Dear All,
(first of all: I hope it is alright that I also inquire about this non-nitrosamine CoC-topic)
In our API we are currently facing a limit discussion for a possible bis-Aryl-Azoxy impurity and I’d greatly appreciate some expert opinions on this.
Background: we have a catalytic hydrogenation of an nitroarene as one of the synthesis steps. According to some literature (e.g. Figueras et al.: https://doi.org/10.1016/S1381-1169(01)00151-0) on this type of reaction, bis-aryl-azoxy compounds might be a possible impurity. So we had to look into this.
However, in the guidelines, only alkyl-azoxy componds are listed as CoC, their aryl-analogues are not mentioned. Some literature (e.g. Galloway et al.: https://doi.org/10.1016/j.yrtph.2013.05.005) states that aryl-azoxy compounds can’t undergo the same metabolic activation and are therefore non-carcinogenic - which means they can be treated according to ICH-Q3A.
In the light of this, we consulted an external toxicologist which gave us ppb-limits for the possible bis-aryl-azoxy-impurity which is in contradiction to our own literature research. Obviously, those ppb-limits pose a significant challenge to our analytical department, also budget-wise. Upon discussion, said toxicologist was not aware of the difference between aryl- and alkyl-azoxy-compounds but still stuck with his suggested limits.
Do you have any opinions or further literature evidence on this topic which we could use as a base for our argument? This would be greatly appreciated!
Thanks a lot
Jan
1 Like
Interesting question!
The alkyl-azoxy compounds (listed in ICH M7 CoC) upon metabolism release the alkyl diazonium ion which is the ultimate mutagen/carcinogen, similar to an alkyl diazonium ion formed from a dialkyl nitrosamine (although through a different metabolic pathway).
On the contrary, metabolism of a bis-aryl-azoxy-compound should result in the formation of an arenium ion which cannot be a potent DNA mutagen/carcinogen owing to the steric hindrance. Same is the case with diaryl-nitrosamines whose carcinogenic potency is significantly lower than the dialkyl-nitrosamines, an argument sufficient to exclude them from the CoC. The Galloway et al. article you have cited can indeed be used in your argument.
Unfortunately, there seems to be a single bis-aryl-azoxy-compound listed in the Lhasa Carcinogenicity Database https://lcdb.lhasacloud.org/study-information/44506744 (3,3’,4,4’-Tetrachloroazoxybenzene) with no reported TD50 for a read-across.
I hope this helps.
2 Likes
Thanks a lot for your input - I was not aware of the 3,3’,4,4’-Tetrachloroazoxybenzene … this definitely complicates our position of argument.
(Edit: do I read the entry in the Lhasa-database correctly that for a 10 mg/kg dose, up to 10/10 mice died?)
In general i am quite surprised that despite the high number of catalytic hydrogenations being applied in API synthesis, this azoxy-topic seems rather underrepresented. This makes me wonder if this is a realistically occuring impurity at all … yet I have still to find some solid evidence about that.
My pleasure.
Lhasa considers the study results on 3,3’,4,4’-Tetrachloroazoxybenzene, as “**Unreliable-**Significant deviations from guidelines”. Lhasa Carcinogenicity Report_Wed, 08 Oct 2025 09_11_45 GMT.pdf (102.2 KB)
The dose response (mortality) in mice does not seem to be linear (study 2 of 4, pg # 5/9). In rats, the mortality was 9/10 & 10/10 at the highest tested concentration of 30 mg/kg per studies 3 & 4 respectively. Since no TD50 has been reported in any of these studies, this data is not useful to assess the carcinogenic potential, not to forget Lhasa considers this data unreliable.
Personally, I would still consider the bis-aryl-azoxy-compounds outside the CoC because they are not listed in the ICH M7 cohort. The possibility of metabolic reduction of a bis-aryl-azoxy compound to an azo compound (possible carcinogen) which can subsequently generate an arylamine (a potential mutagen but outside the cohort of concern) can be explored. The key step would be to identify the potential metabolites of the bis-aryl-azoxy-compound and the mechanistic pathway to mutagenicity if any.
As there is no carcinogenicity data available for aryl-azoxy-compounds, I would say it is unnecessary and unfair to place them in the CoC.
Your insight on the risk of azoxy-compounds in chemical API synthesis is indeed thought provoking!