[CANCER RESEARCH 37, 2553-2556, August 1977]
Epidemiological Features of Canine Renal Neoplasms
Howard M. Hayes, Jr., and Joseph F. Fraumeni,
Jr.
EnvironmentalEpidemiologyBranch, NationalCancerInstitute, NIH. Bethesda,Maryland20014
SUMMARY
A retrospective study of medical records from 13 veteni
nary university
hospitals
and clinics
yielded
73 cases of
confirmed primary neoplasms of the kidney and renal pel
vis. Analysis of the 34 cases of renal carcinoma revealed no
evidence of familial (breed) predisposition, but an excess
risk was identified in males, especially during middle life.
The morphological and epidemiological features of canine
renal carcinoma are similar to those of renal carcinoma in
humans. The dog may be a suitable model for further me
search into the causes of this cancer.
INTRODUCTION
In humans, renal cancer accounts for 2% of all malignant
neoplasms (12), but little is known about etiology. Since the
urinary system in dogs is histologically and functionally
similar to that in humans and responds in a like fashion to
known carcinogenic agents (8), it seems appropriate to
investigate the distribution of spontaneous canine renal
tumors and draw comparison with the lesion in humans.
MATERIALS AND METHODS
Data for this study came from medical abstracts submit
ted by hospitals and clinics of 13 North American veterinary
universities participating in the Veterinary Medical Data
Program. The Veterinary Medical Data Program is a data
registry, sponsored by the National Cancer Institute, to
which participants submit a standardized case abstract,
systematically coded (38), about each clinical episode at
their facility. As of June 1975, there were 1.4 million clinic
visits documented in the registry; approximately 60% were
about dogs.
The Veterinary Medical Data Program abstract contains
descriptive information about the patient (i.e. , identity num
ben, age, breed, and sex), diagnoses, operations, and diag
nostic procedures used during that visit. Also submitted are
data about visits for physical examinations and routine vac
cinations, but not drug therapy. One identity number is
assigned to a patient regardless of the number of subse
quent visits made to the medical facility. When the pedigree
is not available, the determination of breed of the animal is
based upon the phenotypic appearance. The relevant infor
mation collected about tumors includes site, cell type, and
behavioral pattern.
Received October 20, 1976; accepted May 6, 1977.
Dogs with histologically confirmed tumors of the kidney
and renal pelvis were identified, as was a reference popula
tion composed of all dogs seen once for each year during
the study period (hospital-clinic patient-years-at-risk); me
turn visits by the case series were excluded. Both the case
series and the reference population were tabulated by age,
breed, and sex.
The measure of association used in this study was an
estimate of relative risk (R) as approximated by the relative
odds ratio (34). The procedure used to compute the maxi
mum
likelihood
estimate
of R, with
stratification
of control
variables, has been described by Gart (20), and the com
puten procedure for the computation was described by
Thomas (47).
For each summary A value, 95% confidence intervals
were calculated by the method of Gart (20). These intervals
were a measure of the statistical confidence of R; if they did
not include 1, then the calculated A would be considered
significantly different from the R of the comparison group at
the 0.05 level. A less than 1 indicated a lower risk for the
disease in the test group versus the comparison group; R
greater than 1 indicated a higher risk.
RESULTS
Seventy-three dogs were identified with primary renal
neoplasms; 67 involved the kidney, and 6 involved the renal
pelvis. Adenocarcinoma was the most frequent cell type,
followed by carcinoma not otherwise specified and em
bryonal nephmoma (Table 1). Subsequent analyses of age,
breed, and sex risk were limited to “renal
carcinoma―en
compassing the 34 cases of adenocarcinoma and carci
noma (not otherwise specified) of the renal panenchyma.
Review of the medical histories of these dogs revealed a
male with another primary tumor of the urogenital system, a
transitional cell carcinoma of the bladder developing 2
months after the diagnosis of renal carcinoma.
R values were calculated for breeds represented by 3 or
more cases of renal carcinoma. Mixed breed (mongrel) dogs
predominated with 12 cases, followed by 3 cases each
among Labrador retrievers and miniature and toy poodles.
None of these breed groups had a risk significantly different
from that in all breeds combined (Table 2). Fourteen other
breeds not qualifying for risk analysis were represented
among dogs with renal carcinoma.
In both sexes, the risk for renal carcinoma increased with
age (Chart 1). However, males had an overall risk 2.3 times
that of females, based on 23 versus 11 cases, respectively
(95% confidence interval = 1.05 to 4.94). The excess in
males was primarily oven 3 years of age (Chart 2).
AUGUST 1977
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1977 American Association for Cancer Research.
2553
H. M. Hayes, Jr. , and J. F. Fraumeni,
Jr.
Table1
18, 24, 41 , 43), and 70 other tumors of the kidney and renal
pelvis have been reported (6, 11, 17, 35, 36, 45). However,
reference populations for comparison were either lacking
on not suitable for use, so that the reported variations in
diagnosis by age, breed, and sex are uninformative.
The largest canine series reported, which describes the
morphological characteristics of 31 cases of renal carci
noma, was based on material from the Armed Forces Insti
tute of Pathology (5). The authors, in agreement with Willis
(49), concluded that the canine lesion is histologically very
similar to the human counterpart. Because of this similarity,
a comparison of epidemiological features is especially mdc
vant.
In our series, there was no evidence of familial (breed)
predisposition among purebred dogs. A high proportion of
the cases were in mongrel dogs which suggests that, as in
humans (46, 50), genetic determinants of renal carcinoma
are not conspicuous.
Canine renal neoplasms by sex reportedMedicalTotalSexcasesob
to the Veterin
Data Program, March 1964 to July 1975ary
MaleFemaleKidneyAdenocarcinoma
Site/cell type
186Carcinoma
55Embryonal not otherwise
26Fibrosancoma
nephroma
served
24
10
8
specified
25Adenoma
7
22Sarcoma
22Hemangiosarcoma
not otherwise
4
4
specified
30Hemangioma
3
12Fibroma
02Lipoma
11Renal
3
2
2
pelvisSquamous
21Adenocancinoma
cell carcinoma
10Carcinoma
10Fibrosarcoma
not otherwise
3
1
1
1
specified
R
01
10.1
Table 2
Estimated
breedR relative risk (R) of renal carcinoma in dogs by
andstratified
values compared with that in all breeds combined
on age and sex.No.
of
0.87-3.80Labrador
12
3
3
a When
confidence
cantly different
@
dence inter
served
0.70-9.72Miniatune
retriever
and toy poodles
interval
includes
R
1.5
2.7
0.9
0.23-3.19
1 , breed
from that in all breeds combined
R is not
0.01
1
I I I I I I I
I
5
10
0
ROE
studies
and specifically
with
(YERNS)
ft/F
RATIO
A
4.0
A
- CANINE
+
- HUNAN
+
+
++
2.0
÷
+6
++
pet owner
ship. Although the Veterinary Medical Data Program in
cludes medical episodes for non-disease visits and may
therefore approximate the general population-at-risk, our
results cannot be considered as incidence statistics. Esti
mates of relative risk in this study represent hospital-based
prevalence values.
At least 110 canine renal carcinomas (4, 5, 7, 9, 10, 13-15,
22, 23, 27, 30, 31, 33, 40, 48), 22 Wilms' tumors (1, 5, 10, 16,
I I I I I I
15
Chart 1. Estimated relative nisk (R) of canine renal carcinoma by sex and
age. Age interval of 4 to 6 years was used as standard reference (A = 1). A
values for age intervals 0 to 3. 4 to 6, and 7 to 9 were plotted at the midpoint
of each interval. Results for interval @10
years ware plotted at 14 years.
3.0
Our findings, although based upon a well-defined hospi
tal-chinic population of more than 250,000 dogs, are
nevertheless subject to possible biases associated generally
2554
MALE
————FEMALE
at the 0.05 level.
DISCUSSION
retrospective
0.1
signifi
This series contained 6 females and 2 males with Wilms'
tumor (Table 3). Review of the medical histories of these
dogs failed to detect any congenital defects reported with
Wilms' tumor in children (37).
In humans, tumors of the renal pelvis (25) and, possibly,
renal carcinomas (3) have been linked with analgesic drug
use. The medical histories of dogs with these lesions were
reviewed for diseases routinely treated with salicylates. One
of the 6 dogs with a neoplasm of the renal pelvis and 1 of the
34 dogs with renal carcinoma had a history of arthritis. No
other conditions were identified.
with
1.0
95% confi
casesobval―Mixedbneed
Breed group
(A = 1)
1.0
+A
10
+
+
+
70
80
A
+
20
30 40 50 60
AGE IN YEARS
90
Chart 2. Distribution of the male/female ratio of renal carcinoma in hu
mans (12) and dogs from the Veterinary Medical Data Program by midpoint
age intervals using the dog-years to human-years conversion formula (32).
CANCER RESEARCHVOL. 37
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1977 American Association for Cancer Research.
Canine Renal Neoplasms
Table3
dogsBreed
Cases of embryonal
nephroma
breed
Mixed breed
Mixed breed
Female
Female
mos.
1 yr
2-3 yr
Mixed breed
Mixed breed
German shepherd
Female
Male
Female
4-6 yr
7.5 yr
3 yr
Miniature
Female
Female5
Shetland
poodle
sheepdogMale
in
at di
agnosisDischarge
tusMixed groupSexAge
—
reported
sta
Alive
Euthanasia
Euthanasia
Alive
Alive
Alive
6 yr
8 yrEuthanasiaEuthanasia
Our analysis quantifies, for the first time, an excess risk of
renal carcinoma in male dogs. This finding stands in con
trast to the female predominance of canine bladder cancer,
which may be related to the bitch's tendency toward pro
longed urine retention (21). The male predominance of ca
nine renal carcinoma resembles that in humans (Chart 2).
The canine male excess peaks in the 7- to 9-year age group.
Using LeBeau's conversion of dog years to human years
(32), the mean of this age group approximates 49 years of
age in humans. Similarly, the sex ratio of male-to-female
incidence rates in humans is greatest at 45 to 49 years of
age (12).
The primary causes of renal carcinoma in humans are
unclean. A relationship to cigarette smoking has been sug
gested, but the evidence is inconclusive (3, 19, 29, 50). Oc
cupational exposures to polycyciic hydrocarbons among
coke oven workers in the steel industry and to cadmium
have been implicated in some cases (29, 39). Worldwide
trends indicate a correlation of renal carcinoma with con
sumption of coffee (44) and with animal fat and protein (2),
but these hypotheses were not sustained by a case-control
study (3).
The reason for the male predominance of human renal
carcinoma is obscure. Environmental factors are suggested
by the rising incidence reported among American men but
not women (26) and call for further studies to evaluate the
moleof tobacco, nutrition, and occupational determinants.
However, these factors cannot account for the male excess
of kidney cancer in dogs (and similar trends in sex ratio with
age) and suggest the influence of host susceptibility, in
cluding the endogenous production on metabolism of sex
hormones. This possibility is consistent with the evidence
for remission in some human patients with renal carcinoma
using progestins (28, 42).
ACKNOWLEDGMENTS
We wish to thank Karen Beckwith for editorial and technical assistance
and the veterinary university hospitals and clinics contributing data to this
study: Colorado State University, University of California, University of Geor
gia, University of Guelph (Ontario), University of Illinois, Iowa State Univer
sity, Kansas State University, Michigan State University, University of Minna
sota, University of Missouri, Ohio State University, Purdue University (Lafay
etta, Ind.), and University of Saskatchewan.
REFERENCES
1. Appleby, E. C., Hayward, A. H. S., and Ranyand, P. The Use of a
Computenised Record System in the Analysis of 391 Cases of Suspected
Neoplasia. Vet. Record, 89: 547-552, 1971.
AUGUST
2. Armstrong, B., and Doll, R. Environmental Factors and Cancer Incidence
and Mortality in Different Countries, with Special Reference to Dietary
Practices. Intern. J. Cancer, 15: 617-631 , 1975.
3. Armstrong, B., Ganrod, A., and Doll, R. A Retrospective Study of Renal
Cancer with Special Reference to Coffee and Animal Protein Consump
tion. Bnit. J. Cancer, 33: 127-136, 1976.
4. Bannon, C. N., Saunders, L. z. , and Jubb, K. V. Secondary Intraocular
Tumors. Am. J. Vet. Res., 24: 835-853, 1963.
5. Baskin, G., and Depaoli. A. Primary Renal Naoplasms ofthe Dog: Report
of 48 Cases. Vet. Pathol., in press.
6. Ceretto, F. Primary Polymorphocallular Sarcoma of the Kidney with
Metastasas in the Liver, in a Dog. Ann. Fac. Med. Vet. Tonino, 14: 53-69,
1964.
7. CoIlet, P. Tumeur Kystique du Rein chez une Chienne: Nephnactomie:
Guenison. Bull. Soc. Sci. Vet. Lyon, 53: 97-103, 1951.
8. Coplan, M. Report on Carcinogenic Action of Certain Agents on Urinary
Bladderof Dogs.J. Am. Med. Assoc., 172: 1611-1618, 1960.
9. Cotard, J. P., Sevastre, J., and Wyars, M. Observation de un Cas da
Cancer Pnimitif du Rein chez Ia Chian. Rac. Mad. Vet., 151: 343-347,
1975.
10. Cotchin, E. Some Tumours of Dogs and Cats of Comparative Veterinary
and Human Interest. Vat. Record, 71: 1040-1050, 1959.
11. Cnocker, W. J. Three Thousand Autopsies. Cornell Vet., 9: 142-161,
1919.
12. Cutler, S. J., and Young, J. L., Jr. (ads.). Third National Cancer Survey
Incidence Data. NatI. Cancer Inst. Monograph, 41: 100-1 11, 1975.
13. DeKock, G. Pathological Studies on Neoplasms of Dogs in South Africa.
Ondanstepoort J. Vat. Ras., 29: 35-54, 1962.
14. Dorn, C. R., Taylor, D. 0. N., Frye, F. L., and Hibbard, H. H. Survey of
Animal Neoplasms in Alameda and Contra Costa Counties, California. I.
Methodology and Description of Cases. J. NatI. Cancer Inst. , 40: 295305. 1968.
15. Douch, J. Unilateral Naphractomy. Vet. Record, 98: 99, 1976.
16. Drew, R. A., Done, S. H., and Robins, G. M. Canine Embryonal Na
phroma: A Case Report. J. Small Animal Pract., 13: 27-39, 1972.
17. Dunn, W. H. Renal Neoplasm of Dog. J. Am. Vat. Med. Assoc., 84: 119120, 1934.
18. Fiadler, H. H., and Radecker, R. Embnyonales Nephnom bai ainem zehn
jahrigen Hund. Deut. Tiaraarztl. Wochschr., 80: 458-459, 1973.
19. Fraumani, J. F., Jr. Cigarette Smoking and Cancers of the Urinary Tract:
Geographic Variation in the United States. J. NatI. Cancer Inst., 41:
1205-1211, 1968.
20. Gart, J. J. Point and Interval Estimation ofthe Common Odds Ratio in the
Combination of 2x2 Tables with Fixed Marginals. Biometnika, 57: 471—
475, 1970.
21. Hayes, H. M., Jr. Canine Bladder Cancer: Epidemiologic Features. Am. J.
Epidamiol., 104: 673-677, 1976.
22. Hottandonf, G. H., and Hirth, R. S. Lesions of Spontaneous Subclinical
Disaase in Beagle Dogs. Vet. Pathol., 11: 240-258, 1974.
23. Howard, E. B., and Nielsan, S. W. Neoplasia of the Boxer Dog. Am. J.
Vet. Res., 26: 1121-1131, 1965.
24. Jay, C. L. Embryonal Naphroma in a Dog. Southwestern Vet., 13: 234235, 1960.
25. Johansson, S. , Anganvall, L., Bengtsson, U.•
and Wahlqvist, L. Uroapi
thelial Tumors of the Renal Pelvis Associated with Abusa of Phenacatin
containing Analgesics. Cancer, 33: 743-753, 1974.
26. Kenton, A. L. F., Maigs, J. W., Heston, J. F., and Flannery, J. T. Epida
miology of Renal Cell Carcinoma in Connecticut, 1935-1973. J. NatI.
Cancer Inst., 57: 495—500,
1976.
27. Kasbohm, C., and Lattow, E. Kasuistik Intnaabdominalen “Tumonan―
bairn Hund. Berlin Muannch. Tiaraarztl. Wochschr., 74: 34-35, 1961.
28. Kalley, R. M. Progastins. In: J. F. Holland and E. Frai, III (ads.), Cancer
Medicine, pp. [email protected]: Lea and Febigen, 1973.
29. Kolonel, L. N. Association of Cadmium with Renal Cancer. Cancer, 37:
1782-1787, 1976.
30. Knonbengar, H. Zwai Falle von Karzinoman des Harnapparatas des
Hundes mit Lungenmetastasen und Aknopachie. Monatsh. vet. Med.,
14: 609—611,
1959.
31 . Knook, L. A Statistical Investigation of Carcinoma in the Dog. Acts
Pathol. Microbiol. Scand., 35: 407-422, 1954.
32. LeBeau, A. L'age du Chien at CeIui da L'homme. Essaa de Statistique sun
Ia MortalitéCanine. Bull. Acad. Vet. Franca, 26: 229-232, 1953.
33. Lucké,V. M., and Kelly, D. F. Renal Carcinoma in the Dog. Vat. Pathol.,
13: 264-276, 1976.
34. MacMahon, B. , and Pugh, T. F. Epidemiology. Principles and Methods,
pp. 232-233; 269-273. Boston: Little, Brown, and Co., 1970.
35. Mugera, G. M. Canine and Feline Neoplasms in Kenya. Bull. Epizootic
Diseases Africa, 16: 367-370, 1968.
36. Palrninteni, A., Geib, L. W., and Eby, C. H. A Case of Renal Hamangioma.
Mod. Vet. Pnact., 43: 68, 1962.
37. Pandengrass, T. W. Congenital Anomalies in Children with Wilrns' Tu
mon. Cancer, 37: 403-408, 1976.
38. Pniestar, W. A., Adalstein, E. H., and Peters, J. A. (ads.). Standard
1977
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1977 American Association for Cancer Research.
2555
H. M. Hayes, Jr., and J. F. Fraumeni,
Jr.
Nomenclature of Veterinary Diseases and Operations, Ed. 1 revised.
Department of Health, Education, and Welfare. Washington, D. C. : U. S.
Government Printing Office, 1966.
39. Redmond, C. K., Ciocco, A. , Lloyd, W., and Rush, H. W. Long-term
Mortality Study of Steelworkers. VI. Mortality from Malignant Neoplasms
among Coke Oven Workers. J. Occupational Med., 14: 621-629, 1972.
40. Robinson, F. R., and Garner, F. M. Histopathologic Survey of 2,500
German Shepherd Military Working Dogs. Am. J. Vet. Res., 34: 437-442,
1973.
41
.
Sagantz,
J.
W.,
Ayers,
K.
M.,
Cashell,
I.
G.,
and
Robinson,
F.
R.
Malig
nant Embnyonal Nephroma in an Aged Dog. J. Am. Vet. Med. Assoc.,
161: 1658-1660, 1972.
42. Samuels, M. L., Sullivan, P., and Howe, C. D. Medroxyprogesterone
Acetate in the Treatment of Renal Cell Carcinoma (Hypernephroma).
Cancer, 22: 525-532, 1968.
43. Sastny, G. A., and Twiehaus, M. J. A Study of the Animal Neoplasms of
2556
Kansas State. Indian Vet. J., 41: 317-326, 1964.
44. Shenna, D. H. Renal Carcinoma and Coffee Consumption in 16 Coun
tries. Bnit. J. Cancer, 28: 473—474,
1973.
45. Splitter, G. A., Rawlings, C. A., and Casey, H. W. Renal Hamartoma in a
Dog. Am. J. Vet. Ras., 33: 273-275, 1972.
46. Steinberg, S., Brodovsky, H. S., and Goapp, C. E. Renal Carcinoma in
Mother and Daughter. Cancer, 29: 222-225, 1972.
47. Thomas, D. C. Exact and Asymptotic Methods fon the Combination of
2x2 Tables. Computers Biomed. Res., 8: 423-446, 1975.
48. Thrasher, J. P. Neoplasms of Dogs. J. Am. Vet. Mad. Assoc., 138: 27-30,
1961.
49. Willis, R. A. Pathology of Tumours, Ed. 3, p. 459. Washington, D. C.:
Buttenworth, 1960.
50. Wynden, E., Mabuchi, K. , and Whitmone, W. F., Jr. Epidemiology of
Adenocarcinoma of the Kidney. J. NatI. Cancer Inst., 53: 1619-1634,
1974.
CANCER RESEARCHVOL. 37
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1977 American Association for Cancer Research.
Epidemiological Features of Canine Renal Neoplasms
Howard M. Hayes, Jr. and Joseph F. Fraumeni, Jr.
Cancer Res 1977;37:2553-2556.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/37/8_Part_1/2553
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1977 American Association for Cancer Research.