Peak Fronting - again

[Victor]

Please would someone inform me.

What is the status of USP methods? Are they methods which HAVE to be used for the analysis of a particular drug within the pharma industry? Can they be modified, or does any modification have to be approved by some sort of committee before the change is deemed acceptable?

e.g. If I set up a company to manufacture a drug which is out of patent-let me guess one, aspirin. If I want to sell it to customers in the USA, must I test it by a USP method to show it is o.k.?

[Bruce Hamilton]

Please would someone inform me.

What is the status of USP methods? Are they methods which HAVE to be used for the analysis of a particular drug within the pharma industry? Can they be modified, or does any modification have to be approved by some sort of committee before the change is deemed acceptable?

e.g. If I set up a company to manufacture a drug which is out of patent-let me guess one, aspirin. If I want to sell it to customers in the USA, must I test it by a USP method to show it is o.k.?

I'm not that well informed about the US situation, but, simply put, if you want to sell any product ( including generics ), you must have FDA approval.

The US requires that drug testing is either in accordance with the USP, or alternative methods have full validation and equivalency data that demonstrates the methods equal or exceed the USP. Where no USP monograph exists, you have to provide full validation data as part of the approval process.

When the FDA review all applications and manufacturing, they require compendial ( or equivalent ) methods if available, as well as various other tests, such as bioequivalency and process information to show that impuurities generated in your process are removed, or are below levels of identical impurities in the approved product.

It's far easier to follow the published method, but you can tweak conditions very slightly to obtain specified resolution/retention. You can't do antthing major ( like change from a C18 to C8 column, or vary mobile phase composition by changing a solvent ) without revalidating the method ( expensive ).

You have to detail what minor changes you made, the actual conditions used, and for all assays you have to perform a system suitability test to demonstrate your system meets the defined method criteria.

There are many problems when chromatography is required for impurities or assay. In general, the methods may be submitted by manufacturers, and then have to be validated according to USP Section <1225> " Validation of Compendial Methods".

The first problem, are there traceable standards of the drug and impurities?. Recently the USP had to defer some monograph introduction dates because the standards weren't avaliable.

If the methods have been submitted by somebody who has demonstrated to the FDA that they comply with <1225> then that's a head start. But, the USP review panel could choose another method.

There was no requirement that the assay methods be best practice or even stability-indicating, just that they could be replicated, and met the specified quality criteria, such as accuracy, precision, specificity, detection limit, quantitation limit, linearity, range.

Howver, that process has resulted in a lot of methods, eg for steroids, that are not stability-indicating, and the USP is now looking to address such issues by asking for alternative stability-indicating methods.

So, yes, it's very tightly defined, and mobile phases should be prepared exactly as specified, even if organic solvent is mixed prior to setting the pH. If you deviate, an auditor can ask for the veidence to show the changed method still matches the USP method.

If you want more information, you can check out the compliance section of the FDA web site, and/or sections <621> and <1225> in the latest USP.

Hope this helps,

Bruce Hamilton

[Uwe Neue]

Bruce gave a very good and detailed explanation of the reason why the USP procedures are preferred. I would like to add a few more things.

In the old times, the operating conditions were very strictly specified: column L1, 25 cm x 4.6 mm, 10 micron, flow rate 1 mL/min, solvent composition 66.5% methanol, 33.5% 0.05M potassium phosphate buffer pH 2.5, etc. Today, there is some flexibility in the parameters, but the amount of flexibility is fixed. I do not have the actual currently proposed rules in front of me, but one can change the column length, column diameter, particle size, flow rate etc. within reasonable limits, as long as one still meats the system suitability requirements, which are typically specified as the minimum resolution between peaks or approximate retention times.

While the operating conditions are (more or less) well specified, the column used is not available, unless one purchases the reagent booklet from the USP or subscribes to their website. This causes the problem that methods that had been developed on one column do not work on another, newer column of the same type. There are ways around this dilemma. I learned a long time ago that a 3.9 mm x 30 cm L1 column is always a microBondapak C18 column from Waters (nobody else used this dimension), and you did not need to buy the $250 reagent booklet from the USP to understand the column requirement for some 90% of the assays.

Recently, efforts have been made to specify "equivalent" columns, and the USP is in the process of implementing a website that gives you an idea about "equivalency". The address to the website has been posted here before. It should be noted that this is still work in progress, and the progress is not fast.

[Victor]

Bruce Hamilton and Uwe- Neue thank you very much for these very helpful replies.

Thus it would seem it is not possible to change the pH of the mobile phase without showing some equivalence of the changed method.

However, in this case we do not know if following the USP method exactly will eliminate the fronting peaks-maybe Syx will inform us.

[syx]

The problem is solved! Yes, we should make the mobile phase as USP said: adjust pH after organic addition.



Here are the differences:
Substance S:
before: retention time 4.85 minutes, tailing factor 1.25
after: retention time 5.03 minutes, tailing factor 1.38

Substance T:
before: retention time 6.65 minutes, tailing factor 0.63 (fronting)
after: retention time 12.56 minutes, tailing factor 1.13

Now, we are trying to modify the method. We want to adjust pH of aqueous portion of mobile phase to the pH below 5.9 until we get same condition as in USP method.

[Bruce Hamilton]

The problem is solved! Yes, we should make the mobile phase as USP said: adjust pH after organic addition.
...
Now, we are trying to modify the method. We want to adjust pH of aqueous portion of mobile phase to the pH below 5.9 until we get same condition as in USP method.

Why are you trying?. Surely the lesson of this saga is that the USP procedures should be followed as specified, regardless of how far away from current best practice ( as promoted in this group).

If you wish, you can email the USP and suggest they modify the procedure, then wait xx years for the apporved revision.

If you modify the method, you will have to confirm and validate that your changes will have no effect on the results. The method clearly works, so why bother?.

Bruce Hamilton

[Uwe Neue]

Bruce:
I think, he does not want to modify the method, but just figure out what the aqueous pH would be. Maybe because he wants to have a company- or department-internal consistent method for measuring the pH. I do not think that this is a departure form the USP procedure, but rather a clarification.

[Bruce Hamilton]

Bruce:
I think, he does not want to modify the method, but just figure out what the aqueous pH would be. Maybe because he wants to have a company- or department-internal consistent method for measuring the pH. I do not think that this is a departure form the USP procedure, but rather a clarification.

Cool, but is it going to be easy to change the procedure trivially?. The procedure is " Mix 1400 mL of water, 400 mL of acetonitrile, and 2.0 mL of triethylamine in a 2000 ml volumetric flask. Allow to equilibrate to room temperature, and adjust with 0.2N sodium hydroxide or or dilute glacial acetic acid( 1 in 100) to a pH of 5.9+-0.1. Dilute with water to volume....".

It's outside my area of expertise, but if this was a simple 20% acetonitrile : pH=4 0.05M phthalate aqueous buffer mobile phase, the blend pH would be expected to become about 5.0, but the inclusion of TEA, a thermal equilibrium step, and a subsequent pH adjustment before final dilution seems to make an alternative method potentially more difficult to demonstrate equivalency. It's well outside my skill set, and it may actually be trivial to change in a way that any auditor would accept.

Being lazy, I'd stick with what works, and ensure in-house proceduces are subjugated by compendial methods.

Regardless, the OP has solved their problem courtesy of your advice, and that's the main outcome.

Bruce Hamilton

[syx]

I just feel inconvenience with the procedure, especially with pH adjustment after the addition of organic solvent. Anyway, we should modify the method due to small peaks of syrup's placebo, especially around substance S' peak. It is difficult to adjust mobile phase composition if we mix the composition as the USP.