I’m not sure on the discussion about clinical trials.

In any given oral medicine, on average about 90%, and often more, of the material is excipient (usually bulking/binding agents) rather than the active pharmaceutical ingredient being tested. Further, the excipients might have flavours or sweetness themselves, as many are sugars. I’m not sure how much you can conclude about the taste of the tested molecules from clinical trial observations.

Relevant source: https://australianprescriber.tg.org.au/articles/pharmaceutical-excipients-where-do-we-begin.html

I am a relatively recently reformed geneticist/molecular biologist and previously used CRISPR/Cas9 at the bench in an experimental context. I no longer work in the lab and admit am not well-read on the latest literature.

I think this approach is interesting, and theoretically executable, but practically infeasible at the current maturity level of the relevant technologies. I’m not sure such a mission would be a good use of expertise and money at this stage. I share the views of a lot of the top level commenters here about the limited feasibility of the approach on both scientific and societal grounds.

I will not repeat the concerns made by others but have two related comments which, while they have been touched on by yourself and others already, hopefully add to the discussion:

1. I am skeptical of claims that editing techniques have zero/minimal/negligible off-target effects generally, and think you need to have an exceptionally, exceptionally strong evidence base to support this before getting near humans, considering the volume of edits you need to make.

You acknowledge this but I feel you downplay the risk of cancer - an accidental point mutation in a tumour suppressor gene or regulatory region in a single founder cell could cause a tumour. Historically gene therapies (e.g for x-linked SCID) have had issues with this in the past. While the mechanics of this are different, there are obviously similarities which would make me (and any regulatory authority) extremely cautious.

2. You propose that variants in non-coding regions would be preferable to target, since off-target mutations would have less effect in non-coding regions.

Firstly, if you’re this worried about off-targets/incorrect edits then this is quite a major concern for the feasibility of the approach (see #1 above). If you’re not actually worried about off-targets/incorrect edits then you should surely be confident to target coding regions?

Secondly, I don’t follow the logic that choosing non-coding targets would be safer as this would lead to only non-coding off-target mutations. As far as I am aware off-target mutations can take place anywhere in the genome, and it is not the case that having a target in a non-coding region would mean that off-targets were also in non-coding regions. [Unless you are using the term “off-target” to refer to any incorrect edit of the target site, and wider unwanted edits - in my community this term referred specifically to ectopic edits elsewhere in the genome away from the target site.]

I wondered if this assertion was based on there being evidence that editing of the type you propose to use has a greater risk of off-target mutations the closer you are to the target site? But even if that is the case, non-coding and coding regions are adjacent to each other in the genome so a nearby mutation could just as well affect a coding region.