NF1 is associated with tumors of the peripheral and central nervous system (CNS). The most common CNS tumors in NF1 are gliomas, which are seen in approximately 20% of patients
[16][17]. Gliomas usually affect children, with mean age at diagnosis of 4.5 years; the vast majority of such tumors originate within the optic nerves, optic chiasm, and/or hypothalamus. While individuals with NF1 are at higher risk for developing low-grade gliomas compared to high-grade gliomas
[18][19], their risk for high-grade glioma is increased by 50-fold when compared to the general population
[20][21]. Indeed, high grade gliomas are rare tumors and the reported higher risk in children and adults with NF1 is based on epidemiologic studies and several case series
[22].
The World Health Organization (WHO) classification of gliomas has been refined and incorporated molecular parameters, namely 1p/19q codeletion, IDH1/2 mutation, and histone H3-K27M, in addition to histology to define many tumor entities
[23]. In general, low-grade gliomas form a group of WHO grade I and grade II tumors while high-grade gliomas form a group of WHO grade III and IV based on malignancy grade, molecular markers and presumed cell of origin. The most common glioma associated with NF1 is pilocytic astrocytoma, a WHO grade I tumor, with the optic pathway glioma being a hallmark lesion
[24]. Another low-grade astrocytoma that was reported in children with NF1 is pilomyxoid astrocytoma and the grading was suppressed in the revised 2016 WHO Classification to WHO grade I
[25]. In contrast to pilocytic astrocytomas, diffuse astrocytomas, which form WHO grade II, III and IV tumors, are more common in adult individuals with NF with only 12% presenting before the age of 20
[26]. A clinicopathologic study that examined tumors from 100 individuals with NF1 reported pilocytic astrocytoma frequency to be 49% while diffuse astrocytoma to be 27% which included WHO grade II (5%), III (15%), and IV (7%) though this grading used the 2007 WHO Classification
[26]. A recently published comprehensive genomic study performed in 23 high grade and 32 low-grade gliomas in individuals with NF1 demonstrated that children developed mostly low-grade gliomas (i.e., 77% of pediatric gliomas were low grade) whereas 78% of tumors in adults were high-grade
[27]. The study included whole exome sequencing of tumor and matched blood germline DNA to identify germline and somatic single nucleotide variants, small insertions and deletions, and copy number variants. The
NF1 variants observed in germline DNA were typically truncating and led to frameshifts, which did not cluster into specific NF1 protein domains. There was no association between particular patterns of
NF1 genetic variants and the risk of developing glioma. The data supported prior reports that a “second-hit” is required to develop tumors
[28], as loss of heterozygosity in the
NF1 region was detected in the majority of tumors. NF1-associated gliomas were found to have distinct genetic signatures, distinguishing them from those observed in sporadic gliomas, as well as noted to display different genetic landscapes when comparing low- vs. high-grade gliomas. For example, the isocitrate dehydrogenase (IDH) gene mutations (IDH1 and IDH2) are detected in more than 70% of sporadic low-grade gliomas and in the majority of glioblastomas arising from lower grade gliomas
[29]. Indeed, individuals with gliomas harboring IDH mutations have better prognosis than those with IDH wild-type
[30]. IDH mutations were not detected in gliomas associated with NF1 regardless of grade (). This finding may, in part, explain the observation that astrocytomas behave more aggressively than anticipated in adults with NF1
[31]. Another example is that mutations in H3.3 histone genes, frequently found in sporadic pediatric gliomas
[32], were absent in all samples regardless of age. Low-grade tumors exhibited fewer mutations that were over-represented in genes of the MAP kinase pathway, while high-grade tumors were characterized by a higher mutation burden and frequent mutations of
ATRX, typically co-occurring with alterations of
TP53 and cyclin-dependent kinase Inhibitor 2A (
CDKN2). DNA methylation assigned NF1-glioma to LGm6, a poorly defined IDH wild-type subgroup enriched with
ATRX mutations, which may represent a point of therapeutic intervention, as previous studies have shown that loss of ATRX increases sensitivity to DNA-damaging agents
[33][34]. summarizes the common molecular differences between NF1-associated gliomas and the LGm6 subgroup of sporadic gliomas
[27][35].
Summary table listing the frequencies (%) of mutations seen in NF1-glioma as studied by D’Angelo, et al. (Nat Med, 2019), separated by high grade (Grade III–IV) and low grade (I–II). This molecular profile most closely correlated to the LGm6 subgroup of pan-glioma cohort from The Cancer Genome Atlas (TCGA) project (Ceccarelli, et al. Cell 2016), which is an IDH-WT group enriched with ATRX mutations.
TERT = TERT copy number variant gain in NF1-Glioma and TERT promotor expression in LGm6 group. ATRX = inactivation of ATRX from any mutation. CDNK2A = loss of copy number variant. TP53 = frameshift or missense mutation in both groups. PTEN = combination of missense and frameshift mutations in the NF-1 glioma group; missense and loss in LGm6 group. PIK3CA = missense and in-frame indel. NF1 = frameshift/truncating. BRAF = missense in NF1-glioma group, missense and frameshift in LGm6 group.