Rare variants modulating phenotype in NF1 carriers

Abstract

Neurofibromatosis type 1 (NF1) is a rare genetic disorder with highly variable phenotypes, ranging from psychosocial challenges and congenital malformations to benign tumors and even aggressive cancers. We hypothesize that this variability stems from additional rare variants in other genes, in addition to NF1 variants. The analysis of 32 NF1 patients revealed that those with solid cancers carried a higher average of cancer driver variants especially in DNA repair genes compared to those without (p < 0.05). An extended validation study using 217 NF1 carriers (71 cancer and 146 controls) from UK biobank confirmed significant enrichment of pathogenic (P), likely pathogenic (LP) and uncertain significant (VUS) variants in DNA repair genes, in NF1 patients with tumors (FDR ≤ 0.05). Furthermore, P/LP variants in other genes are shown in those patients with NF1 ancillary traits such as cognitive impairments, macrocephaly, and connective defects. This study provides novel evidence suggesting that additional genetic variants in other genes may contribute to the phenotypic variability observed in NF1, indicating that rare secondary mutational events could influence specific manifestations, adding complexity to its variable expressivity.

Fig. 1

Variant burden in NF1 patients versus controls.

Fig. 2

Gene Variants and tumor associations. This Figure illustrates the presence of rare variants classified as P (Pathogenic), LP (Likely Pathogenic) or VUS (Variant of Uncertain Significance) within cancer driver genes across different patients. X-axis: Represents individual patients categorized by tumor type. Y-axis: Lists genes and pathways involved in cancer-related mechanisms. Blue-colored cells: Indicate the presence of a rare P, LP or VUS variant in the corresponding gene for a specific patient. White-colored cells: Indicate the absence of such variants in the respective genes for that patient.

Fig. 3

Results of Gene Set Enrichment Analysis (GSEA) based on cumulative pathogenicity scores.GSEA was performed on a ranked list of genes ordered by the difference in cumulative pathogenicity scores between cases and controls. Positive Normalized Enrichment Scores (NES) indicate pathways enriched in genes more pathogenic in cases, while negative NES reflect enrichment in controls. The color of each bar indicates statistical significance based on False Discovery Rate (FDR): dark blue = FDR ≤ 0.05; light blue = FDR > 0.05; orange = enrichment in controls. DNA repair-related pathways showed the strongest enrichment in the case group.

Fig. 4

Pedigrees with Segregating NF1 Variants. (A) A de novo NF1 case of a 4-month-old child (F6-P). (B) Family 1 was identified in the neonatology unit due to the baby presenting with macrocephaly and polysyndactyly (F1-P). (C) Family 2 was identified in the pediatric unit due to early growth failure and a neurodevelopmental disorder (F2-P). (D) Family 3 was identified in the pediatric-neuropsychiatry unit due to epilepsy and a neurodevelopmental disorder (F3-P). (E) Precocious puberty case. A 7 years-old male with a large umbilical hernia and abdominal neurofibroma (F7-P). (F) Family 4 was identified in the pediatric unit based on skin characteristics. Both the daughter F4-P and the father F4-F are shown. (G) Family 5 was identified during prenatal counseling of the partners (F5-P; F5-F). (H) Sporadic case a 36-year-old man presenting with multiple neurofibromas, freckles, and café-au-lait patches (ID 7). (I) An 11-years old male from family 9 with mosaic NF1 variant (F9-P). (J) A 9-years old female from family 8 with de novo NF1 variant (F8-P). (K) Sporadic case a 44-year-old woman presenting with multiple café-au-lait patches (ID 6).

Fig. 5

Gene and Phenotype Combinations. This figure illustrates the combinations of genes and phenotypes observed in the nine cases where both parents were also genetically characterized.The red box indicates pathogenic or likely pathogenic (P/LP) variants in dominant disorders, the yellow box indicates P/LP variants in recessive disorders, and the beige box represents variants of uncertain significance (VUS) in dominant disorders. The GNAS and H19 gene, boxed in blue, are paternally imprinted, with the disease caused by the inheritance of the mutation on the maternal allele. Gray boxes indicate phenotypes that are not assessable due to the young age of the patient. All the variants in the mentioned genes are reported in Supplementary Table 5.

Fig. 6

Experimental design. The Discovery Cohort UNISI involved 32 NF1 mutated patients underwent clinical data and exome sequencing analysis. Patients were divided into different cohorts based on their clinical characteristics: ‘NF1 Clinical Diagnosis’, ‘ASD’, ‘Solid Cancer’ and ‘Others’. The ‘Solid Cancer’ group and ‘Non-Cancer group’ (composed of patients older than 40 years from other groups) underwent a nested study and analysis using the virtual cancer panel. The Replication Cohort UK Biobank validated the nested study in 217 NF1 patients from the UK Biobank, similarly divided into Solid Cancer (n = 71) and Non-Cancer (n = 146) group. Arrows indicate the study workflow and relationships between groups. ASD  Autism.