Background: Low delivery efficiency of nanoparticles (NPs) to the tumor is a critical barrier in the field of cancer nanomedicine. Strategies on how to improve NP tumor delivery efficiency remain to be determined. Methods: This study analyzed the roles of NP physicochemical properties, tumor models, and cancer types in NP tumor delivery efficiency using multiple machine learning and artificial intelligence methods, using data from a recently published Nano-Tumor Database that contains 376 datasets generated from a physiologically based pharmacokinetic (PBPK) model. Results: The deep neural network model adequately predicted the delivery efficiency of different NPs to different tumors and it outperformed all other machine learning methods; including random forest, support vector machine, linear regression, and bagged model methods. The adjusted determination coefficients (R²) in the full training dataset were 0.92, 0.77, 0.77 and 0.76 for the maximum delivery efficiency (DE_max), delivery efficiency at 24 h (DE₂₄), at 168 h (DE₁₆₈), and at the last sampling time (DE_Tlast). The corresponding R² values in the test dataset were 0.70, 0.46, 0.33 and 0.63, respectively. Also, this study showed that cancer type was an important determinant for the deep neural network model in predicting the tumor delivery efficiency across all endpoints (19– 29%). Among all physicochemical properties, the Zeta potential and core material played a greater role than other properties, such as the type, shape, and targeting strategy. Conclusion: This study provides a quantitative model to improve the design of cancer nanomedicine with greater tumor delivery efficiency. These results help to improve our understanding of the causes of low NP tumor delivery efficiency. This study demonstrates the feasibility of integrating artificial intelligence with PBPK modeling approaches to study cancer nanomedicine. Graphical Abstract:

Machine learning and artificial intelligence approaches have revolutionized multiple disciplines, including toxicology. This review summarizes representative recent applications of machine learning and artificial intelligence approaches in different areas of toxicology, including physiologically based pharmacokinetic (PBPK) modeling, quantitative structure-activity relationship modeling for toxicity prediction, adverse outcome pathway analysis, high-throughput screening, toxicogenomics, big data, and toxicological databases. By leveraging machine learning and artificial intelligence approaches, now it is possible to develop PBPK models for hundreds of chemicals efficiently, to create in silico models to predict toxicity for a large number of chemicals with similar accuracies compared with in vivo animal experiments, and to analyze a large amount of different types of data (toxicogenomics, high-content image data, etc.) to generate new insights into toxicity mechanisms rapidly, which was impossible by manual approaches in the past. To continue advancing the field of toxicological sciences, several challenges should be considered: (1) not all machine learning models are equally useful for a particular type of toxicology data, and thus it is important to test different methods to determine the optimal approach; (2) current toxicity prediction is mainly on bioactivity classification (yes/no), so additional studies are needed to predict the intensity of effect or dose-response relationship; (3) as more data become available, it is crucial to perform rigorous data quality check and develop infrastructure to store, share, analyze, evaluate, and manage big data; and (4) it is important to convert machine learning models to user-friendly interfaces to facilitate their applications by both computational and bench scientists.

Physiologically based pharmacokinetic (PBPK) models are useful tools in drug development and risk assessment of environmental chemicals. PBPK model development requires the collection of species-specific physiological, and chemical-specific absorption, distribution, metabolism, and excretion (ADME) parameters, which can be a time-consuming and expensive process. This raises a need to create computational models capable of predicting input parameter values for PBPK models, especially for new compounds. In this review, we summarize an emerging paradigm for integrating PBPK modeling with machine learning (ML) or artificial intelligence (AI)-based computational methods. This paradigm includes 3 steps (1) obtain time-concentration PK data and/or ADME parameters from publicly available databases, (2) develop ML/AI-based approaches to predict ADME parameters, and (3) incorporate the ML/AI models into PBPK models to predict PK summary statistics (eg, area under the curve and maximum plasma concentration). We also discuss a neural network architecture “neural ordinary differential equation (Neural-ODE)” that is capable of providing better predictive capabilities than other ML methods when used to directly predict time-series PK profiles. In order to support applications of ML/AI methods for PBPK model development, several challenges should be addressed (1) as more data become available, it is important to expand the training set by including the structural diversity of compounds to improve the prediction accuracy of ML/AI models; (2) due to the black box nature of many ML models, lack of sufficient interpretability is a limitation; (3) Neural-ODE has great potential to be used to generate time-series PK profiles for new compounds with limited ADME information, but its application remains to be explored. Despite existing challenges, ML/AI approaches will continue to facilitate the efficient development of robust PBPK models for a large number of chemicals.

Background: Bisphenol A (BPA) is a sealant and flux of plastic materials and has been determined to be an endocrine-disrupting chemical. Prenatal exposure to BPA can lead to substantial adverse effects on fetal growth and development. This study was conducted to assess BPA concentration in pregnant women and umbilical cord blood, and to investigate whether maternal BPA exposure affected fetal outcomes including lower birth weight (LBW), smaller size for gestational age (SGA), and high leptin (HLP) and low adiponectin (LAD) secretion. Methods: We measured the BPA levels of maternal blood (n = 97) and umbilical cord blood (n = 97) with a high-performance liquid chromatography/UV detector. The protein secretion of leptin and adiponectin were separately determined using enzyme-linked immunosorbent assay. A logistic regression was performed to estimate the effects of maternal exposure to BPA on LBW, SGA, and adverse action of adipokines in newborns. Results: The geometric means of BPA concentration in maternal blood and fetal cord blood were 2.5 ng/ml and 0.5 ng/ml, respectively. Elevated risks of LBW (OR 2.42, 95% confidence interval (CI) 1.72-3.36), SGA (OR 2.01, 95% CI 1.39-3.01), and adverse action of leptin (OR 1.67, 95% CI 1.12-2.25) and adiponectin (OR 1.25, 95% CI 1.52-3.97) were observed in male neonates in the highest quartile of maternal BPA exposure. Conclusions: Elevated prenatal BPA exposure increased the risk of LBW, SGA, and adverse actions of adipokines in neonates, especially in male infants. These results provide further evidence that maternal exposure is correlated with adverse birth outcomes.

Low tumor delivery efficiency is a critical barrier in cancer nanomedicine. This study reports an updated version of "Nano-Tumor Database", which increases the number of time-dependent concentration data sets for different nanoparticles (NPs) in tumors from the previous version of 376 data sets with 1732 data points from 200 studies to the current version of 534 data sets with 2345 data points from 297 studies published from 2005 to 2021. Additionally, the current database includes 1972 data sets for five major organs (i.e., liver, spleen, lung, heart, and kidney) with a total of 8461 concentration data points. Tumor delivery and organ distribution are calculated using three pharmacokinetic parameters, including delivery efficiency, maximum concentration, and distribution coefficient. The median tumor delivery efficiency is 0.67% injected dose (ID), which is low but is consistent with previous studies. Employing the best regression model for tumor delivery efficiency, we generate hypothetical scenarios with different combinations of NP factors that may lead to a higher delivery efficiency of >3%ID, which requires further experimentation to confirm. In healthy organs, the highest NP accumulation is in the liver (10.69%ID/g), followed by the spleen 6.93%ID/g and the kidney 3.22%ID/g. Our perspective on how to facilitate NP design and clinical translation is presented. This study reports a substantially expanded "Nano-Tumor Database" and several statistical models that may help nanomedicine design in the future.