16/04/2022
Objectives: Tumors of the central nervous system (CNS) are the most common pediatric solid tumors, where low grade (LGG) and high grade gliomas (HGG) represent up to 55% of CNS tumors. Current molecular classification of these tumors results in a more accurate diagnosis and risk stratification, which ultimately enables individualized treatment strategies. Identifying known alterations is a suitable approach, particularly in developing countries, where NGS approaches are not easily accessible. We sought to assess molecular alterations in BRAF and histone 3 genes. Study design: FISH, IHC and Sanger sequencing were performed in a series of 102 pediatric glial and glioneuronal tumors. We also correlated these results with clinical and histological findings to evaluate their usefulness as diagnostic and/or prognostic tools. Results: We found that the KIAA1549-BRAF gene fusion was a relevant diagnostic tool for pilocytic astrocytoma, but not related to progression free survival (PFS) and overall survival (OS). BRAFV600E mutation was associated with a decreased OS in LGG, and with decreased PFS and OS among pilocytic astrocytomas. All HGG of the midline were H3K27M mutants, while H3G34R mutant cases were located in brain hemispheres. HGG harboring the H3K27M variant were associated with a decreased PFS and OS. Conclusions: Assessing druggable molecular markers with prognostic value is particularly important in those cases where complete resection or further radiation therapy is not possible. These potential diagnostic/prognostic markers may be suitable as further screening tests to reduce the requirement on NGS, which is not available in all laboratories. Furthermore, these results broaden data on BRAF and Histone 3 alterations in children from geographic regions, other than USA and Europe.
Tumors of the central nervous system (CNS) are the most common solid tumors in pediatric patients; all together they account for more than 20% of all childhood cancers [1]. Within CNS tumors, gliomas represent up to 55% of them. The World Health Organization (WHO) defines these pediatric brain tumors into four stages based on their morphology and histological characteristics, where grades I and II (GI and GII) are considered low grade gliomas (LGG) and grades III and IV (GIII and GIV) are considered high grade gliomas (HGH) [2]. This classification reflects the survival odds and generally speaking, grade I have the highest survival odds and grade IV, the lowest. Pediatric LGG include pilocytic astrocytoma (GI), pleomorphic xanthoastrocytoma (GII), diffuse astrocytoma (GII), those entities related to ganglioglioma (GI/II) and dysembryoplastic neuroepithelial tumor (GI). Pediatric HGG are a heterogeneous group of malignant and poorly delimited tumors which extend both macroscopically and radiologically, beyond apparent margins. The most frequent in pediatrics include anaplastic astrocytoma (GIII), glioblastoma (GIV), diffuse intrinsic pontine glioma (GIV) and diffuse midline glioma H3 K27M-mutant (DMG K27M) (GIV).
According to the Argentine Oncopediatric Hospital Registry (ROHA, Registro Oncopediátrico Hospitalario Argentino), approximately 1400 new cases of cancer are diagnosed annually in our country in children under 15 years of age, where CNS tumors are the most frequent solid tumors [3]. Incidence rates in Argentina are similar to those described in other regions [1]. However, in recent years and given the astonishing advances in molecular biology and its implications in diagnosis, prognostic accuracy and tailored-made treatment strategies of pediatric brain tumors, specific molecular markers for each tumor type have been transferred from the bench to the bed-side [4]. More precisely, DMG K27M was defined by WHO based on the molecular characterization of the H3F3A, HIST1H3B and HIST1H3C genes [2]. This entity harbors a H3K27M hallmark somatic mutations in the histone 3 (H3) isoforms H3.3 (coded by H3F3A gene) or isoform H3.1 (HIST1H3B or HIST1H3C genes) in up to 80% of pediatric diffuse midline gliomas, including thalamic glioma and diffuse intrinsic pontine glioma (DIPG) [5]. A different hallmark mutation, H3G34R/V has been described within the same histone genes in gliomas arising from the cerebral hemispheres [6]. These variants have become the most frequently studied in gliomas; more precisely, H3K27M directly disturbs the target site for lysine specific methyltransferases and prevents methylation of the mutated residues, where the original lysine would have been methylated. Additionally, H3G34R/V interferes with the modification of the nearby K36 residue [7]. Given that H3K27M mutant gliomas have a decreased treatment response and a diminished overall survival, in relation to the non-mutant tumors, H3K27M is now considered a tumor driver mutation and is used as a prognostic marker.
In a similar fashion, the V600E mutation in BRAF (coded by the BRAF gene) highly correlates with a poorer prognosis and overall survival across a broad spectrum of pediatric LGG, while the KIAA1549-BRAF gene fusion aids in the diagnosis of pilocytic astrocytomas (PA) [4]. The MAPK signaling pathway regulates various cellular functions, including cell cycle control, cell proliferation, differentiation and apoptosis, in all of which BRAF acts as a stimulatory factor [8]. Moreover, BRAF V600E mutation has been reported to abnormally up-regulate the MAPK signaling pathway [9, 10]. Even though the BRAF V600E mutation responds unfavorably to conventional treatment, it represents a specific druggable target for personalized therapies with kinase inhibitors [11]. Moreover, CDKN2A has been previously shown to influence the response to treatment and outcome in BRAF V600E pediatric LGG [12–14]. On the other hand, the KIAA1549-BRAF gene fusion is formed by a tandem duplication of a 2 MB region (locus 7q34 in BRAF gene) with the subsequent fusion to KIAA1549 gene, which finally results in the constitutive phosphorylation activity by BRAF and up-regulation of the MAPK pathway [15]. Although several translocation break-points were described, the most common is between exon16 of the KIAA1549 gene and exon 9 of BRAF (16–9 fusion); for a comprehensive review refer to [16]. While the BRAF V600E mutation is found in about 60% of pleomorphic xanthoastrocytoma (PXA), 50% of gangliomas, 9% of PA and 2–12% glioblastoma, the KIAA1549–BRAF fusion is frequently found roughly in 50–80% of PA and pilomyxoid astrocytomas (PMA), which is an aggressive variant of PA [16]. Despite all these facts, BRAF molecular profiling has not yet been included in the latest WHO classification of central nervous system tumors.
The molecular assessment of these markers aids in diagnosis, in identifying risk groups and individualizing treatment strategies, a fact of special importance in pediatrics since it may avoid unnecessary sequelae in a developing child. This approach, assessing known alterations by standard molecular techniques is useful, particularly in developing countries, so as to reduce the requirements for NGS or when the latter approach is not accessible.
In the present study, we evaluated the molecular alterations in BRAF and histone 3 genes in pediatric glial and glioneuronal tumors from our institution. We also correlated the results with clinical and histological findings in our pediatric cohort to evaluate their usefulness as diagnostic or prognostic tools in this age group.