deepakpangavhane
☆    

India,
2018-09-03 10:28
(2033 d 04:27 ago)

Posting: # 19205
Views: 4,945
 

 recoveries more than 100% [Bioanalytics]

Hi everyone!

The % recovery was 107.80%, 112.65% and 105.92% for the low quality control (LQC) and middle quality controls (MQC2 and MQC1), respectively. Whereas % recovery for higher quality control (HQC) was 98.25%.

The % recovery was determined by comparing the mean peak area of 6 replicates of extracted quality control samples at High, Middle and Low concentrations against respective mean peak area of 6 replicates of un-extracted quality control samples at High, Middle and Low concentrations. Internal standard used was deuterated.

Kindly confirm recoveries markedly higher than 100% for LQC, MQC2 and MQC1 are acceptable as per USFDA guidelines?? Also specify reason for higher recoveries.
Ohlbe
★★★

France,
2018-09-03 21:21
(2032 d 17:34 ago)

@ deepakpangavhane
Posting: # 19231
Views: 4,353
 

 recoveries more than 100%

Dear Deepak,

❝ The % recovery was determined by comparing the mean peak area of 6 replicates of extracted quality control samples at High, Middle and Low concentrations against respective mean peak area of 6 replicates of un-extracted quality control samples at High, Middle and Low concentrations.


Meaning that what you observe is a mix of extraction recovery and matrix effect. If you want to see the true extraction recovery you have to compare to samples spiked post-extraction. You can combine with matrix effects experiments.

❝ Internal standard used was deuterated.


Makes no difference. You are looking at peak areas, not peak area ratios.

❝ Kindly confirm recoveries markedly higher than 100% for LQC, MQC2 and MQC1 are acceptable as per USFDA guidelines??


Did you see any limit for absolute recovery in either direction in the FDA guidance ?

❝ Also specify reason for higher recoveries.


Multiple. The most probable: analytical variability (your results are well within the 15 % acceptance limits for precision and accuracy). Other possibilities include ion enhancement, sample preparation, calculation errors...

Regards
Ohlbe
Ladi
☆    

Thailand,
2018-09-04 05:51
(2032 d 09:04 ago)

@ deepakpangavhane
Posting: # 19232
Views: 4,278
 

 recoveries more than 100%

Hello Deepak and Ohlbe,

❝ The % recovery was 107.80%, 112.65% and 105.92% for the low quality control (LQC) and middle quality controls (MQC2 and MQC1), respectively. Whereas % recovery for higher quality control (HQC) was 98.25%.


My lab called the 'post-extraction' as 'un-extracted', I am guessing you also mean that?

I observed recovery 100-115% mostly with protein precipitation projects in our lab. My explanation to auditors was that the extracted samples are little bit more concentrated than the post-extracted samples because some protein is precipitated out in extracted samples. While in the post-extracted samples, the measured volume of supernatant from precipitated blank is use to make up a required concentration. Ususally for protein precipitation, recovery is almost 100%. So if extracted is more concentrated than it supposed to be (expected conc.), recovery may cross 100%. However, I normally see similar %recovery across all concentrations.

Regards,
Ladi
Kintan Patel
☆    

India,
2018-11-29 12:48
(1946 d 01:07 ago)

@ Ladi
Posting: # 19659
Views: 3,901
 

 recoveries more than 100%

Hello Ladi and Deepak,

The % recovery across all concentration levels (i.e LOW ,MED add High) should be consistent that means %CV should be within 20% across all levels and also you should mentioned the accepted % recovery range (i.e from 30% to 120%) in your SOP and this high % recovery might be due to ion enhancement , sample preparation error etc. and also it is acceptable as per FDA guideline.(limit for % recovery is not mentioned in FDA Guideline)

Thank You,
Kintan Patel


Edit: Full quote removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post! [Ohlbe]
Ohlbe
★★★

France,
2018-11-30 19:23
(1944 d 18:33 ago)

@ Kintan Patel
Posting: # 19665
Views: 3,900
 

 Origin of figures ?

Dear Kintan Patel,

❝ The % recovery across all concentration levels (i.e LOW ,MED add High) should be consistent that means %CV should be within 20% across all levels


Could you clarify where you took this value of 20 % from ?

❝ and also you should mentioned the accepted % recovery range (i.e from 30% to 120%) in your SOP


Why 30% to 120% ? Where do these values come from ?

Regards
Ohlbe
Kintan Patel
☆    

India,
2018-12-05 11:04
(1940 d 02:51 ago)

@ Ohlbe
Posting: # 19668
Views: 3,869
 

 Origin of figures ?

Dear Ohlbe,

❝ ❝ The % recovery across all concentration levels (i.e LOW ,MED add High) should be consistent that means %CV should be within 20% across all levels


%CV for % recoveries across Low, med and high level shall be within 20 %.

❝ Why 30% to 120% ? Where do these values come from ?


This is just example. You can keep accepted % recovery as 25 % to 125 % etc. in you SOP,but in guideline % recovery criteria no where mentioned but many industries had included acceptance criteria of % recovery in their SOP.

Thanks ,
Kintan Patel


Edit: Full quote removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post! [Ohlbe]
Ohlbe
★★★

France,
2018-12-05 11:30
(1940 d 02:25 ago)

@ Kintan Patel
Posting: # 19669
Views: 3,786
 

 Origin of figures ?

Dear Kintan Patel,

❝ %CV for % recoveries across Low, med and high level shall be within 20 %.


Why 20 % ? Where did you take that value from (apart from your own SOP) ?
How do you calculate the CV: take all individual values ? Or the average at L, average at M and average at H level ?

❝ ❝ Why 30% to 120% ? Where do these values come from ?


❝ This is just example. You can keep accepted % recovery as 25 % to 125 % etc. in you SOP [...]


Right. Then e.g., not i.e.

Regards
Ohlbe
UA Flag
Activity
 Admin contact
22,957 posts in 4,819 threads, 1,639 registered users;
79 visitors (0 registered, 79 guests [including 8 identified bots]).
Forum time: 13:56 CET (Europe/Vienna)

Nothing shows a lack of mathematical education more
than an overly precise calculation.    Carl Friedrich Gauß

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5