Heated tobacco products (HTPs) are promoted as alternatives to conventional cigarettes with a putative reduced-risk profile. The actual impact of HTP use on cancer and other disease risks remains uncertain and requires quantitative evaluation. This study presents a unified, transparent computational framework for toxicological risk assessment of HTPs. The framework aims to integrate chemical emissions data with regulatory compound-specific toxicological thresholds to provide reproducible estimates of both cancer and noncancer risk.
The stated objectives are threefold: (1) systematically review and harmonize existing risk models used in the literature; (2) formulate generalizable mathematical models to estimate lifetime cancer risk, hazard quotients, and margins of exposure while accounting for population demographics, smoking habits, and compound properties; and (3) validate the proposed models by reproducing published results and evaluating sensitivity of risk estimates to key parameters and emission sources.
The proposed framework integrates two primary types of input data: measured or reported chemical emissions from tobacco products, and toxicological thresholds for individual compounds as established by regulatory agencies. Emissions data serve as the exposure metric for each compound, while toxicological thresholds (for example, cancer potency factors or reference doses used by regulators) provide compound-specific benchmarks for risk calculation.
To enable population-relevant estimates, the models incorporate parameters that reflect demographics and smoking behavior. These include factors such as lifetime exposure duration, frequency or intensity of product use, and population distributions when relevant. Compound characteristics such as toxicokinetic or potency differences are included indirectly through the choice of regulatory thresholds and potency factors.
The authors formulate general mathematical expressions that can be applied to individual compounds and aggregated across compounds. Key modeled endpoints are:
Lifetime cancer risk: a quantitative estimate of the lifetime probability of cancer attributable to exposure to carcinogenic compounds in emissions, based on emissions per unit use and regulatory potency values.
Hazard quotients (HQs): compound-specific noncancer comparisons of exposure to the relevant health-based guidance value, used to screen for potential noncancer hazards.
Margin of exposure (MOE): the ratio between a toxicological point of departure (such as a no-observed-adverse-effect level or benchmark dose) and the estimated human exposure; lower MOEs indicate higher concern.
Each mathematical formulation is presented to be generalizable so it can be applied across different products, compounds, or population scenarios. The model forms explicitly allow plugging in different emissions datasets and different regulatory toxicological parameters.
A systematic review of existing risk models in the literature underpins the framework. The authors harmonize model structures and assumptions where possible to reduce heterogeneity in estimates that can arise from methodological differences. Harmonization efforts address choices such as exposure averaging time, assumptions about inhalation dose versus product use patterns, selection of potency factors, and aggregation rules for multiple compounds.
The goal of harmonization is to produce transparent, comparable risk outputs that can reproduce prior published results under comparable assumptions and to provide a consistent baseline for future assessments.
Validation of the framework is performed by attempting to reproduce results previously published in the literature. Where reproduction is possible, the framework demonstrates consistency with earlier findings under the same input assumptions. Additionally, the authors conduct sensitivity analyses to explore how risk estimates change with variation in model parameters and emission sources.
Sensitivity analyses probe the influence of key inputs — for example, differences in emissions datasets, alternative toxicological thresholds, or changes in user behavior parameters — on both per-compound and aggregated risk estimates. These analyses help identify which inputs most strongly drive uncertainty in model outputs.
The framework is applied to emissions data from conventional cigarettes and from HTPs. Using these inputs, the models produce per-compound risk estimates as well as aggregated cancer and noncancer risks for product use scenarios. The authors evaluate the relative change in estimated risk when switching from cigarettes to HTPs, thereby quantifying the relative risk reduction associated with such a switch under the chosen assumptions and datasets.
The abstract does not report numerical results or specific magnitudes of risk reduction; it indicates only that per-compound and aggregated risks were quantified and relative comparisons conducted.
Per-compound results allow identification of specific constituents that contribute most to modeled cancer or noncancer risk. Aggregated estimates combine compound-level risks to deliver product-level risk metrics suitable for comparative assessments. Interpretation depends on the selected toxicological thresholds, the emissions dataset, and assumptions about product use; harmonization and sensitivity analyses are therefore essential to contextualize any reported differences between products.
The authors emphasize reproducibility and transparency as core features. The framework is described as extensible to new nicotine and tobacco products, and adaptable to different regulatory toxicological parameters or alternative emissions datasets. Limitations noted in the abstract include remaining uncertainty about the real-world impact of HTPs on health; specific limitations, numeric uncertainty quantification, or dataset constraints are not detailed in the abstract itself.
Where the source text lacks further methodological or numerical detail, those specifics were not reported in the provided abstract.
By providing standardized, transparent mathematical approaches for cancer and noncancer risk estimation, the framework supports quantitative harm-reduction evaluations. Regulators, researchers, and product developers can apply the models to compare products, prioritize compounds for additional study, and explore how changes in emissions or use patterns may alter population risk. The framework's reproducibility and extensibility make it suitable for emerging products and for incorporating updated toxicological values as they become available.
The study presents a reproducible computational toxicology approach to assessing HTP risks that integrates emissions measurements with regulatory toxicological thresholds and generalizable risk models for lifetime cancer risk, hazard quotients, and margins of exposure. The framework was validated against published results where possible and subjected to sensitivity analyses. According to the abstract, this transparent, extensible approach can be applied to emerging nicotine and tobacco products within harm-reduction paradigms. The abstract does not specify where model code or datasets are hosted or how to access them; those details were not reported in the source text.