If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Pyogenic granuloma (PG) is a common benign vascular skin lesion presenting as a rapidly growing angiomatous papule. The pathogenesis of most sporadic PGs and PGs associated with port wine stains (PWSs) remains elusive. We report that of 10 PGs secondarily arisen on a PWS, 8 showed a BRAF c.1799T>A (p.(Val600Glu)) and 1 a NRAS c.182A>G (p.(Gln61Arg)) mutation. The GNAQ c.548G>A mutation was identified in the PG and in the respective underlying PWS, indicating that PGs originate from cells of the PWS. In contrast to PG, 12 papulonodular lesions, which had developed in the PWSs of seven patients, showed a RAS and BRAF wild-type status. In sporadic PG we identified the BRAF c.1799T>A mutation in 3 of 25, a BRAF c.1391G>A mutation in 1 of 25, and a KRAS c.37G>C mutation in 1 of 25. Mutation-specific immunohistochemical detection of BRAF p.(Val600Glu) confirmed endothelial cells as carriers of the mutation in secondary and sporadic PG.
Our study identifies the BRAF c.1799T>A mutation as a major driver mutation in the pathogenesis of, particularly, secondary PG. These data shed light on the hitherto undetermined genetic basis of PG and classify PG as a benign neoplasm.
According to the International Society for the Study of Vascular Anomalies, vascular anomalies are grouped into tumors and malformations based on their clinical appearance, histopathologic features, and biologic behavior (
). For instance, the occurrence of pyogenic granuloma (PG) within port wine stains (PWSs), which have been shown to result from the activating GNAQ c.548G>A mutation in approximately 90% of cases (
PGs are common benign vascular neoplasms presenting as rapidly growing angiomatous papules or polyps. In most cases male children or young male adults are affected. Histopathologically, PGs evolve from compact proliferations of collapsed vascular structures to vessels with relatively regular lumina and multilobular arrangement. At advanced stages of PG increasing venulization and progressive fibrosis are observed. The differential diagnoses of PG include cherry (senile) angioma (CA) and acquired tufted angioma. Regarding the nature of PG, controversy exists as to whether these vascular lesions should be categorized as reactive or neoplastic. Reports on PG appearing in response to previous injury, hormonal factors, pharmacologic, or laser treatment have led many authors to classify PG as a reactive lesion (
Despite these latest advances, the pathogenetic basis of most sporadic PG and PG associated with PWS remains elusive. Therefore, our primary aim of this study was to test the hypothesis that PGs arising on PWSs may not be mere reactive hyperplasias but rather true benign neoplasms by screening these lesions for activating mutations in the RAS–mitogen-activated protein kinase (MAPK)-signaling pathway.
Results
Material from seven patients who had developed 10 PGs within their PWSs was available for genetic analysis. Three patients had developed two PGs at a time within their PWSs. As elaborated in Table 1, a detailed review of patient files revealed that all secondary PGs had developed spontaneously without a clear coincidence in time between a localized trauma or laser treatment and the formation of the lesion. Except for patient 4, whose PG and PWS showed a GNAQ wild-type sequence at codon 183, the GNAQ c.548G>A mutation was identified in all PGs and in the respective underlying PWSs, indicating that the PGs originate directly from cells of the PWS.
Table 1Genetic analysis and clinical characteristics of PGs on PWSs and nodular PWSs
Of the 10 secondary PG analyzed for mutations in the Ras–MAPK-signaling pathway, 8 tested positive for a BRAF c.1799T>A mutation resulting in a p.(Val600Glu) substitution. In patients 3 and 6 both PGs carried the BRAF c.1799T>A mutation. In patient 5 a BRAF c.1799T>A mutation was detected in the parietal PG, whereas an NRAS c.182A>G (p.Q61R) mutation was identified in the temporal PG. BRAF and NRAS mutations were not detected in the respective underlying PWSs, which indicates their specific pathogenic role in secondary PGs (Figure 1).
Figure 1Clinical, histologic, and genetic analyses of PG (a and b) and nodular hypertrophies (c and d) in PWSs. (a) PG of subject 3 (Table 1) presenting as red polypoid tumor with a subtle collarette on a preexisting PWS. (b) Histologically, the PG shows compact vascular proliferation of solid, largely collapsed vascular structures. The BRAF c.1799T>A mutation is present in the PG, whereas the underlying PWS showed a wild-type sequence at codon 600 of BRAF. The PG and PWS both carry the GNAQ c.548G>A mutation. (c) Multiple papular lesions have developed on a PWS. (d) Histologically, the nodular lesion is composed of thin-walled vascular channels with varying lumina. The node is positive for the somatic GNAQ c.548G>A mutation but shows a wild-type sequence at codon 600 of BRAF. Peaks indicate the DNA antisense strand of the GNAQ gene and the DNA sense strand of the BRAF gene. Scale bar = 2 mm.
The somatic nature of the BRAF, GNAQ, and RAS mutations identified was verified, because perilesional epidermis from 10 samples did not show the mutations identified in the lesional tissues. The specificity of RAS and BRAF mutations in PGs arisen on PWSs is substantiated by the RAS and BRAF wild-type status found in 12 nodular hypertrophies in PWSs from seven patients (Table 1). Of note, we identified a heretofore undescribed GNAQ c.548G>T mutation resulting in a GNAQ p.R183L substitution in the nodular PWS of patient 9 (see Supplementary Figure 1, online). To the best of our knowledge this mutation has not been described before in PWSs. Consulting various functional effect prediction tools [PolyPhen-2 (http://genetics.bwh.harvard.edu/pph2/index.shtml), Mutation Assessor (http://mutationassessor.org), and SIFT (http://sift.jcvi.org)], this mutation is pre-estimated to be damaging.
The observed predominance of RAS and BRAF mutations in PGs arisen on PWSs prompted us to screen sporadic PGs for activating mutations in these genes. In 25 sporadic PGs we identified a BRAF c.1799T>A mutation in 3, a BRAF c.1391G>A mutation in 1, and a KRAS c.37G>C mutation in 1 (Figure 2, Table 2). Corresponding epidermis showed wild-type sequences of the respective codons of BRAF and KRAS, thus proving somaticism. To address the question of which vascular component is the carrier of the activating mutations detected, we investigated the protein expression of BRAF p.(Val600Glu) as the most common activating mutation identified in secondary and sporadic PGs. By performing immunohistochemistry with a BRAF p.(Val600Glu) mutation-specific antibody, strong staining identified endothelial cells as carriers of the BRAF p.(Val600Glu) mutation in five secondary and three sporadic PGs (Figure 3).
Figure 2Histology and corresponding RAS SNaPshot multiplex assay chromatograms of sporadic PG (a) and sporadic CA (b). (a) PG of subject 23 (Table 2) presenting as a dome-shaped lesion with a well-developed lobular architecture. The lesional tissue shows a KRAS c.37G>C mutation, whereas the adjacent epidermis revealed a wild-type sequence at codon 13 of KRAS. (b) CA of subject 18 (Table 3) presents as a dome-shaped lesion with insinuated lobules and thin-walled dilated vascular channels. Genetic analysis revealed a somatic HRAS c.182A>G mutation in the dermal vascular proliferation. Insets show a higher magnification of the respective lesions. Peaks indicate the DNA antisense strand of the respective genes. Scale bar = 2 mm.
Figure 3Immunohistochemical analysis of BRAF p.(Val600Glu) expression and localization in secondary and sporadic PGs. (a) Representative BRAF p.(Val600Glu) staining of a PG (subject 3 in Table 1) arisen on a PWS. Endothelial cells could clearly be identified as carriers of the BRAF p.(Val600Glu) mutation. (b) BRAF p.(Val600Glu) expression in endothelial cells of a sporadic PG (subject 17 in Table 2). (c) Representative negative (left) and positive (right) control staining of a PG with BRAF wild-type sequence and a melanoma with a BRAF p.(Val600Glu) mutation. Scale bar = 100 μm.
Finally, we tested the specificity of RAS and BRAF mutations in sporadic PGs by screening 25 CAs, which may share certain histologic characteristics with PG, for these mutations. Interestingly we identified an HRAS c.37G>C in 3 of 25, a KRAS c.34G>T in 1 of 25, and an HRAS c.182A>G mutation in 1 of 25 CAs (Figure 2, Table 3). Of note, no BRAF mutations were identified in the CA studied. These data demonstrate that on the contrary to the BRAF mutations identified, RAS mutations are not specific for PG but are also involved in the pathogenesis of CA.
The results of this study strongly support the conclusion that the BRAF c.1799T>A mutation plays a pivotal role in the pathogenesis of sporadic and particularly secondary PGs. Alongside sebaceous hyperplasias in patients with MYH-associated polyposis (
), sporadic and secondary PGs thus add to the spectrum of benign nonmelanocytic skin lesions carrying the BRAF c.1799T>A mutation. This hotspot mutation, which accounts for approximately 90% of BRAF mutations in human cancer (
). In the context of our data, a recent report on the occurrence of PG as an adverse effect of the BRAF p.(Val600Glu) specific inhibitor vemurafenib is remarkable (
) and the observation of tumor growth promotion in preexisting RAS mutated lesions by paradoxical activation of MAPK signaling during treatment with BRAF inhibition (
). The detection of HRAS and KRAS mutations in approximately 20% of CAs furthermore portends a common genetic basis of CA and PG and suggests a phenotypic continuum between the two vascular lesions. Phenotypic variability of RAS mutations in these benign vascular lesions may rely on the respective RAS mutation and its level of downstream pathway activation, the cell type affected by the mutation and external stimuli (e.g., traumatization), growth hormones, or additional pathway activation by a drug or a second activating/inhibiting mutation.
In addition to the identification of BRAF c.1799T>A as a driver mutation in PG, its conspicuous association with the GNAQ c.548G>A mutation in secondary PG is the most striking finding of this study. The GNAQ c.548G>A mutation has been detected in approximately 90% of nonsyndromic PWSs and results in a rather low activation of the MAPK-signaling pathway (
). The BRAF c.1799T>A mutation might therefore act as a “second hit” on a GNAQ c.548G>A mutation mediated, mildly activated MAPK-signaling pathway in endothelial cells of PWS. This interpretation is supported by the immunohistochemical data that show a strong homogenous expression of mutant BRAF in basically all endothelial cells of secondary PG.
Although mutant GNAS as a G-protein representative co-occurs with KRAS and BRAF mutations in colon cancer (
), the association of GNAQ and BRAF/KRAS mutations has not been reported in vascular or melanocytic lesions, where the GNAQ c.548G>A has been described in few nevi Ota and uveal melanoma (
). Further studies are needed to investigate a possible association of BRAF c.1799T>A and GNAQ c.548G>A, especially in these melanocytic lesions, because, for example, patients with phacomatosis cesioflammea appear to be at increased risk of developing uveal melanoma (
Phacomatosis pigmentovascularis of cesioflammea type in 7 patients: combination of ocular pigmentation (melanocytosis or melanosis) and nevus flammeus with risk for melanoma.
In tumors BRAF c.1799T>A has been demonstrated to drive angiogenesis by enhancing the expression of several proangiogenic and proinflammatory molecules, including hypxia inducible factor-1α, vascular endothelial growth factors A and C, transforming growth factor-α, IL-1β, and IL-8 (
). RAS mutations in endothelial cells have been shown to induce the angiogenic switch in the phenotype of these cells by driving branching morphogenesis, cell migration, and cell cycle progression (
). Our findings indicate that RAS and BRAF mutations themselves, however, are not sufficient to induce malignancy in human endothelial cells, because CA and PG are benign vascular lesions. Nevertheless, the well-known formation of satellite lesions in PG, as observed in patient 6, whose two PGs in close proximity displayed the same genetic profile, may point to at least some malignant potential of these oncogenes in endothelial cells.
In conclusion, our study identifies the BRAF c.1799T>A mutation as a major driver mutation in the pathogenesis of sporadic and particularly secondary PGs. As in current multistep models of tumor development, it appears to act as a “second hit” on a GNAQ c.548G>A mutation mediated, mildly activated MAPK-signaling pathway in endothelial cells of PWSs, subsequently leading to tumor growth. These data thus shed light on the hitherto undetermined genetic basis of these common benign neoplasms and provide arguments against a classification of PG as a mere reactive hyperplasia.
Methods
Sample and DNA acquisition
Material from seven patients who had developed 10 PGs within their PWSs were retrieved from the histologic files of the Department of Dermatology (University of Regensburg, Germany). Of these seven patients, files were reviewed regarding a possible coincidence in time between a localized trauma or laser treatment and the formation of the lesion. Furthermore, 12 nodular hypertrophies in PWSs from seven patients as well as 25 sporadic PGs and 25 CAs were retrieved from the histologic files noted above. The characteristics of the respective patients are shown in Table 1, Table 2, Table 3. Written informed consent had been obtained from all patients or their parents before excision of the respective PGs, nodular hypertrophies in PWSs, and CAs. The consent procedure and the study, which was performed according to the Declaration of Helsinki, were approved by the local ethics committee.
Manual microdissection was performed under an inverted microscope from paraffin-embedded tissue with a thickness of 8 μm for each section. DNA was isolated using the QIAamp DNA Kit (Qiagen, Hilden, Germany) for paraffin-embedded tissue following the protocol of the manufacturer.
Mutation analyses
Highly sensitive SNaPshot multiplex assays were used to detect HRAS, KRAS, and NRAS mutations as previously described (
). Set 1 covered the following nucleotide positions: 34, 35, 37, 181, 182 of HRAS; 34, 35, 181 of KRAS; and 34, 182 of NRAS. Set 2 covered the following base positions: 38, 183 of HRAS; 37, 38, 182, 183 of KRAS; and 35, 37, 38, 180, 181, 183 of NRAS. The BRAF c.1799T>A and GNAQ c.548G>A mutation were assayed as previously described (
). For amplification and sequencing of BRAF exons 11 and 15 the following primers were used: exon_11_F TCCCTCTCAGGCATAAGGTAA, exon_11_R CGAACAGTGAATATTTCCTTTGAT, exon15_F TCATAATGCTTGCTCTGATAGGA, exon15_R GGCCAAAAATTTAATCAGTGGA. A second independent PCR was used to confirm all detected mutations. PCR conditions can be obtained from the authors upon request.
Immunohistochemistry
Because of tissue limitations, five of eight secondary and three of three sporadic BRAF c.1799T>A mutation-positive PGs could be evaluated for BRAF p.(Val600Glu) localization within the vascular tumors using immunohistochemistry. Fully automated immunostaining was performed with the Ventana Benchmark ULTRA staining system (Ventana Medical Systems, Inc., Tucson, USA), including standard pretreatment for 1 hour with Tris-EDTA buffer and by using mouse monoclonal BRAF p.(Val600Glu) antibody, clone VE1 (Spring Bioscience, Pleasanton, CA), in a dilution of 1:25. The overall BRAF p.(Val600Glu) staining intensity (not the frequency of positive tumor cells) was scored 0 (negative), 1+ (weak), 2+ (strong), and 3+ (very strong) by two individual investigators.
Conflict of Interest
The authors state no conflict of interest.
Acknowledgments
This work was supported by the research grant GR 4610/1-1 from the Deutsche Forschungsgemeinschaft to L.G. We thank all subjects who participated in this study.
Phacomatosis pigmentovascularis of cesioflammea type in 7 patients: combination of ocular pigmentation (melanocytosis or melanosis) and nevus flammeus with risk for melanoma.
Groesser et al. demonstrate that pyogenic granuloma is a RAS pathway-driven tumor. This important observation adds yet another manifestation to the growing spectrum of cutaneous “RASopathies” and raises intriguing questions about the relationship between RAS pathway activation and malignancy.
00026-3&id=gr1.jpg){kind=link}
00026-3&id=gr2.jpg){kind=link}
00026-3&id=gr3.jpg){kind=link}