Scientists identify major safety issue with CRISPR

  • #1
Ygggdrasil
Science Advisor
Insights Author
Gold Member
2019 Award
3,032
2,646

Main Question or Discussion Point

CRISPR-Cas9 technology provides scientists with an easy way to introduce targeted genetic changes to DNA and a number of different companies are looking at ways to use CRISPR to fix genetic diseases in humans. Two groups of researchers, however, have identified a major issue with CRISPR gene editing that has important implications for the safety of the technique.

CRISPR-Cas9 is an enzyme that cuts the DNA inside of the cell at a specific location. After the DNA is cut, scientists can then manipulate the cellular DNA repair systems to introduce mutations at the site of the cut to inactivate a specific gene or introduce foreign DNA sequences at the site of the cut to, for example, fix a disease causing mutation. Normal cells, however, have defenses against DNA damage, including the tumor suppressor protein p53. The researchers found that active p53 proteins inside of the cell greatly reduce the efficiency of gene edits by CRISPR. What this means, however, is that the cells that are successfully edited often have problems with their p53.

Why is this important? p53 helps protect cells against cancer:
The reason why that could be a problem is that p53 dysfunction can cause cancer. And not just occasionally. P53 mutations are responsible for nearly half of ovarian cancers; 43 percent of colorectal cancers; 38 percent of lung cancers; nearly one-third of pancreatic, stomach, and liver cancers; and one-quarter of breast cancers, among others.
https://www.statnews.com/2018/06/11/crispr-hurdle-edited-cells-might-cause-cancer/

If selecting for edited cells selects for cells with defective p53, those cells will likely have an increased propensity to become cancerous in the future.

Small scale studies of CRISPR have not been able to detect increases in cancer, so it is possible that these studies do not tell the whole story. Furthermore, these problems may not occur in some applications of CRISPR (disrupting genes) and may mainly affect cases where scientists are using CRISPR to repair genes. There are also newer technologies that can directly edit DNA without cutting the DNA, which would not be subject to these concerns. Still, these papers point out important limitations in CRISPR technology that need to be considered as CRISPR moves towards clinical applications.

Primary scientific literature:
Haapaniemi et al. 2018 CRISPR–Cas9 genome editing induces a p53-mediated DNA damage response. Nature Medicine. Published online 11 June 2018. https://www.nature.com/articles/s41591-018-0049-z
https://www.biorxiv.org/content/early/2017/08/25/180943

Ihry et al. 2018. p53 inhibits CRISPR–Cas9 engineering in human pluripotent stem cells. Nature Medicine. Published online 11 June 2018. https://www.nature.com/articles/s41591-018-0050-6
https://www.biorxiv.org/content/early/2017/07/26/168443

Popular press:
STAT news: https://www.statnews.com/2018/06/11/crispr-hurdle-edited-cells-might-cause-cancer/
 
Last edited:
  • Like
Likes Demystifier, pinball1970, Filip Larsen and 6 others

Answers and Replies

  • #2
537
146
Yes, let's not fix the disease killing you now because of a risk that might kill you in twenty years.
 
  • Like
Likes General Scientist, Demystifier, BillTre and 2 others
  • #3
atyy
Science Advisor
13,814
2,084
I guess this is not a problem with TALENs?
 
  • Like
Likes BillTre
  • #4
Ygggdrasil
Science Advisor
Insights Author
Gold Member
2019 Award
3,032
2,646
I guess this is not a problem with TALENs?
Because TALENs (and the earlier zinc-finger nuclease technology) also rely on cutting the DNA to generate double-strand breaks, this issue also applies.
 
  • Like
Likes atyy and BillTre
  • #5
russ_watters
Mentor
19,599
5,872
If selecting for edited cells selects for cells with defective p53, those cells will likely have an increased propensity to become cancerous in the future.
As stated, it isn't clear that this concern is meaningful. Is it:
1. Selecting cells with a higher cancer risk results in an increased cancer risk in edited cells vs selecting cells with a lower cancer risk. [Meaningless because you are just selecting and measuring a pre-existing risk without actually altering the risk.]
or:
2. Gene editing increases cancer risk in general, so selecting cells with reduced ability to fight cancer makes it worse.

or something else?
 
  • Like
Likes BillTre
  • #6
Ygggdrasil
Science Advisor
Insights Author
Gold Member
2019 Award
3,032
2,646
As stated, it isn't clear that this concern is meaningful. Is it:
1. Selecting cells with a higher cancer risk results in an increased cancer risk in edited cells vs selecting cells with a lower cancer risk. [Meaningless because you are just selecting and measuring a pre-existing risk without actually altering the risk.]
or:
2. Gene editing increases cancer risk in general, so selecting cells with reduced ability to fight cancer makes it worse.

or something else?
Interpretation #1 is the correct one. While you are correct that this won't result in an increased cancer risk in some applications, there are others where this is a concern. Here are a few ways in which scientists envision using CRISPR:

1) Packaging CRISPR-Cas9 into a virus and applying the virus to the tissue that needs to be repaired (e.g. injection into the eye to treat blindness). As you correctly point out, the concern brought up from the papers does not really apply here. The edits are more likely to go into cells with a pre-existing cancer risk, but likely aren't altering that pre-existing risk. However, these applications are probably furthest from the clinic for other technical and safety reasons (mainly surrounding delivery of the viruses). Furthermore, the fact that active p53 suppresses gene editing suggests that these methods won't be terribly efficient in most of the cells in the body.

2) Scientists extract cells from the individual, edit them in the laboratory, expand them in culture, and re-introduce them into the body. Some of the first planned clinical trials for CRISPR-based therapies are based on this idea. One hopes to engineer immune cells to help them fight cancer and another looks to correct one of the genes involved in sickle-cell anemia. Because scientists select for edited cells in the lab and then grow those cells in the lab before injecting them back into the patients, the concern is that the scientists will be introducing a large population of cancer-prone cells into the patients. This is a particularly important concern when introducing edited stem cells into the body, as envisioned by the sickle-cell anemia therapy.

3) Editing embryos to correct genetic defects at their source. In the process of editing embryos and selection for embryos containing the correct edits, if scientists are not careful, they are likely to also select for embryos with defects in p53. This research makes a strong argument for a moratorium on human germ-line gene editing until we understand more about these issues.
 
Last edited:
  • Like
Likes Grinkle, Greg Bernhardt, BillTre and 1 other person
  • #7
Orodruin
Staff Emeritus
Science Advisor
Homework Helper
Insights Author
Gold Member
16,815
6,610
Yes, let's not fix the disease killing you now because of a risk that might kill you in twenty years.
I do not think anybody advocated that. As with any treatment it is important to investigate possible side effects. However, the fact that side effects exist do not stop us from using the treatments where the severity of the side effects do not outweigh the benefits of the treatment.
 
  • Like
Likes berkeman and jim mcnamara
  • #8
TeethWhitener
Science Advisor
Gold Member
1,752
1,101
@Ygggdrasil I'm curious, because I couldn't read past the abstracts of the papers you posted. The Ihry paper mentions wild-type p53 specifically, so I imagine they probably looked at the effectiveness of CRISPR in the presence of WT versus mutant p53's. The other paper simply mentions "inhibition," and I wonder if simply downregulating or silencing the expression of WT-p53 would increase the effectiveness of CRISPR without adversely selecting for nonfunctional p53 mutants. Does the Haapaniemi paper address this?
 
  • #9
Ygggdrasil
Science Advisor
Insights Author
Gold Member
2019 Award
3,032
2,646
Preprints of both papers are freely available on biorxiv (I've updated the OP to include links to these):
https://www.biorxiv.org/content/early/2017/08/25/180943
https://www.biorxiv.org/content/early/2017/07/26/168443

To answer your question, both papers look at the efficiency of editing in the presence or absence of p53. Both compare p53 wild-type cells to cells in which p53 has been genetically knocked out. Both paper also explore methods to transiently inhibit p53, through either transient expression of MDM2 or a dominant negative verison of p53. More work is necessary to see whether transient inhibition of p53 is harmful. To quote the closing paragraph from Ihry et al.

The toxic response to Cas9 activity has important implications for hPSC-based therapies. P53 inhibition could alleviate toxicity but has the potential to increase off-target mutations and poses a risk for cancer. For ex vivo engineering, Cas9 toxicity combined with clonal expansion could potentially select for P53 mutant cells more tolerant of DNA damage. Although the mutation rate of P53 remains to be determined for other clinically relevant cell types, this is a serious concern for hPSCs. The basal P53 mutation rate in hESCs is significant, and in ref. 14it was found that 3.5% of independent hESC lines and up to 29% of hESCs commonly used in RNA-seq databases have P53 mutations. Before engineering patient cells, the risks and benefits must be fully evaluated. It will be imperative to determine the spontaneous mutation rate of P53 in engineered cells as well as the mutational burden associated with transient P53 inhibition. As hPSC-based cell therapies using genome-edited cells move into the clinic, it will be critical to ensure that patient cells have a functional P53 before and after engineering.
 
  • Like
Likes TeethWhitener
  • #10
537
146
I do not think anybody advocated that. As with any treatment it is important to investigate possible side effects. However, the fact that side effects exist do not stop us from using the treatments where the severity of the side effects do not outweigh the benefits of the treatment.
I was being somewhat facetious, but I can already see headlines and politicians quoting a study like this as a reason not to pursue these gene editing technologies.
 
  • #11
jim mcnamara
Mentor
3,880
2,257
@dipole - you are not adding anything to the thread, other than personal opinion. More of this kind of non-helpful posting is not welcome. Please stop.
 
  • #12
atyy
Science Advisor
13,814
2,084
Because TALENs (and the earlier zinc-finger nuclease technology) also rely on cutting the DNA to generate double-strand breaks, this issue also applies.
Well, I guess we'll find out some time whether it's an issue or not, since TALENs are already in clinical use for CAR-T?
 

Related Threads on Scientists identify major safety issue with CRISPR

Replies
2
Views
2K
  • Last Post
Replies
13
Views
3K
Replies
2
Views
2K
Replies
2
Views
2K
  • Last Post
Replies
1
Views
1K
  • Last Post
Replies
2
Views
855
Replies
17
Views
7K
  • Last Post
Replies
2
Views
2K
  • Last Post
Replies
3
Views
2K
  • Last Post
Replies
3
Views
1K
Top