Abstract
Purpose
Disorders of somatic mosaicism (DoSM) are a heterogeneous group of conditions caused by postzygotic variants in genes within the PI3K/AKT/mTOR and RAS/MAPK signaling pathway. The co-existence of 2 activating variants in this disease group is extremely rare.
Methods
A deep sequencing next-generation sequencing assay for the molecular diagnosis of DoSM was run on 936 individuals with DoSM.
Results
A single pathogenic or likely pathogenic (P/LP) variant was identified in 584 of 617 (94.8%) positive cases; 33 of 617 (5.2%) cases carried 2 P/LP variants. Of these 33 cases, 22 carried 2 P/LP variants in the same gene, including 8 associated with a loss-of-function disease mechanism and 14 associated with a gain-of-function disease mechanism. Eleven cases had P/LP variants in 2 different genes, including PIKC3A variants in 7 cases and 4 cases with 2 P/LP variants in non-PIK3CA genes.
Conclusion
To our knowledge, this is the largest cohort with the co-existence of 2 P/LP somatic variants causing DoSM. The study of the co-existence of 2 clinically significant variants in DoSM requires unique considerations regarding variant allelic fractions, the combination of variants, affected tissue types, and the severity of the disease. Investigations into this unique cohort may further our understanding of the disease mechanism and potential therapeutic options.
Keywords: Disorders of somatic mosaicism, NGS, PI3K/AKT/mTOR, RAS/MAPK
Introduction
Disorders of somatic mosaicism (DoSM) is a heterogeneous group of malformations caused by mosaic genetic alterations. The clinical manifestations of this disease group are highly variable and heterogeneous, depending on the function of the altered genes, allelic diversity, and when and where the alterations occurred during development. These characteristics overlap the features found in different conditions. Therefore, genetic testing plays a significant role in making a molecular diagnosis as well as identifying therapeutic targets. Next-generation sequencing (NGS) has become an important method of genetic testing to improve the molecular diagnosis of genetic disorders. Because of the nature of mosaic genetic alterations in DoSM, high-sensitivity NGS testing plays a significant role in increasing the diagnostic yield with its capability of detecting mosaic genetic alterations with relatively low variant allele fractions (VAFs).
The genetic alterations observed in DoSM usually affect the function and regulation of key genes in cell survival and proliferation pathways, mainly in 2 intracellular signaling pathways: the PI3K/AKT/mTOR and RAS/RAF pathways, both of which also play significant roles in the setting of cancer. Several activating variants in the key genes are associated with both congenital malformations and oncogenesis. For example, activating PIK3CA variants are associated with both breast cancer and PIK3CA-related overgrowth spectrum (PROS), and activating AKT1 variants are associated with both non–small cell lung cancer and Proteus syndrome.1 The RAS/RAF pathway is often referred to as the proliferation pathway. Germline variants in its key players are associated with a group of congenital disorders known as RASopathies. Mosaic variants in RAS genes (HRAS, KRAS, and NRAS) that result in constitutive RAS signaling are widely observed in cancer, whereas they have been recently described in DoSM.2
DoSM are malformation syndromes that share overlapping genetic alterations with cancer.3 Unlike cancer, which is known to be caused primarily by multiple genetic alterations, most known congenital genetic conditions are caused by single-gene events.4 There are no reports to date studying the evaluation of possible gene combinations in this unique disease group. Here, we review the 7 years’ experience in the Clinical Genetics and Genomics Laboratory at WUSM, during which 936 individuals with DoSM were tested using a high-depth NGS gene panel to identify the individuals with 2 pathogenic or likely pathogenic variants in this cohort.
Materials and Methods
Samples from 936 individuals with DoSM as the indication for testing were submitted to the Clinical Genetics and Genomics Laboratory at Washington University in St. Louis between October 2013 and July 2021. DNA extraction was performed on specimens from affected tissues, including fresh tissues, formalin-fixed paraffin-embedded tissue, cultured fibroblasts, and buccal swab specimens. Target enrichment was performed using oligonucleotide-based targeted capture of exonic regions of up to 37 genes known to underlie DoSM. Sequencing of enriched libraries was performed in multiplex, using the paired-end, ∼150-bp configuration with an average unique on-target coverage depth of ∼2000×. Peripheral blood was obtained as an unaffected comparator alongside the primary affected tissue for 23 of the 33 individuals with clinically significant findings. In those cases, peripheral blood was submitted to a clinical reference laboratory for confirmation of the variant of interest by Sanger sequencing. Detection of the variant in peripheral blood indicates either a multitissue distribution of a mosaic alteration or that the variant has a constitutional origin, whereas the absence of the variant in peripheral blood is taken as indicating mosaicism. The detailed descriptions of the DNA isolation, library preparation, sequencing, bioinformatics analysis, and variant interpretation have been described in our previous publication.5 Of note, tissue-profiling NGS assays, such as the assay used for a majority of cases in this study, are not capable of determining whether discrete variants are present in the same cells, or different cells (or cell types), within affected tissues.
Results
In the cohort of 936 individuals with DoSM, 617 (65.9%) tested positive, with at least 1 clinically significant, pathogenic or likely pathogenic (P/LP) variant in the submitted specimen. A single clinically significant variant was identified in 584 of 617 (94.8%) cases, whereas 33 of 617 (5.2%) cases carried 2 clinically significant variants in the submitted specimens, including 8 cases with 2 P/LP variants in the same gene associated with a loss-of-function (LOF) disease mechanism (Table 1), 14 cases with 2 P/LP variants in the same gene associated with a gain-of-function (GOF) disease mechanism (Table 2), and 11 cases with P/LP variants in 2 different genes (Tables 3 and 4, Figure 1). All the P/LP variants were classified based on American College of Medical Genetics and Genomics (ACMG) 2015 guideline, and the detailed variant information, including nomenclature and classification, can be found in Supplemental Table 1.
Table 1.
Eight cases were identified with 2 clinically significant variants in the same gene associated with the loss-of-function disease mechanism
| Case ID | Age/Gender | Clinical Presentation | Tissue for Sequencing | Gene | Variant | Protein Change | Variant Type | VAF (%) | Detected in Blood? | Allele Ratio |
|---|---|---|---|---|---|---|---|---|---|---|
| 1-a | 5 mo/F | Capillary malformation and minimal overgrowth | Fresh-tissue biopsy from the left foot, plantar surface with capillary malformation | RASA1 | NM_002890.3:c.1328_1331dup | p.(Gln444Hisfs∗10) | Frameshift | 4.00 | N | 11.5 |
| NM_002890.3:c.2920del | p.(Asn976Metfs∗20) | Frameshift | 46.0 | Y | ||||||
| 1-b | 2 y/F | Capillary malformation, possible CM-AVM | Fresh-tissue biopsy from right foot | RASA1 | NM_002890.3:c.2603+2dup | NA | Splice site | 4.25 | N | 10.8 |
| NM_002890.3:c.599del | p.(Lys200Serfs∗7) | Frameshift | 46.0 | Y | ||||||
| 1-c | 2 y/M | Capillary malformation and segmental overgrowth | FFPE >80% affected tissue from the AVM region on the back | RASA1 | NM_002890.3:c.934_938del | p.(Glu312Argfs∗14) | Frameshift | 2.84 | N | 1.1 |
| NM_002890.3:c.2925+1del | NA | Splice site | 2.99 | N | ||||||
| 1-d | 9 y/F | Congenital vascular malformation with predominant venulocapillary component within prominent fibroadipose tissue | FFPE >90% affected tissue | PTEN | NM_000314.8:c.388del | p.(Arg130Glufs∗4) | Frameshift | 2.20 | N | 20.9 |
| NM_000314.8:c.723dup | p.(Glu242∗) | Nonsense | 46.0 | Y | ||||||
| 1-e | 13 y/F | AVM of skin on the back and FAVA on the left thigh | FFPE >80% affected tissue from the AVM region on the back | PTEN | NM_000314.8:c.133del | p.(Val45Tyrfs∗9) | Frameshift | 2.20 | N | 21.8 |
| NM_000314.8:c.569del | p.(Pro190Glnfs∗9) | Frameshift | 48.0 | Y | ||||||
| 1-f | 76 y/F | Recent onset eruptive neurofibromas, no other symptoms of neurofibromatosis, history of breast/ovarian/bladder cancer | FFPE skin punch biopsy from the right upper inner breast | NF1 | NM_000267.3:c.5439_5440inv | p.(Gln1814∗) | Nonsense | 11.9 | NA | 1.1 |
| NM_000267.3:c.6652C>T | p.(Gln2218∗) | Nonsense | 13.6 | NA | ||||||
| 1-g | 6 y/F | Hyperpigmentation, plexiform neurofibroma in the neck, no family history of neurofibromatosis | Fresh tissue from the neck | NF1 | NM_000267.3:c.4537C>T | p.(Arg1513∗) | Nonsense | 12.6 | N | 1.1 |
| NM_000267.3:c.7846C>T | p.(Arg2616∗) | Nonsense | 13.3 | N | ||||||
| 1-h | 2 y/M | Hemihypertrophy and port-wine stain | Fresh tissue from the left forearm | TSC1 | NM_000368.5:c.901_902del | p.(Gln301Glufs∗2) | Frameshift | 3.62 | N | 12.9 |
| NM_000368.5:c.441del | p.(Lys148Asnfs∗19) | Frameshift | 47.0 | Y |
AVM, arteriovenous malformation; CM, capillary malformation; F, female; FAVA, fibroadipose vascular anomaly; FFPE, formalin-fixed paraffin-embedded; M, male; N, no; NA, not applicable; VAF, variable allele fraction; Y, yes.
Table 2.
Fourteen cases were identified with 2 clinically significant variants in the same gene associated with the gain-of-function disease mechanism
| Case ID | Age/Gender | Clinical Presentation | Tissue for Sequencing | Gene | Variant | Protein Change | Variant Type | VAF (%) | Detected in Blood? | Allele Ratio | Phasea |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 2-a | 30 y/M | Arteriovenous malformation of left upper limbs | Fresh tissue | TEK | NM_000459.5: c.2690A>G | p.(Tyr897Cys) | Missense | 2.60 | N | 2.54 | In cis |
| NM_000459.5: c.2752C>T | p.(Arg918Cys) | Missense | 6.60 | N | |||||||
| 2-b | 17 y/F | Soft tissue mass left lateral distal thigh, excision, venous malformation | FFPE | TEK | NM_000459.5: c.2740C>T | p.(Leu914Phe) | Missense | 2.10 | NA | 1.26 | Unknown |
| NM_000459.5: c.2545C>T | p.(Arg849Trp) | Missense | 2.64 | NA | |||||||
| 2-c | 11 y/F | Capillary malformation and vascular anomaly with mild overgrowth | Fresh tissue | TEK | NM_000459.5: c.2743C>T | p.(Arg915Cys) | Missense | 1.44 | N | 1.41 | In cis |
| NM_000459.5: c.2689T>A | p.(Tyr897Asn) | Missense | 2.03 | N | |||||||
| 2-d | 13 y/F | Vascular malformation, predominantly venous | FFPE biopsied from right shoulder | TEK | NM_000459.5: c.3336_3337del | p.(Tyr1113Cysfs∗4) | Frameshift | 1.85 | NA | 19.68 | Unknown |
| NM_000459.5: c.2545C>T | p.(Arg849Trp) | Missense | 36.4 | NA | |||||||
| 2-e | 5 y/M | Right-sided congenital vascular malformation causing upper and lower extremity asymmetry | FFPE specimen from the skin and soft tissue of the right neck | TEK | NM_000459.5: c.3339T>A | p.(Tyr1113∗) | Frameshift | 16.7 | NA | 1.06 | Unknown |
| NM_000459.5: c.2545C>T | p.(Arg849Trp) | Missense | 17.7 | NA | |||||||
| 2-f | 11 mo/M | Venous malformation of the lower lip, vascular birthmarks, seizures, developmental delay | FFPE from the lower lip | TEK | NM_000459.5: c.2753G>A | p.(Arg918His) | Missense | 3.00 | N | 14.43 | Unknown |
| NM_000459.5: c.2545C>T | p.(Arg849Trp) | Missense | 43.3 | Y | |||||||
| 2-g | 10 y/M | Congenital vascular hamartomas, including predominate venous malformation with possible lymphatic components affecting the buttock | FFPE tissue of venous malformation of the buttock | TEK | NM_000459.5: c.3339del | p.(Tyr1113∗) | Frameshift | 11.0 | NA | 4.42 | Unknown |
| NM_000459.5: c.2545C>T | p.(Arg849Trp) | Missense | 48.6 | NA | |||||||
| 2-h | 12 y/F | Multifocal vascular malformations, recurrent bleeding from a vascular malformation of the jejunum, segmental bowel resection, possible blue rubber bleb nevus syndrome | FFPE specimen of the jejunal vascular malformation | TEK | NM_000459.5: c.2690A>G | p.(Tyr897Cys) | Missense | 4.80 | N | 5.96 | In cis |
| NM_000459.5: c.2753G>A | p.(Arg918His) | Missense | 28.6 | Y | |||||||
| 2-i | 6 y/F | Vascular malformation, possible blue rubber bleb nevus syndrome | FFPE specimen of the right shoulder with vascular malformation | TEK | NM_000459.5: c.2752C>T | p.(Arg918Cys) | Missense | 13.8 | NA | 1.06 | In cis |
| NM_000459.5: c.2690A>T | p.(Tyr897Phe) | Missense | 14.6 | NA | |||||||
| 2-j | 3 y/F | Congenital vascular malformation | Fresh frozen tissue specimen from skin biopsy | TEK | NM_000459.5:c.3327_3328del | p.(Lys1110Valfs∗7) | Frameshift | 1.07 | N | 1.46 | In cis |
| NM_000459.5:c.3314C>A | p.(Thr1105Asn) | Missense | 1.56 | N | |||||||
| 2-k | 14 y/M | Multiple vascular malformations and telangiectasias | FFPE tissue further on left upper abdomen with venous malformation | TEK | NM_000459.5:c.2689T>C | p.(Tyr897His) | Missense | 5.90 | NA | 1.64 | In cis |
| NM_000459.5:c.2753G>A | p.(Arg918His) | Missense | 9.70 | NA | |||||||
| 2-l | 12 y/M | Lymphangioma circumscriptum of the left flank, sacral agenesis, left leg length discrepancy, clubfoot, constipation, unequal kidney size, autism, ADHD, generalized anxiety disorder, microcystic lymphatic malformation with macrocystic component. nevus on nape | Fresh tissue specimen of hemangioma of left chest | PIK3CA | NM_006218.4:c.1624G>A | p.(Glu542Lys) | Missense | 1.25 | N | 1.65 | Not in cis |
| NM_006218.4:c.1636C>A | p.(Gln546Lys) | Missense | 2.06 | N | |||||||
| 2-m | 31 y/F | Hemangioma of the tongue, with extensive cautery, hemorrhage, fibrosis, and acute inflammation | FFPE sample of the excised lesion | PIK3CA | NM_006218.4:c.263G>A | p.(Arg88Gln) | Missense | 2.09 | NA | 5.52 | Unknown |
| NM_006218.4:c.3140A>G | p.(His1047Arg) | Missense | 11.5 | NA | |||||||
| 2-n | 18 y/F | Arteriovenous malformation | Fresh tissue from the affected skin | KRAS | NM_004985.5:c.34G>A | p.(Gly12Ser) | Missense | 1.98 | N | 1.09 | Not in cis |
| NM_004985.5:c.65A>G | p.(Gln22Arg) | Missense | 2.16 | N |
ADHD, attention-deficit hyperactivity disorder; F, female; FFPE, formalin-fixed paraffin-embedded; M, male; N, no; NA, not applicable; NGS, next-generation sequencing; VAF, variant allele fraction; Y, yes.
Phase: in cis indicates that the 2 variants were observed on the same NGS sequencing read. Not in cis indicates that the 2 variants were observed on different NGS sequencing reads. As those are somatic variants, they could be either in trans in the same cell or in different cell populations. Unknown indicates that the 2 variants are located too far to be able to observed on the same NGS sequencing read. Phase remains unclear.
Table 3.
Seven cases were identified with 2 clinically significant variants in 2 different genes (PIK3CA + X)
| Case ID | Age/Gender | Clinical Presentation | Tissue for Sequencing | Gene | Variant | Protein Change | Variant Type | VAF (%) | Detected in Blood? | Allele Ratio |
|---|---|---|---|---|---|---|---|---|---|---|
| 3-a | 33 y/F | Clinical diagnosis of Proteus syndrome | FFPE specimen: right thigh - dysplastic melanocytic nevus | PIK3CA | NM_006218.4:c.1633G>A | p.(Glu545Lys) | Missense | 0.92 | N | 5.57 |
| BRAF | NM_004333.6:c.1799T>A | p.(Val600Glu) | Missense | 5.10 | N | |||||
| 3-b | 12 y/M | Rapidly involuting congenital hemangioma | Fresh frozen tissue specimen from a right adrenal mass | PIK3CA | NM_006218.4:c.2908G>A | p.(Glu970Lys) | Missense | 35.7 | N | 3.47 |
| GNAQ | NM_002072.5:c.626A>C | p.(Gln209Pro) | Missense | 10.3 | N | |||||
| 3-c | 35 y/F | Venous vascular malformation, clinical history of a port-wine stain | FFPE tissue from the brain’s right temporal | PIK3CA | NM_006218.4:c.3073A>G | p.(Thr1025Ala) | Missense | 2.23 | NA | 5.03 |
| GNAQ | NM_002072.5:c.626A>G | p.(Gln209Arg) | Missense | 11.2 | NA | |||||
| 3-d | 8 mo/M | Capillary malformation, soft tissue hypertrophy, right inner thigh mass, overgrowth, vascular anomaly | Fresh tissue from the right inner thigh | PIK3CA | NM_006218.4:c.2176G>A | p.(Glu726Lys) | Missense | 20.6 | Y, mosaic | 7.72 |
| GNAQ | NM_002072.5:c.626A>C | p.(Gln209Pro) | Missense | 2.67 | N | |||||
| 3-e | 4 y/M | Diffuse capillary malformation and overgrowth of the leg, angiokeratoma | FFPE biopsy from the skin of the left posterior thigh | PIK3CA | NM_006218.4:c.3140A>G | p.(His1047Arg) | Missense | 2.78 | N | 1.57 |
| KRAS | NM_004985.5:c.35G>A | p.(Gly12Asp) | Missense | 1.77 | N | |||||
| 3-f | 10 mo/M | Congenital malformation syndromes involving early overgrowth/port-wine stain | Fresh tissue specimen from right leg | PIK3CA | NM_006218.4:c.1634A>G | p.(Glu545Gly) | Missense | 6.10 | N | 2.93 |
| RASA1 | NM_002890.3:c.2739del | p.(Met914Cysfs∗17) | Frameshift | 2.08 | N | |||||
| 3-g | 10 y/F | Congenital malformation of the peripheral vascular system | Fresh tissue specimen | PIK3CA | NM_006218.4:c.2740G>A | p.(Gly914Arg) | Missense | 8.00 | Y, mosaic | 4.15 |
| TEK | NM_000459.5:c.3340_3347del | p.(Ala1114∗) | Frameshift | 1.93 | N |
F, female; FFPE, formalin-fixed paraffin-embedded; M, male; N, no; NA, not applicable; VAF, variant allele fraction; Y, yes.
Table 4.
Four cases were identified with 2 clinically significant variants in 2 non-PIK3CA genes
| Case ID | Age/Gender | Clinical Presentation | Tissue for Sequencing | Gene | Variant | Protein Change | Variant Type | VAF (%) | Detected in Blood? | Allele Ratio |
|---|---|---|---|---|---|---|---|---|---|---|
| 4-a | 16 y/M | Cafe au lait spots, congenital facial asymmetry, capillary malformation involving left scapula and left arm, nevus spilus on back, right arm with junctional nevi and epidermal nevus, scalp and neck with a linear reddish rough alopecic patch of left temporal scalp | Fresh tissue from nevus of right forearm | BRAF | NM_004333.6: c.1780G>A | p.(Asp594Asn) | Missense | 4.80 | N | 2.36 |
| NRAS | NM_002524.5: c.179G>A | p.(Gly60Glu) | Missense | 2.03 | N | |||||
| 4-b | 16 y/M | Capillary venous lymphatic malformation | FFPE tissue from right dorsal foot | PIK3R1 | NM_001242466.2:c.601A>G | p.(Asn201Asp) | Missense | 13.0 | NA | 1.29 |
| MTOR | NM_004958.4: c.7280T>G | p.(Leu2427Arg) | Missense | 10.1 | NA | |||||
| 4-c | 14 y/F | Clinical diagnosis of Klippel-Trenaunay syndrome, venous malformation, vascular malformation, features of arteriovenous malformation | FFPE specimen from affected left foot | IDH1 | NM_005896.4: c.394C>T | p.(Arg132Cys) | Missense | 9.70 | N | 1.53 |
| PIK3R1 | NM_001242466.2:c.266_276delinsTTCAAGAAAAAAGTTTCTTGAAA | p.(Tyr89_Gln92delinsPheGlnGluLysSerPheLeuLys) | In-frame delins | 6.36 | N | |||||
| 4-d | 34 y/F | Arteriovenous malformation of the right upper arm | Fresh tissue from the affected right arm | RASA1 | NM_002890.3: c.2035C>T | p.(Arg679∗) | Missense | 16.0 | N | 4.00 |
| PDGFRB | NM_002609.4: c.2548G>T | p.(Asp850Tyr) | Missense | 4.00 | N |
F, female; FFPE, formalin-fixed paraffin-embedded; M, male; N, no; NA, not applicable; VAF, variant allele frequency; Y, yes.
Figure 1.
Genetic results summary of all the 936 cases with DoSM. In the cohort of 936 individuals with DoSM, 617 (65.9%) were reported with at least 1 clinically significant variant in the submitted specimen. A single clinically significant variant was identified in 584 of 617 (94.8%) cases, whereas 33 of 617 (5.2%) cases carried 2 clinically significant variants in the submitted specimens. Of the 33 cases with 2 clinically significant variants, 22 cases were identified with 2 clinically significant variants in the same gene, including 8 cases associated with LOF disease mechanism and 14 cases associated with GOF disease mechanism. The remaining 11 cases with 2 clinically significant variants in 2 different genes, including PIKC3A variants observed in 7 cases, and the remaining 4 cases with 2 clinically significant variants in non-PIK3CA genes only. DoSM, disorders of somatic mosaicism; GOF, gain of function; LOF, loss of function.
Two LOF clinically significant variants in the same gene
We identified 8 cases with 2 clinically significant variants detected in the same gene associated with an LOF disease mechanism, including 3 cases with variants in RASA1 (OMIM 139150), 2 with variants in PTEN (OMIM 601728), 2 with variants in NF1 (OMIM 613113), and 1 with variants in TSC1 (OMIM 605284) (Table 1). Germline LOF variants in RASA1 are associated with an autosomal dominant capillary malformation–arteriovenous malformation (CM-AVM; OMIM 608354). A second somatic clinically significant variant of RASA1 has been reported in vascular endothelial cells in CM-AVM, indicating a 2-hit mechanism of disease.6,7 All 3 individuals with 2 clinically significant variants in RASA1 were ≤2 years of age and had capillary malformations. The type 2 segmental mosaicism, a combination of an initial germline mutational event and a second somatic event, was observed in 2 of the 3 cases, with the germline variants detected in peripheral blood specimens by Sanger sequencing. The third case contained 2 somatic RASA1 variants (VAFs: 2.84% and 2.99%). Although the germline variants in RASA1 are known to cause CM-AVM, RASA1 somatic alterations have been reported to cause CM-AVM in the literature. Germline LOF variants in PTEN are associated with the PTEN hamartoma tumor syndrome, a group of disorders that include Cowden syndrome, Bannayan-Riley-Ruvalcaba syndrome, PTEN-related Proteus syndrome, and PTEN-related Proteus-like syndrome.8, 9, 10, 11 Somatic LOF variants in PTEN occur in sporadic primary tumors, with the highest frequencies in endometrial carcinoma and glioblastoma.12 The 2 cases identified in this study with 2 P/LP variants in PTEN were associated with vascular anomalies within fibroadipose tissue. Each of the 2 cases contained 1 germline variant (VAFs: 46.0% and 48.0%) and 1 somatic variant (VAFs: 2.20% and 2.20%). Two cases with 2 P/LP variants in NF1 were observed in a 76-year-old woman and a 6-year-old girl, both of whom had neurofibromas with no other features of neurofibromatosis and no significant family history of neurofibromatosis. Both NF1 variants in these 2 cases were consistent with somatic origin (VAFs: 11.9% and 13.6% in case ID 1-f; 3.3% and 12.6% in case ID 1-g).
Two GOF clinically significant variants in the same gene
Fourteen cases were identified with 2 clinically significant variants in the same gene with GOF as a disease mechanism, including 11 cases with TEK (OMIM 600221) variants, 2 with PIK3CA (OMIM 171834) variants, and 1 with KRAS (OMIM 190070) variants (Table 2).
A total of 22 TEK variants have been identified in those 11 cases. The TEK variants’ distribution is illustrated in Figure 2, including 17 variants found at 5 mutational hotspots located in the tyrosine kinase (TK) functional domain and 5 variants at 3 mutational hotspots located in the C-terminal tail. Among the 9 unique variants, 6 were missense variants, mainly located in the TK functional domain, whereas 1 nonsense and 2 frameshift variants were located downstream of the TK functional domain in the C-terminal tail. Because the nonsense and frameshift variants were located in the C-terminal tail, they may help reserve protein function instead of causing loss of function.
Figure 2.
TEK protein structure and variant distribution in the 11 cases with TEK 2 clinically significant variants. TEK protein functional domains include EGF, FN3, TM, and TK. A total of 11 cases (case IDs from 2-a to 2-k) were listed with variant distribution in 2 major patterns. Variant distribution pattern #1 was observed in 5 cases with one p.Arg849Trp variant along with another variant either at the 3′ end (case IDs: 2-d, 2-e, and 2-g) or in the middle TK functional domain (case IDs: 2-b and 2-f). Pattern #2 was observed in 5 cases with the co-existence of the p.Tyr897Cys variant and another variant in cis at either p.Arg918His (case IDs: 2-a, 2-h, 2-I, and 2-k) or p.Arg915Cys (case ID: 2-c). EGF, epidermal growth factor-like domain; FN3, fibronectin type-III domain; TK, tyrosine kinase domain; TM, transmembrane domain.
Among the 11 TEK cases, 2 patterns were noted. Pattern #1 was observed in 5 cases with one p.Arg849Trp variant along with another variant either at the 3′ end (case IDs: 2-d, 2-e, and 2-g) or in the middle TK functional domain (case IDs: 2-b and 2-f). The p.Arg849Trp variant was observed as a potential germline variant (either detected by Sanger sequencing in the patient’s peripheral blood or assumed based on the VAF) in 3 of the 5 cases. Of note, because the p.Arg849Trp variant and its second hit locate too far away to be observed on the same NGS read, we cannot determine the phase of those 2 variants. Pattern #2 was observed in 5 cases with the co-existence of the p.Tyr897Cys variant and another variant in cis at either p.Arg918His (case IDs: 2-a, 2-h, 2-i, and 2-k) or p.Arg915Cys (case ID: 2-c). The remaining case (case ID: 2-j) contains 2 variants in cis (p.Lys1110ValfsTer7 and p.Thr1105Asn). Four of the 11 cases contained a potential germline TEK variant, either confirmed by Sanger sequencing in the peripheral blood specimen (case IDs: 2-f and 2-h) or assumed based on the VAFs and internal experience (VAFs: 36.4% and 48.6% in case IDs: 2-d and 2-g, respectively). All other variants, with VAFs ranging from 1.07% to 17.7%, are consistent with a somatic origin. Although a total of 9 unique TEK variants were identified in this cohort, the 4 germline variants were located only at 2 loci, 3 at p.Arg849Trp (case IDs: 2-d, 2-f, and 2-g) and 1 at p.Arg918His (case ID: 2-h). Although 4 of 5 cases with pattern #1 were documented to have a venous malformation, because of the lack of comprehensive clinical presentation, the genotype and phenotype correlations among different cases are still unclear.
Two cases (case IDs: 2-l and 2-m) were identified with the co-existence of 2 PIK3CA variants, all of which are unique variants consistent with somatic origin (VAFs range from 1.25% to 11.5%). Manual inspection of the variants in case 2-l showed that the 2 PIK3CA variants (p.Glu542Lys and p.Glu546Lys) were not present on the same NGS reads, suggesting that the 2 variants either occurred on different alleles (in trans) in the same cell population or occurred in 2 different cell populations. The co-existence of 2 activating PIK3CA variants is extremely rare in DoSM; it has been reported in a female patient with multiple congenital lipomas, vascular malformations, and an ovarian cyst.13 The clinical significance of the co-existence of multiple activating PIK3CA variants remains unclear.
One case (case ID: 2-n) was identified with the co-existence of 2 somatic KRAS variants (p.Gly12Ser and p.Gln22Arg) with similar VAFs (1.98% and 2.16%, respectively). Manual inspection of the variants in case 2-n showed that these 2 KRAS variants were present on the same sequencing reads, consistent with the 2 variants occurring in cis. Interestingly, the co-existence of p.Gly12Ser and p.Gln22Arg in cis has been observed mainly in cancer cases, including a patient with intramuscular hemangioma capillary type (IHCT), which is a fast-flow vascular lesion classified as a tumor, although its phenotype overlaps with arteriovenous malformations (AVMs).14,15 The current case in our study was referred for genetic testing with clinical indication of AVM, which is often difficult to distinguish from IHCT clinically because they share several features. Along with a previously reported case, the co-existence of KRAS p.Gly12Ser and p.Gln22Arg in cis has been observed recurrently in patients with the fast-flow vascular anomaly, including AVM and IHCT.15 The pathophysiologic significance of the co-existence has yet to be described.
Co-existence of 2 clinically significant variants in 2 different genes
PIK3CA plus a second gene
Besides the 22 cases with 2 P/LP variants in the same gene, 11 cases were identified in this study with 2 P/LP variants in 2 different genes on the clinical gene panel (Figure 1). Seven of the 11 cases contain a pathogenic variant in PIK3CA (Table 3). Among these 7 cases, 3 cases have an additional pathogenic variant in the gene GNAQ, each of the remaining 4 cases has a pathogenic variant in the genes BRAF, KRAS, TEK, or RASA1, separately. The PIK3CA variants are considered the primary mutational event (with an allele ratio of approximately 3 or higher) in 4 cases (case IDs: 3-b, 3-d, 3-f, and 3-g). Peripheral blood specimens were available from 6 patients in this cohort. Two PIK3CA variants were detected by Sanger sequencing in the peripheral blood specimen with notably reduced VAF, suggesting multitissue mosaicism. The presence of PIK3CA variants in multiple tissues may indicate that these PIK3CA alterations occurred at a relatively earlier stage during development than the lower-VAF mutation. Based on the limited clinical information available, at least 4 cases (case IDs: 3-b, 3-d, 3-f, and 3-g) have associated congenital malformations, which are consistent with the observation of early-stage PIK3CA alterations in this cohort of patients. Of note, when compared with the cases with single clinically significant PIK3CA variants, there was no obvious preference for the location of the PIK3CA variants seen in these 7 cases in the cases with 2 clinically significant variants (Figure 3). Interestingly, none of the second mutational events are in the genes in the PI3K/AKT pathway, which may suggest the intolerance of a secondary clinically significant variant in the PI3K/AKT pathway along with a pathogenic PIK3CA variant.
Figure 3.
PIK3CA protein structure and variant distribution in the 7 cases with clinically significant variants in PIK3CA and another gene. PIK3CA protein functional domains include ABD, C2, Helical, Kinase, and RBD. A total of 7 cases (case IDs from 3-a to 3-g) were listed with 5 partner genes (BRAF, GNAQ, KRAS, TEK, and RASA1). Each of the 7 cases has a unique PIK3CA variant: 3 harboring the PI3K helical domain, and 4 harboring the PI3/4-kinase domain. ABD, adapter binding domain; C2, calcium-dependent phospholipid-binding domain; Helical, PI3K helical domain; Kinase, PI3/4-kinase domain; RBD, Ras binding domain.
Two clinically significant variants in non-PIK3CA genes
The remaining 4 cases contain 2 clinically significant variants in non-PIK3CA genes. (Table 4) In a 16-year-old male with multiple skin and vascular findings (case ID: 4-a), activating variants in BRAF and NRAS were identified as consistent with somatic origin. PIK3R1 variants were identified in 2 cases with multiple vascular malformations (case IDs: 4-b and 4-c), along with additional pathogenic variants in MTOR in case 4-b and IDH1 in case 4-c. In case 4-d, P/LP variants were identified in genes RASA1 and PDGFRB. Because RASA1 is known to have a second affected allele to cause the disease, the PDGFRB alteration may be responsible for the patient’s vascular finding.
Unlike the cases with pathogenic PIK3CA variants, most cases have 2 pathogenic variants in 2 genes within the same pathway, including the PI3K/AKT pathway. It may suggest that 2 pathogenic non-PIK3CA variants are more tolerated than 2 pathogenic PIK3CA variants. The age at the time of testing ranged from 14 to 34 years among these 4 patients. Peripheral blood specimens were available from 3 patients (case IDs: 4-a, 4-c, and 4-d) in this cohort. None of the pathogenic non-PIK3CA variants was detected by Sanger sequencing in the peripheral blood specimen.
Discussion
The coexistence of 2 clinically significant variants in the same tissue in patients with DoSM is uncommon, and the study of this topic is just emerging. In 2019, a group from The Netherlands reported a cohort of patients with coexistence of 2 clinically significant variants, including 6 cases with PIK3CA variants, 9 cases with PTEN2, and 1 case with RASA1, in a cohort of patients with vascular malformations.16 Here, we report an additional 33 individuals with 2 clinically significant variants in the patients with DoSM. There are several considerations to examine, such as whether the 2 mutational events occur at the same time in development or whether they occur in the same cell population. The allelic ratio of the 2 variants can help to determine whether the variants are simultaneous events or successive events. For example, if 2 variants have an allelic ratio close to 1 (approximately the same allelic fraction), the 2 variants could potentially occur as simultaneous events; on the other hand, if the 2 variants have an allelic ratio much greater than 1 (the VAFs are quite different from each other), it suggests the 2 variants occurred at successive events at different time points during the development. However, it is difficult to determine whether 2 variants are in the same cell using short read sequencing unless they are close enough to each other in the same gene. Given the complex considerations and genetic heterogeneity, we discuss these cases based on the disease mechanism and affected genes.
Two LOF clinically significant variants in the same gene
Some of the genes associated with DoSM are also known as tumor suppressor genes. Co-occurring LOF clinically significant variants were identified in 8 cases, involving 4 different genes (RASA1, PTEN, NF1, and TSC1) in the current study. All 4 of these genes are characterized as tumor suppressor genes.17, 18, 19, 20, 21 Similar to the oncogenesis mechanism of tumor suppressor genes in cancer, those genes also require both alleles to be inactivated to cause disease manifestations. This theory is called the 2-hit hypothesis, also known as the Knudson hypothesis, first formulated by Dr Alfred G. Knudson in 1971.19,22 According to the Knudson hypothesis, there are 2 patterns of the 2-hits to explain the nonhereditary (sporadic) and hereditary (constitutional predisposition) forms of diseases. In the nonhereditary pattern, both alleles are affected by somatic alterations; in the hereditary pattern, one allele is affected by germline alteration (heterozygous), and the second allele is affected by a somatic mutational event. This pattern is recognized as type 2 segmental mosaicism.23 Among the 8 cases in our study, 5 cases (case IDs: 1-a, 1-b, 1-d, 1-e, and 1-h) are consistent with constitutional predisposition; the remaining 3 cases (case IDs: 1-c, 1-f, 1-d, and 1-g) are consistent with sporadic diseases. Of note, some patients were referred to the DoSM NGS panel after a negative constitutional genetic evaluation, such as by germline NGS panel or exome sequencing. Because of such patient population bias, the genetic contribution of the constitutional changes may be underestimated. Therefore, a precise evaluation of the genetic patterns (constitutional vs sporadic) in this disease type remains challenging.
TEK
TEK encodes the dimeric endothelial TK receptor Tie2 (or Tek), which regulates embryonic vascular development and endothelial cell functions such as proliferation, survival, and migration.24,25 TEK GOF variants have been associated with the autosomal dominant germline disorder, multiple cutaneous and mucosal venous malformations, and somatic disorders, such as blue rubber bleb nevus syndrome, multifocal venous malformation, and venous malformation.26 The combination of 2 TEK variants, either germline and somatic or both somatic, has been reported in patients with vascular malformations.27,28 In vitro functional studies showed that the p.Tyr897His variant, together with the second variant in cis in the TEK gene, caused ligand-independent hyperphosphorylation. It also suggested that double TEK clinically significant variants had stronger effects than a constituent of a single TEK clinically significant variant. In previously published studies, there was no correlation between clinical phenotype and level of hyperphosphorylation.29 Additional observations have shown that 2 TEK variants in cis are more typically associated with sporadic disorders.27 Although the combination of 2 TEK variants has been reported, a clear genotype-phenotype correlation is still lacking.
PIK3CA
PIK3CA encodes the catalytic subunit of the phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K) protein complex. PROS disorders represent clinically recognizable conditions, typically discerned at birth or during early childhood, and are associated with cutaneous, vascular, skeletal, and cerebral anomalies, as well as focal or segmental overgrowth of the body.30,31 PIK3CA variants associated with PROS overlap those reported as oncogenic variants found in multiple tumor types (COSMIC and cBioPortal Databases).32,33 A large number of PI3K/AKT pathway inhibitors are currently under clinical study, primarily in a cancer setting, but also in the setting of PROS. The co-existence of 2 clinically significant variants in PIK3CA and GNAQ has been seen in 3 cases in this study. A similar combination of variants has been reported in 2 patients in the literature. In 2018, 1 patient presenting with megalencephaly-capillary malformation was reported to carry both a PIK3CA p.Glu45Lys variant and a GNAQ p.Gln209His variant detected in DNA extracted from a congenital hemangioma.3 The other patient was a 58-year-old woman presenting with hepatic small vessel neoplasm. Both GNAQ p.Gln209His and PIK3CA p.His1047Arg variants were detected in an autopsy performed after a pulmonary embolism.34 The co-existence of 2 activating PIK3CA variants has been reported in at least 1 patient with lipoma, capillary malformation, and an ovarian cyst.13 To date, no genotype-phenotype correlation has been established regarding specific variants, or combination of variants, and severity of the disease.
Challenges and limitations
The genotype-phenotype correlation is challenging for DoSM. Similar to the cancer paradigm, the somatic clinically significant variants for this disease group can occur in different tissues at different time points. The same variants may lead to different phenotypes depending on the mutational burden and the timing of the somatic alteration (acquired vs congenital, early developmental stage vs late stage). Therefore, the phenotype is not only dictated by the specific postzygotic somatic variants but also influenced by the VAF, the tissue distribution of variants, modifying constitutional variants, and potentially additional acquired somatic variants.
The limitations of this study include an inability to fully review the phenotype-genotype correlation for every case. An effort to correlate phenotype with genotype was made. However, the clinical information is limited to what has been provided by the ordering clinician as the indication for testing. The genetic testing laboratory usually does not have access to detailed clinical presentations for every case, especially those ordered by external clinicians. Another limitation is that the age of each patient, based on the date of sample collection, may not be consistent with the date of initial clinical diagnosis or evaluation. Third, although constitutional genetic alterations can be detected by this assay, the genetic contribution of the constitutional changes may be underestimated because of patient population bias. Some of the patients with similar clinical presentations may undergo prior genetic testing focusing on constitutional findings. The patients who receive positive constitutional results may not be referred for genetic testing focusing on somatic alterations. Additionally, more subtle presentations may not be appreciated or referred for additional testing. Therefore, it is challenging to determine the clinical significance of each variant, as the VAF and disease onset are variable from case to case.
Another important limitation of this study is the inability to determine the clonal distribution of discrete variants based on all tissue-profiling analyses using NGS technology. Bioinformatics tools are imperfect in their ascertainment of VAF, although clustering of variants present within statistically similar VAF has been widely applied in studies describing the oligoclonality of cancer. This limitation could conceivably be addressed by single-cell omics analysis, and it is a logical next step in understanding the presence and mechanism(s) of cooperativity between cases with multiple pathogenic and likely pathogenic variants in the mix.
Despite the limitations, this study includes the largest cohort of cases with the co-existence of 2 clinically significant variants in DoSM. To date, no study has looked specifically at this unique group of cases. The observation and description of these cases with 2 clinically significant variants, including both clinical presentations and genetic results, will shed light on a better understanding of this unique group of diseases. In addition, this study highlights the importance of broader sequencing in DoSM, which may affect prognosis and treatment responses in those cases. The emerging evidence of the co-existence of multiple clinically significant variants in DoSM in this study also guides future research direction to better understand the genetic mechanism of DoSM.
Data Availability
All relevant data are reported in the article and Figures 1-3 and Supplemental Table 1.
Acknowledgments
The authors thank the patients, families and clinicians involved in these cases. The authors also thank Dr Kilannin Krysiak for her critical review of the manuscript.
Funding
Funding for this project was provided by the Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, MO. No grants or other funding sources were used.
Author Information
Conceptualization: Y.C., J.A.N.; Data Curation: Y.C., M.J.E., M.M.C., M.C.S., J.W.H., J.A.N.; Formal Analysis: Y.C., M.J.E.; Methodology: Y.C., M.J.E., M.M.C., M.C.S., J.W.H., J.A.N.; Supervision: J.A.N.; Visualization: Y.C.; Writing-original draft: Y.C.; Writing-review and editing: J.A.N., Y.C., M.J.E., M.M.C., M.C.S., J.W.H.
Ethics Declaration
This single IRB (sIRB) that approved this study was the Washington University School of Medicine sIRB, #: 202103225. These data were collected during clinical testing, and no separate consent was needed for the deidentified accumulated data from the clinical analyses. The work was conducted in a manner consistent with the Belmont Report and the Declaration of Helsinki.
Conflict of Interest
The authors declare no conflicts of interest.
Footnotes
The Article Publishing Charge (APC) for this article was paid by the Department of Pathology and Immunology, Washington University School of Medicine.
Additional Information
The online version of this article (https://doi.org/10.1016/j.gimo.2023.100807) contains supplementary material, which is available to authorized users.
Contributor Information
Yang Cao, Email: cao.yang@wustl.edu.
Julie A. Neidich, Email: jneidich@wustl.edu.
Additional Information
References
- 1.Lindhurst M.J., Sapp J.C., Teer J.K., et al. A mosaic activating mutation in AKT1 associated with the Proteus syndrome. N Engl J Med. 2011;365(7):611–619. doi: 10.1056/NEJMoa1104017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Al-Olabi L., Polubothu S., Dowsett K., et al. Mosaic RAS/MAPK variants cause sporadic vascular malformations which respond to targeted therapy. J Clin Invest. 2018;128(4):1496–1508. doi: 10.1172/JCI98589. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Siegel D.H., Cottrell C.E., Streicher J.L., et al. Analyzing the genetic spectrum of vascular anomalies with overgrowth via cancer genomics. J Invest Dermatol. 2018;138(4):957–967. doi: 10.1016/j.jid.2017.10.033. [DOI] [PubMed] [Google Scholar]
- 4.Al Hajri Q., Dash S., Feng W.C., Garner H.R., Anandakrishnan R. Identifying multi-hit carcinogenic gene combinations: scaling up a weighted set cover algorithm using compressed binary matrix representation on a GPU. Sci Rep. 2020;10(1):2022. doi: 10.1038/s41598-020-58785-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.McNulty S.N., Evenson M.J., Corliss M.M., et al. Diagnostic utility of next-generation sequencing for disorders of somatic mosaicism: a five-year cumulative cohort. Am J Hum Genet. 2019;105(4):734–746. doi: 10.1016/j.ajhg.2019.09.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Lapinski P.E., Doosti A., Salato V., North P., Burrows P.E., King P.D. Somatic second hit mutation of RASA1 in vascular endothelial cells in capillary malformation-arteriovenous malformation. Eur J Med Genet. 2018;61(1):11–16. doi: 10.1016/j.ejmg.2017.10.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Macmurdo C.F., Wooderchak-Donahue W., Bayrak-Toydemir P., et al. RASA1 somatic mutation and variable expressivity in capillary malformation/arteriovenous malformation (CM/AVM) syndrome. Am J Med Genet A. 2016;170(6):1450–1454. doi: 10.1002/ajmg.a.37613. [DOI] [PubMed] [Google Scholar]
- 8.Marsh D.J., Kum J.B., Lunetta K.L., et al. PTEN mutation spectrum and genotype-phenotype correlations in Bannayan-Riley-Ruvalcaba syndrome suggest a single entity with Cowden syndrome. Hum Mol Genet. 1999;8(8):1461–1472. doi: 10.1093/hmg/8.8.1461. [DOI] [PubMed] [Google Scholar]
- 9.Liaw D., Marsh D.J., Li J., et al. Germline mutations of the PTEN gene in Cowden disease, an inherited breast and thyroid cancer syndrome. Nat Genet. 1997;16(1):64–67. doi: 10.1038/ng0597-64. [DOI] [PubMed] [Google Scholar]
- 10.Marsh D.J., Dahia P.L., Zheng Z., et al. Germline mutations in PTEN are present in Bannayan-Zonana syndrome. Nat Genet. 1997;16(4):333–334. doi: 10.1038/ng0897-333. [DOI] [PubMed] [Google Scholar]
- 11.Marsh D.J., Coulon V., Lunetta K.L., et al. Mutation spectrum and genotype-phenotype analyses in Cowden disease and Bannayan-Zonana syndrome, two hamartoma syndromes with germline PTEN mutation. Hum Mol Genet. 1998;7(3):507–515. doi: 10.1093/hmg/7.3.507. [DOI] [PubMed] [Google Scholar]
- 12.Milella M., Falcone I., Conciatori F., et al. PTEN: multiple functions in human malignant tumors. Front Oncol. 2015;5:24. doi: 10.3389/fonc.2015.00024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lalonde E., Ebrahimzadeh J., Rafferty K., et al. Molecular diagnosis of somatic overgrowth conditions: a single-center experience. Mol Genet Genomic Med. 2019;7(3):e536. doi: 10.1002/mgg3.536. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Miyakura Y., Sugano K., Fukayama N., Konishi F., Nagai H. Concurrent mutations of K-ras oncogene at codons 12 and 22 in colon cancer. Jpn J Clin Oncol. 2002;32(6):219–221. doi: 10.1093/jjco/hyf043. [DOI] [PubMed] [Google Scholar]
- 15.Goss J.A., Konczyk D.J., Smits P.J., et al. Intramuscular fast-flow vascular anomaly contains somatic MAP2K1 and KRAS mutations. Angiogenesis. 2019;22(4):547–552. doi: 10.1007/s10456-019-09678-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Ten Broek R.W., Eijkelenboom A., van der Vleuten C.J.M., et al. Comprehensive molecular and clinicopathological analysis of vascular malformations: a study of 319 cases. Genes Chromosomes Cancer. 2019;58(8):541–550. doi: 10.1002/gcc.22739. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Zhang Y., Li Y., Wang Q., et al. Role of RASA1 in cancer: a review and update. Oncol Rep. 2020;44(6):2386–2396. doi: 10.3892/or.2020.7807. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Parsons R. Discovery of the PTEN tumor suppressor and its connection to the PI3K and AKT oncogenes. Cold Spring Harb Perspect Med. 2020;10(8):a036129. doi: 10.1101/cshperspect.a036129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Hino O., Kobayashi T. Mourning Dr. Alfred G. Knudson: the two-hit hypothesis, tumor suppressor genes, and the tuberous sclerosis complex. Cancer Sci. 2017;108(1):5–11. doi: 10.1111/cas.13116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Patrakitkomjorn S., Kobayashi D., Morikawa T., et al. Neurofibromatosis type 1 (NF1) tumor suppressor, neurofibromin, regulates the neuronal differentiation of PC12 cells via its associating protein, CRMP-2. J Biol Chem. 2008;283(14):9399–9413. doi: 10.1074/jbc.M708206200. [DOI] [PubMed] [Google Scholar]
- 21.Kang Y.J., Lu M.K., Guan K.L. The TSC1 and TSC2 tumor suppressors are required for proper ER stress response and protect cells from ER stress-induced apoptosis. Cell Death Differ. 2011;18(1):133–144. doi: 10.1038/cdd.2010.82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Knudson A.G. Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci U S A. 1971;68(4):820–823. doi: 10.1073/pnas.68.4.820. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Happle R. The concept of type 2 segmental mosaicism, expanding from dermatology to general medicine. J Eur Acad Dermatol Venereol. 2018;32(7):1075–1088. doi: 10.1111/jdv.14838. [DOI] [PubMed] [Google Scholar]
- 24.Saharinen P., Eklund L., Alitalo K. Therapeutic targeting of the angiopoietin-TIE pathway. Nat Rev Drug Discov. 2017;16(9):635–661. doi: 10.1038/nrd.2016.278. [DOI] [PubMed] [Google Scholar]
- 25.Du Z., Zheng J., Zhang Z., Wang Y. Review of the endothelial pathogenic mechanism of TIE2-related venous malformation. J Vasc Surg Venous Lymphat Disord. 2017;5(5):740–748. doi: 10.1016/j.jvsv.2017.05.001. [DOI] [PubMed] [Google Scholar]
- 26.Kangas J., Nätynki M., Eklund L. Development of molecular therapies for venous malformations. Basic Clin Pharmacol Toxicol. 2018;123(suppl 5):6–19. doi: 10.1111/bcpt.13027. [DOI] [PubMed] [Google Scholar]
- 27.Soblet J., Kangas J., Nätynki M., et al. Blue rubber bleb nevus (BRBN) syndrome is caused by somatic TEK (TIE2) mutations. J Invest Dermatol. 2017;137(1):207–216. doi: 10.1016/j.jid.2016.07.034. [DOI] [PubMed] [Google Scholar]
- 28.Soblet J., Limaye N., Uebelhoer M., Boon L.M., Vikkula M. Variable somatic TIE2 mutations in half of sporadic venous malformations. Mol Syndromol. 2013;4(4):179–183. doi: 10.1159/000348327. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Limaye N., Wouters V., Uebelhoer M., et al. Somatic mutations in angiopoietin receptor gene TEK cause solitary and multiple sporadic venous malformations. Nat Genet. 2009;41(1):118–124. doi: 10.1038/ng.272. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Mirzaa G., Timms A.E., Conti V., et al. PIK3CA-associated developmental disorders exhibit distinct classes of mutations with variable expression and tissue distribution. JCI Insight. 2016;1(9) doi: 10.1172/jci.insight.87623. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Keppler-Noreuil K.M., Rios J.J., Ver Parker, et al. PIK3CA-related overgrowth spectrum (PROS): diagnostic and testing eligibility criteria, differential diagnosis, and evaluation. Am J Med Genet A. 2015;167A(2):287–295. doi: 10.1002/ajmg.a.36836. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Tate J.G., Bamford S., Jubb H.C., et al. COSMIC: the catalogue of somatic mutations in cancer. Nucleic Acids Res. 2019;47(D1):D941–D947. doi: 10.1093/nar/gky1015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Gao J., Aksoy B.A., Dogrusoz U., et al. Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci Signal. 2013;6(269):pl1. doi: 10.1126/scisignal.2004088. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Joseph N.M., Brunt E.M., Marginean C., et al. Frequent GNAQ and GNA14 mutations in hepatic small vessel neoplasm. Am J Surg Pathol. 2018;42(9):1201–1207. doi: 10.1097/PAS.0000000000001110. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
All relevant data are reported in the article and Figures 1-3 and Supplemental Table 1.