by Nessa Carey
The fall-out was immediate, and played a huge part in the furore around GM foods and their possible impact on human health. A review of the science by representatives of The Royal Society concluded that the data did not support the conclusion that had been drawn.5 But the damage was done. Even today, when study after study has failed to find any links between GM foods and adverse effects on health, this potato work emerges over and over again in the anti-GM community, with Dr Pusztai cast as a badly treated hero and martyr.
This debacle, from which very few individuals or organisations emerged with credit, demonstrates the damage that can be done to a nascent technology field with a premature doomsday conclusion based on insufficient or flawed data. It wasn’t helped by the journal The Lancet publishing the manuscript that eventually emerged from Dr Pusztai.6 The conclusions drawn from the data were less extreme than the ones previously stated on TV, but critics argued that by publishing the paper the journal was supplying oxygen to an already out-of-control fire. The journal countered that not publishing would amount to censorship, and seems to have decided to put its faith in the self-correcting properties of scientific progress.
And boy, does The Lancet have faith in that precept. Possibly rather naive misplaced faith, unfortunately, as the editors seemed to forget that ‘exciting’ bad science may linger in the collective psyche much longer than the good boring stuff that corrects it. Because it was The Lancet that in 1998 published the notoriously bad manuscript from Andrew Wakefield that claimed an association between the development of autism and the measles/mumps/rubella (MMR) vaccine.7 It’s a paper with embarrassingly small sample sizes, awful techniques and terrible conflicts of interest. Huge analyses, based on hundreds of thousands of children worldwide, have demonstrated unequivocally that there is no link between autism and the MMR vaccine.8 Twelve years after publication (twelve years!) The Lancet finally issued a retraction notice.
But go online and it will take you about twelve seconds to find any number of sites that continue to denounce vaccinations as a cause of autism. Vaccination against childhood diseases has probably been the single most impactful health innovation of the last hundred years, but one badly-timed, erroneously published paper has undermined this. In 2017 over 20,000 people in Europe had measles, with 35 deaths recorded, and the World Health Organization has attributed this to people shunning vaccination.9
This is why the scientific community leapt all over the original paper that claimed gene editing led to hundreds of unexpected and off-target changes in the genome. It’s not because they didn’t like the message. It’s because they believed the experiments were poor and the conclusions were scientifically invalid. And they knew from bitter experience how an entire field can be tainted and damaged if an inappropriate concept takes hold.
The sword is two-edged
None of this, however, means that gene editing should just get a free pass on the safety issue, especially when using it as a therapeutic in humans. There’s one potential problem area that is under a lot of investigation at the moment.
p53 may sound like a suburban bus route but it’s actually one of the most important proteins in our cells, especially when we think about cancer. p53 is sometimes known, rather grandiosely, as a Guardian of the Genome. It’s actually not a bad description, though. The DNA in our cells is always under attack from factors that could damage it, such as radiation or particular chemicals. If repaired incorrectly, these changes can produce mutations that in some cases can eventually lead to cancer. It’s often safer for the organism to kill off the damaged cells, and this is where p53 comes in, by basically triggering a cellular suicide response. If p53 is missing or inactive in a cell, that cell tends to accumulate lots of mutations. This lack of functional p53, and the accumulation of mutations, is a hallmark of most cancers.
The potential problem is that one of the events that takes place in a cell during gene editing is basically cutting the DNA, i.e. damaging it. The cellular machinery has no way of ‘knowing’ that this is something we are deliberately trying to do. It will trigger exactly the same damage-limitation response as any other form of DNA damage, particularly activation of the p53 response.
This could be the reason why the percentage of cells that are successfully edited in any experiment is often considerably below 100%. The cells that aren’t edited successfully may simply be too good at preventing DNA damage, because their p53 is working really well.
In 2018 two groups independently showed that the efficiency of gene editing is indeed influenced by the activity of the p53 machinery.10,11 This leads to a worrying hypothesis. Maybe the cells that are edited most effectively are ones where the p53 pathway is defective. This probably doesn’t matter that much in most experimental situations. But it sure as heck matters if you plan to put those cells into human patients. In this scenario, you edit a population of cells, and choose the ones where the editing has been successful. Then you inject these cells into the human recipient. But what if the reason the editing worked in those cells was because they have a faulty p53 system? You’ve now artificially prioritised those cells with the faulty system and chosen to put these into a patient, in preference to cells where the p53 system is still working well. You might basically be giving the recipient a boosted population of cells that are a bit further down the road towards becoming cancerous.
The authors of both the key publications pointed out very responsibly that this is just a theoretical possibility at the moment. It also only applies to particular sub-classes of gene editing where the aim is to correct a faulty gene, rather than just delete it.
It’s actually really helpful that we understand that there might be a relationship between the efficiency of gene editing and the presence or absence of p53. It will help us to plan better experiments to assess long-term safety of the technology when using it as a disease treatment. We can develop and test hypotheses, and check that cells we use for reimplantation into humans have an intact p53 pathway.
It’s all good. Unless you are a publicly listed company specialising in gene editing. The share price of companies developing the most advanced forms of this technology for treating human diseases dropped between five and thirteen points as news about this story spread.12
The fear that the p53 story sparked in the stock market was in many ways quite ironic, as one of the therapeutic areas where gene editing is likely to have a major impact is in the treatment and cure of cancer. That’s because of a new therapy which is yielding astonishing results in certain tumour types. In this approach, scientists extract a specialised type of immune cell from the cancer patient. They then use genetic modification techniques to alter the cell so that it is capable of attacking the cancer and destroying it. In a 2015 trial in a childhood cancer, 27 out of 30 patients became cancer-free after this treatment.13 These were children who had failed to respond to every other treatment available. This level of response is almost unheard of in oncology.
Given how good gene editing is at creating modifications in DNA, it’s not surprising that it’s now being adapted to create the patient-specific immune cells required for this approach. Both the academic and industrial communities are exploring this very wholeheartedly.14
The science is clear, the money less so
Laboratories throughout the world are exploring the potential for gene editing in a dizzying array of conditions. A seven-year-old child with a devastating skin blistering disease has already had their entire epidermis replaced using an older version of genetic modification,15 and gene editing will almost certainly be used to expand this application. Work is pushing ahead on adapting the new technology to treat the fatal muscle-wasting disease Duchenne muscular dystrophy16,17 and the neurodegenerative condition Huntington’s disease.18 There are families at very high risk of heart attacks and strokes because they can’t control the levels of cholesterol in their blood, and who are insensitive to the statin drugs that are such a mainstay of prevention in cardiovascular disease. Preliminary results using
gene editing in animal models have been encouraging.19
These are all conditions where we know that patients face a lifetime of illness, or early death, or both. It’s quite likely that gene editing will provide the first chance of effective treatment – and in fact cures – that these patients and their families have ever known. The main barrier for the application of this technology is unlikely to be technical. It’s far more likely to be economic. Unless we find ways of bringing down the costs of drug development (and gene editing in humans is essentially a very new form of drug) these therapies may never reach the people who need them. Health economics are extraordinarily complex, and also tied in to the ethics of medical practice. Who has the right to decide when ‘expensive’ becomes ‘too expensive’? But the ethical dilemmas of treating or not treating diseases in living humans post-birth are tiddlers compared with the much bigger issue – should we intervene genetically at the moment of conception, when we will change the genome for all eternity?
Notes
1. https://www.cdc.gov/vaccinesafety/concerns/history/narcolepsy-flu.html
2. Schaefer, K.A., Wu, W.H., Colgan, D.F., Tsang, S.H., Bassuk, A.G., Mahajan, V.B. ‘Unexpected mutations after CRISPR-Cas9 editing in vivo’. Nat. Methods (30 May 2017); 14(6): 547–548.
3. https://www.biorxiv.org/content/early/2017/07/05/159707
4. https://medium.com/@GaetanBurgio/should-we-be-worried-about-crispr-cas9-off-target-effects-57dafaf0bd53
5. Murray, Noreen et al. ‘Review of data on possible toxicity of GM potatoes’. The Royal Society (1 June 1999).
6. Ewen, S.W., Pusztai, A. ‘Effect of diets containing genetically modified potatoes expressing Galanthus nivalis lectin on rat small intestine’. Lancet (16 October 1999); 354(9187): 1353–1354.
7. Wakefield, A.J., Murch, S.H., Anthony, A., Linnell, J., Casson, D.M., Malik, M., Berelowitz, M., Dhillon, A.P., Thomson, M.A., Harvey, P., Valentine, A., Davies, S.E., Walker-Smith, J.A. ‘Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children’. Lancet (28 February 1998); 351(9103): 637–641.
8. http://www.who.int/vaccine_safety/committee/topics/mmr/mmr_autism/en/
9. https://www.bbc.co.uk/news/health-43125242
10. Ihry, R.J., Worringer, K.A., Salick, M.R., Frias, E., Ho, D., Theriault, K., Kommineni, S., Chen, J., Sondey, M., Ye, C., Randhawa, R., Kulkarni, T., Yang, Z., McAllister, G., Russ, C., Reece-Hoyes, J., Forrester, W., Hoffman, G.R., Dolmetsch, R., Kaykas, A. ‘p53 inhibits CRISPR-Cas9 engineering in human pluripotent stem cells’. Nat. Med. (July 2018); 24(7): 939–946.
11. Haapaniemi, E., Botla, S., Persson, J., Schmierer, B., Taipale, J. ‘CRISPR-Cas9 genome editing induces a p53-mediated DNA damage response’. Nat. Med. (July 2018); 24(7): 927–930.
12. https://www.cnbc.com/2018/06/11/crispr-stocks-tank-after-research-shows-edited-cells-might-cause-cancer.html
13. Maude, S.L., Frey, N., Shaw, P.A., et al. ‘Chimeric Antigen Receptor T Cells for Sustained Remissions in Leukemia’. The New England Journal of Medicine (2014); 371(16): 1507–1517.
14. https://www.genengnews.com/gen-news-highlights/mustang-bio-launches-crisprcas9-car-t-collaborations-with-harvard-bidmc/81255233
15. Hirsch, T., Rothoeft, T., Teig, N., Bauer, J.W., Pellegrini, G., De Rosa, L., Scaglione, D., Reichelt, J., Klausegger, A., Kneisz, D., Romano, O., Secone Seconetti, A., Contin, R., Enzo, E., Jurman, I., Carulli, S., Jacobsen, F., Luecke, T., Lehnhardt, M., Fischer, M., Kueckelhaus, M., Quaglino, D., Morgante, M., Bicciato, S., Bondanza, S., De Luca, M. ‘Regeneration of the entire human epidermis using transgenic stem cells’. Nature (16 November 2017); 551(7680): 327–332.
16. Liao, H.K., Hatanaka, F., Araoka, T., Reddy, P., Wu, M.Z., Sui, Y., Yamauchi, T., Sakurai, M., O’Keefe, D.D., Núñez-Delicado, E., Guillen, P., Campistol, J.M., Wu, C.J., Lu, L.F., Esteban, C.R., Izpisua Belmonte, J.C. ‘In Vivo Target Gene Activation via CRISPR/Cas9-Mediated Trans-epigenetic Modulation’. Cell (14 December 2017); 171(7): 1495–1507.
17. Lee, K., Conboy, M., Park, H.M., Jiang, F., Kim, H.J., Dewitt, M.A., Mackley, V.A., Chang, K., Rao,. A., Skinner, C., Shobha, T., Mehdipour, M., Liu, H., Huang, W.C., Lan, F., Bray, N.L., Li, S., Corn, J.E., Kataoka, K., Doudna, J.A., Conboy, I., Murthy, N. ‘Nanoparticle delivery of Cas9 ribonucleoprotein and donor DNA in vivo induces homology-directed DNA repair’. Nat. Biomed. Eng. (2017); 1: 889–901.
18. Dabrowska, M., Juzwa, W., Krzyzosiak, W.J., Olejniczak, M. ‘Precise Excision of the CAG Tract from the Huntington Gene by Cas9 Nickases’. Front. Neurosci. (26 February 2018); 12: 75.
19. King, A. ‘A CRISPR edit for heart disease’. Nature (8 March 2018); 555(7695): S23–S25.
7
CHANGING THE GENOME FOR EVER
When we think about treating someone for a disease, we are usually referring to post-natal treatment. The intervention may begin almost immediately after birth. In the UK, all five-day-old babies are eligible for something called the heel prick test.1 Basically, four drops of blood are taken by pricking the baby’s heel with a needle. These four drops are enough to test for nine rare disorders. These are sickle cell disease, cystic fibrosis, a hormone deficiency and six conditions where the baby is unable to metabolise certain chemicals properly. If medical professionals know early in a baby’s life that she or he is affected with one of these conditions, they can make interventions that increase the chances of survival, and a good quality of life. Babies with cystic fibrosis are very prone to overwhelming lung infections, so early treatment with antibiotics can be the difference between life and death. Babies with congenital hypothyroidism don’t grow properly and are at risk of considerable learning disability. This can be prevented by giving them the key hormone that they lack.
Sometimes the intervention doesn’t require any drugs at all. One in 10,000 babies in the UK is born with a condition called phenylketonuria (PKU). People with this disorder are unable to break down one of the amino acids found in proteins, and it builds up to toxic levels in the brain and blood. Before we understood and could test for this genetic condition, affected individuals grew up with learning disability, behavioural issues and other symptoms such as recurrent vomiting and epilepsy. Now, affected babies are identified soon after birth and are put onto a low-protein diet, plus some supplements of the other amino acids they need. Sticking to this regime, and avoiding artificially sweetened products that contain aspartame (because this is converted into the problematic amino acid in the body), completely overcomes the clinical symptoms associated with this genetic condition.
As our lives continue, we tend to take more medical pharmaceuticals. Painkillers, oral contraceptives, antibiotics, antihistamines, and hormone replacement therapy are commonplace. Even people who are lucky enough to age essentially healthily may find themselves taking statins, low-dose steroids and drugs to overcome erectile dysfunction. We may also find ourselves needing other drugs such as antidepressants, insulin for diabetes, antibodies to treat rheumatoid arthritis, or a range of compounds to cure or control cancer.
Whatever type of drugs we take, and for whatever reason, they all have one thing in common. They are designed to interfere with the action of proteins in dysregulated pathological pathways or to replace proteins that are no longer expressed at high enough levels to do their job. What these drugs are not designed to do is to change the DNA of the individual.
In fact, a vast amount of effort goes into making sure that these drugs leave DNA alone. All new drugs are screened during development so that ones that cause changes to DNA – mutations – are deprioritised. One of the reasons is to minimise the risk that the drugs will create potentially carcinogenic mutational changes in the person being treated. Another is to make sure that they don’t cause any mutations in the germ cells – the ones that create eggs or sperm. It’s very rare for drugs to be licensed now if there is a risk of them causing mutations in the germ cells.
If a drug does induce mutations in germ cells, these mutations may stop the eggs or sper
m from developing normally, leading to potential fertility problems. But just as great a worry is that the mutation will be in an egg or sperm that goes on to become part of a new individual. If this happens, the new person will possess the mutation in all the cells of their body and will pass it on to their own children as well.
This kind of DNA change is actually happening all the time, even in people who never take pharmaceuticals. Even though the germ cells have quite stringent mechanisms in place to control mutations, they are inevitable. This is partly in response to environmental factors, and also because there are a lot of complex DNA events that take place during development of eggs and sperm. The more complex an event, the more likelihood that it will go wrong sometimes. And when you realise that men produce about 1,500 sperm every second,2 the potential for changes to creep in to the genome is obvious.
So when scientists are trying to create new drugs, they are usually trying to make sure that these don’t raise the mutation rate significantly above the existing background level.
Gene editing to treat human diseases is in some ways the complete antithesis of almost all drug discovery to date. In gene editing, the aim is absolutely to change the DNA sequence, albeit in a very controlled and specific manner. Chapter 6 described how this might be exploited to treat a range of disorders for which therapeutic options are inadequate or non-existent. If these approaches are successful, it’s unlikely they will have significant effects on the germ cells. For sickle cell disease, the gene editing will take place outside the body, and the blood progenitors will be reintroduced into the bone marrow. In the case of a condition such as Duchenne muscular dystrophy, it’s likely the gene editing reagents will be targeted direct to the muscles. The affected individual will be treated by changing their DNA, but this will be solely what we call a somatic change to their genome. It will affect certain cells in their body, but leave the germline alone.
