Beyond animal models: using human pluripotent stem cell-derived stem and progenitor cells to predict cytotoxicity in normal tissues

Abstract

Background The clinical efficacy of anticancer drugs is often limited by dose-dependent toxicity to normal cells, particularly to rapidly dividing stem and progenitor cell populations. Traditional preclinical toxicity screening relies on animal models or primary human cells/immortalized lines, both of which have significant limitations regarding scalability, reproducibility, genetic diversity, and translational relevance. Objective This study aims to establish and validate a human pluripotent stem cell -based platform for profiling the toxicity of cytostatic drugs on healthy, dividing cell populations, and to compare their sensitivity to that of cancer cell lines. Methods Two hiPSC lines (male and female) were reprogrammed from umbilical cord blood cells, fully characterized, and differentiated into neural progenitors and mesenchymal derivatives. Dose–response curves were generated using a dilution series of four anticancer drugs and four other pharmacologically active compounds based on cell viability. Toxic half-maximal values (IC50) were determined for each cell type and compared to those obtained from conventional cancer cell lines. Results The highest sensitivity to DNA-damaging agents was exhibited by pluripotent stem cells, followed by neural progenitors, with mesenchymal derivatives being the least sensitive. Notably, except for mesenchymal cells, both pluripotent stem cells and their differentiated derivatives were more sensitive to cytostatics than cancer cell lines. Conclusion Human iPSC-derived stem/progenitor cells recapitulate the in vivo sensitivity of healthy dividing cell populations to chemotherapy. This platform offers unlimited, genetically consistent, and reproducible access to human cells, rendering it as a powerful complementary approach or even an alternative to animal models and primary human cells for early-phase drug safety profiling.