Extra peaks in alfacalcidol formulation

[santech]

Hi
This is related to alfacalcidol analysis by normal phase HPLC. I have been getting a peak in my stability chromatograms which is growing constantly at every interval of stability. But surprisingly assay of alfacalcidol is 100% which confirms there is no degradation even in accelerated stability samples. where is this peak coming from? Is this due to some excipient-excipient interaction? It appears in innovator sample also but relatively smaller in size? I have tried all the possible ways by injecting the placebo, stressed placebo, diluent etc. But still there is no clue. Should this peak be considered as impurity and the product rejected though the assay is ok. or can a reverse phase method help?

[tom jupille]

What you need is an "orthogonal" method; one which has very different selectivity from the one you are using (and, yes, for a normal-phase method, reversed-phase is likely to be orthogonal). You may well have additional impurity / degradant peaks coeluting with your API.

Another possibility is that you are seeing a degradant which has a much higher UV absorbance than your API, so that a small amount gives a disproportionately large response.

[santech]

What you need is an "orthogonal" method; one which has very different selectivity from the one you are using (and, yes, for a normal-phase method, reversed-phase is likely to be orthogonal). You may well have additional impurity / degradant peaks coeluting with your API.

Another possibility is that you are seeing a degradant which has a much higher UV absorbance than your API, so that a small amount gives a disproportionately large response.

Thanks for ur reply. Should that be considered as impurity and the product rejected?

[tom jupille]

If the only criterion is the assay value, and it passes, then it is "acceptable" on that basis. HOWEVER, if a regulatory agency ever looks at your data and sees an additional "unknown" peak which you have ignored, you and your company will be in *deep* trouble.

Your company should have an SOP in place specifying action in the case of unknown/unexplained peaks. Follow it.

[santech]

Hello

Can that peak be hidden by some wavelength programs, as we feel that it does not pose a threat to the product, just to present the regulatory agency.

[tom jupille]

There is a word for that. It is "malfeasance".

[ym3142]

To me, you need a specification for this impurity unless you can demonstrate it is not a degradant.

By the way, how much it is ?