CRISPeR Frenzy asked Luigi Naldini of the San Raffaele Telethon Institute for Gene Therapy in Milan for comment on three studies published in June on the preprint server bioRxiv. The experiments were carried out independently by the groups of Kathy Niakan of the Francis Crick Institute in London, Dieter Egli of Columbia University in New York City, and Shoukhrat Mitalipov of Oregon Health & Science University in Portland. These findings heighten safety concerns about heritable genome editing (see the news item by Heidi Ledford in Nature). Below you can read Naldini’s thoughts.

The discovery of large genetic lesions at the editing site is not surprising because it represents one of the known side effects of DNA breakage, followed by the deletion of a more or less large segment to trigger the repair processes. In most cases, the lesions are limited but can also extend over vast tracts of the genome or entire chromosomal segments. Translocations and other rearrangements may also happen during these processes.

It is possible that the frequency of these events, which depends on the availability of the various repair systems and the state of the involved chromatin, varies with the cell type and the affected locus, and this is one of the surprises of the new studies that would suggest a certain predisposition of human embryos to generate these rearrangements, which of course generates concern.

It should also be added that it is difficult to document these lesions with the traditional techniques used to control editing, as they work by sequencing the surroundings of the affected area to verify mutations. If the entire area is no longer there, the probes will not see anything or will be able to do nothing but capture other intact copies still present in the embryo. In any case, the researcher will not be able to recognize this event.

These analyses further show that we still have to perfect both the knowledge of the editing procedures and their application in ultra-sensitive contexts such as the human embryo. It is therefore certainly premature, even only from a scientific-technological point of view, to carry out interventions whose aim is not research but the generation of vital embryos to be implanted, and this of course without considering the legal and ethical aspects.

One can also wonder what the implications of these results are for the most common applications of somatic cell editing that are beginning to enter clinical trials. It is probable that when we edit a large number of cells to generate a therapeutic product, some cells with large lesions may be generated, and then it will be appropriate to verify at least in principle the possible occurrence and frequency.

Then it will be possible to consider whether these rare cells can represent a risk for the patient or if the presence of chromosomal aberrations hinders engraftment and therefore they are eliminated naturally in the body, as reported in some preliminary studies. In general, in the applications being tested, the safety bar is less high, at least as long as the treatment of serious conditions is considered and the risk-benefit ratio is evaluated.