Genetic Spectrum of Non‐PTPN11 Variants in Noonan Syndrome and Related RASopathies: Findings From a Russian Cohort

ABSTRACT

Noonan syndrome (NS) and related RASopathies form a clinically and genetically heterogeneous group of disorders caused by germline dysregulation of the RAS/MAPK pathway. Pathogenic variants in PTPN11 account for roughly half of classical NS, but the contribution of non‐PTPN11 genes in patients referred with NS or Noonan‐like phenotypes remains incompletely defined in several populations, including Russia. We characterized the spectrum of non‐PTPN11 variants in a large Russian cohort referred for suspected NS or Noonan‐like phenotypes and examined gene‐specific clinical patterns. We analyzed 456 unrelated patients with clinically diagnosed NS or Noonan‐related syndromes using a targeted panel of 23 RAS/MAPK pathway genes. PTPN11 variants (23% of cases) were reported previously; here we focused on the remaining patients. Pathogenic or likely pathogenic variants were identified in 19% of cases (85/456), with NF1, SOS1, SHOC2, and BRAF being the most frequently mutated genes, together accounting for 50% of the non‐PTPN11 diagnoses. Novel variants were observed in 12.1% of cases (7/58 unique variants). Several genotype–phenotype associations emerged: SOS1 mutations were associated with a high prevalence of cardiac defects, especially pulmonary stenosis and atrial septal defects (64% of patients, p = 0.01), whereas NF1 and SPRED1 cases had significantly fewer cardiac anomalies (p < 0.001 and p < 0.05, respectively). Lentigines were predominantly seen in NF1 and SPRED1 patients (p < 0.01), and all three evaluable males with MAP2K1 variants had cryptorchidism (3/3). Conversely, BRAF‐mutated patients showed lower incidence of chest deformity (p < 0.01) but higher rates of developmental delay. NF1, SOS1, SHOC2, and BRAF together accounted for half of the non‐PTPN11 molecular diagnoses in this Russian cohort, and the observed gene‐specific clinical patterns align with previous series, although limited subgroup sizes preclude firm individual associations.

Noonan syndrome and related conditions are caused by variants in multiple genes. We analyzed 456 Russian patients using a 23‐gene panel and found disease‐causing variants in non‐PTPN11 genes in 85 cases. NF1, SOS1, BRAF, and SHOC2 explained half of these diagnoses.

1. Introduction

RASopathies are a group of conditions caused by mutations in genes involved in the RAS‐MAPK signaling pathway. This pathway plays a critical role in cell growth, differentiation, survival, and metabolism in response to extracellular ligands such as growth factors, cytokines, and hormones, which explains the broad range of abnormalities seen in these disorders. Due to the shared disruption of the Ras/MAPK signaling pathway, RASopathies display many overlapping phenotypic characteristics [1].

Noonan syndrome (NS) is one of the most common RASopathies, with a prevalence of 1 in 1000–2500 live births. It is primarily characterized by distinctive facial features, congenital heart defects, hypertrophic cardiomyopathy, skeletal, ectodermal, and hematologic anomalies, and developmental delays [2]. In addition to NS, several related conditions fall under the RASopathy spectrum. For example, LEOPARD syndrome (LS, also known as NS with multiple lentigines) is an autosomal dominant disorder marked by multiple lentigines and café‐au‐lait spots, conduction abnormalities on electrocardiograms, ocular hypertelorism, obstructive cardiomyopathy, pulmonary stenosis, genital abnormalities in males, growth retardation, and hearing loss. LS shares several features with NS, although lentigines and hearing impairment are typically absent in NS [3].

Other Noonan‐like conditions include Costello syndrome (CS) and cardiofaciocutaneous syndrome (CFC), which are also caused by mutations in genes regulating the RAS/MAPK pathway. These conditions manifest with similar developmental and cardiac abnormalities, including distinctive facial features, heart defects (such as hypertrophic cardiomyopathy), growth delays, and varying degrees of intellectual disability. Both syndromes share overlapping phenotypes with NS but also exhibit unique clinical characteristics, such as more pronounced skin anomalies in CFC and a higher risk of malignancies in Costello syndrome [4, 5].

Mutations in more than 20 genes are associated with RASopathies, with PTPN11 being the most common, accounting for around 50% of NS cases [6] [7]. Other frequently mutated genes include SOS1, RAF1, KRAS, and RIT1, each contributing to specific clinical features like hypertrophic cardiomyopathy (RAF1) and milder growth delays (SOS1). Additionally, genes such as BRAF, MAP2K1, and MAP2K2 are linked to more severe forms like cardiofaciocutaneous syndrome. Although NF1‐related disease (neurofibromatosis type 1) is conventionally considered a distinct clinical entity, it shares the same RAS/MAPK pathway as the classical RASopathies and is therefore regularly considered in the differential diagnosis of patients presenting with café‐au‐lait macules, neurofibromas, or other features overlapping with NS. Genetic testing, including techniques like whole‐exome sequencing and targeted gene panels, is crucial in confirming a diagnosis of RASopathies and understanding the specific genotype–phenotype correlations [1].

Here we analyze a Russian cohort referred for suspected NS or a Noonan‐like phenotype, encompassing NS and related RASopathies—including Costello syndrome, cardiofaciocutaneous syndrome, NS with multiple lentigines (LEOPARD syndrome), NS with loose anagen hair, and patients in whom neurofibromatosis type 1 or Legius syndrome was clinically considered—and describe their genetic and clinical spectrum. Notably, the same cohort of patients was previously described, focusing specifically on mutations in the PTPN11 gene [8]. In this work, we aim to expand the analysis to include a broader spectrum of genetic mutations, providing further insight into the phenotypic and genotypic variability observed in these patients.

2. Methods

2.1. Clinical Data

In total, 456 unrelated patients referred with a clinical suspicion of NS or a Noonan‐like phenotype, encompassing NS, and related RASopathies, were included. In a previous study based on the same cohort, we focused on pathogenic variants in the PTPN11 gene, identifying such variants in 106 cases [8]. For the current research, we expanded our analysis to investigate variants in the remaining genes included in our gene panel. Specifically, we examined the 350 patients who did not harbor pathogenic mutations in PTPN11.

Clinical examination data and diagnoses were available for most patients. Most of the patients were diagnosed at the Research and Counseling Department of the Research Centre for Medical Genetics (RCMG), including patients referred from various genetic counseling departments in Russia.

The study was conducted in accordance with the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of the RCMG (approval number 4/1, dated April 19, 2021). All patients provided written informed consent. The experiments were performed following institutional guidelines.

2.2. Genetic Testing

This research utilizes a NS gene panel that includes the following genes: PTPN11, NRAS, RIT1, SHOC2, CBL, HRAS, KRAS, A2ML1, SOS2, SPRED1, MAP2K1, MAP2K2, PPP1R13L, SOS1, WDR35, LZTR1, RAF1, RASA2, IFT80, NEK1, RASA1, BRAF, and NF1. The panel was designed to cover all genes with an established or candidate role in NS and related RASopathies at the time of panel construction, including core NS/CFC/CS/NSML/Legius genes (PTPN11, SOS1, SOS2, KRAS, NRAS, HRAS, BRAF, RAF1, MAP2K1, MAP2K2, SHOC2, CBL, RIT1, LZTR1, A2ML1, SPRED1, NF1) and additional genes implicated in overlapping syndromes or Noonan‐like phenotypes (RASA1, RASA2, PPP1R13L, WDR35, IFT80, NEK1). The panel was designed to cover both the canonical NS/CFC/CS/NSML/Legius genes and additional candidate genes implicated in Noonan‐like phenotypes available at the time of panel construction; this composition was inherited from our earlier study of the PTPN11‐positive subset of the same cohort [8]. Next‐generation sequencing (NGS) was performed on the Ion S5 sequencer (Thermo Fisher Scientific) using the Ion AmpliSeq Library Kit 2.0 according to the manufacturer's instructions. Coverage of the PTPN11 gene (exons and intron–exon junctions) was 98.39%, with a single exon not covered; less than 1% of the coding sequence consists of homopolymeric regions, as reported previously [8]. Equivalent coverage metrics for the remaining genes of the panel are not separately reported here; analyses were restricted to coding regions and canonical splice sites. Reads were processed with Torrent Suite and variants were called and annotated using the in‐house NGSData pipeline (http://ngs‐data.ru). Copy‐number variants were not assessed from the panel data; analyses were restricted to single‐nucleotide variants and small indels within coding regions and canonical splice sites. Variants were described against the following reference transcripts (RefSeq): PTPN11 NM_002834.5, SOS1 NM_005633.4, KRAS NM_004985.5, NRAS NM_002524.5, HRAS NM_005343.4, BRAF NM_004333.6, RAF1 NM_002880.4, MAP2K1 NM_002755.4, MAP2K2 NM_030662.4, SHOC2 NM_007373.4, CBL NM_005188.4, RIT1 NM_006912.6, LZTR1 NM_006767.4, SPRED1 NM_152594.3, and NF1 NM_000267.3. All variants are reported against the GRCh37/hg19 human reference genome. Population allele frequencies were assessed using gnomAD v.2.1.1 (GRCh37) and the RuExAc dataset (http://ngs‐data.ru/vcfdb/). Variants were classified according to the ACMG/AMP 2015 guidelines [9] with ClinGen specifications where available, supplemented by OMIM, HGMD Professional, ClinVar, and the published literature. Candidate causal variants were confirmed by Sanger sequencing on an ABI Prism 3100 instrument (Applied Biosystems) with forward and reverse primers, following the manufacturer's protocol; parental testing was performed when DNA was available, allowing de novo status to be established in a subset of cases (e.g., NF1 c.4555G>C, p.Ala1519Pro). Detailed methods are described in our previous publication on the PTPN11‐positive subset of the same cohort [8].

Statistical analysis: Categorical clinical features were compared between gene‐defined subgroups using gene‐versus‐rest comparisons, restricted to genes with at least three patients with available clinical data. The two‐tailed Fisher's exact test for 2 × 2 contingency tables was applied, as in our previous analysis of the PTPN11‐positive subset of this cohort [8]. Patients with missing data for a given feature were excluded from that comparison only. A p < 0.05 was considered statistically significant. Given the small size of several gene‐defined subgroups, no correction for multiple testing was applied. Cross‐cohort comparisons of overall mutation detection rates were performed using the same test.

3. Results

Out of the total cohort of 456 patients referred for genetic testing with a diagnosis of NS and related disorders, pathogenic and likely pathogenic variants were identified in 191 patients, while variants of uncertain significance (VUS) were detected in 14 cases. Of these, pathogenic and likely pathogenic variants in the PTPN11 gene were identified in 106 cases, which represent 55.5% of the identified mutations [8].

In the remaining cohort of 350 non‐PTPN11 patients, pathogenic and likely pathogenic variants in other NS‐associated genes were identified in 85 cases (44.5% of the identified cases). Additionally, 13 variants were classified as VUS (variant of uncertain significance). For variant c.4555G>C (p.Ala1519Pro) in the NF1 gene, a family analysis was conducted, confirming the relationship between the proband and parents and establishing the de novo origin of the identified variant.

In the cohort of 85 patients with non‐PTPN11 mutations, a total of 58 pathogenic/likely pathogenic variants were identified (see Table 1). All identified mutations were dominant. In total, 7 of the 58 identified variants (12.1%) had not been previously described. Additionally, 3 variants were not reported in publications but were present in ClinVar with a pathogenic annotation. Eleven variants were identified in more than one unrelated family. All the repeated variants had been previously described: c.770A>G, c.1495A>G, and c.1799T>G in the BRAF gene [10, 11]; c.34G>A in the HRAS gene [12]; c.108A>G in the KRAS gene; c.389A>G in the MAP2K1 gene [13]; c.770C>T in the RAF1 gene [13]; c.170C>G in the RIT1 gene [14]; c.4A>G in the SHOC2 gene [15]; c.1654A>G and c.1655G>A in the SOS1 gene [16].

TABLE 1.

Pathogenic and likely pathogenic variants in non‐PTPN11 genes identified in Russian patients with Noonan and Noonan‐like phenotypes.

Gene	Position in cDNA	Exon/Intron	Position in the protein	Number of chromosomes (patients)	Frequency by GnomAD v.2.1.1	HGMD ID/ClinVar ID	Variant pathogenicity (ACMG criteria for previously undescribed variants)	Suspected diagnosis	RefSeq transcript
BRAF	c.770A>G	6	p.(Gln257Arg)	5	—	CM060874	P	3—NS; 2—NS/CS	NM_004333.6
BRAF	c.1495A>G	12	p.(Lys499Glu)	2	—	CM060873	P	2—NS	NM_004333.6
BRAF	c.785A>G	6	p.(Gln262Arg)	1	—	CM076029	P	1—NS	NM_004333.6
BRAF	c.1787G>T	15	p.(Gly596Val)	1	—	CM060877	P	1—CS	NM_004333.6
BRAF	c.1741A>G	14	p.(Asn581Asp)	1	—	CM060878	P	1—NS	NM_004333.6
BRAF	c.1799T>G	15	p.(Val600Gly)	2	—	CM112509	LP (PM2 PM5 PP3 PP5)	1—NS; 1—LP	NM_004333.6
BRAF	c.736G>C	6	p.(Ala246Pro)	1	—	CM060871	P	1—NS	NM_004333.6
BRAF	c.735A>T	6	p.(Leu245Phe)	1	—	CM092081	P	1—NS	NM_004333.6
CBL	c.1111T>A	8	p.(Tyr371Asn)	1	—	CM1710603	P	1—NS	NM_005188.4
HRAS	c.34G>A	2	p.(Gly12Ser)	2	—	CM053283	P	1—NS; 1—NS/CS	NM_005343.4
KRAS	c.108A>G	2	p.(Ile36Met)	2	—	CM070965	P	2—NS	NM_004985.5
KRAS	c.178G>C	3	p.(Gly60Arg)	1	—	CM061080	P	1—NS/CS	NM_004985.5
KRAS	c.458A>T	5	p.(Asp153Val)	1	—	CM061083	P	1—NS/CFC	NM_004985.5
KRAS	c.13A>G	2	p.(Lys5Glu)	1	—	CM073168	P	1—GR	NM_004985.5
LZTR1	c.1356dupA	13	p.(Glu453GlyfsTer216)	1	—	This study (SUB16240513 [SCV pending])	LP (PVS1, PM2)	1—NS	NM_006767.4
LZTR1	c.742G>A	8	p.(Gly248Arg)	1	—	CM154934	P	1—NS/CS	NM_006767.4
LZTR1	c.848G>A	9	p.(Arg283Gln)	1	—	CM191148	P	1—NS	NM_006767.4
MAP2K1	c.389A>G	3	p.(Tyr130Cys)	2	0.000003977	CM061104	P	1—CS; 1—GR	NM_002755.4
MAP2K1	c.199G>A	2	p.(Asp67Asn)	1	—	CM076269	P	1—NS	NM_002755.4
MAP2K1	c.124C>T	2	p.(Leu42Phe)	1	—	CM092997	P	1—NS	NM_002755.4
NF1	c.7127‐1G>C	int 47	splice	1	—	CS1822693	P	1—NS/NF	NM_000267.3
NF1	c.3047_3048delGT	23	p.(Cys1016fs)	1	—	CD063559	P	1—NF/LP	NM_000267.3
NF1	c.1165_1180del15ins16	10	p.(389_395del5ins5fs*1)	1	—	This study (SUB16240513 [SCV pending])	LP (PVS1, PM2)	1—NF	NM_000267.3
NF1	c.4891_4892insAT	36	p.(Asn1631fs*46)	1	—	This study (SUB16240513 [SCV pending])	LP (PVS1, PM2)	1—NF	NM_000267.3
NF1	c.4514delG	33	p.(Arg1505fs)	1	—	SCV002640138	LP (PVS1, PM2)	1—NS	NM_000267.3
NF1	c.3697delT	27	p.(Cys1233fs*7)	1	—	This study (SUB16240513 [SCV pending])	LP (PVS1, PM2)	1—NF	NM_000267.3
NF1	c.5425C>T	37	p.(Arg1809Cys)	1	—	CM077948	P	1—NF	NM_000267.3
NF1	c.1395delT	13	p.(Thr467Hisfs*6)	1	—	This study (SUB16240513 [SCV pending])	LP (PVS1, PM2)	1—NF	NM_000267.3
NF1	c.5205+1G>C	Int 36	splice	1	—	CS2126081	LP (PVS1, PM2, PM5)	1—NS/NF	NM_000267.3
NF1	c.3295A>G	25	p.(Lys1099Glu)	1	—	CM143373	LP (PM2 PM5 PP3 PP5)	1—NS	NM_000267.3
NF1	c.1039C>T	9	p.(Gln347*)	1	—	CM1815865	P	1—NS/NF	NM_000267.3
NF1	c.1185+2T>C	int 10	splice	1	—	SCV001206729	LP (PVS1, PM2)	1—NS/NF	NM_000267.3
NF1	c.4981T>C	36	p.(Cys1661Arg)	1	—	CM137130	P	1—LP	NM_000267.3
NF1	c.4555G>C	34	p.(Ala1519Pro)	1	—	CM143410	LP (PS2 PM2 PP3 PP5)	1—NS	NM_000267.3
NF1	c.5991T>A	40	p.(Ser1997Arg)	1	—	This study (SUB16240513 [SCV pending])	LP (PM2 PM5 PP1 PP2 PP3)	1—NF	NM_000267.3
NRAS	c.182A>G	3	p.(Gln61Arg)	1	—	SCV000490986, SCV000148627	P	1—NCM	NM_002524.5
RAF1	c.770C>T	7	p.(Ser257Leu)	2	—	CM073301	P	2—NS	NM_002880.4
RAF1	c.781C>T	7	p.(Pro261Ser)	1	—	CM073293	P	1—NS	NM_002880.4
RAF1	c.1082G>C	10	p.(Gly361Ala)	1	—	CM1714874	P	1—MCA	NM_002880.4
RIT1	c.246 T>G	5	p.(Phe82Leu)	1	—	CM0136416	P	1—NS	NM_006912.6
RIT1	c.170C>G	4	p.(Ala57Gly)	2	—	CM136413	P	2—NS	NM_006912.6
RIT1	c.244 T>G	5	p.(Phe82Val)	1	—	CM136415	P	1—NS	NM_006912.6
RIT1	c.246 T>A	5	p.(Phe82Leu)	1	—	CM2016871	LP (PM2 PM5 PP3 PP5)	1—CS	NM_006912.6
RIT1	c.284G>C	5	p.(Gly95Ala)	1	—	CM136420	P	1—NS	NM_006912.6
SHOC2	c.4A>G	2	p.(Ser2Gly)	13	0.00003186	CM095445	P	6—NS; 4—NLS with loose anagen hair; 1—CS; 1—CFC; 1—GR	NM_007373.4
SOS1	c.806 T>G	6	p.(Met269Arg)	1	—	CM070275	P	1—NS	NM_005633.4
SOS1	c.1654A>G	10	p.(Arg552Gly)	4	0.000003985	CM070269	P	3—NS; 1—NS/LP	NM_005633.4
SOS1	c.1655G>A	10	p.(Arg552Lys)	2	—	CM070273	P	2—NS	NM_005633.4
SOS1	c.1656G>C	10	p.(Arg552Ser)	1	н/д	CM070274	P	1—NS	NM_005633.4
SOS1	c.1294 T>C	10	p.(Trp432Arg)	1	н/д	CM070287	P	1 – NS/NF	NM_005633.4
SOS1	c.2536G>A	16	p.(Glu846Lys)	1	н/д	CM070272	P	1—NS	NM_005633.4
SOS1	c.1310 T>C	10	p.(Ile437Thr)	1	н/д	CM116029	P	1—NS	NM_005633.4
SOS1	c.2104 T>C	13	p.(Tyr702His)	1	0.000003985	CM070278	P	1—NS	NM_005633.4
SOS1	c.1655G>C	10	p.(Arg552Thr)	1	н/д	CM095735	P	1—NS	NM_005633.4
SOS1	c.508A>G	4	p.(Lys170Glu)	1	н/д	CM086966	P	1—CS/NLS with loose anagen hair	NM_005633.4
SPRED1	c.421C>T	5	p.(Gln141Ter)	1	—	SCV000549916	LP (PVS1, PM2)	1—NS	NM_152594.3
SPRED1	c.841_842del	7	p.(Gln281ValfsTer3)	1	—	This study	LP (PVS1, PM2)	1—NF	NM_152594.3
SPRED1	c.46C>T	2	p.(Arg16Ter)	1	—	CM093927	P	1—NS	NM_152594.3

Abbreviations: CFC, Cardio‐facio‐cutaneous; CM, classification in the HGMD Prof v. database (preferred); CS, Costello syndrome; GR, Growth retardation; LP, LEOPARD; LP, Likely Pathogenic; MCA, Multiple congenital malformations, NLS with loose anagen hair—Noonan‐like with loose anagen hair; NCM, Neurocutaneous melanosis; NF, Neurofibromatosis; NS, Noonan syndrome; P, Pathogenic; SCV, classification in the ClinVar database if the variant is not described in HGMD.

Variants were identified in the following genes: NF1 in 15 cases, SOS1 and BRAF in 14 cases each, SHOC2 in 13 cases, RIT1 in 6 cases, KRAS in 5 cases, MAP2K1 in 4 cases, RAF1 in 4 cases, SPRED1 and LZTR1 in 3 cases each. HRAS mutations were observed in 2 cases, while CBL and NRAS were found in 1 case each (Figure 1).

FIGURE 1.

Distribution of (likely) pathogenic variants by gene across the cohort of 456 unrelated patients referred for suspected NS or a Noonan‐like phenotype. The figure includes 106 previously reported PTPN11‐positive cases [8] and the 85 non‐PTPN11 molecular diagnoses identified in the present study; percentages are calculated over the total cohort (n = 456), and the PTPN11‐positive, non‐PTPN11 diagnosed, and genetically unsolved fractions are indicated separately.

The novel variants identified in this study, as well as those previously described in a single publication, were categorized according to the guidelines of ACMG (Table 1). All variants classified as variants of uncertain significance (VUS) are presented in Table S1.

Of the total 85 probands, 46 were male and 39 were female. Clinical examination data were available for 78 probands carrying pathogenic non‐PTPN11 variants. The clinical diagnoses listed at referral included NS, NS with multiple lentigines (LEOPARD syndrome), Costello syndrome, cardiofaciocutaneous syndrome, neurofibromatosis Type 1, NS with loose anagen hair, growth retardation, and multiple congenital malformations; these reflect the requesting clinician's working impression at testing and were not reassigned after molecular diagnosis. Among these 78 probands, 50 (64.1%) had congenital heart defects, including pulmonary valve stenosis (PVS, 17 cases), patent foramen ovale (PFO, 13 cases), atrial septal defect (ASD, 10 cases), pulmonary artery stenosis (PAS, 8 cases), ventricular septal defect (VSD, 6 cases), patent ductus arteriosus (PDA, 6 cases), supravalvular pulmonary stenosis (SVPS, 5 cases), and aortic regurgitation (AR, 3 cases), with other symptoms occurring in individual and double cases (Table 2; low‐frequency cardiac subtypes observed in only one or two probands are listed in Table S2). Additionally, cardiomyopathy was identified in three patients, further highlighting the range of cardiac involvement within this cohort.

TABLE 2.

Clinical features by gene in non‐PTPN11 probands with available clinical data (n = 78 of 85).

Feature	BRAF	SOS1	NF1	SHOC2	RIT1	KRAS	MAP2K1	RAF1	SPRED1	LZTR1	HRAS	CBL	NRAS
	(n = 13)	(n = 14)	(n = 12)	(n = 12)	(n = 6)	(n = 4)	(n = 4)	(n = 4)	(n = 3)	(n = 2)	(n = 2)	(n = 1)	(n = 1)
Male:Female	5:8	8:6	8:4	6:6	3:3	3:1	3:1	2:2	2:1	0:2	1:1	1:0	1:0
Cardiac	9/13 (69.2%)	13/14 (92.9%)	2/12 (16.7%)	8/12 (66.7%)	5/6	4/4	2/4	3/4	0/3	1/2	2/2	1/1	0/1
Cardiomyopathy	0/13 (0%)	0/14 (0%)	0/12 (0%)	1/12 (8.3%)	0/6	1/4	0/4	1/4	0/3	0/2	0/2	0/1	0/1
Pulmonary valve stenosis	4/13 (30.8%)	5/14 (35.7%)	0/12 (0%)	4/12 (33.3%)	0/6	1/4	0/4	1/4	0/3	0/2	1/2	1/2	0/1
Supravalvular pulmonary stenosis	2/13 (15.4%)	0/14 (0%)	0/12 (0%)	0/12 (0%)	1/6	1/4	0/4	1/4	0/3	0/2	0/2	0/1	0/1
Atrial septal defect	1/13 (7.7%)	4/14 (28.6%)	1/12 (8.3%)	1/12 (8.3%)	1/6	0/4	0/4	2/4	0/3	0/2	0/2	0/1	0/1
Ventricular septal defect	0/13 (0%)	2/14 (14.3%)	0/12 (0%)	2/12 (16.7%)	1/6	0/4	0/4	0/4	0/3	1/2	0/2	0/1	0/1
Patent foramen ovale	2/13 (15.4%)	2/14 (14.3%)	0/12 (0%)	2/12 (16.7%)	2/6	1/4	2/4	0/4 (0.0%)	0/3	0/2	1/2	1/2	0/1
Patent ductus arteriosus	1/13 (7.7%)	1/14 (7.1%)	1/12 (8.3%)	1/12 (8.3%)	0/6	1/4	1/4	0/4 (0.0%)	0/3	0/2	0/2	0/1	0/1
Aortic regurgitation	0/13 (0%)	2/14 (14.3%)	0/12 (0%)	1/12 (8.3%)	0/6	0/4	0/4	0/4 (0.0%)	0/3	0/2	0/2	0/1	0/1
Pulmonary artery Stenosis	1/13 (7.7%)	3/14 (21.4%)	0/12 (0%)	1/12 (8.3%)	1/6	2/4	0/4	0/4 (0.0%)	0/3	0/2	0/2	0/1	0/1
Facial dysmorphism	11/13 (84.6%)	13/14 (92.9%)	9/12 (75.0%)	10/12 (83.3%)	6/6	3/4	4/4	2/4	1/3	2/2	2/2	1/1	0/1
Short neck	6/13 (46.2%)	8/14 (57.1%)	6/12 (50.0%)	9/12 (75.0%)	5/6	3/4	2/4	1/4	1/3	1/2	0/2	1/1	0/1
Lentigines	2/13 (15.4%)	0/14 (0.0%)	8/12 (66.7%)	2/12 (16.7%)	0/6	0/4	0/4	0/4	3/3	0/2	0/2	0/1	0/1
Shield chest	1/13 (7.7%)	8/14 (57.1%)	4/12 (33.3%)	7/12 (58.3%)	3/6	2/4	3/4	2/4	0/3	0/2	1/2	0/1	0/1
Short stature (< 3rd percentile)	3/13 (23.1%)	3/14 (21.4%)	3/12 (25.0%)	5/12 (41.7%)	2/6	1/4	0/4	1/4	1/3	0/2	0/2	1/1	0/1
Global developmental delay or intellectual disability	9/13 (69.2%)	4/14 (28.6%)	4/12 (33.3%)	7/12 (58.3%)	3/6	3/4	4/4	2/4	1/3	1/2	0/2	0/1	0/1
Cryptorchidism	1/5 (20.0%)	4/8 (50.0%)	0/8 (0.0%)	1/6 (16.7%)	2/3	2/3	3/3	0/4 (0.0%)	0/2	0/0	0/1	0/1	0/1
Eye diseases	3/13 (23.1%)	0/14 (0.0%)	0/12 (0.0%)	4/12 (33.3%)	2/6	2/4	1/4	1/4	1/3	0/2	0/2	0/1	0/1
Hearing loss	1/13 (7.7%)	1/14 (7.1%)	0/12 (0.0%)	0/12 (0.0%)	0/6	0/4	0/4	0/4	0/3	0/2	0/2	0/1	0/1
Kidney disease	2/13 (15.4%)	5/14 (35.7%)	3/12 (25.0%)	0/12 (0.0%)	1/6	0/4	1/4	0/4	0/3	0/2	1/2	0/1	0/1
Neurofibromas	0/13 (0.0%)	1/14 (7.1%)	2/12 (16.7%)	0/12 (0.0%)	0/6	0/4	0/4	0/4	0/3	0/2	0/2	0/1	0/1

Note: For each gene, n is the number of clinically characterized patients. Group comparisons were performed using two‐tailed Fisher's exact test in gene‐versus‐rest comparisons, restricted to genes with n ≥ 3, without correction for multiple testing. Low‐frequency cardiac subtypes observed in only one or two probands are listed in Table S2.

Among the 78 probands with available clinical data, 64 (82.1%) exhibited facial dysmorphism, and 43 (55.1%) presented with a short neck. Lentigines were observed in 15 cases (19.2%). Growth below the 3rd percentile was noted in 20 probands (25.6%), and a shield chest was found in 31 cases (39.7%). Global developmental delay or intellectual disability was recorded in 38 probands (48.7%), and cryptorchidism in 13 males (30.2%). Eye anomalies were reported in 14 cases (17.9%), and kidney disease in 13 (16.7%). Hearing loss was confirmed in 2 individuals (2.6%), while neurofibromas and skin spots were present in 3 (3.8%) and 14 cases (17.9%), respectively.

The clinical characteristics of probands and their association with genotype are detailed in Table S1.

4. Discussion

NS is recognized as one of the most genetically heterogeneous developmental disorders, encompassing a wide range of phenotypic manifestations driven by mutations in multiple genes. To date, limited data exist regarding the prevalence and spectrum of NS‐associated genetic variants except PTPN11 mutations among Russian patients. The current study was the first to analyze the frequency of various non‐PTPN11‐related Noonan genes in a large cohort of Russian patients.

Genetic testing using a targeted gene panel, which included 23 genes, was completed on 456 unrelated patients with Noonan‐like syndromes. Previously, we investigated the mutation spectrum and the frequency of PTPN11 variants among these patients. According to the study by Orlova et al., the frequency of PTPN11 mutations was 23% (106/456). In this study, we examined the remaining genes in the same patient cohort. As a result, causative variants were identified in 19% of cases (85/456). Thus, including PTPN11, pathogenic variants were identified in 41.9% of the entire cohort of 456 cases.

Comparative analysis with other international cohorts is inherently challenging due to the differences in the gene panels used, which include varying sets of genes across studies. In this analysis, we excluded genes present in our cohort that were not studied in other cohorts and omitted genes found in other studies that were not included in our panel. The final analysis was focused on the following genes: PTPN11, SOS1, BRAF, SHOC2, KRAS, MAP2K1, RAF1, HRAS, CBL, NRAS, and MAP2K2. The mutation rate in our cohort was 35.96%. When comparing this rate to other international cohorts, significant differences were observed. Specifically, the mutation detection rate in Brazil was 66.53% (p < 0.001), and in Japan, it was 75.00% (p < 0.001), both significantly higher than in our cohort [17, 18]. No significant differences were found when comparing our cohort to Greece (43.02%, p = 0.214) or Italy (43.75%, p = 0.184) [19, 20]. Direct comparison across cohorts is limited by differences in inclusion criteria, gene‐panel composition, and case mix between studies. The results emphasize the need for standardized gene panels to facilitate consistent comparisons and a comprehensive understanding of the genetic landscape of Noonan‐like syndromes across different populations.

In the present study, NF1 was identified as the most frequently mutated gene, with variants detected in 15 cases, underscoring its significant role in the cohort. Although NF1 is often not considered in the context of Noonan‐like disorders, its inclusion may be justified due to overlapping clinical features. SOS1 and BRAF mutations followed, each identified in 14 cases. SHOC2 was also frequently mutated, with 13 cases showing variants. Mutations in RIT1 (6 cases), KRAS (5 cases), MAP2K1 (4 cases), and RAF1 (4 cases) were noted as well, contributing to the spectrum of observed genetic variations. Additionally, SPRED1 and LZTR1 were found in 3 cases each, while rare mutations included HRAS (2 cases), as well as CBL and NRAS, with one case each. Data from previous studies have confirmed that, in addition to PTPN11, SOS1 is also one of the most frequently mutated genes in NS patient cohorts. The frequency of BRAF mutations in our cohort is higher than the less than 2% typically reported for patients with NS and Noonan‐like disorders. The frequency of RAF1 mutations in our study is lower than the 10%–30% reported for NS cases without mutations in PTPN11, KRAS, and SOS1 [21]. These observations highlight the genetic variability within our cohort compared to reported data.

Recurrent variants and gene‐specific peculiarities were observed in eight genes—BRAF, HRAS, KRAS, MAP2K1, RAF1, RIT1, SHOC2, and SOS1. The pathogenic variant c.770A>G (p.(Gln257Arg)) was observed in 36% (5/14) of mutant alleles of the BRAF gene, and the variants c.1495A>G (p.(Lys499Glu)) and c.1799T>G (p.(Val600Gly)) were observed in 14% (1/14) of mutant alleles of this gene. This is consistent with data from Tartaglia et al., who analyzed a cohort of patients from various European countries diagnosed with NS and related disorders, reporting a frequency of 40% for the c.770A>G (p.(Gln257Arg)) variant in the BRAF gene [21]. In HRAS, the c.34G>A (p.(Gly12Ser)) variant was identified in all pathogenic alleles (2/2). This finding aligns with literature data, where the vast majority of these mutations affect residues Gly12 and Gly13, known hot spots for oncogenic somatic mutations [22]. In KRAS, the c.108A>G (p.Ile36Met) mutation appeared in 2 out of 5 pathogenic alleles, while the c.389A>G (p.(Tyr130Cys)) variant in MAP2K1 and the c.770C>T (p.(Ser257Leu)) variant in RAF1 were present in half of the pathogenic alleles (2/4). This corresponds with literature data showing that mutations in the N‐terminal 14‐3‐3 recognition sequence, including Ser257, are among the more frequently observed RAF1 defects [21]. The c.170C>G (p.(Ala57Gly)) mutation in RIT1 was found in 2 of 6 pathogenic alleles. In the SHOC2 gene, the c.4A>G (p.(Ser2Gly)) variant was the most frequent, found on 100% of pathogenic alleles (13/13). This aligns with previous reports indicating that most patients with NS/LAH screened so far share this change, which promotes N‐myristoylation of the protein and causes aberrant targeting of SHOC2 to the plasma membrane and impaired translocation to the nucleus upon growth factor stimulation [15]. Finally, in the SOS1 gene, the c.1654A>G (p.(Arg552Gly)) and c.1655G>A (p.(Arg552Lys)) mutations were identified on 28.6% (4/14) and 14.3% (2/14) of pathogenic alleles, respectively. This observation is consistent with the previous studies showing that substitutions at residue Arg552 account for approximately 30% of total SOS1 mutations [21].

Differentiating NS from related RASopathies and Noonan‐like conditions can be challenging because of overlapping clinical features. In our cohort, the referral diagnosis at testing did not always correspond to the syndrome typically associated with the gene in which a pathogenic variant was identified—for example, NF1‐positive cases were most often referred as neurofibromatosis, whereas SHOC2‐positive cases were referred as Noonan or Noonan‐like with loose anagen hair (Table 1). In our cohort of patients without PTPN11 mutations, most were diagnosed with NS, illustrating its broad diagnostic application (Table 1). As expected, patients with NF1 mutations were more frequently diagnosed with “Neurofibromatosis,” reflecting its well‐known genetic basis. Similarly, SHOC2 mutations were commonly associated with “Noonan/Noonan‐like with loose anagen hair,” consistent with existing data. For patients with KRAS mutations, while NS remained the predominant diagnosis, conditions such as “Costello syndrome” and mixed “Noonan/Costello” presentations were also observed, indicating broader phenotypic variability within this mutation group.

The main clinical findings in our cohort of probands included facial dysmorphism, cardiac anomalies, and short neck, all observed frequently. Global developmental delay or intellectual disability was also common, as was the presence of a shield chest. Growth below the 3rd percentile was notable, and cryptorchidism was observed in a significant number of male probands. Genotype–phenotype correlations revealed statistically significant differences in the presence of clinical manifestations among patients with different gene mutations.

Cardiac anomalies were more frequently observed in patients carrying SOS1 variants (13/14; p = 0.0140), with pulmonary valve stenosis or atrial septal defect in 64.3%, consistent with previous reports [23]. In contrast, cardiac involvement was less frequent in patients with NF1 (p = 0.0004) and SPRED1 (p = 0.0431). Dermatological findings showed that lentigines were significantly more common in NF1 (p < 0.0001) and SPRED1 (p = 0.0060) patients. Given that NF1 is known for such skin manifestations and SPRED1 mutations cause Legius syndrome, which mimics NF1, these results are not unexpected. Shield chest was less frequent in BRAF‐positive patients (1/13; p = 0.0117), and all three evaluable males with MAP2K1 variants had cryptorchidism (3/3; p = 0.0232). The analysis showed that Global developmental delay or intellectual disability was more common in BRAF (69.2%) and MAP2K1 (100%) compared to SOS1 (28.6%). This is supported by the findings of Pierpont et al. suggesting gene‐specific effects within the Ras‐MAPK pathway [7].

5. Limitations

Three points should be considered when interpreting our findings. First, the cohort was assembled from referrals for suspected NS or a Noonan‐like phenotype rather than from patients fulfilling strict consensus criteria for classical NS, and the relative gene frequencies should be read in that context. Second, copy‐number variants were not assessed from the panel data, which may underestimate the contribution of genes such as NF1 and SPRED1. Third, the number of clinically characterized patients per gene was small for several non‐core genes (n ≤ 4 for MAP2K1, RAF1, KRAS, SPRED1, LZTR1, HRAS, CBL, and NRAS), which limits the strength of gene‐specific clinical inferences for those genes.

6. Conclusion

In a cohort of 456 unrelated Russian patients referred for NS or a Noonan‐like phenotype, pathogenic and likely pathogenic non‐PTPN11 variants were identified in 85 cases (19%); together with the previously reported PTPN11‐positive cases, the overall molecular yield was 42%. The largest contributions came from NF1 (n = 15), SOS1 (n = 14), BRAF (n = 14), and SHOC2 (n = 13), which together accounted for half of the non‐PTPN11 molecular diagnoses; this distribution supports the diagnostic value of broad RAS/MAPK pathway gene panels in patients negative for first‐line PTPN11 testing. For these four genes the number of clinically characterized patients was sufficient to evaluate gene‐specific clinical patterns, and our observations agree with previously published series: enrichment of pulmonary valve stenosis and atrial septal defect in SOS1‐positive patients (13/14), the expected predominance of lentigines and the lower frequency of congenital heart defects in NF1‐positive patients (12/12), and a high frequency of the recurrent SHOC2 c.4A>G p.Ser2Gly allele in patients with a Noonan/loose anagen hair phenotype (13/13). The frequency of the recurrent BRAF c.770A>G p.Gln257Arg allele in our cohort (5/14, 36%) is in line with the ~40% reported by Tartaglia et al. [21]. The remaining genes—including MAP2K1, RAF1, KRAS, SPRED1, LZTR1, HRAS, CBL, NRAS, and RIT1—were each represented by a small number of clinically characterized patients in this cohort, and our findings for those genes are reported as case‐level observations consistent with the known clinical spectrum rather than as independent gene–phenotype associations.

Author Contributions

A.P. and O.R. conceived the project and designed the study. D.G. and N.D. collected the clinical data. A.O. conducted genetic testing. A.O. and O.S. analyzed the genomic data. O.S. and A.O. wrote the manuscript. All authors revised the manuscript and approved the final version.

Disclosure

ChatGPT (OpenAI) was not used to generate or interpret primary data, to perform variant classification or statistical analysis, or to draft the Methods, Results, or Tables. All AI‐assisted text was reviewed and verified by the authors, who take full responsibility for the content of the manuscript.

Ethics Statement

The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Ethics Committee of Research Centre for Medical Genetics (the approval number 4/1 from 19 April 2021).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Table S1: Clinical and molecular characteristics of all probands carrying pathogenic or likely pathogenic non‐PTPN11 variants identified in the cohort, together with patients carrying variants of uncertain significance (VUS).

Table S2: cge70211‐sup‐0002‐TableS2.xlsx.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request. Novel variants reported in this study have been submitted to ClinVar (submission ID SUB16240513); ClinVar SCV accession numbers will be added to Table 1 once assigned by ClinVar curators.