Hacker News Reader

Comment by colechristensen | Hacker News Reader

crispr also makes mistakes. The kind that lead to cancer, disability and death. It’s not a magic wand. Being able to do something in cell culture is interesting but very far away from a therapy.

View on HN

Your point is broadly correct (not sure about disability though), but I think it overstates the danger of what will happen in practice. Yes, if you picked a random target and blasted away, you probably wouldn't have a good time. But anything that even approaches clinical trials is going to get substantial engineering put into it to minimize (and characterize) all of the off-target loci. If there's meaningful editing near an oncogene then that's going to be a deal breaker for a particular guide. When the FDA was discussing the Casgevy application, they went into remarkable detail of how Vertex had measured the off-targets - the regulators really, really don't want to approve anything risky.

Frankly just getting it to work at all is the real hurdle in this case.

> If there's meaningful editing near an oncogene then that's going to be a deal breaker for a particular guide.

At this point do we know every gene that had the potential to cause cancer?

We roughly know the space of possible mutations in the human genomes because we have so many sequenced genomes now: if we don't spot a mutation it's probably not good when it happens, survivorship bias.

The problem with CRISPR is that we cannot control where the off-target effects happen, we can currently only optimise the guiding RNA and the Cas enzyme to have as little off-target effects as possible (but not 0, yet). It would be cool to engineer guiding RNAs that bind in those high mutation-rate areas when they have off-target effects, stuff can mutate there and nothing will happen (probably).

> Frankly just getting it to work at all is the real hurdle in this case

This being the point. Many diseases have been cured dozens of times over in a tissue sample in a lab that never make it to actual therapies because the hurdle is elsewhere. The risks with crispr are as stated, especially flooding your entire body with it to target a virus.

> Your point is broadly correct (not sure about disability though), but I think it overstates the danger of what will happen in practice

https://www.nature.com/articles/nrc1122

One thing I was confused by: why did Vertex choose to reactivate HbF instead of attempting to correct the mutation that causes sickle cell disease in the first place?

CRISPR, in the form used by Vertex, is not capable of directly repairing an existing gene. In the case of sickle cell this means directly changing the mutated nucleotide in the HbA gene. CRISPR is very capable of cutting the genome at precise locations. These cuts lead to lossy repair pathways that introduce mutations or deletions that disable the gene at the spot that CRISPR cut. So, the best you can hope for is that the CRISPR cut leads to a loss of function. It's possible to use CRISPR to introduce new sequences into the DNA, by introducing a new DNA sequence alongside the CRISPR proteins, then hoping that DNA repair "accidentally" uses the genetic sequence you put in to repair the break in the DNA. This is even less efficient than just cutting the DNA, and it would not fix the mutated HbA, so it's not really therapeutically relevant for sickle cell.

There are more recent techniques, notably prime editing, that use a modified version of the CRISPR system that can introduce changes to single bases (nucleotides) in the genome. These have some promise of directly fixing diseases caused by single mutations, but there are hurdles in terms of efficiently delivering the prime editor to the right tissues as well as efficiency of the actual repair.

So how is Vertex activating HbF? Did they find some bases to remove that cause the gene to be expressed more than usual?

They’re snipping another gene (BCL11A) that silences, or turns off, HbF. Turning off BCL11A then allows HbF to be expressed

How do Casgevy and Lyfgenia[1] avoid these dangerous mistakes? Is it something that can be generalized to other therapies, or does it need to be recreated for each "edit" that we want to turn into a therapy?

1. https://www.fda.gov/news-events/press-announcements/fda-appr...

According to your link, both of those are applied to extracted stem cells which are then reintroduced. The ability to perform clonal expansion and DNA sequencing makes it possible that screening for off-target edits could be performed. Though I have no idea if it's actually done.

Reply on news.ycombinator.com