Tonia, obviously you have to validate your systems so you need to do some validation ahead of time with non-patient samples. You can validate your instruments and reagents on tissue arrays (you can even get commercial validated arrays for ER, PR and Her2). I suggest Pantomics (www.pantomics.com) for excellent quality arrays of all kinds.
To validate the actual test, once you start up you will have to do some parallel testing by testing in-house and sending paired samples out to be tested. Then document the results of each. When you can show concordance of results over a set of samples (for most antibodies, 10 or more, 5 pos and 5 negative) that test can be considered validated. For ER, PR and Her2 you will need to have concordance on many more samples (25 - 100). I suggest, if possible, taking extra tissue from patient samples and making either tissue arrays or sausage arrays to do this testing. It will drastically lower the number of slides you have to test in house and to send out to be tested. It may be possible to partner with another lab to do this testing. As part of proficiency testing you can ask other labs for samples (wet tissue or unstained slides) to validate your results against theirs. Some labs are looking for partners to do this kind of testing. You then each validate each others tests. Tim Morken Supervisor, Histology / IPOX UCSF Medical Center San Francisco, CA -----Original Message----- From: histonet-boun...@lists.utsouthwestern.edu [mailto:histonet-boun...@lists.utsouthwestern.edu] On Behalf Of tonia.richm...@gracepathology.com Sent: Thursday, May 06, 2010 8:28 AM To: McMahon, Loralee A Cc: histonet@lists.utsouthwestern.edu; thisis...@aol.com Subject: RE: [Histonet] Responses to IHC CAP Validation question If you are a brand new lab, how do you validate IHC if you are no= yet receiving patient specimens? Can the validation be done on cont rol tissues only? Sincerely, Tonia Richmond, AS, HT (ASCP) Chief Operations Officer =aboratory Grace Pathology PH: (501) 765-7367 Email: =]tonia.=chm...@gracepathology.com -----histonet-boun...@lists.utsouthwestern.edu wrote: ----- <=ONT> To: "thisis...@=l.com" <thisis...@aol.com>, "histonet@lists.utsouthwestern.edu" < ;histonet@lists.utsouthwestern.edu> From: "McMahon, Loralee A" <lo=lee_mcma...@urmc.rochester.edu> Sent by: histonet-boun...@lists.u=outhwestern.edu Date: 04/28/2010 02:01PM Subject: RE: [Histonet] Re=onses to IHC CAP Validation question Any inspection that I have under=ne we have used the 25 to 30 case rule. Except for the Er/Pr//Her-2=e use closer to 50 cases. We also use a TMA to make our live=asier. The TMA contains known positives and known negatives. I=ases of t-cell or b-cell markers or cytokeratins. 25 to 30 cases i=asy. But when you are validated for more hard to find markers (SV-@) then fewer cases is acceptable. We always throw in a slide that w=now will not stain for sv-40 like a tonsil - then you can say it has spe=ficity. Any inspector that I have come across is usually understanding= this. But I am sure that there are exceptions to this.........esp ecially if they are not familiar with immunohistochemistry. Loralee McMahon, HTL (ASCP) Immunohistochemistry Supervisor Strong M=orial Hospital Department of Surgical Pathology (585) 275-7210 ______________________ 5F__=F______________ From: histonet-boun...@lis=.utsouthwestern.edu [histonet-boun...@lists.utsouthwestern.edu] On Behalf= thisis...@aol.com [thisis...@aol.com] Sent: Wednesday, April 28, 201:47 PM To: histonet@lists.utsouthwestern.edu Subject: [Histonet] R=ponses to IHC CAP Validation question The following is one respone=ec'd: 1. I asked CAP who told me that they do not currentl=ave a guideline on validating but that they recommend what is in t= following book: Quality Management In Anatomic Pathology, Promoting P=ient Safety Through Systems Improvement and Error by Raouf E. Nakhl=, MD & Patrick Fitzgibbons, MD editors sold by CAP ! Chapter 8-=ality Management in IHC That is what we follow. I. Get a new antib=y and optimize it with your positive control. II. Once optimized you n�d to run it on cases expected to be positive (how many?) "a suffien=ize ..." III. Must also be run on cases expected to be negative. (how=ny? IV. In a situation where you cannot expect a lot of cases or such a case has never been presented in your lab, then you must say just =at. (ex. some of the hormones we just use a pituitary) ___________________ ______________________ 5F__=F__ Histonet mailing list Histonet@lists.utsouthwestern.edu =]http://lists.utsouthwestern.edu/mailman/listinfo/histonet ______________________ 5F__ ______________________ Histo=t mailing list Histonet@lists.utsouthwestern.edu [3]http://l=ts.utsouthwestern.edu/mailman/listinfo/histonet References 1. 3D"mailto:tonia.richm...@gracepathology.com" 2. 3D"http://lists.utsouthwestern.edu/mailman/listinfo/histonet" 3. 3D"http://= _______________________________________________ Histonet mailing list Histonet@lists.utsouthwestern.edu http://lists.utsouthwestern.edu/mailman/listinfo/histonet
_______________________________________________ Histonet mailing list Histonet@lists.utsouthwestern.edu http://lists.utsouthwestern.edu/mailman/listinfo/histonet