Ultrasonography of Osteosarcoma in Dogs
Sonography Atlas of Dogs

Summary
Osteosarcoma,   the  major   form  of  bone
cancer  in  dogs,  is  reviewed.   Incidence  rates
relative  to  breed,  age,  and  sex characteristics
are outlined.  Prediliction sites are also stated.
The  clinical,  radiographic,  and  metasta-
tic   characteristics   of   osteosarcoma   are   ex-
plained.
Theories  on  etiology  including  work  by
Brodey,   Wolke   and   Nielson   are   discussed.
Alternate treatment regiments  are examined,
including  an  indepth  look  at  a  case  at  the
Iowa    State    University    Clinic    which    was
treated using a combination of ostectomy and
allograft, local hyperthermia,  bleomycin, and
levamisole.

A  Review of Osteosarcoma
Osteosarcoma  is  the  major  form  of  bone
cancer  seen  in  the  dog. 6,10,16,22,30  One  study
showed that 85%  of the primary bone tumors
seen    in    dogs.  were    osteosarcoma,    while
chondrosarcomas,  the  second  most  common
primary  bone  tumor,  occurred  only  10%  of
the   time.6
Canine   osteosarcoma   is   nearly
always  malignant,   as   compared   to   a   50%
malignancy   rate   in   cats   and   a   generally
benign situation in cattle and in horses. 22
Several surveys have shown that large and
giant breeds have a much higher incidence of
osteosarcoma    and    are    at    a    significantly

These  sites  are  the  proximal  humerus,  distal
radius,  proximal and  distal femur,  and  prox-
imal  and  distal tib~a.6,7,I6,I7,30In  the  Wolke
and  Nielson  survey,  the  distal  metaphysis  of
the  radius  was  the  most  common  site,  with
23%    of   the    total    cases.    The    proximal
metaphysis  of the  humerus  was  second in  in-
cidence, with 19% of the cases.30
In  some  circumstances  osteosarcoma  may
originate  in  tissues  other  than  bone.  It  has
been reported in  the  esophagus of a  dog,  ad-
jacent  to  a  chronic  lesion  produced  by  the
spirurid   worm,   Spirocerca   lupi.   A   second
extra-osseous  site  is  in  a  mixed  tumor  of  the
mammary gland. 26
One  of the  first  clinical  signs  of osteosar-
coma  in  the  metaphyseal   region  of  a   long
bqne is lameness.  One to two weeks later there
is  generally  a  cool,  palpable  swelling  in th
area of the lesion.  Eventually there is a visible
enlargement  at  the  site  of  the  lesion  that  is
warm  and  painful  due  to  stretching  of  the
periosteum. 22
Radiographically,   this   tumor   is   usually
found  at  the  extremity  of  a  long  bone  and
produces a radiolucent enlargement arising in
the  metaphysis  which  erodes  the  pre-existing
calcified bone of the cortex.26
The destructive process may be  restricted to  the  medulla,  but usually involves the cortex as well,  by the time the tumor is manifested clinically. 27
In     addition     to     cortical     destruction,
another  type of radiographic  change that  oc-
curs with osteosarcoma  is periosteal response.
The degree of periosteal reaction does not de-
pend  on  the  degree  of cortical  destruction.I7
This  periosteal  response  can  lead  to  a  large
soft  tissue  mass  contiguous  to  the  bone.  This
soft  tissue  swelling  around  the  osteosarcoma
lesion is  also  related  to  reactive fibroplasia  in
the  subcutaneous  and  intramuscular  tissues,
which   leads   to   impaired   circulation   and
edema.27
All    osteosarcomas    are    collagenoblastic
tumors   in   which   the   collagen   fibers    are
organized  into  varying  amounts  of  osteoid,
bone,  and cartilage. 22 Depending on which of
these  components  is  dominant,   three  major
subtypes     are      recognized:      osteoblastic,
fibroblastic,  and chondroblastic. 12
The  critical,  identifying  characteristic  of
cells  of  osteosarcQma  is  their  ability  to  pro-
duce   ostedid.26
Osteoid   is   the   collagenous matrix  of  bone,  the  primary  product  of  the metabolic     activity    of    osteoblasts,     which

radius  has  a  much  higher  growth  potential
than   the  proximal  radius  and  also  has  the
higher  incidence  of osteosarcoma  of the  two.
A  similar  situation  exists  with  the  proximal
humerus,   which   exceeds   both   the   growth
potential  and  osteosarcoma  incidence  of  the
distal  humerus.   Brodey  continues  with   the
correlation by showing that the proximal and
distal femur and the proximal and distal tibia
have  nearly  equal  growth  potentials  and   a
nearly equal incidence of osteosarcoma.
Brodey  hypothesizes  this  rapid,  maximal
growth  at  the  metaphysis in giant breed  dogs
leaves  behind  small  foci  of  retained  hyaline
cartilage.  These  foci  have  not  been  seen  in
smaller dogs.  These  foci  may  serve  as  sites of
origin for later tumor growth. 6
Wolke  and  Nielson  consider other factors
to  be  involved  in  the  etiology.  They  suggest
that weight bearing stresses on the metaphysis
of the  long  bones  lead  to  the  development  of
osteosarcoma. They suggest that the re
higher  incidence  in  the  pectoral  limbs  versus
the pelvic limbs is directly proportional to the
relative weight distribution  between the front
and back legs.  They also site the increased in-
cidence  in  heavier  dogs  as  further  proof  of
their  weight-bearing  stress  theory.30  Another
study basically agrees with this theory,  stating
that  repeated  trauma  to  the  growth  plates in
young  giant  breed   dogs  (caused   by  weight
bearing  stresses),   may  partly  be  responsible
for the development of osteosarcomas at these
sites in later life. 14
Another  theory  concerns  the  relationship
of  healed   fractures   to   the   development   of
osteosarcoma.  Bennett,  Campbell and Brown
suggest  that  cartilage  cells  produced  during
the   healing  of  a   fracture   may  persist  long
after  the  fracture  is healed,  potentially form-
ing a  focus  for  neoplastic  development. 4 This
is  similar  to  the   Brodey  theory  of  retained
hyaline  cartilage  cells  providing  the  foci  for
tumor  growth,  differing only in  the  origin  of
the cartilaginous cells.

There  have  been  reports  of dogs  and  cats
that   have   developed   tumors   after   metallic
surgical   implants   were   used   to   treat   bone
fractures. 2
,25 Implanted metals may form cor-
rosive  products  such  as  metallic  salts  or  fine
particles.  The  animal's  response  to  metallic
implants   can   vary   from   inflammation   to
allergic  reaction  to  tumorogenesis.14
A  study of  S   clinical   cases   strongly   supported  
this theory.  AIlS  cases of osteosarcoma arose mid-

side  effects  from  cytotoxic  drugs  in  combina-
tion therapy. 15
Amputation of the diseased limb has been
the  treatment  of choice  for  several years,  but
even   with   amputation   the   survival   rate   is
poor.   Brodey   points   out   that   there   is   no
baseline  data  for  long-term  survival  of  dogs
with   osteosarcoma   that   were   not   treated.
There   are   known   cases   where   dogs   with
osteosarcoma  did  survive  without  treatment,
and it is therefore concluded that not all long-
time survival can be credited to the treatment
under  consideration,   as  some  of  those  dogs
may have lived anyway.5
Chemotherapy    and     immunostimulants
have  been  recent  developments  in  the  fight
against  osteosarcoma.  Methotrexate,  vincris-
tine sulfate,  doxorubicin,  cyclophosphamide,
and bleomycin are some of the many different
chemotherapeutic  agents  that  have  been  or
are being tested.  Thus far  there is insufficient
data to determine if these drugs will be useful
or not.
The   same   observation   is   true   for   im-
munomodulators    such    as    BCG    (bacillus
Calmette-Guerin)    vaccine    and    levamisole.
There is some evidence that  BCG vaccine will
help  delay  metastasis  following  amputation
by   activating   macrophages   non-specifically
and  causing  them  to  recognize  and  destroy
malignant cells. 15.
2o However,  it has also been
demonstrated that BCG vaccine treatment,  at
best,  only  delays  and  does  not  cure  osteosar-
coma.  More specific immunotherapy needs to
be developed.
Thus,  even with therapy,  the prognosis for
a  dog with  osteosarcoma  is  very  poor.  In  one
study of 194 cases of osteosarcoma  85 %
were dead by 8 months,  and of the other 15%, only
one  dog  was  considered  cured. 6
In  another survey  of  65   cases  the   results  were 
similar, with  only  10.7%  of  the  cases  surviving  one
year past the time of diagnosis. 5 There is some
hope  that  combination  therapy  and  earlier
diagnosis will help to improve these figures.

Case Report-No. 582704-
An Experimental Treatment Regime
On June  24,  1980 a  7 year old,  80 pound,
mixed  (Collie-Shepherd)  spayed  female  dog
was  presented   to   the  Iowa  State  University
Small   Animal   Hospital   with   a   history   of
lameness in the left front leg of two days dura-
tion.  The  dog  had  been  in  a  kennel  for  10
days  and  had  not  been  lame  prior  to  board-

The  distal  one  third  to  one  half  of  the
radius,  including the  distal  epiphysis  and  ar-
ticular surface,  were  removed.  Frozen section
histopathology   revealed   that   the   proximal
end  of the  excised  bone  segment  was  free  of
the  tumor.  Two  screws  were  placed  through
the  proximal  radius  into  the  ulna   to  tem-
porarily    stabilize    the    elbow    joint.    Two
Hemovac  tubes  were  inserted  in  the  incision
site  for  local  hyperthermia  treatment.   The
perforated portion of each tube was placed in
the    defect    where    the    distal    radius   had
previously been.  The rationale for  adjunctive
local   hyperthermia   was   the   possibility   of
tumor extension into soft tissues and proximal
to  the  excised  portion  of the  bone  as  well  as
the  fragmentation  of  the  neoplasm  that  oc-
curred at the time of excision.
The  dog  was  given  one  half  bolus,  92mg
or approximately 2.71 mg/kg, of levamisole 3
hours   prior   to   surgery.    Hyperthermia   by
hydrothermic  perfusion  followed   closure  of
the  wound,  synchronized with  10  units of in-
travenous  bleomycin.  The  log for  the  hyper-
thermia  treatment  can  be  found  in  Table  2.
In   the   course   of   the   procedure   the   ther-
mometer was postitioned too close to the skin
on    the   far    side   of   the   leg.    The    tissue
temperature was thought to be too low during
the first  part of the  procedure,  when actually
it  was  probably  too  high.   Consequently  the
tissue readings were in error and the possibili-
ty of thermal burn to the leg was high.

The   dog  was  re-admitted   the   following
Monday,   July   28,   1980.   The   left   leg  was
radiographed  and  the  ulna  had  fractured  at
the  distal screw due to excessive activity while
home.  Bleomycin  and  levamisole  treatments
were  continued  as  called for  by the  protocol.
The   wound   cultured   negative   for   Pseudo-
monas  on  two  cultures,   two  days  apart,   so
systemic   antibiotics   were   discontinued   and
the   burn   was   treated   topically.   The   lym-
phocyte   transformation   test   showed   some
deterioration in the immune status of the dog.
A  urinalysis  and  CBC  were  normal  and  the
alkaline  phosphatase  level  was  still  elevated
with  a  207  lUll.   The  dog  was  again  sent

home  with  the  leg  encased  in  a  Robert-Jones
dressing for protection and support.
On  August  1,  1980  a  whole  cortical  bone
allograft from  a  St.  Bernard  cross  donor  dog
was used to replace the distal one third to one
half of the  radius.  The  radiocarpal joint  was
arthrodesed  and- both  ends  of the  graft  were
stabilized  with  Dynamic  Compression  Plates.
A   cancellous   bone   graft   from   the   greater
tubercle  of  the  humerus  was  packed  at  the
ends    and     around     the     allograph.     The
alignment  was  considered  excellent  on  post·
operative radiographs.  The  thermal burns on
the  lateral  side  of the  leg were  debrided  and
sutured as much as possible.

The  dog was  placed on Ecotrinb for  a  few
days post-operatively to decrease discomfort.
On   August   4,    1980   the   wound   again
cultured positive for Pseudomonas with only a
slight    sensitivity    to    gentocin.     Beginning
August  5,  1980  the  dog  was  treated  daily on
an out·  patient basis.  By August 12,  1980 the
wound  over  the   proximal  entry  site  of  the
infusion tube was starting to granulate in and
the   dog  was   showing  steady  improvement.
Lymphocyte   transformation   tests  continued
to  demonstrate  immunosuppression.   (Table
4) The bleomycin and levamisole therapy was
continued as ordered in the protocol.
The   lymphocyte   transformation   test   of
August   26,    1980    revealed   the   continued
immunosuppression.   (Table   4)   The   dog   is
bearing  nearly full  weight on  the  leg and  the
wounds have nearly closed,  but due to licking
by   the   dog,   constant   bandaging   and   con-
tinued   Pseudomonas   infection,   the   healing
process is slower than normal.

Another   advantage   of   hyperthermia   is
that  it  has  been  shown  to  increase  the  im-
munogenicity of some tumor cells,  perhaps by
unburying  some  of  the  cell  surface  antigens
from surrounding lipids. 24
It  also  seems  relevant  that,  as compared
to  surgical  removal  (which  eliminates  poten-
tial    antigens);    and    radiotherapy,    chemo-
therapy,    and    whole    body    hyperthermia
(which  suppresses  antibody  formation),  local
hyperthermia may cause  a  slow release of an-
tigens  with  no  inhibition,  and  possibly  even
an increase, in antibody formation. 24
Problems possible with local hyperthermia
are  cardiac  arrhythmias,   hepatic  and  renal
dysfunction,  low grade  fever  due  to  necrosis,
and cutaneous burns. 24
Levamisole was used in  this  case in an  at-
tempt to help restore the immune responses of
a      predictably      immunosuppressed      dog.
Though the mechanism of action is unknown,
it  is  well  understood  that  the  best  results  are
obtained  in  immunodeficient  patients.   The
drug  modulates  immune  function  at  2  to  3
mg/kg  of  body  weight.   At  higher  doses,  it
may actually suppress immune function. 8
Though  the  mechanism  of  action  is  un-
known,   levamisole   in   vitro   and   levamisole
therapy  in  vivo  correct  defective  motility  in
phagocytic   cells.   The   drug   also   stimulates
phagocytosis in cultured monocytes.8
Some   immunodeficient   patients   do   not
improve with levamisole treatment.  It may be
due  to  the  inability  of the  individual  tp  pro-
duce levamisole-induced serum factor needed
to  increase  lymphocyte funciton. 8  If the  lym-
phocyte  transformation  test  is  accurate  in  its
assessment   of  the   animal's   immune   status,
then  the  results  of  the  levamisole  therapy  to
date  is  discouraging,  as  the  dog  continues  to
be immunosuppressed.
The reliability of the lymphocyte transfor-
mation  test  is  a  controversial  matter.   Some
feel  it  is  a  good  prognostic  test,  while  others
do not.
9 The work on this project has assumed
the  test to  be reliable  and  will continue  to  do
so.  There are very few good ways to assess im-
munostatus  in  such  a  quantified  manner  as
with this test.
To  close  this  report  no  definite  conclu-
sions can be drawn from  this one clinical,  ex-
perimental  case  that  has  yet  to  run  its  com-
plete  course.   Additional  cases  treated  using
this   therapeutic   protocol,    each   individual
drug  and  other drugs  as  well  as  controls,  are

20.   Owen   LN,   Bostock   DE,   Lavelle   RB:   Studies  on
therapy of osteosarcoma in dogs using BCG vaccine.
J Am  Vet RadiolI8:27-29,  1977.

21.   Peterson   LFA,    jones   jM,   Kelly   Pj,    Pease   GL:
Isolation of osteosarcoma cells from peripheral blood
after biopsy. Mayo Clin Proc 35:443-447,1960.
22.   Pool  RR:  Tumors of bone and cartilage,  Tumors in
Domestic   Animals,    2nd    ed.    Moulton   jE    (ed),
Berkeley,  University of California Press,  pp.  89-149,
1978.
23.   Schneider   PR,   Stowater   jL:   Pathologic   fractures
associated with skeletal metastasis of osteosarcoma in
a dog. JAVMA  175:61-64, 1979.
24.   Short  jG:  Hyperthermia  and  cancer:  a  brief review.
(a  bulletin)  BSD  Corp,  Salt  Lake  City,  pp.   1-13,
1978.
25.   Sinibaldi   K,    Rosen   H,    Liu   SK,    DeAngelis   M:
Tumors    associated    with    metallic    implants    in
animals.   Clin   Ortho  and  Relat  Res  118:257-266,
1976.
26.   Smith    HA,    Jones,    TC,    Hunt    RD:    Neoplasia,
Veterz"nary Pathology,  4th Edition. Lea and Febiger,
Philadelphia, pp 197-201,1972.
27.   Theilen GH,  Madewell BR:  Tumors of the skeleton,
Veterz"nary   Cancer   Medz"cine.    Lea   and   Febiger,
Philadelphia, pp 289-306,1979.
28.   Theilen   GH,   Pool   RR,   Park   RD:   Treatment   of
canine   osteosarcoma   for   limb   preservation   using
osteotomy,        adjuvant       radiotherapy       and
chemotherapy.  VM/ SA C 72: 179-183,  1977.
29.   Tjalma   RA:   Canine  bone  sarcoma:   estimation  of
relative risk as a function of body size. ]  Natl Cancer
Inst36:1137-1150,  1966.
30.   Wolke  RE,   Nielson  SW:   Site  incidence  of  canine
osteosarcoma. ]  8m Anim Pract 7:489-492,  1966.