The Long-Running Nitrosamine Problem

 BY DEREK LOWE

I wrote a few years ago here about the then-current problem with nitrosamine contamination in several commercial drug substances. But that story has never really gone away, as this new paper details.. At the time, the immediate problem was the switch that some manufacturers had made to dimethylformamine (DMF) in a particular synthetic step, and followed that with different conditions for the formation of a tetrazole ring. These produced small amount of nitrous acid in the workup, which reacted with small amounts of dimethylamine left over from the earlier step.

N-nitrosoamines are not something that you want to get exposed to a lot of; that's something that everyone agrees on. At the same time, we are often exposed to small amounts - for example, by consuming meats that have been cured by sodium nitrite/nitrate. As a side note, don't be fooled by the products in that category (sausage, bacon) that advertise as "no added nitrites" or even as "uncured". In the US, anyway, such products are taking advantage of a labeling loophole: they're adding celery extract (or other vegetable extracts) which are high in nitrate, which is reduced to nitrite in situ. You are getting the same nitrite-cured meat you would otherwise (that's why it's red and not gray), and this is providing the same protection against botulism and other bacterial growth that you would get otherwise (which is a big reason nitrite is added to cure meats in the first place), but you're just paying more for it while being given the impression that you're not getting any nitrites at all.

Nitrite in foods can react with endogenous amines to make the N-nitroso species, just as it did in that chemistry process mentioned above. But you don't have to be eating bacon to get exposed: nitrosamines are also produce by fermentation, in foods that have been dried through heating  by reaction of nitrogen oxides from combustion, an example being malted barley used to make beer, and in the process of cooking itself. Some of these can produce notably higher nitrosamine levels than are found in cured meat. Tobacco products are particularly high routes of exposure. The arguing starts, as it generally does in toxicology and has since at least the time of Paracelsus, about what doses are harmful. On a molecular level, the problem is that monoalkyl N-nitrosamines (which are produced from dialkylnitrosamines through oxidative dealkylation) get transformed into reactive diazonium salts. These are undisputed DNA alkylating agents, and a great deal of evidence links high or prolonged exposure to these to several types of cancer in humans and many other animal species.

Dietary sources aside, it seems reasonable to ask that the drug industry not add still more sources of N-nitroso exposure to the list. As it stands, acceptable regulatory limits on exposure through pharmaceutical intake are set about tenfold lower than what people get through their food - that new paper mentioned in the first paragraph goes into a great deal of detail on the various ways to measure exposure, but all of them come down to pretty stringent limits in drug manufacture. The authors are at pains to make clear that pharmaceuticals cannot be made "nitroso-free", any more than the food supply can be. If you have amines and oxygen, you are going to see some level of these things.

And that means that the original problem of things like N,N-dimethyl- or N,N-diethylnitrosamine, which were the contaminants that made the news back in 2018-2019, is nowhere near the end of the story. If you look more closely in the parts-per-billion range, you can find all sorts of N-nitroso things that are produce by oxidation of pharmaceutical ingredients and excipients (and the same goes for food, of course). We definitely want to look out for contamination introduced by chemical processes, but some of these things are just going to form on their own. Right now there's a big push to standardize and validate analytical methods for such things, and another big effort to come up with a standard way to measure how worried we should be about their carcinogenenic potential. The Ames test, for example, is commonly used as a front-line read on mutagenicity, so we need to figure out how well it picks up on these compounds. That's a post here from July of 2002. My writing style does not seem to have changed much! The "to be continued" at the end of it is the link in the next paragraph.

It would be very easy to say "Hey, I know what limit I want for carcinogens in my prescription drugs and in my food: zero. How about that?" But that is not going to happen. It's not possible. And by that I don't mean "It's too expensive" or "We don't wanna"; I literally mean that it's not possible. I also don't mean "It's not possible in our dirty, contaminated, industrial civilization under the conditions of late capitalism", either, in case you might be thinking in that direction. Uncontacted hunter-gatherer tribes foraging for fresh berries in the middle of the jungle are constantly exposed to carcinogens, too. I am sounding a note here that I first did in the early days of this blog, and here's link back to a post from 2002 on these issues.

The potential problem is that if we make the testing regime for N-nitrosos too onerous that some low-margin producers of generic drugs might just decide that some products aren't worth the trouble and cease production. We'll need to do some more risk/reward calculations, to try to understand the harm done in each direction. Eventually, as the new paper linked in the first paragraph says, the industry will have a "new normal" for nitrosamine testing and mitigation - but we're not there yet.

https://www.science.org/content/blog-post/long-running-nitrosamine-problem