[Histonet] RE: New Lab Setup

[Bryan Szpunar]

Hi Jim,
While I do not have quite the length of time in the field that you do, I do
have been through this several times and have some thoughts.

First, to your question about how many samples is acceptable, I would say
you should somewhat base this on your acceptability criteria. For example,
if acceptability for your validation is 95% of the specimens being "overall
acceptable", you should do either 20 or 40 (for your scope, I would pick 20
personally), but NOT somewhere in between those numbers. The reason for
this is you gain nothing in the land of percentages by doing, say 25. At
20, any more than one specimen that is "unacceptable" will throw you below
95%. At 40, you can have two outliers and still be within 95%.

That is just a hypothetical, but hopefully you get the gist. Essentially,
don't give yourself more opportunities to fail the validation if it gains
you nothing in the margin. If your application requires a more robust
validation, go for 40 (a la prognostic IHC).

I would think an IHC/histochemistry validation would be overkill if you are
validating your initial samples in parallel anyway. Just be sure to
document that the results of the specimens were comparable to each other. I
might have a different opinion if you were not using the exact same
processors and doing some form of rapid/microwave processing instead.

Regards,
Bryan





> Date: Wed, 18 Feb 2015 17:27:50 +0000
> From: "Vickroy, James" <jvickroy <@t> SpringfieldClinic.com>
> Subject: [Histonet] New lab setup
> To: "histonet <@t> lists.utsouthwestern.edu"
>         <histonet <@t> lists.utsouthwestern.edu>
> Message-ID:
>         <
> 9B1A1501A800064397369BD8072E6BCA984027 <@t> E2K10DB.springfieldclinic.com>
> Content-Type: text/plain; charset="us-ascii"
>
>
> I am working on setting up  an adequate validation study for tissue
> processing at a new lab.  I want it to be adequate but also not too labor
> intensive or costly.  I have been in Histotechnology for over 36 years and
> have done this before in the lab I can from.   Unfortunately sometimes I
> make things more complicated than others.  I realize the standards for
> testing new protocols and antibodies are defined in the CAP checklist but
> routine tissue processing  just says to document the procedure and to run
> parallel samples as a blind study.
>
> We are going to run only GI biopsies at first and I had planned on running
> parallel samples with the department from my  previous employer.   I also
> thought that down the road we may have more than a rapid biopsy program so
> I should do a few larger samples at a regular processing schedule to
> validate both schedules (rapid and normal).  I also thought that I should
> pick out a couple of blocks for  representative special stains and IHC
> stains tha we would compare even though the new lab will send everything to
> the old lab for special stains and IHC stains.
>
> We have two processors and know I will also have to run a small validation
> of each new processor.  I figured as I was validating the tissue processing
> programs I would also be validating the first tissue processor.
>
> Both of our two processors will be VIP6(s) and the machines from my former
> employer were also VIP6(s) which were of course validated.
>
> Can anyone share how many samples are adequate for tissue processing
> program validation and new processor validation? In other words what have
> you done and what was accepted by CAP since the wording of the question
> does not state any particulars?   Also can you tell me if you think the
> special stains and IHC stains comparison is necessary, given that all of
> the stains will be done at the old lab and the only difference will be
> where the tissues were processed.
>
> My original design of the study had 25 parallel samples but I am wondering
> if that is overkill.
>
> Thanks
>
> Jim Vickroy
>
> Jim Vickroy
> Histology Manager
> Springfield Clinic, Main Campus, East Building
> 1025 South 6th Street
> Springfield, Illinois  62703
> Office:  217-528-7541, Ext. 15121
> Email:  jvickroy <@t> SpringfieldClinic.com<mailto:
> jvickroy <@t> SpringfieldClinic.com>
>
>
> This electronic message contains information from Springfield Clinic, LLP
> that may be confidential, privileged, and/or sensitive. This information is
> intended for the use of the individual(s) or entity(ies) named above. If
> you are not the intended recipient, be aware that disclosure, copying,
> distribution, or action taken on the contents of this information is
> strictly prohibited. If you have received this electronic message in error,
> please notify the sender immediately, by electronic mail, so that
> arrangements may be made for the retrieval of this electronic message.
> Thank you.
>