Reporting Impurity values

[kspansheria]

Dear friends

I have dual component product and both API has their individual impurities. My problem is related to reporting of impurities. For API (A) LOQ value is 0.09% and same LOQ is there for other component i.e. API (B). Now when I report individual impurities of APIs it has following LOQ values

API (A) LOQ API (B) LOQ
Impurity 1 0.10% Impurity 5 0.09%
Impurity 2 0.08% Impurity 6 0.09%
Impurity 3 0.10% Impurity 7 0.09%
Impurity 4 0.09% Impurity 8 0.09%

Now, I am giving LOQ value for total is 0.09%. My all impurities are BLQ, for impurity1 I got 0.095% which is BLQ but as my total impurity LOQ value is different then I got confused, whether i should consider it or not.

Please let me know how I can give LOQ value for total impurities as well as for single maximum unknown impurity.

Regards
Kamlesh

[krickos]

Hi

Please check the ICH Q3B guideline, it gives recommendations including examples with real values including how to handle like 0,095 values.

For total impurities (Not more than 0,5% is not uncommon for APIs/drug substances), it is usually OK to have 0,1% or 0,10% as LOQ/lower report limit. Hower note that ICH Q3B and ICH Q3A(drug substances) has daily dose related recommendations.

www.ich.org click Q in top right corner.

[gtma]

For development activities, you may want to report impurities values below the LOQ for trending purposes. For other activities (e.g. commercial activities), you'll have to set up your method/chromatographic software to exclude data below the LOQ. You may find the stability trends to be quite variable with this exclusion approach.

[tom jupille]

For development activities, you may want to report impurities values below the LOQ for trending purposes.

If your quantitation is sufficiently precise that you can identify trends, then you are, by definition above the LOQ.

[krickos]

For development activities, you may want to report impurities values below the LOQ for trending purposes. For other activities (e.g. commercial activities), you'll have to set up your method/chromatographic software to exclude data below the LOQ. You may find the stability trends to be quite variable with this exclusion approach.

Agreed. I see those guidelines as minimum requirements, for development (when final dose might be unknown etc) and trending (not just stability, you just might want to spot a potential issue a tad earlier).
For drug substances, impurity profiling also need to be kept in mind.

[gtma]

when ICH refers to ICH reporting limit of 0.1%, it implies 0.10% or 0.055% to 0.144%? And ICH ID limit of 0.2% refers to 0.20%? Also, would it be acceptable to have two unknown peaks co-eluting as long as they are below the ICH ID limit?

Thanks in advance.

[krickos]

Well as an analytical devolpment scientist you do need to keep an Eye on ICH Q2 validation, ICH Q3A-C and coming D, but also ICH Q6a. Plus market specific PGI guidelines if appliceble.

REading those all together, any limit below 1,0% would be with 2 decimal places, ie 0,50-0,10% etc as per todays standard.

ICH minimum report limits seems/are absolute ie no rounding up, as shown in examples in ICH Q3B.

[gtma]

Thanks krickos. I also interpret it as such. Interestingly, I found it difficult to get clarity even from my colleagues regarding what the ICH reporting limit of 0.1% implies....does it refer to 2 decimal places? Validation at 0.055% can be more challenging than at 0.10% in some cases.....assuming the ICH reporting limit is 0.1%. I saw a slide from the Canadian agency a while ago referring to the ICH reporting limit as 0.10% (2 dec places). I find it interesting that ICH recommends reporting to 2 dec places below 1.0% but use 1 dec place for the ICH limits. I hope someone can explain this to me.

[krickos]

Quick answer is likely no.

Different approaches it seem. FDA seems to go for 2 decimal places while EU countries seems to accept one decimal place at least in some cases.

In Ph Eur, when applying ICH Q3A principles, 0,10% for any unspecified impurity was introduced for new and old drug substance monographs.
For old monographs specified impurities retained one decimal place, new monographs have 2 decimal places below 1,0%.

So my guess is that the trend is 2 decimalplaces especially for new drugs.

As for validation, especially if you revalidate some old drug Product with existing procedure, yes I agree 0,05% could be challeging in some instaces, would recommend to aim at 0,05% and use 0,10% as fallback position especially for exsiting procedures, if you are developing then you have usually have a larger freedome to work more with the sensitivity.

[sisay]

Hi all

I found this interesting topic regarding the reported data of impurities in drug product. Currently I faced the similar situation but the qualification threshold. Here I use a proposed example similar to Q3B example to describe as below:

CASE 1
MDD:300 mg
Reporting threshold:0.05% (assumed case)
Qualification threshold:0.2%
Raw result:0.2477%
Reported result:0.25%(same decimals to Reporting threshold)
Actual reported result: 0.3% (0.25%) (i.e. second time to round into one decimal to consistent with Qualification threshold and report two values) or just report 0.25%?
TDI:1 mg
Qualification (not exceed 0.2%) :None

CASE2
MDD:300 mg
Reporting threshold:0.1%
Qualification threshold:0.2%
Raw result:0.2477%
Reported result:0.2%(same decimals to Reporting threshold)
Actual reported result: 0.2%
TDI:1 mg
Qualification (not exceed 0.2%):Yes

In these two cases, different reporting threshold will result in different qualification result. Am I right? What I confused is that generally how should I report the data when report and qualification threshold are different in decimals. Because Q3B did not use percentage in Qualification threshold as an example.

Hope any of you still follow this topic and provide me any information. Thank you in advance!

[gtma]

I would recommend reporting results to 2 dec places but avoid double rounding. So in your example, report "0.25% (0.2%)" or if you're using LIMS, report to 4 dec places to avoid confusion.