|
Background
Some
tumours, notably carcinoma of the breast and prostate, are often
responsive to hormones, a property which has been exploited by the use of
endocrine surgery and medically using drug therapy to influence hormone
levels or inhibit the effects of hormones on tumour cells.
Steroid
hormones bind with high specificity and affinity to intracellular
receptors. These steroid receptors belong to a 'superfamily' of proteins
whose function is to control the transcription of a repertoire of other
cellular genes. 
Steroid receptors such as oestrogen andprogesterone receptor are located
in the cell nucleus. Hormone is believe to diffuse or be transported to
the nucleus where a steroid-receptor complex is formed with receptor
dimerization. Some of the genes regulated by steroid receptors are
involved in controlling cell growth and it is currently believed that
these effects are the most relevant to oestrogen receptor influences on
the behaviour and treatment of breast cancer.
Approximately
30% of unselected patients with breast cancer will respond to hormone
therapy such as oophorectomy (or chemical castration) or tamoxifen
treatment. The demonstration that radio labelled oestradiol bound to some
breast cancer specimens and that this effect was related to response to
hormone ablation
led to the development of hormone receptor assays directed at
identification of patients suitable for hormone therapy. By assay of
oestrogen receptor status alone, using the standard radioligand binding
assay on tissue cytosol samples, a response is seen in between 50 and 60%
of patients with oestrogen receptor positive tumours, in contrast to a
<10% response observed in patients with oestrogen receptor negative
tumours.
The threshold for designation of oestrogen receptor positivity is usually
10 fmol/mg cytosol protein. Levels of response are recognized to increase
to over 80% in patients with tumours having high receptor levels of
several hundred or more fmol/mg protein. Prediction of response can be
refined further by combining oestrogen receptor assay with progesterone
receptor assay.
Patients with ER positive PR positive tumours have a 78% response, those
with ER positive PR negative a 34% response, those with ER negative PR
positive a 45% response and ER negative PR negative tumours have a 10%
response.
It should be noted that these data relate to ligand binding assay systems
and results from combined ER and PR immunocytochemical assay may differ.
Assay
methods
The
cytosol ligand binding assay has until recently been the standard assay
method but has a number of disadvantages. In particular it requires a
relatively large amount of tissue homogenate and is affected by bound
oestrogen receptor from high endogenous levels of oestradiol in
premenopausal women. The development of monoclonal antibodies specific for
the receptor protein
in the 1980's led the way to both enzyme immunoassay
and immunocytochemical assay development.
More recently, the immunocytochemical methods, despite being less readily
quantifiable, have superseded both the ligand binding and enzyme immuno
biochemical assays, as they require less tissue, allow formal histological
assessment thereby reducing sampling error,
and may be used on very small samples such as fine needle aspirates.
All these forms of assay correlate well
but there is some evidence that there is a closer association with
response to hormone therapy using immunocytochemical assay results
probably because of avoidance of sampling error and lack of influence of
tumour cellularity and type.
Immunocytochemical
Assay
The
early immunocytochemical methods developed required frozen material but
more recently the antibodies have been successfully applied to routinely
processed formalin fixed tissue. Initially many of the methods employed
required enzyme predigestion and overnight incubation of the tissue
sections.
Microwave antigen retrieval has been found to enhance immunohistochemical
staining by several authors
and is becoming increasingly popular.
These
techniques allow successful use of antibodies such as 1D5 on tissues
routinely processed in any histopathology laboratory giving the potential
for evaluation of oestrogen receptor status as part of the process of
histological assessment of breast carcinoma. In our centre this method
has now been adopted for all cases of primary breast carcinoma allowing
inclusion of oestrogen receptor status as part of the standard
histological report.
Assessment
of ER status using the 1D5 antibody with microwave pre-treatment of
tissues gives comparable results with other means of assessment but has
many advantages particularly its applicability to routine histology
samples,
as well as methodological ease and speed. The microwave technique is
technically easy and rapid to perform, not requiring an overnight
incubation procedure which may be seen as a disadvantage of the methods
described for the H222 antibody. 1D5 is our favoured antibody for these
reasons. It gives an assessment of ER status which correlates well with
the assessment made using Abbott H222 on both frozen and formalin fixed
tissue.
Oestrogen
receptor is a thermolabile unstable nuclear protein which is water soluble
and has a short half life after surgical removal of tumour. Most tissue
fixatives can be used to preserve receptor reactivity if microwave
predigestion is used. It is essential however to ensure that rapid
fixation occurs and tumour specimens should be incised immediately after
resection to ensure rapid penetration of fixative.
Controls
Use
of control tissue is essential in hormone receptor assays particularly
because of the risk of false negative classification. Positive control
tissues should include not only a block of a known strongly positive
tissue but also a block of tissue showing weak expression to ensure
sensitivity is maintained; ideally the test block should include normal
breast lobules and ducts to provide an internal control population of
cells since a proportion of these should show positive reactivity. Use of
internal control cells in this fashion protects against the effects of
poor fixation.
Assessment
of staining
Oestrogen
and progesterone receptor are located in the nucleus of breast epithelial
and carcinoma cells. There is currently no internationally accepted
scoring system for hormone receptor immunocytochemical assay. The
proportion of tumour cells showing positive reactivity (Abbott
Laboratories Datasheet), their intensity of reactivity, combinations of
both of these
and a simple categorical
have all been promoted. The staining methods currently used with
antibodies such as 1D5 have been reviewed
and assistance is available from the commercial suppliers of these
antibodies.
In
our laboratory we currently provide a result in the form an H-score and
the percentage of tumour cells showing reactivity. The H-score is based on
a summation of the proportion of tumour cells showing different degrees of
reactivity; no reactivity =0, weak =1, moderate =2, strong =3. This gives
a maximum total score of 300 if 100% of tumour cells show strong
reactivity. An example is shown below:
|
Category
|
%
cells
|
Score
|
|
0
|
10
|
0
|
|
1
|
30
|
30
|
|
2
|
50
|
100
|
|
3
|
10
|
30
|
|
Total
|
100
|
160
|
|
Conclusion:
|
Oestrogen
Receptor status: Positive
|
|
|
% cells
staining = 90
|
|
|
H score = 160
|
An
alternative simplification of this method has been proposed by the San
Antonio group .
This is based on an estimate of the proportion of positive tumour staining
cells (0, none; 1, <1/100; 2, 1/100 to 1/10; 3, 1/10 to 1/3; 4, 1/3 to
2/3; and 5, >2/3) combined with an average intensity score (0, none; 1,
weak; 2, intermediate; 3, strong). Summation of scores of 0-2 are regarded
as negative; scores of > 3 positive. Adjuvant endocrine therapy has no
survival benefit in patients with negative tumours.
Clinical
application
Use
of semiquantative methods, such as the H-score, which produce a numerical
score influenced by the intensity of reactivity, have an association with
the amount of receptor present as assessed by biochemical methods.
Informed clinicians are now beginning to use not just a standard result
based on a common cut off point or threshold but some form of added
quantitation of known sensitivity and specificity. It is possible to
assess the sensitivity and specificity of an assay at different cut off
points giving an ability to choose differing thresholds for different
clinical situations.
For example, when selecting for adjuvant hormone therapy sensitivity is
required and a low threshold may be appropriate; in contrast elderly
patients or patients with advanced tumours being selected for
primary/first line hormone treatment require specificity and a higher
threshold is needed.
None
of the hormone receptors assays are absolute in their ability to predict
response. A proportion (0-10% in different studies) of ER negative tumours
are found to respond to hormone therapy.
It has been postulated that oestrogen receptor expression is stimulated by
low levels of available oestrogen
and it is possible that down regulation of the ER gene to
immunohistochemically undetectable levels may occur in some tumours due to
high circulating levels of endogenous oestrogen. This does not necessarily
mean that they will not respond to hormone therapy which has itself been
shown to up-regulate ER expression in normal breast tissue.
The
observation that a proportion of tumours exhibit heterogeneous reactivity
is well documented.
The reasons for its occurrence are not clear. It may be due to
heterogeneous expression of ER by tumours as described by Poulson.
Similar heterogeneity is observed in normal tissue
and may be related to different physiological states within the cell
population.
|
