—	—		Y
			NF1	—	—		—	—	—		Y
							CTCF				N
			TET1								N
II: oncogene-induced senescence	p53	TP53	TP53	TP53	—	Y	TP53	TP53	—	Y	Y
	p53	CDKN2A	CDKN2A	CDKN2A	—		—	—	—		Y
	p53		PPM1D	PPM1D	—	Y	—	—	—		Y
	Rb	RB1		RB1	—		—	—	—		Y
	RTK/Ras/PI(3)K	BRAF		BRAF	—	Y	—	—	—		Y
		CDKN2B	CDKN2B	—	—		—	—	—		Y
			ACVR1	—	—		—	—	—		Y
III: stressed growth	p53	ATM		ATM	ATM	Y	—	—			Y
	p53	ATR		ATR	—	Y	—	—	—		Y
		MYC		—	—		—	—	—		Y
		CDK4		—	—		—	—	—		Y
		MDM2		—	—		—	—	—		Y
IV: replicative senescence/crisis		CHD5									N
		TREX1									N
							Terra				N
		RB1	—	RB1	—		—	—	—		Y
	WNT	TERT	TERT	TERT	TERT		—	—	—		Y
	p53	TP53	TP53	TP53	—	Y	TP53	TP53	—	Y	Y
			ATRX	-	—		ATRX	—	—		Y
			—	DAXX	—	Y	DAXX	—	—		Y
V: immortalisation and dedifferentiation		OLIG2									N
		SOX2									N
GB-SNVs	G-proteins		—	GNAS	GNAS		—	—	GNAS		Y
	NOTCH		—	NOTCH1	—		—	—	—		Y
	NOTCH		—	NOTCH2	—		—	—	—		Y
	p53		—	BRCA1	BRCA1	Y	—	—	—		Y
	p53		—	BRCA2	BRCA2		—	BRCA2	—		Y
	p53		—	BRPF3	—		—	—	—		Y
	p53		—	MDM4	—		—	—	—		Y
	p53		—	MSH2	—		—	MSH2	—		Y
	p53		—	MSH6	—	Y	—	MSH6	—	Y	Y
	p53		—	RAD50	—		—	—	—		Y
	RTK/Ras/PI(3)K		—	ALK	—		—	—	—		Y
	RTK/Ras/PI(3)K		—	CDH1	—		—	CDH1	—		Y
	RTK/Ras/PI(3)K		—	CSF1R	CSF1R		—	CSF1R	—		Y
	RTK/Ras/PI(3)K		—	FGFR2	—	Y	—	—	—		Y
	RTK/Ras/PI(3)K		—	FGFR3	—		—	—	—		Y
	RTK/Ras/PI(3)K		—	FGFR4	—		—	—	—		Y
	RTK/Ras/PI(3)K		—	FOXO3	—		—	—	—		Y
	RTK/Ras/PI(3)K		—	JAK2	—	Y	—	—	—		Y
	RTK/Ras/PI(3)K		—	KDR	KDR		—	—	—		Y
	RTK/Ras/PI(3)K		—	KLK1	—		—	KLK1	—	Y	Y
	RTK/Ras/PI(3)K		—	LZTR1	—		—	—	—		Y
	RTK/Ras/PI(3)K		—	MYB	—	Y	—	—	—		Y
	RTK/Ras/PI(3)K		—	NTRK2	—		—	—	—		Y
	RTK/Ras/PI(3)K		—	TSC2	—	Y	—	TSC2	—	Y	Y
	SHH		—	PTCH1	—	Y	—	—	—		Y
	SHH		—	PTCH2	—		—	—	—		Y
	SHH		—	SMO	—	Y	—	—	—		Y
	WNT		—	APC	—		—	APC	—		Y
	WNT		—	CREBBP	—	Y	—	CREBBP	—	Y	Y
	WNT		—	DICER1	—		—	—	—		Y
	WNT		—	KLF4	—		—	—	—		Y
	WNT		—	KMT2D	—		—	—	—		Y
Risk mutations related to heritable diseases (Barthel et al. [8])		TERC									N
		OBFC1									N
		POT1									N
		RTEL1									N
		TERT	TERT	TERT	TERT		—	—	—		Y
		TP53	TP53	TP53	—	Y	TP53	TP53	—	Y	Y
		NF1	—	—	—		—	—	—		Y
		NF2	—	—	—		—	—	—		Y
		CHK2 (CHEK2)	—	CHEK2	—	Y	—	—	—		Y

Also included is a list of risk mutations related to heritable diseases. Genes identified with VUS that were possibly pathogenic in the GBM cohort are highlighted in bold.

Table 4

Summary of the proportion of initial and recurrent of IDH-wildtype and IDH-mutant glioblastoma patient tumours that had SNVs that could be assigned as potentially clinically actionable.

Gene	IDH-wildtype				IDH-mutant				Frequency in GBM (Sahm et al.)	Targeted agent (clinical trial)
	Initial tumour		Recurrent tumour		Initial tumour		Recurrent tumour
	N	%	N	%	N	%	N	%	%
PIK3CA	2/38	5	0/7	0	0/3	0	0/1	0	6.3	mTOR inhibitor; everolimus (NCT02449538); BKM120/everolimus (NCT01470209)
PIK3R1	2/38	5	0/7	0	0/3	0	0/1	0		mTOR inhibitor
EGFR	6/38	16	1/7	14	0/3	0	0/1	0	34	ABBV-221 (NCT02365662); naratinib (NCT01953926); AZD9291 (NCT02465060); EGFR-targeting antibodies, vaccines, TK inhibitors, osimertinib, poziotinib
PDGFRA	2/38	5	0/7	0	0/3	0	0/1	0	11	Dasatinib; nilotinib/Pazopanib (NCT02029001); MGCD516 (NCT02219711)
BRAF	1/38	3	0/7	0	0/3	0	0/1	0		Vemurafenib; MEK inhibitor
PTEN	11/38	29	1/7	14	0/3	0	0/1	0	32	INC280/BKM120 (NCT01870726); everolimus (NCT02449538); erlotinib, everolimus or dasatinib (NCT02233049); GSK2636771 (NCT01458067); BMN673 (NCT02286687); BKM120/everolimus (NCT01470209)
BRCA1	7/38	18	1/7	14	0/3	0	0/1	0		Olaparib
BRCA2	7/38	18	1/7	14	1/3	33	0/1	0		Olaparib
PTCH1	1/38	3	0/7	0	0/3	0	0/1	0		SMO inhibitor, sonidegib and vismodegib
SMO	3/38	8	0/7	0	0/3	0	0/1	0		SMO inhibitor, sonidegib and vismodegib
ATR	3/38	8	0/7	0	0/3	0	0/1	0		ATR inhibitor (BAY1895344)
MSH6	4/38	11	0/7	0	1/3	33	0/1	0	4.3	MK-3475 (NCT01876511)
TP53	7/38	18	0/7	0	3/3	100	0/1	0
CDKN2A	3/38	8	0/7	0	0/3	0	0/1	0
RB1	1/38	3	0/7	0	0/3	0	0/1	0
ATM	6/38	16	1/7	14	0/3	0	0/1	0

For particular genetic alterations, the proportion of glioblastomas (n = 47) with alterations in those genes, as recorded by Sahm et al. [30], is also provided. Also summarised are available and new therapeutic agents currently on trial in clinical studies targeting molecular aberrations.

3.3. SNVs Detected in Initial and Recurrent IDHMUT Glioblastomas

SNVs detected in IDHMUT initial (n = 12) and recurrent tumours (n = 1; Tables , and ) impacted IDH1 and 10 genes across 5 biological pathways (Figures 1 and 2; Table 2). Majority of initial tumours had SNVs in genes in the RTK/Ras/PI(3)K (66%; 2/3), followed by p53 (100%; 3/3) and WNT signalling pathway (33%; 1/3). All initial IDHMUT tumours (100%; 3/3) harboured at least one potentially actionable variation in TP53 (100%; 3/3), BRCA2 (33%; 1/3), and MSH6 (33%; 1/3; Table 4). Just 7 SNVs in 6 genes were VUS that were possibly pathogenic in IDHMUT initial tumours. These included IDH1 and the p53 pathway genes MSH6 and TP53 and the RTK/Ras/PI(3)K genes KLK1 and TSC2 and the CREBBP gene in the WNT pathway (Table 3). The KLK1 variant was potentially pathogenic in IDHMUT but not in IDHWT. The recurrent IDHMUT tumour had SNVs in p53, WNT signalling, and G-protein pathway genes. Matched analysis revealed that seven genes had SNVs in the initial that were not observed in the recurrent tumour (Figure 2). The recurrent tumour had SNVs in one gene not recorded in the initial (GNAS). No genes had SNVs that were potentially actionable in the recurrent IDHMUT tumour (Table 4).

3.4. CNVs in IDHWT and IDHMUT Glioblastomas

CNVs were detected in IDHWT tumours only (). The results for CNVs in the corresponding genes in TCGA-GBM are presented in . For sample 36, there appears to be a hemizygous deletion in BRCA2 in the initial, but a CNV gain in the recurrent tumour. Both trends were identified in TCGA-GBM, but predominantly BRCA2 had shallow deletions. There were CNV gains in GNAS for recurrent sample 3b. TCGA-GBM results also predominantly indicate CNV gains for GNAS. In recurrent samples 1b and 7b, TERT appeared to have hemizygous deletions. TCGA-GBM had both TERT CNV losses and gains with no predominant trend evident. For SMARCA4, there appears to be a CNV gain in initial sample 1 but a hemizygous deletion in the recurrent sample. TCGA-GBM had mostly CNV gains with some losses for SMARCA4. Significant CNV gains in EGFR were observed for initial and recurrent sample 2 and similarly in TCGA-GBM cases.

3.5. Investigation of the Corresponding Genes (with Mutations and CNVs in the GBM Cohort) in the TCGA-GBM and GDC Datasets

The results of investigations in the TCGA-GBM and GDC datasets for the 21 genes identified with VUS that were possibly pathogenic in the GBM cohort are presented in . A summary of SNVs identified from those corresponding genes in the TCGA-GBM dataset is provided in . TCGA-GBM cases in the mutation data included 6 verified and 2 ambiguous IDH-mutant individuals; however, majority of cases are unannotated. PTEN was the gene most impacted by mutations (34.86%) and shallow or deep deletions (; ). EGFR had mutations (26.97%) and CNV gains. FGFR2 (1.53%), JAK2 (1.27%), MYB (1.27%), and ATM (2.04%) had fewer mutations and mostly shallow or deep deletions. Both BRAF (2.54%) and SMO (1.02%) had fewer mutations and mostly low level CNV gains. TP53 (31.55%), PIK3CA (10.18%), and PIK3R1 (10.94%) had relatively high mutations and a mixture of CNV gains and deletions. IDH1 (6.62%), BRCA1 (2.8%), PTCH1 (3.56%), CREBBP (3.56%), MSH6 (3.05%), DAXX (2.29%), TSC2 (2.04%), PPM1D (1.78%), KLK1 (0.51%), and CHEK2 (0.25%) had low rate of mutations and a mixture of CNV low level gains and losses. BRCA1 (2.8%) had low rate of mutations and both CNV low level gains and shallow or deep deletions. The results for the 12 NOTCH, SHH, and WNT pathway genes identified to be impacted in the GBM cohort investigated in the TCGA-GBM and GDC datasets are presented in and . The WNT pathway genes DICER1 (2.29%), KLF4 (0.25%), and CREBBP (3.56%) had mutations and CNV shallow deletions, as well as low level gains and high level amplifications. TERT (2.80%) and KMT2D (3.05%) had mutations and CNV shallow gains and losses as well as deep deletions. APC (4.58%) and TCF4 (0.76%) had mutations, low level gains, and shallow deletions. The SHH genes, PTCH1 (3.56%), PTCH2 (1.78%), and SMO (1.02%) were impacted by mutations. Whilst the SMO gene had CNV gains, by comparison, the PTCH1 and PTCH2 genes had both CNV gains and losses. NOTCH genes, NOTCH2 (4.07%) and NOTCH1 (0.25%), had mutations and were impacted also by gains and losses in CNV.

3.6. Impact of SNV Burden on Survival in IDHWT GBM Patients

The number of tumour SNVs was prognostic for survival in methylated GBM patients (log rank = 7.63, 95% CI = 6.90–27.10; P value = 0.006, two-sided). Median survival for methylated GBM with ≤ 4 SNVs was 23 months compared to a median survival of 10 months for a tumour with ≥ 5 SNVs (Figure 3; ). For unmethylated GBM patients, the number of tumour SNVs was not prognostic for survival (log rank = 3.393, 95% CI = 9.441–12.559; P value = 0.065). Median survival was 13 months for unmethylated GBMs with ≤ 4 SNVs, compared to a median survival of 11 months for ≥ 5 SNVs (Figure 4; ). Sample sizes were relatively small in these survival analyses; therefore, the observed trends would need to be confirmed using a larger cohort.

Figure 3

Comparison of survival for IDHWT glioblastoma MGMT methylated patients with high versus low total number of tumour SNVs, based on a median split. Kaplan–Meier analysis indicates that IDH GBM patients with a greater tumour SNV burden have significantly a shorter overall survival (P=0.006).

Figure 4

Comparison of survival for IDHWT glioblastoma MGMT unmethylated patients with high versus low total number of tumour SNVs, based on a median split. Kaplan–Meier analysis indicates that IDH GBM patients with a greater tumour SNV burden have a shorter overall survival; however, this trend was not significant (P=0.065).

4. Discussion

The mutational landscape of the GBM subtypes in this cohort raises the possibility of new combinations of therapeutic approaches for individual GBM patients. Potentially actionable variation was most often identified in EGFR, PTEN, BRCA1/2, and ATM. These genetic alterations could be targeted by novel approaches with EGFR-targeting antibodies, tyrosine kinase inhibitors, and DNA damage repair inhibitors either singly or in combination. In particular, the BRCA1/2 mutations raise the possibility that DNA damage repair agents may be an option for small numbers of GBM patients in combination with other agents. Administering olaparib PARP (poly (ADP-ribose) polymerase) inhibitor, developed for BRCA1/2 mutated ovarian cancer, in combination with TMZ has shown promising results for treating relapsed glioblastoma patients in a phase I clinical trial (NCT01390571) [44]. However, patient selection to date has not been based on detailed molecular profiling with HTS. In this study's GBM cohort, both IDHWT and IDHMUT GBM had VUS that were predicted to be pathogenic in MSH6 [45-47], CREBBP [48-52], TP53 [17, 47], and TSC2 [36–43, 53]. In particular, MSH6 (MutS homolog 6) is a DNA mismatch-repair protein that has been identified as a putative driver gene in glioma [45, 47]. Similarly, MSH6 may be involved in acquired resistance to alkylating agents [46]. Moreover, CREBBP (CREB binding protein gene/CBP) activates the DNA damage response and repair pathway by acetylating factors involved in base excision repair, nucleotide excision repair, nonhomologous end joining, and double-strand break repair (e.g., PARP-1, H2AX, and NBS1) [49].

4.1. IDHWT Glioblastomas

In IDHWT glioblastomas, SNVs impacted genes in the RTK/Ras/PI(3)K (79%), p53 (61%), WNT (58%), SHH (16%), NOTCH (8%), Rb (5%) and G-protein (5%) pathways. Potentially actionable mutations detected from initial IDHWT tumours included EGFR, PTEN, BRCA1, BRCA2, ATM, and ATR [54-56]. Therapies for this subtype might include the EGFR-targeting antibodies, EGFR-targeting vaccines, TK inhibitors, erlotinib, and DNA damage repair inhibitors including olaparib and ATR inhibitors. Anti-EGFR-targeting antibodies to date have not shown clinical efficacy in GBM although trials are ongoing [57]. Similarly, trials of DNA damage repair inhibitors are underway, and the results are anticipated; however, patients have not been selected for these trials using molecular profiling with HTS.

Interestingly, in this cohort, a high proportion of IDHWT tumours was impacted by BRCA1 (18%) and BRCA2 (18%) mutations. This trend was not observed in the TCGA-GBM dataset (2.8%; 2.3%); however, the IDH status of patients is not confirmed in most cases [58]. Only one variant from the GBM cohort (BRCA1 : Ch17: 41246062) was identifiable amongst the TCGA-GBM dataset BRCA1 (n = 16) and BRCA2 (n = 39) variants. The well-known breast cancer specific germline mutations in BRCA1 (185delAG; Chr17: 43124030–43124031 and 5382insC; Chr17: 43057065) and BRCA2 (6174delT; Chr13: 32340301) were not amongst the variants identified in either the GBM cohort or the TCGA-GBM cohort. In this GBM cohort, amongst the BRCA2 variants were confirmed somatic mutations in haemangioblastoma (BRCA2 : COSM3753648, COSM5019704) [43], which is a rare, benign tumour that typically occurs in the cerebellum [3]. Many IDHWT tumours had alterations impacting WNT [59-63] signalling pathway genes (58%) including CREBBP(4), KLF4(2) [64, 65], TERT(2) [17], and APC(3) [66-70]; however, targeting this pathway is currently challenging. Initial IDHWT tumours also showed predicted pathogenic variation in NOTCH (11%) [71] and SHH (13%) pathways [72] including PTCH1 (PATCHED-1) and SMO (Smoothened) [73-75]. The Hedgehog antagonist GDC-0449 (vismodegib) has been trialled in recurrent GBM (NCT00980343) and childhood brain tumours with varying success to date.

4.2. Recurrent IDHWT Glioblastomas

Interestingly in this cohort, no tumours exhibited a TMZ-induced hypermutated phenotype. Tumours did not have mutations in TERT promoter regions. Kim et al. found that a TMZ-induced hypermutated phenotype was rare in IDH-wildtype primary glioblastomas [76]. Acquired resistance in glioma has been attributed to dysregulated pathways (signalling and DNA repair), persistence of cancer stem cell subpopulations, and autophagy mechanisms [77]. In this cohort, only the RTK/Ras/PI(3)K, p53 DNA damage repair, WNT signalling, and G-protein pathways were impacted by genetic alterations and not the SHH, NOTCH, and Rb pathways, despite their association with glioma resistance. Whilst fewer pathways were impacted, intertumour heterogeneity between initial and recurrent IDH wildtype tumours was nevertheless observed, similar to previous studies [76, 78]. Indeed, recurrent tumours can diverge to such an extent that they are no longer recognised as lineal descendants of the dominant clone identified initial at diagnosis [78, 79]. Potential signatures of IDHWT recurrent tumour resistance included VUS that were possibly pathogenic in PTEN. PTEN mutations cause activation of the PI3K/AKT survival pathway and chemoresistance in GBM [80]. Other possible signatures of recurrent tumour resistance in this GBM cohort included CNV gains in the genes (chromosome), BRCA2 (Chr13), GNAS (Guanine nucleotide-binding protein G(s) subunit alpha; Chr20), and EGFR (Chr7). Copy number gains are thought to impact driver genes to initiate tumourigenesis. The oncogene EGFR is located on chromosome 7, which frequently has CNV gains in IDH-wildtype glioblastomas (∼70%) [5, 6]. Gains in the chromosome 20 arm containing GNAS are frequently observed in pituitary brain tumours (adenomas) and may exert a mitogenic influence on the WNT signalling pathway via cAMP activation, which may provide a proliferative advantage for resistance [81]. However, GNAS has not been identified as a prognostic in dicator implicated in GBM [82]. CNV losses observed in the GBM cohort included SMARCA4 (Chr19) [47] and TERT (Chr5). CNV losses may be concordant with gene expression downregulation [83].

4.3. IDHMUT Glioblastomas

Results for IDHMUT glioblastomas comprised three initial and one recurrent case only. Pathways impacted by genetic alterations included the RTK/Ras/PI(3)K (66%), p53 (100%), and WNT pathways (33%). Possibly pathogenic VUS identified herein included those co-mutated in both subtypes as well as KLK1 (kallikrein1). The kallikreins KLK6, KLK7, and KLK9 have been shown to have higher protein levels in Grade IV glioma compared to Grade III tumours and consequently may have utility as prognostic markers for patient survival [84]. All initial IDHMUT tumour samples harboured potentially actionable variation in at least one of the genes TP53, BRCA2, and MSH6. The recurrent tumour had fewer pathways (p53, WNT, and G-protein) impacted by genetic alterations. Matched analysis revealed intertumour heterogeneity. The recurrent IDHMUT tumour lacked potentially actionable variation that could be targeted. Given the small sample size for this subtype all trends reported here would need to be confirmed in a larger cohort.

5. Conclusion

Our study reveals that matched initial and recurrent GBM samples harbour potentially actionable variations, and these were most often identified in EGFR, PTEN, BRCA1/2, and ATM. These genetic alterations could potentially be targeted by novel approaches with EGFR-targeting antibodies, tyrosine kinase inhibitors, and DNA damage repair inhibitors either singly or in combination. This study underlines the need for detailed genetic analysis of GBM patients to identify individuals that might benefit from novel therapeutic approaches that are becoming available in the near future. This information is also important for patient recruitment to clinical trials.