help on accurately preparing 1ug/ml drug in oil mixture

[Peter Apps]

Apart from a rounding error; 0.3366 should be 0.3367, your calculation is correct.

But you still do not have a calibration for the analysis. You cannot measure the quantity of API in the capsules by HPLC unless you have a calibration.

Can you detect 3.36 ug of API dissolved in whatever volume of TFA and ethanol you use to extract the capsules ?

If I understnad correctly your capsules can hold a miximum of 1 ml of carrier, calculate the volume of 1g of carrier plus API and you will see that there is a problem.

Peter

[darasc0211]

Hi, Peter:

Now the client agreed to increase the dose to 3ug/ml so I though I could directly weigh and dilute. Here is what I did:

1. prepare a solution containing 8.9g OA and 10mg BHT, 2. decide the density of the solution using volumetric flask, 3. weight 0.3mg API and add API along with weighing boat in the container, 4. add density x 100 ml of the solution prepared in step 1 to the API.

But this doesn't give me a very good result. I have 109%, 85%, 102% so they are not consistent. Maybe determine density with volumetric flask was not accurate enough??

Next I will completely ignore density determination and follow what we've discussed by diluting from a higher concention solution by w/w. i.e.:
1. weigh 3mg and add a solution containing 8.9g OA and 10mg BHT in solution, w/w=3mg/(8.9g OA+0.01g BHT+0.003gAPI) =3/8.913=0.3366mg/g=336.6ug/g (Could you check this calculation for me? I mentioned in my previous email but you didn't address it. This will directly affect my calculation on HPLC assay% result. Thanks).
2. then weigh 1g of this solution and add up to total of 100g.--this should be 3ug/ml formulation, so w/w=336.6/100=3.36ug/g.
3. weigh 1g of 3.36ug/g into a capsule, I think I can accurately weigh 1g so my final API in capsule shouldn't differ too much.
4. Add 1 capsule to volumetric flask and extract with 0.2% TFA and Ethanol and analyze by HPLC. I am also in the process of developing a stability indicating method, regarding to your question if it's doable to such a impurity level, I don't know and we'll discuss with client as it goes.

Could you confirm with my calculation again.

I appreciate your help.

So, to summarize so far:

Your client wants you to put exactly 1ug of API into each capsule, then take it out again and assay it for API and impurities. And the customer is always right !

You can break this down into two parts: part 1 putting the API and carrier into the capsule, part 2 assaying API and imputities.

The API that goes into the capsule has to be dissolved in oil+BHT. As long as there does not have to be exactly 1 ml of oil in the capsule you can make gravimetric dilutions similar to the scheme we have already sorted out, recording actual weights at each stage, calculate the mass fraction of API in carrier, and then weigh an appropriate amount of API in carrier into each capsule to give you 1 ug of API inthe capsule. That is just simple arithmatic.

If you want to assay the API in each capsule you have to calibrate the analysis, which is not part of your current plan.

What concentrations of impurities do you want to detect ? Say 1% for argument's sake, you then have to detect 1% of 1ug of impurity = 10 ng of impurity in each sample. Can you do that ? And of course that analysis has to be calibrated as well.

Peter

[darasc0211]

Hi, Peter:

You are right. My capsule can only hold 1ml, so instead of weighing 1g of 3.37ug/g, I should weigh 3ug/3.37ug/g=0.893g into each capsule.

As I mentioned in my previous email, I will add 1 capsule (contains 3ug API) to a 25mL volumetric flask, then add 5ml of 1% TFA and QS with Ethanol. This gives me a concentration of 3ug/25ml=0.12ug/ml. The solution seems clear. My peak area of this concentration is around 20. My other problem is that I get lots of peaks from oleic acid which interferes with my API band so next step I need to resolve these peaks from each other. (I posted another thread seeking for solutions on getting rid of oleic acid peaks, do you have any great dieas?)

I am working on checking if the method I am trying to develop can get me a 97-103% recovery at this low dose, and I need to validate linearity, precision etc and see if it' doable at this low dose. But if it doesn't work I'll have to talk to the client to increase their dose strength. So that's the current game plan.

I am not too confident about this but we'll play by ear.

Thanks a lot.

Apart from a rounding error; 0.3366 should be 0.3367, your calculation is correct.

But you still do not have a calibration for the analysis. You cannot measure the quantity of API in the capsules by HPLC unless you have a calibration.

Can you detect 3.36 ug of API dissolved in whatever volume of TFA and ethanol you use to extract the capsules ?

If I understnad correctly your capsules can hold a miximum of 1 ml of carrier, calculate the volume of 1g of carrier plus API and you will see that there is a problem.

Peter

[darasc0211]

Hi, Peter:

There is no purity factor for this API, so in order to accurately quantify the assay%, I should use the API that's used for the actual formulation as reference standard, correct?

I am asking because the client wanted me to use a USP reference standard to bracket my sample because he doesn't have enough API, but because the purity factor of the API is unknown, this could cause some calculation error. The purity factor maybe ignored if I were to quantitate a higher sample concentration, but I am working on a 0.12ug/ml so this could cause big calculation error.

Please let me know if I am thinking this right. If so I need to negiciate with my client for more drug to work with.

Thanks a lot for your advise.

[Peter Apps]

Analysis of drugs in a regulated lab is not my area, but as a simple matter of logic (not even chemistry) you cannot assay a material by using that material itself as a reference material - by definition the assay must always come out at 100% if you do.

So the client is correct - you have to have a reference standard that is independent of your samples.

Forgive me for saying so, but I think that you need to work in detail through the whole procedure - starting at the end with a clear and unambiguous statement of what question you are trying to answer - and make sure that there are no loose ends. Having done that, draw up a detailed protocol and have it approved both by the client and by your line manager.

Peter