N-Nitrosonornicotine (NNN) is believed to play a critical role in the development of cancer in tobacco users and was therefore classified by the International Agency for Research on Cancer (IARC) as a Group 1 carcinogen. Hence, urinary NNN is commonly determined within the toxicological risk assessment of new nicotine delivery products (e.g. heat-not-burn, smokeless tobacco). Recent studies have shown that NNN can be formed by nitrosation of nornicotine in saliva. Our own investigations have also shown occasionally high levels of NNN in smokers, which could not be explained by their exposure. Thus, we systematically investigated the artefactual formation of NNN in urine. We performed incubation experiments with non-smoker urine samples by adding the precursors for nitrosation – nornicotine and nitrite/nitrate. We found that pH was most critical for artefactual NNN formation while varying the added amounts of precursors had a rather low impact. High concentrations of NNN (> 50 pg/mL) were formed within minutes after addition of the precursors at acidic conditions (pH = 2-5) with increasing formation over time. Even at neutral pH (pH = 6-7), NNN formation was observed albeit at much lower concentrations (2-4 pg/mL). Our results indicate that artefactual NNN formation can occur due to the presence of nornicotine in urine even though no NNN exposure from the product was expected (e.g. from e-cigarettes). Hence, we developed an LC-MS/MS method for NNN in human plasma and saliva in order to circumvent NNN analysis in urine. The validated method is characterized by very high sensitivity (LOQ: 0.5 pg/mL), broad linear range (0.5-1000 pg/mL) and high throughput. The new method is thus well suited for large biomonitoring and PK studies, and will be used in future investigations to explore the suitability of NNN as a biomarker of exposure in plasma as an alternative matrix to urine.