Christiane Verena Loehr, Jens Peter Teifke, Eugen Weiss
Institut fuer Veterinaer-Pathologie, Justus-Liebig-Universitaet
Giessen, Germany
Address for correspondence:
Christiane V. Loehr, D.V.M.
Institut fuer Veterinaer-Pathologie
Justus Liebig Universitaet
Frankfurter Str. 96
D-35392 Giessen, Germany
Tel: 0049-641-9938200
Fax: 0049-641-9938209
E-mail: Jens.Teifke@vetmed.uni-giessen.de
ABSTRACT
The analysis of proteins associated with nucleolus organizer
regions (NOR) by a silver staining (AgNOR method) is proposed as a
marker of cellular proliferation and an additional procedure for
the prognostic evaluation of human tumors. For specific staining
of these acid non-histone proteins the AgNOR-method developed by
Ploton and coworkers [9] and standardized by the AgNOR committee
is used [1,5].
In this study 69 canine mammary tumors were analyzed and stained
for AgNORs following the standardized protocol by image analysis
(KS400, Zeiss-Kontron; objective 40). The evaluation of 8 AgNOR
parameters and the follow-up data over at least 12 months in a
multivariate analysis revealed, that the quantification of AgNOR-
parameters is of some value in the estimation of the prognosis of
dogs with mammary tumors. An elevated AgNOR cluster number, AgNOR
area and/or nucleus area are statistically significant risk
factors for the survival and recurrence-free interval. However,
there is a remarkable overlap for the majority of AgNOR-parameters
comparing benigne and malignant tumors of the same histological
type.
KEYWORDS: Dog, mammary tumors, multivariate analysis, prognosis,
standardized AgNOR method
INTRODUCTION
An increasing number of neoplasms occurring in domestic animals is
submitted to histopathological diagnosis due to a growing interest
of patient owners about the prognosis of their animal. Therefore
follow up studies in combination with additional parameters in
order to increase the understanding about the biological behaviour
of tumors are inevitable in veterinary medicine. Mammary tumors
represent with a percentage of 42.9 the most frequently diagnosed
tumors of the dog in bioptic material investigated at the Institut
fuer Veterinaer-Pathologie, Giessen [3]. The variety of canine
mammary tumors according the classification system recommended by
the World Health Organization (WHO) complicates the estimation of
their biological behaviour. This fact emphasized even more the
investigation of additional tumor characteristics like
proliferation patterns [7]. An inexpensive, easy and
retrospectively performable procedure to study the proliferative
activity of tumors is the AgNOR method. This silver staining
visualizes acid non-histone proteins, which are associated with
the ribosomal DNA loops encoding the ribosomal RNA [8]. Their
quantity is therefore believed to be directly correlated to the
proliferative activity of cells [3]. The major draw back of the
AgNOR-method, the lack of a standardized protocol, is resolved
[1,5].
MATERIAL AND METHODS
We selected for this study 69 tissue samples obtained from bitches
which underwent surgery for mammary gland neoplasias at the
Ambulatorische und Geburtshilfliche Veterinaerklinik, Giessen with
a follow up data of at least 12 months. Histopathological
examination was performed according to the classification
recommended by the WHO.
Silver staining of AgNORs was performed following the protocol of
the standardized AgNOR method [1,5]. Briefly after dewaxing and
rehydration of routinely processed formalin-fixed tissue sections
the incubation was carried out on the slides for 17 to 20 minutes
in a moist chamber. The staining solution, containing one part 2 %
gelatine in 1 % formic acid and two parts of 50 % silver nitrate
solution, was freshly prepared. For a prolonged durability of
AgNORs and an increased contrast the slides were postfixed for
5 minutes in 5 % sodium thiosulfate solution [4].
Assessment of AgNORs was performed using the image analysis system
KS400 and a 40x objective [1]. AgNOR cluster number and AgNOR
area/cell were automatically, the nucleus area interactively
analyzed, the ratio (AgNOR area/cell:nucleus area) and the CV
values of these parameters calculated.
The exploratory data analysis was performed as a multivariate
analysis to estimate the relevance of the investigated parameters
for the diagnostic procedure and in predicting the prognosis of
bitches with mammary tumors. A multiple stepwise linear regression
was carried out in the case of the interval scaled dependend
variables survival time, time until a local tumor reappearance,
histopathological tumor groups and grades of malignancy (interval
scaled, BMDPLR(R)). A multiple logistic regression was done for
the categorial dependend variables survival rate, rate of
reappearing tumor growth, lymph node status and grade of
malignancy (dichotom, BMDP2R(R)). As independend parameters were
investigated the AgNOR parameters, tumor volume, involvement of
epithelial and myoepithelial cells, grade of malignancy and lymph
node status.
RESULTS
Staining results revealed with the standardized AgNOR-method with
postfixation were of excellent quality (see 971-111.pcx). The mean
AgNOR area/cell was never higher than 2 um,showing only a moderate
mean variation (Table 1). The mean CV values differed considerably
between but not within the AgNOR parameters.
Two calculations were performed. The first calculation was done
only with the AgNOR parameters, the second calculation included
further tumor parameters as listed above.
Considering only AgNOR parameters the survival time and the time
until reappearance were well described by the total AgNOR
area/cell. The histologically estimated tumor groups and
biological behaviour correlated with the mean AgNOR number/cell or
its CV. To calculate the survival, the rate of reappearing tumor
growth and the lymph node status, it was sufficient to take into
account the nucleus area.
Running the model with all parameters the survival time and the
time until recurrence had to be excluded due to the low number of
cases.
The tumor volume proved to be in the cases of the biological
behaviour and the tumor groups the only required parameter. In the
description of the survival could be noted the expected shift to
the lymph node status and tumor volume, but interestingly the
AgNOR number/cell was included too. The recurrency rate correlated
only at a low significance, but exclusively with the CV value of
the AgNOR ratio. No changes occurred in regard to the lymph node
status.
DISCUSSION
Due to the small number of 69 tumor cases investigated until now,
the results have to be understood as a trend analysis. But still
there could be shown a remarkable association of AgNOR parameters
and the prognosis of dogs which underwent surgery for mammary
gland tumors. AgNOR parameters have been able to compete with well
established parameters as the tumor volume and the lymph node
status, though being inferior. The CV values of AgNOR parameters
have not been of crucial importance in predicting the prognosis in
canine mammary gland tumors in contrast to oberservations
concerning breast neoplasias in women [5].
ACKNOWLEDGEMENT
We would like to thank Dr. K. Failing for his support in running
the statistics. These investigations have been financially
supported by "Ewald und Hilde Berge Stiftung", Giessen.
REFERENCES
1. Aubele M, Biesterfeld S, Derenzini M, Hufnagl P, Martin H,
Oefner D, Ploton D, Rueschoff J (1994) Guidelines of AgNOR
quantitation. Zentrbl Pathol 140:107-108
2. Derenzini M, Pession A, Trere D (1990) The quantity of
nucleolar silver stained proteins is related to proliferating
activity in cancer cells. Lab Invest 63:137-140
3. Eskens U (1993) Statistische Untersuchungen ueber nach den
Empfehlungen der Weltgesundheitsorganisation (WHO)
klassifizierten Tumoren des Hundes unter besonderer
Beruecksichtigung der Mamma- und Hauttumoren. Vet Med Diss, Giessen
4. Loehr CV, Driemeier D, Teifke JP (1994) Die Polyethylenglykol-
Thiosulfat (PEG-Th)-F"rbung - eine Modifikation der AgNOR-Methode.
Zentrbl Pathol 140:29-32
5. Oefner D, Aubele M, Biesterfeld S, Derenzini M, Gimenez-Mas JA,
Hufnagl P, Ploton D, Trere D, Rueschoff J (1995) Guidelines of
AgNOR quantitation - first update. In: Hofstaedter F, Knuechel R,
Rueschoff J, Cell proliferation assessment in oncology. Virchows
Archiv 427:340-341
6. Oefner D, Rueschoff J, Trere D, Aubele M, Derenzini M, Schmid KW
(1995) Standardized morphometric silver stained nucleolus
organizer region associated protein (AgNORs) analysis - basic
concepts. Path Res Pract 191:236
7. O'Reilly SM, Richards MA (1992) Clinical aspects of assessing
cell proliferation. In: Hall PA, Levison DA, Wright NA (eds.)
Assessment of cell proliferation in clinical practice, Springer,
London
8. Pardue ML, Hsu TC (1975) Locations of 18 S and 28 S ribosomal
genes on the chromosomes of the Indian mintjac. J Cell Biol
64:251-254
9. Ploton D, Menager M, Jeannesson P, Himber G, Pigeon F, Adnet JJ
(1986) Improvement in the staining and in the visualization of
argyrophilic proteins of the nucleolar organizer region at the
optical level. Histochem J 18:5-14
Table 1 Results of morphometry on AgNORs of canine mammary tumors
given by the mean and standard deviation
----------------------------------------------------------------
parameter nucleus AgNOR AgNOR AgNOR n
tumor groups area number/cell area/cell ratio 69
um2 um2 um2/um2
________________________________________________________________
hyperplasia 15,51 1,33 0,69 0,05 4
+/-1,4 +/-0,22 +/-0,12 +/-0,01
simple adenoma 18,4 2,76 0,26 1,06 6
+/-2,1 +/-0,14 +/-0,27 +/-0,01
simple carcinoma 22,4 3,54 0,36 1,21 4
+/-3,01 +/-0,13 +/-0,93 +/-0,04
solid carcinoma 22,48 1,67 1,11 0,05 8
+/-7,22 +/-0,37 +/-0,49 +/-0,01
inv/anapl. carcinoma 32,8 4,1 0,49 1,52 7
+/-5,85 +/-0,41 +/-0,41 +/-0,01
compl. adenoma 20,6 2,28 0,21 0,85 13
+/-3,54 +/-0,09 +/-0,39 +/-0,01
compl. carcinoma 22,7 3,06 0,33 1,15 9
+/-4,04 +/-0,26 +/-0,54 +/-0,02
malig.susp. adenoma 22,36 1,36 1,00 0,05 6
+/-3,35 +/-0,18 +/-0,17 +/-0,004
benign mixed tumor 16,2 2,38 0,26 1,45 6
+/-1,53 +/-0,08 +/-0,06 +/-0,01
malig. mixed tumor 21,4 2,98 0,34 1,26 6
+/-3,63 +/-0,13 +/-0,11 +/-0,01
----------------------------------------------------------------
AgNOR ratio = (AgNOR area/cell) / nuclear area