[CANCER RESEARCH 51. 3059-3061. June 1. 1991]
Advances in Brief
Localization of Parathyroid Hormone-related Protein in Breast Cancer Métastases:
Increased Incidence in Bone Compared with Other Sites'
Gerard J. Powell,2 Justine Southby, Janine A. Danks,1 Ross G. Stillwell, John A. Hayman,
Michael A. Henderson,4 Richard C. Bennett, and T. John Martin5
I'nirersily of Melbourne Department of Surgery ¡(j.J. P., M. A. //., R. C. B.¡,St. lïncenl'sInstitute of Medical Research and L'niversity of Melbourne Department of
Medicine [J. S.. J. A. D., T. J. M.], and Department oj Anatomical Pathology /R. G. S.J, St. I 'incenl's Hospital, Melbourne, I'¡doria3065, Australia, anil L'niversity of
Melbourne Department of Pathology, Repatriation Cenerai Hospital, Heidelberg, Victoria 3081, Australia [J. A. H.J
Abstract
Parathyroid hormone-related protein (FI'HrP) has recently been iden
tified in 60% of a series of primary breast cancers. The detection of a
bone-resorbing factor in tumors with a propensity to metastasize to bone
prompted study of PTHrP in breast cancer metastasis. PTHrP »as
localized by immunohistology in 12 of 13 (92%) breast cancer métastases
in bone and in 3 of 18 (17%) métastases in non-bone sites. The statistical
difference »ashighly significant (P < 0.0001). Production of PTHrP as
a bone-resorbing agent may contribute to the ability of breast cancers to
grow as bone métastases.
Introduction
PTHrP6 is recognized as the principal agent responsible for
the increased bone résorptionand reduced calcium excretion
accompanying the humoral hypercalcemia of malignancy (1).
In addition, it has been identified in normal human skin (2), as
well as in a series of squamous cell cancers in normocalcemic
subjects (2). We have demonstrated PTHrP immunohistochem-
ically in 60% of primary tumors from 101 unselected patients
with breast cancer (3). Seven of these patients have developed
bone métastasesin the early period of follow-up and all have
had primary tumors that have stained positively for PTHrP.
These observations and the frequency with which breast cancer
metastasizes to bone (4) prompted the present investigation, in
which immunohistochemistry has been used to determine the
incidence of PTHrP staining in a retrospective series of breast
cancer métastasesin bone and non-bone sites. The metastatic
process is a complex, nonstochastic sequence of essential steps,
in which a metastatic cancer cell must interact with the microen-
vironment of the target organ in a manner that allows tumor
growth if this process is to be completed. PTHrP is a powerful
stimulator of bone résorption(5). This has led us to consider
whether PTHrP produced by breast cancers might contribute
to their ability to establish and grow in bone.
Materials and Methods
Paraffin blocks of breast cancer métastaseswere obtained from
pathology archives for all biopsy-proved métastases. Those arising from
Received 2/26/91; accepted 4/11/91.
The costs of publication of this article were defrayed in part by the payment
of page charges. This article must therefore be hereby marked advertisement in
accordance with 18 L'.S.C. Section 1734 solely to indicate this fact.
1Supported by the Anti-Cancer Council of Victoria and the National Health
and Medical Research Council of Australia.
2 Recipient of a Cockburn Research Scholarship of the Royal Australasian
College of Surgeons Foundation and the R. J. Fletcher Research Scholarship of
the University of Melbourne.
' Recipient of an Australian Research Council Individual Fellowship.
4 Recipient of a Florance Cancer Research Fellowship of the Royal Australa
sian College of Surgeons Foundation.
* To whom requests for reprints should be addressed.
6 The abbreviation used is: PTHrP. parathyroid hormone-related protein.
breast primaries were positively identified and métastasesto regional
lymph nodes and local recurrences were excluded from study. Post
mortem specimens were also excluded inasmuch as these have previ
ously been found to be unsuitable for immunohistochemistry due to the
effects of autolysis. The patient's records were used to confirm the
findings and to provide relevant clinical data. A total of 42 biopsy-
proved, distant breast cancer métastaseswere identified from 40 pa
tients over the period 1986 to 1990 for which there were complete
records. Of these, sufficient material was available for immunohisto-
chemical study in 31 cases.
Tumor Samples. In the case of soft tissue métastases,the tumor
tissue had been formalin fixed, routinely processed, and embedded in
paraffin. Tissue samples taken from bone had been decalcified in citric
(2.5%)/formic acid (8.75%) solution (6) following fixation before proc
essing and paraffin embedding. Sections (5 urn thick) at either end of
those cut serially were stained in hematoxylin and eosin and reviewed
by a pathologist to ensure that representative tumor was present in the
sections.
Immunohistochemistry. The primary antibody was a polyclonal anti-
serum raised against the amino-terminal portion of the PTHrP mole
cule by injecting synthetic PTHrP(l-34) into New Zealand White
rabbits. The antiserum (255.5) was characterized as described previ
ously (2. 3) and demonstrated no detectable cross-reactivity with para
thyroid hormone. This antiserum compares favorably with other
PTHrP antisera used previously in ¡mmunohistochemistry (2, 7). The
peroxidase-antiperoxidase method of tissue antigen localization used
was modified from that described by Danks et al. (2), based on the
technique of Sternberger el al. (8).
Controls. To ensure specificity of staining for PTHrP, antibody and
method controls were performed (2) including (a) deletion of alternate
layers of the antibody sandwich (primary antiserum, secondary anti-
serum, and PAP complex) and (b) preabsorption of anti-PTHrP(l-34)
antiserum overnight with 0.5 mg/ml synthetic PTHrP(l-34) and hu
man PTH(l-34), respectively, prior to immunohistochemistry on skin
and breast cancer specimens previously found to stain positively for
PTHrP. To ascertain that the décalcificationtechnique did not alter
PTHrP staining, a fresh parathyroid adenoma was placed in décalcifi
cation solution, processed, and used as a positive control for immuno-
staining (7). All experiments included samples of human skin as a
positive control (2) and the substitution of primary antibody with
preimmune rabbit serum as a negative control.
Evaluation of Staining. The stained sections were reviewed by three
independent observers and compared with controls and hematoxylin
and eosin-stained slides. Tumor métastaseswere called positive if any
of the tumor cells stained brown. Sections involving bone marrow were
reviewed by a pathologist to ensure that marrow stem cells were not
confused with tumor. There was concordance of opinion in all cases
except one specimen from bone marrow which was equivocal and
therefore judged to be negative.
Results
Of the 31 specimens obtained for study. 13 were from bone
and 18 were from various non-bone sites (Table 1). The 13 bone
métastasesincluded 6 diagnostic trephines, reamings from 4
3059
Research.
on August 31, 2017. © 1991 American Association for Cancer cancerres.aacrjournals.org Downloaded from