Back to the Top
Dear all,
I have a very specific question for the analysis routine in bioanalytical labs.
I would like to know about the use of smooth for chromatograms in bioanalysis using LC-MS/MS technique. As far as I know the smooth is a tool provided by software instrument for helping the analyst to integrate into the appropriate way all the chromatograms throughout an analytical batch. Is there any guidance from any regulatory body for using this tool?
Thanks in advance,
Rafael E. Barrientos Astigarraga, PhD, MBA, BChem
MAGABI Pesq. Clin. Farm. Ltda.
Back to the Top
The following message was posted to: PharmPK
For a validated method, as long as you keep consistent, and not change from day to day, it should be acceptable.
Back to the Top
The following message was posted to: PharmPK
Dear Rafael,
You are absolutely right for the use of smoothness in chromatogram integration.
There are no regulatory guidelines for that. The only thing you have to keep in mind that, choose smoothness factor correctly so that
1. Two closely eluting peaks must get separated
2. Your chromatogram data point must content more than 15 to 20 points, for a good chromatogram. In this case Dwell time also plays a major role. You can choose Dwell time by using the formula Dwell Time= 2000/ No. of MRM
Thanks,
Sudipta Basu
Senior Research Scientist
Sai Advantium Pharma Ltd, Pune, India.
Back to the Top
The following message was posted to: PharmPK
Dear Rafael,
Regulators do not restrict the way use integration parameters. But they
do insist that integration parameters should be same throughout the
batch. However following points are important:
1. Use minimum possible smoothing.
LCMS peaks are typically scatchy peaks
Gas parameters and spray position adjustments help to produce
more smoother peaks, so that you do not need to use too much
smoothing using software.
2.Use minimum possible bunching factor. Typical adjacent peaks get
merged with high bunching factor
Regards
Vinayak
Back to the Top
Dear Rafael,
This is an interesting question and many of the bioanalytical scientists don't pay much attention to the different peak integration parameters provided by different manufacturer's software.
I have not come across any regulatory guidance document which states about the dos and donts of peak integration and it is more governed by an in-house SOP in a GLP lab set up. In my opinion, these features are very much software centric and different chromatography data softwares have different peak integration parameters which makes it more difficult for the Regulatory agencies to have a control over this. What is of paramount importance is consistency while applying the integration parameters to the entire data set of a particular batch. In fact, it is the best practice to fix the peak integration parameters based on the initial pre-validation and/or validation results and apply the same integration parameters during the routine analysis for that method (of course to the extent possible). But it is mandatory from regulatory perspective and without doubt that all the integration parameters are kept uniform throughout a batch quantitation. However, if one or two chromatograms are not integrated correctly wi
Back to the Top
The following message was posted to: PharmPK
Dear all,
In my opinion smoothing should not influence the signal to noise ratio
of your crude data. In other words: if noise in a chromatogram
disappears as a result of heavy smoothing, you're actually changing
sensitivity of your assay. This should not be the case.
Sincerely,
Rob ter Heine
Back to the Top
Dear Neel, Vinayak and Sudipta,
Thank you very much for your contributions on this specific issue. I do agree with most of them such as: to use the same smooth value within a batch, smooth depends on the HPLC peak behavior, the bunching factor is also important for the integration process, data point across the peak, etc.
But I still have a couple of questions as follow:
- what is "the minimum possible smooth"? For each instrument and software there is a specific scale for smooth. Also the algorithm is different for each manufacturer. I do understand that higher smooth can cover some odds situation in HPLC (peak splitting for instance) but how can I determine the minimum value for smooth? You may consider that an LC-MS/MS chromatogram use to show a single peak.
- in our lab, we use Analyst software and the scale for smoothing range from 3 - 41 (0-3-5-7-9-11-13 and so on) using the IntelliQuan algorithm. Is there any way to prove that smooth 3 better than 9 for example? Do I should always use smooth 3?
I know that this is a difficult question but just to clarify: we have got some questions from the regulatory agency (ANVISA) asking why we used smooth 9 for the integration of a specific analytical batch. Important detail: we use the same smooth for the whole sequence and I do not need to perform any reintegration or manual integration.
Best regards,
Rafael E. Barrientos Astigarraga
MAGABI Pesq. Clin. Farm. Ltda.
Back to the Top
Rafael,
There is nothing in the FDA or the European guidelines about the smoothing or the bunching factor. The only thing you need to do is to keep both consistent throughout validation and sample analysis.
The following message was posted to: PharmPKBack to the Top
Rafeal,
I normally start with smooth 2, bunching 2. When peak shows a small
split, (this is not chromatographioc split, but due to scratchy peak
from mass spec, which is quite normal) then I increase smooth stepwise,
may be upto 5. If this does not help, I will increasing bunching factor.
Increasing bunching is not preferred as it will bunch adjacent peaks
within RT range specified by the factor.
You will know why you chose 9 instead of 3. If this was just a generic
choice, you may reintegrate using 3 and show that there is no difference
in results. That should satisfy regulator.
Vinayak
Back to the Top
The following message was posted to: PharmPK
Dear Rafel:
We do not allow any smoothing in our lab. There is no need for
smoothing if the chromatography is good and sensitivity is adequate.
Extensive smoothing (like 9) risks covering up the near co-eluding
peaks like metabolites.
TK Chen
Director DMPK,
Cylene Pharmaceuticals
Back to the Top
The following message was posted to: PharmPK
Dear Rafael,
It's quite uncommon to have question like this from regulatory agency. I have few questions to you,
1. Do you have used different smooth in different batches?
2. Do you have any Standard Operating Procedure (SOP) regarding integration parameters?
3. What is the S/N ratio in the raw data?
4. Is there any co-eluting or late eluting peak near to the analyte Rt?
I'm asking these so to understand why regulatory agency asked this question.
Solution:
1. You can change your smooth to 3 to atleast and see that is there any change in the result. If not you can justify. Hopefully it will not change, as you have applied the same smooth to all and also if there is no co-eluting or late eluting peak near to the analyte Rt.
2. Check your S/N ratio at 3 smooth and in 9 smooth. If S/N ratio increased by increasing smooth, you can justify. It will happen only when the base line noise is very less otherwise ratio will decrease. 3. You can also check the No. of data point at smooth 3 and smooth 9. If data point increased you are safe. For good chromatogram atleast 15 to 20 data points are necessary. If you are using Analyst software, First go to "open data file" then select the individual peak after that go to "view" window and then click "graph information window" option. The software will show No. of data point.
Actually, by increasing number of smooth we are simply manipulating the data. By increasing smooth we are compromising selectivity and that's why the regulatory body have asked the question (as per me). If you can justify that by selecting smooth 9 your selectivity and sensitivity will not affect, you are safe.
Hope I could answer to your query to certain extent.
Sudipta Basu
Senior Research Scientist-DMPK
Sai Advantium Pharma Ltd, Pune India.
Want to post a follow-up message on this topic?
If this link does not work with your browser send a follow-up message to PharmPK@boomer.org with "Smooth in bioanalysis" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)