Re: [Histonet] RE: negative IHC controls

From:Hadi Yaziji


--Apple-Mail-2--243860447
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=ISO-8859-1;
	format=flowed

Thanks for sharing your CV with us.

You seem to be fixated on how slides should be filed. It's quite easy=20
in our laboratory (and any other laboratory) to find the positive QC's,=20=

because they're filed chronologically (is there any other way to file=20
slides?). First you find out the date the case in question was run, and=20=

you go to the QC file cabinet and pull the positive QC's that belong to=20=

the same day. Negative QC's (and again, one slide per tissue block) are=20=

filed with the case.

As to your "evidence based good practice", please don't generalize. Two=20=

hospitals across the street from each other can represent two extremes=20=

of quality of care. Actually, I happened to have spent a month in a=20
prestigious Histopathology Department in London nine years ago. The=20
quality of IHC in their lab was superb, but it was miserable in the=20
hospital right across the street from them. The same applies to the US=20=

and any other country.

Finally, if my last email was unprofessional and insulting to you,=20
please accept my apology. Please remember that it was your unnecessary=20=

use of the terms "dam fool"  and "poor quality control by any=20
definition of the term" in your previous email that initiated my=20
response to you.

Best wishes,

Hadi Yaziji, M.D.
PhenoPath Laboratories


On Monday, October 13, 2003, at 03:00  PM, Nick Kirk wrote:

> Hadi
> =A0
> I am sorry but I must take issue with your very condescending and if I=20=

> may say so insulting tone.
> =A0
> For your information, I have over 18 years of experience doing IHC and=20=

> my lab does a considerable number of slides per day.
> I have set up the IHC service in two labs and have received much=20
> praise for doing it.
> I have also done extensive=A0lab based=A0trials work for both DAKO and=20=

> Novocastra here in the UK as well, so I think my credentials are=20
> better than most.
> I'm not at issue with the interpretation issue at all, that is indeed=20=

> a very skilled task that takes many years of study and experience to=20=

> become competent in.
> What I am taking issue with is the issues around clinical audit and=20
> best practice.
> Now I don't know what clinical audit systems you have in your=20
> establishment, but in ours it is regarded as poor clinical governance=20=

> i.e. bad practice by both myself, the Head Biomedical Scientist of the=20=

> department and all of my Pathologist colleagues not to have separate=20=

> controls per case.
> This is especially the case when you are reviewing tumour antigenicity=20=

> on samples that may come out of the archive.
> All our slides are stored in case order so to have positive controls=20=

> for each case actually makes it much easier the extract all the=20
> relevant slides per case for later review. Having slides for the same=20=

> case filed in separate places introduces the potential for error,=20
> especially when being initially filed.
> I know this to be true because it has happened to me on several=20
> occasions where slides filed in different places have been mis-filed=20=

> and become difficult to retrieve.
> In the UK we use a system of evidence based good practice and evidence=20=

> suggests that is precisely what we are doing.
> =A0
> I hope that now explains it to you.
> =A0
> Nick Kirk BSc (Hons)=A0FIBMS (I've got letters too)
> Head Biomedical Scientist
> Histopathology
> Hinchingbrooke Hospital
> Huntingdon
> Cambridgeshire
> England
> =A0
>
> -----Original Message-----
> From: histonet-admin@lists.utsouthwestern.edu=20
> [mailto:histonet-admin@lists.utsouthwestern.edu]On Behalf Of Hadi=20
> Yaziji
> Sent: 11 October 2003 16:22
> To: histonet@pathology.swmed.edu
> Subject: Re: [Histonet] negative IHC controls
>
> Dear Nick,
>
> Thanks for the email. You definitely misread my email. First, where in=20=

> my message did I advocate not to use negative and positive controls=20
> per case? Of course it's poor patient's care not to do that. Is it=20
> necessary to run + and - controls on every antibody? I don't think so=20=

> provided you have the necessary experience to interpret them. To=20
> address your other point about reviewing cases, we keep our positive=20=

> and negative controls indefinitely. You can always pull the controls=20=

> that pertain to the run of the particular day of the case in question.
>
> You will be 'dam fool' (I am copying your words) to run IHC if you=20
> have no experience. That's much more dangerous legally and ethically.=20=

> I've seen slides misinterpreted where the positive control is present=20=

> on the same slide.
>
> Bottom line: one + control per run is adequate (provided you have the=20=

> experience, which I'm not sure you have based on your reply). One -=20
> control per case is more than sufficient.
>
> I hope this addresses your concerns.
>
> Hadi Yaziji, M.D.
> PhenoPath Laboratories
>
>
> On Saturday, October 11, 2003, at 12:16 AM, Nick Kirk wrote:
>
> A cautionary note
> =A0
> I agree that false positives are rare but they do happen. We had a=20
> couple a while ago when it turned out that the Teflon coat on the=20
> probe of our DAKO Autostainer had perished and was carrying over=20
> reagents from one slide to another and also contaminating our negative=20=

> control solution with primary antibody.
> Also they are very useful in checking that there is no endogenous=20
> biotin activity (more common than people think) and also that your=20
> hydrogen peroxide solution hasn't gone off when quenching endogenous=20=

> peroxidase activity.
> =A0
> As for the argument about only one positive control per antibody per=20=

> run rather than per patient -
> Well the question I would pose is this. If in 5 years time you wish to=20=

> review the immuno for a particular case, how do you ensure that the=20
> staining quality was adequate at the time if there aren't any positive=20=

> controls for that case?
> On medico-legal grounds alone you should be doing positive controls=20
> for each case, especially if those results end up with the patient=20
> having a particularly aggressive treatment like chemotherapy or a=20
> particularly invasive surgical procedure such as a colectomy or a=20
> mastectomy. If someone wants to sue your organisation at a later date=20=

> for inappropriate treatment you need every bit of proof that your part=20=

> of the investigation was above reproach and controls is one thing that=20=

> will definitely be asked about. I certainly wouldn't like to stand up=20=

> in court and try and defend myself against that one!
> =A0
> Incidentally, we use Surgipath's Control slides which allow you (most=20=

> of the time depending on the size of the test section) to have the=20
> positive control and test section on the same slide, which saves a lot=20=

> of space on the immunostainer and neatly solves the problem of where=20=

> do you store the positive control if you use the single control per=20
> run model and you have multiple cases with the same antibody. It also=20=

> acts as a check that the slide has received the correct primary=20
> antibody and that someone hasn't loaded the immunostainer wrongly.
> =A0
> At the end of the day I still go by the analogy someone else made here=20=

> earlier about airbags and seat belts in cars. Would you drive in a car=20=

> without them? They may never be needed in your entire driving life,=20
> but you would be a dam fool not to have=A0 them wouldn't you?
> =A0
> And I=A0think we will have to agree to=A0disagree Hadi, with your last=20=

> statement, it is poor quality control by any definition of the term,=20=

> not to use both negative and positive controls for each case.
> =A0
> Nick Kirk
> Histopathology
> Hinchingbrooke Hospital
> Huntingdon
> England
>
> -----Original Message-----
> From: histonet-admin@lists.utsouthwestern.edu=20
> [mailto:histonet-admin@lists.utsouthwestern.edu]On Behalf Of Hadi=20
> Yaziji
> Sent: 11 October 2003 03:07
> To: histonet@pathology.swmed.edu
> Subject: Re: [Histonet] negative IHC controls
>
> Theoretically, I agree with you that (a) running multiple negative=20
> control sections to cover the different antibodies/pretreatments is=20
> ideal and (b) money should be well spent on QC (I can't emphasize that=20=

> enough). However, I honestly don't remember when was the last time=20
> when I had a false positive staining on a patient's slide that the=20
> negative control was the only way that allowed me to recognize the=20
> false positive signal on the real slide. And we look at 100s of IHC=20
> stains every day.
>
> Experienced pathologists and technologists can still recognize in >=20
> 99% of the times a false positive signal from a real signal on the=20
> tissue section of a given antibody without even having to look at=20
> negative controls. I don't think it should be a CAP requirement at=20
> all. In fact, I'd say the same thing on positive controls. All you=20
> need is one positive control per antibody, but not one control/per=20
> antibody/per patient. In many cases positive internal controls are=20
> present on the same slide, so you can tell whether your antibody=20
> worked simply by evaluating the expected positive internal controls.=20=

> In fact, if your positive 'external' control worked and internal=20
> controls didn't, then you need to repeat your antibody test regardless=20=

> of the positive external control. External controls are different=20
> tissues, fixed differently, processed differently and the tissue ages=20=

> differently (in terms of its antigenicity).
>
> Individuals who perform and interpret IHC studies must have the=20
> knowledge to recognize the expected sub-cellular localization of every=20=

> antibody on every type of tissue, including aberrant signals. For=20
> instance, you can easily dismiss a granular cytoplasmic TTF-1 signal=20=

> as not positive, because TTF-1 is a nuclear transcription factor and=20=

> we know it's expressed in the nuclei. However, hepatocellular=20
> carcinomas can give you a consistent and reproducible granular and=20
> cytoplasmic TTF-1 signal, and if you subsequently run confirmatory=20
> markers (HepPar1, polyclonal CEA, CD10, etc.) they will be positive.=20=

> Such (granular cytoplasmic) signal should be recognized by the=20
> interpreter (tech or pathologist) as a specific signal for tumors with=20=

> hepatoid phenotypes. This is just one of many examples..
>
> The bottom line: in real life, one negative control per case is more=20=

> than sufficient. And to say the least, it's inaccurate to state this=20=

> is poor patient care and lousy quality control.
>
> I look forward to any constructive criticism.
>
> Hadi Yaziji, M.D.
> PhenoPath Laboratories
>
> On Thursday, October 9, 2003, at 07:03 AM, Horn, Hazel V wrote:
>
> Not only should you be running a negative control for each patient=20
> slide.=A0=A0 That negative control should be treated just as your =
antibody=20
> is.=A0=A0 If the antibody is rabbit and antigen retrieved, so should =
your=20
> control.=A0=A0 If another antibody on the same patient is mouse and =
not=20
> retrieved another negative control should be run with this same=20
> protocol.=A0=A0 In the United States, labs that are inspected by the =
CAP,=20
> are required to run these controls.=A0=A0=A0 MONEY should never be=20
> considered as a reason to stop doing a part of a procedure.=A0=A0 It's=20=

> poor patient care and lousy quality control.=A0=A0 IMHO.
> =A0
> Hazel Horn, HT/HTL (ASCP)
> Histology Supervisor
> Arkansas Children's Hospital
>
> Phone - 501.364.4240
> Fax - 501.364.3912
>
>
> -----Original Message-----
> From: vermast [mailto:vermast@rogers.com]
> Sent: Wednesday, October 08, 2003 3:57 PM
> To: histonet@lists.utsouthwestern.edu
> Subject: [Histonet] negative IHC controls
>
> =A0
> I would like to get a feel for how many out there are running negative=20=

> control slides for IHC.=A0
> =A0
> In our lab we do just a handful of antibodies and initially I had been=20=

> running a negative control slide with each patient slide. =A0 After =
much=20
> discussion with our pathologists, we decided to omit these negatives=20=

> (which were conistently negative)=A0and continue to just run a =
positive=20
> control with each primary antibody for the run.=A0 We use the Dako=20
> autostainer and prediluted primaries.=A0 The decision to stop running=20=

> negatives also coincided with Dako's decision to sell the negative=20
> control sera separately from the primaries (they used to come packaged=20=

> together).=A0 Perhaps I assumed that discontinuing to pair these=20
> reagents together meant that few labs were using the negatives.
> =A0
> Anyhow, after having reviewed the last QMPLS (Canada) survey committee=20=

> comments, I believe the committe would like a negative control run=20
> with each patient tissue slide in order to evaluate background=A0 =
(they=20
> have used NCCLS guide pages as reference).=A0 Incidentally we weren't =
a=20
> part of the survey due to a technicality.
> Any help or advice would be appreciated.
> =A0
> L. Vermast
> Stratford, Ont.
>
>
>
>
>
> 
>
>
> The information contained in this message may be privileged and=20
> confidential and protected from disclosure. If the reader of this=20
> message is not the intended recipient, or an employee or agent=20
> responsible for delivering this message to the intended recipient, you=20=

> are hereby notified that any dissemination, distribution or copying of=20=

> this communication is strictly prohibited. If you have received this=20=

> communication in error, please notify us immediately by replying to=20
> the message and deleting it from your computer. Thank you. Arkansas=20
> Children's Hospital
>
>

--Apple-Mail-2--243860447
Content-Transfer-Encoding: quoted-printable
Content-Type: text/enriched;
	charset=ISO-8859-1

Thanks for sharing your CV with us.


You seem to be fixated on how slides should be filed. It's quite easy
in our laboratory (and any other laboratory) to find the positive
QC's, because they're filed chronologically (is there any other way to
file slides?). First you find out the date the case in question was
run, and you go to the QC file cabinet and pull the positive QC's that
belong to the same day. Negative QC's (and again, one slide per tissue
block) are filed with the case.


As to your "evidence based good practice", please don't generalize.
Two hospitals across the street from each other can represent two
extremes of quality of care. Actually, I happened to have spent a
month in a prestigious Histopathology Department in London nine years
ago. The quality of IHC in their lab was superb, but it was miserable
in the hospital right across the street from them. The same applies to
the US and any other country.=20


Finally, if my last email was unprofessional and insulting to you,
please accept my apology. Please remember that it was your unnecessary
use of the terms "dam fool"  and "poor quality control by any
definition of the term" in your previous email that initiated my
response to you.


Best wishes,


Hadi Yaziji, M.D.

PhenoPath Laboratories



On Monday, October 13, 2003, at 03:00  PM, Nick Kirk wrote:


=
Arial0000,0000,FFFFHadi

=A0

=
Arial0000,0000,FFFFI
am sorry but I must take issue with your very condescending and if I
may say so insulting tone.

=A0

=
Arial0000,0000,FFFFFor
your information, I have over 18 years of experience doing IHC and my
lab does a considerable number of slides per =
day.

=
Arial0000,0000,FFFFI
have set up the IHC service in two labs and have received much praise
for doing it.

=
Arial0000,0000,FFFFI
have also done extensive=A0lab based=A0trials work for both DAKO and
Novocastra here in the UK as well, so I think my credentials are
better than most.

=
Arial0000,0000,FFFFI'm
not at issue with the interpretation issue at all, that is indeed a
very skilled task that takes many years of study and experience to
become competent in.

=
Arial0000,0000,FFFFWhat
I am taking issue with is the issues around clinical audit and best
practice.

=
Arial0000,0000,FFFFNow
I don't know what clinical audit systems you have in your
establishment, but in ours it is regarded as poor clinical governance
i.e. bad practice by both myself, the Head Biomedical Scientist of the
department and all of my Pathologist colleagues not to have separate
controls per case.

=
Arial0000,0000,FFFFThis
is especially the case when you are reviewing tumour antigenicity on
samples that may come out of the archive.

=
Arial0000,0000,FFFFAll
our slides are stored in case order so to have positive controls for
each case actually makes it much easier the extract all the relevant
slides per case for later review. Having slides for the same case
filed in separate places introduces the potential for error,
especially when being initially filed.

=
Arial0000,0000,FFFFI
know this to be true because it has happened to me on several
occasions where slides filed in different places have been mis-filed
and become difficult to retrieve.

=
Arial0000,0000,FFFFIn
the UK we use a system of evidence based good practice and evidence
suggests that is precisely what we are =
doing.

=A0

=
Arial0000,0000,FFFFI
hope that now explains it to you.

=A0

=
Arial0000,0000,FFFFNick
Kirk BSc (Hons)=A0FIBMS (I've got letters =
too)

=
Arial0000,0000,FFFFHead
Biomedical Scientist

=
Arial0000,0000,FFFFHistopathology

=
Arial0000,0000,FFFFHinchingbrooke
Hospital

=
Arial0000,0000,FFFFHuntingdon

=
Arial0000,0000,FFFFCambridgeshire

=
Arial0000,0000,FFFFEngland

=A0


Tahoma-----Original Message-----

From: histonet-admin@lists.utsouthwestern.edu
[mailto:histonet-admin@lists.utsouthwestern.edu]On Behalf Of
Hadi Yaziji

Sent: 11 October 2003 16:22

To: histonet@pathology.swmed.edu

Subject: Re: [Histonet] negative IHC controls


Dear Nick,


Thanks for the email. You definitely misread my email. First, where in
my message did I advocate not to use negative and positive controls
per case? Of course it's poor patient's care not to do that. Is it
necessary to run + and - controls on every antibody? I don't think so
provided you have the necessary experience to interpret them. To
address your other point about reviewing cases, we keep our positive
and negative controls indefinitely. You can always pull the controls
that pertain to the run of the particular day of the case in question.


You will be 'dam fool' (I am copying your words) to run IHC if you
have no experience. That's much more dangerous legally and ethically.
I've seen slides misinterpreted where the positive control is present
on the same slide.


Bottom line: one + control per run is adequate (provided you have the
experience, which I'm not sure you have based on your reply). One -
control per case is more than sufficient.


I hope this addresses your concerns.


Hadi Yaziji, M.D.

PhenoPath Laboratories



On Saturday, October 11, 2003, at 12:16 AM, Nick Kirk wrote:


A cautionary note

=A0

I agree that false positives are rare but they do happen. We had a
couple a while ago when it turned out that the Teflon coat on the
probe of our DAKO Autostainer had perished and was carrying over
reagents from one slide to another and also contaminating our negative
control solution with primary antibody.

Also they are very useful in checking that there is no endogenous
biotin activity (more common than people think) and also that your
hydrogen peroxide solution hasn't gone off when quenching endogenous
peroxidase activity.

=A0

As for the argument about only one positive control per antibody per
run rather than per patient -

Well the question I would pose is this. If in 5 years time you wish to
review the immuno for a particular case, how do you ensure that the
staining quality was adequate at the time if there aren't any positive
controls for that case?

On medico-legal grounds alone you should be doing positive controls
for each case, especially if those results end up with the patient
having a particularly aggressive treatment like chemotherapy or a
particularly invasive surgical procedure such as a colectomy or a
mastectomy. If someone wants to sue your organisation at a later date
for inappropriate treatment you need every bit of proof that your part
of the investigation was above reproach and controls is one thing that
will definitely be asked about. I certainly wouldn't like to stand up
in court and try and defend myself against that one!

=A0

Incidentally, we use Surgipath's Control slides which allow you (most
of the time depending on the size of the test section) to have the
positive control and test section on the same slide, which saves a lot
of space on the immunostainer and neatly solves the problem of where
do you store the positive control if you use the single control per
run model and you have multiple cases with the same antibody. It also
acts as a check that the slide has received the correct primary
antibody and that someone hasn't loaded the immunostainer wrongly.

=A0

At the end of the day I still go by the analogy someone else made here
earlier about airbags and seat belts in cars. Would you drive in a car
without them? They may never be needed in your entire driving life,
but you would be a dam fool not to have=A0 them wouldn't you?

=A0

And I=A0think we will have to agree to=A0disagree Hadi, with your last
statement, it is poor quality control by any definition of the term,
not to use both negative and positive controls for each case.

=A0

Nick Kirk

Histopathology

Hinchingbrooke Hospital

Huntingdon

England


-----Original Message-----

From: histonet-admin@lists.utsouthwestern.edu
[mailto:histonet-admin@lists.utsouthwestern.edu]On Behalf Of
Hadi Yaziji

Sent: 11 October 2003 03:07

To: histonet@pathology.swmed.edu

Subject: Re: [Histonet] negative IHC controls


Theoretically, I agree with you that (a) running multiple negative
control sections to cover the different antibodies/pretreatments is
ideal and (b) money should be well spent on QC (I can't emphasize that
enough). However, I honestly don't remember when was the last time
when I had a false positive staining on a patient's slide that the
negative control was the only way that allowed me to recognize the
false positive signal on the real slide. And we look at 100s of IHC
stains every day.


Experienced pathologists and technologists can still recognize in >
99% of the times a false positive signal from a real signal on the
tissue section of a given antibody without even having to look at
negative controls. I don't think it should be a CAP requirement at
all. In fact, I'd say the same thing on positive controls. All you
need is one positive control per antibody, but not one control/per
antibody/per patient. In many cases positive internal controls are
present on the same slide, so you can tell whether your antibody
worked simply by evaluating the expected positive internal controls.
In fact, if your positive 'external' control worked and internal
controls didn't, then you need to repeat your antibody test regardless
of the positive external control. External controls are different
tissues, fixed differently, processed differently and the tissue ages
differently (in terms of its antigenicity).


Individuals who perform and interpret IHC studies must have the
knowledge to recognize the expected sub-cellular localization of every
antibody on every type of tissue, including aberrant signals. For
instance, you can easily dismiss a granular cytoplasmic TTF-1 signal
as not positive, because TTF-1 is a nuclear transcription factor and
we know it's expressed in the nuclei. However, hepatocellular
carcinomas can give you a consistent and reproducible granular and
cytoplasmic TTF-1 signal, and if you subsequently run confirmatory
markers (HepPar1, polyclonal CEA, CD10, etc.) they will be positive.
Such (granular cytoplasmic) signal should be recognized by the
interpreter (tech or pathologist) as a specific signal for tumors with
hepatoid phenotypes. This is just one of many examples..


The bottom line: in real life, one negative control per case is
more than sufficient. And to say the least, it's
inaccurate to state this is poor patient care and lousy quality
control.


I look forward to any constructive criticism.


Hadi Yaziji, M.D.

PhenoPath Laboratories


On Thursday, October 9, 2003, at 07:03 AM, Horn, Hazel V wrote:


Not only should you be running a negative control for each patient
slide.=A0=A0 That negative control should be treated just as your =
antibody
is.=A0=A0 If the antibody is rabbit and antigen retrieved, so should =
your
control.=A0=A0 If another antibody on the same patient is mouse and not
retrieved another negative control should be run with this same
protocol.=A0=A0 In the United States, labs that are inspected by the =
CAP,
are required to run these controls.=A0=A0=A0 MONEY should never be
considered as a reason to stop doing a part of a procedure.=A0=A0 It's
poor patient care and lousy quality control.=A0=A0 IMHO.

=A0

Hazel Horn, HT/HTL (ASCP)

Histology Supervisor

Arkansas Children's Hospital


Phone - 501.364.4240

Fax - 501.364.3912



-----Original Message-----

From: vermast [mailto:vermast@rogers.com]

Sent: Wednesday, October 08, 2003 3:57 PM

To: histonet@lists.utsouthwestern.edu

Subject: [Histonet] negative IHC controls


=A0

I would like to get a feel for how many out there are running negative
control slides for IHC.=A0

=A0

In our lab we do just a handful of antibodies and initially I had been
running a negative control slide with each patient slide. =A0 After much
discussion with our pathologists, we decided to omit these negatives
(which were conistently negative)=A0and continue to just run a positive
control with each primary antibody for the run.=A0 We use the Dako
autostainer and prediluted primaries.=A0 The decision to stop running
negatives also coincided with Dako's decision to sell the negative
control sera separately from the primaries (they used to come packaged
together).=A0 Perhaps I assumed that discontinuing to pair these
reagents together meant that few labs were using the negatives.

=A0

Anyhow, after having reviewed the last QMPLS (Canada) survey committee
comments, I believe the committe would like a negative control run
with each patient tissue slide in order to evaluate background=A0 (they
have used NCCLS guide pages as reference).=A0 Incidentally we weren't a
part of the survey due to a technicality.

Any help or advice would be appreciated.

=A0

L. Vermast

Stratford, Ont.






<



The information contained in this message may be privileged and
confidential and protected from disclosure. If the reader of this
message is not the intended recipient, or an employee or agent
responsible for delivering this message to the intended recipient, you
are hereby notified that any dissemination, distribution or copying of
this communication is strictly prohibited. If you have received this
communication in error, please notify us immediately by replying to
the message and deleting it from your computer. Thank you. Arkansas
Children's Hospital



=

--Apple-Mail-2--243860447--


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


<< Previous Message | Next Message >>