Type “revolutionary new cancer” into your favorite search engine and you can almost hear the hype machines engage.
Headlines promise revolutionary vaccines, tests, drugs and imminent clinical trials. But despite the many extraordinary advances in this golden age of oncology, few are truly revolutionary. Then there is CRISPR, a technology that has emerged only in the last decade, empowering us to edit the genetic code of almost any living thing. The technology, which uses software to mimic a geneediting process that bacteria use to fight off viruses, can make a gene stop working, repair damaged DNA, or even insert an entirely novel piece of DNA. Science, the flagship journal of the American Association for the Advancement of Science, published a special issue this August on revolutionary technologies. CRISPR led off its short list.
Testing of CRISPR in humans is just beginning, but the possible uses of the technology are vast, with an array of potential benefits, pitfalls and ethical implications. CRISPR is already being used by a food company to give bacterial cultures used in cheese and yogurt more immunity against viruses. In the future, the technology could be used to create better crops or livestock, make synthetic copies of microbes used to manufacture drugs, modify the fertility of pest species so they become less prevalent, eliminate diseases in animals, people or societies, or even design babies with specific characteristics, such as enhanced intelligence or height, in some cases creating traits passed on to future generations.
LUKAS DOW, Ph.D., uses CRISPR to build genetic models of cancer on which medicines can be tested. - COURTESY CORNELL / PHOTO BY ROBYN WISHNA
While genetic engineering has been feasible in the laboratory for some time, CRISPR makes it comparatively easy. “We never really had tools that were powerful enough to do it in a reasonable time frame at a reasonable expense,” says Lukas Dow, Ph.D., assistant professor of biochemistry in medicine at the Sandra and Edward Meyer Cancer Center at Weill Cornell Medicine in New York City. That explains why, in the time it typically takes someone to earn a Ph.D., CRISPR has not just emerged but infiltrated labs worldwide. Its capabilities for genetic manipulation are more sophisticated than even those celebrated in last year’s biggest cancer news, the FDA’s approval of Kymriah (tisagenlecleucel) and Yescarta (axicabtagene ciloleucel), known as chimeric antigen receptor (CAR)-T cell therapies. These expensive but powerful medicines use genetic engineering to enhance the human immune system by training T cells to selectively seek and destroy cancer. That’s a revolution, too, yet the engineering itself is much less refined than CRISPR technology.
“There is so much anticipation and expectation that some people seem to have formed the impression that CRISPR is already curing diseases and in the clinic,” says Michel Sadelain, M.D., Ph.D., of Memorial Sloan Kettering Cancer Center in New York City. Sadelain hopes it won’t be long before that happens. He is constructing what could be the first U.S. trial using CAR-T cells designed with the help of CRISPR, scheduled to begin in 2019. “There is no doubt that CRISPR is a major advance,” says Sadelain, “but there is still a whole body of work to do on its safety.”
CRISPR has had, and will continue to have, growing pains. Early concerns about precision, called off-target effects, have been allayed as techniques improve and more assessment data is collected. “We have been lulled into the view that editing is small and local and controllable,” Allen Bradley of the Wellcome Sanger Institute in England told the journal Nature Methods in July. “(T)he reality of DNA repair in a cell is much more complex.”
MICHEL SADELAIN, M.D., Ph.D., has made it his life's work to engineer human immune cells. - COURTESY MEMORIAL SLOAN KETTERING CANCER CENTER
In June 2018, two articles sharing study findings in Nature Medicine said that the anticancer gene P53 often prevents CRISPR from successfully editing human cells, and when the technology does make an edit, it may be in cells that lack P53. When those edited cells multiply in the body, their malfunctioning P53 increases the risk that cancer might develop.
Scientists who work with CRISPR say that part of the reason for the research was to reveal any such complexities and increase understanding of how the technology works.
Still, the studies hit the mainstream media, including an article in the New York Post with the headline “Futuristic gene-editing technology may cause cancer.”
“That really couldn’t be further from the truth,” Dow says with a sigh. “There’s been a lot of hype about CRISPR’s potential to cure disease. Some people even say (it could) cure all disease. The same hype cycle drives fear.”
HOW CRISPR WORKS
Because fungi fight bacteria, we discovered antibiotics, which may be the most important medical breakthrough of the 20th century. Because bacteria are under constant attack from viruses, we now have CRISPR, perhaps the most intriguing medical breakthrough of this century.
The new technology takes advantage of a natural immune response of the same name that bacteria use to fend off viruses. The mechanism was first seen in E. coli in 1987. You may recall that nucleic acids — DNA and RNA — are each made from just four molecules. Scientists assign each of these a letter, so your genetic code can be read as a 3 billion–character string of just these four letters. Viruses are also constructed largely of RNA or DNA, and they do damage by hijacking the DNA of their host. Inside bacteria, CRISPR works to ward off these viral interlopers, even saving samples of viral DNA, like mug shots, to allow the bacteria to recognize future intruders.
CRISPR stands for Clustered Regularly Interspaced Short Palindromic Repeats. It refers to palindromes, or repeating patterns, of DNA that are found in most single-celled organisms known as archaea and in nearly half of bacteria. The palindromes are like bookmarks for the virus snippets that the bacteria capture and keep “on file” so they are able to recognize any dangerous viruses that come along later. When the bacteria detect such a virus, they send a Cas (CRISPR-associated) enzyme to destroy it by cutting it up.
Our limited ability to decipher DNA meant that we didn’t even notice these curious palindrome patterns in bacterial DNA until the late 1980s. It took another 20 years to see the purpose in the pattern. Now that scientists understand how bacteria use CRISPR to repel viruses, they are harnessing the technique for use in the lab and clinic. In the decade since the technology emerged, CRISPR techniques have proliferated, spurring a host of research, investments, patent battles and sci-fi scenarios.
CRISPR’s importance in cancer is fundamental. The disease is first and foremost a corruption of the body’s DNA, a cruel hitch in our genetic code. In the past 20 years, our ability to read that code has expanded exponentially, driving many advances in treatment and understanding. CRISPR now makes it feasible to write, and rewrite, that code.
To do so, scientists use computers equipped with special software to indicate what they would like to accomplish with CRISPR and are given instructions on the materials they should mix in a test tube to get the desired reaction. According to MIT Technology Review, to use the technology on people, several methods have been proposed that will be tested in clinical trials: gels or creams that go on the skin; drinks or foods; ear injections; CRISPR-engineered skin grafts; and a technique like that used in Kymriah and Yescarta that removes cells from a patient, engineers them in a lab and reinfuses them.
CRISPR IN THE LAB
Dow’s lab focuses on gastrointestinal tumors, primarily colon and pancreatic cancers, and searches for the genetic drivers of disease. There can be thousands of mutations in a tumor, but many of them have little or no role (so-called passengers). What mutations drive a tumor to form and grow? What makes it susceptible to a drug and then able to evolve resistance to that drug?
“If you take apart a car and want to understand what every component of the engine does, it’s very difficult, because often you just take a piece out and it stops working,” Dow says. Tumors are the same way. For several decades now, cancer scientists have been painstakingly building genetic models of cancer in mice and yeast and from cancer cells themselves. They’ll simplify a cancer by knocking out noncritical parts — something akin to the air conditioner. Then they’ll systematically remove and restore other genes to try to figure out whether they might be the cancer’s spark plug or accelerator.
“It’s a way for us to home in on the individual parts of a very complex machine and find out exactly what those things do,” says Dow.
CRISPR allows that work to proceed far more quickly and with much more precision. “By building these more accurate models, our hope and our goal is that we will build systems that are more predictive of human disease,” Dow says. Being able to predict response or resistance patterns means that researchers can more precisely target drugs: “The goal has always been to develop better treatments with the hope of getting curelike responses.”
Cancer is a disease of enormous complexity. In 2000, Douglas Hanahan, Ph.D. (who developed one of the first mouse models of cancer), and Robert Weinberg, Ph.D. (who discovered the Ras cancer-causing gene), published “The Hallmarks of Cancer.” One of the most cited papers in oncology, it outlined six biological processes defining cancer; in 2011 they updated it to 10. In 2018, a group of Australian cancer researchers looked at CRISPR research through the “Hallmarks” lens and found active CRISPR research in all 10 areas of interest. “This system has the potential to dramatically accelerate progress in cancer research,” they concluded.
CRISPR feels so powerful right now because it comes on the heels of a tremendous effort to sequence cancer genomes. Researchers now have easy access to the complete genetic code of thousands of tumors of nearly every cancer type. “We now know in many different tumor types all of the mutations that occur,” says Kathryn O’Donnell, Ph.D., assistant professor at the University of Texas Southwestern Medical Center Department of Molecular Biology and Simmons Cancer Center in Dallas. “Finding the ones that really matter — sifting the drivers from the passengers — is really important.”
KATHRYN O'DONNELL, Ph.D., uses CRISPR to speed up research into how lung cancer develops and grows. - O'DONNELL: COURTESY UT SOUTHWESTERN
O’Donnell’s lab focuses on mechanisms of tumorigenesis in lung cancer, and CRISPR cuts months, even years, off discovery timelines. Not too long ago, the development of a mouse model might have taken researchers years of collaboration. Now it can be accomplished in months.
Some of O’Donnell’s work right now focuses on an overactive cell surface protein in lung cancer called Protocadherin-7, encoded by the PCDH7 gene. Using CRISPR to knock out that gene in a lung cancer mouse model shows a significant reduction in tumorigenesis. Now O’Donnell and her colleagues are trying to figure out how to turn this knowledge into a target for potential therapy. It can be tough to balance her excitement with her necessary critical thinking. “We as scientists need to be cautious and cautiously optimistic,” she says. “I think more research is needed, and we always need to be rigorous and careful as we proceed.”
CRISPR IN PEOPLE
Genetic engineering has long been the dream of cancer researchers. In the early 1990s, Sadelain began his life’s work of engineering human T cells while he was a fellow at the Massachusetts Institute of Technology in Cambridge. His advisers found his vision without merit, and it took twoand- a-half years to achieve even limited proof of concept. With CRISPR, he says, he could now teach a high school student how to engineer a batch of T cells in an afternoon.
Kymriah and Yescarta are both created using a retrovirus to introduce the desired genetic modification to the T cells. Though powerful enough, the technique has limited precision. Genes won’t even necessarily wind up on the same chromosome. That works well enough; the successes with T cell engineering so far show that. But location does matter to genes. Genes work in a context, and that context is their chromosome, their neighborhood. If you insert the same genes in different places, they don’t work in the same way, explains Sadelain. In fact, there are rare cases of genetic insertion leading to leukemia because it occurred next to a critical growth gene.
Sadelain has been working with CRISPR to better target the modification. “We expected to find chromosomal position effects, and we did,” he says. But they were pleasantly surprised to discover just how big a difference it made: In a mouse model of acute lymphoblastic leukemia, CAR-T cells generated by CRISPR “vastly” outperformed the conventionally generated CAR-T cells at every dose level. The next question is, will it work the same way in humans?
Meanwhile, Khalid Shah, Ph.D., director of the Center for Stem Cell Therapeutics and Imaging at Brigham and Women’s Hospital in Boston, wants to use CRISPR to help cancer cells to kill cancer.
Some cancer cells that circulate in blood have a selfhoming tendency that causes them to return to tumor sites. Harnessing this behavior could deliver a Trojan horse inside the walls of a difficult-to-reach tumor. Working with mice, Shah and his colleagues used CRISPR to edit these brain and breast cancer cells, inserting molecules that would initiate cell death. The mice showed a “marked survival benefit.” Shah hopes to bring that technique to clinical trials in humans in the next three to four years.
“Cellular based therapy — CAR-T cell therapy, cancer cell therapy — may become the norm in the next 20 years,” Shah predicts.
CRISPR may also help realize the long-held dream of early cancer diagnosis by detecting miniscule amounts of tumor DNA in a patient’s blood serum or urine.
But most patients who benefit from CRISPR technology won’t get such paradigm-breaking treatments and probably won’t even know. Their doctors will make simple decisions, informed by CRISPR, to guide treatment at an early stage. If it hasn’t happened already, it probably will soon, and it may even go unnoticed by patients. It won’t be long before every branch of cancer research is built on conclusions derived from CRISPR clues.
If CRISPR clears the safety bar set by clinical trials, many more exciting developments could usher in a new era. CRISPR has encouraged blue-sky thinking about treatment for many cancer types. Several trials are already under way in China, where T cell–engineering drugs created with CRISPR have been used in patients. One lab has even repaired a gene in embryos that otherwise would have caused Marfan syndrome, providing “proof of principle for the technical feasibility of gene therapy,” the researchers stated in a paper explaining the findings.
“CRISPR enables precision genetic engineering,” Sadelain says. “That allows you to do more precise experiments, more precise engineering, and we all hope that it’s going to make more precise medicine.”
We’ve never been completely comfortable with genetic engineering as a society, and that’s probably as it should be. Our modest accomplishments continue to set off loud debates across a range of human enterprises: medicine, war, conservation, agriculture, religion.
CRISPR raises the stakes, moving us from mostly speculative fiction to scientific reality. What changes can we make? What changes should we make? What changes dare we make? Might a monster someday be created? “Genetic engineering and claims about genetic disease and genetic problems have a deep, dark history,” says Arthur Caplan, Ph.D., the Drs. William F. and Virginia Connolly Mitty Professor of Bioethics at New York University. “Genetics has been the subject of gross and horrific abuse. We can’t ignore that fact.”
While many people believe humans have no business messing with genetic code, Caplan is not one of them. “I don’t worry about the slippery slope. I think eliminating and preventing diseases makes a lot of sense. I think it would be almost impossible to argue against it.”
But CRISPR is particularly challenging because of the rapid pace of discovery it’s enabling. “Speed is important, but speed kills,” he cautions. While some people worry that bioethics may slow progress, Kaplan doesn’t agree. In fact, history suggests that ethical prudence has significant value. In September 1999, scientists at the University of Pennsylvania broke several ethical guidelines, leading to the death of a patient in an early gene therapy clinical trial. That put the field on hold for 10 years, says Caplan.
He believes CRISPR is a long way from proving that it can efficiently and reliably edit embryos. Consider the Chinese Marfan repair: At what point do we allow that embryo to come to term? Caplan has some thoughts about what shortcuts not to take: “You have to demonstrate that you can do it safely in animals,” he says. “The offspring have to be healthy. And you do have to be able to show that your tool is as accurate as you say it is.”
Caplan would like further safeguards to backstop CRISPR’s rapid rise. A global registry of genetic engineering efforts is needed, because the technology is already widespread. And he hopes that journal editors will not reward rash or racing behavior. A bioethics framework should also strengthen mechanisms for emergencies and desperate measures, tools that can benefit some cancer patients. “We’ve done that with drugs; we should be able to do it with gene editing,” he says. We also need to settle some outstanding business questions. Two major institutions are already embroiled in an epic patent dispute over CRISPR techniques. As more companies and institutions add their incremental improvements, freely using the technology could become more difficult.
And then there is the cost: Even the doctors involved in CAR-T cell research were stunned when Novartis set its initial price point for Kymriah at nearly a half million dollars. CRISPR may bring prices down, but will it be a meaningful price cut?
“Do not tell me about helping humanity until you tell me how you are going to make it affordable,” says Caplan. “Justify your prices. Tell me how you got to them.” And he suggests that this accounting should probably recognize that these discoveries are largely built upon generous public funding of genome mapping. “Paying twice never made me happy,” he says.