Re: My First PCR Results

[Guest guest]

Hi Efrem,

The number means little without knowing the baseline of the lab. Since

all labs have their own numbers, you need to find out what your lab's

baseline value is, then you can compare your number to see what kind of

a log reduction you've had.

Each time the decimal point moves, it's one log so if your lab's

baseline was 30.00 then your count of 3.55 would be a one log

reduction. If your lab's baseline was 300.00 then your count would be

a two log reduction.

It can take time (sometimes even years) to get to the magical three log

reduction so don't worry if you're not there yet.

Your next PCR in three months will be important because it'll give you

something to compare this one to. In the end, it's the trend that's

important, not any one single test.

Tracey

>

> I was diagnosed on 1/4/07 and have been on since late January.

> I recently had my first PCR test and I tested positive with a

number of 3.55.

>

> Does anyone have any insight to what that means.

> I did not do a PCR when I was diagnosed only a FISH test..

>

>

>

[Guest guest]

Just to add to Tracey's notes:

I have heard that from a larger center (OHSU I think) an " average " PCR

ratio value for newly diagnosed patients was found to be 50%.

This was an average so some people might be diagnosed at a higher or

lower.

The goal of treatment will be to get about a 3 log reduction in the

BCR/ABL ratio or a value of 0.05% (if the person started at 50%, then

0.05% would be 3 log).

[Guest guest]

What does 50% mean as far as the ratio. My results were 3.55. Which were 3.55

healthy cells to every 1 nonhealthy. Does that mean my percentage is 21.9

percent??? (1 out of every 4.55 is nonhealthy)

<timothyfarley16@...> wrote: Just to add to Tracey's

notes:

I have heard that from a larger center (OHSU I think) an " average " PCR

ratio value for newly diagnosed patients was found to be 50%.

This was an average so some people might be diagnosed at a higher or

lower.

The goal of treatment will be to get about a 3 log reduction in the

BCR/ABL ratio or a value of 0.05% (if the person started at 50%, then

0.05% would be 3 log).

[Guest guest]

There a bit of confusion here. A PCR (quantitative RT-PCR) report shows

different numbers and is somewhat cryptic. I suspect the 3.55 number to be

the percentage (3.55/100) of Ph+ cell detected by the test. But if you give

us more details on what is on the report we can give you a better answer.

Supposing this number is the percentage of ph+ cells, it would mean the test

is finding a ratio of 0.0355 ph+ cells. The ratio is always the number of

ph+ cells over the number of normal cells. It is always between 0 (PCRU) and

1 (100% of ph+ cells). Some labs give the ratio as is (my does), some give

the percentage (ratio x 100). It should be mentioned in the report what the

given number means.

The lab report should also give the log reduction from baseline (my lab

gives me a baseline of 100%, some use a lower value like 50%). The precise

value of the baseline doesn't matters so much for us as it matters for the

clinical studies as they need to have consistency in the reporting to

compare results from a cohort of patients. PCR values of 0.01% or 0.005% are

both pretty good, even if some would call the first MMR and some would

accept only the second as MMR.

Cheers,

Marcos.

On 6/5/07, fischer <efrem_fischer@...> wrote:

>

> What does 50% mean as far as the ratio. My results were 3.55. Which were

> 3.55 healthy cells to every 1 nonhealthy. Does that mean my percentage is

> 21.9 percent??? (1 out of every 4.55 is nonhealthy)

>

> <timothyfarley16@... <timothyfarley16%40>> wrote:

> Just to add to Tracey's notes:

>

> I have heard that from a larger center (OHSU I think) an " average " PCR

> ratio value for newly diagnosed patients was found to be 50%.

>

> This was an average so some people might be diagnosed at a higher or

> lower.

>

> The goal of treatment will be to get about a 3 log reduction in the

> BCR/ABL ratio or a value of 0.05% (if the person started at 50%, then

> 0.05% would be 3 log).

>

>

[Guest guest]

I'll let Tim answer your question because clearly math isn't my

strong suit but I will say this much, some labs report PCR results

using a percentage (%) and other labs use whole numbers. Was your

3.55 an actual percentage or just a value given?

Tracey

Just to add to

Tracey's notes:

>

> I have heard that from a larger center (OHSU I think) an " average "

PCR

> ratio value for newly diagnosed patients was found to be 50%.

>

> This was an average so some people might be diagnosed at a higher

or

> lower.

>

> The goal of treatment will be to get about a 3 log reduction in the

> BCR/ABL ratio or a value of 0.05% (if the person started at 50%,

then

> 0.05% would be 3 log).

>

>

>

>

>

>

>

[Guest guest]

I agree with what Tracey and Marcos wrote.

These reports can be confusing even to doctors and nurses. As Marcos

wrote, if your ratio is expressed as a number it would be 0.0355 and if

a percentage than 3.55 %.

If the baseline for a CML population the testing lab was 50%, this

would equal 0.5 To convert percentages to a whole number, move the

decimal 2 spots to the left.

The initial goal is a 3 log reduction which as Tracey explained is

moving the decimal place 3 spots to the left. This is also called a

Major Molecular Response (MMR). Depending on the lab, this number is

either 0.05% (aka 0.0005) or it could be 0.01% and (aka 0.0001).

[Guest guest]

On 6/5/07, <timothyfarley16@...> wrote:

> I agree with what Tracey and Marcos wrote.

>

> These reports can be confusing even to doctors and nurses. As Marcos

> wrote, if your ratio is expressed as a number it would be 0.0355 and if

> a percentage than 3.55 %.

> If the baseline for a CML population the testing lab was 50%, this

> would equal 0.5 To convert percentages to a whole number, move the

> decimal 2 spots to the left.

>

> The initial goal is a 3 log reduction which as Tracey explained is

> moving the decimal place 3 spots to the left. This is also called a

> Major Molecular Response (MMR). Depending on the lab, this number is

> either 0.05% (aka 0.0005) or it could be 0.01% and (aka 0.0001).

If the second set of numbers are referring to a 100% baseline you are

off a zero :

for a 50% baseline you are right : MMR is 0.05% (0.5/10^3 = 0.5/1000 = 0.0005)

for a 100% baseline MMR is 0.1% (1/10^3 = 1/1000 = 0.001)

Marcos.

>

>

>

>

--

Marcos Perreau Guimaraes

Suppes Brain Lab

Ventura Hall - CSLI

Stanford University

220 Panama street

Stanford CA 94305-4101

650 329 9920 x 305

650 630 5015 (cell)

marcospg@...

montereyunderwater@...

www.stanford.edu/~marcospg/

[Guest guest]

Hi all,

This is long.

Let me try and make this a little clearer (or a lot muddier). The PCR number

that OHSU uses for a newly diagnosed patient is 50. The lab that I use in

Montreal uses a value of 10 for a newly diagnosed CML patient. As Tracey

pointed out, a PCR test is not usually done at diagnosis time, so an average

value for that institution is used.

Once we have a starting point, then we can measure how well or how bad we

are doing. It is very common to report the difference in terms of a log

reduction.

Actually it really doesn't matter how you report it. As long as you have two

values, you can easily see the difference.

Let's take an example. Suppose you were an OHSU patient and after 3 months

on Gleevec your PCR value was 5. This is then very simple to express. You

can say that you had a reduction of 45 in your PCR value. Or you can say

that you had a 1 log reduction in your PCR.

Now it is 6 months later and your PCR test returned a value of 0.5. You can

compare this latest result with the initial value of 50 and come up with a

reduction of 2 logs since you began treatment. Or you can compare it with

the 3 month result and see that you have a 1 log reduction over the past 3

months.

Over time, your PCR result tends to stabilize. So after a few years on

Gleevec, you PCR might reach 0.05 and stay there. This will be a 3 log

reduction from your PCR at diagnosis time.

Or, it might return a value of 0.0000. At this point, the PCR equipment

being used cannot detect any CML chromosome in the sample. You would then be

considered to be PCRU or PCR undetectable.

Now let's complicate your life. Suppose your result came back as 0.06. We

know what the log reduction is if the value was .05 compared to a starting

point of 50 or 5, but how do we calculate the log reduction for a value of

0.06? You need a calculator or a slide rule. The answer is a 2.92log

reduction. You get it by evaluating 0.06/50; ans = 0.0012 and looking up the

log of 0.0012.

The following web site http://wolf.galekus.com/viewpage.php?page_id=10 will

do the calculation for you. Enter 10 for the log base, then the number and

press on Evaluate.

If your PCR was 0.04, then the calculation is 0.04/50 = 0.0008 and looking

up the log of 0.0008 we get a 3.0969 reduction.

Zavie

[ ] Re: My First PCR Results

Just to add to Tracey's notes:

I have heard that from a larger center (OHSU I think) an " average " PCR

ratio value for newly diagnosed patients was found to be 50%.

This was an average so some people might be diagnosed at a higher or

lower.

The goal of treatment will be to get about a 3 log reduction in the

BCR/ABL ratio or a value of 0.05% (if the person started at 50%, then

0.05% would be 3 log).

[Guest guest]

To be clear, you numbers are percent, right ? Maybe we could write % or

percent when it's a percentage, as some labs reports give directly the

ratio, as mine does. Also some labs use a 100% baseline for some people, I

know at least one case : myself.

Marcos.

On 6/5/07, Zavie miller <zmiller@...> wrote:

>

> Hi all,

>

> This is long.

>

> Let me try and make this a little clearer (or a lot muddier). The PCR

> number

> that OHSU uses for a newly diagnosed patient is 50. The lab that I use in

> Montreal uses a value of 10 for a newly diagnosed CML patient. As Tracey

> pointed out, a PCR test is not usually done at diagnosis time, so an

> average

> value for that institution is used.

>

> Once we have a starting point, then we can measure how well or how bad we

> are doing. It is very common to report the difference in terms of a log

> reduction.

>

> Actually it really doesn't matter how you report it. As long as you have

> two

> values, you can easily see the difference.

>

> Let's take an example. Suppose you were an OHSU patient and after 3 months

> on Gleevec your PCR value was 5. This is then very simple to express. You

> can say that you had a reduction of 45 in your PCR value. Or you can say

> that you had a 1 log reduction in your PCR.

>

> Now it is 6 months later and your PCR test returned a value of 0.5. You

> can

> compare this latest result with the initial value of 50 and come up with a

> reduction of 2 logs since you began treatment. Or you can compare it with

> the 3 month result and see that you have a 1 log reduction over the past 3

> months.

>

> Over time, your PCR result tends to stabilize. So after a few years on

> Gleevec, you PCR might reach 0.05 and stay there. This will be a 3 log

> reduction from your PCR at diagnosis time.

>

> Or, it might return a value of 0.0000. At this point, the PCR equipment

> being used cannot detect any CML chromosome in the sample. You would then

> be

> considered to be PCRU or PCR undetectable.

>

> Now let's complicate your life. Suppose your result came back as 0.06. We

> know what the log reduction is if the value was .05 compared to a starting

> point of 50 or 5, but how do we calculate the log reduction for a value of

> 0.06? You need a calculator or a slide rule. The answer is a 2.92log

> reduction. You get it by evaluating 0.06/50; ans = 0.0012 and looking up

> the

> log of 0.0012.

>

> The following web site http://wolf.galekus.com/viewpage.php?page_id=10will

> do the calculation for you. Enter 10 for the log base, then the number and

> press on Evaluate.

>

> If your PCR was 0.04, then the calculation is 0.04/50 = 0.0008 and looking

> up the log of 0.0008 we get a 3.0969 reduction.

>

> Zavie

>

> [ ] Re: My First PCR Results

>

> Just to add to Tracey's notes:

>

> I have heard that from a larger center (OHSU I think) an " average " PCR

> ratio value for newly diagnosed patients was found to be 50%.

>

> This was an average so some people might be diagnosed at a higher or

> lower.

>

> The goal of treatment will be to get about a 3 log reduction in the

> BCR/ABL ratio or a value of 0.05% (if the person started at 50%, then

> 0.05% would be 3 log).

>

>

[Guest guest]

iT WAS A VALUE GIVEN

Tracey <traceyincanada@...> wrote: I'll let Tim answer your

question because clearly math isn't my

strong suit but I will say this much, some labs report PCR results

using a percentage (%) and other labs use whole numbers. Was your

3.55 an actual percentage or just a value given?

Tracey

Just to add to

Tracey's notes:

>

> I have heard that from a larger center (OHSU I think) an " average "

PCR

> ratio value for newly diagnosed patients was found to be 50%.

>

> This was an average so some people might be diagnosed at a higher

or

> lower.

>

> The goal of treatment will be to get about a 3 log reduction in the

> BCR/ABL ratio or a value of 0.05% (if the person started at 50%,

then

> 0.05% would be 3 log).

>

>

>

>

>

>

>

[Guest guest]

The number 50 is not a percentage. It is a value. The same for 5 and the

others.

Zavie

[ ] Re: My First PCR Results

>

> Just to add to Tracey's notes:

>

> I have heard that from a larger center (OHSU I think) an " average " PCR

> ratio value for newly diagnosed patients was found to be 50%.

>

> This was an average so some people might be diagnosed at a higher or

> lower.

>

> The goal of treatment will be to get about a 3 log reduction in the

> BCR/ABL ratio or a value of 0.05% (if the person started at 50%, then

> 0.05% would be 3 log).

>

>

[Guest guest]

If it's the only number your dr gave you then it's a percentage, 3.55% (in

average 3.55 leukemia cell in one hundred cells), what's good at 6 months

(you didn't tell us your percentage on the FISH test at dx, so I can't tell

you the reduction). You are doing pretty well, don't worry and wait for the

next PCR.

I had 7% at 6 months (started at dx with 100% on the FISH test) and I am

doing fine 2 years later. My last PCR was .5%, a 2.x log reduction from

initial 100% value, not yet MMR but as other mentioned that's not the end of

the world.

Cheers,

Marcos.

On 6/5/07, fischer <efrem_fischer@...> wrote:

>

> iT WAS A VALUE GIVEN

>

> Tracey <traceyincanada@... <traceyincanada%40>> wrote: I'll

> let Tim answer your question because clearly math isn't my

>

> strong suit but I will say this much, some labs report PCR results

> using a percentage (%) and other labs use whole numbers. Was your

> 3.55 an actual percentage or just a value given?

>

> Tracey

>

> Just to add to

> Tracey's notes:

> >

> > I have heard that from a larger center (OHSU I think) an " average "

> PCR

> > ratio value for newly diagnosed patients was found to be 50%.

> >

> > This was an average so some people might be diagnosed at a higher

> or

> > lower.

> >

> > The goal of treatment will be to get about a 3 log reduction in the

> > BCR/ABL ratio or a value of 0.05% (if the person started at 50%,

> then

> > 0.05% would be 3 log).

> >

> >

> >

> >

> >

> >

> >

[Guest guest]

You mean it's the number with a unit like ng/ìL of BCR/ABL RNA (don't have

the test report there) ?

My last PCR had actually the result with the old and new units as the

standard changed. But they always translate it into a ratio of ph+ cells

over normal cells (I multiply by 100 to have the percentage), and also give

me the log reduction from my initial percentage (100%). I have to say the

units and explanation on how to get the ratio and log reduction is always

written on my reports.

Although some standardization of the lab reports wouldn't be a bad thing.

When I showed my PCR reports to my 2nd Dr in Paris she was grumbling about

that.

Marcos.

On 6/5/07, Zavie miller <zmiller@...> wrote:

>

>

> The number 50 is not a percentage. It is a value. The same for 5 and the

> others.

>

> Zavie

>

> [ ] Re: My First PCR Results

> >

> > Just to add to Tracey's notes:

> >

> > I have heard that from a larger center (OHSU I think) an " average " PCR

> > ratio value for newly diagnosed patients was found to be 50%.

> >

> > This was an average so some people might be diagnosed at a higher or

> > lower.

> >

> > The goal of treatment will be to get about a 3 log reduction in the

> > BCR/ABL ratio or a value of 0.05% (if the person started at 50%, then

> > 0.05% would be 3 log).

> >

> >

[Guest guest]

> (you didn't tell us your percentage on the FISH test at dx, so I

can't tell

> you the reduction).

*******************************

Marcos, how can you compare a FISH test from diagnosis to a PCR test 6

months later? Isn't that like comparing chocolate muffins to chocolate

bars? They both contain chocolate but other than that, you can't say

much about them.

Tracey

[Guest guest]

Here an explanation of PCR testing from the cml support web I used in the

past :

http://www.cmlsupport.org.uk/?q=pcrresults

Marcos.

On 6/5/07, Marcos Perreau Guimaraes <montereyunderwater@...> wrote:

>

> You mean it's the number with a unit like ng/ìL of BCR/ABL RNA (don't have

> the test report there) ?

> My last PCR had actually the result with the old and new units as the

> standard changed. But they always translate it into a ratio of ph+ cells

> over normal cells (I multiply by 100 to have the percentage), and also give

> me the log reduction from my initial percentage (100%). I have to say the

> units and explanation on how to get the ratio and log reduction is always

> written on my reports.

> Although some standardization of the lab reports wouldn't be a bad thing.

> When I showed my PCR reports to my 2nd Dr in Paris she was grumbling about

> that.

> Marcos.

>

> On 6/5/07, Zavie miller <zmiller@...> wrote:

> >

> >

> > The number 50 is not a percentage. It is a value. The same for 5 and the

> > others.

> >

> > Zavie

> >

> > [ ] Re: My First PCR Results

> > >

> > > Just to add to Tracey's notes:

> > >

> > > I have heard that from a larger center (OHSU I think) an " average " PCR

> > > ratio value for newly diagnosed patients was found to be 50%.

> > >

> > > This was an average so some people might be diagnosed at a higher or

> > > lower.

> > >

> > > The goal of treatment will be to get about a 3 log reduction in the

> > > BCR/ABL ratio or a value of 0.05% (if the person started at 50%, then

> > > 0.05% would be 3 log).

> > >

> > >

[Guest guest]

Marcos:

Are you sure your 100% value is not a FISH or karyotyping cytogenetics

result? It just seems to be such a

Most every person at diagnosis is 100% PH+ by cytogenetics and a pretty

high number if a FISH is done (70-90%). The PCR value can vary

depending on lab technique and stage of diagnosis.

I plugged a formula in MS-EXCEL that I think works.

Put 1.04 in cell A1. Put 0.0059 in cell B1. In cell C1 write the

formula =LOG10(B1/A1). This gave a result of -2.246 and that seems to

be the reduction they put on the report. I believe some labs will use

the standard calculation (1.04 in Stanford's case) if the patient did

not have their personal baseline PCR result for future comparison.

I have heard that the PCR equivalent of a Complete Cytogenetic Response

is between 1.5 and 2.0 log reduction.

Hope this helps

[Guest guest]

My question is what the 0.0059 represents. In physical science a number

always represents something, even numbers without dimensions like this

ratio. A ratio is in general between two numbers which have the same unit

and measure something.

PCR :

If my understanding is right a quantitative RT-PCR measures the number of

copies of two gene sequences :

- the BCR-ABL sequence which only leukemic blood stem cells contain

- and some some other (control, sometime called housekeeping) gene every

blood stem cell (leukemic and normal) has.

The ratio given in the PCR report is the ratio between these two numbers.

The log change is the log in base 10 of the ratio between this ratio and

the average ratio at dx. This was the unclear part, why 1.04, greater than

100% ? From what I understand the measure of the number of copies of BCR-ABL

has some false positives in it, and they correct it by adding 0.04 to the

baseline. But I can be wrong.

There are a lot of things the lab needs to control to have an accurate

result and they don't use all the same housekeeping gene, so results for a

sample sample may slightly vary from one lab to another.

FISH :

measure the same thing by other ways and is less accurate (what doesn't

matter when I was near 100%). Instead od the 1/100000 minimum power of

resolution of PCR, it looks at only 100 or 200 cells.

Conclusion : I have the ratio (or percentage) of ph+ in my blood at dx and

later. Even if they are not measured with the same tool I should be able to

compare them. If you ask a friend to count the money in your wallet, you

should have the same or a very close amount than if you looked yourself (if

it's really a friend), and you can compare the value with what you counted

yourself a week before. Your friend or you may have forgot a cent, but that

shouldn't matter so much.

Cheers,

Marcos

On 6/6/07, <timothyfarley16@...> wrote:

>

> Marcos:

>

> Are you sure your 100% value is not a FISH or karyotyping cytogenetics

> result? It just seems to be such a

>

> Most every person at diagnosis is 100% PH+ by cytogenetics and a pretty

> high number if a FISH is done (70-90%). The PCR value can vary

> depending on lab technique and stage of diagnosis.

>

> I plugged a formula in MS-EXCEL that I think works.

> Put 1.04 in cell A1. Put 0.0059 in cell B1. In cell C1 write the

> formula =LOG10(B1/A1). This gave a result of - 2.246 and that seems to

> be the reduction they put on the report. I believe some labs will use

> the standard calculation (1.04 in Stanford's case) if the patient did

> not have their personal baseline PCR result for future comparison.

>

> I have heard that the PCR equivalent of a Complete Cytogenetic Response

> is between 1.5 and 2.0 log reduction.

>

> Hope this helps

>

>

>

--

Marcos Perreau Guimaraes

Suppes Brain Lab

Ventura Hall - CSLI

Stanford University

220 Panama street

Stanford CA 94305-4101

650 329 9920 x 305

650 630 5015 (cell)

marcospg@...

montereyunderwater@...

www.stanford.edu/~marcospg/

[Guest guest]

Hi Marcos,

Now for my pet peeve.

" There are a lot of things the lab needs to control to have an accurate

result and they don't use all the same housekeeping gene, so results for a

single sample may slightly vary from one lab to another. "

" Even if they are not measured with the same tool I should be able to

compare them. "

The results can vary dramatically from one lab to another. In fact, they

will vary within the same lab.

For my past 2 PCR tests, I have had a sample sent to two different labs for

analysis. One lab returned a value of a 2.9 log reduction and the other lab

returned a value of > 4 log reduction.

So who do you believe? There is no standardardization when it comes to PCR

testing for CML. There are several international efforts underway to

standardize PCR testing but it looks like it will be 5-10 years before there

will be an international standard.

My advice is to find one lab and keep getting you PCR testing done there.

Also, you need to check with them constantly to see if they have changed

their technique or their eqipmentment.

Zavie

Re: [ ] Re: My First PCR Results

My question is what the 0.0059 represents. In physical science a number

always represents something, even numbers without dimensions like this

ratio. A ratio is in general between two numbers which have the same unit

and measure something.

PCR :

If my understanding is right a quantitative RT-PCR measures the number of

copies of two gene sequences :

- the BCR-ABL sequence which only leukemic blood stem cells contain

- and some some other (control, sometime called housekeeping) gene every

blood stem cell (leukemic and normal) has.

The ratio given in the PCR report is the ratio between these two numbers.

The log change is the log in base 10 of the ratio between this ratio and

the average ratio at dx. This was the unclear part, why 1.04, greater than

100% ? From what I understand the measure of the number of copies of BCR-ABL

has some false positives in it, and they correct it by adding 0.04 to the

baseline. But I can be wrong.

There are a lot of things the lab needs to control to have an accurate

result and they don't use all the same housekeeping gene, so results for a

sample sample may slightly vary from one lab to another.

FISH :

measure the same thing by other ways and is less accurate (what doesn't

matter when I was near 100%). Instead od the 1/100000 minimum power of

resolution of PCR, it looks at only 100 or 200 cells.

Conclusion : I have the ratio (or percentage) of ph+ in my blood at dx and

later. Even if they are not measured with the same tool I should be able to

compare them. If you ask a friend to count the money in your wallet, you

should have the same or a very close amount than if you looked yourself (if

it's really a friend), and you can compare the value with what you counted

yourself a week before. Your friend or you may have forgot a cent, but that

shouldn't matter so much.

Cheers,

Marcos

On 6/6/07, <timothyfarley16@...> wrote:

>

> Marcos:

>

> Are you sure your 100% value is not a FISH or karyotyping cytogenetics

> result? It just seems to be such a

>

> Most every person at diagnosis is 100% PH+ by cytogenetics and a pretty

> high number if a FISH is done (70-90%). The PCR value can vary

> depending on lab technique and stage of diagnosis.

>

> I plugged a formula in MS-EXCEL that I think works.

> Put 1.04 in cell A1. Put 0.0059 in cell B1. In cell C1 write the

> formula =LOG10(B1/A1). This gave a result of - 2.246 and that seems to

> be the reduction they put on the report. I believe some labs will use

> the standard calculation (1.04 in Stanford's case) if the patient did

> not have their personal baseline PCR result for future comparison.

>

> I have heard that the PCR equivalent of a Complete Cytogenetic Response

> is between 1.5 and 2.0 log reduction.

>

> Hope this helps

>

>

>

--

Marcos Perreau Guimaraes

Suppes Brain Lab

Ventura Hall - CSLI

Stanford University

220 Panama street

Stanford CA 94305-4101

650 329 9920 x 305

650 630 5015 (cell)

marcospg@...

montereyunderwater@...

www.stanford.edu/~marcospg/

[Guest guest]

Even if they are not measured with the same tool I should be able to

> compare them.

************************************

No Marcos, you can't compare a FISH test to a PCR test because they

are completely different. In the FISH test, they look for the

bcr/abl rearrangement, plain and simple. In the PCR test they use a

control gene to compare the bcr/abl cells with normal cells and the

control gene they use varies from lab to lab. In fact, you can't

even compare one PCR to another if you use a different lab for both

tests.

I've had the exact same blood sample go to two different labs and

both came back with vastly different results. When I compare my

various samples over time at the same lab though, I see that they are

more or less stable so you really need to stick with the same lab

when doing PCR's.

With cytogenetics and FISH tests, you can go to any lab and get the

same results because those tests are standardized so you can compare

them but you just can't compare them with PCR's.

It will be a great day for everyone when they get around to

standardizing PCR tests but unfortunately, I don't see it happening

any time soon.

Take care,

Tracey

[Guest guest]

Hi Zavie,

I am past my 2h/week being with cml so I ll just say that I agree with

what you said, my lab is not very helpful when it comes to giving extra

information. This is all about the difference between precision and

accuracy. The second quote you pasted just says that if both numbers measure

the same quantity, even by different methods, you should be able to compare

them in theory, although it may not be accurate. But if that's the best I

can get I still want the information, keeping in mind the inaccuracy of the

result.

Now back to planning my next week-end dives, the forecast looks good

Marcos.

On 6/6/07, Zavie miller <zmiller@...> wrote:

>

> Hi Marcos,

>

> Now for my pet peeve.

>

> " There are a lot of things the lab needs to control to have an accurate

> result and they don't use all the same housekeeping gene, so results for a

> single sample may slightly vary from one lab to another. "

>

> " Even if they are not measured with the same tool I should be able to

> compare them. "

>

> The results can vary dramatically from one lab to another. In fact, they

> will vary within the same lab.

>

> For my past 2 PCR tests, I have had a sample sent to two different labs

> for

> analysis. One lab returned a value of a 2.9 log reduction and the other

> lab

> returned a value of > 4 log reduction.

>

> So who do you believe? There is no standardardization when it comes to PCR

> testing for CML. There are several international efforts underway to

> standardize PCR testing but it looks like it will be 5-10 years before

> there

> will be an international standard.

>

> My advice is to find one lab and keep getting you PCR testing done there.

> Also, you need to check with them constantly to see if they have changed

> their technique or their eqipmentment.

>

> Zavie

>

> Re: [ ] Re: My First PCR Results

>

> My question is what the 0.0059 represents. In physical science a number

> always represents something, even numbers without dimensions like this

> ratio. A ratio is in general between two numbers which have the same unit

> and measure something.

>

> PCR :

> If my understanding is right a quantitative RT-PCR measures the number of

> copies of two gene sequences :

> - the BCR-ABL sequence which only leukemic blood stem cells contain

> - and some some other (control, sometime called housekeeping) gene every

> blood stem cell (leukemic and normal) has.

> The ratio given in the PCR report is the ratio between these two numbers.

> The log change is the log in base 10 of the ratio between this ratio and

> the average ratio at dx. This was the unclear part, why 1.04, greater than

> 100% ? From what I understand the measure of the number of copies of

> BCR-ABL

> has some false positives in it, and they correct it by adding 0.04 to the

> baseline. But I can be wrong.

> There are a lot of things the lab needs to control to have an accurate

> result and they don't use all the same housekeeping gene, so results for a

> sample sample may slightly vary from one lab to another.

>

> FISH :

> measure the same thing by other ways and is less accurate (what doesn't

> matter when I was near 100%). Instead od the 1/100000 minimum power of

> resolution of PCR, it looks at only 100 or 200 cells.

>

> Conclusion : I have the ratio (or percentage) of ph+ in my blood at dx and

> later. Even if they are not measured with the same tool I should be able

> to

> compare them. If you ask a friend to count the money in your wallet, you

> should have the same or a very close amount than if you looked yourself

> (if

> it's really a friend), and you can compare the value with what you counted

> yourself a week before. Your friend or you may have forgot a cent, but

> that

> shouldn't matter so much.

>

> Cheers,

> Marcos

>

> On 6/6/07, <timothyfarley16@...<timothyfarley16%40>>

> wrote:

> >

> > Marcos:

> >

> > Are you sure your 100% value is not a FISH or karyotyping cytogenetics

> > result? It just seems to be such a

> >

> > Most every person at diagnosis is 100% PH+ by cytogenetics and a pretty

> > high number if a FISH is done (70-90%). The PCR value can vary

> > depending on lab technique and stage of diagnosis.

> >

> > I plugged a formula in MS-EXCEL that I think works.

> > Put 1.04 in cell A1. Put 0.0059 in cell B1. In cell C1 write the

> > formula =LOG10(B1/A1). This gave a result of - 2.246 and that seems to

> > be the reduction they put on the report. I believe some labs will use

> > the standard calculation (1.04 in Stanford's case) if the patient did

> > not have their personal baseline PCR result for future comparison.

> >

> > I have heard that the PCR equivalent of a Complete Cytogenetic Response

> > is between 1.5 and 2.0 log reduction.

> >

> > Hope this helps

> >

> >

> >

>

> --

> Marcos Perreau Guimaraes

> Suppes Brain Lab

> Ventura Hall - CSLI

> Stanford University

> 220 Panama street

> Stanford CA 94305-4101

> 650 329 9920 x 305

> 650 630 5015 (cell)

> marcospg@... <marcospg%40csli.stanford.edu>

> montereyunderwater@... <montereyunderwater%40gmail.com>

> www.stanford.edu/~marcospg/

>

>