Resident Short Reviews Inflamed conjunctival nevi (ICN) may suggest malignancy because of their rapid growth and atypical histology. The objective of this study was to characterize the diagnostic features of ICN. A retrospective, nonrandomized study of 13 patients with ICN was conducted. A scoring method was developed based on histology and immunopathological parameters. The presence of epithelial solid or cystic inclusions and the preservation of goblet cells are consistent with a benign melanocytic lesion. Periodic acid Schiff stain and immunohistochemistry to identify the epithelial component prove helpful in the differential diagnosis for melanoma. Polyclonal lymphoid infiltrate and benign cytological features of ICN exclude a diagnosis of lymphoma. Despite the presence of immunoglobulin 4 positive plasma cells in the lesions, ICN does not meet the diagnostic criteria for immunoglobulin 4 related disease. Most patients with ICN are young. The treatment for ICN is complete excision, and the prognosis is excellent. (Arch Pathol Lab Med. 2014;138:1242 1246; doi: 10.5858/arpa.2013-0245-RS) Among the benign intraepithelial melanocytic proliferations, conjunctival nevi (CN) are the most common conjunctival neoplasm. 1 Conjunctival nevi manifest as both intraepithelial and compound melanocytic lesions. Grossly, the lesion is located in the bulbar conjunctiva and has various pigmentation and shapes, including melanotic or amelanotic and flat or raised nodules. The CN are often biopsied because of their rapid growth rate and malignant potential. Zamir et al 2 studied 63 patients who presented with a history of moderately or rapidly growing lesions or congested conjunctival lesions. The study findings suggested that growth of the lesions was due to inflammatory infiltration and cystic enlargement. Thus, histopathological Accepted for publication September 24, 2013. From the Department of Pathology and Genomic Medicine, Houston Methodist Hospital, Houston, Texas (Drs Choi and Chévez-Barrios); and the Department of Pathology and Laboratory Medicine, Weill Cornell Medical College, New York, New York (Dr Chévez-Barrios). The authors have no relevant financial interest in the products or companies described in this article. Reprints: Patricia Chévez-Barrios, MD, Department of Pathology and Genomic Medicine, Houston Methodist Hospital, 6565 Fannin St, Ste M227, Houston, TX 77030 (e-mail: pchevez-barrios@ houstonmethodist.org). Inflamed Conjunctival Nevi Histopathological Criteria Eunice K. Choi, MD; Patricia Chévez-Barrios, MD examination of the lesion to determine malignancy status is of vital importance. In accordance with the findings by Zamir et al, CN are histologically associated with intraepithelial and stromal melanocytic proliferations, solid and cystic inclusions, goblet cells, and chronic inflammatory infiltrates that include eosinophils and plasma cells. These inflammatory features may define inflamed CN (ICN), and to the pathologist unfamiliar with nevi of the eye, they may suggest malignancy because of their atypical histology. The lesions manifest most commonly in young adults; thus, the condition is commonly called juvenile CN. However, CN can arise at any age. 3 To date, diagnostic features of ICN have not been systematically evaluated in a case series. The objectives of this study were to review the current histopathological criteria and to characterize the diagnostic features of ICN in a case series of 13 patients. CLINICAL PRESENTATION General pathologists do not commonly encounter CN in day-to-day practice. However, among conjunctival melanocytic lesions, CN are the most common. In a study 4 of 1643 patients with conjunctival lesions, 28% of the lesions were CN, making them the most frequent. However, the second most common lesion was conjunctival melanoma (13%). 4 These statistics illustrate the importance of differentiating these 2 prognostically different lesions. Inflamed CN may manifest as a rapidly growing lesion, which is concerning for both patients and treating physicians. Because of malignant potential, the lesion is surgically removed and submitted for histopathological diagnosis. We report 13 cases with Table 1. Clinical Features of Patients With Inflamed Conjunctival Nevi Case No. Age, y/sex Eye 1 8/F Right 2 18/M Left 3 5/M Left 4 8/M Left 5 14/M Right 6 40/M Right 7 12/M Left 8 11/M Not provided 9 15/M Left 10 9/F Left 11 16/M Left 12 2/M Left 13 12/M Left 1242 Arch Pathol Lab Med Vol 138, September 2014 Inflamed Conjunctival Nevi Choi & Chévez-Barrios
Figure 1. Hematoxylin-eosin stain, original magnification 340. A through C, Infiltrate of eosinophils. D through F, Lymphocytic inflammatory infiltrate. 1þ present (A and D), 2þ present (B and E), and 3þ present (C and F). histological diagnosis of ICN from the Houston Methodist Hospital, Houston, Texas, during the period January 1, 2007, to June 30, 2013. This study was approved by the Institutional Review Board of the Houston Methodist Research Institute (protocol No. IRB (2)0709-0099). Patient clinical records were reviewed for age, sex, and laterality of eye in which the nevi were located. The mean age at the time of excision was 13 years (age range, 2 40 years). The male predominance (85%) had also been found in the previously reported case series by Zamir et al 2 (53%) that studied juvenile ICN. A predominance of left eye locus (75%) was also seen (Table 1). HISTOPATHOLOGY AND ANCILLARY STUDIES Histologically, CN are intraepithelial and stromal melanocytic proliferations, with solid and cystic inclusions, goblet cells, and chronic inflammatory infiltrates that include eosinophils and plasma cells. We developed a Arch Pathol Lab Med Vol 138, September 2014 Inflamed Conjunctival Nevi Choi & Chévez-Barrios 1243
Case No. Intraepithelial Component Table 2. Stromal Component Histological and Immunopathological Parameters of Inflamed Conjunctival Nevi a Solid Inclusions Cystic Eosinophilic Lymphoid Lymphoid Inclusions Infiltrate Infiltrate Follicle CD3 CD20 PAS IgG4 IgG IgG4 to IgG Ratio, % 1 3 1 1 0 1 1 Absent 1 3 0 3 16 18.75 2 2 2 2 1 3 3 Present 1 3 3 24 24 100 3 3 2 2 1 2 3 Present 2 2 1 1 3 33.33 4 2 1 1 3 1 3 Present 2 3 3 1 50 2 5 2 3 2 1 1 3 Present 2 2 2 54 97 55.67 6 3 1 1 3 1 2 Present 1 1.5 2 219 219 100 7 2 2 1 2 2 2 Absent 3 1 2 31 89 34.83 8 1 3 3 1 2 1 Absent ND ND ND ND ND ND 9 2 3 1 1 1 2 Present 2 1 1 ND ND ND 10 3 1 1 2 1 1 Absent ND ND ND ND ND ND 11 2 2 1 1 0 3 Present ND ND ND ND ND ND 12 3 2 1 3 2 1 Present 3 2 3 ND ND ND 13 2 2 2 2 2 3 Present ND ND ND ND ND ND Abbreviations: IgG, immunoglobulin; IgG4, immunoglobulin G4; ND, not done because blocks were unavailable; PAS, periodic acid Schiff. a The single numerals represent a scoring method based on histology and immunopathological parameters (range, 0 to 3þ). scoring method based on histology and immunopathological parameters: using a routine hematoxylin-eosin stain, we evaluated the relative amount of intraepithelial versus stromal nevomelanocytic component (range, 0 to 3þ), where 0 indicates the absence of melanocytes and 3þ indicates nevomelanocytes occupying most of the structure. We also evaluated the amount of epithelial inclusions (either solid or cystic) using a similar grading system, where 0 indicates the absence of epithelial inclusions and 3þ indicates a predominant epithelial inclusion component. For the inflammatory components, we applied the scoring system to the eosinophilic (Figure 1, A through C) and lymphocytic (Figure 1, D through F) infiltrates, as well as the presence or absence of lymphoid follicle formation. Using periodic acid Schiff (PAS) stain, we also scored the amount of positive goblet cells in the intraepithelial component (range, 0to3þ), where 0 indicates the absence of goblet cells and 3þ indicates a predominant goblet cell component in the epithelial inclusions. In addition, to confirm the components evaluated by routine stain, we studied the intraepithelial and stromal melanocytic components by immunohistochemistry using an S100 stain, solid and cystic inclusions using a cytokeratin stain, and lymphoid infiltrates using stains to evaluate CD3 and CD20 positivity of lymphocytes. The CD3 and CD20 stains were also evaluated using the 0 to 3þ scoring system similar to that described for the inflammatory infiltrate by hematoxylin-eosin stain in Figure 1. We also evaluated the immunoglobulin 4 (IgG4) to IgG ratio of plasma cells per 10 high-power fields (original magnification 3400) (Table 2). The results from our case series demonstrated that ICN have more prominent S100-positive intraepithelial (mean, 2.3þ) melanocytic components than stromal (mean, 1.9þ) components (Figure 2). All of the nevi had epithelial inclusions, solid (mean, 1.5þ) and cystic (mean, 1.6þ), in the stroma associated with the melanocytic nests (Figure 3). In all cystic inclusions, PAS stain positive goblet cells (mean, 1.9þ) were present (Figure 4). All cases had dense inflammation consisting of plasma cells, lymphocytes (Figure 5, A), and eosinophils (Figure 5, B) within the stroma and focally infiltrating the epithelium. Eosinophils comprised a mild to moderate (mean, 1.5þ) amount of the infiltrate in the epithelium and stroma. Of the plasma cells, an average of 49% (range, 2% 100%) of IgG-positive cells showed IgG4 positivity. Nine cases (69%) had lymphoid follicle formation. The CD3-positive (mean, 1.9þ) and CD20-positive (mean, 2.1þ) tissues demonstrated a slight predominance of B cells (Table 2). The 2 patients with a 100% IgG4 to IgG ratio were the oldest in the group (18 years and 40 years). DIFFERENTIAL DIAGNOSIS ICN Versus Melanoma Architecturally, ICN can seem concerning because they frequently show reverse maturation, which is rarely seen in benign skin nevi. Unlike melanoma, ICN have a lymphoplasmacytic infiltrate mixed with eosinophils. This was seen in all 13 cases in our study. Conjunctival epithelial inclusions are commonly seen in CN. In the present study, 92% of cases showed epithelial cysts, and all 13 cases demonstrated solid elements. These benign lesions aid in the diagnosis of ICN because melanoma destroys the inclusions. 3 These cystic inclusions retained PAS stain positive goblet cells, which are also a feature of benign lesions. Cytologically, melanocytes in all 13 cases of ICN showed no cytological atypia. Cytological features that support a diagnosis of melanoma are marked nuclear pleomorphism, hyperchromasia, the presence of nucleoli and spindle cell morphology, and mitotic activity in the stromal component. 1 ICN Versus Lymphoma In the present study, all 13 cases had dense inflammation. This dense inflammatory infiltrate may lead pathologists to misdiagnose ICN as lymphoma, especially in older patients. 3 Both CD3 and CD20 stains were performed to illustrate the reactive nature of these lesions. Both B cells and T cells were present (ratio, 1.1), and no evidence of cytological atypia was seen in the lymphocytes. Many of these cells were forming lymphoid follicles. ICN Versus IgG4-Related Disease In October 2011, a group of 35 IgG4-related disease experts met in Boston, Massachusetts, to provide guidelines for the diagnosis of IgG4-related disease. This diagnosis is now in the differential diagnosis of many lesions with plasma cells; thus, we included the consensus findings and applied them to the staining patterns of ICN in our 13 cases. In true ICN cases, we consider that the reactivity seen in the IgG4 and IgG stains is in response to the main lesion, ICN, rather than an indicator of IgG4-related disease. The IgG4- related diseases are characterized by histopathological 1244 Arch Pathol Lab Med Vol 138, September 2014 Inflamed Conjunctival Nevi Choi & Chévez-Barrios
appearances and elevated IgG4-positive to IgG-positive plasma cell ratios. 5 The 3 major histopathological features are dense lymphoplasmacytic infiltrate, fibrosis arranged in a storiform pattern (at least focally), and obliterative phlebitis. 5 The minor features are phlebitis (without obliteration of the lumen) and increased numbers of eosinophils. 5 The presence of 3 of the 3 major histopathological features is required for diagnosis of IgG4-related disease, and in isolation minor features have no contribution to the diagnosis. 5 Researchers have suggested greater than 40% as a cutoff value for the IgG4-positive to IgG-positive plasma cell ratio. 5 Lesions such as orbital necrobiotic xanthogranuloma are associated with systemic IgG4-related disease, 6 and all of our 13 cases had dense lymphoplasmacytic infiltrate and increased numbers of eosinophils. This might cause pathologists to diagnose ICN as IgG4-related disease. However, the cases herein lacked fibrosis and obliterative phlebitis. Despite the fact that our cases did not fulfill the requirement for diagnosis of IgG4-related disease, we performed IgG and IgG4 immunohistochemical stains. An average of 49% (range, 2% 100%) of the IgG-positive plasma cells showed IgG4 positivity. The IgG4-related disease has been described in almost every organ system, including periorbital tissue, such as in the lacrimal gland. 5,7 9 However, to our knowledge IgG4-related disease has not been described in conjunctival lesions, and our 13 cases did not meet the requirements for the diagnosis of IgG4-related disease. The results highlight the presence of IgG4-positive plasma cells in other entities not related to IgG disease. In our series, the older patients (18 years and 40 years) had a 100% IgG4 to IgG ratio. The significance of this in our small series is unknown but may be related to the age of the patients and the response of the conjunctiva to different processes. CURRENT TREATMENT AND PROGNOSIS The treatment for ICN is complete excision, which is also the treatment for CN. Because of its benign behavior, no other therapeutic modality is recommended. However, ICN may recur after a long follow-up time. Twelve percent of studied nevi were re-excised specimens in a reported case series by Thiagalingam et al. 1 These recurrent lesions had no characteristic features that distinguished them from other nonrecurrent nevi. 1 In our series, we included no recurrent nevi. Figure 2. Intraepithelial melanocytic components. Cytokeratin stains the epithelial component red, and S100 stains the nevomelanocytes brown. Lymphocytes comprised the unstained portion in the epithelium. Double-staining immunohistochemistry with cytokeratin (Texas red) and S100 protein (diaminobenzidine); hematoxylin-eosin stain, original magnification 340. Figure 3. Cysts (arrow) and epithelium associated with nevus cells and lymphocytes (hematoxylin-eosin stain, original magnification 310). Figure 4. Periodic acid Schiff stain (original magnification 320) of goblet cells (arrow) in the epithelial component of the nevus. CONCLUSIONS For the diagnosis of ICN, our results and a review of the literature show that there is a slightly more intraepithelial component than intrastromal nevomelanocytes. We also found that all of the cases had intraepithelial inclusions, which are more frequently of a cystic type. If a cystic component is present, PAS stain positive goblet cells are always present in a moderate amount (mean, 1.9þ). Most of the cases had an infiltrate of eosinophils in a moderate amount (mean, 1.5þ), and all of the cases had a lymphocytic infiltrate that was also moderate (mean, 2.1þ). Nine of 13 cases demonstrated lymphoid follicle formation. The amount of B lymphocytes (mean, 2.1þ) is approximately the same as the amount of T lymphocytes (mean, 1.9þ). Although all of the cases that stained for IgG4 to IgG ratios showed some degree of positivity, the full criteria for IgG4- related disease were not fulfilled. This finding leads us to believe that ICN may represent one of the entities that Arch Pathol Lab Med Vol 138, September 2014 Inflamed Conjunctival Nevi Choi & Chévez-Barrios 1245
Figure 5. Inflammatory component. A, Notice the lymphoid infiltrate in the superficial stroma (hematoxylin-eosin stain, original magnification 310). B, Higher magnification of areas with eosinophils that typically infiltrate the epithelium and nevus cells (hematoxylin-eosin stain, original magnification 340). express IgG4-positive plasma cells but is not part of the IgG4-related disease spectrum. Pathologists should be aware of this potential diagnostic pitfall. In summary, the criteria for diagnosis of ICN should include a lack of melanocytic atypia, the presence of epithelial inclusions in association with nevomelanocytes, an infiltrate of eosinophils and lymphocytes with plasma cells, and goblet cells in the cystic epithelial inclusions. References 1. Thiagalingam S, Johnson MM, Colby KA, Zembowicz A. Juvenile conjunctival nevus: clinicopathologic analysis of 33 cases. Am J Surg Pathol. 2008;32(3):399 406. 2. Zamir E, Mechoulam H, Micera A, Levi-Schaffer F, Pe er J. Inflamed juvenile conjunctival naevus: clinicopathological characterisation. Br J Ophthalmol. 2002;86(1):28 30. 3. Maly A, Epstein D, Meir K, Pe er J. Histological criteria for grading of atypia in melanocytic conjunctival lesions. Pathology. 2008;40(7):676 681. 4. Shields CL, Demirci H, Karatza E, Shields JA. Clinical survey of 1643 melanocytic and nonmelanocytic conjunctival tumors. Ophthalmology. 2004; 111(9):1747 1754. 5. Deshpande V, Zen Y, Chan JK, et al. Consensus statement on the pathology of IgG4-related disease. Mod Pathol. 2012;25(9):1181 1192. 6. Singh K, Rajan KD, Eberhart C. Orbital necrobiotic xanthogranuloma associated with systemic IgG4 disease. Ocul Immunol Inflamm. 2010;18(5):373 378. 7. Stone JH, Zen Y, Deshpande V. IgG4-related disease. N Engl J Med. 2012; 366(6):539 551. 8. Dahlgren M, Khosroshahi A, Nielsen GP, Deshpande V, Stone JH. Riedel s thyroiditis and multifocal fibrosclerosis are part of the IgG4-related systemic disease spectrum. Arthritis Care Res (Hoboken). 2010;62(9):1312 1318. 9. Masaki Y, Dong L, Kurose N, et al. Proposal for a new clinical entity, IgG4- positive multiorgan lymphoproliferative syndrome: analysis of 64 cases of IgG4- related disorders. Ann Rheum Dis. 2009;68(8):1310 1315. 1246 Arch Pathol Lab Med Vol 138, September 2014 Inflamed Conjunctival Nevi Choi & Chévez-Barrios