AI Insight
Researchers developed InFiniteLungDT, a digital twin system trained on in vitro biological data to predict chronic neutrophilic lung inflammation caused by inhaled materials, without requiring knowledge of the materials' intrinsic physicochemical properties. The model was validated against a dataset spanning 49 nanomaterials and 7 inhalation studies, covering approximately 3,094 mouse and 364 rat exposures, achieving predictive accuracy of 93% for instillation and 84% for inhalation exposure routes. Critically, the system delivers results including dose-response relationships, inflammation onset timing, and duration within less than one week, compared to months required for standard animal studies.
Why it matters
This method represents a potential regulatory-grade alternative to animal testing for assessing the pulmonary safety of inhaled materials, including nanomaterials, which could substantially reduce both the time and cost of occupational and environmental safety evaluations. If validated through peer review and accepted by regulatory agencies, it could meaningfully reduce the number of animals used in inhalation toxicology studies.
⚠️ Preprint – Noch nicht peer-reviewed
Dieser Artikel wurde noch nicht von unabhängigen Experten begutachtet. Die Ergebnisse sind vorläufig und sollten mit Vorsicht interpretiert werden.
Until now, there has been no animal-free alternative method for predicting chronic inflammation and delivering the associated dose responses, the timing of onset, and the duration of inflammation, as required by regulatory agencies. Here, we demonstrate that the in-vitro-learned digital twin of neutrophilic lung inflammation (InFiniteLungDT) can obtain regulatory-relevant information about chronic neutrophilic lung inflammation by measuring dynamics of early biological effects in vitro induced by respirable materials without knowing their intrinsic properties. We constructed the digital twin(s) for each of the material, for which we have in vivo exposure data, primarily in vivo instillation data set comprising 49 different nanomaterials and 7 in vivo inhalation data set (compliant with OECD TG 412) encompassing total about 3094 single mouse exposures and 364 rat exposures (and approx. 775/225 non-exposed mouse/rat controls), to predict concentration-dependent time evolved neutrophil influx into the lung, and assess the reproducibility as well as the accuracy (predictive capacity) of both endpoints and LOAELs, that were determined at 93% for instillation, and 84% for inhalation exposure. Taking into account the time-to-deliver-result being less than 1 week, this proves that the effect of inhaled material from acute to chronic conditions can be assessed orders of magnitude faster and cheaper than in a reference animal study.