For the millions living with age-related macular degeneration and other RPE-related blinding diseases, cell-based transplantation therapies have long promised restoration of sight. A fundamental question has quietly haunted that promise: are the lab-grown replacement cells truly equivalent to the native tissue they are meant to replace? New three-dimensional epigenomic mapping of the human retinal pigment epithelium now provides the most granular answer yet — and the findings complicate the optimism surrounding iPSC-derived therapies.
Using multi-scale chromatin architecture profiling, the researchers constructed a comprehensive 3D epigenomic atlas of native human RPE, capturing how DNA is physically organized within the nucleus — including topologically associating domains, enhancer-promoter loops, and accessibility patterns across the genome. When iPSC-derived RPE (iPSC-RPE) was mapped against this native reference, meaningful divergences emerged in chromatin conformation and regulatory element activity, suggesting that transcriptional equivalence at the gene expression level does not necessarily translate to equivalent epigenomic identity. Specific loci associated with RPE-critical functions showed differential chromatin states between native and iPSC-derived cells.
This work matters well beyond ophthalmology. The broader field of regenerative medicine has largely relied on transcriptomic similarity — matching RNA expression profiles — as the benchmark for confirming that stem-cell-derived tissues are ready for therapeutic use. This study challenges that paradigm by demonstrating that 3D genome architecture adds a distinct, non-redundant layer of cellular identity that transcriptomics alone cannot capture. For RPE transplantation programs currently in clinical trials, this raises legitimate questions about long-term functional durability and immunological behavior of grafted cells. The study is limited by the cross-sectional nature of epigenomic profiling and cannot yet determine whether the identified divergences translate to measurable functional deficits in vivo. Nevertheless, as a reference atlas published in a top-tier journal, this represents a genuinely significant contribution — establishing a high-resolution baseline that future iPSC optimization and quality-control protocols will likely need to meet.