Background

The basic strategy of gene therapy has long been to change the DNA sequence itself. CRISPR-Cas9 is a prime example. This approach, which involves cutting or correcting target genes through double-strand breaks (DSBs), is powerful but carries safety concerns such as unintended DNA recombination or large-scale chromosomal rearrangements. Most importantly, DNA sequence alterations are irreversible. Once cut, they cannot be reversed.

However, the way cells regulate gene expression is not limited to changing the DNA sequence. Approximately 900 types of chromatin-modifying factors and 1,600 types of transcription factors precisely control the on/off state of genes through DNA methylation, histone acetylation, phosphorylation, and other mechanisms. If these chemical tags could be directly rewritten, could gene expression be precisely tuned without altering the DNA blueprint?

This idea, proposed about 20 years ago by Marianne Rots of the University Medical Center Groningen in the Netherlands, is now entering clinical trials.

Key Findings

Epigenome editing is a system that combines a Cas protein (dead Cas) with a guide RNA and an epigenetic effector, removing its cleavage function. Depending on the type of effector, genes can be turned on or off. VP64 (a transcription factor derived from herpes simplex virus) or TET1 (a demethylase) are used for activation, while the KRAB domain or a DNA methyltransferase complex is used for repression. The 'hit and run' technique, reported in 2016 by Angelo Lombardo of the San Raffaele Telethon Institute for Gene Therapy in Milan, was a turning point in this field. It demonstrated that long-lasting effects can be achieved even with only temporary expression of the effector.

The most advanced preclinical data came from a study inhibiting the PCSK9 gene in hepatocytes. nChroma Bio, co-founded by Lombardo and based in Boston, reduced PCSK9 protein levels by more than 98% in mice, and this effect was maintained for over a year. The effect persisted even under conditions where hepatocytes divide and regenerate after partial hepatectomy. In cynomolgus monkeys, PCSK9 was reduced by 90%, and LDL cholesterol was reduced by 70%. Crucially, the ability to reactivate repressed genes by removing methyl marks has also been demonstrated, providing reversibility.

Clinical trials have also begun. Tune Therapeutics, co-founded by Charles Gersbach of Duke University, is developing a therapy that epigenetically silences the hepatitis B virus genome. According to results presented at the European Association for the Study of the Liver (EASL) in May 2026, viral biomarkers were undetectable for up to 17 months in the high-dose group. The Gersbach group also screened more than 11,000 guide RNAs in a 2025 study of Prader-Willi syndrome, discovering a 'master switch' that reactivates a maternally silenced gene across 5-6 megabases on chromosome 15 by demethylating only a single genomic locus.

It is also applicable to CAR-T cell engineering. In an experiment in which the CAR gene was expressed while simultaneously epigenetically silencing two additional genes, 80% of the cultured cells showed all three changes simultaneously. This demonstrates the feasibility of multiplexing existing gene editing and epigenome editing.

Significance and Prospects

The key advantage of epigenome editing is that gene expression can be tuned on a continuous spectrum, rather than a binary on/off switch. It is possible to achieve dose-dependent regulation like a drug, while also expecting long-term effects with a single administration. Currently, more than 12 companies are exploring this technology, and several early-stage clinical trials are underway.

However, as Rots frankly admits, 'We still don't fully understand the rules. We cannot predict the final outcome of biological experiments.' The complexity of the epigenome, in which DNA methylation, acetylation, phosphorylation, and biotinylation act simultaneously, still requires trial and error, and the optimization of viral vectors or lipid nanoparticles for in vivo delivery remains a challenge. Nevertheless, the fact that the concept of reprogramming gene expression without altering the DNA sequence has been repeatedly validated in preclinical and early clinical studies suggests that gene therapy is shifting its paradigm from 'cutting and correcting' to 'tuning and restoring'.