[Histonet] Formalin fixation time for breast specimens

[Morken, Tim]

Nice wrench, eh?

They point out that they used known high expressing tumors to do this experiment. High expressors will stain with almost anything, even poor AR or weak antibody. So, the question has to be: what about low expressors? They conclude that the fixation paradigm should be rethought (now 6 - 72 hours), but I don't think the experiment is complete without testing the full range of expression. 

Oyama, et al (Breast Cancer 14:182-188, 2007) was experimenting with over-fixation (up to 3 weeks) but started at 3 hours as the "under fixation" for comparison. In the one low-expressing case his group tested they got negative (Allred score) at 3 hours fixation, +2 Allred at 6 hours and +3 Allred at 24 hours. But for higher expressors (6 Allred and up) they also found that there was either no difference or a slight increase in Allred score when fixation was 3, 6 or 24 hours, or even 3 weeks. 

Considering that pathologists will often score ER as positive when the percentage score is between 1% to 10% expression, that does not leave much room for error if under-fixation causes false negatives. 

And this paper does not even mention the parallel testing that is normally done on the same tissue block for PgR and Her2. These days that should be factored into any discussion about fixation of breast samples. Or shall fix a different sample for each test (may happen!!). 

They also used a different processing method (rapid sakura microwave) than the method used in the papers they compare against. Not necessarily bad, but they did not control for the different methods with parallel testing. They also did not control for their criticism of another paper storing samples in 100% alcohol after fixation, but before processing (to enable processing all samples at once). I'm not aware of any evidence either way that storage in 100% alcohol up to 7 days will affect ER staining. These authors imply that it did affect staining. 

I suggest reading the accompanying editorial by Neal Goldstein after reading the paper itself.


Tim Morken
Supervisor, Histology / IPOX
UCSF Medical Center
San Francisco, CA  
 

-----Original Message-----
From: histonet-bounces <@t> lists.utsouthwestern.edu [mailto:histonet-bounces <@t> lists.utsouthwestern.edu] On Behalf Of Richard Cartun
Sent: Thursday, May 06, 2010 8:58 AM
To: Histonet
Subject: [Histonet] Formalin fixation time for breast specimens

There has been a lot of discussion recently regarding recommendations for formalin fixation of breast specimens.  If you are interested in this topic please read the following article published in the May 2010 issue of the American Journal of Clinical Pathology by Ibarra JA, et al., "Fixation time does not affect the expression of estrogen receptor".

Richard

Richard W. Cartun, Ph.D.
Director, Histology & Immunopathology
Director, Biospecimen Collection Programs
Assistant Director, Anatomic Pathology
Hartford Hospital
80 Seymour Street
Hartford, CT  06102
(860) 545-1596 Office
(860) 545-2204 Fax



_______________________________________________
Histonet mailing list
Histonet <@t> lists.utsouthwestern.edu
http://lists.utsouthwestern.edu/mailman/listinfo/histonet