A Structure-Based Grouping Approach for Evaluating the Toxicity of E-Cigarette Flavor Ingredients: A 5-Week Inhalation Study in A/J Mice

Question: Most flavors are generally recognized as safe for use in foods. However, limited toxicological information is available for evaluating the potential hazard of flavors delivered via inhalation. Because toxicity testing of each individual compound or formulation is not feasible, we sought to evaluate flavors by using a structure-based grouping approach in a short-term inhalation study.

Methods: Structurally related flavor compounds were clustered into groups, and 38 representatives were selected—one from each structural group (Flavor Group Representative, FGR)—on the basis of known and in silico predicted toxicological information. The selected FGRs were combined to create a full "toolbox" flavor mixture. This mixture was then used in a dose-range-finding study with emphasis on subacute toxicity, respiratory tract irritation, and inflammation to select appropriate concentrations of flavor compounds from the "toolbox" to be used in a future chronic inhalation study. A/J mice were whole-body exposed to the following for 6 hours per day, 5 days per week, for 5 weeks: fresh air (sham); aerosol from propylene glycol (PG) and vegetable glycerol (VG) with nicotine (N); aerosol from PG, VG, and N with flavors (F) at low, medium, and high concentrations; or mainstream smoke (MS) from the 3R4F reference cigarette.

Results: The aerosols were well tolerated by the mice, without signs of severe acute toxicity post-exposure. Exposure to the flavored aerosols, even at the highest flavor concentration, did not cause lung inflammation, as evidenced by the lack of immune cell infiltrates in bronchioalveolar lavage fluid and histopathological findings. In contrast, exposure to MS resulted in lung inflammation and moderate to severe adaptive changes in nasal and laryngeal epithelia. Most of the latter effects were absent in mice exposed to flavored e-vapor aerosols, and, when present, they were significantly less severe than in MS-exposed mice.

Conclusion: The tested flavor concentrations did not cause severe subacute toxicity or respiratory tract irritation/inflammation and can be considered suitable for use in future chronic inhalation studies for assessing chronic toxicity and lung tumorigenesis in A/J mice.