Sage Journals: Discover world-class research

Abstract

Twenty-four canine cutaneous nodules, diagnosed as mast cell tumors by fine-needle aspiration biopsy and confirmed by histopathologic analysis by staining with hematoxylin and eosin (HE) and toluidine blue, were analyzed by computerized nuclear morphometry on panoptic- and HE-stained cytopathology slides. Two hundred nuclei per lesion were examined. The morphometric parameters investigated were nuclear area, mean diameter, perimeter, regularity factor, and ellipticity factor. Lesions were graded as I (well differentiated), II (intermediate differentiation), or III (poorly differentiated) according to the following morphologic features: invasiveness, cellularity and cellular morphology, mitotic index, and stromal reaction. Nuclear morphometric results were then compared with histopathologic grades. Values of nuclear area, mean diameter, and perimeter increased with increase in histopathologic grade, but statistical analysis revealed significant differences only between grades II and III and between grades I and III when HE was used (P, 0.01) and between grades I and III with panoptic stain (P, 0.05). The ellipticity factor and regularity factor did not reveal significant differences between histopathologic grades. The results indicate that nuclear morphometric analysis, in combination with the rapid and inexpensive cytopathology technique, can help in mast cell tumor grading, thus contributing to the establishment of a more precise prognosis and treatment.

Keywords

Cytology cytopathology dogs histopathologic grading histopathology mast cell mast cell tumor morphometry

Mast cell tumors are potentially malignant neoplastic processes belonging to the group of round cell tumors, which correspond to 7–21% of the skin neoplasias affecting dogs, with no sexual predisposition.²⁶,³¹ The breeds considered to be most susceptible are Boxer, Boston and Bull terriers, and Labrador Retriever, and the mean age of affected dogs is 8.5 years.²,¹⁴,²¹,²⁴,²⁶

Several histopathologic classifications have been proposed to identify the degree of differentiation of canine mast cell tumors.²,¹⁴,²⁴ Among them, the classification proposed by Patnaik et al.²⁴ is considered to be the most complete and is the one most frequently used.¹¹,¹⁶,¹⁸,²¹,²³,²⁶,²⁸,³⁰ However, histopathologic grading is based on the use of subjective parameters such as invasiveness, cellularity and cellular morphology, and mitotic index and often proves to be inconsistent for the evaluation of borderline cases.²⁶,³⁰,³² Although histopathologic grading is considered to be a good method for the prediction of long-term tumor behavior, reproducibility problems lead to as much as 50–60% discordance between experienced pathologists.¹,¹¹,¹⁷,²²,²⁹,³⁰

In addition to mast cell tumor classification by traditional histopathologic criteria, other methods have been tested as potential prognostic markers in an attempt to improve the prediction of the behavior of this neoplasia in terms of recurrence, metastasis, and animal survival, including cell proliferation markers (AgNORs, PCNA, and Ki-67) and markers of the expression of tumor suppressor genes such as p53.¹,³,¹¹,¹⁶,¹⁸,³⁰

One of the solutions found over the last few decades to minimize subjective interobserver variations and consequently aid diagnostic decisions was the development of computerized morphometric techniques.⁵,⁷,⁸,¹²,¹⁵,¹⁷,²² An additional relevant method is cytopathologic examination, which is an efficient, practical, and rapid technique for the diagnosis of mast cell tumors.⁶,⁷,¹³,¹⁹ Using cytopathology is also advantageous in that the cell is spread on the slide so that its image is a complete projection of its cytoplasm, thereby eliminating the tridimensional problems of the various sections of the same cell encountered in histopathology.¹⁰

The objective of the present study was to determine whether mast cell tumors of different histopathologic grades show different nuclear morphometric patterns by combining the precision of computerized morphometric analysis with the rapidity and practicality of cytopathologic examination.

Materials and Methods

Tumors

The study was conducted on 24 skin mast cell tumors from 14 dogs obtained from the Department of Surgery, Faculdade de Medicina Veterinária e Zootecnia, Universidade de São Paulo (FMVZ-USP), from February 1999 to October 2000. The diagnosis was made by fine-needle aspiration biopsy (FNAB) and was confirmed by histopathologic analysis. Seven dogs were males (50%) and seven were females (50%). The most frequent breed was Boxer (35.7%, 5/14), followed by Schnauzer and mongrel dogs, with the numbers of both being the same (14.2%, 2/14). The remaining dogs were a Pinscher, a Poodle, a Brazilian Fila, a Dachshund, and a Doberman Pinscher. Age was determined for 13 dogs, with the mean age being 8.5 years (range, 4–13 years). Six dogs (42.8%) had more than one mast cell tumor, and the remaining dogs had a single lesion (57.2%, 8/14) each.

Cytologic and histopathologic processing

Two cytopathologic slides obtained by FNAB were air-dried and stained by the panoptic method (HEMATOCOR®, Biolog®, Biológica Comercial Ltda., São Paulo, SP, Brazil), and two additional slides obtained by scraping the cut surface of the surgical piece were fixed in absolute alcohol for 24 hours and stained with hematoxylin and eosin (HE). The material to be used for histopathology was fixed in 10% formaldehyde and was routinely processed. Two 5-μm sections were obtained from each tumor and were stained with HE and toluidine blue, respectively.²⁵

Histopathologic grading

The 24 tumors were classified as grades I to III according to the level of cellular differentiation on the basis of the histopathologic grading criteria of Patnaik et al.²⁴ Three observers, who were experienced anatomic pathologists and had no preliminary clinical information about the tumors, performed this grading independently. The histopathologic grade established by the consensus of two or more observers was considered to be the final diagnosis.

Nuclear morphometric analysis

The material obtained by cytopathologic processing was analyzed with an Olympus® BX 50 Trinocular microscope (Melville, NY) coupled to a video camera system (SONY®, CCD-IRIS model, San Jose, CA) and to a computer equipped with the Image ProPlus® analysis system (Media Cybernetics, Silver Spring, MD, version 3.0.01.00 for Windows 95/NT). The computer used was equipped with a 550-MHz Pentium III® Intel® processor with 128 Megabytes of RAM and a 17-in. monitor. The images created by the computer system were formatted as tif files. The measurements were calibrated with the aid of a micrometer ruler (Olympus®) and a 100× magnification oil-immersion objective.

For each tumor, 200 nuclei (100 stained by the panoptic method and 100 with HE) of cells presenting sharp nuclear and cytoplasmic limits that did not overlap were contoured by tracing their margins with the aid of a mouse. Because of the quality of slides or number of slides available (or both), of the 24 tumors, 23 tumors stained with HE (tumor Nos. 1–22 and 24) and 16 tumors stained by the panoptic method (tumor Nos. 1, 4, 5, 7, 10–12, 14–16, 18–20, and 22–24) were available for nuclear morphometric analysis on cytopathology slides. For the tumors stained with HE, we used 700 cells from grade I lesions, 1,000 from grade II lesions, and 600 from grade III lesions (7, 10, and 6 tumors, respectively); for the tumors stained by the panoptic method, we used 500 cells from grade I lesions, 600 from grade II lesions, and 500 from grade III lesions (5, 6, and 5 tumors, respectively).

The parameters available in the tools offered by the Image ProPlus® software and used in the present study were as follows: Area, equivalent to the area of the object; ellipticity factor, equivalent to the major/minor axis ratio of an ellipse referring to the object; average diameter, equivalent to the mean measurement of the diameters measured at 5° intervals around the center of the object; perimeter, equivalent to the measurement of the perimeter of the object; and regularity factor, representing the regularity of the contour of the object according to the formula (perimeter²)/(4 × π × area).

Statistical analysis

The results obtained by computerized cytomorphometry were compared and correlated with the grades obtained by histopathologic examination. Data were analyzed for normality by the Kolmogorov-Smirnov test using the GraphPad InStat® software (version 3.01 for Windows 95, GraphPad Software, San Diego, CA, www.graphpad.com), with the level of significance set at P < 0.05. Because the data did not show normal distribution, comparison of histologic grades was performed by one-way analysis of variance (ANOVA) followed by the Tukey-Kramer multiple comparisons test.

Results

Histopathologic grading

The histopathologic grades for each tumor, distributed according to the observer, are listed in Table 1. Seven tumors (29.2%) confined to the dermis and interfollicular spaces, which contained well-differentiated mast cells, arranged in rows or small groups, and showed no mitotic figures, were classified as grade I (Fig. 1); 11 tumors (45.8%) that were moderately to highly cellular and infiltrated the dermis and subcutaneous tissue, were formed by moderately differentiated mast cells with intracytoplasmic granules of variable morphology, and showed rare mitotic figures were classified as grade II (Fig. 3); 6 neoplasms (25%), which replaced the dermis and subcutaneous tissue and extended to adjacent tissues, had pleomorphic cells with indistinct cytoplasms arranged in packed sheets, and showed mitotic figures more commonly, were classified as grade III (Fig. 5). The number of eosinophils was different between tumors of the same histopathologic grade. There was agreement between the diagnoses made by the three pathologists on 12 occasions (50%), with at least two observers grading the tumor similarly in all cases.

Table 1.

Distribution of the mast cell tumors analyzed according to the classification proposed by Patnaik et al.24 as a function of the observers consulted (São Paulo, 1999–2000).


		Observer
Tumor No.	Veterinary Hospital No.	1	2	3	Final Diagnosis

1	52875-3	II	II	II	II
2	100118-4	II	II	II	II
3	93715	I	II	II	II
4	87119-3	I	II	II	II
5	97565-1	I	I	I	I
6	94868-3	I	I	II	I
7	41770	I	II	II	II
8	100118-3	II	II	II	II
9	94868-2	I	I	II	I
10	93906-2	I	I	I	I
11	93131-4	III	III	III	III
12	85729	II	III	III	III
13	100118-2	II	II	II	II
14	93131-3	II	III	III	III
15	98104-1	II	I	I	I
16	93131	III	II	III	III
17	100118-1	II	II	II	II
18	94516	I	II	II	II
19	90233-2	I	II	II	II
20	52875-1	III	III	III	III
21	93131-2	III	III	III	III
22	93906-1	I	I	I	I
23	90233-1	I	II	II	II
24	79801	I	I	I	I

Fig. 1.

Skin, Brazilian Fila, tumor No. 24. Grade I mast cell tumor. Well-differentiated mast cells, with distinct and granular cytoplasm (arrowhead), arranged in rows. Uniform and condensed chromatin, with no mitotic cells. HE. Bar = 15 μm.

Fig. 2.

Skin, Boxer, tumor No. 10. Grade I mast cell tumor. Note the well-defined and uniform nuclei and cytoplasm of well-differentiated mast cells. HE. Bar = 5.5 μm.

Fig. 3.

Skin, Boxer, tumor No. 19. Grade II mast cell tumor. Moderately pleomorphic cells, most with distinct cytoplasm, arranged in groups. Round to oval nuclei, with scattered chromatin and single nucleoli. HE. Bar = 15 μm.

Fig. 4.

Skin, Boxer, tumor No. 1. Grade II mast cell tumor. Moderately pleomorphic cells with scattered chromatin. HE. Bar = 5.5 μm.

Fig. 5.

Skin, Doberman Pinscher, tumor No. 11. Grade III mast cell tumor. Highly pleomorphic mast cells with indented irregular nuclei in dense sheets. Some cells show multiple nucleoli and mitotic figures (arrowhead). HE. Bar = 15 μm.

Fig. 6.

Skin, Doberman Pinscher, tumor No. 16. Grade III mast cell tumor. Highly pleomorphic mast cells. HE. Bar = 5.5 μm.

Nuclear morphometric analysis

Nuclear morphometric results were then compared with histopathologic grades. Values of nuclear area, mean diameter, and perimeter increased with increase in histopathologic grade, but statistical analysis showed significant differences only between grades II and III and between grades I and III with HE (P < 0.01) and between grades I and III when the panoptic stain was used (P < 0.05). The ellipticity and regularity factors did not reveal statistically significant differences between histopathologic grades. Tables 2 and 3 present the karyocytometric data for the parameters analyzed as a function of histopathologic grading using HE and panoptic staining, respectively. Figures 1–6 display the major morphologic features observed for the various grades.

Table 2.

Distribution of mean, standard deviations, and range of the data concerning the nuclear morphometric parameters obtained for cytopathologic material stained with HE, as a function of histopathologic grading (São Paulo, 1999–2000).


Grade	Area (μm²)	Perimeter (μm)	Average Diameter (μm)	Ellipticity Factor	Regularity Factor

I (n = 7)∗
Mean ± SD	47.25 ± 3.76†	25.31 ± 1.02†	7.55 ± 0.30†	1.23 ± 0.07	1.09 ± 0.02
Minimum	43.35	24.24	7.23	1.16	1.06
Median	45.13	24.76	7.44	1.21	1.09
Maximum	52.43	26.71	7.95	1.37	1.13
II (n = 10)∗
Mean ± SD	49.98 ± 5.04†	25.88 ± 1.36†	7.78 ± 0.38†	1.19 ± 0.06	1.08 ± 0.01
Minimum	42.86	23.93	7.19	1.13	1.06
Median	50.00	25.92	7.80	1.17	1.08
Maximum	59.22	28.07	8.40	1.36	1.11
III (n = 6)∗
Mean ± SD	62.65 ± 12.03	28.87 ± 2.61	8.71 ± 0.79	1.16 ± 0.01	1.0-7 ± 0.02
Minimum	52.29	26.29	8.00	1.14	1.05
Median	58.37	28.14	8.43	1.17	1.07
Maximum	85.67	33.70	10.21	1.19	1.11

∗ n = Number of tumors analyzed.

^† Statistically significant difference (P, 0.01) in relation to grade III—one-way ANOVA/Tukey-Kramer test.

Table 3.

Distribution of mean, standard deviations, and range of the data concerning the nuclear morphometric parameters obtained for cytopathologic material stained with panoptic stain, as a function of histopathologic grading (São Paulo, 1999–2000).


Grade	Area (μm²)	Perimeter (μm)	Average Diameter (μm)	Ellipticity Factor	Regularity Factor

I (n = 5)∗
Mean ± SD	64.35 ± 7.35 †	29.07 ± 1.79 †	8.80 ± 0.51 †	1.25 ± 0.03	1.06 ± 0.01
Minimum	55.02	26.72	8.13	1.21	1.05
Median	64.87	28.87	8.80	1.25	1.05
Maximum	74.91	31.69	9.53	1.29	1.08
II (n = 6)∗
Mean ± SD	69.26 ± 6.44	30.33 ± 1.63	9.14 ± 0.36	1.26 ± 0.07	1.07 ± 0.03
Minimum	62.58	28.57	8.72	1.20	1.04
Median	69.04	30.10	9.18	1.24	1.06
Maximum	80.03	33.15	9.69	1.41	1.13
III (n = 5)∗
Mean ± SD	74.33 ± 1.43	31.62 ± 0.19	9.50 ± 0.05	1.22 ± 0.05	1.08 ± 0.01
Minimum	72.95	31.30	9.44	1.18	1.06
Median	74.16	31.71	9.52	1.21	1.08
Maximum	76.25	31.76	9.57	1.32	1.10

∗ n = number of tumors analyzed.

^† Statistically significant difference (P, 0.05) in relation to grade III—one-way ANOVA/Tukey-Kramer test.

Discussion

The clinical data obtained in the present study confirm literature reports about a higher incidence of mast cell tumors in the Boxer race and a mean age of involvement of 8.5 years.¹,²,¹¹,¹⁴,¹⁶,²¹,²³,²⁴,³² Multiple tumors were detected in six dogs (42.8%), in contrast to the 3–14% rate reported in the literature.²¹,²³,³⁰

Histopathologic grading

Considering the proposal that multiple mast cell tumors should be classified according to the lesion of highest grade,¹⁶ the distribution of histopathologic grading among the 14 animals studied was as follows: 35.7% (5/14) tumors of grade I, 42.9% (6/14) of grade II, and 21.4% (3/14) of grade III. The results appear to be similar to those obtained by Patnaik et al.²⁴

Although widely adopted, the histopathologic grading proposed by Patnaik et al.²⁴ presents some difficulties, including counts of mitotic figures per 40× magnification field. In the present study, some highly cellular tumors presented extremely anaplastic cells with few granules that invaded adjacent tissue, where areas of necrosis were also observed. These tumors were classified as grade III, even though they did not have three to six mitoses per field, as proposed by Patnaik et al.²⁴ Some investigators have stated that mitotic counts require greater standardization and therefore lose reproducibility because of factors such as the short duration of the M phase, presence of pyknotic nuclei, and even differences in field area in objectives from different manufacturers.¹,⁹,²⁷ Thus, we believe that, although reported with fixed values, the evaluation of mitotic index may not be a determinant for grading but should be considered only as an additional parameter of anaplasia, and the presence of mitosis counts to a greater or lesser extent should be taken into account in each case.

Another factor to be considered for a tumor classification system is interobserver variability.¹,¹¹,²¹,²³,²⁶,³⁰,³² In the present study, we decided to treat this problem by submitting the material for histopathologic grading to three experienced anatomic pathologists who worked independently, without any previous clinical information. The results obtained showed interobserver variations in 50% of the tumors (Table 1). Even considering that the basis for grading was the same, we may propose that this discrepancy in the results obtained was because of different observer experience or training. The other possibility could be the difficulty in evaluating borderline cases between two grades because of conflicting criteria such as large and pleomorphic nuclei, characteristic of grade II or III tumors, and small number of mitoses per field, characteristic of grade I or II tumors. The problems of interobserver reproducibility found by us support the observations reported by others, i.e., 50–60% discordance between experienced pathologists, and emphasize the importance of the definition of more precise criteria for the classification of this neoplasia.¹⁷,²²

Morphometric analysis

In the histopathologic descriptions used to differentiate mast cell tumors in terms of degree of malignancy, mast cell dimensions and shapes were mentioned. Quantitative measurements such as cell and nuclear diameters were reported by Hottendorf and Nielsen,¹⁴ whereas Bostock² preferred using the nucleus/cytoplasm ratio for characterizing cells of different grades. Nevertheless, the methods by which these results were obtained were not described in detail, and the values were imprecise, which prevented speculations about the morphometric approach used. In the classification used in the study by Patnaik et al.,²⁴ the descriptive text did not contain any objective/quantitative result in an attempt to establish cytomorphometric parameters, with the consequent subjectivity of traditional histopathology classifications.

The present results demonstrate that the area, mean diameter, and perimeter of the nucleus are the most appropriate nuclear morphometric parameters for differentiating between the histopathologic grades of canine skin mast cell tumors. These parameters revealed significant differences between grades II and III and between grades I and III with HE (P < 0.01). With the panoptic stain, we obtained less expressive results, showing significant differences only between grades I and III (P < 0.05). The increase observed in mean area, perimeter, and diameter confirmed the increase in mast cell nuclear size previously observed by others.²,¹⁴,²⁴ On the other hand, the higher values for area, perimeter, and mean diameter obtained by the panoptic method can be explained by the fixation process for the cytopathology slides. For panoptic staining, the smears are air-dried, which aids cell spreading, whereas for HE staining, fixation is performed by immersion in absolute alcohol for 24 hours, with a greater cell retraction resulting from the dehydrating action of the fixative.²⁰

In contrast, among the parameters concerning nuclear shape, the ellipticity factor and the regularity factor revealed no significant differences between grades for either stain. The mean value of the ellipticity factor decreased with increasing histopathologic grade, i.e., the cells had less elliptic nuclei in the more undifferentiated mast cell tumors. The idea of ellipticity, the major/minor axis ratio, is not present in the various descriptions used to classify the degree of malignancy. Thus, when we read “round” in most descriptions, we understand that what is being expressed is the characteristic of regularity of the contour and the presence or absence of indentations and not the extent of ellipticity of the nucleus.

The irregularity of the nuclear contour is better expressed morphometrically by the regularity factor. The grading reports available in the literature show that less differentiated tumors tend to show indentation, lobulations, or nuclear irregularities.²,¹⁴,²⁴ Regardless of the staining method used for measurement (HE or the panoptic method), the regularity factor did not correlate with tumor grade. Descriptions reported in the literature, based on observations on histologic sections, favor the results obtained with panoptic staining in which the values of the regularity factor increase with tumor differentiation.

The differences in nuclear shape between cytopathology and histopathology, despite using the same staining method (HE), may be explained by the characteristics of fixation and processing of the material.⁴ In histopathology, the tissue undergoes countless chemical reactions capable of deforming both the nuclear and cytoplasmic morphology, and cell overlap also occurs. In contrast, the cell density is lower and the chemical influences are fewer in cytopathology.

Therefore, the results obtained in our study regarding the regularity and ellipticity factors corroborated the data obtained by Marchevsky and Erler,²² who stated that although these parameters do indicate significant irregularities/differences in shape parameters, they are not good indicators when there are small deviations from circularity, a fact that was observed for mast cells.

We did not determine the nucleus/cytoplasm ratio or cytoplasmic measurements because of the difficulty in delimiting the cellular margins with both cytopathology stains. When stained with HE, the cells presented ill-defined contours in contrast to the image background, and rapid panoptic staining caused the stained cytoplasmic granules to be a great obstacle in the delimitation of the cytoplasmic margin. We believe that the manual contour of these cells depends a lot on the ability of the operator, a fact that may increase the variability among users of image analysis programs, which makes the procedure more subjective and, therefore, goes against one of the major objectives of the study.

In conclusion, expressive results were obtained for the different histopathologic grades in terms of parameters such as area, perimeter, and mean diameter. Because grading affects prognosis and the need for ancillary therapy (such as surgery alone or the use of chemotherapy or radiation therapy²³), the necessary therapeutic measures could be taken in a more effective manner by using cytopathology and morphometric analysis. However, when comparing the values obtained for each group for the parameters considered, we still observe overlap between them, a fact that may not be an obstacle after the compilation of more complete data.

Considered separately, the results do not permit, as of now, a precise classification of mast cell tumors in terms of prognosis because the data concerning the evolution of each tumor have not yet been analyzed. Further studies are under way in our laboratory to assess morphometry as a prognostic method. We trust that, after the conclusion of these studies, the morphometric differences between grades may permit us to establish “fixed” minimum and maximum values for each parameter investigated in order to classify mast cell tumors in a rapid and precise manner in terms of degree of differentiation, which makes this a method of great interest to the clinician/surgeon and to the animal's owner.

Footnotes

Acknowledgements

We thank Dr. Júlia Maria Matera, Departamento de Cirurgia, FMVZ-USP, for submitting surgical specimens; Dr. Omar Miguel for his assistance with the statistical analysis; and Fundação de Amparo à Pesquisa do Estado de São Paulo for financial support (grants nos. 97/09459-7 and 96/4903-8).

This work is part of the MSc dissertation presented by Ricardo De Francisco Strefezzi to the Departamento de Patologia, Faculdade de Medicina Veterinária e Zootecnia, Universidade de São Paulo.

References

1 Abadie

Amardeilh

Delvedier

: Immunohistochemical detection of proliferating cell nuclear antigen and Ki-67 in mast cell tumors from dogs. J Am Vet Med Assoc 215: 1629–1634, 1999

2 Bostock

: The prognosis following surgical removal of mastocytomas in dogs. J Small Anim Pract 14: 27–40, 1973

3 Bostock

Crocker

Harris

Smith

: Nucleolar organizer regions as indicators of post surgical prognosis in canine spontaneous mast cell tumors. Br J Cancer 59: 915–918, 1989

4 Dardick

Caldwell

McCaughey

WTE

Al-Jabi

: Nuclear morphologic and morphometric analyses of large noncleaved cell and immunoblastic non-Hodgkin's lymphomas. Hum Pathol 15: 965–972, 1984

5 Dardick

Payne

: Rationale and potential for image analysis in histopathology: an overview. Ultrastruct Pathol 16: 97–98, 1992

6 Duncan

Prasse

: Cytology of canine cutaneous round cell tumors. Vet Pathol 16: 673–679, 1979

7 Eich

Whitehair

Moroff

Heeb

: The accuracy of intraoperative cytopathological diagnosis. J Am Anim Hosp Assoc 36: 16–18, 2000

8 Eldar

Sabo

Cohen

Misselevich

Abrahamson

Cohen

Kelner

Boss

: The value of histomorphometric nuclear parameters in the diagnosis of well differentiated follicular carcinomas and follicular adenomas of the thyroid gland. Histopathology 34: 453–461, 1999

9 Ellis

PSJ

Whitehead

: Mitosis counting-a need to reappraisal. Hum Pathol 12: 1, 3–4, 1981

10.

10 Gil

Barba

: Principles of stereology: computerized applications to anatomic pathology. In: Image Analysis: A Primer for Pathologists, ed. Marchevsky

and Bartels

, pp. 79–124. Raven Press, New York, NY, 1994

11.

11 Ginn

Fox

Brower

Gaskin

Kurzman

Kubilis

: Immunohistochemical detection of p53 tumor-suppressor protein is a poor indicator of prognosis for canine cutaneous mast cell tumors. Vet Pathol 37: 33–39, 2000

12.

12 Hamilton

Allen

: Morphometry in histopathology. J Pathol 175: 369–379, 1995

13.

13 Hart

Wurm

Kremmer

Mischke

: Lymphnoten-diagnostik mittels Feinnadelaspirationszytologie beim Hund. Kleintierpraxis 40: 869–876, 1995

14.

14 Hottendorf

Nielsen

: Pathologic survey of 300 extirpated canine mastocytomas. Zentralbl Veterinarmed [A] 14: 272–281, 1967

15.

15 Ikeguchi

Oka

Saito

Kondo

Tsujitani

Maeta

Kaibara

: Computerized nuclear morphometry: a new morphologic assessment for advanced gastric adenocarcinoma. Ann Surg 229: 55–61, 1999

16.

16 Jaffe

Hosgood

Taylor

Kerwin

Hedlung

Lopez

Davidson

Miller

Paranjepe

: Immunohistochemical and clinical evaluation of p53 in canine cutaneous mast cell tumors. Vet Pathol 37: 40–46, 2000

17.

17 Koss

: Diagnostic Cytology and its Histopathologic Bases, 4th ed., vol. 2. JB Lippincott, Philadelphia, PA, 1992

18.

18 Kravis

Vail

Kisseberth

Ogilvie

Volk

: Frequency of argyrophilic organizer regions in fineneedle aspirates and biopsy specimens from mast cell tumors in dogs. J Am Vet Med Assoc 209(8):1418–1420, 1996

19.

19 Larkin

: Veterinary cytology-fine needle aspiration of masses or swellings on animals. Ir Vet J 47: 65–73, 1994

20.

20 Lee

Tolmachoff

Marchevsky

: DNA content analysis (“ploidy”) by image analysis. In: Image Analysis: A Primer for Pathologists, ed. Marchevsky

and Bartels

, pp. 261–307, Raven Press, New York, NY, 1994

21.

21 Macy

: Canine mast cell tumors. Vet Clin North Am Small Anim Pract 15: 783–803, 1985

22.

22 Marchevsky

Erler

: Morphometry in pathology. In: Image Analysis: A Primer for Pathologists, ed. Marchevsky

and Bartels

, pp. 125–179, Raven Press, New York, NY, 1994

23.

23 O'Keefe

: Canine mast cell tumors. Vet Clin North Am Small Anim Pract 20: 1105–1115, 1990

24.

24 Patnaik

Ehler

Macewem

: Canine cutaneous mast cell tumor: morphologic grading and survival time in 83 dogs. Vet Pathol 21: 469–474, 1984

25.

25 Prophet

Mills

Arrington

Sobin

: Laboratory Methods in Histotechnology. Armed Forces Institute of Pathology, Washington, DC, 1992

26.

26 Pulley

Stannard

: Skin and soft tissues. In: Tumors in Domestic Animals, ed. Moulton

, 3rd ed., pp. 38–44. University of California Press, Los Angeles, CA, 1990

27.

27 Quinn

Wright

: The clinical assessment of proliferation and growth in human tumors: evaluation of methods and applications as prognostic variables. J Pathol 160: 93–102, 1990

28.

28 Rogers

: Common questions about diagnosing and treating canine mast cell tumors. Vet Med 88(3):246–250, 1993

29.

29 Séguin

Leibman

Bregazzi

Ogilvie

Powers

Dernell

Fettman

Withdrow

: Clinical outcome of dogs with grade-II mast cell tumors treated with surgery alone: 55 cases (1996-1999). J Am Vet Med Assoc 218: 1120–1123, 2001

30.

30 Simoes

JPC

Schoning

Butine

: Prognosis of canine mast cell tumors: a comparison of three methods. Vet Pathol 31: 637–647, 1994

31.

31 Tyler

Cowell

Baldwin

Morton

: Introduction. In: Diagnostic Cytology and Hematology of the Dog and Cat, ed. Cowell

and Tyler

, 2nd ed., pp. 1–19. Mosby Inc., St. Louis, MO, 1999

32.

32 Yager

Scott

: The skin and appendages. In: Pathology of Domestic Animals, ed. Jubb

KVF

Kennedy

Palmer

, 4th ed., vol. 1, pp. 531–738. Academic Press, San Diego, CA, 1993

Morphometry of Canine Cutaneous Mast Cell Tumors

Abstract

Keywords

Materials and Methods

Tumors

Cytologic and histopathologic processing

Histopathologic grading

Nuclear morphometric analysis

Statistical analysis

Results

Histopathologic grading

Nuclear morphometric analysis

Discussion

Histopathologic grading

Morphometric analysis

Footnotes

Acknowledgements

References