Pathology of Canine Melanocytic Neoplasms

Not only can melanocytic neoplasms be challenging to treat, they can also be challenging to diagnosis and to prognosticate. Providing a definitive diagnosis for poorly pigmented and amelanotic melanocytic neoplasms is often difficult based on microscopic examination alone, as these tumors can closely mimic poorly differentiated malignant neoplasms such as carcinomas, soft tissue sarcomas, and round cell tumors. It is, however, of the utmost importance to accurately diagnose melanocytic neoplasms, as prognosis and therapy vary greatly between differentials. Immunohistochemical (IHC) markers that are highly sensitive and 100% specific for detecting melanocytes are often required to confirm the diagnosis. Melan-A, PNL-2, tyrosine reactive protein (TRP)-1, and TRP-2 have been found to fulfill these criteria. These markers can be used for IHC either individually or as a cocktail, but the cocktail is more efficient and cost-effective. In addition, a greater percentage of neoplastic cells labels with the cocktail than with each individual antibody resulting in a greater intensity overall. IHC labeling with these markers is most commonly identified within nests of intra-epithelial neoplastic cells that are located adjacent to the underlying mass (Figure 1). Thus, the likelihood of a correct diagnosis increases dramatically when biopsy samples contain ample non-ulcerated, overlying and adjacent epithelium.


Figure 1. Canine oral amelanotic malignant melanoma. Intraepithelial nests of neoplastic cells exhibit lentiginous spread and show strong cytoplasmic red labeling with a melanocytic diagnostic cocktail that contains antibodies against Melan-A, PNL-2, tyrosine reactive protein (TRP)-1, and TRP-2.

Accurately determining the prognosis of canine melanocytic tumors, especially for those of the oral cavity, is also difficult. Historically, location has been used as a strong predictor of prognosis for canine melanocytic neoplasms. In the past it was believed that all melanocytic neoplasms of the oral cavity are malignant and that cutaneous forms are almost always benign. Melanocytic neoplasms of the digit and lip, as a group, have been shown to have survival times somewhere in between those reported for cutaneous and oral neoplasms. While many canine oral melanocytic neoplasms are indeed malignant, recent evidence suggests that a subset of these tumors may have a more favorable prognosis than historically thought. This subset has been termed histologically well differentiated melanocytic neoplasms (HWDMNs). It is important to identify these more benign HWDMNs because, aside from excision, no additional therapy is needed in most cases.

Dogs with HWDMNs are usually older with a mean age of 10.5 years in one study. Golden Retrievers, Labrador Retrievers, and Retriever crosses were overrepresented in two studies of affected dogs, but this may be due to breed popularities. No sex predilection has been noted. These neoplasms are grossly characterized by their small size (generally less than 1cm diameter) heavy pigmentation, lack of ulceration, and raised non-invasive nature. Many are pedunculated with a narrow base (Figure 2). They are generally easy to completely excise. Histologically, they resemble blue nevi in humans.


Figure 2. Raised, heavily pigmented, non-ulcerated, superficial, canine histologically well-differentiated melanocytic neoplasm (HWDMN).

Dogs with HWDMNs have a favorable prognosis and prolonged survival times compared to the more aggressive malignant melanomas (MM). One study reported a mean survival of 23.4 months for dogs that were alive or had died of unrelated causes by the end of the study period and had only been treated via local excision. The mean survival of all dogs still alive at the end of the study was 31.3 months. In that study, only 3 of 64 dogs died of tumor-related causes and all three dogs had neoplasms in the oral cavity as opposed to the lip. The risk of post-surgical local recurrence appears to be greater for neoplasms located in the oral cavity compared to those on the lips but the risk of tumor-associated death was not statistically different in one study. These survival times greatly exceed the 4 to 8 months survival for MMs of the lips and oral cavity reported in other studies. Canine oral MMs are locally invasive with a high degree of metastatic potential and a poor prognosis for long-term survival, similar to human malignant melanomas. They also have similar morphologic features as human mucosal and acral lentiginous malignant melanomas, the most notable being prominent lentiginous spread. Other common features include ulceration, an extensive vertical growth phase, and frequent lymph node metastasis.

Thus, the challenge is to identify more benign oral/lip melanocytic neoplasms and more malignant cutaneous neoplasms. To do this, we use a combination of histologic features and Ki67 index, which is a measure of growth fraction of neoplastic cells. Nuclear atypia, number of mitoses in 10 hpf, and Ki67 index, are the most useful histopathologic parameters to predict prognosis of canine melanocytic neoplasms. Nuclear atypia, however, is subject to inter-observer variation if one does not adhere to strict criteria and it is difficult to assess in some tumors, especially spindloid ones. Mitotic index is also helpful in predicting prognosis but it is not as predictive as nuclear atypia or Ki67 index and it can also be difficult to determine in highly pigmented tumors without bleaching. Ki67 index, on the other hand, can be assessed equally well in heavily or non-pigmented lesions when using a red chromogen and is a more objective measure than histologic criteria. For oral melanocytic neoplasms, the Ki67 index is determined as the average number of positively labeled neoplastic cell nuclei per area of a 1cm2 optical grid reticle at 400 X (5 grid areas counted).


Figure 3. Canine oral malignant melanoma with large numbers of cells showing nuclear labeling for Ki67 (red) on the left. Canine oral histologically well-differentiated melanocytic neoplasm with only a small number of cells with nuclear labeling for Ki67 on the right.

In one study, the mean Ki67 index was significantly higher in oral melanocytic neoplasms from dogs that died within 1 year of diagnosis than the mean for neoplasms from dogs that were still alive at 1 year (p < 0.0005) (Figure 3). A cut-off value for Ki67 index was determined statistically by ROC analysis as 19.5 positive nuclei per grid reticle. The sensitivity and specificity of this threshold value was 87.1% and 85.7%, respectively. In cutaneous neoplasms, the Ki67 index is determined as the percentage of positive cells per 500 neoplastic cells. In one study, a high Ki67 index was associated with a significantly poorer 2-year survival rate. An index greater than 15% was considered high. This method of assessing Ki67 index had a 97% predictive value in this study.

The following flow-chart is helpful for determining the prognosis of canine melanocytic neoplasms based on histologic features and Ki67 index.

Canine Melanocytic Neoplasms*

  1. Assess nuclear atypia in three fields if MC is close to cutoff. Also, if the diagnosis is histologically well-differentiated melanocytic neoplasm of the oral cavity or lip or malignant cutaneous melanoma, consider nuclear atypia.
  2. If MC (per 10 hpf) is 4 or greater for oral/lip or 3 or greater for cutaneous melanoma, then we recommend repeating the count and assessing nuclear atypia. Bleaching may be needed.
  3. Mitotic Count (MC): For oral/lip melanomas, locate a field containing at least one mitotic figure, if possible, and begin counting there; count mitotic figures in 10 consecutive hpf (400X). For cutaneous melanomas, count mitotic figures in 10 consecutive high-powered fields starting in a random location. Avoid ulceration, necrosis and inflammation.
    3    3a
  4. Ki67: Start counting positive nuclei in areas with the highest labeling. Avoid ulcers, necrosis and inflammation when performing counts. For oral melanomas, using a 1 cm2 grid reticle count the number of positive nuclei in 5 grid areas at 40X. For cutaneous melanomas, determine the percentage of positive nuclei per 500 cells.
    #4  this one

*Modified from: Smedley, RC., et al. (2011). Prognostic markers for canine melanocytic neoplasms: A comparative review of the literature and goals for future investigation. Vet Pathol 48: 54-72.

Histologic and IHC classification of canine melanocytic neoplasms have improved, but the genetic features of these neoplasms are still poorly understood. A recent trend in human medicine is to integrate both morphologic and genetic features in order to improve melanoma classification and to determine proper targeted therapies. Genetic features commonly found in human melanomas (e.g. mutations in NRAS, BRAF, c-Kit and G-proteins of the Gαq family of GTPases) appear to be rare in canine melanocytic neoplasms. However, because canine oral malignant melanomas have similar morphologic features, clinical behavior, and poor response to therapy as certain subtypes of human melanomas that have been shown to have c-KIT mutations, the role of c-Kit in canine melanocytic neoplasms has recently been explored in more detail. A small number of published c-Kit mutations have been identified in canine melanocytic neoplasms, but most are silent mutations with no significance to tumorigenesis. Only 2 missense mutations have been identified in two separate studies, each in a single canine melanoma and one (L579P) was an activating mutation equivalent to human KIT L576P, the most common activating c-KIT mutation found in human melanomas. We recently performed whole exome sequencing of c-Kit in a set of canine melanocytic neoplasms and identified nine significant c-Kit mutations, but none were likely to be driving mutations (unpublished data).

IHC expression of KIT has also been examined in canine melanocytic neoplasms in a limited number of cases and it appears to vary considerably between cutaneous and oral/lip melanocytic neoplasms. One study reported a greater percentage and intensity of KIT expression in dermal melanocytomas compared to cutaneous malignant melanomas but did not find any association between KIT expression in cutaneous malignant melanomas and any of the examined histopathologic criteria, except between labeling intensity and degree of tumor pigmentation. Similarly, in a recent study of canine oral melanocytic neoplasms, there was no association between KIT expression or mutation status and any particular histopathologic feature (unpublished data). Another study demonstrated strong KIT labeling in 14 of 14 dermal melanocytomas and reported that dermal melanocytomas were more likely to label for KIT than malignant mucosal melanomas; dermal malignant melanomas and oral HWDMNs were not evaluated in this study, however. No studies have found any significant associations between KIT labeling and mutation status or survival. It is also important to report the location of positively labeling melanocytes within a neoplasm as labeling in the in situ component of mucosal melanomas seems to be consistently diffuse and strong and therefore not related to prognosis or c-KIT mutation in status in humans or dogs. In some studies it is not clear where the labeling was located, which makes comparisons between studies difficult. Significant KIT labeling of the submucosal component of canine oral/lip HWDMNs and oral MMs actually appears to be rare (unpublished data).

So the question is, is there any evidence to support the use of tyrosine kinase inhibitor (TKI) therapy that target c-Kit mutations in dogs with malignant melanoma? Based on the current data, there is no indication for generalized use of TKIs that target c-Kit mutations in these neoplasms. However, additional studies are needed to evaluate the potential effectiveness of TKIs in treating melanomas that do have significant KIT expression, as there may be upstream effects that could be targeted.

Submitted by:

Rebecca C. Smedley, DVM, MS, Dipl ACVP

Matti Kiupel, Dr. vet. med. Dr. habil., PhD, Dipl ACVP


Bastian BC. The Molecular Pathology of Melanoma: An Integrated Taxonomy of Melanocytic Neoplasia. Annu Rev Pathol Mech Dis 2014; 9:239-271.

Bergin IL, Smedley RC, Esplin DG, Spangler WL, Kiupel MK: Prognostic evaluation of canine oral melanomas.

Bergman, P., and Wolchok, J. (2008). Of Mice and Men (and Dogs): development of a xenogeneic DNA vaccine for canine oral malignant melanoma. Cancer Ther. 6, 817–826.

Bergman, P.J., Kent, M.S., and Farese, J.P. (2013). Melanoma. In Withrow and MacEwen’s Small Animal Clinical Oncology. S.J., Withrow, D.M., Vail, and R.L., Page, eds. (St. Louis, MO: Elsevier/Saunders), pp. 321–334.

Broekaert SM, Roy R, Okamoto I, van den Oord J, Bauer J, Garbe C, Barnhill RL, Busam KJ, Cochran AJ, Cook MG, Elder DE, McCarthy SW, Mihm MC, Schadendorf D, Scolyer RA, Spatz A, Bastian BC. Genetic and morphologic features for melanoma classification. Pigment Cell Melanoma Res. 2010 Dec;23(6):763-770.

Chu PY, Pan SL, Liu CH, Lee J, Yeh LS, Liao AT. KIT gene exon 11 mutations in canine malignant melanoma. Vet J. 2013 May;196(2):226-230.

Curtin JA, Fridlyand J, Kageshita T, Patel HN, Busam KJ, Kutzner H, Cho KH, Aiba S, Bröcker EB, LeBoit PE, Pinkel D, Bastian BC. Distinct sets of genetic alterations in melanoma. N Engl. J. Med. 2005 353:2135-2147.

Esplin, DG. (2008). Survival of dogs following surgical excision of histologically well-differentiated melanocytic neoplasms of the mucous membranes of the lips and oral cavity. Vet Pathol 45: 889-896.

Gomes J, Queiroga FL, Prada J, Pires I. Study of c-kit immunoexpression in canine cutaneous melanocytic tumors. Melanoma Res. 2012 Jun;22(3):195-201.

Gramer I, Kessler M, Geyer J. Detection of novel polymorphisms in the ckit gene of canine patients with lymphoma, melanoma, haemangiosarcoma, and osteosarcoma. Vet Res Commun Vet Res Commun. 2016 Jun;40(2):89-95.

Laprie C, Abadie J, Amardeilh MF, et al. MIB-1 immunoreactivity correlates with biologic behaviour in canine cutaneous melanoma. Vet Derm 2001, 12: 139-147.

Murakami A, Mori T, Sakai H, Murakami M, Yanai T, Hoshino Y, Maruo K. Analysis of KIT expression and KIT exon 11 mutations in canine oral malignant melanomas. Veterinary Comp Oncol 2011; 9(3):219-224.

Newman SJ, Jankovsky JM, Rohrbach BW, LeBlanc AK. C-kit expression in canine mucosal melanomas. Vet Pathol. 2012 Sep;49(5):760-765

Prasad, M.L., Patel, S.G., Huvos, A.G., Shah, J.P., and Busam, K.J. (2004). Primary mucosal melanoma of the head and neck: a proposal for microstaging localized, Stage I (lymph node-negative) tumors. Cancer 100, 1657–1664.

Simpson RM, Bastian BC, Michael HT, Webster JD, Prasad ML, Conway CM, Prieto VM, Gary JM, Goldschmidt MH, Esplin DG, Smedley RC, Piris A, Meuten DJ, Kiupel M, Lee CC, Ward JM, Dwyer JE, Davis BJ, Anver MR, Molinolo AA, Hoover SB, Rodriguez-Canales J, Hewitt SM. Sporadic naturally occurring melanoma in dogs as a preclinical model for human melanoma. Pigment Cell Melanoma Res. 2014 Jan;27(1):37-47.

Smedley RC, Lamoureux J, Sledge DG, Kiupel M: Immunohistochemical diagnosis of canine oral amelanotic melanomas. Vet Pathol 2011. 48: 32-40.

Smedley RC, Spangler WL, Esplin DG, Kitchell BE, Bergman PJ, Ho H-Y, Bergin IL, Kiupel M.   Prognostic Markers for Canine Melanocytic Neoplasms: A Comparative Review of the Literature and Goals for Future Investigation Vet Pathol 2011. 48: 54-72.

Spangler WL, Kass PH. The histologic and epidemiologic bases for prognostic considerations in canine melanocytic neoplasia. Vet. Pathol. 2006. 43(2): 136-142

Posted in ,