StudyRareStudyRare

RASopathies & RAS-MAPK Signaling

Log in to star

Last updated 2mo ago

Log in to add personal notes on this page.

The RAS-MAPK pathway anchors one of the most clinically central signaling families in genetics because germline activating mutations along its length produce a phenotypically overlapping family of conditions called the RASopathies. Together these are the most common single-gene cause of congenital heart disease (Noonan syndrome alone affects ~1/1,000–2,500 births). Recognizing that NF1, Noonan, Costello, CFC, and LEOPARD/Legius are all variations on a single mechanistic theme (germline activation of RAS-MAPK signaling, but tuned-down enough to be compatible with viable development) is one of the unifying concepts in dysmorphology.

The same pathway is the most commonly somatically-activated pathway in cancer (KRAS, BRAF, HRAS, NRAS mutations). The germline RASopathies are essentially "constitutive RAS-MAPK signaling, but mild enough to survive," which is why patients have an increased cancer risk.

Growth factor → RTK → GRB2 / SOS → RAS-GTP → RAF → MEK → ERK → transcription
                                       ↑
                             RAS-GAPs turn it OFF:
                             NF1 (neurofibromin), SPRED1

Key principles:

  • RAS (HRAS, KRAS, NRAS) is a small GTPase. Active when bound to GTP, inactive when bound to GDP. It's a binary on/off switch.
  • RAS-GAPs (GTPase-activating proteins) accelerate GTP hydrolysis, switching RAS off. NF1's neurofibromin is a RAS-GAP. SPRED1 (Legius syndrome) regulates neurofibromin.
  • RAS-GEFs (guanine-exchange factors) like SOS load GTP onto RAS, switching it on. SOS1 mutations cause Noonan.
  • PTPN11/SHP2 is a phosphatase upstream of RAS that amplifies RTK signaling. ~50% of Noonan cases have PTPN11 mutations.
  • The downstream cascade (RAF → MEK → ERK) is also a mutation hotspot. BRAF activates → CFC and Costello-like phenotypes.

A useful frame: anything that pushes the pathway "more on" (gain-of-function in RAS, RAF, MEK, SHP2, SOS, or loss of RAS-GAPs like NF1, SPRED1) produces a RASopathy.

SyndromeGene(s)MechanismCardinal features
Noonan syndromePTPN11 (~50%), SOS1, RAF1, KRAS, NRAS, RIT1, SOS2Gain-of-functionShort stature, pulmonic stenosis, HCM, broad/webbed neck, low-set rotated ears, pectus, downslanting palpebral fissures, lymphedema, mild ID variable
Noonan with multiple lentigines (LEOPARD)PTPN11 (different, dominant-negative variants), RAF1Loss of phosphatase activityLentigines, ECG conduction abnormalities, ocular hypertelorism, pulmonic stenosis, abnormal genitalia, retarded growth, deafness
Cardiofaciocutaneous (CFC)BRAF, MAP2K1 (MEK1), MAP2K2 (MEK2), KRASDownstream activationSevere ID, ectodermal abnormalities (sparse curly hair, hyperkeratosis), feeding difficulty, pulmonic stenosis. More severe than Noonan
Costello syndromeHRAS (de novo, paternal age effect)RAS gain-of-functionCoarse facies, papillomata around mouth/nose, cutis laxa, severe ID, increased malignancy risk (rhabdomyosarcoma, neuroblastoma)
Neurofibromatosis type 1 (NF1)NF1Loss-of-function (RAS-GAP)Café-au-lait macules, axillary/inguinal freckling, neurofibromas, optic glioma, Lisch nodules, sphenoid wing dysplasia, scoliosis, learning disability
Legius syndromeSPRED1Loss-of-functionCafé-au-lait macules + axillary freckling; looks like NF1 minus the tumors. ~2% of NF1-suspected patients
Capillary malformation-AVMRASA1RAS-GAP lossMultifocal capillary malformations + arteriovenous malformations
Mazzanti / Noonan-like with loose anagen hairSHOC2Pathway scaffoldSlow-growing easily plucked hair, ectodermal features, Noonan-like

Cardiac signature: pulmonic valve stenosis is the canonical RASopathy CHD (Noonan, CFC, LEOPARD). HCM is also common. Compare with Williams syndrome (ELN) which gives supravalvular aortic stenosis; pathway and gene are completely different even though both are valvular issues in syndromic kids.

The clinical features cluster because RAS-MAPK signaling is required for the same developmental processes across syndromes:

  • Cardiac valve and septal development → pulmonic stenosis, HCM, AV canal
  • Lymphatic development → lymphedema, cystic hygroma in fetal life (causing the broad neck appearance)
  • Bone growth and craniofacial morphogenesis → short stature, facial coarseness, pectus
  • Skin/hair (ectodermal) → café-au-lait, sparse curly hair (CFC), papillomata (Costello)
  • CNS development & learning → variable ID, learning disability
  • Hematopoiesis → juvenile myelomonocytic leukemia (JMML) susceptibility in Noonan with PTPN11 mutations

RAS-MAPK is the most-mutated pathway in human cancer. Germline RASopathy patients are at increased cancer risk because their cells already start with elevated baseline pathway activity:

  • Costello (HRAS): highest cancer risk. Rhabdomyosarcoma (1st 5 years of life), neuroblastoma, transitional cell carcinoma. Surveillance protocol with abdominal ultrasound + urinalysis.
  • Noonan with PTPN11: JMML, ALL, neuroblastoma. Can have transient myeloproliferative-like disorder in infancy.
  • NF1: optic pathway gliomas (10–20% in childhood, often indolent), MPNST (malignant peripheral nerve sheath tumor) in adulthood, breast cancer in women, pheochromocytoma, GIST.
  • CFC, LEOPARD: modestly increased risk; less clear surveillance protocol.

Germline mutations are tuned by selection: too much pathway activation is embryonic-lethal, which is why germline RAS mutations are missense (not truncating) and target specific residues.

  • Phenotypic recognition first: characteristic facies (especially Noonan's broad-tip/upturned-pointed nose, downslanting palpebral fissures, low-set rotated ears) is often more sensitive than genetic testing.
  • Multigene panel rather than single-gene testing: a Noonan/RASopathy panel covers ~80–95% of clinical cases.
  • Echocardiogram at diagnosis (pulmonic stenosis, HCM) and follow-up.
  • Cancer surveillance depends on specific gene (most rigorous for Costello and NF1).
  • NF1 is autosomal dominant with very high penetrance but variable expressivity. ~50% of cases are de novo (NF1 has the highest spontaneous mutation rate of any human gene, likely related to the gene's enormous size, ~280 kb).
  • Noonan is autosomal dominant; many cases are de novo, with paternal age effect for some genes.
  • Costello is essentially always de novo HRAS mutation with strong paternal age effect.
  • Recurrence after a de novo case is generally low (<1%) but parental mosaicism can elevate it; gonadal mosaicism makes "0% recurrence" counseling unsafe.

"NF1 ≈ Legius": same skin findings (CAL macules, freckling), opposite tumor risks. Test SPRED1 if NF1 panel is negative and only skin findings are present.

"Pulmonic = Noonan": pulmonic valve stenosis is the canonical RASopathy CHD. Compare with supravalvular aortic = Williams (different gene, ELN).

"HRAS = high-risk Costello": the worst tumor risk in the RASopathy family.

"Pathway thinking": when you see "this child has features of NF1 but no neurofibromas; what gene next?" → SPRED1 (Legius). Same pathway, different node.

  • Pulmonic stenosis + short stature + webbed neck + downslanting palpebral fissuresNoonan. Test PTPN11 first.
  • Café-au-lait macules + axillary freckling + neurofibromas + Lisch nodulesNF1. Diagnostic criteria are clinical first; testing for atypical or de novo cases.
  • Café-au-lait macules + axillary freckling but no other NF1 features → consider Legius (SPRED1). Same pathway, no tumors.
  • Coarse facies + papillomata + cutis laxa + ID + early childhood tumorsCostello (HRAS). Rhabdomyosarcoma surveillance.
  • Lentigines + cardiac conduction abnormalities + pulmonic stenosis + deafnessLEOPARD. PTPN11 (specific dominant-negative variants).
  • Severe phenotype with Noonan-like features but more pronounced ID and ectodermal involvementCFC. BRAF most common.
  • HCM in a syndromic child with broad neck and downslanting palpebral fissuresNoonan (RAF1, PTPN11) > consider familial HCM workup.
  • The RASopathies are collectively the most common single-gene cause of CHD.
  • NF1 has the highest spontaneous mutation rate of any human gene: ~50% of cases are de novo.