1 INTRODUCTION

RASopathies include a heterogeneous group of genetic multisystemic disorders characterized by distinctive facial features, developmental delay, learning difficulties, short stature, congenital heart diseases (CHDs), renal anomalies, lymphatic malformations, and bleeding disorders (Aoki, Niihori, Inoue, & Matsubara, 2016; Rauen, 2013; Romano et al., 2010; Tartaglia & Gelb, 2010). They include different syndromes. The most common disorders are Noonan syndrome (NS; OMIM #163950) CardioFacioCutaneous Syndrome (CFCS; OMIM #115150), Costello syndrome (CS; OMIM #218040), and NS with multiple lentigines (NSML; also known as LEOPARD syndrome; OMIM #151100). These syndromes overlap cardiac and extracardiac features and share germline mutations of the RAS/MAPK pathway.

Heart defects, both CHDs and cardiomyopathies (CMP), were reported in 60–90% of patients affected by RASopathies. Specific types of CHDs and correlation with the specific gene mutations have been described (Calcagni et al., 2017; Calcagni et al., 2018).

Typical CHDs (such as pulmonary valve stenosis—PVS, atrioventricular septal defect—AVSD, atrial septal defect—ASD) have been reported (Digilio et al., 2013; Linglart & Gelb, 2020; Marino, Digilio, Toscano, Giannotti, & Dallapiccola, 1999). Hypertrophic cardiomyopathy (HCM) in the obstructive or nonobstructive form, is the most common CMP associated with this group of disorders (Calcagni et al., 2018). Dilated CMP was reported as well, due to a pathological different mechanism in the RAS/MAPK cascade (Gelb, Roberts, & Tartaglia, 2015).

Recently, we reported a multicentric study on 371 patients followed up in the CARNET (CArdiac Rasopathies NETwork) database, including the interventional and surgical results, genotype–phenotype correlations together with the prognostic assessment (Calcagni et al., 2017a; Calcagni et al., 2017b). Atypical Cardiac Defects (ACDs) were reported also in these patients, but no specific cardiac outcome has been described.

The aim of the present study was to update the CARNET experience and to analyze the prevalence, gene correlations, and outcome in RASopathies' patients presenting with ACDs.

2 METHODS

3 RESULTS

3.1 Study population

Among the 440 molecularly characterized patients followed in CARNET centers, we excluded those who did not present any cardiac defect (83 cases). Among these remaining 357 patients, we then excluded 312 subjects presenting with common CHDs (PVS, ASD, and AVSD) or HCM. The remaining 45 patients were included in this study and analyzed in terms of type of CHD, mutated genes and cardiac outcome. NS, NSML and CFCS were present in 36, 5 and 4 patients, respectively. No CS patient was present in the remaining population. Figure 1 illustrates the cohort of patients and the subgroup with ACDs.

Median age at last follow-up was 20.1 years (range 6.9–47 years). Twenty patients were female (44.4%). Different gene mutations were found in our population, the most common of which affected the PTPN11 gene, followed by SOS1 and BRAF genes. Mutations in other genes of the RAS/MAPK cascade were rarer. Table 1 summarizes demographic, clinical and molecular characteristics of included patients.

Table 1. Demographic and molecular characteristics of patients

Age at last follow up (median and range)	20.1 years (6.9–47 years)
Female sex	20 (44.4%)
Syndrome (n = 45)
NS	36 (80.0%)
LS	5 (11.1%)
CFC	4 (8.9%)
Mutated genes (n = 45)
PTPN11	27 (60.0%)
SOS1	5 (11.1%)
BRAF	4 (8.9%)
SHOC2	2 (4.4%)
KRAS	1 (2.2%)
NRAS	1 (2.2%)
RAF1	1 (2.2%)
CBL	1 (2.2%)
LZTR1	1 (2.2%)
MEK1	1 (2.2%)
SOS2	1 (2.2%)

4 DISCUSSION

RASopathies are a heterogeneous group of genetic disorders with similar cardiac and extracardiac features.

Typically, mutations in genes involved in the RAS/MAPK cascade have been reported since the PTPN11 gene mutation was discovered in 2001 (Tartaglia et al., 2001). This mutation occurs in about 50% of patients with NS. An additional 30% can be explained by other gene mutations in the RAS/MAPK pathway genes, including SOS1, RAF1, RIT1, KRAS, SHOC2, NRAS, SOS2, BRAF, A2ML1, LZTR1, MYST4, RASA2, RRAS, SPRY1 and SYNGAP1 (Aoki et al., 2013; Tidyman & Rauen, 2016a; Tidyman & Rauen, 2016b; Yamamoto et al., 2015; Yu et al., 2019).

CHDs and CPM are very common features in patients with RASopathies, particularly in NS, in which a prevalence of 50–80% is reported (Linglart & Gelb, 2020).

Classically, PVS, HCM, AVCD, and ASD are considered distinct heart defects in these disorders. Minor defects have been also reported (Linglart & Gelb, 2020; Marino et al., 1999), but a specific correlation between these anomalies and gene mutations and their prognostic assessment are not yet available.

The present study reviews 45 RASopathy patients enrolled from seven centers with high experience in RAS/MAPK disorders. In general, ACDs present in these subjects are prognostically benign, well tolerated and do not required specific treatment. In fact, in our cohort, only five ACD patients needed surgical procedures, mainly due to valvular involvement.

These results are in line with a previous study from CARNET on a large cohort of children and adolescents with RASopathies, suggesting that the need for either cardiac surgery or catheterization, was less than 50% (Calcagni, Limongelli, et al., 2017a). Similar data were reported by other groups (Colquitt & Noonan, 2014; Prendiville et al., 2014). Thus, cardiac treatment could be restricted to a few selected cases, presenting with mitral or aortic valve diseases. While worsening of mitral and aortic valve disease has been reported in RASopathies patients with CHDs or HCM was previously reported (Calcagni, Limongelli, et al., 2017a), the present study argues that these valvular diseases may have a major clinical impact also in the absence of a coexisting typical heart defect. Mitral valve prolapse particularly with valve regurgitation may need surgical treatment in patients with severe valve dysplasia. Nevertheless, in RASopathy patients with significant valvular regurgitation, mitral valve plasty could be ineffective, due to anatomic characteristics of mitral valve, as an abnormal and dysplastic anterior leaflet insertion (Digilio, Marino, Giannotti, & Dallapiccola, 1997; Marino et al., 1995; Marino, Digilio, Gagliardi, Giannotti, & Dallapiccola, 1996). Therefore, isolated mitral or aortic valve disorders in these patients could not be regarded as minor defects.

Surgical treatment is also mandatory for selected cases manifesting other “structural” ACDs, such as aortic coarctation. Preliminary data preceding exhaustive gene screening, disclosed left sided obstructive defects in particular aortic coarctation in NS (Digilio et al., 1997; Digilio et al., 1998).

Coronary artery anomalies in RASopathies are not a rare entity. PTPN11 gene mutations are usually associated with these defects, particularly in NSML patients (Pacileo et al., 2006). Mutations in other genes have been also reported (Calcagni, Baban, De Luca, et al., 2016). Coronary ectasia secondary to HCM is well established, likely as a consequence of increased mass, diastolic dysfunction and obstruction of the outflow tract of the left ventricle (Limongelli et al., 2007). In the present cohort of patients, no individual with coronary artery ectasia was affected by HCM or other common cardiovascular defects. This vascular phenotype could be related to the RAS pathway defect itself, independently from an underlying HCM and/or cardiovascular defects. Cardiologists should be aware of the dilation of the coronary arteries in these patients, for preventing the risk factors of myocardial infarction, such as arterial hypertension and hypercholesterolemia that could accelerate the atherosclerotic coronary artery disease. Once coronary anomaly is recognized by echocardiography, further evaluation with coronary CT angiography or MR angiography should be undertaken, especially after pediatric age, in order to monitor the dilation's progression. Since the long-term outcome of coronary arteries ectasia in patients with RASopathy is unknown and that the dilated coronary arteries predispose to thrombosis, the use of antiplatelet or anticoagulant treatment must be considered also in these patients.

Two of our patients displayed dilation of the ascending aorta. Although rare, aortic root dilation, aortic dissection and giant aneurysms of the Valsalva sinuses, have been described in adult patients with PTPN11 gene mutations. Histological analysis of the surgically removed giant aneurysms has shown deficiency of the medial elastin and nonspecific myxoid degeneration, suggesting a connective tissue disorder-like change in the wall of aorta (Morgan, Coupe, Honey, & Miller, 1989; Power, Lewin, Hannibal, & Glass, 2006; Purnell, Williams, Von Oppell, & Wood, 2005; Shachter, Perloff, & Mulder, 1984). Aortic dilation may be unnoticed before the clinical manifestation of aortic dissection, thus recommending a strict long-term follow-up, notably interval aortic measurements (Power et al., 2006).

We found dilation of the left atrial appendage in three of our patients. Only one adult individual with features of CFCS and NS has been described, presenting with congenital aneurysm of the left atrial appendage (DeSena et al., 2010). Although RASopathy patients are predisposed to hematological disorders, such as abnormal bleeding or bruising caused by coagulation defects, they may also develop thrombi in the enlarged left atrial appendage or atrial fibrillation over the years. Hence, a careful evaluation of the coagulation system is mandatory whenever antiaggregant or anticoagulant therapy is demanded in these patients. In young patients with RASopathies and occurrence of cryptogenic stroke, recognition of an enlarged left atrial appendage may prompt to primary oral anticoagulation therapy.

Although only a few of our patients required surgical treatment, their lesions were benign in terms of morbidity. Furthermore, it is important to highlight these data to confirm the major role of echocardiography both for screening at the time of first diagnosis and during the clinical follow up of individuals with RASopathies, even in absence of CHDs or CMPs. No patients in our cohort died, arguing that despite the common cardiac involvement in these syndromes, their lesions are usually benign in terms of mortality. Nevertheless, accurate follow up based on clinical evaluation, ECG and echocardiogram is highly recommended, as previously suggested (Roberts, Allanson, Tartaglia, & Gelb, 2013; Romano et al., 2010; Wilkinson et al., 2012).

In agreement with published results, PTPN11 was the most commonly mutated gene also in ACDs individuals, followed by SOS1 and BRAF. These results overlap personal and published experience of mutated gene prevalence in patients with typical heart defects. Among the genes more recently associated with RASopathies (Aoki et al., 2013), we found one patient with RIT1 mutation. In agreement with personal and published results, this could suggest a higher penetrance for cardiac disease in these individuals (Calcagni, Baban, Lepri, et al., 2016; Kouz et al., 2016).

5 STUDY LIMITATIONS

The retrospective design of the study represents per se a limitation. Moreover, RASopathy patients in whom no gene mutations were detected, may have biased our results. Finally, another limitation concerns the absence of specific statistical analysis in terms of Kaplan–Mayer for mortality or morbidity. Since no patients died for cardiac nor extracardiac disorders, and because of limited number of patients undergoing reintervention, we were unable to extrapolate these results.

6 CONCLUSIONS

Patients with RASopathies need a lifetime cardiac follow-up, not only for monitoring the most common CHDs and HCM, but also for ACDs. No data regarding dilation CA or left atrial appendage prognosis have prospectively reported in this population. Therefore, these apparently minor cardiac conditions may require cardiac surgery or percutaneous procedures in patients with worsening of valvulopathies. Medical management aimed to prevent complications due to coronary artery dilation and enlarged left atrial appendage might be needed. With a deeper understanding of cardiac characteristics and genotype–phenotype correlations of patients with RASopathies, it will be possible to offer to families appropriate counseling in terms of prognosis and tailored management for optimal patient outcome.

CONFLICT OF INTEREST

The authors report no relationships that could be construed as a conflict of interest.

AUTHOR CONTRIBUTIONS

All the authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.