AI limit for 1-Nitrosopiperazine

, ,

Hello all, I am trying to get a better understanding for the derivation of some nitrosamines which have carcinogenic data. As presented in the FDA and EMA guidelines for nitrosamines regarding derived AI limits, it is my understanding that an AI limit can be derived for a nitrosamine if carcinogenic data is available and “sufficient/reliable” (i.e., well-defined, multiple doses, etc.).

When I looked up the AI limit for 1-nitrosopiperazine, both the FDA and EMA stated that the AI limit for this compound is 400 ng/day based on CPCA. HOWEVER, carcinogenic data with derived TD50 values are available for 1-nitrosopiperazine via the Lhasa Carcinogenic Database which appear to be based on reliable data collectively (see attached picture)

I’m trying to understand: why is the AI limit of 400 ng/day for nitrosopiperazine via CPCA supported, rather than based on its TD50 value?

The Love 1977 study on 1-nitroso-piperazine is not meeting ICH M7(R2) robustness criteria and therefore a broader weight of evidence would likely be sought by regulators to derive a compound specific AI for 1-nitroso-piperazine.

There are also the historically observed TD50 trends within the nitrosopiperazine class (different substitutions) which are not always easy to explain (or were not always linked to the most robust studies) (and maybe elevate the burden on the need for a WoE).

Nonetheless, already in 1992, it has been argued by Lijinsky that mono-nitrosopiperazine might have reduced potency due to the basic properties of the remaining free nitrogen in the 6-membered ring and therefore can be different compared to the class as whole. (Meaning that the acceptance of a higher AI of nitroso-piperazine doesn’t per definition open the way for surrogate use for read-across with fully N-substituted NDSRIs, which is probably what you are thinking about rather than needing a limit for nitroso-piperazine itself).

Whereas 1-nitroso-piperazine was not one of the CPCA validation compounds of Kruhlak 2024, the data on this compound is discussed in the paper.

400 ng/day for 1-nitroso-piperazine (and others of the class) was an advancement, as originally read-across with small nitrosamines was applied for some nitroso-piperazines! For example, for 1-methyl-4-nitrosopiperazine EMA guided AI of 26.5 ng/day based on NDEA readacross (on 1 July 2023 revised to a 400 ng/day CPCA-based limit). Also the fact that there are no differences in CPCA workflow for piperidines compared to piperazines shows some indication that the designers tend to agree that nitrosopiperazines might not be typically and consistently more potent than nitrosopiperidines.

In Dobo 2022 structural group 10 consists of N-nitroso-piperazines and was assigned an overall AI of 153 ng/day based on the lowest robust TD50 in the structural group (1,2,6-trimethyl-4-nitrosopiperazine), suggesting that it would be prudent to assign an AI of 153 ng/day to the N-nitroso-piperazine class.

Bercu 2023 has endorsed this concept and evaluated a 150 ng/day limit for 1-methyl-4-nitrosopiperazine, while this nitrosamine is missing the beta-methyl substituent as present in the surrogate used.

dos Santos 2022 evaluates this to be a highly conservative limit based on a worst-case approach, considering that readacross with the most structurally related compound (N-nitroso-piperazine) would at least suggest an AI of around 5510 ng/day, while predicting that the AIs of N-nitroso-piperazines would typically range between 100 and 3000 ng/day.

The Bercu 2023 monographs also include an AI proposal for N-nitroso-piperazine of 28500 ng/day based on component-specific TD50 data (Love 1977).

Overall, for the N-nitroso-piperazine class TD50 data studied in the Dobo 2022 investigation, it became clear that the only alternative study having a similar robustness compared to the 1,2,6-trimethyl-4-nitrosopiperazine study (at least 2 dose groups, and at least 20 animals per group) was the Love study on 1-nitrosopiperazine.

It remains uncertain if the higher AI proposals (5510 ng/day, 28500 ng/day) will become accepted, but I’m not confident it is the way to surrogate use due to the free NH.

Another pragmatic approach could be applying 150 ng/day for N-nitroso-piperazines as a class and applying MW correction when deriving a limit for a NDSRI, which often will be just below 400 ng/day based on CPCA.

I suggest to start reading here:

Dobo, K. L., Kenyon, M. O., Dirat, O., Engel, M., Fleetwood, A., Martin, M., Mattano, S., Musso, A., McWilliams, J. C., Papanikolaou, A., Parris, P., Whritenour, J., Yu, S., & Kalgutkar, A. S. (2022). Practical and Science-Based Strategy for Establishing Acceptable Intakes for Drug Product N‑Nitrosamine Impurities. Chemical Research in Toxicology, 35(3), 475–489. https://doi.org/10.1021/acs.chemrestox.1c00369.

Bercu, J. P., Masuda-Herrera, M., Trejo-Martin, A., Sura, P., Jolly, R., Kenyon, M., Thomas, R., Ponting, D. J., Snodin, D., Tuschl, G., Simon, S., De Vlieger, K., Hutchinson, R., Czich, A., Glowienke, S., Reddy, M. V., Johanssen, S., Vock, E., Claude, N., & Weaver, R. J. (2023). Acceptable intakes (AIs) for 11 small molecule N-nitrosamines (NAs). Regulatory Toxicology and Pharmacology, 142, 105415. Redirecting.
(For the monograph: see monograph 11 in SI)

dos Santos, C. E. M., Dorta, D. J., & de Oliveira, D. P. (2022). Setting limits for N-nitrosamines in drugs: A defined approach based on read-across and structure-activity relationship for N-nitrosopiperazine impurities. Regulatory Toxicology and Pharmacology, 136, 105288. Redirecting.

Love, L. A., Lijinsky, W., Keefer, L. K., Garcia, H. (1977). Chronic oral administration of 1-nitrosopiperazine at high doses to MRC rats. Zeitschrift für Krebsforschung und Klinische Onkologie, 89, 69-73.

Lijinsky, W. (1992). Chemistry and Biology of N-nitroso compounds. Cambridge Monographs on Cancer Research. Cambridge University Press, Cambridge and New York, 464 p.

Kruhlak, N. L., Schmidt, M., Froetschl, R., Graber, S., Haas, B., Horne, I., Horne, S., King, S. T., Koval, I. A., Kumaran, G., Langenkamp, A., McGovern, T. J., Peryea, T., Sanh, A., Siqueira Ferreira, A., van Aerts, L., Vespa, A., & Whomsley, R. (2024). Determining recommended acceptable intake limits for N-nitrosamine impurities in pharmaceuticals: Development and application of the Carcinogenic Potency Categorization Approach (CPCA). Regulatory Toxicology and Pharmacology, 150, 105640. Redirecting.

4 Likes

This is fully in line with my recent experience I had when I had to evaluate NO-ofloxacin as NDSRI. The here stated Lhasa values for 1-nitrosopiperazine do not really fullfill the criterium of “sufficient substance specific animal carcinogenicity data” (EMA/409815/2020 Rev.22) & thus we had to argue via a holistic “weight-of-evidence” approach (ICH Q9, M7).

2 Likes

Thank you ccdw and OCarter. This information was useful as I am trying to navigate the most appropriate AI limit for the nitrosopiperazine I am currently investigating (it has a cylcoproylmethyl- group attached to its nitrosopiperzine). I might reach out to you too further if I have any follow up questions.

Something important on limit definitions, and trying to take alternative approaches: Ask yourself if the data are you trying to use or generate is under a OECD method or protocol (https://www.oecd.org/en/topics/sub-issues/testing-of-chemicals/test-guidelines.html)

2 Likes

I’m not sure how this fits this discussion, is it because of the repeated mentioning of expert assessment using a weight of evidence approach? It is not abnormal that (published) Industry proposed WoE approaches use combinations of data linked and not linked to OECD protocols.

Considering the rapidly changing field of nitrosamine science and the evolutions on the development of novel in vitro and in vivo mutagenicity assays for nitrosamines, it is not abnormal that data from newer assays (not yet having OECD protocols) would be used as well, depending on the state of the art science at the time of submission. Some (regulatory) experts would argue that the (scientific quality of the) method (and its suitability to tackle the scientific problem at hand) is more important than being OECD TG validated assays and suggest wider acceptance of non-OECD methods if the science already proves their superiority. There is no global consensus though and a framework for approaching such new data or the WoE approach could be useful (something in between there being nothing and the OECD phase; or clearly addressing how much a door can be opened with non-OECD TG based data in the mix). For the time being I think non-OECD TG-built proper data has its place in a WoE approach, at the very least to address the issue of such novel methodologies being the way forward, but having no OECD TG yet and a getting a framework confirmed. If such novel methodology data has value and the design can be justified, I would not hide it because it does not fit an OECD boat. (On the other hand having only non-OECD elements in a WoE approach could be a bit uncommon as well, but is maybe justified based on limitations for the specific case.)

On the other hand there is the increased and very valuable use of quantum mechanical tools in the WoE approaches. I would also not ignore those insights in a WoE approach until we have a validated QSAR model.

I would argue also that the CPCA is not validated based on “OECD method or protocol” originating data only…

Scientific arguments should allow derogating from OECD guidance.

1 Like

FDA has recommended an AI limit of 1300 ng/day based on read-across analysis using N-nitrosopiperidine as surrogate.

1 Like

FDA has recommended an AI limit of 1300 ng/day based on read-across analysis using N-nitrosopiperidine as surrogate.

1 Like

You are right, Erica probably wanted to cite HC and EMA instead of FDA and EMA.

Justification of this read-across probably links to the piperidine – piperazine parallels per CPCA and the basic feature of NPIZ I was discussing.

  • As CPCA design already treats piperidine and piperazine the same, NPZ to NPIP read-across is not fully unexpected.
  • NPZ is anticipated to have reduced potency compared to NPIP due to the basic properties of the remaining free nitrogen in the 6-membered ring, and therefore read-across to NPIP is conservative. (Which is also quantitatively seen in the 1300 ng/day value being lower than the 5510-28500 ng/day direct NPZ limits that have been derived.)
4 Likes