R advice/requirements for degradants?

[Flappytango]

Hi,

Im trying work out a single method for GMP use that will quantitate an active along with 7 of its degradants.

two of the degradants are oxides cis+trans, they are the problem..

I can get baseline resolution (they are the first two peaks to elute) but a big goal for this method is short runtime.

what is the minimal resolution i need on these two peaks?

Is it be allowable to run them out as one peak and quantitate them together?

[shaun78]

Minimum resolution required by USP/EP/BP/JP/ICH/CEDR is 2.0. Some people will split a hair, say resolution must be 2 (notice no decimal) and thus allow for a resolution of anything larger than 1.5 to be acceptable. How that is delt with is best determined by your QA department.


As for running the peaks together, that depends upon your process parameters and if it is important to know how much of each isomer is around. A lot of times one isomer is more toxic/less effective than another and people want to know exactly how much of each is present. However, if this is not important then there is no problem with combining the cis/trans isomers as one peak.

[mbicking]

While a resolution of 2.0 will be adequate in many cases, if the size ratio of the two peaks becomes too large, significant errors could result. If you need to determine the amount of each isomer present, then I would set a resolution goal of 3 to minimize errors, as long as one peak is at least 10% of the other one.

If you don't need to know each separately, then combine the peaks, as noted above.

And if run time is important then you should investigate the use of one of the "fast LC" columns that are available. You should be able to get the same stationary phase but in a smaller particle size - e.g., switch from 5 um to 3 um. The column length is usually less, and is operated at a higher flow rate, so you will see a significant decrease in retention time, but similar resolution.

Of course, your LC must be in good working order and have low extra-column volume if you want to be successful.

[Dan]

I have seen instances where an incomplete resolution of isomers has been acceptable. You would need to check with your regulatory affairs and QA groups (if you are in a regulated area) to check on their opinion. It will depend on the specifications for the degradants which is in turn set by the toxicology of the compounds (the two isomers may not necessarily have the same toxicity).

However, as you have already shown that you can get resolution, then the RA and QA groups may not allow for a combined peak/quantitation. The FDA, EMEA, etc would also want to see the peaks resolved if it can be done.

As others have noted, a resolution of 2.0 is recommended.

I will second the comment from M. Bicking, there are ways to keep the resolution while getting a shorter run time.

Regards,
Dan

[Flappytango]

Thanks for the replies...

what has been said pretty much confirms what i suspected in that the peaks may be combined but some may frown upon doing so.

we have a good deal of experience with these compounds, understand their toxicity, and the amouts that they are present in our products.

we are already working with small particles and low dwell volume instruments.

The current challancge is to elute these isomers together (with good peak shape) while keeping k reasonable.

[mbicking]

If you are not getting adequate resolution with your current system, and you don't want to increase your run time, then your only real option is to change your selectivity - that is, chainge the "chemistry" of the system. That means change either the mobile phase or stationary phase.

Since your two isomers are a cis/trans pair, you could look for a stationary phase that has some shape selectivity. I believe there is a C30 phase that claims such a property, and perhaps others here could suggest some alternatives. But many other conventional phases may also work, including polar embedded, phenyl, etc.