Mutational Analysis of BRAF Inhibitor–Associated Squamoproliferative Lesions
2015; Elsevier BV; Volume: 17; Issue: 6 Linguagem: Inglês
10.1016/j.jmoldx.2015.05.009
ISSN1943-7811
AutoresBritt Clynick, Tania Tabone, Kathryn A. Fuller, Wendy N. Erber, Katie Meehan, Michael Millward, Benjamin A. Wood, Nathan T. Harvey,
Tópico(s)HER2/EGFR in Cancer Research
ResumoIn recent years, there has been increasing use of BRAF-inhibiting drugs for the treatment of various malignancies, including melanoma. However, these agents are associated with the development of other nonmelanoma skin lesions, in particular squamoproliferative lesions such as keratoacanthomas (KAs), squamous cell carcinomas, and BRAF inhibitor–associated verrucous keratoses. The molecular pathogenesis of these lesions is of interest, not only for therapeutic reasons, but also for the insight it might provide into the development of similar lesions in a sporadic setting. We used next-generation sequencing to compare the mutational profiles of lesions after treatment with a BRAF inhibitor, with similar lesions arising sporadically. HRAS mutations were common among the BRAF inhibitor–induced lesions, being identified in 56%, compared with 14% of lesions in the sporadic group (P = 0.002). Thus, despite similar histomorphological appearances, the underlying molecular mechanisms may be different. In addition, within the BRAF inhibitor–associated group, the lesions designated as KAs and BRAF inhibitor–associated verrucous keratoses had a similar mutational profile (mutations in PIK3CA, APC, and HRAS), which was distinct to that seen in squamous cell carcinomas (FGFR3, CDKN2A, and STK11). We have previously noted histological overlap between KAs and BRAF inhibitor–associated verrucous keratoses, and this finding supports the notion that they may represent morphological or temporal variants of a single lesion type. In recent years, there has been increasing use of BRAF-inhibiting drugs for the treatment of various malignancies, including melanoma. However, these agents are associated with the development of other nonmelanoma skin lesions, in particular squamoproliferative lesions such as keratoacanthomas (KAs), squamous cell carcinomas, and BRAF inhibitor–associated verrucous keratoses. The molecular pathogenesis of these lesions is of interest, not only for therapeutic reasons, but also for the insight it might provide into the development of similar lesions in a sporadic setting. We used next-generation sequencing to compare the mutational profiles of lesions after treatment with a BRAF inhibitor, with similar lesions arising sporadically. HRAS mutations were common among the BRAF inhibitor–induced lesions, being identified in 56%, compared with 14% of lesions in the sporadic group (P = 0.002). Thus, despite similar histomorphological appearances, the underlying molecular mechanisms may be different. In addition, within the BRAF inhibitor–associated group, the lesions designated as KAs and BRAF inhibitor–associated verrucous keratoses had a similar mutational profile (mutations in PIK3CA, APC, and HRAS), which was distinct to that seen in squamous cell carcinomas (FGFR3, CDKN2A, and STK11). We have previously noted histological overlap between KAs and BRAF inhibitor–associated verrucous keratoses, and this finding supports the notion that they may represent morphological or temporal variants of a single lesion type. In recent years, there has been increasing use of several novel agents that specifically target V600 BRAF mutations (predominantly V600E and V600K) in melanoma and other malignancies. These drugs, which include vemurafenib (RG7204; Roche, Nutley, NJ) and dabrafenib (GSK2118436; GlaxoSmithKline, Philadelphia, PA), are showing encouraging results in clinical trials, with improved rates of overall and progression-free survival.1Chapman P. Hauschild A. Robert C. Haanen J. Ascierto P. Larkin J. Dummer R. Garbe C. Testori A. Maio M. Improved survival with vemurafenib in melanoma with BRAF V600E mutation.N Engl J Med. 2011; 364: 2507-2516Crossref PubMed Scopus (6115) Google Scholar, 2Hauschild A. Grob J.J. Demidov L. Jouary T. Gutzmer R. Millward M. Rutkowski P. Blank C. Miller Jr., W. Kaempgen E. Dabrafenib in BRAF-mutated metastatic melanoma: a multicentre, open-label, phase 3 randomised controlled trial.Lancet. 2012; 380: 358-365Abstract Full Text Full Text PDF PubMed Scopus (2357) Google Scholar However, these agents are also associated with a variety of cutaneous adverse effects that can result in significant morbidity, including the development of other nonmelanoma skin lesions.3Anforth R. Blumetti T. Kefford R. Sharma R. Scolyer R. Kossard S. Long G. Fernandez-Peñas P. Cutaneous manifestations of dabrafenib (GSK2118436): a selective inhibitor of mutant BRAF in patients with metastatic melanoma.Br J Dermatol. 2012; 167: 1153-1160Crossref PubMed Scopus (149) Google Scholar, 4Boyd K. Vincent B. Andea A. Conry R.M. Hughey L. Nonmalignant cutaneous findings associated with vemurafenib use in patients with metastatic melanoma.J Am Acad Dermatol. 2012; 67: 1375-1379Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar, 5Chu E. Wanat K. Miller C. Amaravadi R. Fecher L. Brose M. McGettigan S. Giles L. Schuchter L. Seykora J. Diverse cutaneous side effects associated with BRAF inhibitor therapy: a clinicopathologic study.J Am Acad Dermatol. 2012; 67: 1265-1272Abstract Full Text Full Text PDF PubMed Scopus (142) Google Scholar, 6Harvey N. Millward M. Wood B. Squamoproliferative lesions arising in the setting of BRAF inhibitors.Am J Dermatopathol. 2012; 34: 822-826Crossref PubMed Scopus (37) Google Scholar The tumors that most frequently develop include squamoproliferative lesions, such as keratoacanthomas (KAs), squamous cell carcinomas (SCCs), and other lesions showing wart-like features we prefer to call BRAF inhibitor–associated verrucous keratoses (BAVKs).3Anforth R. Blumetti T. Kefford R. Sharma R. Scolyer R. Kossard S. Long G. Fernandez-Peñas P. Cutaneous manifestations of dabrafenib (GSK2118436): a selective inhibitor of mutant BRAF in patients with metastatic melanoma.Br J Dermatol. 2012; 167: 1153-1160Crossref PubMed Scopus (149) Google Scholar, 6Harvey N. Millward M. Wood B. Squamoproliferative lesions arising in the setting of BRAF inhibitors.Am J Dermatopathol. 2012; 34: 822-826Crossref PubMed Scopus (37) Google Scholar The molecular pathogenesis of these lesions is of interest, not only for therapeutic reasons, but also for the insight it might provide into the development of similar lesions in a sporadic (ie, non–BRAF inhibitor–associated) setting. Current theoretical understanding of their pathogenesis cites a paradoxical stimulatory effect of BRAF inhibitors on cells that do not carry the target BRAF mutation. As expected, in melanoma cell lines that carry a V600E BRAF mutation, BRAF-inhibiting drugs will block activation of the downstream molecule extracellular signal–related kinase (ERK). However, in cells with wild-type BRAF, these drugs have the opposite effect, with an increase in ERK activation.7Hatzivassiliou G. Song K. Yen I. Brandhuber B. Anderson D. Alvarado R. Ludlam M. Stokoe D. Gloor S. Vigers G. Morales T. Aliagas I. Liu B. Sideris S. Hoeflich K. Jaiswal B. Seshagiri S. Koeppen H. Belvin M. Friedman L. Malek S. RAF inhibitors prime wild-type RAF to activate the MAPK pathway and enhance growth.Nature. 2010; 464: 431-435Crossref PubMed Scopus (1280) Google Scholar, 8Heidorn S. Milagre C. Whittaker S. Nourry A. Niculescu-Duvas I. Dhomen N. Hussain J. Reis-Filho J. Springer C. Pritchard C. Kinase-dead BRAF and oncogenic RAS cooperate to drive tumor progression through CRAF.Cell. 2010; 140: 209-221Abstract Full Text Full Text PDF PubMed Scopus (1191) Google Scholar, 9Poulikakos P. Zhang C. Bollag G. Shokat K. Rosen N. RAF inhibitors transactivate RAF dimers and ERK signalling in cells with wild-type BRAF.Nature. 2010; 464: 427-430Crossref PubMed Scopus (1396) Google Scholar, 10Su F. Viros A. Milagre C. Trunzer K. Bollag G. Spleiss O. Reis-Filho J.S. Kong X. Koya R.C. Flaherty K.T. RAS mutations in cutaneous squamous-cell carcinomas in patients treated with BRAF inhibitors.N Engl J Med. 2012; 366: 207-215Crossref PubMed Scopus (866) Google Scholar This effect is potentiated in cells harboring an activating RAS mutation, as demonstrated by increased proliferation of tissue culture cells that have been transfected with mutant RAS genes, as well as accelerated development of cutaneous SCCs in a Ras-driven mouse model of skin carcinogenesis.10Su F. Viros A. Milagre C. Trunzer K. Bollag G. Spleiss O. Reis-Filho J.S. Kong X. Koya R.C. Flaherty K.T. RAS mutations in cutaneous squamous-cell carcinomas in patients treated with BRAF inhibitors.N Engl J Med. 2012; 366: 207-215Crossref PubMed Scopus (866) Google Scholar, 11Holderfield M. Lorenzana E. Weisburd B. Lomovasky L. Boussemart L. Lacroix L. Tomasic G. Favre M. Vagner S. Robert C. Ghoddusi M. Daniel D. Pryer N. McCormick F. Stuart D. Vemurafenib cooperates with HPV to promote initiation of cutaneous tumors.Cancer Res. 2014; 74: 2238-2245Crossref PubMed Scopus (29) Google Scholar Indeed, several previous studies have shown that RAS mutations are commonly identified in squamoproliferative lesions from patients undergoing treatment with BRAF-inhibiting drugs,10Su F. Viros A. Milagre C. Trunzer K. Bollag G. Spleiss O. Reis-Filho J.S. Kong X. Koya R.C. Flaherty K.T. RAS mutations in cutaneous squamous-cell carcinomas in patients treated with BRAF inhibitors.N Engl J Med. 2012; 366: 207-215Crossref PubMed Scopus (866) Google Scholar, 12Anforth R. Tembe V. Blumetti T. Fernandez-Peñas P. Mutational analysis of cutaneous squamous cell carcinomas and verrucal keratosis in patients taking BRAF inhibitors.Pigment Cell Melanoma Res. 2012; 25: 569-572Crossref PubMed Scopus (49) Google Scholar, 13Oberholzer P. Kee D. Dziunycz P. Sucker A. Kamsukom N. Jones R. Roden C. Chalk C. Ardlie K. Palescandolo E. RAS mutations are associated with the development of cutaneous squamous cell tumors in patients treated with RAF inhibitors.J Clin Oncol. 2012; 30: 316-321Crossref PubMed Scopus (345) Google Scholar and more recently RAS mutations have also been documented in noncutaneous tumors arising in this setting.14Boussemart L, Girault I, Mateus C, Thomas M, Routier E, Cazenave H, Tomasic G, Wechlser J, Kamsu-Kom N, Roy S, Favre M, Lacroix L, Eggermont A, Vagner S, Robert C: BRAF inhibitors induce skin and extra-cutaneous tumors via paradoxical activation of the MAPK pathway: molecular study of 66 tumors and visualization of BRAF/CRAF protein dimers [abstract 934]. Presented at the 105th Annual Meeting of the American Association for Cancer Research, 2014, April 5–9, San Diego, CA.Google Scholar Thus, in patients undergoing treatment with BRAF inhibitors, there is the potential for paradoxical activation of the mitogen-activated protein kinase (MAPK) pathway in cells that do not harbor a BRAF mutation. Concomitant RAS activation may combine with BRAF inhibition to drive tumorigenic growth.15Robert C. Arnault J.P. Mateus C. RAF inhibition and induction of cutaneous squamous cell carcinoma.Curr Opin Oncol. 2011; 23: 177-182Crossref PubMed Scopus (105) Google Scholar To further elucidate the biological basis of these secondary cutaneous lesions, we used next-generation sequencing (NGS) to assess the mutational profiles of lesions after treatment with a BRAF inhibitor. We compared these with the profiles of histologically similar lesions arising in a sporadic setting. We then used immunohistochemistry to assess all these lesions for evidence of MAPK pathway activation. Formalin-fixed, paraffin-embedded biopsy material was obtained from the pathology archives of PathWest Laboratory Medicine, QEII Medical Centre (Nedlands, WA, Australia). A total of 41 biopsy samples were collected, originating from 11 different patients undergoing treatment with either vemurafenib or dabrafenib. The lesions had been previously categorized as KA (n = 10), SCC (n = 11), and BAVK (n = 20).6Harvey N. Millward M. Wood B. Squamoproliferative lesions arising in the setting of BRAF inhibitors.Am J Dermatopathol. 2012; 34: 822-826Crossref PubMed Scopus (37) Google Scholar In addition, 21 biopsy samples of similar squamoproliferative lesions were collected from 21 patients who were not undergoing treatment with BRAF inhibitors (ie, sporadic lesions). These included KA (n = 10) and SCC (n = 11). The diagnostic categories for both the BRAF inhibitor–associated and the sporadic cases were determined by a consultant histopathologist (N.T.H. or B.A.W.). Ethical approval for this study was obtained from the Sir Charles Gairdner Hospital (Nedlands, WA, Australia) Human Research Ethics Committee. A RM2125 RTS microtome (Leica Bioystems, North Ryde, Australia) was used to obtain two sections (10 μm thick) from each formalin-fixed, paraffin-embedded tissue block for subsequent tissue digestion and DNA extraction. Representative sections (4 μm thick) obtained before and after tissue sectioning were stained with hematoxylin and eosin to assess the proportion of tumor present in the final purified DNA sample. Before DNA extraction, the paraffin was first removed via a standard series of xylene and graded ethanol washes. DNA was extracted using a commercial magnetic bead separation method (ChargeSwitch gDNA Micro Tissue Kit; Life Technologies, Tullamarine, Australia), according to the manufacturer's instructions. The yield of purified genomic DNA was estimated using the Qubit 2.0 Fluorometer and Qubit dsDNA HS Assay Kit (Life Technologies), according to the manufacturer's instructions. The NGS platform used in this study was the Ion Personal Genome Machine (PGM) Sequencer (Life Technologies).16Henson J. Tischler G. Ning Z. Next-generation sequencing and large genome assemblies.Pharmacogenomics. 2012; 13: 901-915Crossref PubMed Scopus (83) Google Scholar, 17Pareek C. Smoczynski R. Tretyn A. Sequencing technologies and genome sequencing.J Appl Genet. 2011; 52: 413-435Crossref PubMed Scopus (470) Google Scholar, 18Singh R. Patel K. Routbort M. Reddy N. Barkoh B. Handal B. Kanagal-Shamanna R. Greaves W. Medeiros L. Aldape K. Clinical validation of a next-generation sequencing screen for mutational hotspots in 46 cancer-related genes.J Mol Diagn. 2013; 15: 607-622Abstract Full Text Full Text PDF PubMed Scopus (264) Google Scholar Library preparation for each sample was performed using the Ion AmpliSeq Library Kit 2.0 and Ion AmpliSeq Cancer HotSpot Panel version 2 (Life Technologies), following the manufacturer's instructions. This targeted cancer panel sequences hot spot mutations from 50 oncogenes and tumor-suppressor genes implicated in various cancers: these genes are summarized in Table 1. Unique Ion Xpress Barcode Adapters 1-32 (Life Technologies) were ligated to the amplicons and subsequently purified using Agencourt AMPure XP Reagent (Beckman Coulter, Lane Cove, Australia). The amplicons underwent a second round of PCR amplification to complete their linkage with the adapters, with another purification step using Agencourt AMPure XP Reagent (Beckman Coulter). The amplified library was then quantified using the Qubit 2.0 Fluorometer (Life Technologies), according to the manufacturer's instructions. The final library concentrations were standardized to 100 pmol/L in Ion AmpliSeq Low TE buffer (Life Technologies).Table 1The 50 Target Genes Amplified with the Ion AmpliSeq Cancer HotSpot Primer PanelGene namesABL1EGFRGNASKRASPTPN11AKT1ERBB2GNAQMETRB1ALKERBB4HNF1AMLH1RETAPCEZH2HRASMPLSMAD4ATMFBXW7IDH1NOTCH1SMARCB1BRAFFGFR1IDH2NPM1SMOCDH1FGFR2JAK2NRASSRCCDKN2AFGFR3JAK3PDGFRASTK11CSF1RFLT3KDRPIK3CATP53CTNNB1GNA11KITPTENVHL Open table in a new tab Ten uniquely barcoded library samples (100 pmol/L each) were pooled, and altogether five pools were generated for the BRAF inhibitor–associated lesions and three pools for the sporadic lesions. The final concentration of each pool was adjusted to 9 pmol/L diluted in nuclease-free water, and then clonally amplified onto ion sphere particles by emulsion PCR with biotinylated primers using the Ion PGM Template OT2 Reagents 200 Kit (Life Technologies) and OneTouch 2 System (Life Technologies), according to the manufacturer's instructions. Each pool was loaded onto an Ion 318v2 Chip (one pool per chip; Life Technologies) for single-end sequence analysis using the Ion PGM Sequencer using 500 flows (125 cycles) for 200-base-read-sequencing. Data collected from the PGM were initially processed using the Ion Torrent platform-specific pipeline software Torrent Suite version 3.6.2 (ThermoFisher Scientific, Waltham, MA) to generate sequence reads and to filter and remove poor signal-profile reads. In particular, this software was used to align reads to the reference genome (human genome hg19) and generate run metrics, including chip loading efficiency, total coverage of reads, and total read counts and quality. ANNOVAR (Biobase Biological Databases, Wolfenbuettel, Germany) was used to annotate the detected variants, and provide details about the possible functional consequences of the variant [eg, nonsynonymous single-nucleotide variant (SNV), synonymous SNV, and frameshift insertion/deletion]. ANNOVAR was also used to identify a subset of variants previously reported in publically available databases [namely, the single-nucleotide polymorphism database and the 1000 Genomes project (http://www.1000genomes.org, last accessed May 18, 2015)]. Formalin-fixed, paraffin-embedded skin biopsy specimens were sectioned (4 μm thick) and placed on positively charged slides for immunohistochemistry (IHC) staining using a BOND RX automated immunostainer (Leica Biosystems). All slides were dewaxed in Bond dewax solution (Leica Biosystems) before undergoing heat-induced epitope retrieval at pH 6 for 20 minutes (bond epitope retrieval solution 1). Antibodies that specifically recognize the phosphorylated form of ERK (pERK), either MAPK-YT (Abcam, Melbourne, Australia) or ERK1/2 (Cell Signaling Technologies, Danvers, MA), were then incubated on tissue sections for 30 minutes at a dilution of 1:100 and 1:200, respectively. The two antibodies showed equivalent staining of control tissue during the optimization phase, and the results represent a combination of the two. The antibodies were diluted with Bond Primary Antibody Diluent (Leica Biosystems). Antigen binding was detected using an alkaline phosphatase BOND Polymer Refine Red Detection kit (Leica Biosystems). Sections were counterstained in hematoxylin, dehydrated, and mounted in Ultramount 4 mounting media (Hurst Scientific Pty Ltd, Perth, WA, Australia) using an automated glass coverslipper (Leica CV5030; Leica Biosystems). BRAF-mutated melanoma tissue was used as a positive control. The adjacent nonlesional epidermis of each section was used as a negative control in this study and has previously been demonstrated to be a suitable negative control for pERK staining.10Su F. Viros A. Milagre C. Trunzer K. Bollag G. Spleiss O. Reis-Filho J.S. Kong X. Koya R.C. Flaherty K.T. RAS mutations in cutaneous squamous-cell carcinomas in patients treated with BRAF inhibitors.N Engl J Med. 2012; 366: 207-215Crossref PubMed Scopus (866) Google Scholar A Fisher's exact test and χ2 analysis using contingency tables were used for statistical analysis. The intrapatient and interpatient distribution of the BRAF inhibitor–associated lesions is summarized in Table 2. They occurred on both sun-exposed (19 of 41) and non–sun-exposed (22 of 41) sites in equivalent proportions. Although there appeared to be a tendency for non–sun-exposed sites to yield BAVKs rather than KAs or SCCs, this trend did not reach statistical significance in our sample (P > 0.05).Table 2Characteristics of the BRAFi-Associated Lesions Including the Age and Sex of Each Patient, the Type of BRAFi the Patient Was Receiving, the Site of the Lesion, the Histopathological Diagnosis, and the Mutations IdentifiedPatientAge, years/sexDrug treatmentSiteDiagnosisMutation type161/MaleVemurafenibChestKAPIK3CA S690F, FBXW7 P205S, APC A1582P, HRAS G12D, TP53 R248Q,TP53 P177SShinKAHRAS G12DCalfKAPIK3CA E542K, HRAS G13VCalfBAVKPIK3CA E542K, HRAS Q61LChestBAVKHRAS G12DShinBAVKPIK3CA S323F, PIK3CA E542K, APC A1582P, TP53 E294X252/FemaleVemurafenibThighBAVKAPC A1582P, HRAS G12D, TP53 P82SNeckBAVKHRAS Q61L350/MaleVemurafenibForearmKANot enough tissue for sequencingLower trunkKAPIK3CA E542K, HRAS Q61LChestBAVKTP53 E285K, TP53 R248W, TP53 K132X, HRAS Q61L, ATM G3051R,APC A1582PUpper trunkBAVKHRAS Q61L, FGFR3 L760F, PIK3CA E542K, PIK3CA A1066T469/FemaleVemurafenibThighBAVKPIK3CA E542K, HRAS G12DThighBAVKPIK3CA H1047R, HRAS G12DLegBAVKPIK3CA E542K, FGFR3 G380R, HRAS G12D, STK11 A347VThighBAVKPIK3CA E542K, PIK3CA H1047R, APC A1582P, PTEN T131I, HRAS G12D,KRAS G12DBackBAVKAPC A1582P, HRAS G12CThighBAVKHRAS G13V, APC A1582P, PIK3CA E545K577/MaleVemurafenibThighKANo mutations detectedThighSCCNo mutations detectedShinSCCRET T636M, FGFR3 A374TCalfSCCFGFR3 H675Y, FGFR3 Q682X, SMO G393D, ABL1 S367L, AKT1 D44N,TP53 M246I, TP53 R213X, STK11 V63M, GNA11 V204M, JAK3 A732T,JAK3 P725SHandSCCCDKN2A R58X, TP53 R248Q, TP53 V218E666/MaleVemurafenibShoulderSCCNo mutations detected742/FemaleDabrafenibChestBAVKAPC A1582T, HRAS G12D868/FemaleDabrafenibChestBAVKKRAS G12V, HRAS Q61LChestBAVKHRAS Q61L, TP53 R213GAnterior neckBAVKHRAS Q61L, APC A1582P968/MaleDabrafenibUpper armKAPIK3CA E453KCalfKANo mutations detectedShinKASTK11 P38S, TP53 R196Q, FGFR3 Q758X, SMO M525I, NOTCH1 G1572STempleSCCSTK11 D358N, SMO G529S, CDKN2A R80X, FGFR3 A374VShinSCCNo mutations detectedChestSCCTP53 A347V, KRAS C51F, HNF1A A269V, BRAF R31X, CDKN2A P114L,CDKN2A R80X, KDR I256T, PDGFRA R558CForearmSCCRB1 Q344X, RB1 R455X, TP53 E204X, ERBB4 P170SElbowSCCNo mutations detectedForearmKACDKN2A R80X, TP53 E286K1050/FemaleDabrafenibThighBAVKPIK3CA E545K, HRAS G12DChestBAVKFGFR3 P250S, HRAS G12DChestSCCHRAS G13R1149/MaleDabrafenibLipBAVKTP53 H179Y, TP53 L130I, TP53 A76TBAVK, BRAF inhibitor-associated verrucous keratoses; BRAFi, BRAF inhibitor; KA, keratoacanthoma; SCC, squamous cell carcinoma. Open table in a new tab BAVK, BRAF inhibitor-associated verrucous keratoses; BRAFi, BRAF inhibitor; KA, keratoacanthoma; SCC, squamous cell carcinoma. Sequencing data analysis revealed a mean coverage depth of 1034 reads per nucleotide position within the target region. The distribution of reads covering the 207 amplicons was consistent across each sample, with average amplicon uniformity coverage of 95.41%. Target capture was effective, with an average of 99.83% of the sequence reads mapped to targeted gene regions (aligned to human genome reference 19). Successful sequencing of the samples was measured by using a minimum of 250,000 reads with a quality score of AQ20, which corresponds to a predicted error rate of 1% (one misaligned base per 100 bases). Other parameters adopted by our study allowing a variant to be considered true include a sequencing coverage of 1000×, as well as a variant frequency of at least 10% in a background of wild-type alleles. Within the BRAF inhibitor–induced lesions, 40 of the 41 lesions were successfully sequenced, with insufficient DNA extracted from one specimen. A total of 783 variants in 41 genes were identified, with an average of 19 variants detected per lesion (range, 12 to 49 variants). Stringent mutation detection criteria were used to identify somatically acquired mutations: any variants present in the population with a minor allele frequency >5%, according to the 1000 Genomes project, were excluded and considered a potential common polymorphism or passenger mutation; similarly, intronic mutations and synonymous exonic mutations were excluded because these do not alter the protein function and are considered less likely to be disease-causing mutations. A total of 163 variants were predicted to cause nonsynonymous coding changes in 28 different genes, 17 variants were predicted to cause stop-gain coding shifts in six different genes, and six variants were predicted to cause splicing alterations in five different genes. From these 186 variants, 185 were not previously described in the single-nucleotide polymorphism database or the 1000 Genomes project. Within the sporadic lesions cohort, all 21 lesions were sequenced, with a total of 304 variants identified in 31 genes, with an average of 14 variants detected per lesion (range, 11 to 20 variants). Similar stringent mutation detection criteria to those described above were also used to identify somatically acquired mutations within this cohort. As a result, a total of 47 variants were predicted to cause nonsynonymous coding changes in 15 different genes, five variants were predicted to cause stop-gain coding shifts in four different genes, and three variants were predicted to cause splicing alterations in two different genes. From these 55 variants, 53 were not previously described in the single-nucleotide polymorphism database or the 1000 Genomes project. Figure 1A displays the frequency of each mutation identified in both the BRAF inhibitor and sporadic subgroups, and clear differences between the two groups can be appreciated. As expected, HRAS mutations were common among the BRAF inhibitor–induced lesions, being identified in 23 (56%) of the 41 samples. These mutations included 12 cases showing activating mutations in codon 12 (G12D, G12C), three cases showing activating mutations in codon 13 (G13V, G13R), and eight cases showing activating mutations in codon 61 (Q61L). PIK3CA, TP53, and APC were also commonly mutated in this group, whereas mutations in STK11, FGFR3, CDKN2A, and SMO were also seen, albeit less frequently. The specific mutations found in the BRAF inhibitor–associated lesions are summarized in Table 2. In contrast to the BRAF-inhibitor subgroup, HRAS mutations were much less commonly seen in the sporadic lesions (Figure 1A), being present in 3 (14%) of 21 lesions in this group. This difference was statistically significant (P = 0.002). Perhaps not surprisingly, mutations in TP53 were the most commonly identified abnormality in this group. The lesions in the sporadic group also showed mutations in PIK3CA and APC. Mutations in these genes were seen at lower rates than in the BRAF-inhibitor group; however, these differences did not reach statistical significance (P > 0.05). The lesions in this group also showed low levels of several other mutations that were not seen in the BRAF-inhibitor group (Figure 1A). The specific mutations found in the sporadic lesions are summarized in Table 3.Table 3Characteristics of the Sporadic Control Lesions, Including the Age and Sex of Each Patient, the Site, the Histopathological Diagnosis, and Mutations IdentifiedPatientAge, years/sexSiteDiagnosisMutation type159/MaleChestKAFGFR3 R399H282/MaleForearmKAPTPN11 S499F, TP53 H47Y360/FemaleUpper armKATP53 Y88N469/MaleForearmKAAPC E1317Q553/MaleLegKAMET P84S, HRAS G13D683/FemaleForearmKASMO P645L, CDKN2A P81L, TP53 G147E, TP53 Q104X, SRC T511I770/MaleCalfKACDKN2A P114L, TP53 G134R, TP53 Y104N879/FemaleShinKANo mutations detected977/FemaleNeckKATP53 P20Q, TP53 P20S1076/FemaleArmKAVHL R82C, PIK3CA E418K, HRAS G12S1176/MaleNeckSCCKIT W582X, CDKN2A P114L, CDKN2A R58X1286/FemaleLegSCCPIK3CA E542K, PIK3CA S690F, TP53 P146L, TP53 R116W1370/FemaleEarlobeSCCMET T1010I, CDKN2A D84N, TP53 P146S, TP53 S127F, SMAD4 D512N1476/FemaleLegSCCERBB4 G319R, ABL1 M244I, TP53 P45L1586/FemaleForearmSCCPIK3CA Q546P, TP53 P146S1668/FemaleArmSCCNo mutations detected1778/FemaleTempleSCCHRAS G12S, TP53 C44W1872/MaleEarSCCNo mutations detected1987/FemaleLegSCCPIK3CA E542K, EZH2 Y602F, EZH2 Y602N, CDKN2A H83N, TP53 P146L, TP53 P177L2046/MaleTempleSCCSMO N202K, CDKN2A R58X, TP53 H47Y, TP53 S127F, STK11 D207G2142/MaleLipSCCERBB4 W171X, TP53 R116WKA, keratoacanthoma; SCC, squamous cell carcinoma. Open table in a new tab KA, keratoacanthoma; SCC, squamous cell carcinoma. Figure 1B examines the mutations seen in the BRAF-inhibitor group only, subdivided by the histological classification of the lesions. HRAS mutations were most predominant in the BAVKs, in which 18 (90%) of 20 lesions contained a mutation. Interestingly, only one SCC from this group had a HRAS mutation. Indeed, within the BRAF-inhibitor–associated group, we noted that overall the lesions designated as KAs and BAVKs had a similar mutational profile, which was different to the SCCs. The mutational profile that seems to characterize KAs and BAVKs comprises mutations in PIK3CA, APC, HRAS, and TP53. Alternatively, SCCs were characterized more by mutations in TP53, FGFR3, CDKN2A, and STK11. The SCCs were also notable in that they showed a much broader range of mutations than either the BAVKs or the KAs. Of the 41 BRAF inhibitor–associated lesions, 35 (85%) showed positivity for pERK with IHC (Figure 2), defined as at least focal strong labeling within the nucleus of >5% of lesional keratinocytes. There was no statistically significant difference between HRAS mutant and wild-type lesions with regard to their pERK staining: most lesions with a HRAS mutation were positive (21/23, 91%), as were most of the lesions without an HRAS mutation (14/18, 78%). Of the 14 pERK-positive lesions that were negative for HRAS mutations, five showed no targeted mutations from our panel, five had other potential MAPK-activating mutations (FGFR3, ERBB4, KRAS, and BRAF), and four had mutations in genes from other signaling pathways (APC, PIK3CA, CDKN2A, and TP53). In many cases, the staining was noted to be heterogeneous throughout the lesion, often being strongest in the more differentiated keratinizing cells, with the basal layer showing less staining. A similar pattern of staining was noted in the sporadic lesions, in which all 21 lesions were positive (data not shown). The melanoma used as a positive control stained positively in all runs, and in all cases the adjacent epidermis was negative, apart from a weak blush within the stratum granulosum. Curiously, both endothelial cells and smooth muscle cells within the dermis showed consi
Referência(s)