Paraffin-embedded or fresh-frozen samples were collected from 62 pediatric patients with various grade I (n = 45) or grade II (n = 17 LGGs, as follows: pilocytic astrocytoma (n = 34), ganglioglioma (n = 8), diffuse astrocytoma (n = 6), pleomorphic xanthoastrocytoma (n = 6), pilomyxoid astrocytoma (n = 3), subependymal giant cell astrocytoma (n = 2), astroblastoma (n = 1), gangliocytoma (n = 1), oligodendroglioma (n = 1). Data on patient demographics, tumor location, treatment, response to treatment, and survival are shown in Table 1.

Polymerase chain reaction (PCR) DNA analysis for IDH1 and IDH2 was performed in 45 and 34 samples, respectively. All findings were negative. Of the 62 samples, 14 did not show a sequence for either gene (N/A) and 7 were evaluated for only one gene (N/P) (Table 1).

Real-time PCR (RT-PCR) was performed on 9 grade II pediatric glioma samples. A deletion was detected in exon 1 and exon 2 of CDKN2A in one sample (no. 54) of pleomorphic xanthoastrocytoma. The cycle threshold (Ct) values are shown in Figure 1.

Two samples of grade II LGG were initially analyzed for MYB translocation using rapid amplification of cDNA ends (RACE): one diffuse astrocytoma (no. 51) and one pilomyxoid astrocytoma (no. 55) (Figure 2A–D). When primers 1 + 3 were used, the RACE gel showed bands at 700, 600, and 450 k in both samples. With the primer combination 2 + 3, two bands were detected, at 200 and 380 bp.

Although we detected RACE products of translocation in the gel, direct sequencing showed only the normal MYB sequence, probably because the bands were weak (Figure 2D). The PCR results with a combination of 3 primers included MYB (forward + reverse)[(primers (1) + (2)], MYB (forward) + PCDHGA1(reverse)[primers(1) + (4)], PCDHGA1(forward + reverse) [primers (3) + (4)] in samples 53 and 51. A light band was detected in sample 53 using the 1 + 4 [MYB (forward) + PCDHGA1(reverse)] primer combination.

RT-PCR with a different combination of primers was used as an alternative method to detect MYB translocation with PCDHGA1 in 10 samples (Table 2). In samples without MYB translocation, the reactions with primers 1 + 2 or 3 + 4 led to a product, and in samples with MYB translocation with exon 2 of PCDHGA1, the reactions with primers 1 + 4 led to a product. Product was detected in sample 53 with primers 1 + 2 and 3 + 4 and also in sample 51 (Figure 3A–C). Five samples were positive for translocation of exon 9 of MYB with exon 2 of PCDHGA1: nos. 53, 55 (pilomyxoid astrocytoma), no. 62 (oligodendroglioma), and no. 60 (pleomorphic xanthoastrocytoma), and 4 were positive for translocation of exon 9 of MYB with exon 3 of PCDHGA1 (primer combination 5 + 6): nos. 51 (diffuse astrocytoma), 50 (ganglioglioma), 62 (oligodendroglioma), and 60 (pleomorphic xanthoastrocytoma).

Of the 10 samples, 5 were positive for translocation of MYB with exon 2 of PCDHGA1, and 4 were positive for MYB translocation with exon 3 of PCDHGA1. All samples positive for a translocation were negative for BRAF except sample 51.

To determine whether MYBL1 duplication-truncation was present in grade II pediatric LGGs, PCR for genomic DNA and RT-PCR were used. Ten samples of grade II gliomas were analyzed at the DNA level (Tables 3 and 4, Figure 3). A band at 350 bp on the gel represents a PCR product of the reaction. Eight samples were positive, as follows: no. 50 (ganglioglioma), 52, 51 and 49 (diffuse astrocytoma), 55 and 53 (pilomyxoid astrocytoma), 54 (pleomorphic xanthoastrocytoma), and 62 (oligodendroglioma). The remaining sample, no. 60 (pleomorphic xanthastrocytoma), was negative.

When PCR products of genomic DNA were run on gel, we detected a band at 350 bp in 6 samples: nos. 49, 51 and 52 (diffuse astrocytoma), 50 (ganglioglioma, and 55 and 53 (pilomyxoid astrocytoma). In one sample (no. 60, pleomorphic xanthoastrocytoma), the band was undetected, and in one (no. 54, pleomorphic xanthoastrocytoma), the results were equivocal. Sample 62 (oligodendroglioma) was negative in the gel but positive on RT-PCR. The data are summarized in Table 4.

It was clear that there were discordant results for MYBL1, so we appllied several different methods in an attempt to reach conclusive results.

Analysis of mRNA (converted to cDNA) was performed. The presence of amplification suggested positive MYBL1-trunc2 (Figures 4A,B). Four samples were positive for MYBL1-trunc2: nos. 50 (ganglioglioma), 58 (pilocytic astrocytoma), 60 (pleomorphic xanthoastrocytoma), and 52 (diffuse astrocytoma) (Table 4). Four samples were negative for MYBL1-trunc2: nos. 54 (pleomorphic xanthoastrocytoma), 62 (oligodendroglioma), 53 (pilomyxoid astrocytoma), and 61 (pleomorphic xanthoastrocytoma).

RNA sequencing results for BRAF showed no genetic alteration in 6 samples examined, probably because of low coverage in the area. On Sanger sequencing, 2 patients had the BRAF V600E genetic alteration.

RNA sequencing analysis of CDKN2A revealed one genetic alteration with high coverage quality, rs11515, in 5 of 6 patients. RT-PCR was performed on 9 grade II LGG samples. Deletions were detected in one of them (no. 54) in both exon 1 and exon 2 of CDKN2A (Figure 5).

RNA sequencing of IDH1 and IDH2 revealed no genetic alteration in IDH1 and 13 potential single nucleotide polymorphisms (SNPs) in IDH2 (n = 6). Eight of the SNPs were known intron variants with no pathogenic effect, 3 were unknown intron variants and were found to be heterozygous, and 2 were homozygous unknown intron variants and are under investigation.

Archival samples of pediatric LGGs (WHO grades I–II) were obtained from the Department of Pathology, Rabin Medical Center. Fresh tissues were also collected and analyzed after informed consent was obtained from the parents. Inclusion criteria were patient age <18 years and sufficient paraffin-embedded or frozen tissue for analysis. Tumor classification and cellularity were reviewed by a neuropathologist. Demographic and clinical data were collected from the medical files. The study was approved by the national and institutional review boards.

Paraffin-embedded tissues or frozen tissues were used for DNA isolation. In brief, 10 µm sections were stained with hematoxylin-eosin, and the slides were reviewed by a pathologist. The areas with an estimated content of more than 75% tumor cells were then separated by microdissection (no. 11 surgical blade) from 5 consecutive 10-µm unstained paraffin-embedded sections of each block. The tissues were deparaffinized and incubated overnight in 1% sodium dodecyl sulfate and proteinase K 0.5 mg/mL. DNA was purified by phenol chloroform extraction and ethanol precipitation and dissolved in 50 µL of distilled water, as described previously (13). All fresh tissues removed at surgery were taken from areas immediately adjacent to the tumor that was submitted to pathology.

Tumor tissue was snap-frozen in liquid nitrogen. Total RNA was isolated using TRIzol reagent (Invitrogen, Life Technologies, Carlsbad, CA) according to the manufacturer’s protocol and then reverse transcribed into complementary DNA (cDNA) using random hexamers (Amersham Biosciences, Buckinghamshire, United Kingdom) and Moloney murine leukemia virus reverse transcriptase (Promega, Madison, WI).

IDH1 and IDH2 analysis using specific designed primers for Sanger analysis.

RT-PCR was used to check deletions in CDKN2A. The ratio between the CDKN2A gene Ct and the reference gene RNase Ct was calculated. A ratio of <0.5 was considered a deletion. The primers for exons 1 and 2 of CDKN2A and the reference gene RNase are shown in Table 5 (24).

Oncogenic genetic alterations in MYB and MYBL1 were analyzed with rapid amplification of cDNA ends (RACE) using specific primers designed for the RACE reaction. The method entails initiation of first strand cDNA synthesis at the poly(A) tail of mRNA using the adapter primer and then destroying the original mRNA template with RNase. Specific cDNA is amplified by PCR using a gene-specific primer, which anneals in a region of known exon sequences, and an adapter primer that targets the 5´ terminus. The primers are listed in Table 6.

To identify MYB translocation and MYBL1 rearrangement, RNA was extracted from the tissue and converted to cDNA, as described (13). Triplicates of cDNA from LGG samples were analyzed. The mastermix contained 1 µg (1 µL), forward primer 0.5 µL (10 µM), reverse primers 0.5 µL (10 µM), Syber Fast 5 µL and H₂O 3 µL. The total reaction volume (10 mL) was loaded into a 96-well plate and run on StepOne System software, v 2.2 (Applied Biosystems, Foster City, CA, USA) in real time. The primers are listed in Table 2. The Ct levels of product expression were calculated.

To investigate whether MYB and BRAF genetic alterations are mutually exclusive, we analyzed MYB and BRAF rearrangements in the same samples.

After DNA extraction, each PCR amplification was performed in a 50 µL reaction volume containing 150 ng of sample DNA as a template using specific primers for the relevant genes. PCR parameters were as follows: denaturation for 5 min at 95°C; 35 cycles of 1 min at 95°C; and annealing for 1 min at 56–60°C and for 1 min at 72°C with Taq polymerase. The PCR product was amplified on 2% agarose gel and visualized with ethidium bromide staining.

Direct sequencing of the PCR products was performed with reagents and an analyzer (Big Dye Terminator Cycle Sequencing and ABI PRISM 3700 DNA Analyzer; Applied Biosystems).

For RNA sequencing, we used 5 samples of grade II LGGs (2 pilomyxoid astrocytoma, 2 diffuse astrocytoma, 1 pleomorphic xanthoastrocytoma) and one sample of grade I LGG (pilocytic astrocytoma).