Hazard assessment of hexagonal boron nitride and hexagonal boron nitride reinforced thermoplastic polyurethane composites using human skin and lung cells.

Hexagonal boron nitride (hBN) is an emerging two-dimensional material attracting considerable attention in the industrial sector given its innovative physicochemical properties. Potential risks are associated mainly with occupational exposure where inhalation and skin contact are the most relevant exposure routes for workers. Here we aimed at characterizing the effects induced by composites of thermoplastic polyurethane (TPU) and hBN, using immortalized HaCaT skin keratinocytes and BEAS-2B bronchial epithelial cells. The composite was abraded using a Taber® rotary abraser and abraded TPU and TPU-hBN were also subjected to photo-Fenton-mediated degradation mimicking potential weathering across the product life cycle. Cells were exposed to the materials for 24 h (acute exposure) or twice per week for 4 weeks (chronic exposure) and evaluated with respect to material internalization, cytotoxicity, and proinflammatory cytokine secretion. Additionally, comprehensive mass spectrometry-based proteomics and metabolomics (secretomics) analyses were performed. Overall, despite evidence of cellular uptake of the material, no significant cellular and/or protein expression profiles alterations were observed after acute or chronic exposure of HaCaT or BEAS-2B cells, identifying only few pro-inflammatory proteins. Similar results were obtained for the degraded materials. These results support the determination of hazard profiles associated with cutaneous and pulmonary hBN-reinforced polymer composites exposure.

Iron-carbohydrate complexes treating iron anaemia: Understanding the nano-structure and interactions with proteins through orthogonal characterisation.

Intravenous (IV) iron-carbohydrate complexes are widely used nanoparticles (NPs) to treat iron deficiency anaemia, often associated with medical conditions such as chronic kidney disease, heart failure and various inflammatory conditions. Even though a plethora of physicochemical characterisation data and clinical studies are available for these products, evidence-based correlation between physicochemical properties of iron-carbohydrate complexes and clinical outcome has not fully been elucidated yet. Studies on other metal oxide NPs suggest that early interactions between NPs and blood upon IV injection are key to understanding how differences in physicochemical characteristics of iron-carbohydrate complexes cause variance in clinical outcomes. We therefore investigated the core-ligand structure of two clinically relevant iron-carbohydrate complexes, iron sucrose (IS) and ferric carboxymaltose (FCM), and their interactions with two structurally different human plasma proteins, human serum albumin (HSA) and fibrinogen, using a combination of cryo-scanning transmission electron microscopy (cryo-STEM), x-ray diffraction (XRD), small-angle x-ray scattering (SAXS) and small-angle neutron scattering (SANS). Using this orthogonal approach, we defined the nano-structure, individual building blocks and surface morphology for IS and FCM. Importantly, we revealed significant differences in the surface morphology of the iron-carbohydrate complexes. FCM shows a localised carbohydrate shell around its core, in contrast to IS, which is characterised by a diffuse and dynamic layer of carbohydrate ligand surrounding its core. We hypothesised that such differences in carbohydrate morphology determine the interaction between iron-carbohydrate complexes and proteins and therefore investigated the NPs in the presence of HSA and fibrinogen. Intriguingly, IS showed significant interaction with HSA and fibrinogen, forming NP-protein clusters, while FCM only showed significant interaction with fibrinogen. We postulate that these differences could influence bio-response of the two formulations and their clinical outcome. In conclusion, our study provides orthogonal characterisation of two clinically relevant iron-carbohydrate complexes and first hints at their interaction behaviour with proteins in the human bloodstream, setting a prerequisite towards complete understanding of the correlation between physicochemical properties and clinical outcome.

Environmental and Health Impacts of Graphene and Other Two-Dimensional Materials: A Graphene Flagship Perspective.

Two-dimensional (2D) materials have attracted tremendous interest ever since the isolation of atomically thin sheets of graphene in 2004 due to the specific and versatile properties of these materials. However, the increasing production and use of 2D materials necessitate a thorough evaluation of the potential impact on human health and the environment. Furthermore, harmonized test protocols are needed with which to assess the safety of 2D materials. The Graphene Flagship project (2013-2023), funded by the European Commission, addressed the identification of the possible hazard of graphene-based materials as well as emerging 2D materials including transition metal dichalcogenides, hexagonal boron nitride, and others. Additionally, so-called green chemistry approaches were explored to achieve the goal of a safe and sustainable production and use of this fascinating family of nanomaterials. The present review provides a compact survey of the findings and the lessons learned in the Graphene Flagship.

Boron Nitride Nanosheets Induce Lipid Accumulation and Autophagy in Human Alveolar Lung Epithelial Cells Cultivated at Air-Liquid Interface.

Hexagonal boron nitride (hBN) is an emerging 2D material attracting significant attention due to its superior electrical, chemical, and therapeutic properties. However, inhalation toxicity mechanisms of hBN in human lung cells are poorly understood. Here, cellular interaction and effects of hBN nanosheets is investigated in alveolar epithelial cells cultured on porous inserts and exposed under air-liquid interface conditions for 24 h. hBN is taken up by the cells as determined in a label-free manner via RAMAN-confocal microscopy, ICP-MS, TEM, and SEM-EDX. No significant (p > 0.05) effects are observed on cell membrane integrity (LDH release), epithelial barrier integrity (TEER), interleukin-8 cytokine production or reactive oxygen production at tested dose ranges (1, 5, and 10 µg cm-2 ). However, it is observed that an enhanced accumulation of lipid granules in cells indicating the effect of hBN on lipid metabolism. In addition, it is observed that a significant (p < 0.05) and dose-dependent (5 and 10 µg cm-2 ) induction of autophagy in cells after exposure to hBN, potentially associated with the downstream processing and breakdown of excess lipid granules to maintain lipid homeostasis. Indeed, lysosomal co-localization of lipid granules supporting this argument is observed. Overall, the results suggest that the continuous presence of excess intracellular lipids may provoke adverse outcomes in the lungs.

Quaternary ammonium-based coating of textiles is effective against bacteria and viruses with a low risk to human health.

While the global healthcare system is slowly recovering from the COVID-19 pandemic, new multi-drug-resistant pathogens are emerging as the next threat. To tackle these challenges there is a need for safe and sustainable antiviral and antibacterial functionalized materials. Here we develop an 'easy-to-apply' procedure for the surface functionalization of textiles, rendering them antiviral and antibacterial and assessing the performance of these textiles. A metal-free quaternary ammonium-based coating was applied homogeneously and non-covalently to hospital curtains. Abrasion, durability testing, and aging resulted in little change in the performance of the treated textile. Additionally, qualitative and quantitative antibacterial assays on Staphylococcus aureus, Pseudomonas aeruginosa, and Acinetobacter baumanii revealed excellent antibacterial activity with a CFU reduction of 98-100% within only 4 h of exposure. The treated curtain was aged 6 months before testing. Similarly, the antiviral activity tested according to ISO-18184 with murine hepatitis virus (MHV) showed > 99% viral reduction with the functionalized curtain. Also, the released active compounds of the coating 24 ± 5 µg mL-1 revealed no acute in vitro skin toxicity (IC50: 95 µg mL-1) and skin sensitization. This study emphasizes the potential of safe and sustainable metal-free textile coatings for the rapid antiviral and antibacterial functionalization of textiles.

Uncovering the dynamics of cellular responses induced by iron-carbohydrate complexes in human macrophages using quantitative proteomics and phosphoproteomics.

Iron-carbohydrate complexes are widely used to treat iron deficiencies. Macrophages play a crucial role in the uptake and fate of these nanomedicines, however, how complexed iron carbohydrates are taken up and metabolized by macrophages is still not fully understood. Using a (phospho-)proteomics approach, we assessed differences in protein expression and phosphorylation in M2 macrophages triggered by iron sucrose (IS). Our results show that IS alters the expression of multiple receptors, indicative of a complex entry mechanism. Besides, IS induced an increase in intracellular ferritin, the loss of M2 polarization, protective mechanisms against ferroptosis, and an autophagic response. These data indicate that macrophages can use IS as a source of iron for its storage and later release, however, the excess of iron can cause oxidative stress, which can be successfully regulated by the cells. When comparing IS with ferric carboxymaltose (FCM) and iron isomaltoside-1000 (IIM), complexes with a higher carbohydrate ligand stability, we observed that FCM and IIM are metabolized at a slower rate, and trigger M2 polarization loss to a lower extent. These results indicate that the surface characteristics of the iron-carbohydrate complexes may influence the cell responses. Our data show that the application of (phospho-)proteomics can lead to a better understanding of metabolic processes, including the uptake, biodegradation and bioavailability of nanomedicines.

Critical nanomaterial attributes of iron-carbohydrate nanoparticles: Leveraging orthogonal methods to resolve the 3-dimensional structure.

Intravenous iron-carbohydrate nanomedicines are widely used to treat iron deficiency and iron deficiency anemia across a wide breadth of patient populations. These colloidal solutions of nanoparticles are complex drugs which inherently makes physicochemical characterization more challenging than small molecule drugs. There have been advancements in physicochemical characterization techniques such as dynamic light scattering and zeta potential measurement, that have provided a better understanding of the physical structure of these drug products in vitro. However, establishment and validation of complementary and orthogonal approaches are necessary to better understand the 3-dimensional physical structure of the iron-carbohydrate complexes, particularly with regard to their physical state in the context of the nanoparticle interaction with biological components such as whole blood (i.e. the nano-bio interface).

Release and toxicity assessment of carbon nanomaterial reinforced polymers during the use and end-of-life phases: A comparative review.

The research on carbon-based nanomaterial (C-NM) composites has increased in the last two decades. This family of functional materials shows outstanding mechanical, thermal and electrical properties, and are being used in a variety of applications. An important challenge remains before C-NM can be fully integrated in our production industries and our lives: to assess the release of debris during production, use, and misuse of composites and the effect they may have on the environment and on human health. During their lifecycle, composites materials can be subjected to a variety of stresses which may release particles from the macroscopic range to the nanoscale. In this review, the release of debris due to abrasion, weathering and combustion as well as their toxicity is evaluated for the three most used C-NM: Carbon Black, Carbon Nanotubes and Graphene-related materials. The goal is to stimulate a Safe-By-Design approach by guiding the selection of carbon nano-fillers for specific applications based of safety and performance.

Advanced and Smart Textiles during and after the COVID-19 Pandemic: Issues, Challenges, and Innovations.

The COVID-19 pandemic has hugely affected the textile and apparel industry. Besides the negative impact due to supply chain disruptions, drop in demand, liquidity problems, and overstocking, this pandemic was found to be a window of opportunity since it accelerated the ongoing digitalization trends and the use of functional materials in the textile industry. This review paper covers the development of smart and advanced textiles that emerged as a response to the outbreak of SARS-CoV-2. We extensively cover the advancements in developing smart textiles that enable monitoring and sensing through electrospun nanofibers and nanogenerators. Additionally, we focus on improving medical textiles mainly through enhanced antiviral capabilities, which play a crucial role in pandemic prevention, protection, and control. We summarize the challenges that arise from personal protective equipment (PPE) disposal and finally give an overview of new smart textile-based products that emerged in the markets related to the control and spread reduction of SARS-CoV-2.

Gene expression profiling of human macrophages after graphene oxide and graphene nanoplatelets treatment reveals particle-specific regulation of pathways.

Graphene and its derivatives are attractive materials envisaged to enable a wealth of novel applications in many fields including energy, electronics, composite materials or health. A comprehensive understanding of the potential adverse effects of graphene-related materials (GRM) in humans is a prerequisite to the safe use of these promising materials. Here, we exploited gene expression profiling to identify transcriptional responses and toxicity pathways induced by graphene oxide (GO) and graphene nanoplatelets (GNP) in human macrophages. Primary human monocyte-derived macrophages (MDM) and a human macrophage cell line, i.e. differentiated THP-1 cells, were exposed to 5 or 20 µg/mL GO and GNP for 6 and 24 h to capture early and more persistent acute responses at realistic or slightly overdose concentrations. GO and GNP induced time-, dose- and macrophage type-specific differential expression of a substantial number of genes with some overlap between the two GRM types (up to 384 genes (9.6%) or 447 genes (20.4%) in THP-1 or MDM, respectively) but also a high number of genes exclusively deregulated from each material type. Furthermore, GRM responses on gene expression were highly different from those induced by inflammogenic material crystalline quartz (maximum of 64 (2.3%) or 318 (11.3%) common genes for MDM treated with 20 µg/mL GO and GNP, respectively). Further bioinformatics analysis revealed that GNP predominantly activated genes controlling inflammatory and apoptotic pathways whereas GO showed only limited inflammatory responses. Interestingly, both GRM affected the expression of genes related to antigen processing and presentation and in addition, GO activated pathways of neutrophil activation, degranulation and immunity in MDM. Overall, this study provides an extensive resource of potential toxicity mechanisms for future safety assessment of GRM in more advanced model systems to verify if the observed changes in gene expression in human macrophages could lead to long-term consequences on human health.

(Re)Conceptualizing decision-making tools in a risk governance framework for emerging technologies-the case of nanomaterials.

The utility of decision-making tools for the risk governance of nanotechnology is at the core of this paper. Those working in nanotechnology risk management have been prolific in creating such tools, many derived from European FP7 and H2020-funded projects. What is less clear is how such tools might assist the overarching ambition of creating a fair system of risk governance. In this paper, we reflect upon the role that tools might and should play in any system of risk governance. With many tools designed for the risk governance of this emerging technology falling into disuse, this paper provides an overview of extant tools and addresses their potential shortcomings. We also posit the need for a data readiness tool. With the EUs NMP13 family of research consortia about to report to the Commission on ways forward in terms of risk governance of this domain, this is a timely intervention on an important element of any risk governance system.

Mild COVID-19 induces early, quantifiable, persistent troponin I elevations in elder men.

Importance: Elderly patients, especially men, are at risk of increased morbidity from coronavirus disease 2019 (COVID-19). Long-term data on troponin I levels in longitudinal observational studies of outpatients with mild to moderate COVID-19 are scarce. Objective: This controlled cohort study aimed to evaluate the course of troponin I concentrations over a long period in convalescent COVID-19 outpatients with mild to moderate symptoms. Setting and participants: In this cohort study, individuals with PCR-confirmed, mild to moderate SARS-CoV-2 infection as well as control individuals with confirmed negative PCR and negative SARS-CoV-2 serology were included. Study visits were performed from April 2020 through July 2021 (initialized during the first wave of the corona pandemic in Switzerland). A study visit in patients comprised blood draws every week in the first month and additionally after 8 weeks. This course was repeated in patients observed long-term. Results: This study enrolled 278 individuals from the Canton of St. Gallen, Switzerland, aged 12-92 years (59.5% women), who had mild to moderate COVID-19 symptoms (outpatients only) and a diagnosis confirmed by positive RT-PCR. Fifty-four of the participants with confirmed SARS-CoV-2 infection were followed for 14 months with repeat cycles of the testing protocol. In addition, 115 symptomatic patients that were PCR and serology negative were enrolled in the same time period as a control group. In COVID-19 patients, low-level troponin I concentrations (cTnI) were significantly increased from baseline until week 9 after positive RT-PCR diagnosis in men older than 54 years [ΔcTnI = 5.0 ng/L (median); 95% CI 4.1-6.0; p = 0.02]. The troponin I concentration remained elevated throughout 14 months in men older than 54 years within the cohort with a prolonged observation period. This statistically significant change in troponin I concentration was not dependent on co-morbidities in this group. ALT, Creatinine, BNP, and D-Dimer values after convalescence did not differ in comparison to the control cohort. Conclusion: In this analysis of individuals with confirmed SARS-CoV-2 infection, hs troponin I levels of men aged 54 or older significantly increased after infection. They remained elevated for at least 14 months after diagnosis. This suggests the possibility of an ongoing, long-term, low-grade myocardial injury. Further studies with focus on elderly patients and a prolonged observational period are necessary to elucidate whether the phenomenon observed is associated with detectable structural changes to the heart muscle or is without further clinical consequences.

Structure-activity relationship of graphene-related materials: A meta-analysis based on mammalian in vitro toxicity data.

To support a safe application of graphene-related materials (GRMs) it is necessary to understand the potential negative impacts they could have on human health, in particular on the lung - one of the most sensitive exposure routes. Machine learning (ML) approaches can help analyse the results of multiple toxicity studies to understand the structure-activity relationship and the effect of experimental conditions, thus supporting predictive nanotoxicology. In this work we collected in vitro cytotoxicity data obtained from studies using lung cells; we then fitted multiple regression models to predict this endpoint based on the material properties and experimental conditions. Moreover, the data set was used to calculate the Benchmark Dose Lower Confidence Interval (BMDL), a dose descriptor widely used in risk assessment. Regression and classification models were applied for the prediction of the BMDL value and BMDL range. The analyses show that both cytotoxicity and the BMDL range can be predicted well (Q2 = 0.77 and accuracy = 0.71, respectively). Both physico-chemical characteristics such as the lateral size, number of layers, and functionalization, and experimental conditions such as the assay and media used were important predicting features, confirming the need for thorough characterization and reporting of these parameters.

Airborne emissions from combustion of graphene nanoplatelet/epoxy composites and their cytotoxicity on lung cells via air-liquid interface cell exposure in vitro.

Graphene nanoplatelet (GNP) as a nanofiller improves the mechanical strength, electrical conductivity, and flame retardancy of the polymers significantly. With an increasing number of GNP-reinforced products, a careful safety assessment is needed to avoid social and economic setbacks. However, no study has addressed the effects of combustion-generated emissions from GNP-reinforced products in the lung, the most sensitive exposure route to airborne particles. Therefore, we studied the influence of GNP as a nanofiller on the emitted particles and polycyclic aromatic hydrocarbons (PAHs), and cytotoxicity of the emissions from the combustion of pure epoxy (EP) and GNP-reinforced epoxy (EP-GNP). GNP was not detected in the airborne emissions. PAHs were found in airborne particles of both emissions from EP and EP-GNP, with some differences in their concentrations. A first hazard assessment was performed on human alveolar epithelial cells exposed to the airborne emissions at air-liquid interface conditions. At 24 h and 96 h after the exposure, similar responses were observed between EP and EP-GNP except an acute transient decrease in mitochondrial activity after exposure to the emissions from EP-GNP. Both emissions from EP and EP-GNP had no acute effects on membrane integrity, cell morphology or expression of anti-oxidative stress markers (HMOX1 and SOD2 genes). Meanwhile, both emissions induced the activation of the aryl hydrocarbon receptor (CYP1A1 gene) and a transient (pro-) inflammatory response (MCP-1), but the effects between EP and EP-GNP were not significantly different.

A novel inactivated virus system (InViS) for a fast and inexpensive assessment of viral disintegration.

The COVID-19 pandemic has caused considerable interest worldwide in antiviral surfaces, and there has been a dramatic increase in the research and development of innovative material systems to reduce virus transmission in the past few years. The International Organization for Standardization (ISO) norms 18,184 and 21,702 are two standard methods to characterize the antiviral properties of porous and non-porous surfaces. However, during the last years of the pandemic, a need for faster and inexpensive characterization of antiviral material was identified. Therefore, a complementary method based on an Inactivated Virus System (InViS) was developed to facilitate the early-stage development of antiviral technologies and quality surveillance of the production of antiviral materials safely and efficiently. The InViS is loaded with a self-quenched fluorescent dye that produces a measurable increase in fluorescence when the viral envelope disintegrates. In the present work, the sensitivity of InViS to viral disintegration by known antiviral agents is demonstrated and its potential to characterize novel materials and surfaces is explored. Finally, the InViS is used to determine the fate of viral particles within facemasks layers, rendering it an interesting tool to support the development of antiviral surface systems for technical and medical applications.

Evaluation of fiber and debris release from protective COVID-19 mask textiles and in vitro acute cytotoxicity effects.

Since the start of the current COVID-19 pandemic, for the first time a significant fraction of the world's population cover their respiratory system for an extended period with mostly medical facemasks and textile masks. This new situation raises questions about the extent of mask related debris (fibers and particles) being released and inhaled and possible adverse effects on human health. This study aimed to quantify the debris release from a textile-based facemask in comparison to a surgical mask and a reference cotton textile using both liquid and air extraction. Under liquid extractions, cotton-based textiles released up to 29'452 ± 1'996 fibers g-1 textile while synthetic textiles released up to 1'030 ± 115 fibers g-1 textile. However, when the masks were subjected to air-based extraction scenarios, only a fraction (0.1-1.1%) of this fiber amount was released. Several metals including copper (up to 40.8 ± 0.9 µg g-1) and iron (up to 7.0 ± 0.3 µg g-1) were detected in acid dissolved textiles. Additionally the acute in vitro toxicity of size-fractionated liquid extracts (below and above 0.4 µm) were assessed on human alveolar basal epithelial cells. The current study shows no acute cytotoxicity response for all the analyzed facemasks.

Hazard assessment of abraded thermoplastic composites reinforced with reduced graphene oxide.

Graphene-related materials (GRMs) are subject to intensive investigations and considerable progress has been made in recent years in terms of safety assessment. However, limited information is available concerning the hazard potential of GRM-containing products such as graphene-reinforced composites. In the present study, we conducted a comprehensive investigation of the potential biological effects of particles released through an abrasion process from reduced graphene oxide (rGO)-reinforced composites of polyamide 6 (PA6), a widely used engineered thermoplastic polymer, in comparison to as-produced rGO. First, a panel of well-established in vitro models, representative of the immune system and possible target organs such as the lungs, the gut, and the skin, was applied. Limited responses to PA6-rGO exposure were found in the different in vitro models. Only as-produced rGO induced substantial adverse effects, in particular in macrophages. Since inhalation of airborne materials is a key occupational concern, we then sought to test whether the in vitro responses noted for these materials would translate into adverse effects in vivo. To this end, the response at 1, 7 and 28 days after a single pulmonary exposure was evaluated in mice. In agreement with the in vitro data, PA6-rGO induced a modest and transient pulmonary inflammation, resolved by day 28. In contrast, rGO induced a longer-lasting, albeit moderate inflammation that did not lead to tissue remodeling within 28 days. Taken together, the present study suggests a negligible impact on human health under acute exposure conditions of GRM fillers such as rGO when released from composites at doses expected at the workplace.

Approach toward In Vitro-Based Human Toxicity Effect Factors for the Life Cycle Impact Assessment of Inhaled Low-Solubility Particles.

Today's scarcity of animal toxicological data for nanomaterials could be lifted by substituting in vivo data with in vitro data to calculate nanomaterials' effect factors (EF) for Life Cycle Assessment (LCA). Here, we present a step-by-step procedure to calculate in vitro-to-in vivo extrapolation factors to estimate human Benchmark Doses and subsequently in vitro-based EFs for several inhaled nonsoluble nanomaterials. Based on mouse data, the in vitro-based EF of TiO2 is between 2.76 · 10-4 and 1.10 · 10-3 cases/(m2/g·kg intake), depending on the aerodynamic size of the particle, which is in good agreement with in vivo-based EFs (1.51 · 10-4-5.6 · 10-2 cases/(m2/g·kg intake)). The EF for amorphous silica is in a similar range as for TiO2, but the result is less robust due to only few in vivo data available. The results based on rat data are very different, confirming the importance of selecting animal species representative of human responses. The discrepancy between in vivo and in vitro animal data in terms of availability and quality limits the coverage of further nanomaterials. Systematic testing on human and animal cells is needed to reduce the variability in toxicological response determined by the differences in experimental conditions, thus helping improve the predictivity of in vitro-to-in vivo extrapolation factors.

Combined in vitro-in vivo dosimetry enables the extrapolation of in vitro doses to human exposure levels: A proof of concept based on a meta-analysis of in vitro and in vivo titanium dioxide toxicity data.

Evaluating the potential risks of nanomaterials on human health is fundamental to assure their safety. To do so, Human Health Risk Assessment (HHRA) relies mostly on animal studies to provide information about nanomaterials toxicity. The scarcity of such data, due to the shift of the nanotoxicology field away from a phenomenological, animal-based approach and towards a mechanistic understanding based on in vitro studies, represents a challenge for HHRA. Implementing in vitro data in the HHRA methodology requires an extrapolation strategy; combining in vitro dosimetry and lung dosimetry can be an option to estimate the toxic effects on lung cells caused by inhaled nanomaterials. Since the two dosimetry models have rarely been used together, we developed a combined dosimetry model (CoDo) that estimates the air concentrations corresponding to the in vitro doses, extrapolating in this way in vitro doses to human doses. Applying the model to a data set of in vitro and in vivo toxicity data about titanium dioxide, we demonstrated CoDo's multiple applications. First, we confirmed that most in vitro doses are much higher than realistic human exposures, considering the Swiss Occupational Exposure Limit as benchmark. The comparison of the Benchmark Doses (BMD) extrapolated from in vitro and in vivo data, using the surface area dose metric, showed that despite both types of data had a quite wide range, animal data were overall more precise. The high variability of the results may be due both to the dis-homogeneity of the original data (different cell lines, particle properties, etc.) and to the high level of uncertainty in the extrapolation procedure caused by both model assumptions and experimental conditions. Moreover, while the surface area BMDs from studies on rodents and rodent cells were comparable, human co-cultures showed less susceptibility and had higher BMDs regardless of the titanium dioxide type. Last, a Support Vector Machine classification model built on the in vitro data set was able to predict the BMD-derived human exposure level range for viability effects based on the particle properties and experimental conditions with an accuracy of 85%, while for cytokine release in vitro and neutrophil influx in vivo the model had a lower performance.