Tumor size as a predictor of lymphatic invasion in oral melanomas of dogs

Kenneth A. Carroll 1Department of Small Animal Medicine and Surgery, College of Veterinary Medicine, North Carolina State University, Raleigh, NC 27606.

Search for other papers by Kenneth A. Carroll in
Current site
Google Scholar
PubMed
Close
 BVSc
,
Charles A. Kuntz 2Southpaws Specialty Surgery for Animals, Moorabbin, VIC 3189, Australia.

Search for other papers by Charles A. Kuntz in
Current site
Google Scholar
PubMed
Close
 DVM, MS
,
Jane Heller 3School of Animal and Veterinary Sciences, Charles Sturt University, Wagga Wagga, NSW 2650, Australia.

Search for other papers by Jane Heller in
Current site
Google Scholar
PubMed
Close
 BVSc, PhD
,
Andrew Peters 3School of Animal and Veterinary Sciences, Charles Sturt University, Wagga Wagga, NSW 2650, Australia.

Search for other papers by Andrew Peters in
Current site
Google Scholar
PubMed
Close
 BVSc, PhD
,
Randi Rotne 3School of Animal and Veterinary Sciences, Charles Sturt University, Wagga Wagga, NSW 2650, Australia.

Search for other papers by Randi Rotne in
Current site
Google Scholar
PubMed
Close
 BVSc, PhD
, and
Andrew Dunn 4Adelaide Veterinary Specialist and Referral Centre, Norwood, SA 5067, Australia.

Search for other papers by Andrew Dunn in
Current site
Google Scholar
PubMed
Close
 BVSc

Abstract

OBJECTIVE

To investigate putative associations between oral melanoma size and variables of histologic grade such as mitotic index, nuclear atypia, junctional activity, ulceration, lymphatic invasion, and degree of pigmentation.

SAMPLE

59 samples of oral melanomas from dogs sourced from 6 diagnostic laboratories within Australia.

PROCEDURES

The size of each melanoma was microscopically measured, and each sample was evaluated for variables of histologic grade including mitotic index, nuclear atypia, junctional activity, ulceration, lymphatic invasion, and degree of pigmentation by a veterinary pathologist. The association between tumor size and histologic outcomes was then statistically evaluated.

RESULTS

A significant relationship was identified between the size of oral melanomas and a single variable of histologic grade, lymphatic invasion, with larger tumors more likely to show lymphatic invasion. Further analysis revealed 2 applicable size thresholds for different clinical scenarios. Results indicated lymphatic invasion can confidently be ruled out for tumors < 6.5 mm in diameter (100% sensitivity) and ruled in for tumors ≥ 24.5 mm in diameter (100% specificity).

CONCLUSIONS AND CLINICAL RELEVANCE

An association was found for oral melanomas of dogs between tumor size and lymphatic invasion.

Abstract

OBJECTIVE

To investigate putative associations between oral melanoma size and variables of histologic grade such as mitotic index, nuclear atypia, junctional activity, ulceration, lymphatic invasion, and degree of pigmentation.

SAMPLE

59 samples of oral melanomas from dogs sourced from 6 diagnostic laboratories within Australia.

PROCEDURES

The size of each melanoma was microscopically measured, and each sample was evaluated for variables of histologic grade including mitotic index, nuclear atypia, junctional activity, ulceration, lymphatic invasion, and degree of pigmentation by a veterinary pathologist. The association between tumor size and histologic outcomes was then statistically evaluated.

RESULTS

A significant relationship was identified between the size of oral melanomas and a single variable of histologic grade, lymphatic invasion, with larger tumors more likely to show lymphatic invasion. Further analysis revealed 2 applicable size thresholds for different clinical scenarios. Results indicated lymphatic invasion can confidently be ruled out for tumors < 6.5 mm in diameter (100% sensitivity) and ruled in for tumors ≥ 24.5 mm in diameter (100% specificity).

CONCLUSIONS AND CLINICAL RELEVANCE

An association was found for oral melanomas of dogs between tumor size and lymphatic invasion.

The oral cavity is a common location for the development of neoplasia in small animals.1,2 Oral tumors can be difficult to manage surgically because of their close association with local periodontal structures including bone. As the most common malignant oral tumor in dogs, oral melanoma typically occurs on the gingiva and is characterized by rapid growth, local invasion, and early metastasis.3,4

Because of their variable biological nature, oral melanomas in dogs are best characterized by clinical stage and histologic grade.5–7 Clinical stage refers to the anatomic extent of disease and is determined by findings on physical examination, laboratory analyses, and diagnostic imaging techniques, whereas grade is defined by the histologic appearance of a neoplasm.

In dogs with oral melanomas, determination of clinical stage and histologic grade provide independent prognosticators related to duration of remission and MST.5,8–10 However, their combined effect on the predicted clinical course of disease is not well established in the veterinary literature. The aspect of clinical stage most easily measured in a clinical setting is tumor size; however, to our knowledge, the use of tumor size to help predict the histologic behavior of oral melanomas has not been determined for dogs.

The purpose of the study reported here was to investigate the putative associations between oral melanoma size in dogs and continuous (ie, MI and nuclear atypia) and dichotomous (ie, MI, nuclear atypia, junctional activity, ulceration, lymphatic invasion, and degree of pigmentation) variables of histologic grade.

Materials and Methods

Selection of oral melanoma samples

Fifty-nine oral melanoma samples of dogs were obtained from 6 veterinary diagnostic centers across Australia, including the ASAP Laboratory, Melbourne, VIC (n = 30); Gribbles Veterinary Pathology, Melbourne, VIC (10); Veterinary Laboratory Services at the University of Queensland, Gatton, QLD (8), Gribbles Veterinary Pathology, Adelaide, SA (5); Idexx Laboratories, Melbourne, VIC (5); and the Veterinary Diagnostic Laboratory at Charles Sturt University, Wagga Wagga, NSW (1). At all laboratories, electronic records for a 10-year period (January 2006 to January 2016) were searched in the canine database with the terms oral melanoma and oral malignant melanoma.

Fulfillment of the following criteria was necessary for tumor inclusion in the study: a definitive histopathologic diagnosis of oral melanoma, as confirmed by a board-certified veterinary pathologist; complete excisional biopsy of the tumor with clean surgical margins of ≥ 1 cm (as confirmed by the referring surgeon and included in the histopathology report); availability of a full histopathology report written by a board-certified veterinary pathologist; and availability of clinical information for the patient, including age, breed, sex, and anatomic location of the tumor within the oral cavity. Once approved for inclusion in the study, the laboratories supplied the oral melanoma tissue blocks and clinical data to the Veterinary Diagnostic Laboratory at Charles Sturt University. All tissue blocks were cut at their largest diameter as 4-μm-thick sections and stained with routine H&E in preparation for histologic analysis.

Data collection

All relevant clinical data including age, breed, and sex of the patient and anatomic location of the tumor were initially collated into a template. Histologic evaluation of oral melanoma samples was undertaken by a veterinary pathologist (AP) without knowledge of the clinical outcome for the patient.

Size measurement of oral melanomas— Size of each oral melanoma was recorded directly from the tissue section. As each sample was a full excisional biopsy of the tumor with adequate surgical margins of ≥ 1 cm, each section was first scanned at low power (magnification, 40×) for the spread of neoplastic growth. The tumor margins were measured in 3 dimensions directly from the section with a ruler. The largest measurement was used as the final tumor size in millimeters.

Histologic analysis of oral melanomas—Oral melanomas were evaluated for the following histologic variables of malignancy: junctional activity, lymphatic invasion, ulceration, MI, nuclear atypia, and degree of pigmentation. These histologic variables were selected in the study because they were part of the current recommendations for the prognostication of melanocytic neoplasms in dogs.8

Junctional activity, ulceration, and lymphatic invasion were all assessed for their presence or absence (yes or no). Junctional activity referred to the nest-like proliferation of neoplastic cells at the dermoepidermal junction (Figure 1). Ulceration was defined as partial or complete loss of the epidermis and was often accompanied by focal areas of hemorrhage and edema (Figure 2). Lymphatic invasion was characterized by the presence of neoplastic cells within lymphatic vessels (Figure 3).

Figure 1—
Figure 1—

Photomicrographs of sections of oral melanomas from dogs. A—When dermoepidermal junctional activity is absent, the highly pigmented melanocytes do not penetrate the overlying epidermis. B—In the presence of junctional activity, neoplastic melanocytes form nest-like structures at the dermoepidermal junction. H&E stain; bars = 500 μm.

Citation: Journal of the American Veterinary Medical Association 256, 10; 10.2460/javma.256.10.1123

Figure 2—
Figure 2—

Photomicrographs of sections of oral melanomas from dogs. A— When ulceration is absent, the epidermis is intact. B—In the presence of ulceration, notice the partial loss of the overlying epidermis with evidence of hemorrhage and edema. H&E stain; bars = 200 μm.

Citation: Journal of the American Veterinary Medical Association 256, 10; 10.2460/javma.256.10.1123

Figure 3—
Figure 3—

Photomicrographs of sections of an oral melanoma undergoing lymphatic invasion from a dog. A and B—Melanocytes form clusters within the lumen of a lymphatic vessel. Mitotic figures are evident (B; arrows). C and D—Neoplastic melanocytes are present in the lumen of a lymphatic vessel. Notice that cells appear to be budding from the lymphatic wall (C). H&E stain; bars = 200 μm (A and C) and 100 μm (B and D).

Citation: Journal of the American Veterinary Medical Association 256, 10; 10.2460/javma.256.10.1123

A differential cell counter was used in determination of the MI. The number of mitotic figures in 10 consecutive nonoverlapping fields at high power (magnification, 400×) was counted. Counts commenced in the area of highest mitotic activity and least ulceration. Variation in the appearance of cell nuclei was assessed through evaluation of nuclear atypia. Melanocytic cells were classified as typical or atypical with a previously established grading system.9 Typical melanocytes were defined by a small nucleus, single, centrally orientated nucleolus, and minimal clumping of chromatin. Atypical melanocytes were defined by larger and often multiple nucleoli with less regular shape that were eccentrically located in the nucleus (Figure 4). Within 1 hpf (magnification, 400×), 200 neoplastic cells were evaluated with a differential cell counter and subjectively classified as either typical or atypical on the basis of the appearance of their nuclei. The degree of pigmentation was assessed subjectively at low power (magnification, 40×). A modification of a previously established histologic scale was used to evaluate and characterize histologic sections as having high (≥ 50%) or low (< 50%) pigment.10

Figure 4—
Figure 4—

Photomicrographs of sections of an oral melanoma with nuclear atypia from a dog. A—Atypical melanocytes form nest-like structures, which are surrounded by connective tissue. B—Melanocytes appear atypical, as evidenced by frequent multiple nucleoli. C—Atypical melanocytes contain multiple nucleoli that are eccentrically located within the nucleus (arrows). D—Mixture of typical (vertical arrows) and atypical (horizontal arrows) melanocytes. The typical melanocytes possess a small nucleus, single centrally orientated nucleolus, and minimal clumping of chromatin. H&E stain; bars = 100 μm (A and B) and 50 μm (C and D).

Citation: Journal of the American Veterinary Medical Association 256, 10; 10.2460/javma.256.10.1123

Statistical analysis

A convenience sampling technique was used to select the cases for inclusion in this study. The sample size calculation was conducted post hoc by use of an online epidemiological calculator to check for adequate power.a To detect a size difference of 12 mm in diameter between variables of histologic grade with a power of 80% and confidence interval of 95%, a sample size of 10 (5 samples in each group) was required.

Clinical data were summarized by use of descriptive statistics (ie, median and mean ± SD) and compared between groups with a combination of graphical and univariable analyses; oral melanoma size was assessed in terms of sex with an independent t test, and anatomic location was assessed with a 1-way ANOVA.

Variables of histologic grade were evaluated with a combination of descriptive statistics and graphical and univariable analyses. Nuclear atypia and MI were assessed on a continuous and nominal scale. Continuous assessment involved Pearson correlation analyses between size and each variable with graphical representation on a scatterplot. Along with other variables of histologic grade (ie, junctional activity, ulceration, lymphatic invasion, and degree of pigmentation), nuclear atypia and MI were also assessed for their relationship to tumor size dichotomously by use of an independent t test and boxplot graphs. For any significant findings, an ROC analysis was performed. These curves helped determine threshold tumor measurements that provided the highest specificity and sensitivity for each variable. All statistical analyses were performed with a commercially available software program,b and values of P < 0.05 were considered significant.

Results

Oral melanoma samples included in this study were from 26 breeds of dog, including both purebred and mixed-breed dogs. Labrador Retrievers (n = 8), Staffordshire Bull Terriers (7), Cocker Spaniels (6), and Golden Retrievers (4) were the most common breeds, whereas the Bloodhound (1), German Shepherd Dog (1), and Great Dane (1) breeds were uncommon. The largest mean tumor size was found in Golden Retrievers (18.75 mm in diameter), and the smallest mean tumor size was found in mixed-breed dogs (8.67 mm in diameter).

Dogs ranged in age from 4 to 16 years, with a mean and median age of 11.8 and 12 years, respectively. Of 59 dogs, 30 (51%) were females and 29 (49%) were males. The gingiva was the most common location for oral melanomas (n = 36), followed by the buccal mucosa (8), inner surface of the lip (5), and tongue (5). Less common sites included the hard palate (n = 3) and tonsils (2). Tumor size ranged from 2 to 26 mm (median, 12 mm; mean, 14.07 ± 5.83 mm) in diameter. Tumors were largest on the buccal mucosa (mean, 16.50 ± 5.61 mm in diameter) and tonsils (mean, 16.50 ± 6.36 mm in diameter) and smallest on the hard palate (mean, 8.67 ± 2.52 mm in diameter). No significant differences in tumor size were detected related to sex (P = 0.824) or location (P = 0.454).

In the 59 samples, mean and median MI were 46.2 ± 41.7 and 40 mitotic figures/10 hpf, respectively. The highest MI was 224 mitotic figures/10 hpf, which was observed in a tumor from a 12-year-old Golden Retriever. When MI was assessed linearly on a continuous scale, no significant (P = 0.569) association was found between tumor size and MI, and only 0.6% of the total variance of MI could be explained by the size of the tumor. When MI was assessed dichotomously, no significant (P = 0.917) difference was found in the size of oral melanomas with an MI < 4 mitotic figures/10 hpf (n = 7; mean, 14.29 ± 7.41 mm in diameter) and oral melanomas with an MI ≥ 4 mitotic figures/10 hpf (52; mean, 14.04 ± 5.67 mm in diameter; Table 1).

Table 1—

Dichotomous analysis of oral melanoma size (mm) and variables of histologic grade in 59 tumors from dogs.

Histologic variableResultSample size (n)Mean size (mm)SDSEMt Test resultP value
Junctional activityYes4414.685.430.821.3980.167
 No1512.276.731.74  
UlcerationYes5414.465.700.781.7420.087
 No59.806.062.71  
Lymphatic invasionYes4515.025.750.862.3420.023*
 No1411.005.111.37  
Nuclear atypia (%)< 301413.575.881.57−0.3620.718
 ≥ 304514.225.870.87  
MI (per 10 hpf)< 4714.297.412.80−0.1050.917
 ≥ 45214.045.670.79  
Degree of pigmentation (%)< 505214.045.670.790.1050.917
 ≥ 50714.297.412.80  

Significant (P < 0.05) difference between tumor sizes for dichotomous outcome.

No significant (P = 0.280) association was found between oral melanoma size and nuclear atypia when assessed on a linear scale. Mean and median nuclear atypia was 34.5 ± 10.8% and 35.6%, respectively. The highest score of nuclear atypia was 60%, and this was observed in a tumor from a 13-year-old Staffordshire Bull Terrier. Also, when nuclear atypia was categorized as < 30% (n = 14; mean size, 13.57 ± 5.88 mm in diameter) or ≥ 30% (45; mean size, 14.22 ± 5.87 mm in diameter), no significant (P = 0.718) association was found between oral melanoma size and nuclear atypia (Table 1).

When assessed dichotomously, no significant associations were found between tumor size and junctional activity (P = 0.167), ulceration (P = 0.087), or degree of pigmentation (P = 0.917). However, a significant (P = 0.023) association was detected between tumor size and lymphatic invasion. A 4.02-mm difference in diameter was found in the mean size of invasive tumors (n = 45; mean, 15.02 ± 5.75 mm) and noninvasive tumors (14; mean, 11.00 ± 5.11 mm; Table 1).

Results of ROC analysis revealed acceptable threshold values of tumor size to distinguish between tumors with and without lymphatic invasion. A sensitivity of 100% was seen at a tumor size ≥ 6.5 mm in diameter, and specificity of 100% was obtained at a size ≥ 24.5 mm in diameter. Although significant (P = 0.047), the area under the ROC curve for tumor size in predicting lymphatic invasion was 0.678 (95% confidence interval, 0.521 to 0.835) and was considered poor. The Youden index was used to determine the optimal tumor size with the highest combined sensitivity and specificity to detect lymphatic invasion. The Youden index was maximized at a tumor size of ≥ 16.5 mm in diameter, with a sensitivity and specificity of 42.2% and 92.9%, respectively.11

Discussion

The relationship between size and histologic grade of tumors is of considerable relevance to veterinary clinicians. In the study presented here, potential associations between a variable that can be assessed at clinical examination (ie, tumor size) and histologic outcomes for dogs with oral melanomas were evaluated. A significant association was found between tumor size and the histologic variable of lymphatic invasion. These findings demonstrated that larger tumors are more likely to invade the lymphatic system.

Findings in the present study agreed with epidemiological findings of previous reports.6,12–14 The tumors were mostly from purebred animals, and Labrador Retrievers (8/59 [14%] dogs) and Staffordshire Bull Terriers (7/59 [12%] dogs) were most common in the population. It is difficult to assess the bias associated with submission of the tumor samples, as the distribution of referral versus nonreferral cases was unknown. However, a Labrador Retriever and Staffordshire Bull Terrier breed predisposition for oral melanomas has been established.2,5,7 In agreement with findings in other studies,2,4,6,12–14 the mean age of dogs for tumor development was 11.8 years, tumors were most common on the gingiva (36/59 [61%] tumors), and sex distribution was almost equal between males (49%) and females (51%).

Histologic variables were assessed in the study presented here. When compared with tumor size, no significant results were detected for junctional activity, ulceration, nuclear atypia, MI, and degree of pigmentation. Three of these histologic variables, nuclear atypia, MI, and degree of pigmentation, have been shown to be useful prognostic indicators for dogs with oral melanomas, with significant relationships to MST and remission duration.8 The only variable found to have a significant association to tumor size in the present study was lymphatic invasion.

Results of ROC analysis of tumor size revealed acceptable threshold values for sensitivity and specificity for the histologic variable of lymphatic invasion. A single cutoff value for tumor size that was useful to optimize both sensitivity and specificity for lymphatic invasion could not be identified in the present study. The Youden index was maximized at a tumor size of ≥ 16.5 mm in diameter, but with a sensitivity and specificity of only 42.2% and 92.9%, respectively. However, the use of different cutoff values for varied clinical scenarios may make it possible to optimize either sensitivity or specificity at a single point. To optimize sensitivity (100%) in which lymphatic invasion could confidently be ruled out, a cutoff value for tumor size of 6.5 mm in diameter was required. Although this measurement was low in specificity, a negative result (ie, tumor size < 6.5 mm in diameter) would indicate that the patient does not have lymphatic invasion. To obtain 100% specificity, which requires certainty that lymphatic invasion had occurred, a cutoff value for tumor size of 24.5 mm in diameter was required. As specificity detects the proportion of nondiseased animals that test negative, any positive results (ie, tumors ≥ 24.5 mm in diameter) would be 100% predictive for lymphatic invasion. Unfortunately, for tumors between 6.5 and 24.5 mm in diameter, no definitive information regarding lymphatic invasion was determined in the study presented here.

Lymphatic invasion has been described as the gold standard for identification of malignancy of melanocytic neoplasms of the mouth.7,8,15 In 2002, a study16 confirmed that lymphatic invasion is negatively associated with patient survival in dogs with oral melanomas. Of the 52 melanomas investigated, a significant (P = 0.014) difference was observed in the MST of dogs with (7.3 months) and without (10.5 months) lymphatic invasion.

However, it must be stressed that no conclusions on prognosis can be made from the study presented here, as clinical outcomes were not assessed. Although tumor size thresholds for predicting lymphatic invasion were found, follow-up data for the patient population were not available. More accurately, the present study demonstrated that tumor size is simply predictive of lymphatic invasion activity in oral melanomas, and future prospective studies investigating clinical outcome are necessary to determine the prognostic importance of this relationship.

Several limitations are associated with the study presented here. As it was a retrospective study and samples were sourced from 6 veterinary diagnostic laboratories, variation within the oral melanoma samples across the different laboratories may have been present. Differences may have occurred because of variation within the sample population, surgical technique at the referring veterinary hospitals, and processing at the diagnostic facility. Because of the retrospective design, difficulty existed in obtaining clinical data from the referring veterinarians. Although epidemiological information was readily available, only a single case (1/59 subjects) contained data regarding regional lymph node involvement and distant metastases. If this information were available for all cases, the World Health Organization staging system for oral melanomas could have been incorporated into the study, and this would have allowed for better comparison to other findings within the veterinary literature. In addition, the study was constrained by a limited sample size of 59 oral melanomas from dogs.

Another limitation of the study presented here involved the measurement of tumor size. As gross size measurements were not readily available from each referring veterinarian, a microscopic measurement of the tumor's largest diameter was taken directly from the tissue section. Tumor size measurements must be accurate, as small discrepancies can affect clinical staging and subsequent treatment. Within the human literature, microscopic measurement is regarded as the gold standard for the estimation of tumor size, and the College of American Pathologists recommends that this technique always be used for the measurement of tumor size for clinical staging.17–19 However, because of the size of the microscope slides, this technique also had an upper measurement limit of 75 mm. Tumors smaller than this size were measured with an exact diameter at their largest dimension, whereas melanomas larger than 75 mm would be too large to fit on the section. It is also noted that microscopic measurements may differ from the actual gross size as estimated by veterinary clinicians.

Microscopic measurements may underestimate the size of a tumor because of the effects of formalin processing.19 Formalin actively diffuses through tissues, binds to amino groups, and causes formation of cross-linked proteins, which can cause extensive histologic changes to the tissue. A study20 that examined the change in size of tumors on oral tissues from dogs following excision, formal fixation, and slide preparation found substantial shrinkage in tumor margins, ranging from 34.5% to 47.3%. Numerous studies in humans demonstrate considerable shrinkage in tumor size before and after formalin fixation in oral (25%),21 esophageal (10%),22 prostatic (4.5%),23 and renal (11.4%)24 tumors. The shrinkage effects of formalin processing cannot be underestimated, and this must be considered when reporting tumor size, as it may result in understaging of dogs with oral melanomas.

To the authors' knowledge, the study presented here was the first in the veterinary literature that has compared and found an association between tumor size and a variable of histologic grade for oral melanomas in dogs. Although several limitations have been identified for the study reported here, the important findings warrant further evaluation in a larger prospective study. In the present study, 2 tumor size threshold measurements were determined for a high sensitivity or specificity outcome in terms of prognostication; however, many oral melanomas in dogs would fall between the identified 2 tumor size thresholds. In the future, a larger study may provide stronger threshold values and may also establish more relationships between tumor size and other variables of histologic grade that were not identified in the present study.

On the basis of the findings of the study presented here, no conclusions on prognosis of oral melanomas in dogs can be made, as clinical outcomes were not investigated. However, the documented association between tumor size and lymphatic invasion may be useful in providing the blueprint for further prospective investigations on oral melanomas in dogs. Inclusion of clinical outcomes (MST and remission duration) would identify the true prognostic potential of tumor size and its relationship to lymphatic invasion and thus be clinically useful for veterinary clinicians, pathologists, and oncologists alike.

Acknowledgments

This study was not supported by a grant. The authors declare that there were no conflicts of interest.

ABBREVIATIONS

MI

Mitotic index

MST

Median survival time

ROC

Receiver operating characteristic

Footnotes

a.

Epitools Epidemiological Calculator, Ausvet Pty Ltd. Available at: epitools.ausvet.com.au. Accessed May 10, 2019.

b.

SPSS, version 2.2, IBM Corp, Armonk, NY.

References

  • 1. Uzal FA, Plattner BL, Hostetter JM. Alimentary system. In: Maxie G, ed. Jubb, Kennedy & Palmer's pathology of domestic animals. 6th ed. St Louis: Elsevier Health Sciences, 2015;2627.

    • Search Google Scholar
    • Export Citation
  • 2. Ramos-Vara JA, Beissenherz ME, Miller MA, et al. Retrospective study of 338 canine oral melanomas with clinical, histologic, and immunohistochemical review of 129 cases. Vet Pathol 2000;37:597608.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 3. Modiano JF, Ritt MG, Wojcieszyn J. The molecular basis of canine melanoma: pathogenesis and trends in diagnosis and therapy. J Vet Intern Med 1999;13:163174.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 4. Nishiya AT, Massoco CO, Felizzola CR, et al. Comparative aspects of canine melanoma. Vet Sci 2016;3:710.

  • 5. Hahn KA, DeNicola DB, Richardson RC, et al. Canine oral malignant melanoma: prognostic utility of an alternative staging system. J Small Anim Pract 1994;35:251256.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6. Bergman PJ. Canine oral melanoma. Clin Tech Small Anim Pract 2007;22:5560.

  • 7. Smith SH, Goldschmidt MH, McManus PM. A comparative review of melanocytic neoplasms. Vet Pathol 2002;39:651678.

  • 8. Smedley RC, Spangler WL, Esplin DG, et al. Prognostic markers for canine melanocytic neoplasms: a comparative review of the literature and goals for future investigation. Vet Pathol 2011;48:5472.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9. Spangler WL, Kass PH. The histologic and epidemiologic bases for prognostic considerations in canine melanocytic neoplasia. Vet Pathol 2006;43:136149.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 10. Bergin IL, Smedley RC, Esplin DG, et al. Prognostic evaluation of Ki67 threshold value in canine oral melanoma. Vet Pathol 2011;48:4153.

  • 11. Fluss R, Faraggi D, Reiser B. Estimation of the Youden index and its associated cutoff point. Biom J 2005;47:458472.

  • 12. Bergman PJ, Kent MS, Farese JP. Melanoma. In: Withrow SJ, Vail DM, Page RL, eds. Withrow and MacEwen's small animal clinical oncology. 5th ed. St Louis: Elsevier Health Sciences, 2012;321334.

    • Search Google Scholar
    • Export Citation
  • 13. Dennis MM, Ehrhart N, Duncan CG, et al. Frequency of and risk factors associated with lingual lesions in dogs: 1,196 cases (1995–2004). J Am Vet Med Assoc 2006;228:15331537.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 14. Todoroff RJ, Brodey RS. Oral and pharyngeal neoplasia in the dog: a retrospective survey of 361 cases. J Am Vet Med Assoc 1979;175:567571.

    • Search Google Scholar
    • Export Citation
  • 15. Goldschmidt MH. Histological classification of epithelial and melanocytic tumors of the skin of domestic animals. Washington, DC: Armed Forces Institute of Pathology, 1998;331340.

    • Search Google Scholar
    • Export Citation
  • 16. Millanta F, Fratini F, Corazza M, et al. Proliferation activity in oral and cutaneous canine melanocytic tumours: correlation with histological parameters, location, and clinical behaviour. Res Vet Sci 2002;73:4551.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 17. Behjatnia B, Sim J, Bassett LW, et al. Does size matter? Comparison study between MRI, gross, and microscopic tumor sizes in breast cancer in lumpectomy specimens. Int J Clin Exp Pathol 2010;3:303309.

    • Search Google Scholar
    • Export Citation
  • 18. Fitzgibbons PL, Page DL, Weaver D, et al. Prognostic factors in breast cancer. Arch Pathol Lab Med 2000;124:966978.

  • 19. Verma VP, Kaur N, Agarwal N, et al. Intra-operative measurement of tumour size in breast cancer and its comparison with other methods: a prospective study. Ecancermedicalscience 2008;2:96.

    • Search Google Scholar
    • Export Citation
  • 20. Johnson RE, Sigman JD, Funk GF, et al. Quantification of surgical margin shrinkage in the oral cavity. Head Neck 1997;19:281286.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 21. Mohiyuddin SMA, Padiyar BV, Suresh TN, et al. Clinicopathological study of surgical margins in squamous cell carcinoma of buccal mucosa. World J Otorhinolaryngol Head Neck Surg 2016;2:1721.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 22. Siu KF, Cheung HC, Wong J. Shrinkage of the esophagus after resection for carcinoma. Ann Surg 1986;203:173176.

  • 23. Jonmarker S, Valdman A, Lindberg A, et al. Tissue shrinkage after fixation with formalin injection of prostatectomy specimens. Virchows Arch 2006;449:297301.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 24. Tran T, Sundaram CP, Bahler CD, et al. Correcting the shrinkage effects of formalin fixation and tissue processing for renal tumors: toward standardization of pathological reporting of tumor size. J Cancer 2015;6:759766.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Figure 1—

    Photomicrographs of sections of oral melanomas from dogs. A—When dermoepidermal junctional activity is absent, the highly pigmented melanocytes do not penetrate the overlying epidermis. B—In the presence of junctional activity, neoplastic melanocytes form nest-like structures at the dermoepidermal junction. H&E stain; bars = 500 μm.

  • Figure 2—

    Photomicrographs of sections of oral melanomas from dogs. A— When ulceration is absent, the epidermis is intact. B—In the presence of ulceration, notice the partial loss of the overlying epidermis with evidence of hemorrhage and edema. H&E stain; bars = 200 μm.

  • Figure 3—

    Photomicrographs of sections of an oral melanoma undergoing lymphatic invasion from a dog. A and B—Melanocytes form clusters within the lumen of a lymphatic vessel. Mitotic figures are evident (B; arrows). C and D—Neoplastic melanocytes are present in the lumen of a lymphatic vessel. Notice that cells appear to be budding from the lymphatic wall (C). H&E stain; bars = 200 μm (A and C) and 100 μm (B and D).

  • Figure 4—

    Photomicrographs of sections of an oral melanoma with nuclear atypia from a dog. A—Atypical melanocytes form nest-like structures, which are surrounded by connective tissue. B—Melanocytes appear atypical, as evidenced by frequent multiple nucleoli. C—Atypical melanocytes contain multiple nucleoli that are eccentrically located within the nucleus (arrows). D—Mixture of typical (vertical arrows) and atypical (horizontal arrows) melanocytes. The typical melanocytes possess a small nucleus, single centrally orientated nucleolus, and minimal clumping of chromatin. H&E stain; bars = 100 μm (A and B) and 50 μm (C and D).

  • 1. Uzal FA, Plattner BL, Hostetter JM. Alimentary system. In: Maxie G, ed. Jubb, Kennedy & Palmer's pathology of domestic animals. 6th ed. St Louis: Elsevier Health Sciences, 2015;2627.

    • Search Google Scholar
    • Export Citation
  • 2. Ramos-Vara JA, Beissenherz ME, Miller MA, et al. Retrospective study of 338 canine oral melanomas with clinical, histologic, and immunohistochemical review of 129 cases. Vet Pathol 2000;37:597608.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 3. Modiano JF, Ritt MG, Wojcieszyn J. The molecular basis of canine melanoma: pathogenesis and trends in diagnosis and therapy. J Vet Intern Med 1999;13:163174.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 4. Nishiya AT, Massoco CO, Felizzola CR, et al. Comparative aspects of canine melanoma. Vet Sci 2016;3:710.

  • 5. Hahn KA, DeNicola DB, Richardson RC, et al. Canine oral malignant melanoma: prognostic utility of an alternative staging system. J Small Anim Pract 1994;35:251256.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6. Bergman PJ. Canine oral melanoma. Clin Tech Small Anim Pract 2007;22:5560.

  • 7. Smith SH, Goldschmidt MH, McManus PM. A comparative review of melanocytic neoplasms. Vet Pathol 2002;39:651678.

  • 8. Smedley RC, Spangler WL, Esplin DG, et al. Prognostic markers for canine melanocytic neoplasms: a comparative review of the literature and goals for future investigation. Vet Pathol 2011;48:5472.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9. Spangler WL, Kass PH. The histologic and epidemiologic bases for prognostic considerations in canine melanocytic neoplasia. Vet Pathol 2006;43:136149.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 10. Bergin IL, Smedley RC, Esplin DG, et al. Prognostic evaluation of Ki67 threshold value in canine oral melanoma. Vet Pathol 2011;48:4153.

  • 11. Fluss R, Faraggi D, Reiser B. Estimation of the Youden index and its associated cutoff point. Biom J 2005;47:458472.

  • 12. Bergman PJ, Kent MS, Farese JP. Melanoma. In: Withrow SJ, Vail DM, Page RL, eds. Withrow and MacEwen's small animal clinical oncology. 5th ed. St Louis: Elsevier Health Sciences, 2012;321334.

    • Search Google Scholar
    • Export Citation
  • 13. Dennis MM, Ehrhart N, Duncan CG, et al. Frequency of and risk factors associated with lingual lesions in dogs: 1,196 cases (1995–2004). J Am Vet Med Assoc 2006;228:15331537.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 14. Todoroff RJ, Brodey RS. Oral and pharyngeal neoplasia in the dog: a retrospective survey of 361 cases. J Am Vet Med Assoc 1979;175:567571.

    • Search Google Scholar
    • Export Citation
  • 15. Goldschmidt MH. Histological classification of epithelial and melanocytic tumors of the skin of domestic animals. Washington, DC: Armed Forces Institute of Pathology, 1998;331340.

    • Search Google Scholar
    • Export Citation
  • 16. Millanta F, Fratini F, Corazza M, et al. Proliferation activity in oral and cutaneous canine melanocytic tumours: correlation with histological parameters, location, and clinical behaviour. Res Vet Sci 2002;73:4551.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 17. Behjatnia B, Sim J, Bassett LW, et al. Does size matter? Comparison study between MRI, gross, and microscopic tumor sizes in breast cancer in lumpectomy specimens. Int J Clin Exp Pathol 2010;3:303309.

    • Search Google Scholar
    • Export Citation
  • 18. Fitzgibbons PL, Page DL, Weaver D, et al. Prognostic factors in breast cancer. Arch Pathol Lab Med 2000;124:966978.

  • 19. Verma VP, Kaur N, Agarwal N, et al. Intra-operative measurement of tumour size in breast cancer and its comparison with other methods: a prospective study. Ecancermedicalscience 2008;2:96.

    • Search Google Scholar
    • Export Citation
  • 20. Johnson RE, Sigman JD, Funk GF, et al. Quantification of surgical margin shrinkage in the oral cavity. Head Neck 1997;19:281286.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 21. Mohiyuddin SMA, Padiyar BV, Suresh TN, et al. Clinicopathological study of surgical margins in squamous cell carcinoma of buccal mucosa. World J Otorhinolaryngol Head Neck Surg 2016;2:1721.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 22. Siu KF, Cheung HC, Wong J. Shrinkage of the esophagus after resection for carcinoma. Ann Surg 1986;203:173176.

  • 23. Jonmarker S, Valdman A, Lindberg A, et al. Tissue shrinkage after fixation with formalin injection of prostatectomy specimens. Virchows Arch 2006;449:297301.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 24. Tran T, Sundaram CP, Bahler CD, et al. Correcting the shrinkage effects of formalin fixation and tissue processing for renal tumors: toward standardization of pathological reporting of tumor size. J Cancer 2015;6:759766.

    • Crossref
    • Search Google Scholar
    • Export Citation

Advertisement