Final Report Summary - INDIVUHEART (Individualized early risk assessment for heart diseases)
Human induced pluripotent stem cells (hiPSC) are an ethically uncritical and unlimited source of differentiated human cells for drug safety screening, regenerative medicine and the study of human diseases. An exciting perspective is to use hiPSC as “patients-in-the-dish” addressing the integrated influence of the entire genome on function. While consequences of single gene mutations can be relatively easily examined in isogenic hiPSC-pairs using CRISPR/Cas9 gene editing, the patient-in-the-dish approach requires that functional parameters of control hiPSC-cardiomyocytes (hiPSC-CM) exhibit normal means±SD and that patient-derived hiPSC-CM differ in a meaningful manner. Two important limitations to this end are variability and immaturity of these cells.
IndivuHeart tested the hypothesis that these limitations can be overcome by using 3D-engineered heart tissue (EHT), optimized experimental procedures and standard operating procedures (SOP). The following results were obtained. (i) We developed an SOP-based process from skin biopsy to mature EHTs and optimized protocols for cardiac differentiation of hiPSC, generation and measurements of EHTs. (ii) We established quality criteria and standards in hiPSC culture aiming at eliminating potential confounders affecting cardiac in vitro function. (iii) hiPSC-CMs in the EHT generated under these conditions showed improved structural, electrophysiological and metabolic maturation. (iv) A systematic prospective comparison of >500 hiPSC-EHTs from 3 selected healthy probands, 3 clones per proband and 3 CM differentiation batches per clone was utilized to analyse the reliability of the EHT system. At this point, we found that assay reliability ranged from an excellent coefficient of variation of 0.05-0.2 (drug responses), 0.29 (baseline force), to almost 2 (arrhythmia assessment). Statistical analysis of the sources of variability showed that the CM differentiation introduces the dominant source of variation while the clone and donor level contributed less.
The most ambitious part of the IndivuHeart project was to analyze EHT function from healthy controls (n=39) and patients with dilated or hypertrophic cardiomyopathy (DCM n= 9/HCM n=11) in a fully blinded manner. The study was designed as a prospective clinical trial with pre-specified goals, inclusion and exclusion criteria (NCT02417311 ClinicalTrials.gov). In the end, EHTs from 45 probands/patients fulfilled the prespecified quality criteria. Their contractile phenotype was evaluated over an average of 4 weeks under baseline conditions and under physiological (force-frequency-relation, post-rest-potentiation) and pharmacological stress conditions (isoprenaline, carbachol, calcium sensitizer EMD and the hERG blocker E4031). RNA sequencing data is under current evaluation. Final correlation of the hiPSC-EHT functional data with the clinical (history, ECG, echo, MRI), biomarker (e.g. CK, CKMB, cTnT, NT-proBNP) and genetic data (Illumina Panel Sequencing with the TruSight Cardio Sequencing) is still ongoing. Significant differences between hiPSC-EHTs groups were observed between DCM and healthy controls with DCM-EHTs showing more frequent arrhythmias and lower force. hiPSC from DCM patients depicted lower cardiac differentiation potential, which may point to a more general disease phenotype in these patients. Taken together, the first goal of IndivuHeart is fulfilled with a robust definition of normal values with low SD for a whole panel of healthy control hiPSC-EHT function. The finding that baseline phenotypes of hiPSC-EHTs varied more between probands and patients than responses to physiological or pharmacological stress tests is unexpected and should guide future studies. The 5-year follow-up clinical examination is to be done in autumn 2019 and will answer the second IndivuHeart question whether hiPSC-EHT function has value in predicting the clinical course.
IndivuHeart tested the hypothesis that these limitations can be overcome by using 3D-engineered heart tissue (EHT), optimized experimental procedures and standard operating procedures (SOP). The following results were obtained. (i) We developed an SOP-based process from skin biopsy to mature EHTs and optimized protocols for cardiac differentiation of hiPSC, generation and measurements of EHTs. (ii) We established quality criteria and standards in hiPSC culture aiming at eliminating potential confounders affecting cardiac in vitro function. (iii) hiPSC-CMs in the EHT generated under these conditions showed improved structural, electrophysiological and metabolic maturation. (iv) A systematic prospective comparison of >500 hiPSC-EHTs from 3 selected healthy probands, 3 clones per proband and 3 CM differentiation batches per clone was utilized to analyse the reliability of the EHT system. At this point, we found that assay reliability ranged from an excellent coefficient of variation of 0.05-0.2 (drug responses), 0.29 (baseline force), to almost 2 (arrhythmia assessment). Statistical analysis of the sources of variability showed that the CM differentiation introduces the dominant source of variation while the clone and donor level contributed less.
The most ambitious part of the IndivuHeart project was to analyze EHT function from healthy controls (n=39) and patients with dilated or hypertrophic cardiomyopathy (DCM n= 9/HCM n=11) in a fully blinded manner. The study was designed as a prospective clinical trial with pre-specified goals, inclusion and exclusion criteria (NCT02417311 ClinicalTrials.gov). In the end, EHTs from 45 probands/patients fulfilled the prespecified quality criteria. Their contractile phenotype was evaluated over an average of 4 weeks under baseline conditions and under physiological (force-frequency-relation, post-rest-potentiation) and pharmacological stress conditions (isoprenaline, carbachol, calcium sensitizer EMD and the hERG blocker E4031). RNA sequencing data is under current evaluation. Final correlation of the hiPSC-EHT functional data with the clinical (history, ECG, echo, MRI), biomarker (e.g. CK, CKMB, cTnT, NT-proBNP) and genetic data (Illumina Panel Sequencing with the TruSight Cardio Sequencing) is still ongoing. Significant differences between hiPSC-EHTs groups were observed between DCM and healthy controls with DCM-EHTs showing more frequent arrhythmias and lower force. hiPSC from DCM patients depicted lower cardiac differentiation potential, which may point to a more general disease phenotype in these patients. Taken together, the first goal of IndivuHeart is fulfilled with a robust definition of normal values with low SD for a whole panel of healthy control hiPSC-EHT function. The finding that baseline phenotypes of hiPSC-EHTs varied more between probands and patients than responses to physiological or pharmacological stress tests is unexpected and should guide future studies. The 5-year follow-up clinical examination is to be done in autumn 2019 and will answer the second IndivuHeart question whether hiPSC-EHT function has value in predicting the clinical course.