Uveal Melanocytoma without GNAQ/GNA11 Mutations

Abstract

Introduction

Uveal melanocytomas are benign melanocytic proliferations characterized histologically by large polyhedral cells, abundant intracytoplasmic melanin, small centrally located nuclei, and low nuclear-to-cytoplasmic ratios (N:C). Activating missense changes of the G-protein subunits, GNAQ/GNA11, drive uveal melanocytic proliferation and are characteristic molecular mutations found in intraocular nevi including melanocytomas, as well as malignant uveal melanomas. Sequencing of uveal melanocytic proliferations from biopsies or enucleations identifies these mutations and complements traditional histochemical approaches to classify and diagnose the uveal melanocytic neoplasm as well as offering valuable clinical prognoses about patient outcomes.

Case Presentation

A 56-year-old female presented with gradual, painless vision loss in the left eye. On exam, a large (11.1 × 12.8 × 10.6 mm by ocular ultrasound), hyperpigmented ciliochoroidal mass was found in the superior nasal quadrant abutting the lens. The patient elected for enucleation given the tumor’s large size and vision loss in the affected eye. An fine needle aspiration (FNA) biopsy of the fresh mass, collected immediately after enucleation, revealed no mutations in the 7-gene DecisionDX-UMSeq panel. Histologic and immunohistochemical evaluation of pupil-optic nerve cross-sections from the formalin fixed paraffin embedded enucleation revealed a ciliary body melanocytoma. DNA extracted from the melanocytoma enucleation cross-sections was sequenced for 197-clinically actionable tumor genes through the Stanford’s Actionable Mutation Panel for Solid Tumors (STAMP). The sequencing results confirmed no mutations in GNAQ, GNA11, and the other uveal melanoma genes tested via the DecisionDX-UMSeq panel but identified missense variants of unknown significance in three genes previously not reported in uveal melanocytic neoplasms.

Conclusion

To our knowledge, this is the first reported uveal melanocytoma lacking GNAQ/GNA11 oncogenic variants or other known uveal melanocytic neoplasm driver mutations. This case supports that there are additional to-be-identified molecular pathways and genes that drive uveal melanocyte proliferations.

Plain Language Summary

Mutations in GNAQ or GNA11 trigger uveal melanocytic proliferation and are found in the vast majority of intraocular nevi, melanocytomas, and melanomas. Here, we report a patient who underwent enucleation for a large pigmented melanocytic neoplasm arising from the ciliary body. Histopathologic examination of the enucleation showed that the ciliary body neoplasm was comprised of bland heavily pigmented melanocytes compatible with a melanocytoma. Surprisingly, a commercial targeted sequencing test (DecisionDx-UMseq) on an FNA sample collected from the tumor was negative for mutations in seven genes linked to uveal melanomas, including GNAQ and GNA11. Next generation sequencing of 197 cancer genes from the entire enucleation confirmed the results of the FNA sequencing test with absence of mutations in GNAQ, GNA11, and the other known uveal melanoma genes. Incidentally, the large cancer gene panel sequencing test identified a novel variant of unknown significance in NTRK2, a gene that shares common downstream signal transduction pathways with GNAQ/GNA11. The genetic profile of this current melanocytoma was unique when compared with the genetic profiles of 19 previously published uveal melanocytomas. In summary, our case strongly supports the existence of additional undocumented driver genes beyond GNAQ and GNA11 in the pathogenesis of uveal melanocytomas.

Introduction

In recent years, molecular testing has become important for classification and prognostication of uveal melanocytic neoplasms [1]. GNAQ/GNA11 mutations are found in both benign and malignant uveal melanocytic proliferations (Table 1) [2–4]. GNAQ and GNA11 encode for intracellular G-protein alpha subunits and are oncogenic, when mutated, by activating downstream Protein Kinase C mitogen-activated kinase, PI3-kinase/Akt, and YES-associated protein-related pathways [5–8]. By these mechanisms, GNAQ/GNA11 mutations are considered major contributors in the development of uveal melanocytic neoplasms where they are reported in 80–90% of cases [5, 9]. Less frequently in cases lacking GNAQ/GNA11 mutations, variants in genes closely associated to GNAQ/GNA11, such as the associated G-protein coupled receptor, CYSLTR2, and downstream effector, PLCβ4, have been identified [10–12].

Table 1.

Uveal melanocytoma cases including genetic analysis

Study	Cases	Location	GNAQ*	CYSLTR2*	Extended panel†	Other gene(s)*
Solomon et al. [2] 2022	16	Ciliary body	3/4	–	0	–
		Iridociliary	11/11	–	0	–
		Iris	1/1	–	0	–
Mudhar et al. [3] 2013	2	Ciliochoroidal	1/1	–	Lim/0	–
		Choroidal	1/1	–	Lim/0	–
Francis et al. [4] 2016	1	Iridociliary	1/1	–	Lim/0	–
Present Case 2024	1	Ciliary body	0	0	0	ABL1, NTRK2, TACC3

* Gene(s) of variant/mutation.

^† Gene variants of: BAP1, SF3B1, EIF1AX, SRSF2, PTEN, PLCB4, BRAF, CREBBP, KRAS, NRAS, and NF1.

Lim, some, but not all genes of category were investigated.

0, not present from tested criteria (negative result).

–, criteria not investigated.

Here, we report a case of ciliary body melanocytoma that lacks all previously reported driver mutations for uveal melanocytic neoplasms as determined by commercial DecisionDx-UMseq testing of melanocytoma fine needle aspiration (FNA) biopsy and confirmed by whole exome sequencing of the subsequent enucleated specimen. To our knowledge, this is the first melanocytoma reported without mutations in GNAQ, GNA11, CYSLTR2, or PLCβ4 and other characteristic uveal melanocytic proliferation genes [6, 10–12]. This finding supports that additional driver genes can promote uveal melanocytic neoplasms.

Case Information

A 56-year-old female presented with painless and progressive vision changes in the left eye (OS), worsening over 2 years. There was no significant past medical or family history. On ocular examination, her best-corrected visual acuity was 20/20 right eye (OD) and 20/60 OS. Her pupils were 4 mm, reactive to light, without relative afferent pupillary defect in both eyes. Intraocular pressures were 19 mm Hg OD and 20 mm Hg OS by applanation tonometry. Anterior segment examination with slit-lamp biomicroscopy revealed nuclear sclerosis grade I OD and grade III with opacification of the posterior lens in the superior nasal quadrant OS. Anterior chambers were deep and quiet in both eyes (OU). Dilated fundus examination revealed a large, hyperpigmented ciliochoroidal mass in the superior nasal quadrant abutting the lens OS. No subretinal fluid or orange pigment was observed. Optic disc examination showed cup disc ratio of 0.5 without disc swelling OU. Ocular ultrasound measured a large 11.1 × 12.8 × 10.6 mm ciliochoroidal mass with mid-to-high internal reflectivity, including cystic spaces. A clinical diagnosis of ciliochoroidal melanoma was made. The patient elected for enucleation of the left eye. Prior to formalin fixing of the enucleated eye for histopathology evaluation, a FNA of the pigmented mass from the enucleation was performed and concomitantly frozen; the cellular aspirate was then sent for genetic testing for uveal melanoma mutations and metastatic potential (DecisionDX-UM and DecisionDx-UMSeq, Castle Biosciences).

The enucleated eye measured 22.5 mm antero-posteriorly, with an oval shaped, hyperpigmented ciliary body mass abutting the posterior surface of the lens (Fig. 1a). After formalin fixation, paraffin embedding, sectioning and staining including bleach preparations, microscopic evaluation revealed a large ciliochoroidal tumor consisting of heavily pigmented cells with small, bland, round-oval nuclei (Fig. 1b). No subretinal fluid or intra/extrascleral extension of the melanocytic lesion was identified. Melanin bleaching using potassium permanganate, oxalic acid, and hydrogen peroxide revealed polyhedral cellular morphology and further confirmed benign cytologic features including centrally located small, bland nuclei, normal N:C ratio, and no mitotic figures (Fig. 1c). Immunohistochemistry confirmed the tumor’s melanocytic lineage: nuclei were positively stained for SOX10 (Fig. 1d) and cytoplasm with Melan-A (Fig. 1e). To exclude ciliary body adenocarcinoma and epithelial origins, PAX8 (Fig. 1f) and broad-spectrum cytokeratin staining, AE1/AE3/CAM5.2 (Fig. 1g), were used respectively with negative results.

Fig. 1.

Histology of enucleated eye and tumor. a Low-power magnification of globe with hematoxylin and eosin (H&E) staining. b High-power magnification H&E showing sheets of heavily pigmented cells. c High-powered field post-melanin bleaching reveals plump polyhedral cells with bland, oval nuclei. d Staining of SOX10 with positive, red nuclear pigmentation. e Cytoplasmic staining by Melan-A (red) among pigmented cells. f Absence of PAX8 nuclear staining. g Absence of broad-spectrum cytokeratin staining using AE1/AE3 and CAM5.2. h Preservation of choriocapillaris (white arrows) and Bruch’s membrane (red arrows); also, cystic space with colloid-like staining (blue arrow).

Additionally, ancillary staining showed retained nuclear BAP1 and absence of preferentially expressed antigen in melanoma (PRAME). Collectively, these findings confidently supported the diagnosis of a benign ciliary body melanocytoma.

Other points of histologic interest involved the choriocapillaris which was preserved, and an intact Bruch’s membrane (Fig. 1h). Additionally, numerous cystic spaces were observed, some containing eosinophilic colloidal-like staining (Fig. 1a,h), similar to those reported in prior ciliary body melanocytomas [13].

From the FNA biopsy collected at the time of enucleation, two commercial molecular tests were performed (DecisionDX-UM and DecisionDX-UMSeq, Castle Biosciences, Friendswood TX). The DecisionDX-UM test showed a Class 1 genetic prognostic profile [14]. Surprisingly, the DecisionDX-UMSeq test found no mutations in any of the seven genes on the panel, all linked to pathogenesis and prognosis of uveal melanocytic neoplasms: GNAQ, GNA11, CYSLTR2, PLCβ4, SF3B1, EIF2AX, and BAP1 (Table 2). Given the absence of uveal melanocytic mutations in the molecular testing (FNA), despite the identification of a large melanocytoma by histology in the enucleation, DNA was extracted from formalin fixed paraffin embedded sections of the enucleation and sent for an exome sequencing panel of 197 genes via Stanford’s Solid Tumor Actional Mutation Panel (STAMP) v4.3.0; https://stanfordlab.com/test-details/s/STAMPT.html (Supplemental, S1). Consistent with the DecisionDX-UMseq results, no mutations were found in GNAQ/GNA11 or the other uveal melanoma genes assessed in the DecisionDX-UMSeq panel. However, three single nucleotide missense variants of unknown significance were identified in ABL1 1985C>T (NM_005157.4), NTRK2 1480G>A (NM_006180.3), and TACC3 1793C>G (NM_006342.2) (Supplemental, S2), resulting in amino acid substitutions of P662L, A494T, and A598g, respectively (Table 3). Bioinformatic analyses tools, PolyPhen-2, Sorting Intolerant From Tolerant (SIFT), and v1.7 of Combined Annotation-Dependent Depletion (CADD), were used to assess the influence of each variant on the respective protein product (Table 4) [15]. Of the three variants of unknown significance, NTRK2 (c.1480 G>A) was predicted to encode a deleterious protein, i.e., tropomyosin receptor kinase B (TrkB), by all three models (Table 4).

Table 2.

Molecular sequencing profile via Castle Biosciences

Specimen	Analysis	Genes tested	Results	Comment
Fine-needle aspiration	DecisionDx-UMSeq	GNAQ, GNA11, CYSLTR2, PLCB4, SF3B1, EIF2AX, BAP1	No mutation identified	Absence of mutation in GNAQ, GNA11, CYSLTR2, or PLB4 does not rule out a uveal melanoma diagnosis. While absence of mutation in BAP1 has been associated with lower risk of metastasis
Fine-needle aspiration	DecisionDx-UM	CDH1, ECM1, EIF1B, FXR1, HTR2B, ID2, LMCD1, LTA4H, MTUS1, RAB31, ROBO1, SATB1	Class 1A; No mutation identified	Class 1 molecular signature is associated with a low risk of near term (within 5 years) clinical metastasis. A discriminant value ≥0.1 is reported with normal confidence *MRPS21, RBM23, SAP130 were used as controls

Genomic positions are given in reference to GRCh37.

Table 3.

Stanford actionable mutation panel for solid tumors (STAMP)

Variant	Amino acid change	Variant Allele fraction	Transcript	Pathogenicity
ABL1 c.1985 C>T	P662L	47%	NM_005157.4	Unknown significance
NTRK2 c.1480 G>A	A494T	43%	NM_006180.3	Unknown significance
TACC3 c.1793 C>G	A598G	39%	NM_006342.2	Unknown significance

Genomic positions are given in reference to GRCh37.

Table 4.

Predictive modeling of variant’s influence on protein product

Variants	PolyPhen-2	SIFT	CADD (v1.7)
ABL1 c.1985 C>T	Benign	Tolerated	Likely benign
NTRK2 c.1480 G>A	Probable damage	Deleterious	Deleterious
TACC3 c.1793 C>G	Probable damage	Tolerated	Likely benign

Genomic positions are given in reference to GRCh37 (hg19).

No pathogenic or likely pathogenic variants identified.

Estimated tumor mutation burden (TMB) 1.1 mutations per megabase.

Discussion

Our case highlights the importance and value of molecular testing in evaluating uveal melanocytic neoplasms. The commercial tests, DecisionDX-UM and DecisionDX-UMseq, on the FNA biopsy as well as the 197-clinically actionable tumor gene whole exome sequencing of the enucleation supported the favorable prognosis of the large ciliary body mass. The retained BAP1 and negative PRAME immunostains also supported favorable prognosis for the lesion. The lesion had classic histopathologic features and immunohistochemical profile for uveal melanocytoma, but surprisingly, no mutations in any of the seven genes, typically found in uveal melanocytic neoplasms, were detected via sequencing of an FNA sampling of the mass at the time of enucleation, and of the entire mass extracted from formalin fixed paraffin embedded sections of the enucleation. We compared the molecular profile of our current melanocytoma with the other 19 uveal melanocytomas whose molecular profiles have been previously reported in the literature. In a molecular genetic analysis of 16 uveal melanocytomas, 15 out of 16 had identifiable GNAQ mutations. All 16 melanocytomas were negative for five other common uveal melanocytic tumor mutations including GNA11, BAP1, SF381, EIF1AX, and BRAF, but mutations in CYSLTR2 were not tested in that study [2]. Interestingly, the single uveal melanocytoma case in that study without GNAQ/GNA11 mutations also originated from the ciliary body like our present case (Table 1); however, the possibility exists for a CYSLTR2 mutation in that ciliochoroidal melanocytoma, given the omission of CYSTLTR2 in their sequencing panel. Molecular profiles of three uveal melanocytomas reported in other studies were all found to have GNAQ mutations [3, 4]. From this, we suspect our current uveal melanocytoma carries a novel genetic profile – absence of mutations in all seven uveal melanoma genes – compared to the 19 uveal melanocytoma lesions whose molecular profiles were previously evaluated. Interestingly, one case of non-pigmented epithelioid and spindle cell choroidal melanoma evaluated via “cBioPortal For Cancer Genomics (TCGA)” also lacked GNAQ/GNA11 mutations. However, while this melanoma (TCGA-VD-A8KJ) had numerous chromosomal copy number variations, our currently reported case did not, as documented with the STAMP analysis [16].

Our whole exome sequencing of 197 tumor genes in the enucleation identified three single nucleotide variants of unknown significance that were not previously reported in any uveal melanocytic neoplasms. These could be passenger mutations commonly found by sequencing. However, the NTRK2 (A494T) variant protein product was predicted to be deleterious by multiple different computational algorithms: PolyPhen-2, SIFT, CADD (v1.7). In cell culture experiments, products of NTRK2 and GNAQ/GNA11 activate PKC and the Protein Kinase C mitogen-activated kinase pathway, raising the possibility that NTRK2 may play a similar oncogenic role in driving uveal melanocytic neoplasms [5, 17–19]. Future studies are required to distinguish whether the NTRK2 variant is a passenger mutation or a causal oncogene in uveal melanocytic proliferation.

Conclusion

We described a patient carrying a large ciliochoroidal pigmented mass that raised clinical concerns for melanoma. Enucleation was performed and histopathologic evaluation identified a ciliary body melanocytoma. An FNA biopsy post-enucleation was commercially sequenced using Castle Bioscience DecisionDX-UM and DecisionDX-UMseq, which supported a benign melanocytic process. Surprisingly, no mutations typically linked to uveal melanocytic proliferations were detected. Whole exome sequencing of the enucleation confirmed the absence of mutations in the seven uveal melanoma genes, but identified three variants of unknown significance, namely ABL1 (P662L), TACC3 (A598G), and NTRK2 (A494T), the last of which was predicted to be deleterious in computational modeling algorithms. Our case supports that other unknown mutations may be involved with driving the pathogenesis of uveal melanocytic neoplasms.

Acknowledgments

The attributes of this case would not have been possible without Stanford’s Actionable Panel for Solid Tumors (STAMP) and the staff involved.

Statement of Ethics

This study was performed in accordance with the Declaration of Helsinki. This human study was approved by the Research Compliance Office at Stanford University. Approval: IRB#54760. All adult participants provided written informed consent for publication in this report. Written informed consent was obtained from the individual(s) for publication of the details of their medical case and any accompanying images.

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

This study was supported by The Alan and Irene Adler Ocular Cancer Research Initiative.

Author Contributions

S.T.K.: wrote/edited text, created tables/figures, conducted predictive modeling of variants. R.F.: text review, editing and literature review. C.G.: immunohistochemistry image prep. M.B.: text and literature review. D.S.M.: text review. P.M.: patient care and wrote and edited case. J.H.L.: wrote and edited text, provided histopathological analysis, and supervised the project.

Funding Statement

This study was supported by The Alan and Irene Adler Ocular Cancer Research Initiative.

Data Availability Statement

Due to the HIPAA Privacy Rule, no additional data has been made available outside of this report. All pertinent data have been shared within this case report.