In USP, unknown impurities are usually expressed as % area of the active drug. For known impurities listed in monographs, there are usually STDs available. How are the (upper) limits set up for the impurities, the structures of which are known but which are not listed in the monograph?

Many thanks in advance.

Normally the limits are set by analysing some of the sample batches of the product but in any case it should not be more than the limit given in USP for any individual impurity.

As a general guideline known inpurities are set limits of less than or equal to 0.3 % and unknowns are set at less than or equal to 0.1 %.
In the generic drug industry you may source your material from a company that uses a different synthetic route to the one used in the monograph, so not all of the monography impurities are relevant. From my experience (Europe/non FDA) as long as we know what the impurity we give it a limit of 0.3 %, even if it is not listed in the monograph,

Suggestion: Since you know the structure of the
compounds can you go to your toxicology department
and get TIEL values and set your anayltical
methods/specifications around them?

What does percentage mean? Is that percentage of HPLC peak area? If the impurties are not UV active or max. absobance wavelengh is not the same as API, how can you tell if there is an impurity?

I got confused and thank you in advance for the explaination.

I have seen different ways of calculating the %. Some will use a standard area count for a reference others will use the % of the main component peak.
It is assumed that the impurities are of similar absorbance as the main component and are calculated off of the main components area count for %. If the identity of the unknown peak is determined then an effort must be made to quantitate it seperate from the calculation of the main component. Unless, they both have the same molar absorbtivity. We have several methods that run at two wavelengths due to the differing UV max. But, these peaks have been identified and are quantitated off a in house standard.
The best way to detect if there are any impurties is using a diode array detector and examine the whole spectrum.

As y'ou're under USP/FDA reglementation all the impurities that are representing more than 0,1% of the total amount must be identified, accuretly quantified and controlled in process (QC).I mean 0,1% measured for example with a chromatogram drawn on Max plot (every point is traced at the maximum absorbance from the spectral image obtained with a diode array detector)

This means you need to use the standard to quantitate the impurity at the maximum absorbance of this impurity compound. What happens if the standard is not available? Can one use the area% of main peak? I know if they have similar absorbance profiles it may work. But if their max absorbance wavelength are apart, how will you do? Thanks.

The FDA wants to see that you are doing your best, that you are putting forth an effort. Even if it may not be absolutly correct. Some assumptions can be made.
If you know what the impurity is but can not get a standard, than you need to make one. You find the cleanist product you can and call that a standard. Inject it and use the percentage of area on the chromatogram report as the potency, minus mobile phase peaks. Then use this standard to calculate the impurity in the product.
If the impurity is not known, than assumptions need to be made about the molar absorbance at the wavelength your main component is being examined at. You have to say they are one to one, hence you can determine if the impurity is greater than 0.1%.
If using this method you determine that the impurity is greater than 0.1% but do not know what it is, than you assign it a name/number and continue on with your validation etc... You want to know what it is as soon as possible, but once again you need to show effort in identifing the impurity. The identification of the impurity is another matter all together.

I have a reference stating that identification of impurities greater than 0.1% is applicable for only synthetic routes and for semi synthetic route product it is from 0.3%

Anonymous: Which is the difference between synthetic route and semi synthetic route ?

usually depends on the type of impurities?as process or degradation.if it is process one can not exceed the limit of 0.1%but if it is degradation one can set the limit based on the stability data available for sucessive lots of the drug product& also knowing the toxicity of the perticuler impurity.

Can i have your comments on establishing the response factor of impurities with respect the main analyte.

The impurity spc is 0.2%. Should we need to take the conc of impurity of 0.2% wrt the main analyte working concentration say main analyte concentration 0.5mg/ml.
0.2% of 0.5mg/ml is 0.001mg/ml.

0.001mg/ml of impurity and main analyte blend solution will be taken.
or should we take 0.5mg/ml of main analyte and 0.001mg of impurity for response factor estimation.

if the individual impurity limit is 0.1% and the total impurity is 0.5 or some times upto 1% is the product acceptable?

I am getting a peak sliced at the apex. It is not due to high concentration or attenuation problem. Is it due to detector time constant? Thanks for the suggestion

Why do you think it is NOT due to attenuation problems?

There is a comprehensive guideline on acceptable levels of impurities issued by the International Conference on Harmonization (ICH).

It is Q3 at http://www.ifpma.org/ich5q.html

Tim Kearsley.

My response to Mr. Shyam Sundar's message:

Respose factor determination is recommeded over the linear dynamic range of the impurity for the chromatographic and detection technique used.

A single point determination does not give assurance beyond precision. But accuracy is critical here, isn't it.

Some recently acquired wisdom from FDA investigators indicate response factors of impurities are determined as a ratio of slopes obtained for the main component and the impurity concerned. Each concentration in the linearity range is injected at least in duplicate and the entire linearity exercise is done in duplicate.

The response factor is calculated as a ratio of the slopes of the impurity versus the main component. The response factor is the average of the two values obtained. You may affix appropriate acceptance criteria. Linearity range includes LOQ to 120% of specification limit. LOQ should be at least 10-25% of the specification limit.

Finally, take a cue from the BP 2001. If your expected response is in between 0.9-1.1, you may delete response factors from your calculations.

Coming to the actual topic of the conversation:

In-house/process impurities are covered under the heading any (other)impurity in most monographs. However, if the impurity is characterised (structurally and toxicologically), the limit may be relaxed up to a limit justifiable by the firm. Care, of course is taken to keep the competitor in mind. Under no circumstance, an unknown impurity can exceed 0.1%, even if mentioned in the monograph. Queries relating to this are sureshot during regulatory submissions.

Thanks,