N-Nitrosamine Formation in Pharmaceutical Drug Products: Development and Validation of a Biased, Conservative, Predictive Model

Abstract

Following the discovery of N-nitrosamines in a range of medicinal products, regulatory agencies have required marketing authorization holders to undertake risk assessments for the presence of N-nitrosamines in their drug products. This work focusses on a solution phase kinetic model of secondary amine nitrosation that may be applied through the concept of a saturated solution layer to consider the formation of N-nitrosamines in a solid drug product. The conservative assumptions made in defining the reactant concentrations bias the model to overpredict the level of N-nitrosamine formed. This overprediction is demonstrated when model predictions are compared to testing data for the formation of nitroso-4-phenyl piperidine in a model drug product. Additionally, comparison of the model predictions to product testing data for two Nitrosamine Drug Substance Related Impurities (NDSRIs), a nitrosated β-blocker and the nitrosamine of a dialkylamine related substance impurity, further confirm the tendency of the model to overpredict. These results demonstrate that the conservative model has utility in the context of a drug product nitrosamine risk assessment. Extension of the model to consider competing nitrosation reactions occurring within a drug product is discussed alongside the impact of reactant availability on the predicted rate of nitrosation.

Introduction

Following the discovery in 2018 of dialkyl-N-nitrosamines in products containing members of the Sartan family of drugs1 there have been further findings of dialkyl-N-nitrosamines in a range of medicinal products.1,2,3 These findings have led to product recalls and requests from regulatory authorities that marketing authorization holders assess their products for the presence of N-nitrosamines.4,5 These calls for review, and subsequent guidance,6,7,8 requested that the risk of formation and or presence of nitrosamines in drug substances and drug products be assessed.

There was little evidence at the time to show that nitrosamine formation could occur within the matrix of a solid drug product, which led a group of pharma companies to collaborate through the IQ Consortium9 to investigate factors that could lead to nitrosation occurring within a drug product. These experimental investigations10 and other independent studies11,12 have demonstrated that N-nitrosamine formation can occur if vulnerable amines13 are combined with nitrite containing excipients within the matrix of a solid drug product and aided the identification of factors that could give rise to a risk of nitrosamine formation in a solid drug product.14,15,16
Having demonstrated that nitrosation can occur in certain cases, it is then informative to consider how a reaction may occur between nitrite and an amine within a drug product. Aqueous nitrosation of amines by species derived from nitrite is a well-studied reaction that has recently been reviewed,17 as part of an evaluation of the risk of N-nitrosamine formation due to trace nitrite present in the water used during drug substance processing. Based on a review of the published literature a general kinetic model was proposed for the aqueous nitrosation of dialkylamines by three nitrosating agents (dinitrogen trioxide, nitrosyl chloride18 and protonated nitrous acid) derived from nitrous acid in water (Scheme 1). The conservative nature of this model has been demonstrated versus experimental studies of a small number of dialkylamines in aqueous solution over a range of pHs.19,20,21 The model may also be extended to consider alkyl anilines22 by substitution of the appropriate rate constants.19 It is not suitable for aminopyridines and related structures as their nitrosation has been shown to follow a different rate law.23,24,25

An extensive investigation of nitrosamine formation in model solid drug products has been reported,10 which laid the foundations for the following proposal. For nitrosation to occur within the matrix of a solid drug product multiple events need to take place: nitrite contained in at least one of the excipients is protonated to generate nitrous acid, nitrous acid reacts with at least one other species to generate a nitrosating agent and finally the nitrosating agent nitrosates the reactive free base form of the amine to form a N-nitrosamine. Alternatively, a nitrite precursor present in an excipient needs to react to liberate nitrite which can then be transformed into a nitrosating agent. Given the multistep/ multicomponent nature of nitrosation it is highly unlikely that it could occur as a true solid-state reaction at the interface between two particles within a solid drug product. While such reactions do occur,26 there is little or no evidence that processes requiring three or more components can occur in this manner. An alternative explanation for how nitrosation could occur may be found in the application of the saturated solution layer concept27 that was proposed to explain the dependence of an acid-catalyzed decomposition reaction on temperature and humidity. This concept has been successfully applied to the degradation of a range of salts of procaine in the solid state,28 where the extent of reaction was found to be consistent with the solubility of the salt, the pH of a saturated solution and the solution state pH dependent hydrolysis kinetics of procaine. The formation of a saturated solution layer that allows bridging between the drug substance and excipient particles within a solid drug product would provide a means by which free nitrite or nitrite from precursors in the excipients could be mobilized, equilibrate to form a nitrosating agent and react with a secondary or primary amine29 contained within the drug substance or an excipient. Such a process is illustrated (Figure 1) for nitrosation by N2O3.

Nitrosation within a saturated solution layer that bridges the drug substance and excipient particles within a solid drug product has the potential to be modelled kinetically by applying the published aqueous solution nitrosation model,17,19 which has been demonstrated to overpredict.19,20, 21 For this to be possible a number of initial conditions and parameters for the simulation would need to be defined:

• pH of the saturated solution layer,
• initial nitrite concentration in the saturated solution layer,
• initial amine concentration in the saturated solution layer,
• initial halide concentration in the saturated solution layer,
• pKa(s) of the amine,
• simulation time and temperature.

Download the full article as PDF here N-Nitrosamine Formation in Pharmaceutical Drug Products

or continue reading here

Ian W. Ashworth, N-Nitrosamine Formation in Pharmaceutical Drug Products: Development and Validation of a Biased, Conservative, Predictive Model, Journal of Pharmaceutical Sciences, 2025, 104067, ISSN 0022-3549, https://doi.org/10.1016/j.xphs.2025.104067.

Read more interesting articles on N-Nitrosamine here:

• Formation of N-Nitrosamine Drug Substance Related Impurities in Medicines: A Regulatory Perspective on Risk Factors and Mitigation Strategies
• N-nitrosamine Mitigation with Nitrite Scavengers in Oral Pharmaceutical Drug Products
• Risk Assessment and Management Strategy of Two New Nitrosamine Drug Substance Related Impurities (NDSRIs) in a Pharmaceutical Drug Product for the Treatment of a Rare Disease- from Prediction to Control