IntroductionIntraocular melanocytoma is a rare variant of melanocytic nevus, and belongs to the group of non-epithelial associated melanocytic lesions, which include cutaneous blue nevi and primary leptomeningeal melanocytic tumors.1Reidy J.J. Apple D.J. Steinmetz R.L. et al.Melanocytoma: nomenclature, pathogenesis, natural history and treatment., 2Esmaili D.D. Mukai S. Jakobiec F.A. Kim I.K. Gragoudas E.S. , 3Murali R. McCarthy S.W. Scolyer R.A. Blue nevi and related lesions: a review highlighting atypical and newly described variants, distinguishing features and diagnostic pitfalls., 4van de Nes J. Gessi M. Sucker A. et al.Targeted next generation sequencing reveals unique mutation profile of primary melanocytic tumors of the central nervous system. This neoplasm is composed of large polyhedral cells with abundant, intensely pigmented cytoplasm and relatively small, oval nuclei with few nucleoli, best seen in bleached sections.5Melanocytes, Melanocytic Nevi, and Melanocytomas. Typically, intraocular melanocytoma presents as a mass on or adjacent to the optic disk with characteristic homogeneous, jet-black appearance.Less commonly, melanocytoma involves the uveal tract with somewhat overlapping clinical features with other uveal nevi or small melanomas. 6Kathil P. Milman T. Finger P.T. Characteristics of anterior uveal melanocytomas in 17 cases.,7LoRusso F.J. Boniuk M. Font R.L. Melanocytoma (magnocellular nevus) of the ciliary body: report of 10 cases and review of the literature. Iris melanocytoma has a distinct clinical morphology, and presents as a darkly-pigmented nodule with feathery surface, typically lacking internal vasculature or ectropion uvea.7LoRusso F.J. Boniuk M. Font R.L. Melanocytoma (magnocellular nevus) of the ciliary body: report of 10 cases and review of the literature.,8Demirci H. Mashayekhi A. Shields C.L. Eagle Jr., R.C. Shields J.A. Iris melanocytoma: clinical features and natural course in 47 cases. Iris melanocytoma usually behaves in a benign manner, but may show growth, necrosis (due to lack of internal vasculature), glaucoma, and seeding to the iridocorneal angle and sclera.7LoRusso F.J. Boniuk M. Font R.L. Melanocytoma (magnocellular nevus) of the ciliary body: report of 10 cases and review of the literature.,8Demirci H. Mashayekhi A. Shields C.L. Eagle Jr., R.C. Shields J.A. Iris melanocytoma: clinical features and natural course in 47 cases. The typical management of suspected iris melanocytoma is observation, while sector iridectomy and, less frequently, enucleation are performed in cases with persistent glaucoma, clinical concern for melanoma, or for a blind painful eye.8Demirci H. Mashayekhi A. Shields C.L. Eagle Jr., R.C. Shields J.A. Iris melanocytoma: clinical features and natural course in 47 cases.,9Shields J.A. Shields C.L. Eagle Jr., R.C. Melanocytoma (hyperpigmented magnocellular nevus) of the uveal tract: the 34th G. Victor Simpson lecture. Ciliary body melanocytomas may pose a greater challenge in the diagnosis due to overlapping features with ciliary body melanoma. These tumors are more likely to be managed with iridocyclectomy, plaque radiotherapy, or enucleation depending on the size and clinical concern for malignancy.7LoRusso F.J. Boniuk M. Font R.L. Melanocytoma (magnocellular nevus) of the ciliary body: report of 10 cases and review of the literature.,9Shields J.A. Shields C.L. Eagle Jr., R.C. Melanocytoma (hyperpigmented magnocellular nevus) of the uveal tract: the 34th G. Victor Simpson lecture.Clinical differentiation of melanocytoma from melanoma may be difficult, and failure to recognize pigmented iris and ciliary body tumors as melanocytoma may lead to unnecessary treatment and oncologic surveillance for metastases. Fine needle aspiration biopsy (FNAB) has been reported as one of the diagnostic approaches to differentiate melanocytoma from melanoma.8Demirci H. Mashayekhi A. Shields C.L. Eagle Jr., R.C. Shields J.A. Iris melanocytoma: clinical features and natural course in 47 cases.,10Char D.H. Miller T.R. Crawford J.B. Cytopathologic diagnosis of benign lesions simulating choroidal melanomas.,11El-Harazi S.M. Kellaway J. Font R.L. Melanocytoma of the ciliary body diagnosed by fine-needle aspiration biopsy. FNAB coupled with various molecular tests has been widely used in the diagnosis and prognostication of uveal melanoma.12Angi M. Kalirai H. Taktak A. et al.Prognostic biopsy of choroidal melanoma: an optimised surgical and laboratory approach. Molecular genetic studies may potentially be similarly useful in the diagnosis of melanocytoma, especially in cases with borderline cytomorphologic features. However, our knowledge about the molecular alterations beyond GNAQ and GNA11 mutations in ocular melanocytoma is limited. Published series examining the genetics of uveal melanocytomas consist of two ciliochoroidal cases, tested by mutational analysis of a single exon in three target genes (GNAQ, GNA11, BRAF) and six cases (1 ciliary body and 5 optic nerve head) tested by targeted sequencing of GNAQ/GNA11, BAP1, and SF3B1 only.13Mudhar H.S. Doherty R. Salawu A. Sisley K. Rennie I.G. Immunohistochemical and molecular pathology of ocular uveal melanocytoma: evidence for somatic GNAQ mutations.,14Francis J.H. Wiesner T. Milman T. et al.Investigation of Somatic GNAQ, GNA11, BAP1 and SF3B1 Mutations in Ophthalmic Melanocytomas. While no other alterations have been reported to date, whether GNAQ/GNA11 mutations are the sole driver alteration in uveal melanocytoma is not clear. Furthermore, the presence and extent of chromosomal copy number alterations in melanocytomas are not well-studied. Thus, it is crucial to determine the molecular pathogenesis of uveal melanocytomas, whether they harbor additional mutations and copy number variations, and whether molecular testing can be used to distinguish them from uveal melanomas when histopathologic findings are equivocal.

In this study, we evaluated the genetic alterations of iris and ciliary body melanocytoma and melanoma using a capture-based targeted next-generation sequencing panel, which can detect single nucleotide variants, select gene fusions, and chromosomal copy number variations, and evaluated the association between the molecular findings and clinicopathologic features.

Materials and MethodsThis study was conducted as a retrospective, cross-sectional study in accordance with the Declaration of Helsinki and Good Clinical Practice Guidelines, under the approval of the University of California San Francisco (UCSF) and Wills Eye Hospital (WEH) Institutional Review Boards. We searched pathology archives for melanocytic tumors involving the iris and/or ciliary body of adult patients (18 years of age or older) managed with local resection or enucleation between 1994 and 2018. Four ophthalmic pathologists (TM, RCE, MMB, MP) reviewed all cases with available pathology material and a consensus diagnosis of melanocytoma, melanocytoma with atypia (concerning for melanoma arising in melanocytoma), or melanoma was achieved for all cases. Each tumor was assessed for the amount of pigmentation and mitotic activity, and the presence of small or prominent nucleoli, intranuclear cytoplasmic pseudoinclusions, and necrosis. Presence of each cell type (large polyhedral, epithelioid, spindle) were categorized as absent, present (50% or less), and predominant (more than 50%). Tumors comprised of polyhedral intensely pigmented cells with or without spindle cells demonstrating low nucleus-to-cytoplasmic ratio, finely dispersed chromatin, and small or inconspicuous nucleoli on bleached preparations were classified as melanocytoma (Figure 1a-b). Tumors composed entirely of any proportion of spindle B and epithelioid melanoma cells demonstrating significant cytologic atypia, large nucleoli, significant mitotic activity, or infiltrative growth pattern were classified as melanoma (Figure 1d). Tumors with overall morphologic features of melanocytoma but containing focal cytologic features concerning for melanoma were classified as melanocytoma with atypia (Figure 1e-f). All melanomas were classified and staged in accordance with the pathologic staging criteria per American Joint Committee on Cancer (AJCC) 8th edition.15

Kivelä T SE, Grossniklaus HE, Jager MJ, Singh AD, Caminal JM. Uveal melanoma. In: MK AMESGFBDBRW, ed. AJCC cancer staging manual. 8th edition ed. New York: Springer; 2017: 805-817.

Extent of involvement of the adjacent tissues (trabecular meshwork, choroid, and extrascleral extension) was recorded. Margin status was assessed. Tumors with sufficient tissue for molecular analysis were included in the study. Tumors with only necrotic material (two melanocytomas), one metastatic cutaneous melanoma, and five tumors that failed sequencing due to insufficient or poor quality of the extracted DNA (two melanomas, one melanocytoma with atypia, two melanocytomas) were excluded from the study.

Figure 1Representative example of an anterior uveal melanocytoma composed of polyhedral intensely pigmented cells on H&E-stained sections. Nuclear details are difficult to see on unbleached hematoxylin & eosin-stained sections (a) while inconspicuous nucleoli (arrow) can be seen on bleached sections (b). Representative examples of anterior uveal melanocytoma with atypical features (c, d). Melanocytoma demonstrates lighter pigmentation where nucleoli (blue arrows) can be seen on hematoxylin & eosin-stained sections (c) or mitotic activity (blue arrow) can be seen on bleached sections (d). Areas of necrosis (blue arrow) can be seen in melanocytoma and do not represent malignancy (e). Representative example of a mixed spindle and epithelioid cell melanoma, with large nucleoli (black arrows) essentially negative for pigmentation, and mitotic activity (blue arrows) (f).

Targeted Next-Generation SequencingThe UCSF500 Cancer Panel is a clinically validated next-generation sequencing assay performed at the CLIA-certified UCSF Clinical Cancer Genomics Laboratory. The assay is validated for both formalin-fixed paraffin embedded (FFPE) tissues and for fresh samples such as fine-needle aspiration materials.16Afshar A.R. Damato B.E. Stewart J.M. et al.Next-Generation Sequencing of Uveal Melanoma for Detection of Genetic Alterations Predicting Metastasis. The test is available for all providers, including non-UCSF, and can be performed either as a tumor-only or tumor-normal paired test. For this study, genomic DNA was extracted from FFPE tumor tissue using the QIAamp DNA FFPE Tissue Kit (Qiagen) according to the manufacturer’s protocol. The regions with highest tumor content, specifically including areas of cytologic atypia if applicable, were carefully macrodissected for analysis. Relative tumor content of the lesional tissue used for DNA extraction and sequencing analysis was quantitatively estimated by microscopic assessment. Capture-based next-generation DNA sequencing was performed using an assay that targets all coding exons of 479 cancer-related genes, select introns and upstream regulatory regions of 47 genes to enable detection of structural variants including gene fusions, and DNA segments at regular intervals along each chromosome to enable genome-wide copy number and zygosity analysis, with a total sequencing footprint of 2.8 Mb (UCSF500 Cancer Panel; Supplementary Table 1).16Afshar A.R. Damato B.E. Stewart J.M. et al.Next-Generation Sequencing of Uveal Melanoma for Detection of Genetic Alterations Predicting Metastasis.,17Kline C.N. Joseph N.M. Grenert J.P. et al.Targeted next-generation sequencing of pediatric neuro-oncology patients improves diagnosis, identifies pathogenic germline mutations, and directs targeted therapy. Specifically, this assay covers all exons of the GNAQ, GNA11, SF3B1, EIF1AX, BAP1, SRSF2, and PLCB4 genes frequently altered in uveal melanocytic neoplasms.18Robertson A.G. Shih J. Yau C. et al.Integrative Analysis Identifies Four Molecular and Clinical Subsets in Uveal Melanoma.

Multiplex library preparation was performed using the KAPA Hyper Prep Kit (Roche) according to the manufacturer’s specifications. Hybrid capture of pooled libraries was performed using a custom oligonucleotide library (Nimblegen SeqCap EZ Choice). Captured libraries were sequenced as paired-end 100 bp reads on an Illumina HiSeq 2500 instrument. Sequence reads were mapped to the reference human genome build GRCh37 (hg19) using the Burrows-Wheeler aligner (BWA). Recalibration and de-duplication of reads was performed using the Genome Analysis Toolkit (GATK), enabling accurate allele frequency determination and copy number assessment. Coverage and sequencing statistics were determined using Picard CalculateHsMetrics and Picard CollectInsertSizeMetrics. Single nucleotide variant and insertion/deletion mutation calling was performed with Mutect, Unified Genotyper, and Pindel. Variant annotation was performed with Annovar. Single nucleotide variants, insertions/deletions, and structural variants were visualized and verified using Integrated Genome Viewer. Genome-wide copy number analysis based on on-target and off-target reads was performed by CNVkit and visualized using Nexus Copy Number and NxClinical (Biodiscovery). According to the test performance data generated during initial clinical validation, for samples with at least 25% lesional cells, the sensitivity for detection of copy number changes is 100%, and the sensitivity and specificity for single nucleotide variants and small indels (≤40 bp) are >99% and >98%, respectively, with sequencing coverage of >500x.

Single nucleotide variants and small insertions-deletions were included in this analysis only if there were at least five mutant reads and when they were deemed likely to be oncogenic and contributing to tumorigenesis. Variants of uncertain significance (missense variants with uncertain functional or oncogenic significance) were excluded. Cases with mutational profile predominated by C>T/G>A transitions and containing the pathognomonic CC>TT/GG>AA dinucleotide substitutions were considered to harbor mutational signature associated with ultraviolet light exposure (UV mutational signature)19Alexandrov L.B. Nik-Zainal S. Wedge D.C. et al.Signatures of mutational processes in human cancer..Statistical analysis

Biostatistical analysis was performed in Stata Version 16 (StataCorp, College Station, Texas, USA). Comparison of clinical and histopathologic features stratified by molecular alterations were performed using Fisher’s Exact test or Mann-Whitney test, as appropriate. Local progression-free survival was analyzed by Kaplan-Meier curves using Log-Rank test. Tumors were divided into two groups as those with GNAQ or GNA11 mutations as the sole oncogenic alteration (n=20), and those with additional likely oncogenic mutations and/or chromosomal copy number alterations known to be associated with melanomagenesis beyond GNAQ/GNA11 mutations (n=13). A p-value cutoff of 0.05 was used to assess statistical significance.

ResultsPatient DemographicsA total of 35 melanocytic tumors primarily involving iris and ciliary body (34 local resection and 1 enucleation specimen) were included in the study. Clinical, pathologic, and molecular features are summarized in Figure 2. There were eight melanocytomas (seven ciliary body and one iris) from three males and five females who were diagnosed at a median age of 51 years (range 30-79 years). There were eight melanocytomas with focal atypia (six ciliary body and two iris) from four males and four females who were diagnosed at a median age of 54 years (range 26-57 years). Additionally, there were 19 melanomas (11 ciliary body and 8 iris) from six males and 13 females who were diagnosed at a median age of 57 years (range 19-78 years).

Figure 2Oncoprint summary table of the clinical, pathologic, and genetic features of the anterior uveal melanocytic lesions

Pathologic FeaturesClinical and pathologic features based on the histologic diagnosis are summarized in Table 1. Detailed pathological information is presented in Supplementary Table 2. Briefly, all eight melanocytomas consisted predominantly of large, polyhedral cells with abundant, intensely pigmented cytoplasm and small, oval nuclei with only small nucleoli. Large nucleoli and mitotic activity were absent.

Table 1Clinical and pathologic features of all anterior uveal melanocytic tumors in the study cohort

* Statistical analyses were performed comparing melanocytoma (with or without atypia) versus melanoma

ˆ 40 HPF: 40 high-power field refers to 40 consecutive fields using 400x magnification, which corresponds to a total area of 10 square millimeters.

# pTNM Stage, according to AJJC 8th edition, is based on size (basal diameter and height) for ciliary body tumors, and is based on size, ciliary body involvement and presence of glaucoma for iris tumors. A “presumed” stage is provided for melanocytomas for comparison purposes.

Percentages may add up to >100% due to rounding

ϕ : Statistical analysis was performed using Mann-Whitney test.

χ: Statistical analysis performed by Fisher’s Exact test.

Among eight cases of melanocytoma with atypia, intense pigmentation was patchy in four cases with variable amounts of large polyhedral cells. In the areas of atypia suspicious for a melanoma component, seven (88%) had predominantly spindle cells and one had mixed spindle and epithelioid cells (case MC-A-01). Mitotic figures (ranging from 1 to 4 mitoses in 40 high-power fields) were identified in three tumors (38%). All tumors involved both iris and ciliary body, but tumor epicenter was considered as iris in three tumors corresponding to stage pT2 per AJCC 8th edition15

Kivelä T SE, Grossniklaus HE, Jager MJ, Singh AD, Caminal JM. Uveal melanoma. In: MK AMESGFBDBRW, ed. AJCC cancer staging manual. 8th edition ed. New York: Springer; 2017: 805-817.

.Melanomas consisted of a variable proportion of spindled and epithelioid cells with no or only rare polyhedral cells with dense cytoplasmic pigment. Anaplastic features with prominent cytologic atypia were seen in seven (37%) and large nucleoli were present in six (32%) cases. Eight melanomas (42%) were limited to iris and two tumors (11%) were limited to ciliary body. Seven of eight iris melanomas involved inferior quadrant. Nine tumors (47%) involved both iris and ciliary body, and tumor epicenter was interpreted as ciliary body for all 15

Kivelä T SE, Grossniklaus HE, Jager MJ, Singh AD, Caminal JM. Uveal melanoma. In: MK AMESGFBDBRW, ed. AJCC cancer staging manual. 8th edition ed. New York: Springer; 2017: 805-817.

. One ciliary body melanoma (case MM-11) showed extension into the adjacent choroid as well as a 7-mm anterior extrascleral extension. Among eight primary iris tumors, six were staged as pT1a, one as pT1b, and one with elevated intraocular pressure was staged as pT1c. Among 11 primary ciliary body tumors, nine (82%) were stage pT1b, one was stage pT3b, and one with a large extrascleral extension was stage pT4e.Genetic Analysis

Detailed results of the genetic analysis are presented in Supplementary Table 3. The median tumor content in the study cohort was 60% (range 20-90%) with only two cases (one melanocytoma and one melanoma) estimated to have less than the target >25% tumor fraction. Melanomas had only marginally higher tumor content than melanocytomas (63% +/- 3% vs 57% +/- 3%, p=0.053). Targeted next-generation sequencing generated an average of 19.3 million unique reads per tumor sample with an average mean on-target coverage of 366x. This analysis demonstrated mutually exclusive GNAQ or GNA11 mutations in 32/35 tumors (91%), which were uniformly known activating/oncogenic missense variants at codon p.Q209. There were seven cases with GNA11 mutations, which were uniformly found in cases with histologic diagnosis of melanoma.

Seven of eight melanocytomas (88%) and all eight melanocytomas with atypia contained GNAQ p.Q209L hotspot mutation, while no GNAQ or GNA11 mutation was identified in the remaining melanocytoma. None of the 16 melanocytomas showed any additional likely oncogenic alterations, including variants involving BAP1, SF3B1, EIF1AX, SRSF2, PLCB4, BRAF, KRAS, NRAS, or NF1. CYSLTR2, which is mutated in some uveal melanoma without GNAQ/GNA11 or PLCB4 mutations, was not included in our panel. None of the tumors showed focal amplifications or deep deletions, and all tumors had balanced diploid genomes without chromosomal copy number gains or losses.

Seventeen of 19 melanomas (90%) demonstrated GNAQ p.Q209P (n=10) or GNA11 p.Q209L (n=7) hotspot mutations. Thirteen of these melanomas, showed additional oncogenic mutations (n=9) and/or copy number alterations known to be recurrent in uveal melanoma including monosomy 3p or trisomy 6p or 8q (n=8). Representative examples of genome-wide copy number plots are provided in Supplementary Figure 1. One ciliary body melanoma (MM-18) showed BAP1 p.R114fs mutation with monosomy 3 without a detectable alteration in GNAQ or GNA11 despite adequate depth of coverage. One iris melanoma (MM-19, confirmed to be primary) without a detectable alteration in GNAQ or GNA11 harbored pathogenic mutations in BRAF (p.V600K), NF1, NFKBIE, RB1, and TERT promoter with an associated UV mutational signature, resembling cutaneous melanoma.19Alexandrov L.B. Nik-Zainal S. Wedge D.C. et al.Signatures of mutational processes in human cancer. UV mutational signatures were also identified in three additional iris melanomas and one ciliary body melanoma with GNAQ/GNA11 mutations. All four iris melanomas with UV mutational signature were located in the inferior quadrant, and the one ciliary body melanoma with UV mutational signature was located in the temporal quadrant.

Among seven iris melanomas with GNAQ/GNA11 mutations, additional oncogenic alterations were seen in five tumors, including 3 different EIF1AX exon 1 mutations in two tumors, a truncating EP300 mutation in one, an inactivating missense PTEN mutation in one, and gains of chromosomes 2, 4, and 6 in one tumor. The remaining two tumors did not demonstrate any detectable oncogenic alterations beyond GNAQ/GNA11. Among 10 ciliary body melanomas with GNAQ/GNA11 mutations, additional oncogenic alterations were seen in eight tumors, including truncating BAP1 mutations associated with monosomy 3 (n=2), EIF1AX p.P2L mutation associated with monosomy 3 and UV mutational signature (n=1), EIF1AX p.P2H mutation (n=1), SRSF2 p.Y92_D97del small in-frame deletion, and CREBBP p.M481fs frameshift mutation (n=1), and chromosomal copy number changes without additional oncogenic single nucleotide variants involving genes beyond GNAQ/GNA11 (n=3).

Overall, 20 tumors harbored GNAQ/GNA11 mutations as the solitary oncogenic alterations. These were termed “tumors with solitary GNAQ/GNA11 mutation”. Twelve cases with GNAQ/GNA11 mutations harbored additional oncogenic mutations and/or chromosomal copy number alterations frequently seen in uveal melanomas (i.e. monosomy 3 and/or gain of 6p or 8q). One tumor without an identifiable GNAQ/GNA11 mutation harbored BAP1 mutation and monosomy 3, and together this group (n=13) was termed “tumors with GNAQ/GNA11 mutation plus additional oncogenic alterations.” The first group (“tumors with solitary GNAQ/GNA11 mutation”) contained all melanocytomas and five melanomas, whereas the second group (“tumors with GNAQ/GNA11 mutation plus additional oncogenic alterations”) contained 13 melanomas and no melanocytomas. The tumors with additional oncogenic alterations were less likely to contain large polygonal cells and diffuse, intense pigmentation, and more likely to contain anaplasia, large nucleoli, elevated mitotic activity, and positive margin. Clinical and pathologic features of these two groups are summarized in Table 2. One melanocytoma without a detectable GNAQ/GNA11 alteration, and one iris melanoma genetically resembling cutaneous melanoma were excluded from this analysis.

Table 2Clinical and pathologic features of anterior uveal melanocytic tumors with only GNAQ/GNA11 mutations versus tumors with GNAQ/GNA11 mutations plus additional oncogenic alterations.

* The group “tumors with GNAQ/GNA11 mutation plus additional oncogenic alterations” includes one tumor (MM-18) harboring a BAP1 mutation without an identifiable GNAQ/GNA11 alteration.

$ One melanocytoma (MC-08) with no detectable GNAQ/GNA11 or other oncogenic alterations was excluded from the above analysis, as was one melanoma (MM-19) due to having a cutaneous melanoma-like genetic profile with BRAF p.V600E, TERT promoter, and other mutations.

ˆ 40 HPF: 40 high-power field refers to 40 consecutive fields using 400x magnification, which corresponds to a total area of 10 square millimeters.

# pTNM Stage, according to AJJC 8th edition, is based on size (basal diameter and height) for ciliary body tumors, and is based on size, ciliary body involvement and presence of glaucoma for iris tumors. A “presumed” stage is provided for melanocytomas for comparison purposes.

ϕ : Statistical analysis was performed using Mann-Whitney test.

χ: Statistical analysis performed by Fisher’s Exact test.

Treatment and Clinical Outcome

Thirty-three tumors were managed with iridocyclectomy (94%), one with sector iridectomy (3%), and one eye harboring ciliary body melanoma with extrascleral extension underwent enucleation (3%). The median follow-up period was 78 months for patients with melanocytoma (range 12-197 months), 14 months (range 0-80 months) for melanocytoma with atypia, and 38 months (range 0-144 months) for melanoma. None of the melanocytomas, with or without atypia, showed recurrence/progression or metastasis in the follow-up period. Five melanomas (1 iris, 4 ciliary body) showed local recurrence with a median time to recurrence of 17.4 months (range 3.4- 62 months), and four were treated with enucleation and one with proton beam irradiation at time of local recurrence. Metastasis (to liver) was documented in only one ciliary body melanoma (case MM-01) at 86.2 months, which had previously recurred locally at 17.4 months for which secondary enucleation had been performed. This tumor was mixed spindled and epithelioid histologic subtype with a pathologic stage of pT1b at the time of iridocyclectomy, and harbored GNAQ hotspot mutation and a BAP1 truncating mutation with loss of the wild-type allele due to monosomy 3.

Melanoma diagnosis was associated with worse local progression-free survival (LPFS) in comparison to melanocytomas (p=0.033, Log Rank test), and there was no difference in the LPFS between melanocytomas with or without atypia given the lack of recurrence in either group (Figure 3a). LPFS was not different between tumors with or without positive margins (p=0.144, Log Rank test), or between tumors that would have qualified for pTNM stage 1 or higher (p=0.372). Overall, tumors with GNAQ/GNA11 mutation plus additional oncogenic alterations had worse LPFS in comparison to tumors with solitary GNAQ/GNA11 mutations (p=0.0029, Log Rank test, Figure 3b). Among tumors with histologic diagnosis of melanoma, the presence of additional oncogenic alterations beyond GNAQ/GNA11 was associated with a trend towards worse LPFS, but this did not reach statistical significance (p=0.076, Log Rank test, Figure 3c).

Figure 3Kaplan-Meier survival results for local progression-free survival (LPFS) in anterior uveal melanocytic neoplasms. Stratification of the entire patient cohort based on the histologic diagnoses as melanocytoma, melanocytoma with atypia, or melanoma (a). Stratification of the entire patient cohort (excluding MC-08 and MM19) based on tumors with solitary GNAQ/GNA11 mutation versus tumors with GNAQ/GNA11 mutation plus additional oncogenic alterations (including BAP1, SF3B1, EIF1AX, PTEN, EP300 mutations and/or chromosomal copy number alterations such as monosomy 3, gain of 6p or 8q), irrespective of histologic diagnosis (b). Stratification of patients with histologic diagnosis of melanoma based on solitary GNAQ/GNA11 mutation versus those with GNAQ/GNA11 mutation plus additional oncogenic alterations (c).

DiscussionIn this study, we have analyzed the genetic features of iris and ciliary body melanocytoma and melanoma as diagnosed by histopathologic criteria, and evaluated the value of molecular testing for the diagnosis and prognostication of these lesions. We have identified a GNAQ or GNA11 mutation in 32 of 35 cases (91%), which is similar to the previous reports of high incidence of these mutations in uveal melanocytic neoplasms.13Mudhar H.S. Doherty R. Salawu A. Sisley K. Rennie I.G. Immunohistochemical and molecular pathology of ocular uveal melanocytoma: evidence for somatic GNAQ mutations.,18Robertson A.G. Shih J. Yau C. et al.Integrative Analysis Identifies Four Molecular and Clinical Subsets in Uveal Melanoma.,20Scholz S.L. Moller I. Reis H. et al.Frequent GNAQ, GNA11, and EIF1AX Mutations in Iris Melanoma., 21Van Raamsdonk C.D. Bezrookove V. Green G. et al.Frequent somatic mutations of GNAQ in uveal melanoma and blue naevi., 22Van Raamsdonk C.D. Griewank K.G. Crosby M.B. et al.Mutations in GNA11 in uveal melanoma.Prior studies demonstrated that uveal nevi and melanomas with wildtype GNAQ/GNA11 harbor alterations in CYSLTR2 or PLCB4.23Moore A.R. Ceraudo E. Sher J.J. et al.Recurrent activating mutations of G-protein-coupled receptor CYSLTR2 in uveal melanoma., 24Nell R.J. Menger N.V. Versluis M. et al.Involvement of mutant and wild-type CYSLTR2 in the development and progression of uveal nevi and melanoma., 25Johansson P. Aoude L.G. Wadt K. et al.Deep sequencing of uveal melanoma identifies a recurrent mutation in PLCB4. CYSLTR2 is a G-protein coupled receptor signaling via Gαq and G11, the proteins encoded by GNAQ and GNA11, and PLCB4 encodes phospholipase C β4, an immediate downstream effector of Gαq and G11. All four alterations lead to constitutive activation of the Gαq->PKC pathway.26Ma J. Weng L. Bastian B.C. Chen X. Functional characterization of uveal melanoma oncogenes. While the employed sequencing panel included PLCB4, which was wildtype in all cases in this study, CYSLTR2 was not included, and thus may be involved in the tumorigenesis of the couple remaining tumors in our cohort without identifiable GNAQ, GNA11, or PLCB4 mutations. Given the limited amount of tissue, unfortunately w