Review of Mitalipov paper CRISPR’ing human embryos: transformative work on the edge

In the same way perhaps that some excited relatives or parents-to-be both gush and worry about a baby before it is even born, our field has been transfixed for a week by the Mitalipov paper on CRISPR’ing human embryos even though the paper just now came out.

Now that the paper is out, we can take a closer look at this “baby” and for us scientists that involves giving it a critical review. In science, “critical” often means a thorough once over with a somewhat skeptical eye, but not necessarily a negative one. Indeed, my overall take on this paper is positive. It is quite strong technically and has many elements that are innovative even though 3 previous studies have already tried CRISPR gene editing in human embryos of various kinds.

This new paper “Correction of a pathogenic gene mutation in human embryos” is in a different category than the other ones in its approach and implications. It is quite rigorous and contains generally very thorough analyses. There are still some very important open questions and I believe there are some issues with perhaps some small overstatements, but by and large this paper is top-notch.

What are the key take homes from this study? Let’s look at my big five questions and now some attempts at answers.

1. Off-target activity? They didn’t detect any. Overall some statements in the paper are perhaps a bit overexuberant such as statements of “abolished mosaicism” (they actually did find a mosaic embryo). I also do not believe they can be quite so confident about “no off-target activity”, when as best as I can tell they did not look thoroughly in enough embryos and cells and in an unbiased manner at the whole genome to really be sure about this. Still their finding of no detectable off-targets so far is impressive.
2. Indels? The Indels present at the target locus even under ideal circumstances in just under 30% of embryos are a big deal and remain a major problem. See part of Figure 3A above in the experiment where they found the 27% of embryos having Indels.
3. Clinical intent overall and NAS report on human gene editing? These folks make no bones about their hope to one day use this kind of technology for human reproduction with specific clinical goals. While they also included some appropriately cautionary statements about future clinical use, at the same time some language such as envisioned possible “rescue” of embryos was potentially concerning. I am highly skeptical that gene editing in the human germline can make sense as a safe and more effective approach than embryo screening by PGD and PGS.
4. Will this paper embolden others to dive in to this space too? Perhaps it will catalyze more research on CRISPR in human embryos. That could be both good in the sense of learning more, but also risky in terms of not everyone doing such a good job as these authors did in considering ethical implications and even in the technological level they used. Also, where will everyone get eggs and sperm for studies?
5. Will this paper lead to a negative, perhaps political reaction? I do think it is better than 50-50 that there will be some kind of political reaction from conservatives about this development and possibly some kind of proposed restrictions.

And more questions pop up now that I’ve read the paper.

10,000 eggs or embryos? What if to get to a clear answer on whether this technology is safe and effective it takes 1,000 or 10,000 human embryos, and hence eggs? CRISPR’ing human embryos at that scale might be needed to get clearer answers on efficacy and safety. Does the hypothetical potential benefit of pursuing human germline editing justify that? These are not every day run of the mill cells. Procurement and use of human eggs and embryos requires extra consideration.

What about epigenetics? Does CRISPR’ing human embryos lead to epigenetic effects that have biological outcomes, some of which may be negative?

Flying blind OK? Another thing to keep in mind is that if this technology were taken in a reproductive direction, you could not analyze the embryos in depth like they did in this paper. You’d have to largely fly blind. At best you could pluck a few blastomeres off for analysis, but you have to leave most of the embryo behind to actually get a baby.

Overall, this is an impressive paper, but one that also raises the stakes on future CRISPR use in humans.