
Neurofibromatosis type 1 (NF1) is one of the most common autosomal dominant tumor predisposition syndromes, affecting approximately 1 in 3,000 individuals worldwide [1-4]. Individuals with NF1 are born with one mutated, nonfunctional copy of the
Patients typically exhibit café-au-lait macules (CALMs), skinfold freckling, Lisch nodules, and tumors including peripheral nerve tumors such as cutaneous neurofibromas, plexiform neurofibromas (PNs), and malignant peripheral nerve sheath tumors (MPNSTs), as well as brain tumors like optic nerve gliomas and brainstem gliomas [1,6-8]. Additional symptoms include skeletal abnormalities, vascular issues, learning disabilities, attention deficits, an increased risk of autism, and social and behavioral problems [7,9]. Cognitive and behavioral disorders impact up to 80% of children diagnosed with NF1 and are among the most significant clinical manifestations for patients and their families [9,10]. Hematopoietic neoplasms, such as juvenile myelomonocytic leukemia, and pheochromocytomas are also common in NF1 patients [11].
In this review, we comprehensively summarize recent research knowledge that significantly enhances the development of NF1 therapeutics.
Since the discovery of the
Loss of NF1 function results in constitutive Ras/Raf/MEK/ERK signaling activation, a primary driver of tumorigenesis in NF1 patients [14,20-23]. In 2011, a pivotal phase I clinical trial demonstrated that selumetinib, a selective MEK1/2 inhibitor, could reduce PNs in NF1 patients [24]. This was a significant breakthrough because PNs are often inoperable and can cause severe pain, disfigurement, and other complications. Subsequent phase II trials confirmed these findings, showing tumor reduction in 68% of patients and improved quality of life. Selumetinib, became the first Food and Drug Administration (FDA)-approved treatment for PNs in pediatric patients in 2020 [25-27]. In 2024, mirdametinib received orphan drug, fast track, and rare pediatric disease designations from both the FDA and European Medicines Agency [28-30].
Despite these advancements, several challenges remain in developing therapeutics for NF1. Current MEK inhibitors show limited efficacy against more aggressive tumors, such as MPNSTs and high-grade gliomas. Additionally, PNs is a heterogeneous disease, with patients displaying variable responses to MEK inhibitors, highlighting the need for personalized treatment approaches. The long-term safety profiles and potential resistance to MEK inhibitors require further investigation to ensure sustained efficacy and patient safety. Exploring combination therapies that target multiple pathways simultaneously could enhance treatment outcomes. Furthermore, identifying new therapeutic targets beyond the MEK pathway is crucial for developing more effective treatments for NF1.
Currently, other MEK inhibitors (trametinib, binimetinib, etc.) and the tyrosine kinase inhibitors (cabozantinib, etc.) that target a broad range of pathways, are being investigated in clinical trials for children and adults with PNs (Fig. 1) [26,28,31-33].
To date, identifying specific
Up to now, four clinically confirmed genotype-phenotype correlations have been reported out of more than 3,197 identified constitutional
Approximately 4.7-11% of NF1 patients have microdeletions involving 14 protein-coding genes, including NF1, and 4 microRNA genes [34-36]. These microdeletions are classified into four types based on their size.
Type-1 deletions: the most common, span 1.4 Mb and are usually caused by maternal meiosis-related non-allelic homologous recombination (NAHR) between NF1-REPa and NF1-REPc. These deletions include 14 protein-coding genes and 4 microRNA genes [37,38].
Type-2 deletions: cover 1.2 Mb and include 13 protein-coding genes including
Type-3 deletions: span 1.0 Mb and involve 9 protein-coding genes, mediated by NAHR between NF1-REPb and NF1-REPc.
Type-4 deletions: lack recurrent breakpoints, exhibit variable gene loss, and can be either constitutional or postzygotic.
The extent of the deletion and the accompanying loss of adjacent genes can significantly influence the severity of NF1 deletion syndrome compared to NF1 arising from intragenic mutations. Furthermore, the mechanism of deletion, particularly in type-2 deletions, can result in mosaicism, which inherently contributes to phenotypic variability [34].
Patients with type-1 deletions display a more severe phenotype, with increased cutaneous, subcutaneous, plexiform, and spinal neurofibromas, and a significantly higher burden of internal neurofibromas [35,39]. They have a fourfold increased risk of MPNSTs, particularly when
Studies have demonstrated a positive correlation between
The genotype-phenotype correlation of the p.Met992del variant was first described in 2007 in relation to an in-frame deletion at codon 992, and it has been further characterized through multiple study cohorts [42]. Patients with the p.Met992del mutation typically exhibit a milder phenotype, primarily consisting of café-au-lait spots and axillary freckling, without the characteristic cutaneous and PNs of NF1. Other clinical features include learning disabilities (17-38.8%), thoracic abnormalities (16%), short stature (11%), scoliosis (10%), pulmonary stenosis (9%), macrocephaly (9%), and symptomatic spinal neurofibromas (2%). Although 4.8% of individuals were found to have non-optic pathway tumors, these were mostly low-grade and asymptomatic, with no detectable cutaneous or PNs. The overall prevalence of lipomas was 5.5%. The molecular mechanisms underlying this mutation remain unknown. This mild phenotypic spectrum overlaps with the clinical features observed in Legius syndrome, caused by pathogenic variants in SPRED1. However, Legius syndrome patients differ from NF1 patients in that they do not exhibit Lisch nodules [43].
Patients with NF1 carrying the p.Arg1809Cys mutation, resulting from a missense change NF1 c.5425C>T, which affects an arginine residue within the pleckstrin homology (PH) domain of neurofibromin, presented with café-au-lait spots and freckles, macrocephaly, thoracic abnormalities, growth retardation, and learning disabilities [44]. Similar to those with the p.Met992del mutation, these patients did not exhibit overt cutaneous, spinal, or PNs, optic pathway gliomas, other malignancies, or skeletal abnormalities. Approximately 25% of individuals displayed Noonan-like features, with pulmonary stenosis and short stature being significantly more prevalent in this group compared to the classical NF1 cohort [45].
Patients with missense mutations affecting one of the codons 844-848 within the cysteine-serine rich (CSR) domain exhibit a severe phenotype [46]. These patients show a high prevalence of plexiform or spinal neurofibromas, symptomatic and asymptomatic optic pathway gliomas, malignant neoplasms, and skeletal abnormalities. This severe phenotype is observed in 75% of adult NF1-affected individuals with these variants, clearly indicating that missense mutations outside the GAP-related domain (GRD) can result in severe clinical presentation. Notably, 25% of patients with NF1 carrying these variants do not present with the typical severe phenotype.
In alignment with genotype-phenotype correlations, the identification of mosaicism and loss of heterozygosity (LOH) events through deep sequencing can more effectively associate specific genetic alterations with clinical outcomes. This capability enhances the prediction of disease progression and the customization of individualized treatment strategies.
Mosaic neurofibromas or segmental NF1 occur when the
LOH, which occurs due to somatic loss, refers to the loss of the normal (wild type) allele in cells that already possess one mutated NF1 allele.
The first report of NF1 LOH was published in 2007 by researchers investigating the genetic mechanisms in patients with pheochromocytoma and NF1 [51]. They found that 67% of patients with NF1-related tumors had somatic loss of the unmutated allele at the
In melanocytes, LOH in NF1 increases melanogenic gene expression and activates the MAP kinase pathway, resulting in pigmentary abnormalities such as CALMs and freckling. This loss of neurofibromin disrupts the Kit-Mitf signaling axis, essential for melanocyte development and differentiation [52]. In Schwann cells, LOH leads to the formation of PNs and cutaneous neurofibroma [53-56]. LOH in astrocytes contributes to the development of optic pathway gliomas, low-grade tumors that can impair vision [53]. In hematopoietic cells, LOH can result in hematologic malignancies such as juvenile myeloid leukemia [56].
Recent studies focus on identifying genetic variations in LOH through WGS as well as understanding the mechanisms of LOH.
While not discussed in detail here, pathophysiological studies have shown that benign tumors such as PNs and low-grade gliomas (LGGs) in NF1 patients can transform into MPNSTs and high-grade gliomas, respectively. Atypical neurofibromatous neoplasms (ANNUBPs) represent an intermediate stage between neurofibromas and MPNSTs. ANNUBPs are associated with the loss of the
The involvement of the tumor microenvironment (TME) in NF1 pathogenesis is increasingly recognized. Recent advances in single-cell RNA sequencing have delineated the intricate cellular composition and signaling dynamics within mature neurofibromas, elucidating the complexity of the TME [60,61].
Research has demonstrated that schwann cells-immunes (mast cells, T cells) in PNs [62-67], microglia and T cells in LGGs [68-71], and hematopoietic cells in MPNSTs [72,73], play a significant role in promoting the development of NF1-associated tumors. These interactions between immune cells and the TME present potential targets for therapeutic intervention.
In the beginning,
Research has shown that
In a mouse model of neurofibromas, the CXCL10/CXCR3 axis involving dendritic cells and T cells plays a critical role in the early development (approximately 2 months) of neurofibromas. Macrophages contribute to neurofibroma formation through STAT3 and CSF-1/CSF1R signaling. These findings suggest that inhibitory molecules targeting STAT3, CSF-1/CSF1R, and pegylated interferon alpha 2b may have the potential to suppress tumor growth [67].
Approximately 15-20% of children with NF1 develop LGGs along the optic pathway, leading to visual impairment in 30-50% of these cases [68,69]. In the microenvironment of NF1 glioma tumors, increased c-Jun N-terminal kinase (JNK) activity in
These findings highlight the importance of microglia and immune interactions in NF1 tumor development, offering potential targets for therapeutic intervention to prevent visual decline.
NF1-related MPNSTs are aggressive tumors that develop from PNs in about 5-10% of NF1 patients [72]. A recent study compared the tumor environments of two NF1-related MPNST mouse models, one with and one without
This enhanced understanding of the TME has been instrumental in the development of targeted therapies, including combination approaches, and has propelled the application of personalized medicine for NF1-associated malignancies. Progress in TME research has unveiled new avenues for diagnostic strategies, risk assessment, and therapeutic development, ultimately aiming to improve clinical outcomes for patients with NF1-associated tumors.
The nomenclature for NF1 isoforms was updated in 2017 to provide a unified annotation system for the
The most abundant form of NF1 mRNA contains 57 exons (not including 4 alternatively spliced exons) and encodes the 2818 amino acid neurofibromin protein [74]. Several transcript variants resulting from alternative splicing of the NF1 pre-messenger RNA have been identified [74-76]. To date, five alternatively spliced exons leading to the expression of NF1 isoforms have been specifically studied: 9a, 10a-2, 23a, 43, and 48a (according to the previous nomenclature) [74-77].
It is important to note that studies have shown that alternative splicing of exon 23a impacts the GTPase activity of neurofibromin encoded by the
The isoform 2 transcriptome exhibits differential expression between isoform 1 and various tissues, with isoform 2 being preferentially expressed in differentiated cells [74-76]. Embryonic stem cells expressing exclusively isoform 2 exhibit increased activity of the Ras/ERK pathway, which is crucial for NF1 cell proliferation and survival, while cAMP levels remain unchanged. This observation is consistent with previous studies in yeast, confirming a specific role for isoform 2 in regulating the Ras/ERK pathway without affecting cAMP levels [82]. Neurons expressing neurofibromin isoform 2 also demonstrate higher Ras/ERK signaling. Nguyen et al. [80] generated a mouse model that exclusively expresses isoform 2 in all tissues. These mice, referred to as NF1 23a/23a, were viable, fertile, and exhibited no physical abnormalities. However, the brains of NF1 23a/23a mice showed higher Ras/ERK pathway activity compared to wild-type mice and impairments in both short-term and long-term spatial memory [77,79,80].
Cryo-electron microscopy studies have demonstrated that NF1 isoforms 1 and 2 (exon 23a) exhibit different functional states through their 3D structures [83]. Isoform 1 typically exists in an open state, allowing it to interact with Ras and perform its GAP function effectively. Isoform 2 can exist in both closed (self-inhibited) and open states. The closed state is stabilized by zinc binding, inhibiting Ras binding, while the open state allows Ras interaction.
The distinct functional roles of NF1 isoforms underscore their involvement in various cellular processes and disease mechanisms, making these differences crucial considerations for developing novel therapeutic strategies.
Over the past century, research has significantly advanced our understanding of neurofibromas in NF1. Key discoveries include the identification of NF1 LOH in Schwann cells within neurofibromas, but not in fibroblasts. Mouse models have confirmed that the loss of NF1 in Schwann cells leads to the formation of neurofibromas, highlighting the crucial role of the TME in NF1 pathogenesis.
Recent studies suggest that NF1 LOH occurs in multipotent precursors, explaining the diverse phenotypes observed in NF1 patients. High-throughput technologies, such as next-generation sequencing (NGS) and whole-exome sequencing (WES), are revolutionizing clinical research. These technologies facilitate the development of novel therapeutics and advance precision medicine. Enhanced genotyping and phenotyping methodologies, combined with meticulous research approaches, are anticipated to elucidate the genetic complexity, underlying molecular mechanisms, and heterogeneous phenotypes observed in NF1 patients. Despite these advancements, further research is essential to develop specific therapeutic strategies tailored to the diverse mutation types observed in NF1 patients.
At this point in time, utilizing induced pluripotent stem (iPS) cells derived from patients with NF1 mutations allows researchers to study the disease in a patient-specific manner. This technology enables the development of personalized therapeutic strategies by providing a detailed understanding of NF1 pathology and testing potential treatments. iPS cells can be differentiated into relevant cell types, offering a powerful tool for advancing personalized medicine and targeted therapies for NF1.
None.
This research was supported by Brain Pool Program funded by the Ministry of Science and ICT through the National Research Foundation of Korea (RS-2024-00402454).
![]() |
![]() |