The two studies, both published in Nature Medicine, showed a connection between the success of CRISPR genome editing and the activity of the tumour suppressor gene p53. These findings raised concerns that genome editing may increase cancer risk, and have impacted stock market values of biotech companies working on CRISPR-based medical treatments.
However, the implications of the findings may be limited to therapies based on CRISPR-edited stem cells. 'In the normal, mature cells, gene disruption can happen even when p53 is activated,' said Emma Haapaniemi, lead author of the first study. Sickle-cell disease, thalassaemia and certain forms of blindness are all targets for CRISPR therapies in mature human cells.
The first study, conducted by the Karolinska Institute in Sweden, reported that CRISPR/Cas9 based genome editing is much less efficient in retinal cells than in cancer cells. This because the double-stranded DNA breaks caused by Cas9 as part of the editing process activate p53's tumour suppressor response.
In turn, this may prevent the cell from multiplying, resulting in cell death. In retinal cells lacking p53, genome editing was much more successful.
The authors caution that p53 and related genes 'should be monitored when developing cell-based therapies utilising CRISPR-Cas9'. Many cancer cells either lack p53 or contain a non-functional mutated version. P53 mutations are implicated in 30 to 50 percent of ovarian, colorectal, lung, pancreatic, stomach and liver cancers.
In the second study, researchers from Novartis Institute for Biomedical Research in Cambridge, Massachusetts, saw success rates 17 times higher in CRISPR editing of pluripotent stem cells with an inactivated P53 gene, compared with control cells.
Professor Robin Lovell-Badge, research group leader at the Francis Crick Institute in London, who was not involved in either study, stressed the differences between the application of CRISPR in animals or humans compared with the cell culture experiments carried out in both publications. 'It is therefore not apparent that the problems identified by these two papers, and certainly the scary press releases, are entirely justified – they may be relatively specific to the two studies themselves.'
Darren Griffin, professor of genetics at the University of Kent, who was also not involved in either study, said: 'This paper provides reason for caution, but not necessarily alarm. […] Almost any treatment that has the power to do good, has the power to do harm and this finding should be considered in this broader context.'