I would like to offer the following information regarding HER2 immunostaining:

1.  We currently use DAKO's pAb "c-erbB-2" to demostrate the Her-2/neu
"protein" (not Her-2/neu which is the gene) in formalin-fixed,
paraffin-embedded tissue.  We use the antibody at a dilution of 1:20,000
overnight with LSAB+ detection (DAKO) and DAB+ (DAKO) chromogen.  The
antibody can be diluted 1:2,000 for 30' incubations.  Please keep in mind
that this is an old lot that may not be available anymore.  Yes, there are
many commercially available antibodies (both polyclonal and monoclonal)
that yield satisfactory results; however, we have selected this one based
on our experience with it over a 5 year period where we have stained over a
thousand cases.  In addition, it is the antibody that is used in DAKO's
HercepTest kit.

2.  I have compared DAKO's HercepTest kit with our "in-house" method on a
small series of cases and can honestly tell you that their kit works very
well.  I like to think that our immunoperoxidase stains are "first class",
but the HercepTest kit produced slides that were cleaner and easier to
interpret (most likely due to the non-avidin/biotin detection system).  We
also compared the immunostaining results to molecular analysis using a
differential PCR method and there was good correlation. The results of this
study have been accepted for presentation at next year's International
Academy of Pathology in San Francisco.  Yes, the kit is expensive, but the
decision to use the kit may not be entirely ours; pressure to use it may
come from clinical oncologists and 3rd party payors.  Also, the kit gives
small laboratories the capability to perform this test since all major
reagents (as well as positive control cell lines) are included.  Maybe a
new CPT code can be invented for this test (i.e., thin prep cytology)?

3.  Yes, it's true that one can obtain good immunoreactivity for HER2
without heat-induced epitope retrieval (especially for 3+ tumors), but low
to moderate level expression may be "masked" following harsh formalin
fixation and, therefore, will benefit from HIER.  We perform HER2 staining
on tissues from many different laboratories and have found HIER to reduce
interlaboratory variation with regard to fixation and tissue processing.
Obviously, tissues fixed in non-formalin fixatives or combination fixatives
(e.g., zinc-formalin) may require a modified pretreament.

4.  With regard to HER2 immunoreactivity of benign tissue elements I would
like to say that we occassional see it, but it is rare.  Most of the
staining in benign epithelium that we see is due to endogenous biotin and,
as such, should be present in your negative control.  This is where the
HercepTest kit "shines" since it does not use avidin-biotin detection.  I
recently read in the book "HER2" that normal/benign cells may express up to
50,000 receptors (below the threshold of IHC detection), but overexpressing
tumors can have as many as 5 million receptors!!!  The book "HER2" is not a
scientific reference, but the author must have researched this information.

5. I am willing to help those who are having difficulties with HER2
immunostaining by providing positive control sections as long as the
requests are not overwelming.  I may be able to provide stained sections as
well.  Also, feel free to e-mail, telephone, or fax me with questions.
This is a very important prognostic marker and it is imperative that we do
everything possible to make sure that we are accurately identifying
patients for appropriate adjuvant chemotherapy and/or Herceptin therapy.

I apologize for this lengthy message, but I realize that this is a "hot"
topic which has generated much interest and many questions.  I hope this
information proves useful.

Richard W. Cartun, Ph.D.
Director, Immunopathology
Co-Director, Histology
Hartford Hospital
Hartford, CT  06102

(860) 545-1596
(860) 522-0045 Fax
e-mail: rcartun@harthosp.org