Method validation

[brbabu_1979]

Hi every one, I have one doubt about method validation. If you have three dosage forums (like 10 mg/ml, 15 mg/ml and 20 ml/mg), at the time of analytical method validation by HPLC which form we should select for the validation? Why? Are there any recommendation guidelines for this? Please explain me. Thanks in advance.

Babu

[Consumer Products Guy]

Hi every one, I have one doubt about method validation. If you have three dosage forums (like 10 mg/ml, 15 mg/ml and 20 ml/mg), at the time of analytical method validation by HPLC which form we should select for the validation? Why? Are there any recommendation guidelines for this? Please explain me. Thanks in advance.

Babu

I'm a lab scientist and not a validations pro, but here's what I'd do:
I'd do a single validation that covers from 80% of your low dosage to 120% of your highest dosage. LOD & LOQ, robustness, specificity should relate to your lowest dosage.

At this point are you adjusting the sample sizes so that all dosage forms will net out as about the same concentration, and then use a single calibration standard for all three, or will you be keeping your sample prep exactly the same and just using your extended linearity studies to validate that? You should decide that first.

[rindtr]

Hi,
I am a food engineer working for a food control lab and have no experience on pharmaceuticals but the question should be asked about your doubt is what you expect your method. is the method for detecting a banned substance or is it for controlling the exact amount of substance which is in a formula of a spesific drug? if it is for a banned substance you should define your LOQ and do your validation around it. if definite doses are the case you should focus on those amounts and form a calibration curve which includes them and do your validation studies at the top bottom and middle point of your calibration curve. I suggest to the limits of the calibration has no importance as far a your calibration is linear.

good luck

[michaelcarolus]

Hi

I agree with the "Consumer Products Guy" with regard to the range for the validation. Hopefully the different the 3 different dosages have the same excipients. If the excipients are different you will need to do 3 validations. If they are dose proportional then you can do all 3 dosage forms together. Under the Method Precision part of the validation we would do Method Precison for all 3 dosage forms. Although, it might not be required, we feel that doing it for all 3 dosage forms gives you the confidence that the method will work for all 3.

I hope this helps. Cheers

Mike

[brbabu_1979]

Hi, Thanks to Tom, rindtr and Mike for your valuable replays. Ours is oral solution which is having dose proportional. I can understand what you guys were telling. What I am trying to know is when we come across to choose one of the dosage forms from out of there or four for method validation; simply we do choose any dosage form and perform the validation or any guide lines tell us to select this dosage form and do. And also do some limit parameters to other dosage forms. I tried to find in ICH but I couldn’t. We you guys come across please help.

Thank in advance
Babu

[hareshami]

By logic, you have to choose middle level (15mg/ml) doses.

Regards,
Haresh B. Patel

[analyte]

I am not a validation expert but I think you have to make validation for each of them separately. Although they have similar excipient profile they are different formulations, and you have to prove that your method works on each of them.

[Karen01]

I am not a validation expert but I think you have to make validation for each of them separately. Although they have similar excipient profile they are different formulations, and you have to prove that your method works on each of them.

If the method can be shown to be specific for all formulations in the presence of the highest amount of each excipient then I think it should still be one validation (albeit with a bit more work).

I agree that you should adjust the sample prep to give about the same concentration of API for all dosage forms and validate the method that way... it would still be a heck of a lot less work than separate validations and it COULD be a huge time (and thus money saver) to be able to to test all the dosage forms in a single run/sample set/sequence (whatever terminology you use) raher than having to do separate runs for them with their own standards and QC's and paperwork!

In pharma method development, those practical considerations are almost as important as the science. And it will make the QC group happy with you .. which is very important!


Karen (who used to manage Pharma Analytical Development)

[analyte]

I agree that shortcuts are important in analytical method development and validation, but it is also important to prove that the proposed shortcut works in every condition as a robust alternative to the “long way”.

As all of us already know that especially in pharmaceutical field, the most important thing is not the lovely and happy looks of your QC team, or the money or time you gain; the robustness of your method and closeness of your results to the true values come first.

I think that if someone omits some validation experiments to make “shortcuts”, one day it is possible to see that the method be lack of required analytical quality especially for batch-to-batch assays, related substances or bioavailability studies.

I do not say to perform excessive or unnecessary experiments, but if there are some requirements to be met and some experiments to be done, we have to do them.

Otherwise, perform just a single set of std. addition at 100% (instead of 3 sets x3 different percentages) for accuracy, a set of 5 independent experiments (instead of 3 sets x 10 exp.) for precision, a single set of 3 points calibration (instead of 3 sets x 5 points) for linearity, etc., and go… Is it short enough to make happy all of us?

[miro2009]

Hello experts,
I'm having the same case of brbabu, I need to perform validation of an HPLC method for quantification of main analyte, we have 2 main sosage forms 7.5 and 15 ug/ml, in addition the methos will be used to quantify final bulks during the manufacturing process (the starting material from which dilution is made to 7.5 and 15 ug/ml), and this final bulk usually has variable concentrations, but usually in the range of 60-70 ug/ml.
Can anyone suggest any protocol fot this? So far I know the following, our working standard has a conc of 100 ug/ml, so I'll make acal. curve with the highest level possible to obtain a linear curve, to cover the 60-70 range (linearity done)
For the range and accuracy, can anyone tell me what's the difference between them? Shouldn't the range cover also accuracy? I usually perform 3 replicates in the 50%, 100% and 150% of target. In this case, is there a suggestion to cover the 2 dosage forms and the final bulk in minimized steps? For eg. 3 replicates of the following:
1)50% of the 7.5ug/ml dose,
2) 150% of the 15 ug/ml dose
3)100% of 7.5ug/ml, 100% of 15ug/ml, and 100% of 70 ug/ml

Would this be enough to cover range and accuracy?

Thank you in advance!

[AA]

Myself, I would be validating all 3 dosage forms. Why, because the risk of failing an audit and getting slapped with a warning letter or, even worse, having to recall product because of some issue with a dosage form I didnt validate for, far outweighs the effort of validating all 3 dosage forms. It comes down to the risk you (or your company) are willing to take.

[pliva]

Myself, I would be validating all 3 dosage forms. Why, because the risk of failing an audit and getting slapped with a warning letter or, even worse, having to recall product because of some issue with a dosage form I didnt validate for, far outweighs the effort of validating all 3 dosage forms. It comes down to the risk you (or your company) are willing to take.

If the excipients and process of manufacturing are the same, you can do it for the lowest value. I dont know about other regulatory, MHRA accepted my project in 2009. All the best. After all , it is all about taching chance:) and how much your boss want to negotiate to take the risk.

[miro2009]

I also agree that all concentrations should be validated and not only focus on 1 dosage, this is why I suggested to take a point 50 or 70% lower than the smaller dose, and a point 130 or 150% of the higher dose, in addition to 100% of each dose, shouldn't this cover all the range of interest? So in total, I would do 3 replicates of 4 different controls. As for the final bulk, this is an inhouse requirement in the lab so is not of major concern now.

Another seperate question about validation protocols generally:
What is the difference between range and accuracy in terms of actual procedure to be done for these 2 items? Generally, for the range, we perform 3 replicates of 50%, 100% and 150% of target and calculate % recovery. What additional should be done foe the accuracy item??

[mac]

To reply to the OP, I personally would validate each dosage form separately, three separate validated methods. It's better in the long run as it eliminates any doubt about your methods. If the sample matrix of the three dosage forms is the same, this shouldn't be that much hassle technically, just a bit of extra testing. Your documentation should also be very similar. Best of luck!