J. Comp. Path. 1996 Vol.

114, 165-174

F e l i n e F i b r o s a r c o m a s at V a c c i n a t i o n S i t e s a n d
Non-vaccination Sites
E D. Doddy*++w L. T. Gllckman*, N. W. G l l c k m a n t and
E. B. Janovitz +
* Department of Veterinary Pathobiology and I" Centrefor Applied Ethology and Human-Animal
Interaction, Purdue University, West Lafayette, IN 47907, and ,t Indiana Animal Disease
Diagnostic Laborato~.y, West Lafayette, IN 47907, USA

Summary
A retrospective study of 195 feline sarcomas diagnosed histologically between
July 1988 and June 1994 showed that 170 (87"2%) were fibrosarcomas.
Cats with vaccination site (VS) fibrosarcomas were younger (8"6_ 3"9 years;
median = 8 years) than cats with non-vaccination site (NVS) fibrosarcomas
(10"2_ 4"7 years; median= 11 years) (P=0-03), but there was no such as-
sociation with breed, sex, or "neuter status". Microscopical features more
characteristic of VS fibrosarcomas than of NVS fibrosarcomas were (1)
subcutaneous location (P<0.001), (2) necrosis (P<0"001), (3) inflammatory cell
infiltration (P<0"001), (4) increased mitotic activity (P<0-02), (5) pleomorphism
(P<0"001), and (6) variability in the density of the extracellular matrix
(P<0"001). When these data were fitted to a logistic regression model, younger
age (P= 0"003), subcutaneous location of the fibrosarcoma (P= 0"0002), and
the presence of inflammation (P=0"017) were more characteristic of VS
fibrosarcomas than of NVS fibrosarcomas. The study showed that in the
absence of any vaccination history, the age of a cat, coupled with certain
histological characteristics (e.g., tumour location in skin, and inflammation),
may help in distinguishing VS fibrosarcomas from NVS fibrosarcomas. The
characteristic histological features of VS fibrosarcomas, such as necrosis,
increased mitotic activity and pleomorphism, are those of aggressive
tumours. 9 1996 W.B. Saunders Company Limited

Introduction
Since 1991, an increase in the incidence of sarcomas in the skin of cats at
vaccination sites (VSs), and convincing epidemiological evidence of a link
between vaccination and subsequent tumour development have been reported
(Hendrick and Goldschmidt, 1991; Hendrick et al., 1992; Esplin et al., 1993;
Kass et al., 1993; Hendrick et al., 1994b). O f VS sarcomas, fibrosarcoma is
the most common (Hendrick and Goldschmidt, 1991; Esplin et al., 1993; Kass
et al., 1993; Hendrick and Brooks, 1994), but malignant fibrous histiocytoma
(Esplin et al., 1993; Hendrick and Brooks, 1994), osteosarcoma (Esplin et al.,
1993; Hendrick and Brooks, 1994), myofibroblastic sarcoma (Dubielzig et al.,
1993), chondrosarcoma (Hendrick and Brooks, 1994) and rhabdomyosarcoma

wCurrent address: Harvard Medical School, New England Regional Primate Research Center, One Pine
Hill Drive, P.O. Box 9102, Southborough, MA 01772-9102, USA.

00214975/96/020165+ 10 $12.00/0 9 1996 W.B. Sannders Company Limited

166 F.D. Doddy et al.
(Hendrick and Brooks, 1994) have also been described. Some of these tumours
have been evaluated and classified immunohistochemically and electron mi-
croscopically, as well as by light microscopy.
The phenomenon of sarcoma formation at VSs is apparently unique to
cats. The development of VS sarcomas 3 months to 3 years after vaccination
suggests that the inflammatory reaction to vaccines sometimes results in
uncontrolled growth of transformed mesenchymal cells. Consistent with this
hypothesis is the development of primary intraocular sarcomas in some cats
after ocular trauma or chronic uveitis (Dubielzig, 1984; Peiffer et al., 1988;
Dubielzig et al., !990; Hakanson et al., 1990). Chronic localized inflammatory
reactions to vaccines have been reported in h u m a n beings (Fawcett and Smith,
1984), dogs a n d cats (Schmeitzel et al., 1986; Hendrick and Dunagan, 1991;
Macy, 1994)..In man, the association of certain malignancies with chronic
inflammation, chronic irritation, and mechanical trauma, has been recognized
(Zajicek, 1985; Matsubara et al., 1988; Preston-Martin et al., 1990; Gordon
and Weitzman, 1993). Additionally, an aluminum oxide ceramic hip prosthesis
has been implicated as the cause of a h u m a n soft-tissue sarcoma (Ryu et al.,
1987).
Because of the increased awareness of VS sarcomas by veterinarians, cat
owners, and vaccine m~nufacturers, this study was undertaken to identify
factors associated with feline VS fibrosarcomas, and to define the histological
characteristics of such tumours.

Materials and Methods

Selection of Cases
A search of feline necropsy and biopsy case records at the Indiana Animal Disease
Diagnostic Laboratory (ADDL) for the period ofJuly 1988 toJune 1994 was undertaken
to identify cases of the following, types of neoplasia: fibrosarcoma, osteosarcoma,
malignant fibrous histiocytoma, gaant celt tumour of soft parts, myofibroblastic sar-
coma, rhabdomyosarcoma, leiomyosarcoma, chondrosarcoma, sarcoma, and un-
differentiated sarcoma. Tissue sections and the case history from each accession were
reviewed to verify the original diagnosis, and to obtain the age, breed, sex, relevant
history, and physical findings.
The topographical site from which each fibrosarcoma (n= 170) was removed was
specified in writing on the form submitted with the tumour sample by the referring
veterinarian. Vaccination sites (VSs) were defined as the interscapular and scapular
regions, flank and paralumbar regions, dorsolateral thorax, dorsal back and neck,
and femoral regions. Non-vaccination sites (NVSs) were any other location such as
the head, distal parts of the limbs, bone or tail. Because the ventrolateral aspect of
the neck Was regarded as a dubious VS, fibrosarcomas (n = 5) at this location were
excluded from the analyses.

Histopathological Examination
The tissues had been routinely processed for histopathology. In addition to staining
with haematoxylin and eosin (HE), selected tissue sections were also stained with
Masson's t/ichrome (n = 18) and phosphotungstic acid-haematoxylin (n = 9). The clas-
sification of each sarcoma was based on histological characteristics.
Tissue sections were examined by one of the authors (FDD), without knowledge of
the reported location of the tumour, for the following microscopical features: location
 Feline F i b r o s a r c o m a s at V a c c i n a t i o n Sites 167
in skin (dermis, subcutis, both, or not applicable); necrosis (mild, moderate, or severe);
inflammatory infiltrates (present or absent); granulation tissue (present or absent, or
uncertain); binucleated or multinucleated tumour cells, or both (few, moderate num-
bers, or many); local invasion (present, absent, or unknown); average number of
mitoses per 10 microscopical ( x 400) fields; pleomorphism (mild, moderate, or severe);
stromal component in addition to collagen (osseous, chondroid, or myxomatous); and
variability in the density of the extracellular matrix (none, mild, or marked). It was
not always possible to evaluate accurately each microscopical feature, for one or more
of the following reasons: the tissue specimen available for microscopical examination
was too small in size; the tumour boundaries were incomplete; the location of the
tumour in skin was uncertain. Ambiguous results were excluded from the final analyses.
VS fibrosarcomas were compared with NVS fibrosarcomas, in respect of each of the
10 microscopical features.

Statistical Methods
Information from the ADDL case records was coded and entered into a computerized
database. Data were then transferred to the SAS| System for Windows version 6"08
(SAS Institute, Cary, NC, USA) for all statistical analyses. A chi-square test was used
to determine if the occurrence of fibrosarcomas, as compared with other sarcomas,
was associated with VS. A Pvalue of <0-05 was considered significant for all analyses.
Due to the relatively small number of sarcomas other than fibrosarcoma (n=25,
12"8%), further statistical analyses were restricted to fibrosarcomas.
Cats with VS fibrosarcomas and those with NVS fibrosarcomas were compared in
terms of host factors (age, gender, and "neuter status") and histological characteristics
of the tumour; a chi-square test was used for categorical variables and a t-test for
continuous variables.
Histological characteristics and host factors that were statistically associated (P<0"05)
with VS fibrosarcoma in the univariate analyses were included in a multivariate
unconditional logistic regression analysis. Two histological characteristics, local in-
vasion and granulation tissue, were significantly associated with VS fibrosarcomas in
the univariate analysis, but were not included in the multivariate analysis since data
on these variables were uncertain for 51% and 89% of the cats, respectively. The
log-likelihood statistic was used to assess the final fit of the regression model. The
significance of each explanatory variable was assessed with the Wald chi-square
statistic. Parameter estimates for each explanatory variable were "exponentiated" to
obtain the odds ratio, which is a measure of the strength of the association between
each explanatory variable and the dependent variable, VS.

Results

Anatomical Distribution
Feline sarcomas classified by a n a t o m i c a l location are shown in T a b l e 1, a n d
fibrosarcomas similarly classified are shown in T a b l e 2. T h e o c c u r r e n c e o f
fibrosarcomas at VSs c o m p a r e d with that o f o t h e r sarcomas was not significant
( P = 0 - 1 2 1 ) (data not shown). T h e most f r e q u e n t V S at which fibrosarcomas
o c c u r r e d was the interscapular a n d scapular regions ( 3 0 / 1 7 0 ; 17.6%), a n d
the most f r e q u e n t N V S was the h e a d ( 3 1 / 1 7 0 ; 18"2%) (Table 2).

Age, Sex, and Breed

T h e average age o f cats with V S fibrosarcomas was 8 " 6 _ 3"9 years (median =
8 years) versus 10"2-t-4"7 years ( m e d i a n = l l years) for cats with N V S
168 E D. D o d d y e t td.
Table 1
S i t e s at w h i c h t h e d i f f e r e n t t y p e s o f f e l i n e s a r c o m a s o c c u r r e d

Sa~ortla lyp~ Number (and %) of cases in which the tumour occurredat

VS NVS dubious VS

Fibrosarcoma* 104 (84"6) 61 (92-4) 5 (83"3)

Malignant fibrous histiocytoma 16 (13"0) 0 (0) 0 (0)
Osteosarcoma 0 (0) 2 (3"0) 1 (16"7)
Neurofibrosarcoma 0 (0) 1 (1"5) 0 (0)
Undifferentiated sarcoma~ 3 (2.4) 2 (3"0) 0 (0)

Total 123 (100) 66 (100) 6 (100)

VS, vaccination site; NVS, non-vaccination site

* Includes three cats each with more than one fibrosarcoma; one tumour from each animal was included
by random selection or first occurrence.
t Includes one cat with a cutaneous undifferentiated sarcoma, but excludes later occurrence of visceral
fibrosarcomas.

Table 2
S i t e s at w h i c h f e l i n e f i b r o s a r c o m a s o c c u r r e d

Location Number (and %) of

f~msarcom~

Vaccination sites
Interscapular/scapular 30 (17"6)
Flank/paralumbar 27 (15"9)
Dorsolateral thorax 19 (11"2)
Dorsal area of back and neck 17 (10"0)
Femoral region 11 (6"5)

Non-vaccination sites
Head* 31 (18"2)
Limbs~ 13 (7-6)
Other++ 10 (5"9)
Bone 3 (1-8)
Tail 4 (2-4)

Dubious vaccine site

Ventrolateral neck 5 (2-9)

Total 170 (100)

* "Head" includes oral mucous membranes.

t "Limbs" excludes the scapulae and outer thigh.
+ "Other" includes perineum, nasopharynx, mesentery, intestine, pelvic
inlet, tongue, and perivulvar region.

fibrosarcomas ( P = 0"03). Ages were unknown for 10 (6" 1%) of the 165 cats.
There were no significant differences in sex or neuter status between cats
with VS fibrosarcoma and those with NVS fibrosarcoma. Gender was unknown
for three (1-8%) of the 165 cats; 73 (45-1%) were male and 89 (54-9%) were
female. There were 74 (45-7%) intact and 88 (54"3%) neutered cats.
Breed did not appear to be associated with site of fibrosarcoma. Breed was
unknown for four (2"4%) of the 165 cats with such tumours, 151 (91"5%) cats
were of mixed breed, and 10 (6" 1%) were purebred.
 Feline Fibrosarcomas at Vaccination Sites 169
Table 3
O c c u r r e n c e of feline s a r c o m a s in different y e a r s

Tumour type and Number (and %) of sarcomas in

location
1989 1990 1991 1992 1993 1994

Fibrosarcomas
All 21 (100) 21 (100) 39 (100) 20 (100) 36 (100) 33 (100)
VS 7 (33"3) 10 (47"6) 22 (56"4) 13 (65) 26 (72-2) 26 (78.8)
NVS 13 (61'9) 10 (47-6) 15 (38-5) 7 (35) 10 (27-8) 6 (18'2)
DVS 1 (4-8) 1 (4'8) 2 (5-1) 0 (0) 0 (0) 1 (3"0)
Other sarcomas
All 0 (0) 2 (100) 7 (100) 0 (0) 3 (100) 13 (100)
VS 0 (0) 2 (100) 7 (100) 0 (0) 3 (100) 7 (53"8)
NVS 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 5 (38.5)
DVS 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1 (7.7)

VS = vaccination site; NVS = non-vaccination site; DVS = dubious VS.

Table 4
Feline a c c e s s i o n s , fibrosaxcoma morbidity, and "ratio of location" in different y e a r s

Parameter Data for

1989 1990 1991 1992 1993 1994

Accessions* 1855 1244 1314 1215 1184 1129

FPM~" 1"13 1"69 2"97 1"65 3-04 2"92
VS: NVS
fibrosarcomas 0"54 1"00 1"47 1"86 2-6 4'33

* Total feline accessions at ADDL.

~"Fibrosarcoma proportionate morbidity (FPM)= the ratio of the number of fibrosarcomas to the total
number of feline accessions.

Temporal Trends
Feline sarcomas were submitted to the A D D L with increasing frequency
between July 1988 and J u n e 1994 and this was associated with an increase in
the proportionate morbidity for fibrosarcomas (the ratio of the number of
fibrosarcomas to the number of total feline accessions) (Tables 3 and 4). The
ratio of VS fibrosarcomas to NVS fibrosarcomas increased > 8-fold over this
period. (Data were not analysed statistically.)

Histology
The histology of 165 fibrosarcomas (104 from VSs and 61 from NVSs) was
reviewed and analysed (Table 5). When the location of VS fibrosarcomas
within the skin was compared with that of NVS fibrosarcomas in the skin, the
former were found more commonly in the subcutis (P<0-001).
Necrosis was more common in VS fibrosarcomas than in NVS fibrosarcomas
(P<0"001). Necrotic areas in approximately 25% (n=27) of the VS fibro-
sarcomas had cavitated centres infiltrated by variable numbers ofmacrophages
admixed with fewer neutrophils and nuclear debris. Five of these cavitated
lesions were associated with vascular thrombosis.
170 E D. D o d d y e t al.
Table 5
Unlvaxiate analysis of 165 feline fibrosarcomas
(VSs vs NVSs)

Micwscopicalfeature P Value

Location in skin <0-001"

Tumour necrosis <0-001"
Inflammation <0'001 *
Granulation tissue 0.042*
Binucleated and multinucleated giant cells 0.08*
Local invasion 0'014"
Mitotic activity/10 ( • 400) fields 0-021'f
Pleomorphism <0-001 *
Other stromal component(s) 0"147"
Variability in density of extracellular matrix <0-001"

* Chi-squared statistic.
I" t-Test statistics.

Inflammation was more commonly associated with VS fibrosarcomas than

NVS fibrosarcomas (P<0"001). Lymphocytes constituted the predominant
inflammatory cell type (data not shown) in 72 (69"2%) of 104 VS fibrosarcomas
and in 15 (25%) of 60 NVS fibrosarcomas, but macrophages were also found
in tissues around the VS tumours. Granulation tissue at the basal or basolateral
margins of 19 fibrosarcomas was infiltrated by inflammatory cells. Granulation
tissue was more common in VS fibrosarcomas (n= 16; 15"4%) than in NVS
fibrosarcomas (n= 3; 4"9%) (P=0"042). Multinucleated tumour cells (two to
five nuclei) were observed, but their numbers did not differ significantly
between VS and NVS fibrosarcomas (P= 0"08). Local invasion occurred more
commonly in VS (n=46; 44"2%) than NVS (n= 11; 18%) fibrosarcomas (P=
0.014), and vascular invasion was observed in two VS malignant fibrous
histiocytomas. Mitotic figures were more common in VS than NVS fibro-
sarcomas (P=0"021). Cellular pleomorphism was severe in 64-4% of VS
fibrosarcomas as compared with 36-1% of NVS fibrosarcomas (P<0-001).
Although both VS and NVS fibrosarcomas had collagenous extracellular
matrix, only VS fibrosarcomas had osteoid (n=2, 1"9%), chondroid (n=2;
1.9%) or myxomatous (n=7; 6"7%) matrix (P=0-147). The organization of
spindle cells and extracellular matrix was less uniform (i.e., had greater
variation in density) in VS than NVS fibrosarcomas (P<0.001).
Age of the cat and six histological characteristics that were statistically
different between VS and NVS fibrosarcomas were included in a multivariate
unconditional logistic regression analysis (Table 6). Missing information pre-
cluded the inclusion of local invasion and granulation tissue. The fit of the
regression model was statistically significant (P= 0"0001). With increasing age
there was a decreased risk of developing a fibrosarcoma at a VS (odds ratio
[OR] = 0"98; P = 0"003). Fibrosarcomas that occurred at VSs were more likely
than those at NVSs to show inflammation ( O R = 5-28;P= 0"017) and to be
situated in the subcutis ( O R = 25"44; P = 0"0002).
 Feline Fibrosarcomas at V a c c i n a t i o n Sites 171
Table 6
Logistic regression analysis to identify histological characteristics of 109 fibrosarcomas that
were a s s o c i a t e d with vaccination sites

Variable Odds 95% Confidence Wald chi-squared

ratio limits (P value)

Age (months) 0-98 0"97, 0"99 8-78 (0-003)

Mitotic activity* 0-99 0-91, 1"07 0.06 (0"813)
Pleomorphism~ 1-16 0'46, 2"9l 0-10 (0-750)
Necrosis~ 1"28 0"74, 2'23 0"78 (0"375)
Variability of extracellular matrix~" 2-07 0"80, 5"33 2"26 (0-133)
Inflammation (present vs absent) 5-28 1.35, 20"64 5-73 (0-017)
Location in skin (subcntis vs dermis) 25-44 4"56, 141'90 13"62 (0-0002)

* Mitotic activity is total number of mitoses per 10 ( x 400) microscopical fields.

I" Odds ratio for each one unit increase in level of variable. See text for explanation of coding of
variables.

Discussion
Fibrosarcoma, a malignant tumour of fibroblasts, is the most common malig-
nant mesenchymal tumour in cats, being reported to represent 12-25% of all
feline skin tumours in recent surveys (Gross et al., 1992). In cats, the majority
of subcutaneous fibrosarcomas arise on the trunk and distal parts of the limbs,
and the majority of dermal fibrosarcomas arise in the pinnae and digits (Gross
et al., 1992). The average age of cats with fibrosarcomas is 12 years (Gross et
al., 1992); the tumours often recur locally but rarely metastasize.
The rate of submission of feline fibrosarcomas at the Indiana ADDL from
July 1989 to June 1994 was relatively constant, but there was an increase in
the proportionate morbidity for fibrosarcomas. A reversal in the ratio of VS
to NVS fibrosarcomas occurred in 1991, possibly due to an increased awareness
of VS-associated fibrosarcomas in cats.
The study indicated that the morphology of VS and NVS fibrosarcomas
differed, and that cats with VS fibrosarcomas were generally younger than
those with NVS fibrosarcomas. The pathogenesis of fibrosarcoma associated
with feline sarcoma virus and feline leukaemia virus has been elucidated
(Hoover and Mullins, 1991), but the pathogenesis of non-retroviral fibro-
sarcomas in cats is unknown. The results of this study and others (Hendrick
et al., 1992) suggest that VS and NVS fibrosarcomas arise as a consequence
of different oncogenic mechanisms. In particular, the pathogenesis of vaccine-
induced fibrosarcomas is probably secondary to chronic inflammation with
neoplastic transformation of mesenchymal cells which participate in the
connective tissue repair process (Hendrick et al., 1992). Apparent transition
from chronic inflammation to tumour development has been observed in
surgical biopsy specimens from VS lesions (Hendrick et al., 1994a). The
association of inflammation with VS fibrosarcomas in our study is in agreement
with this observation, but the immunological response to tumour antigens
may also contribute to the inflammatory response. The more frequent presence
of granulation tissue at the periphery of VS than of NVS fibrosarcomas is
further support for a role of inflammation and healing in tumorigenesis. A
172 E D. Doddy et al.
similar pathogenesis has been proposed for primary ocular sarcomas in cats
(Dubielzig et al., 1990).
VS fibrosarcomas occurred more frequently in the subcutis than in the
dermis, while NVS fibrosarcomas occurred more frequently in the dermis.
This finding correlated with the increasing use of the subcutaneous route of
vaccination in cats (Hendrick et al., 1992). VS fibrosarcomas are not limited
to the skin; they have also been reported in the muscles of the thigh (Dubielzig
et al., 1993; Hendrick and Brooks, 1994).
Necrosis, which was more common in VS than in NVS fibrosarcomas, may
have been attributable to infarction; thrombosis was observed in five VS
fibrosarcomas in association with necrotic foci. However, other necrotic foci
may be due to the tumour outgrowing its blood supply, or to the production
of subcutaneous fat necrosis and inflammation by the initial vaccine injection.
Hendrick et al. (1994b) reported that VS were larger than NVS fibrosarcomas.
Rapid growth and local invasion are more characteristic of VS fibrosarcomas,
suggesting that they are more biologically aggressive. In a previous report,
VS were more likely than NVS fibrosarcomas to recur after surgical excision
(Hendrick et al., 1994b). Although metastasis of VS fibrosarcomas is probably
rare, such metastasis to the mediastinum and lungs was recently reported
(Rudmann et al., 1994); and in the present study vascular invasion by apparently
malignant multinucleated giant cells in two malignant fibrous histiocytomas
was observed.
The difference in phenotypic expression of tumour type may be the result
of differentiation of a mesenchymal stem cell. There is in-vitro evidence to
suggest that cytokines play a role in phenotypic expression of some cell types
(Gabbiani, 1992). Within the group of VS fibrosarcomas we observed marked
variability in the density of the extracellular matrix, similar to that in
fibrosarcomas of infants and children (Schofield et al., 1994). None of the
morphological patterns observed in children were predictive of biological
behaviour of the tumour.
In conclusion, distinction between VS and NVS fibrosarcomas by their
morphological features is possible. From this study and other reports, it would
appear that VS fibrosarcomas are locally invasive but not likely to metastasize.
Since the incidence of VS fibrosarcomas is probably low (Kass et al., 1993),
vaccination against feline leukaemia and rabies should not be discouraged.
Further research is needed, however, to clarify the pathogenesis ofVS tumours,
in the hope of improving safety.

Acknowledgments
We thank Ms J. Hewitt, Ms M. F. Nelson, and Ms L. Bakowski for their assistance
in obtaining accession records from the Indiana ADDL, and MsJaneice Samman for
the preparation of histological slides. This paper is submitted as journal article no.
14511, Agricultural Experiment Station, Purdue University.

References
Dubielzig, R. R. (1984). Ocular sarcomas following trauma in three cats. Journal of
the American VeterinaTy Medical Association, 184, 578-581.
 Feline Fibrosarcomas at Vaccination Sites 173
Dubielzig, R. R., Everitt, J., Shadduck, J. A. and Albert, D. M. (1990). Clinical and
morphologic features of post-traumatic ocular sarcomas in cats. VeterinaryPathology,
27, 62-65.
Dubielzig, R. R., Hawkins, K. L. and Miller, P. E. (1993). Myofibroblastic sarcoma
originating at the site of rabies vaccination in a cat. Journal of Veterinary Diagnostic
Investigation, 5, 637-638.
Esplin, D. G., McGill, L. D., Meininger, A. C. and Wilson, S. R. (1993). Postvaccination
sarcomas in cats. Journal of the American VeterinaryMedicalAssociation, 202, 1245-1247.
Fawcett, H. A. and Smith, N. P. (1984). Injection-site granuloma due to aluminum.
Archives of Dermatology, 120, 1318-1322.
Gabbiani, G. (1992). The biology of the myofibroblast. Kidney International, 41,530-532.
Gordon, L. I. and Weitzman, S. A. (1993). Inflammation and cancer. The Cancer
Journal, 6, 257-261.
Gross, T. L., Ihrke, P.J. and Walder, E.J. (1992). Fibrocytic tumors. In: Veterinary
Dermatopathology, Mosby-Year Book, Inc., St Louis, pp. 413-415.
Hakanson, N., Shively, J. N., Reed, R. E. and Merideth, R. E. (1990). Intraocular
spindle cell sarcoma following ocular trauma in a cat: case report and literature
review. Journal of the American Animal Hospital Association, 26, 63--66.
Hendrick, M.J. and Brooks, J. J. (1994). Postvaccinal sarcomas in the cat: histology
and immunohistochemistry. Veterinary Pathology, 31, 126-129.
Hendrick, M.J. and Dunagan, C. A. (1991). Focal necrotizing granulomatous pan-
niculitis associated with subcutaneous injection of rabies vaccine in cats and dogs:
10 cases (1988 1989). Journal of the American Veterinary Medical Association, 198,
304-305.
Hendrick, M. J. and Goldschmidt, M. H. (1991). Do injection site reactions induce
fibrosarcomas in cats? Journal of the American VeterinaryMedical Assoication, 199, 968.
Hendrick, M. J., Goldschmidt, H. M., Shofer, F. S., Wang, Y. and Somlyo, A. P.
(1992). Postvaccinal sarcomas in the cat: epidemiology and electron probe
microanalytical identification of aluminium. Cancer Research, 52, 5391-5394.
Hendrick, M.J., Kass, P. H. McGill, L. D. and Tizard, I. R. (1994a) Postvaccinal
sarcomas in cats. Journal of the National Cancer Institute, 86, 341-343.
Hendrick, M. J., Shofer, F. S., Goldschmidt, M. H., Haviland, J. C., Schelling,
S. H., Engler, S.J. and Gliatto,J. M. (1994b). Comparison of fibrosarcomas that
developed at vaccination sites and at nonvaccination sites in cats: 239 cases
(1991-1992). Journal of the American Veterinary Medical Association, 205, 1425-1429.
Hoover, E. A. and Mullins, J. I. (1991). Feline leukemia virus infection and diseases.
Journal of the American VeterinaryMedical Association, 199, 1287-1297.
Kass, P. H., Barnes, W. G., Spangler, W. L., Chomel, B. B. and Culbertson, M. R.
(1993). Epidemiologic evidence for a causal relation between vaccination and
fibrosarcoma tumorigenesis in cats. Journal of the American Veterinary Medical As-
sociation, 203, 396-405.
Macy, D. (1994). Vaccine associated sarcomas. In: Proceedings of the 12th American College
of Veterinary Internal Medicine Forum, San Francisco, CA, pp. 854-856.
Matsubara, O., Tan-Liu, N. S., Kenney, R. M. and Mark, E.J. (1988). Inflammatory
pseudotumors of the lung: progression from organizing pneumonia to fibrous
histiocytoma or to plasma cell granuloma in 32 cases. Human Pathology, 19,
807-814.
Peiffer, R. L., Monticello, T. and Bouldin, T. W. (1988). Primary ocular sarcomas in
the cat. Journal of SmaU Animal Practice, 29, 105-116.
Preston-Martin, S., Pike, M. C., Ross, R. K., Jones, P. A. and Henderson, B. E.
(1990). Increased cell division as a cause of human cancer. Cancer Research, 50,
7415-7421.
Rudmann, D. G., Van Alstine, W. G., Doddy, F., Sandusky, G. and Janovitz, E. B.
(1994). Pulmonary and mediastinal metastases of a post vaccinal sarcoma in a cat.
In: 3 7th Annual Meeting of the American Association of Veterinary Laboratory Diagnosticians,
Grand Rapids, MI, p. 121.
174 r . D . Doddy et al.
Ryu, R. K. N., Bovill, E. G., Skinner, H. B. and Murray, W. R. (1987). Soft tissue
sarcoma associated with aluminum oxide ceramic total hip arthroplasty. Clinical
Orthopaedics and Related Research, 216, 207-212.
Schmeitzel, L. P., Loeffier, D. and Bass, M. C. (1986). Focal cutaneous reactions at
vaccination sites in a cat and four dogs. In: Proceedingsof the American Academy of
VeterinaryDermatology~American College of VeterinaryDermatology, 2, 39.
Schofield, D. E., Fletcher, J. A., Grier, H. E. and Yunis, E.J. (1994). Fibrosarcoma
in infants and children. American Journal of Surgical Pathology, 18, 14-24.
Zajicek, G. (1985). Inflammation initiates cancer by depleting stem cells. Medical
Hypotheses, 18, 201-219.

Received,January 5th, 1995 ]

Accepted, November 3rd, 1995]