Saudi Journal of Ophthalmology (2019) 33, 214–218 Original Article Search for histopathological characteristics of inflammatory juvenile conjunctival nevus in conjunctival nevi related to age: Analysis of 33 cases L. Nebot ⇑; A. Sáez; P. Serret; C. Caral; R.B. García-Chamón; M.R. Bella Abstract Purpose: Conjunctival nevi in young individuals can correspond to the entity named Inflammatory Juvenile Conjunctival Nevus (IJCN), presenting clinically as a rapid growth lesion, and showing at the histopathological study an inflammatory infiltrate sur- rounding the lesion. All these findings can suggest a diagnosis of malignancy. Due to a case of IJCN diagnosed in our Pathology department, we realized that this entity is rarely reported in the literature and histopathological diagnostic criteria are not well defined. The aim of our study is to compare the histopathological characteristics of conjunctival nevi in patients aged thirty years or less to those in patients above 30 years, looking for the findings described in IJCN. Methods: All the excisional specimens of resected conjunctival nevus in a tertiary hospital from 2000 to 2018 were retrieved from the Pathology department archives. Demographic data were recorded, and histopathological variables (histological type of nevus, lymphocytic infiltration, eosinophilic infiltration, presence of lymphoid follicles, stromal nevomelanocytic component, intraepithe- lial nevomelanocytic component, epithelial inclusions, quantity of goblet cells in epithelial inclusions, cellular atypia, mitoses and maturation of the lesion) were evaluated by three independent observers. Statistical analysis was performed comparing the two age groups. Results: The study determined a significant predominance of the lymphocytic and eosinophilic infiltration in the group of patients aged thirty years or less respect to the elderly group. The percentage of stromal component of the lesion is larger in patients over thirty years compared to the younger group. There was no correlation between epithelial inclusions, maturation or cytological aty- pia and age groups. Conclusion: We found some histopathological differences in conjunctival nevi related to young age, some of them coincident with the ones described in IJCN, which histopathologically could lead to a misleading diagnosis. However, we did not find significant differences related to age in many of the described histopathological findings described in IJCN. Larger series with a greater num- ber of cases would be of interest to characterize more precisely this lesion. Keywords: Conjunctival nevus, Inflammatory juvenile conjunctival nevus, Melanocytic conjunctival lesion, Lymphocytic infiltration Ó 2019 The Authors. Production and hosting by Elsevier B.V. on behalf of Saudi Ophthalmological Society, King Saud University. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). https://doi.org/10.1016/j.sjopt.2019.07.011 Received 31 March 2019; received in revised form 16 July 2019; accepted 25 July 2019; available online 9 September 2019. Department of Pathology, Parc Taulí Hospital Universitari, Institut d’Investigació i Innovació Parc Taulí I3PT, Universitat Autònoma de Barcelona, Sabadell, Spain ⇑ Corresponding author. e-mail address: lnebot@tauli.cat (L. Nebot). Peer review under responsibility Access this article online: of Saudi Ophthalmological Society, www.saudiophthaljournal.com King Saud University Production and hosting by Elsevier www.sciencedirect.com Search for histopathological characteristics of inflammatory juvenile conjunctival nevus 215 Introduction pare the histopathological characteristics of young patients (aged thirty or less) to patients aged over thirty. The most frequent tumours of the conjunctiva (50%) are melanocytic proliferations, 28% of these melanocytic lesions Materials and Methods are conjunctival nevus [1]. This neoplasm appears most fre- quently in the bulbar conjunctiva (72%) [2]. Progression to This is a retrospective observational study that analyses melanoma has only been described in 1% of cases [3]. the histopathological characteristics of conjunctival nevus. Inflammatory juvenile conjunctival nevus (IJCN) is a benign Two age groups of patients were compared: children/young lesion located most often on the juxtalimbal conjunctiva, that adults up to 30 years old and adults over thirty years old. All occurs in children and adolescents (average age of surgery patients with diagnosis of conjunctival nevus in the database 11–12 years). In children is usually a lightly pigmented or of Corporació Sanitaria Parc Taulí, Hospital, Sabadell (Spain), amelanotic lesion that become pigmented at puberty or who were excised between years 2000 and 2018 were pregnancy [4]. Clinically, it may grow rapidly or increase its included in the study. The specimens fixed with formalde- pigmentation, becoming suspicious of malignancy. In these hyde and paraffin embedded were sectioned at 3 mm thick. cases, excision of the lesion is recommended. It is suggested Slides stained with Hematoxylin-Eosin were reviewed by that the growth of the lesions may be due to inflammatory three independent observers (two pathologists and a pathol- infiltration and cystic degeneration. IJCN are associated with ogist trainee). systemic allergy, allergic conjunctivitis and vernal conjunctivi- The histopathological features evaluated were: histologi- tis [5,6]. Levi-Schaffer et al. studied this issue and suggested cal type (compound, junctional or subepithelial), lymphocytic an association between IJCN and allergic inflammation. They infiltration, eosinophilic infiltration, presence of lymphoid fol- described an increased presence of nerve growth factor licles, subepithelial nevomelanocytic component, intraepithe- (NGF), eosinophils and mast cells in IJCN and demonstrated lial nevomelanocytic component, epithelial inclusions (solid higher production of NGF by fibroblasts related to the lesion or cystic), amount of goblet cells in epithelial inclusions, cel- respect normal fibroblasts, which modulate eosinophil prop- lular atypia, mitoses and maturation of the lesion. The fea- erties through NGF [6,7]. tures were evaluated semiquantitatively (0 to 3) or Histologically, most IJCN are described as compound nevi dicotomically (Yes/No) (Table 1). (97%) with intraepithelial and subepithelial melanocytic nests The definite value for each item was the one determined and solid or cystic epithelial inclusions, the latter being more by most of the observers. Statistical analysis was carried out frequently cystic with PAS stain–positive goblet cells. They using Statistical Package for the Social Sciences (SPSS) soft- show a remarkable stromal inflammatory infiltrate with lym- ware. Descriptive statistics were presented as percentages phocytes, plasma cells and eosinophils [4]. Some authors and frequencies. Chi-square test (w2). was applied to com- describe a ‘‘reverse’’ subepithelial maturation in this entity, pare the categorical variables of each histopathological with melanocytes displaying nuclear and cytoplasmic size greater in depth than in the junctional component, a pattern Table 1. Histopathological features. of confluent growth in the junctional component, and certain degree of atypia [8]. Sometimes the prominent inflammatory Histological type of nevus Compound component produces a distortion of the architecture and an Junctional Subepithelial apparent cytological atypia. It is described as a benign entity with some ‘‘atypical’’ features that can be associated with Lymphocytic infiltration Grade 0: Absent Grade 1: Isolated melanoma in cutaneous melanocytic lesions [4]. inflammatory cells Classically, it corresponds to an enlarging lesion in the bul- Grade 2: Groups of few bar conjunctiva. Young age and cystic component are indica- cells tors of a benign lesion. The differential diagnosis of IJCN Grade 3: Prominent includes malignant conjunctival melanoma, lymphoma and groups of cells primary acquired melanosis. The most important clinical Eosinophilic infiltration Grade 0: Absent and histological differential diagnosis is conjunctival malig- Grade 1: Isolated inflammatory cells nant melanoma, whose clinical presentation can be similar Grade 2: Groups of few to IJCN. However, cysts are rarely seen in melanomas, and cells unlike malignant melanoma, in IJCN there is no marked cyto- Grade 3: Prominent logical atypia or mitotic activity in the stromal component [3]. groups of cells The lymphocytic inflammatory infiltrate may suggest the Lymphoid follicles Yes/No diagnosis of a conjunctival lymphoma, but it is rare in young Stromal/intraepithelial 0%/100% population, and the lymphocytic population in IJCN corre- nevomelanocytic component 25%/75% sponds to an admixture of B and T cells. [4] 50%/50% 75%/25% We received an excisional biopsy of a conjunctival melano- 100%/0% cytic lesion in a 28 year old man, with a diagnosis compatible Solid epithelial inclusions Grade 0–3 with IJCN. It is an uncommon lesion and there are few case Cystic epithelial inclusions Grade 0–3 series reported in the literature. The aim of this study is to Maturation of nevus cells deeply Yes/No describe the features of conjunctival nevi received in our Goblet cells in epithelial inclusions Grade 0–3 Pathology department (Corporació Sanitaria Parc Taulí Cellular atypia Yes/No Hospital, Sabadell, Spain) in the last eighteen years, and com- Atypical or deep mitoses Yes/No 216 L. Nebot et al. feature between the two age groups (30 or >30 years old). nificant differences in the distribution of the type of nevus The significance level was fixed as 5% for each analysis. A p- according to the age group (p = 0.191). value less than 0.05 was considered statistically significant. Lymphocytic infiltration in the group of younger patients was considered grade 3 in 9 cases (52.9%), grade 2 in 3 cases Results (17.6%), grade 1 in 4 cases (23.5%) and grade 0 in 1 case (5.9%). In the group of older patients, the lymphocytic infiltra- Thirty-three cases of conjunctival nevus were identified, 32 tion was grade 2 in 3 cases (20%), grade 1 in 9 cases (60%) of them being evaluable histologically, while a remaining and grade 0 in 3 cases (20%), with any case of grade 3, with case was considered insufficient for its valuation. The series statistically significant differences in lymphocytic infiltration included 17 cases of patients aged thirty years or less (range: between the two age groups (p = 0.008). 5–30 years old), and 15 cases of patients over thirty years old Eosinophilic infiltration was found only in the group of (range: 31–71 years old). The histopathological features of younger patients, in 8 of 17 cases (47.1%), with statistically each age group are summarized in Table 2. significant difference (p = 0.024); it was considered grade 3 There are 17 cases in the group of patients aged thirty in 2 cases, grade 2 in 1 case, and grade 1 in 5 cases. In our years or less with histopathological characteristics of conjunc- series, only two cases showed lymphoid follicles and both tival nevus. Of these, 15 cases corresponded to compound were patients under 30 years of age. nevi, 1 to a junctional nevus and 1 to subepithelial nevus. Five cases showed no epithelial inclusions, solid or cystic. Of the 15 cases over thirty years old, 11 were compound nevi Four of these belong to the group until thirty years and and 4 corresponded to subepithelial nevi. There were no sig- one case belong to the oldest group. Solid and cystic epithe- Table 2. Histopathological features. Group age* Significance p-value 30 y (n = 17) >30 y (n = 15) No. (%) No. (%) Histological type of nevus Compound 15 (88.2%) 11 (73.3%) NS Junctional 1 (5.9%) 0 (0%) Subepithelial 1 (5.9%) 4 (26.7%) Lymphocytic infiltration 0 1 (5.9%) 3 (20%) p = 0.008 1 4 (23.5%) 9 (60%) 2 3 (17.6%) 3 (20%) 3 9 (52.9%) 0 (0%) Eosinophilic infiltration 0 9 (52.9%) 15 (100%) p = 0.024 1 5 (29.4%) 0 (0%) 2 1 (5.9%) 0 (0%) 3 2 (11.8%) 0 (0%) Lymphoid follicles Yes 2 (11.8%) 0 (0%) NS No 15 (88.2%) 15 (100%) Stromal/intraepithelial nevomelanocytic component 0%/100% 1 (5.9%) 0 (0%) 25%/75% 2 (11.8%) 0 (0%) 50%/50% 6 (35.3%) 1 (6.7%) NS 75%/25% 7 (41.2%) 10 (66.7) 100%/0% 1 (5.9%) 4 (26.7) Stromal nevomelanocytic component 50% 9 (52.9%) 1 (6.7%) p = 0.006 >50% 8 (47.1%) 14 (93.3%) Solid epithelial inclusions 0 7 (41.2%) 8 (53.3%) NS 1 7 (41.2%) 5 (33.3%) 2 1 (5.9%) 2 (13.3%) 3 2 (11.8%) 0 (0%) Cystic epithelial inclusions 0 5 (294%) 1 (6.7%) NS 1 4 (23.5%) 6 (40%) 2 5 (29.4%) 3 (20%) 3 3 (17.6%) 5 (33.3%) Maturation of nevus cells deeply Yes 13 (75.5%) 15 (100%) NS No 4 (23.5%) 0 (0%) Goblet cells in epithelial inclusions 0 8 (47.1%) 3 (20%) NS 1 6 (35.3%) 8 (53.3%) 2 2 (11.8%) 2 (13.3%) 3 1 (5.9%) 2 (13.3%) Cellular atypia Yes 2 (11.8%) 1 (6.7%) NS No 15 (88.2%) 14 (93.3%) Atypical or deep mitoses Yes 0 (0%) 0 (0%) NS No 17 (100%) 15 (100%) *Age in years at time of surgery. NS: No significance. Search for histopathological characteristics of inflammatory juvenile conjunctival nevus 217 lial inclusions were present in 10 and 12 cases respectively in the group until thirty years, and in 7 and 14 cases respectively in group of patients older than thirty years. In the group until thirty years, grade 3 solid and cystic epithelial inclusions were present in 2 (11.8%) and 3 cases (17.6%) respectively, while in the older group there wasn’t any case of solid epithelial inclusion grade 3, although there were 5 cases (33.3%) of cystic inclusions grade 3. In our series, older patients had a greater stromal nevome- lanocytic component with statistically significant difference (p = 0.006). The stromal component constitutes more than 50% of the lesion in 93.3% of cases in the oldest group. Cellular atypia was evaluated considering the usual atypia in nevus of young patients. Only 3 cases presented greater grade of atypia than expected, two of them in the group of patients until thirty years and the other in the group older than thirty. Deep maturation of the lesion was absent in 4 cases (23.5%) in the youngest group and in zero cases (0%) in the oldest group of patients. The statistical analysis showed no differences between age and the presence of cystic inclusions, solid inclusions, goblet cells, maturation or cytological atypia. (Figs. 1 and 2) Discussion Inflammatory juvenile conjunctival nevi are described as compound nevus with intraepithelial and stromal nevome- lanocytic component [4,8]. In our series, in the group of younger patients (30 years old), compound nevus repre- sented 88.2% of cases, junctional nevus 5.9%, and subepithe- lial nevus 5.9%, while in the group of patients over thirty years old, compound nevus represented 73.3%, junctional nevus 0% and subepithelial nevus 26.7% of cases. The group of patients under thirty years had mainly compound nevus, while in the group of patients over thirty years there was a higher percentage of nevus of the subepithelial type. How- ever, analyzing statistically these results we did not find sta- tistically significant differences in the distribution of the types of nevus between the two age groups. It has been described that conjunctival nevi appear in the first and second decades of life with the nevomelanocytic cell nests located in the epidermal-stromal junction, and during the second and third decade the nests migrate to the stroma Fig. 1. A, Dense lymphocytic infiltrate grade 3 (Original magnifica- tion  200 Hematoxylin and Eosin). B, Presence of a lymphoid follicle forming the compound nevus. In the third and fourth decade formation (Original magnification  200 Hematoxylin and Eosin). C, the lesion usually is totally located in the subepithelial stroma Eosinophilic infiltrate grade 3 (Original magnification  400 Hematoxylin [1,3]. This fact could explain why in our series the older group and Eosin). presented higher percentage of stromal component in the lesion. Comparing the two age groups, there are significant dif- (47.1%). No eosinophilic infiltrate was found in patients older ferences related to inflammatory infiltrate between them. than 30 years age. There is a statistically significant high level Nearly fifty-three percent (52.9%) of patients aged thirty of eosinophilic infiltration among the age groups. Zamir et al. years or less had severe lymphocytic infiltrate with prominent found that 75% of conjunctival nevus of a serie of 63 patients cell aggregates (grade 3), and two cases (11.8%) showed lym- younger than 20 years were compound nevus with a promi- phoid follicles, while any patient of the group of older nent inflammatory infiltrate constituted by lymphocytes, eosi- patients had severe lymphocytic infiltrate. Furthermore, nophils and plasma cells with features of inflamed patients aged over thirty years old had more frequently only conjunctival nevus. Of these cases, 30% presented germinal isolated lymphocytes (60%) than the group of patients ages centres and 77 % eosinophilic infiltrate [5]. until thirty years old (23.5%). So the cases of the youngest Cystic inclusions are considered a sign of chronicity and group showed more prominent lymphocytic infiltrates com- are caused by the migration of subepidermal melanocytes pared to the group over thirty years. Patients under 30 years [3]. Of the five cases that do not show intraepithelial inclu- of age presented eosinophilic infiltration in eight cases sions (solid or cystic), four correspond to the group of 218 L. Nebot et al. the conjunctiva. None of the three observers detected mito- tic activity, as the cases of inflammatory juvenile atypical nevus reported in the literature [1,2]. Finally, no significant correlation was obtained between the age group and the aty- pia or maturation of the lesion. To summarize, in an attempt to contribute to the charac- terization of IJCN, we described the histopathological char- acteristics of all conjunctival nevi received in our Pathology department from 2000 to 2018, looking for differences between two age groups. The younger group aged thirty or less, most frequently presented lymphocytic and eosino- philic infiltrates. We noticed also greater stromal component of nevus cells in the oldest age group. After reviewing the scant studies and reports about IJCN, we realized that there are not clearly defined diagnostic criteria for this entity. Most of the authors define this entity as a lesion with ‘‘reverse’’ maturation, confluent growth pattern in the junctional com- ponent, chronic inflammation and variable but low grade of cytologic atypia [3,8,9]. After the review of our cases and con- sidering the lack of unification of diagnostic criteria reported in the literature, it is unclear if any of our cases could finally be diagnosed as IJCN. Our study is limited due to the small sam- ple size. Larger series with a greater number of cases would contribute to a best characterization of this tricky and scar- cely described lesion. Acknowledgement The authors would like to extend thanks to Dr. Jordi Costa Fig. 2. A, Predominant component of goblet cells in epithelial inclusions Pueyo for his assistance with statistical analysis. (Original magnification  100 Hematoxylin and Eosin). B, Prominent epithelial inclusions (Original magnification  100 Hematoxylin and References Eosin). 1. Costea CF, Turliuc MD, Dimitriu G, Bogdănici CM, MoT ß oc A, Chihaia MA, et al. Inflammatory juvenile compound conjunctival nevi. A clinicopathological study and literatura review. Rom J Morphol patients of thirty or less years old. However, we did not find a Embryol. 2017;58(3):739–47. statistically significant association between age and epithelial 2. Colarossi C, Milazzo M, Paglierani M, Massi D, Memeo L, Canzonieri V. inclusions grade. A juvenile case of conjunctival atypical nevus. Diagn Pathol 2013;8:64. Cellular atypia higher than the atypia expected in a con- 3. García de Oteyza G, Betancourt J, Benedetti Sandner M, Vázquez- Romo KA, Hernández-Ayuso I, Ramos-Betancourt N. Inflammatory ventional nevus was found in 2 cases in the youngest group juvenile compound nevi: A melanoma is not all that it appears to be. and in one case of the oldest group. In addition, the two Arch Soc Esp Oftalmol 2019;94(2):90–4. lesions belonging to the group of patients until thirty years 4. Thiagalingam S, Johnson MM, Colby KA, Zembowicz A. Juvenile old presented absent deep maturation of nevomelanocytic conjunctival nevus: clinicopathologic analysis of 33 cases. Am J Surg Pathol 2008;32(3):399–406. cells. Maturation of nevomelanocytes was the evaluated trait 5. Zamir E, Mechoulam H, Micera A, Levi-Schaffer F, Pe’er J. Inflamed with the most discordance among the three different juvenile conjunctival naevus: clinico pathological characterisation. Br J observers. Ophthalmol. 2002;86(1):28–30. Colarossi et al. described a case of atypical juvenile con- 6. Alkatan H, Al-Arfaj K, Maktabi A. Conjunctival nevi: clinical and junctival compound nevus with marked atypia and inflamma- histopathologic features in a Saudi population. Ann Saudi Med 2010;30(4):306–12. tion. It corresponded to a compound nevus with atypical cells 7. Levi-Schaffer F, Micera A, Zamir E, Mechoulam H, Puxeddu I, in deep nests, with focal pagetoid spread but without mitotic Piliponsky AM, et al. Nerve growth factor and eosinophils in activity. The final diagnosis was juvenile conjunctival atypical inflamed juvenile conjunctival nevus. Invest Ophthalmol Vis Sci nevus [2]. Costea et al. also made the final diagnostic of 2002;43(6):1850–6. 8. Dadras SS, Zembowicz A. Conjunctival melanocytic nevi. In: Cassarino inflammatory juvenile atypical compound nevus in a lesion SS, Dadras SS, editors. Diagnostic pathology: neoplastic with chronic inflammation, lack of maturation in the deep of dermatopathology. 2nd ed. Philadelphia: Elsevier; 2017. p. 806–9. the lesion, pagetoid spread and atypical nevomelanocytes 9. Jakobiec FA, Sandhu H, Bhat P, Colby K. Bilateral conjunctival without mitotic activity [1]. We consider that our three cases melanocytic nevi of simultaneous onset simulating conjunctivitis in a with certain degree of atypia did not have enough cytological child. Cornea 2010;29(8):937–40. atypia to make the diagnosis of atypical compound nevus of