Gross, histologic, and immunohistochemical characteristics of cutaneous chromatophoromas in captive bearded dragons

Abstract

Chromatophoromas are neoplasms that develop from the dermal pigment-bearing and light-reflecting cells (chromatophores) in the skin of reptiles, fish, and amphibians. Seventeen cutaneous chromatophoromas were identified from 851 bearded dragon submissions (2%) to a private diagnostic laboratory in a 15-y period. No sex predilection was found. Ages ranged from 9 mo to 11 y. Chromatophoromas most commonly were single, raised, variably pigmented masses or pigmented scales on the trunk, and less commonly the extremities or head. Microscopically, iridophoromas, melanophoromas, mixed chromatophoromas, and nonpigmented chromatophoromas were identified. Neoplasms were often ulcerated and invaded deep into the subcutis and muscle. Most commonly, nuclear atypia was mild-to-moderate, and mitotic count was low. Six neoplasms had abundant, periodic acid-Schiff–positive, mucinous stroma. Histologic examination was often adequate to diagnose chromatophoromas in bearded dragons given that 11 of 17 had some degree of pigmentation, although it was often scant. IHC for S100 and PNL2 could be helpful to diagnose poorly pigmented neoplasms. No lymphatic invasion or metastases at the time of excision were noted in any of the cases. Follow-up data were available for 6 cases, with no reports of recurrence or neoplasia-related death. Two cases had elevated mitotic counts and nuclear pleomorphism, which has been associated with metastasis in other reports of bearded dragon chromatophoromas. In general, it appears that aggressive surgical excision is often curative, but monitoring may be warranted for cases in which tumors had high mitotic count and nuclear pleomorphism.

Chromatophoromas arise from dermal-pigmented and light-reflecting cells (chromatophores) in reptiles, fish, and amphibians.¹ Chromatophores are responsible for the wide variety of color patterns observed in these species, and are abundant in species capable of rapid color change.¹ In reptiles, chromatophoromas are subdivided into 3 subtypes based on the pigmented cell origin.¹ Iridophoromas arise from the crystalline purine-producing cells (iridophores), melanophoromas or melanomas arise from the melanin-producing cells (melanophores), and xanthophoromas arise from carotenoid- and pteridine-producing cells (xanthophores).¹ Iridophoromas are differentiated from the others histologically given that they are birefringent when viewed with polarized light microscopy.^1,22

Bearded dragons (Pogona spp.) are native to Australia.²⁸ Central bearded dragons (Pogona vitticeps), more commonly referred to as bearded dragons, are popular pets and zoologic exhibit animals throughout the world.^21,26 Neoplasia appears to be relatively common in bearded dragons.^{4,5,7,8,10-14,16,18,31,33,37} There is a single case report of a chromatophoroma in an individual dwarf bearded dragon,⁵ and 2 case series of reptile neoplasia that include bearded dragons.^10,14 To our knowledge, there are no large retrospective studies regarding the histologic and immunohistochemical characteristics of cutaneous chromatophoromas in bearded dragons. We describe here the gross, histologic, and immunohistochemical findings in 17 cases of chromatophoromas in captive bearded dragons.

Materials and methods

Animals

Bearded dragons submitted to Northwest ZooPath (Monroe, WA, USA) from 2000 to 2015 were considered for our study. Criteria for inclusion were masses or plaques with gross and/or histologic features consistent with chromatophoromas. The samples included in our study were obtained via surgical biopsy or collected at postmortem examination. Location, sex, age, type of collection, size of masses, number of masses, presence of metastases, recurrence of masses following excision, and any relevant history or follow-up information were recorded.

Histologic findings

Tissue samples were fixed in 10% neutral-buffered formalin prior to being processed routinely for histologic examination. When required, demineralization of specimens with 12% HCl solution was performed for ~1 d (case 11). Paraffin-embedded tissues were sectioned at 5 µm, mounted on frosted glass slides, and stained with hematoxylin and eosin (H&E). Periodic acid–Schiff (PAS) stain (cases 5, 6, 10, 11, 13, 15, 17) was also applied for mucinous tumors. Some heavily pigmented tumors were bleached with 10% H₂O₂ (cases 13, 15). Tumor size was assessed by measuring the greatest dimension of the mass on the frosted glass slide. Chromatophoromas were reexamined histologically and categorized based on the predominant type of intracellular pigment by 2 of the authors (C.F. Monahan, M. Kiupel). Histologic subtypes included melanophoromas (i.e., cells containing dark-brown to black pigment), iridophoromas (i.e., cells containing distinct birefringent crystals using polarized light), mixed chromatophoromas (i.e., containing 2 different chromatophore types), and nonpigmented chromatophoromas (i.e., tumors with cellular morphology consistent with chromatophoromas, but with no pigmentation or birefringence or only single cells containing a few pigment granules). We identified no xanthophoromas (characterized by carotenoid- or pteridine-producing cells).

Chromatophoromas were microscopically evaluated according to the histologic factors proposed for canine cutaneous melanocytic neoplasms, including lymphatic invasion, mitotic count, degree of pigmentation, nuclear atypia, level of infiltration, and ulceration.³⁵ A high-magnification (40×) objective with a field number of 22 mm was used to count mitotic figures in 10 representative fields (i.e., densely populated areas of the tumor that excluded areas of necrosis, hemorrhage, histologic artifacts, ulceration, and edema). Thus, the area of the field in which mitotic figures were counted was 2.37 mm².¹⁹ The degree of pigmentation was categorized as low (<30% of neoplastic cells), moderate (30–60% of neoplastic cells), or high (>60% of neoplastic cells). Well-differentiated neoplastic cells contained a single small nucleus with a single, centrally located nucleolus and evenly dispersed chromatin material.³⁶ More undifferentiated or atypical cells contained larger, irregularly shaped nuclei with variably clumped chromatin; nucleoli were larger, peripheralized, and often multiple.³⁶ Nuclear atypia was classified as mild (<30% of neoplastic cells), moderate (30–60% of neoplastic cells), or marked (>60% of neoplastic cells).

Immunohistochemistry

An immunohistochemical panel was performed on 5-mm serial sections for each case. The panel included antibodies to detect protein S100, melanocytic antigen PNL2, and melan A. IHC was performed on a Bond autostainer automated system (Leica), using a biotin-free, polymeric alkaline phosphatase–linked antibody detection system with fast red chromogen and hematoxylin counterstain (Bond Polymer AP red detection kit; Leica) as described previously.³⁴ For negative controls, antibodies were replaced with homologous immune sera.

Results

A search of Northwest ZooPath’s archives for bearded dragons revealed 851 surgical or postmortem submissions between 2000 and 2015. Of these submissions, 17 (2%) potential chromatophoromas were identified. One case included original and follow-up surgical submissions, the latter with wider margins. Four bearded dragons belonged to zoologic collections and 13 were privately owned. One case was diagnosed via postmortem examination; all other cases were excisional biopsies of cutaneous masses. Eight cases were female, 6 were male, and 3 were of unknown sex. The age was known for 14 cases and included 1 juvenile (<1 y old) and 13 adults (≥2 y old), with an age range of 9 mo to 11 y (average 4.8 y).

Of these 17 cases, 12 had a single mass and 5 cases had multiple masses at the time of sample collection, for a total of 22 masses evaluated histologically. All masses, except 1 of the masses in case 5, were re-examined histologically and with IHC. Cases with multiple masses included 2 cases of melanophoromas (2 of 5 total), 2 cases of mixed chromatophoromas (1 of 3 total), and 1 case of nonpigmented chromatophoroma (1 of 6 total).

When gross descriptions were provided, the majority were described as masses or nodules (14 cases; 15 masses) ranging from white to pigmented black to yellow. In 2 cases (3 masses), lesions were described as black scales or dark spots of thickened skin (Fig. 1). The masses were 3–21 mm in greatest dimension after fixation. The anatomic location was known in 15 cases (17 masses). The most common location was the trunk (8 masses, 47%), followed by extremities (5 masses, 29%) and head (4 masses, 24%).

Figures 1–7.

Cutaneous chromatophoromas in captive bearded dragons. Figure 1. Multiple black, plaque-like lesions on the trunk (mixed chromatophoroma). Figure 2. Iridophoroma. H&E with polarized light. Figure 3. Melanophoroma. H&E. Figure 4. Mixed chromatophoroma (iridophoroma and melanophoroma components). H&E. Figure 5. Nonpigmented chromatophoroma. H&E. Figure 6. Nonpigmented chromatophoroma with mucinous stroma and osseous and cartilaginous metaplasia. H&E. Figure 7. Melanophoroma with mucinous stroma (asterisk). H&E.

All chromatophoromas spanned the dermis and frequently extended up to the epidermal interface, but without junctional activity. Masses in 4 cases (6 masses) also extended into the subcutis, and into the muscle layer (6 cases, 7 masses). Chromatophoromas were fairly well-demarcated to poorly demarcated, nonencapsulated, and moderately to highly cellular. Neoplastic cells were arranged in streams, whorls, and interlacing bundles, and were supported by fine fibrous stroma. Neoplastic cells were typically spindloid (12 cases; 14 masses) or had mixed polygonal and spindloid (5 cases; 7 masses) morphology. Neoplastic cells had variably distinct cell borders and moderate amounts of eosinophilic cytoplasm. Intracytoplasmic pigment varied from brown to amber and birefringent with polarized light (iridophoromas) to dark-brown or black (melanophoromas). Variants included a combination of the 2 pigment types (mixed chromatophoromas) and absence of pigment (nonpigmented chromatophoromas). The chromatophoromas were further characterized as iridophoromas (3 cases), melanophoromas (5 cases; 6 masses), mixed chromatophoromas (3 cases; 5 masses), or nonpigmented chromatophoromas (6 cases; 7 masses; Figs. 2–5). The degree of pigmentation was none-to-low for 13 cases (14 masses), moderate for 2 cases (3 masses), and high for 2 cases (4 masses).

Nuclei were round-to-ovoid, finely to coarsely stippled, and contained 1–6 small nucleoli. Nuclear atypia was mild in 7 cases (8 masses), moderate in 7 (9 masses), and marked in 3 (4 masses) with frequent multinucleate giant cells and karyomegalic nuclei (1 melanophoroma, 2 nonpigmented chromatophoromas). The mitotic count was 0–26 mitotic figures in ten 400× fields and was generally low, with a mean of 4.

The overlying epithelium was ulcerated in 6 cases (8 masses) and often associated with heterophilic inflammation. In 4 cases (5 masses), the area of ulceration was covered by a serocellular crust and/or had a secondary bacterial infection. Eleven cases (12 masses) were mild to markedly inflamed, with infiltrates of heterophils and lymphocytes scattered throughout the neoplasm.

A subset of the tumors (6 cases: 4 melanophoromas, 1 mixed chromatophoromas, 1 nonpigmented chromatophoroma; 8 masses) had abundant mucinous and PAS-positive stroma (Fig. 6). These tumors were poorly demarcated, nonencapsulated, and had low cellularity. Neoplastic cells were polygonal-to-spindloid. Nuclei were elongate, finely stippled to euchromatic, and had 0–1 nucleoli. Two of the 6 cases (4 masses) with mucinous stroma were heavily pigmented (melanophoromas). One of the mucinous tumors (nonpigmented chromatophoromas) had multiple areas of osseous and cartilaginous metaplasia (Fig. 7, Table 1).

Table 1.

Histologic features of chromatophoroma subtypes from captive bearded dragons.

	Iridophoroma	Melanophoroma	Mixed chromatophoroma	Nonpigmented chromatophoroma
No. of cases	3	5	3	6
Level of infiltration	1 dermal, 1 subcutis, 1 muscle	1 dermal, 2 subcutis, 2 muscle	0 dermal, 1 subcutis, 2 muscle	4 dermal, 1 subcutis, 1 muscle
Morphology	2 mixed, 1 spindloid	2 mixed, 3 spindloid	1 mixed, 2 spindloid	0 mixed, 6 spindloid
Pigmentation
No	0	0	0	6
Low	2	4	1	0
Moderate	1	0	1	0
High	0	1	1	0
Nuclear atypia
Mild	2	4	0	1
Moderate	1	0	3	3
Marked	0	1	0	2
Mitotic count	2 low, 1 moderate	4 low, 1 high	3 low	3 low, 2 moderate, 1 high
Mucinous stroma	0	4	1	1

There was no evidence of lymphatic invasion by any of the neoplasms. No junctional activity, lentiginous spread, or intranuclear pseudoinclusions were observed in any of the neoplasms.

Overall, 16 of 17 cases had scattered-to-diffuse cytoplasmic-to-rare nuclear immunoreactivity for S100 (Fig. 8). The one case (iridophoroma) that was not immunoreactive for S100 had cytoplasmic immunoreactivity for PNL2. Three cases had scattered-to-diffuse cytoplasmic labeling for PNL2 (Fig. 9). The 3 cases with PNL2 labeling were 1 iridophoroma, 1 melanophoroma, and 1 nonpigmented chromatophoroma. None of the neoplasms was immunoreactive for melan A (Table 2).

Figure 8, 9.

Immunohistochemistry of cutaneous chromatophoromas in captive bearded dragons. Figure 8. Neoplastic cells have strong brown cytoplasmic and scattered nuclear immunolabeling for S100. 3,3′-diaminobenzidine (DAB). Figure 9. Neoplastic cells have scattered strong red cytoplasmic immunolabeling for PNL2. Vector red.

Table 2.

Immunohistochemistry of chromatophoroma subtypes from captive bearded dragons.

	Iridophoroma	Melanophoroma	Mixed chromatophoroma	Nonpigmented chromatophoroma
No. of cases	3	5	3	6
Melan A	0	0	0	0
PNL2	1	1	0	1
S100	2	5	3	6

Follow-up information was available for 6 cases. For the cases with follow-up, tumors were not reported to recur after initial excision, and the cause of death was not related to the tumor. One case had full postmortem results, and no metastatic lesions were found in any of the tissues submitted.

Discussion

Chromatophoromas accounted for 2% of bearded dragon submissions to Northwest ZooPath. This low prevalence is in agreement with most of the reports in bearded dragons,^5,14 although one study reported a higher prevalence, perhaps because the relative prevalence was compared to other reptile species with chromatophoromas in their study, rather than to the total number of bearded dragon submissions to their laboratory.¹⁰ Also, all of those animals were privately owned,¹⁰ hence differences in husbandry, genetics, or environment could have played a role.

In domestic animals, dogs, horses, and certain breeds of pigs most commonly develop cutaneous melanocytic neoplasms.^9,27,35 Similar to dogs and horses, neoplasms in our study typically developed in adults (≥2 y old), and no sex predilection was observed.^27,35 Chromatophoromas in our study typically were seen as single, mass-like, or nodular lesions that were pigmented to some degree but of various colors. It should be noted that some neoplasms were black scales or dark spots of thickened skin, hence more subtle gross lesions should be biopsied for histologic examination.

For canine melanocytic cutaneous neoplasms, the established prognostic scheme found correlation between poor prognosis and tumors with high mitotic count (>3 mitotic figures/10 HPF), ulceration, deep invasion, infiltrative growth, poor pigmentation, and an elevated Ki67 index.³⁵ This scheme does not appear to be relevant to bearded dragon pigmented tumors in our series, given that many of the tumors had a high mitotic count (according to this scheme), were ulcerated, invaded deep into the subcutis and muscle layers, and were poorly pigmented, but with no evidence of lymphatic invasion or reported malignant behavior. None of the chromatophoromas in our series had junctional activity or lentiginous spread, likely because the cell of origin in reptiles is dermal rather than epidermal.¹

As described in other reports of chromatophoromas in bearded dragons, a subset of the neoplasms in our series was associated with PAS-positive, mucinous stroma.¹⁰ This variant was not correlated with a specific subset, although one of these neoplasms also had osseous and cartilaginous metaplasia, which has not been described in other reports in bearded dragons but has been described rarely in human and canine melanomas.^2,3,17,25,32

In all of the cases in our series, metastasis was not noted at the time of excision, and lymphatic invasion was not observed. For the 6 cases in which follow-up clinical information was available, there were no reports of recurrence or neoplasia-related death. Unfortunately, follow-up was not obtained for the few cases that had an increased mitotic count or pleomorphism, although there was no report of metastasis at the time of surgery. As a result, the clinical importance of pleomorphism in these cases is unknown. In a previous study, 2 of 15 cases had an increased mitotic count and pleomorphism, and had reports of visceral metastasis, hence these parameters may be markers of more aggressive behavior.¹⁰ Similar to cutaneous melanocytic neoplasms in dogs and horses, and unlike cutaneous chromatophoromas in snakes, the majority of cutaneous chromatophoromas in bearded dragons may be invasive but do not metastasize, and surgical excision is typically curative.^23,27,35

Although the majority (65%) of the neoplasms in our series had some degree of pigmentation to confirm chromatophore origin, IHC was performed because it is commonly helpful in the diagnosis of poorly pigmented melanocytic neoplasms in mammals.^29,30,34 The majority (16 of 17; 94%) of the cases in our study labeled with S100, and the single case without immunoreactivity was identified as a chromatophoroma based on the intracytoplasmic pigment and PNL2 labeling; this is similar to results obtained from previous studies of chromatophoromas in various reptile species.^10,23 S100 has been found to be a sensitive, but nonspecific, marker for melanocytic neoplasms in mammals.^15,29,30,34 Given that S100 is not specific to pigmented cells, tumors such as peripheral nerve sheath tumors, which have been described in bearded dragons,^16,20 can also label with S100. PNL2 has been shown to be a highly sensitive and specific marker for melanocytes in multiple mammal species.^6,29,30,34 In our case series, PNL2 was much less sensitive than S100, and labeled only 3 of 17 cases. Melan A is also a highly specific marker for melanocytic tumors in humans and dogs.^24,29,34 Previous studies in reptiles^5,10 showed some degree of melan A immunoreactivity in all of their cases, but none of the neoplasms, nor the normal dermal melanocytes in our study, were immunoreactive for melan A. This difference is thought to be the result of variance in antibody concentration, antigen retrieval, and/or time of fixation, because the same monoclonal antibody clone was used. Based on our results and those of others who evaluated chromatophoromas in snakes,²³ the sensitivity of melan A in reptiles appears to be low. IHC was not found to aid differentiation of the different subsets of chromatophoromas.

Typically, histomorphologic features are adequate to diagnose chromatophoromas in bearded dragons. IHC for S100 and potentially PNL2 may aid in the diagnosis of poorly pigmented or nonpigmented chromatophoromas. Pigmented cutaneous lesions in bearded dragons should be removed and submitted for histologic examination. Aggressive surgical excision is often curative, but monitoring of tumors with increased mitotic count, nuclear atypia, and pleomorphism may be warranted.