SBIR-STTR Award

Medical devices have been documented to contain toxic chemicals that can leach and cause acute contact dermatitis (ACD) after repeated exposure or prolonged contact of the skin to these toxins. ACD is credited for 10-15% of all occupational illnesses and is also the second highest reported occupational hazard. Given its prevalence, ACD is also a great public health burden with combined yearly costs of up to $1 billion, which spans including medical costs, worker’s compensation and lost working time due to workplace absence. To this end, the U.S. Food and Drug Administration has mandated that all medical devices must be evaluated for possible skin sensitization using in vivo animal assays, which includes the Guinea pig maximization test (GPMT). Although GPMT tests provide valuable data on the skin sensitization effects of potential toxins, these assays are time-consuming and expensive. Moreover, the Interagency Coordinating Committee on the Validation of Alternative Methods (ICCVAM) recently published a Strategic Roadmap, calling for the development of alternative approaches to reduce animal testing of chemical and medical agents. Thus, there is a stated need to modernize safety evaluation of medical devices to reduce animal testing and shorten the regulatory review time, which would ultimately bring safer devices to the market faster. To address this unmet need, the key objectives of our FDA Phase I SBIR project are to (i) produce rigorously validated computational models for the GPMT assay integrating data obtained in human, mouse, and in vitro assays; and (ii) integrate these models into a software product termed PreSS/MD (Predictor of Skin Sensitization for Medical Devices). Our specific aims for this study include: 1) collecting, curating, and integrating the largest publicly available dataset for GMPT; 2) creating and validating novel computational models for GMPT data; 3) developing the PreSS/MD web server to allow users to make predictions of skin sensitization potential in medical devices. We will also develop a model for mixtures, including compounds tested jointly in different concentrations, using an approach that we developed previously. Finally, we will implement novel approaches to help users of our PreSS/MD platform interpret the developed models in terms of key chemical features responsible for skin sensitization. In addition, we will employ biomedical knowledge graphs to elucidate Adverse Outcome Pathways (AOPs) for skin sensitizers. Successful execution of this Phase I project will yield in the development of PreSS/MD as a centralized resource to evaluate the skin sensitization potential for medical devices. We expect this software-as-a-service web server platform will be of great value for companies and sponsors seeking regulatory approval of medical devices.

Public Health Relevance Statement:
PROJECT NARRATIVE Given that medical devices have been documented to contain toxic chemicals that may lead to allergic contact dermatitis, the US Food and Drug Administration requires that all devices be evaluated for possible skin sensitization effects using in vivo assays such as the Guinea pig maximization test. In the effort to modernize skin sensitization safety evaluation methods to reduce in vivo animal testing, herein we propose to develop a software product, PreSS/-MD (Predictor of Skin Sensitization caused by Medical Devices), as an innovative and unique in silico alternative with the potential to better predict human response compared to the existing approaches for skin sensitization assessment. Successful execution of the objectives described in this project will result in a centralized web server platform to evaluate the skin sensitization potential for medical devices, which will be of significant value for companies and sponsors seeking regulatory approval of medical devices.

Project Terms:
Acute; Address; Advanced Development; adverse outcome; Allergic Contact Dermatitis; Animal Testing; Animals; Bayesian Method; Bayesian Modeling; Biological Assay; Cavia; chemical release; Chemical Structure; Chemicals; Computer Models; Computer software; Consumption; Contact Dermatitis; cost; Data; Data Set; Databases; Detection; Development; Devices; Economics; Evaluation; experience; Feedback; Generations; Human; Immune response; in silico; in vitro Assay; in vivo; innovation; Instruction; Interagency Coordinating Committee on the Validation of Alternative Methods; International; Knowledge; knowledge graph; Lead; lymph nodes; machine learning algorithm; Medical; Medical Care Costs; Medical Device; Methods; model development; Modeling; Modernization; Mus; novel; novel strategies; Occupational; occupational hazard; Online Systems; operation; Pathway interactions; Phase; phase 1 study; Poison; Prevalence; Prostheses and Implants; Public Health; Publishing; Pythons; Quantitative Structure-Activity Relationship; Reaction; Reporting; Resources; response; Safety; Skin; skin patch; Small Business Innovation Research Grant; software as a service; Structure; success; systemic toxicity; Test Result; Testing; Time; tool; Toxic effect; Toxin; United States Food and Drug Administration; Validation; web portal; web server; Workers' Compensation; Workplace

1 Medical devices contain chemicals that can leach and cause adverse effects. International standards (ISO 2 10993) require the evaluation of such chemicals for specific toxicity endpoints, including skin sensitization, 3 irritation, and cytotoxicity. Short-terms assays commonly used for this task are time-consuming, expensive, and 4 require the sacrifice of many animals. Emerging FDA directives call to restrict and, eventually, eliminate animal 5 testing of medical and cosmetic products and develop alternative methods including computational tools. To 6 address this unmet need, in Phase I of this project we have created the largest carefully curated and publicly 7 available Guinea Pig Maximization Test (GPMT) dataset and developed first-in-class machine learning models 8 that predict the GPMT outcome. We implemented our models within the fully operational Predictor of Skin 9 Sensitization for Medical Devices (PreSS/MD) web portal. In Phase II, we will create new models and software10 modules for reliable assessment of chemicals found in medical devices for sensitization, irritation, and11 cytotoxicity per ISO 10993 guidance. These modules will be both available for licensing as standalone tools or12 web applications as well as integrated into novel Predictor of Toxicity for Medical Devices (PredTox/MD) web13 portal. The proposed R&D studies are structured around the following Specific Aims: Specific Aim 1: Develop14 a highly curated, comprehensive PredTox/MD database. We will collect, thoroughly curate, and integrate15 public data for all human, in vivo, and in vitro regulatory assays for skin sensitization, irritation/corrosion, and16 cytotoxicity. We will extend our database to include all available data on chemical mixtures and develop special17 curation workflows to handle mixtures of any composition. Specific Aim 2: Develop validated computational18 models to predict sensitization, irritation, and cytotoxicity for chemicals leaching from medical devices.19 We will employ our widely accepted predictive Quantitative Structure-Activity Relationship (QSAR) modeling20 workflow fully compliant with OECD model validation principles. Consensus ensemble models will be developed21 with several descriptor types and machine learning algorithms, including deep and active learning and a22 Bayesian model integrating multiple individual assay models to predict the overall chemical safety. Specific Aim23 3: Develop software modules for assessing medical device toxicity and incorporate these modules into24 PredTox/MD portal. Models and workflows developed in Aim 2 will be programmed as autonomous software25 modules that will be integrated into PredTox/MD platform and available for individual licensing to enable rapid26 multi-point toxicity assessment for extractables and leachables found in medical devices. Successful27 completion of Phase II studies will result in the novel computational toolkit and web-based resource to28 evaluate toxicity of medical devices as required by ISO 10993 guidance.

Public Health Relevance Statement:
There are emerging FDA directives to restrict and, eventually, eliminate animal testing of medical and cosmetic products and integrate alternative New Approach Methods (NAMs) into regulatory programs. However, such evaluation still relies on time-consuming and expensive animal testing. To address this yet unmet need, in Phase II of this project, we will create novel computational toolkit and web-based resource to evaluate the toxicity of medical devices for sensitization, irritation, and cytotoxicity.

Project Terms: