Project Grant F30HL189252

AI FEEDBACK STRATEGY FOR REPRODUCIBLE CARDIOVASCULAR MICRO-ORGAN DEVELOPMENT - PROJECT ABSTRACT AN ESTIMATED 90% OF CLINICAL TRIALS FOR INVESTIGATIONAL DRUGS FAIL; APPROXIMATELY 45% OF FAILURES ARE ATTRIBUTABLE TO TOXICITIES NOT APPARENT IN ANIMAL MODELS, WITH CARDIOTOXICITY BEING A COMMON CULPRIT. THUS, NEW APPROACH METHODOLOGIES (NAMS) TO MODELING HEART FUNCTION ARE NEEDED TO COMPLEMENT RESEARCH IN EXPERIMENTAL ANIMALS. ADDITIONALLY, AS CARDIOVASCULAR DISEASE IS THE LEADING CAUSE OF MORTALITY WORLDWIDE, SPECIES-SPECIFIC AND SCALABLE MODELS OF HUMAN HEART FUNCTION ARE URGENTLY NEEDED TO INVESTIGATE DISEASE MECHANISMS AND ACCELERATE THE DRUG SCREENING PROCESSES. CARDIAC ORGANOIDS HOLD SIGNIFICANT PROMISE FOR STUDYING HUMAN HEART DEVELOPMENT AND DISEASE AND HAVE BEEN INCREASINGLY USED FOR COMMERCIAL DRUG SCREENING. HOWEVER, CARDIAC ORGANOID MODELS OFTEN HAVE SEVERAL LIMITATIONS, INCLUDING LACK OF VASCULAR PERFUSION, SPHEROID RATHER THAN TUBULAR MORPHOLOGY, AND INABILITY TO SELF-GENERATE FLOW. ADDITIONALLY, TO PERMIT HIGH-THROUGHPUT CARDIAC ORGANOID USE IN RESEARCH STUDIES, KEY CHALLENGES IN THE FIELD OF ORGANOID BIOLOGY INCLUDING LOW-YIELD AND HETEROGENOUS MORPHOLOGIES MUST BE OVERCOME. OUR LAB HAS DEVELOPED A MICROFLUIDIC CARDIOVASCULAR CHIP (MCVCHIP) TO MIMIC HUMAN HEART DEVELOPMENT AND FUNCTION. THE MCVCHIP CONSISTS OF HUMAN INDUCED PLURIPOTENT STEM CELLS (IPSCS) CULTURED IN 3-DIMENSIONAL WELLS TO PROMOTE BIOMIMETIC TUBE-LIKE ORGANOID MORPHOLOGIES THAT BETTER MODEL IN VIVO PROCESSES; ON OCCASION, WE HAVE OBSERVED TUBULAR ORGANOIDS THAT CONTRACT AND GENERATE OSCILLATORY FLOW. HOWEVER, THESE ARE CHALLENGING TO CONSISTENTLY REPRODUCE WITH NO PUBLISHED PROTOCOL TO-DATE RELIABLY ACHIEVING TUBE-LIKE CARDIAC ORGANOIDS CAPABLE OF SELF-GENERATING FLOW. TO ADDRESS THIS PROBLEM, OUR LAB HAS BUILT A CUSTOM ROBOTIC DEVICE TO AUTOMATE THE PROCESS OF SEEDING IPSCS INTO THE CHIP, THUS MINIMIZING VARIABILITY IN CELL DISTRIBUTION DURING THE INITIAL SEEDING PROCESS. HOWEVER, CONTROL OF INITIAL CONDITIONS IS NOT SUFFICIENT TO ENSURE LONGER TERM CONSISTENCY. WE HYPOTHESIZE THAT FEEDBACK SYSTEMS, MIMICKING THOSE OF THE EMBRYO, ARE NECESSARY TO ACHIEVE CONSISTENT AND PHYSIOLOGICALLY RELEVANT CARDIOGENESIS. THUS, TO INCREASE CARDIAC ORGANOID BIOFIDELITY AND REPRODUCIBILITY, WE PROPOSE AI-ENABLED ROBOTIC SYSTEMS WITH AUTOMATED CLOSED- LOOP FEEDBACK CONTROL IN REAL-TIME. THIS PROPOSAL CONSISTS OF 3 AIMS: (1) TO DEVELOP AI-TECHNOLOGY TO CONTINUOUSLY MONITOR AND CLASSIFY DEVELOPING ORGANOIDS (2) TO PREDICT CARDIAC ORGANOID DEVELOPMENTAL TRAJECTORIES AND MAKE COURSE-CORRECTING INTERVENTIONS AND (3) TO APPLY OUR NEW TECHNOLOGIES IN ORDER TO INVESTIGATE BIOPHYSICAL DISEASE MECHANISMS BEHIND SINGLE VENTRICLE CONGENITAL HEART DISEASE. THOUGH THE PRIMARY FOCUS OF THIS STUDY IS ON CARDIAC ORGANOIDS, THE PLATFORM WILL BE DESIGNED TO BE GENERALIZABLE TO OTHER SYSTEMS AND TRANSFERRABLE TO OTHER INVESTIGATOR LABORATORIES FOR MAXIMAL USABILITY AND IMPACT.