Editing Humanity: The CRISPR Revolution and the New Era of Genome Editing

CHAPTER 11 OVERNIGHT SUCCESS

The world got a prime time glimpse of the renaissance of gene therapy on a 2017 episode of the television show America’s Got Talent. Christian Guardino, a genial sixteen-year-old from Long Island, New York, stunned Simon Cowell and the other judges with his rendition of the Jackson 5 hit, “Who’s Lovin’ You.” While the audience cheered the teenager’s amazing voice, the bigger story was out of sight. As an infant, Guardino was diagnosed with a form of Leber’s congenital amaurosis (LCA type 2)—a genetic disorder that causes an inexorable degeneration of cells in the retina. A Fox News report made light of Guardino’s medical ordeal. “When Christian Guardino was young, he learned that he would lose his sight. Fortunately, thanks to some gene therapy, he later regained the gift of sight. In the interim, he turned to music and stuck with it.” The report irked geneticist Ricki Lewis, author of The Forever Fix. “Christian didn’t just order up gene therapy like a side of fries,” she grizzled.¹

Indeed not. Christian was twelve years old when his parents heard about an experimental gene therapy being developed by Jean Bennett, a colleague of Wilson’s at the University of Pennsylvania, and enrolled him. The results were almost literally off the charts. David Dobbs described the transformation: “Guardino could see. Everything that had posed an obstacle before—light and dark, steel and glass, the mobile and the immovable—now brought him pleasure. The world had opened before him.”² That pleasure was shared with millions of TV viewers when Guardino won something called the “Golden Buzzer.” Bennett proudly shows the clip in her talks as Guardino disappears from view under a shower of confetti.

Bennett has dedicated her career to finding a therapy for LCA. In the early ’80s, she’d spent time in the lab of French Anderson, who advised her to “go to Harvard Medical School, learn about the diseases you want to treat, go back to the lab and start working on them.”³ She followed his advice: at Harvard she met her future husband, Al Maguire, a retinal surgeon. In 1990, inspired by Anderson’s historic gene therapy trial, Bennett and Maguire discussed the prospects of transferring genes to the retina. There was just one small problem: there were no known genes, no animal models, no natural history, no vectors, no surgical methods, and no outcome measures.

Bennett’s naïveté proved to be a blessing. After moving to Penn in the early 1990s, she realized that “the retina is pretty cool and that it would actually be fun to look at retinal gene therapy.”⁴ Wilson and his colleagues were building new facilities and developing novel vectors. Eyes are “post-mitotic,” meaning cells don’t divide, which would stop the delivered transgene from becoming diluted. And there was little risk of triggering an immune response. (The eye is an “immune privileged organ,” a mini isolation chamber with a blood barrier and no lymph system.⁵) The effects of the therapy could be tested in various ways and could be measured against the ideal control—the patient’s untreated eye.

The pieces finally came together in 1997, when researchers identified the gene mutated in one of many forms of LCA. The defective gene, RPE65, was one of hundreds of mutated genes that constitute hereditary blindness. There are some 7 million blind people in the United States, including 700,000 children. With only about 1,000 LCA patients in the United States, tackling this disorder was a drop in the bucket. But Bennett had to start somewhere. She learned that the Penn veterinary college housed a breed of dogs with the same gene mutation. A blind four-year-old briard sheepdog mix named Lancelot joined the quest to cure LCA. (Lancelot’s role was essential as the viral vector didn’t work in mice.)

In dogs as in humans, the pace of the retinal degeneration is slow, giving the researchers time to assess the benefits of their treatment. Maguire injected the therapeutic RPE65 gene, packaged in AAV, under the retina of Lancelot and two other dogs. Within a few weeks, Lancelot’s demeanor changed: he began to see, watching and following the veterinary staff.⁶ He became something of a canine celebrity, even visiting the United States Congress. He produced a large family with his sibling, Guinevere, and was eventually adopted by Bennett.

That early momentum stalled as the entire gene therapy community reeled from the Gelsinger tragedy. But in July 2005, Katherine High, a colleague at Children’s Hospital of Philadelphia (CHOP), walked into Bennett’s office and made her an offer she couldn’t refuse: “How would you like to run a clinical trial at CHOP?” Five months later, Bennett and High addressed the Recombinant Advisory Committee (RAC) of the NIH. They were proposing to treat children—a controversial precedent for the gene therapy community. A pivotal moment came when patient advocates Betsy and David Brint recounted their daily struggle to help their youngest son, Alan, who has LCA, who needs the help of a dozen people to get through a day at school. Gene therapy, not to be overly dramatic, was Alan’s only hope. The RAC voted unanimously to grant approval.

The first of a dozen patients in the phase I LCA2 trial were treated in Naples in October 2007, led by Francesca Simonelli. During the procedure, the retinal surgeon inserted a canula the width of an eyelash, through which the genetically modified virus was delivered. The procedure results in a localized retinal detachment, which typically flattens within a few hours.

Maguire warned his wife not to read too much into the early vision results. But one test gave her hope: she performed a pupillometry—a light-mediated reflex—on a patient one month after the injection. Bennett recalled: “When the retina is functioning, it sends a signal through the optic nerve to the brain, which then sends a signal back to the muscle that controls the iris and causes it to constrict. Nobody can constrict a pupil at will, so I couldn’t contain myself when I saw this crystal-clear result.”⁷ She also called the test the bane of her existence, requiring hours measuring pupil diameters and analyzing spreadsheets strewn across her dining room table. But the results on the first three patients were unequivocal, and validated in the New England Journal of Medicine.⁸

Bennett’s youngest patient was Corey Haas. He entered the hospital in 2008 walking with a cane, holding his parents’ hands. After treatment, in a video that has done the rounds at medical conferences, Haas is asked to navigate a makeshift obstacle course comprised of junk retrieved from Bennett’s basement. With his treated eye patched, Corey couldn’t stop bumping into obstacles. But when the patch was switched, he navigated the maze without difficulty. Soon he was riding a bike, playing video games, and throwing a baseball, just like any healthy nine-year-old boy. Bennett’s other patients could suddenly see the moon, the stars, and their own faces. (“Mamma mia!” cried an Italian patient staying with Bennett on seeing her reflection for the first time.) Parents started clamoring for their child to have their other eye injected.

With no more dogs readily available, Bennett donated $10,000 to the lab of veterinary ophthalmologist Kristina Narfstrom to breed some new lines. Bennett picked up the puppies in person and adopted Venus and Mercury. Experiments showed little risk of immune response, allowing Bennett and Maguire to inject both eyes in patients in the Phase III trial. After twelve months, the control group was allowed to cross over to receive the drug as well. By all measures, the response a year later was as good as with the initial cohort of patients.⁹

In 2013, High cofounded Spark Therapeutics, and licensed Bennett and Maguire’s original patent. Four years later, an FDA advisory committee unanimously recommended the approval of therapy, Luxturna. Final FDA approval for the first in vivo gene therapy drug^I came four months after approval of Kymriah, Novartis’s CAR-T cell therapy for acute lymphoblastic leukemia. In March 2018, hospitals in Boston, Miami, and Los Angeles administered Luxturna as a prescription drug for the first time. With a hefty list price of $425,000 per eye, it was no surprise that Spark should attract a big pharma suitor. Roche acquired Spark for $4.3 billion.

Delivering a keynote lecture at a conference in 2018, Bennett ended her talk on a curious note: she had just received a phone call from French Anderson, the father of gene therapy. Anderson had just been released on parole after serving twelve years of a fourteen-year prison sentence, having been convicted of sexual molestation of the teenage daughter of his senior lab director.¹⁰ Bennett insisted there was “abundant evidence of his innocence” and said he hoped to return to the field he’d helped launch almost three decades earlier. “I hope you’ll welcome him back,” she said.

That’s unlikely, but the significance of Bennett’s trailblazing work is not. Hundreds of patients have received gene therapy for ocular diseases since the LCA trial. Bennett hopes that her model will help to develop therapies for other blindness disorders, named after notable ophthalmologists such as Karl Stargardt, Friedrich Best, and Charles Usher. Bennett and Maguire continue their research at the new Center for Advanced Retinal and Ocular Therapeutics, or CAROT for short.

Luxturna was not the first gene therapy to make it through the roller-coaster ride to approval. The first drug was actually Glybera, which was approved in Europe for an ultra-rare disorder called lipoprotein lipase deficiency, resulting in patients essentially having heavy cream in their bloodstream. Manufactured by UniQure, the therapy earned the inglorious mantle of “world’s most expensive drug.” The European market could not support a $1.5 million price and it was subsequently withdrawn, having been administered to just one patient in Berlin.

But drugs like Strimvelis for severe combined immune deficiency (SCID) and Zolgensma for spinal muscular atrophy (SMA), coupled to success in developing CAR-T immunotherapies (Kymriah and Yescarta), signal a new era for gene and cell therapy. There were more than nine hundred registered gene therapies with the FDA at the beginning of 2020. UC Berkeley’s David Schaffer put it nicely: “After twenty years, gene therapy is an overnight success.”¹¹

In late 2017, I invited Shakir Cannon, an African American patient advocate, to write a personal essay for the inaugural issue of a new science journal dedicated to all things CRISPR.^II I had heard Cannon speak at the first CRISPRcon meeting at Berkeley about his struggles with sickle-cell disease (SCD) and his hopes that CRISPR might one day prove an effective therapy, if not a cure. His personal motto was, “Any day without pain is a good day.” He accepted my invitation, signing his email “Thankful.” After a few weeks, I reached out for an update. My emails went unanswered.

Then I heard the awful news. Shakir had died suddenly on December 5, 2017, of pneumonia, just thirty-four years of age. Shakir was one of 100,000 Americans and an estimated 20 million affected worldwide, primarily in Africa and parts of Asia. Together with other mutations in the beta-globin gene resulting in beta thalassemia, these are some of the most common genetic diseases on the planet. Each year 300,000 SCD babies are born.

SCD is a recessive disorder caused by the inheritance of a faulty gene from each parent. Hemoglobin, the protein that carries oxygen in our body, is made up of a quartet of peptide chains—a pair of alpha chains and a pair of beta chains. A tiny mutation, the switch of a T for an A in the beta-globin gene, produces a misformed protein that clumps together. This results in the red blood cells, normally a beautiful flexible biconcave shape, deforming to form rigid sickle-shaped cells that are prone to aggregate and block normal blood flow. Across Africa, SCD has various names—Ahututuo, Chwecheechwe, Nuidudui, Nwiiwii. Roughly translated, they mean “beaten up,” “body biting,” or “body chewing.” About 30 percent of adults with SCD experience debilitating pain every day, in some cases requiring heavy doses of prescription painkillers. “It’s like having your hand slammed in a car door, but instead of it lasting for a few seconds, it lasts for weeks,” said one patient.

Shakir’s short life is a case in point. At age three, Shakir had a stroke, which he overcame with years of physical therapy. Once a month he skipped school to have a blood transfusion. Every night, he received a subcutaneous injection of a drug called Desferal. A portacath was implanted in his chest to help the injections (classmates joked it was his third nipple). He received growth hormone injections because of his short stature. While attending a basketball game with a friend, Shakir experienced a pain crisis so severe he could barely breathe or talk. His parents rushed him to the emergency room at Albany Medical Center, where he stayed for a week.¹² Despite this, Shakir cofounded the Minority Coalition for Precision Medicine and accepted an invitation from the Obama administration to present at the White House.¹³

While the average SCD patient lifespan in the United States is about forty, in Africa most sickle-cell children don’t reach double digits. So why is this deadly disease so prevalent? Sickle-cell carriers (one mutant, one normal beta-globin gene) are inherently resistant to malaria, which kills 500,000 people in Africa each year. This “heterozygous advantage” provides a life-saving selective advantage like a built-in vaccine for SCD carriers and ensures that the sickle-cell gene continues to thrive in areas of the world prone to malaria.

“Blood is by far the most common cell in the body, so it’s not surprising that critters want to feed on blood,” says Merlin Crossley, a geneticist at the University of New South Wales in Sydney. There are about 100 trillion mosquitoes spread across most inhabited areas of the planet. Only a few species spread disease however, and of those, it is only the female mosquitoes that enjoy sucking human blood. In so doing, Crossley’s critters, chiefly Anopheles gambiae, transmit parasites such as Plasmodium falciparum, which causes malaria.

Recent analysis suggests that the SCD mutation first arose in a newborn in West Africa some 7,300 years ago, during the African humid period.¹⁴ That infant unknowingly possessed a stealthy superpower: (s)he would be resistant to malaria, greatly increasing the chances of reaching reproductive age—and a 50:50 chance of passing the same trait onto his or her children. Over the subsequent 250 or so generations to the present day, that single mutation has spread around the globe, especially in Africa, the Mediterranean and Asia, areas devastated by malaria.^III Today it is estimated that 5 percent of the world population carries the sickle-cell trait or another mutation in the beta-globin gene.

We know more about SCD than almost any of the other 6,000 or more documented genetic disorders. The disease was first described by Chicago physician James Herrick in 1910.¹⁵ Herrick reported “freakish” cells in a blood sample from “an intelligent negro of 20”—actually Walter Noel, a dental student from Grenada. Subsequent case reports with multiple affected siblings pointed to a genetic basis. The editors of the Journal of the American Medical Association made a striking pronouncement in 1947:

The most significant feature of sickle cell anemia is not its characteristic bizarre deformation of erythrocytes but the fact that it is apparently the only known disease that is completely confined to a single race… [SCD] is independent of either geography or customs and habits. Its occurrence depends entirely on the presence of Negro blood, even though in extremely small amounts.¹⁶

Two years later, Nobel laureate Linus Pauling discovered that extracted red blood cell proteins from SCD patients ran differently in a gel than healthy controls, predicting (correctly) that the hemoglobin molecule in SCD patients (HbS) carried two additional positive charges. Pauling proposed that SCD was “a disease of the hemoglobin molecule”—the first molecular disease. His prediction was confirmed when a South African physician scientist, Anthony Allison, showed that sickle-cell carriers were resistant to the malaria parasite.¹⁷

Vernon Ingram^IV was a German national who immigrated to the United Kingdom as a teenager one year before World War II. In 1957, Ingram was able to zoom into the amino-acid sequence of the globin chains to pinpoint the molecular aberration in SCD predicted by Pauling—a single amino-acid alteration (glutamic acid to valine) in the beta globin chain. Ingram made his breakthrough at the Cavendish Laboratory in Cambridge, where Crick and Watson had assembled the double helix four years earlier, although Ingram’s lab was a converted bicycle shed.¹⁸ A decade later, Makio Murayama showed how the appearance of that rogue valine residue creates a hydrophobic surface that enables the sickle chains to clump together forming stiff polymers.

DNA sequencing was not invented for another twenty years, but once fellow Cavendish biochemist Fred Sanger developed his eponymous Nobel-worthy sequencing method, it was natural to work out the sequence of the HbS gene mutation. One year after the mutation was genetically spelled out in 1977, Yuet Kan and Andrée Dozy reported the use of a polymorphic DNA marker adjacent to the beta-globin gene to perform prenatal genetic diagnosis for pregnant women with a family history of SCD.

Despite knowing the molecular basis of SCD for more than sixty years, a treatment has remained elusive, dreams of a cure a mirage. That may be about to change: a Bay Area biotech called Global Blood Therapeutics in 2019 had a drug called Voxelotor approved by the FDA; it binds to the mutant hemoglobin and increases its oxygen affinity, although further studies are needed to ensure the drug significantly reduces pain crises.

Several gene therapy biotechs have set their sights on treating SCD and beta-thalassemia. With a large proportion of beta-thalassemia patients lacking any beta-globin production, there are two main strategies for genetic therapy. The most straightforward would be to replace the defective gene by restoring copies of the healthy beta globin gene. On Boxing Day, 2017, a twenty-eight-year-old African American, Jennelle Stephenson, arrived at the NIH clinical center for the beginning of a major clinical trial. Asked to describe the pain she experiences on a scale of 1–10, she said it went beyond a 10, a sharp stabbing pain affecting her shoulders, back, elbows, arms, cheekbones—her entire body.¹⁹ Once she collapsed in a hospital emergency room, only to be accused by hospital staff of faking her distress to get narcotic drugs.

A team led by hematologist John Tisdale purified Stephenson’s stem cells and inserted the correct version of the beta globin gene. To ferry the gene into her cells, Tisdale and his collaborators at Bluebird Bio in Kendall Square chose a modified lentivirus vector. After chemotherapy to cripple her immune system, Stephenson received an infusion of her modified stem cells. (Bluebird’s first SCD patient, a French teenager, was treated with LentiGlobin three years earlier.)²⁰

A few months later, Tisdale compared magnified images of Stephenson’s blood. Before, the sickled blood cells are plainly visible like a biology textbook photo. Tisdale meticulously scans the new slide but comes up empty. “Her blood looks normal,” he says. Stephenson is able to run, swim, and take judo classes, experiencing an endorphin high for the first time. NIH director Francis Collins, whose interest in the genetics of blood diseases traces back to meeting a sickle-cell patient as a young medical student in the 1970s, told 60 Minutes: “I’ve got to be careful, but from every angle that I know how to size this up, this looks like a cure.”²¹

Several other strategies are being tested to treat SCD patients, including tinkering with the regulatory switches of globin production. During pregnancy, fetuses produce a special form of hemoglobin, appropriately called fetal hemoglobin (HbF). This form has a higher affinity for oxygen than adult hemoglobin, all the better to pull oxygen from the maternal bloodstream. HbF is made up of four globin chains—a pair of alpha chains that also exist in adult hemoglobin, along with two gamma (γ-) chains. A few days after birth, production of γ-globin shuts down, replaced by beta globin. Reactivating the fetal form of hemoglobin in SCD and beta thalassemia patients offers a promising strategy. But where’s the switch?

In a 1948 study, Brooklyn pediatrician Janet Watson observed fewer sickle cells in two hundred “Negro newborn infants” than in older patients. “Fetal hemoglobin thus appears to lack the sickling properties of adult hemoglobin,” she concluded.²² In the 1950s, physician Richard Perrine was puzzled that some SCD patients at the Aramco oil company in Saudi Arabia had low levels of anemia, with only mild pain episodes.²³ He too suspected an increase in levels of fetal hemoglobin compensated for the disease. Decades later, Collins began studying a benign genetic disorder called hereditary persistence of fetal hemoglobin (HPFH) in which, as the name suggests, fetal globin production curiously persists after birth into adulthood. Collins discovered a pair of mutations in HPFH patients located just in front, or upstream, of the start of the γ-globin gene. Collins had effectively located the site of a regulatory switch that tells the γ-globin gene to shut down. Disrupt that signal and γ-globin stays on, offering a lifeline to thalassemia and SCD patients.

It took more than twenty years to identify the genetic circuitry leading to the γ-globin switch. In 2008, researchers including Vijay Sankaran and Stuart Orkin at Boston Children’s Hospital conducted a genome-wide association study to identify gene variants associated with high levels of HbF. One of the biggest hits incriminated a gene for a zinc finger transcription factor called BCL11A, which governs this “fetal switch” by shutting down HbF by clamping onto the DNA in the regions identified by Collins’s sleuthing two decades earlier. Sankaran and Orkin nominated BCL11A as an attractive therapeutic target, suggesting that “directed down-regulation of BCL11A in patients would elevate HbF levels and ameliorate the severity of the major β-hemoglobin disorders.”²⁴

How would one go about suppressing BCL11A? Disrupting the gene itself was the simplest route but that won’t work: BCL11A regulates many other genes including some that are expressed in the brain—patients with inherited mutations in BCL11A are on the autism spectrum. Orkin and Daniel Bauer came up with an alternative strategy. They identified a critical regulatory element that enhances BCL11A activity; if disrupted, BCL11A is switched off specifically in red blood cell precursors. By crippling the enhancer sequence, Orkin’s team would silence BCL11A selectively in the cells that give rise to mature red blood cells. In turn, this would remove the brake on HbF production while simultaneously dialing down sickle globin production.

In May 2018, twenty-one-year-old Emmanuel “Manny” Johnson Jr. became the first patient treated in a clinical trial conducted at Dana Farber Cancer Institute, led by Orkin’s colleagues David Williams and Erica Esrick. The Boston team isolated stem cells from Manny’s blood and, using a modified lentivirus, engineered the genes so that when the hematopoietic stem cell (HSC) becomes a red blood cell, the gene switch is automatically turned on. Williams’s team uses RNA interference, a Nobel Prize–winning technology. Following chemotherapy to allow the replacement cells to take hold, Manny’s modified cells were infused intravenously, setting up shop in his bone marrow, pumping out healthy red blood cells.

After seventeen years of monthly blood infusions—Manny suffered a stroke at the age of four—Manny is hoping for a cure not just for himself but also his younger brother Aiden, who has the same disease. “I’m doing this so that my brother might not need all the years of treatment I’ve had to go through,” he says. Six months later, Williams shows Manny before-and-after photographs of his magnified blood showing a frame of healthy round blood cells. “Oh wow, I’ve never seen this before, this is fantastic.”²⁵ Manny hasn’t needed a blood transfusion in the nine months since the therapy. Williams hopes for similar results in his next patients, including twenty-six-year-old Brunel Etienne Jr., who started chemotherapy shortly after attending the Super Bowl. The tickets were a gift from New England Patriots star Devin McCourty, who has a relative with SCD.

Early success stories like these from medical centers in Boston or Stanford or the NIH are wonderful, but have little relevance for the millions of sufferers in Africa, where the impact of SCD is felt the hardest. In October 2019, the NIH and the Gates Foundation announced a $200 million program over four years to produce affordable therapies for patients in Africa. The goal, Collins says, is “to make sure everybody, everywhere has the opportunity to be cured, not just those in high-income countries.” But he readily admits: “This is a bold goal.”²⁶

In her final year at college, Shani Cohen gave birth to her first child, a beautiful baby girl named Eliana (“God has answered me” in Hebrew). Around eight months of age, Shani noticed that her daughter couldn’t stand up in her crib like other infants her age. Eliana was eventually diagnosed with SMA type 2, often called “floppy baby,” caused by a mutation in a gene called SMN. Shani was devastated, the only silver lining being that type 2 is not the most severe form of SMA. The motor neurons die, leaving patients basically paralyzed, many dying of respiratory failure, often before their first birthday. “Think of it as ALS [amyotrophic lateral sclerosis] but in infants,” said Sean Nolan, the CEO at the time of AveXis.²⁷

For Jerry Mendell, success treating patients with SMA and other forms of muscular dystrophy finally came after two decades of persistence. Following the Gelsinger tragedy, he built a group at Nationwide Children’s Hospital in Columbus, Ohio, and set about developing new AAV vectors. In 2008, his colleague Brian Kaspar reported success with a newly engineered virus called AAV9 that had traversed the blood-brain barrier in mice.²⁸ Following more encouraging results in monkeys, Mendell believed they were ready to begin a clinical trial, but no drug companies wanted to take the risk. Kaspar cofounded his own company called BioLife (later renamed AveXis), and licensed the rights to Mendell’s SMA program. But the high AAV doses required to ensure delivery to all the affected tissues, including the diaphragm and the heart, had experts worried. One even warned Kaspar: “You’re going to kill someone—this is going to be Jesse Gelsinger all over again.” Mendell wasn’t about to stop now. “I’m sick of watching kids die,” was his justification.

Mendell’s trial launched in 2014 with fifteen children; within a few weeks, he saw an unwelcome spike in levels of liver enzymes in the first patient—warning signs of inflammation or liver damage. After consulting the FDA, Mendell changed the trial design, administering steroids before the gene therapy. Most complications disappeared. As reported in 2017, all patients showed rapid increases in motor function attributed to elevated levels of the SMN protein.²⁹ Most of the children can now sit unassisted, an unprecedented result. “What used to be called a science experiment, gene therapy, is becoming a reality,” Nolan said, narrating a video of an SMA child sporting a Spider-Man backpack skipping out of hospital unaided. One of Mendell’s first patients was Evelyn Villarreal, whose younger sister Josephine died before the therapy was available. At age three, Evelyn is dancing, doing pushups, and enthralling Senators and staffers on Capitol Hill.

The SMA success story proved irresistible to the Swiss pharma giant Novartis, which acquired AveXis for $8.7 billion in 2018. Zolgensma became the second FDA-approved gene therapy drug in May 2019, but how would Novartis price a potential one-time therapy? CEO Vas Narasimhan had hinted the price could be as high as $5 million. In that light, the final “responsible” list price—$2.1 million—seemed almost reasonable, but Zolgensma still claimed the dubious title of “the world’s most expensive drug” in history. Even the SMA trade paper seemed taken aback: “At $2.125 million, a 60-minute intravenous infusion of Zolgensma costs more than a 2,000-square-foot apartment in Paris with a view of the Eiffel Tower, a brand-new 2019 Aston Martin One-77—among the fastest cars ever made—or a Cirrus Vision SF50 private jet.”³⁰ Novartis allows patients to defray payments over five years, and will offer some sort of money-back guarantee if, for example, the patient dies.

Most experts however felt the exorbitant price was justified: it fell within the guidelines set by the Institute for Clinical and Economic Review, a drug pricing nonprofit, and it could work out cheaper than the full course of a rival drug, Biogen’s Spinraza, over five to ten years. In Novartis’s case, the price had to recoup not only the cost of developing the therapy—an estimated $550 million—but the billions of dollars to acquire AveXis. While there are 10,000–20,000 SMA patients in the United States, there are only about seven hundred children under two years of age, the cutoff for Zolgensma approval.

SMA patient groups welcomed the drug’s approval as priceless. Nathan Yates, an economics professor who suffers from SMA, declared there should not be a price tag on life. He thought of his parents receiving the devastating news that their child has SMA and that there is no cure. “The price of Zolgensma seems insignificant now, don’t you think?”³¹ It wasn’t to the parents of Eliana Cohen. Shani and her husband Ariel despaired as their health insurer denied coverage for Zolgensma and spurned their appeals. With Eliana’s second birthday just a week away, the Cohens launched a desperate crowdsource campaign. Miraculously, they raised $2 million in just five days.

On Maundy Thursday 2019, the New England Journal of Medicine published a study in which a team at the St. Jude Children’s Research Hospital in Memphis, Tennessee, successfully treated eight infant boys with “bubble boy” disease (X-linked SCID), the same disorder that Alain Fischer had treated two decades earlier.³² “We believe that the patients are cured,” said team leader Ewelina Mamcarz. “They’re living normal lives, and they have normal, functional immune systems.” They had returned home, were starting daycare, and making antibodies in response to vaccines.³³ The results were hailed as a complete fix for these patients, although investigators would be monitoring to ensure there were no adverse effects. “The data are extraordinary for every single patient,” said Manny Lichtman, CEO of Mustang Bio, which licensed the rights to the therapy. But how will patients’ families afford MB-107, which is designed to be a one-time treatment? Lichtman suggests a deferred payment scheme over a period of about ten years (assuming the therapy still works).

Gene therapies are expensive to develop and to manufacture, and the biotech companies that take the risks to develop these drugs deserve to recoup their investment. Skyrocketing prices are happening across all areas of the pharmaceutical industry, not just gene therapy. To address the soaring cost of drugs for rare diseases, some economists have floated the idea of healthcare loans, or “mortgages for medicines.”³⁴ It is not clear whether current economics will sustain gene therapies and genome editing treatments. The arrival of new, improved therapies should lead to stiffer competition and reduced prices, but in practice, precision medicines and reformulated generics often justify higher prices. “Certainly, there must be a price that is too high,” said philanthropist John Arnold. $5 million? $20 million? $100 million? “How should society answer that question?”

“I don’t want my legacy to be the most expensive drugs in history,” George Church told me. “We’ve brought down the price of the genome from $3 billion now to ‘zero dollars.’ That I’m proud of. I’m much more excited about that than I am about my contribution to expensive therapy.”³⁵ This is not scalable. Five percent of live births have a Mendelian genetic disorder—the long tail of thousands of rare or orphan diseases. “We’re not going to be spending $2 million on 5 percent of births!” Church says. He estimates that the total cost, including opportunity losses and caregiver costs, is a catastrophic $1 trillion worldwide per year, not to mention the collective pain and suffering.

To appease the sticker shock over its Zolgensma pricing, Novartis executives came up with a lottery. In December 2019, the company announced it would offer one hundred free doses of Zolgensma annually—four names drawn every fortnight—for patients outside the United States. Winners must undergo testing for AAV9 antibodies before AveXis sends the magic dose to the relevant hospital. “Imagine parents putting a child in a draw every two weeks to see if their life can be saved,” sighed Lucy Frost, mother of an SMA child. “I think it could have been done much better.”³⁶

The molecular arsenal to combat cancer and tackle genetic diseases is expanding far beyond gene therapy. Cell therapy, RNA interference, and phage therapy all show promise. Former President Jimmy Carter is the poster boy for CAR-T therapy. A pair of FDA approved drugs, Kymriah and Yescarta, have produced near-miraculous results in some patients, although the effects are short-lived in many others.

Recently, an ancient therapy dating back more than a century has returned to the headlines. Bacteria, as we have seen, evolved their CRISPR defense systems to nullify phage invasions. But in diseases caused by antibiotic-resistant bacteria, we can weaponize these phages to become a new form of precision medicine.

In 1915, physician Frederick Twort discovered a bacteria-killing extract that he said contained an “ultra-microscopic virus.”³⁷ Around the same time, Félix d’Hérelle at the Pasteur Institute was studying dysentery among French cavalry when he noted the bacteria in one of his cultures “dissolved away like sugar in water.” He suspected an invisible microbe, “a virus parasitic on bacteria”³⁸ and coined the phrase bacteriophage (from the Greek phagin, “to eat”). D’Hérelle later performed the first successful phage therapy experiment, treating dysentery patients with phage isolated from another patient’s stool samples. But the reaction of most of his peers ranged from indifference to scorn. D’Hérelle forged a collaboration with George Eliava in Tbilisi, the capital of the Republic of Georgia, who had independently discovered phages that kill cholera samples. Founded in 1923, the Eliava Institute in Tbilisi became the last refuge for decades for phage therapists.³⁹

In September 2017, fifteen-year-old cystic fibrosis patient Isabelle Holdaway underwent a lung transplant at the Great Ormond Street Hospital in London. Although successful, pockets of antibiotic-resistant bacteria seized her liver and her surgical wound. Isabelle’s health deteriorated as bacterial nodules broke through her skin. Her doctor, Helen Spencer, feared the worst. Isabelle’s mother suggested a Hail Mary pass—phage therapy. The medical team contacted Graham Hatfull at the University of Pittsburgh, who had amassed a trove of 15,000 phage strains, housed in a pair of six-foot-tall freezers.⁴⁰ Hatfull identified a trio of phages from his subzero stockpile, named Muddy (isolated from a rotting eggplant),^V BPs (from a storm drain), and ZoeJ (in a soil sample).⁴¹ In June 2018, Isabelle received her first infusion of about one billion phages. Within six weeks, her liver infection had cleared up and the skin lesions were under control. A similarly miraculous outcome occurred in San Diego, when a phage cocktail saved the life of Tom Strathdee, who suffered a serious multidrug-resistant bacterial infection.

Gene therapy is on course to become a mainstream part of 21st-century medicine. Novartis bought facilities to manufacture the large quantities of engineered virus to deliver Zolgensma and other therapies—patients with a neuromuscular disorder require ten times more vector than needed for a localized disease in the eye. Sarepta Therapeutics licensed the rights to two other muscular dystrophy gene therapy programs developed by Mendell’s team. “Our goal is to make Columbus the center of the universe for gene therapy,” said Sarepta CEO Ed Kaye.⁴² Meanwhile, Amicus Therapeutics, led by CEO John Crowley (portrayed by Brendan Fraser in the Harrison Ford film Extraordinary Measures), licensed ten programs for lysosomal storage disorders. Spark Therapeutics is one of several companies developing a gene therapy for hemophilia. Fulvio Mavilio, an executive with Audentes Therapeutics, reported initial success of gene therapy for boys with the incurable disease X-linked myotubular myopathy (XLMTM). Children previously unable to sit up, let alone walk, can now take their first steps unaided, and speak after being taken off a ventilator.

Terry Flotte, the dean of the University of Massachusetts Medical School, is leading a trial for Tay-Sachs disease, which is most common in Ashkenazi Jews. Investigators used gene therapy to supply the missing enzyme in two children, injecting the virus into the brain either directly or via the spinal fluid. And in New York, Ron Crystal, another gene therapy veteran, is launching an ambitious trial to treat Alzheimer’s patients, building on trailblazing work by Allen Roses two decades ago. Roses discovered an association between a rare version of the apolipoprotein E gene (APOE4) and the risk of Alzheimer’s disease. Crystal’s strategy is to deliver a different form of ApoE, which in principle will mop up the harmful variant. “If you’re a mouse, we can cure you of your amyloid plaques,” Crystal tells me. In San Francisco, one of David Schaffer’s new-and-improved AAV vectors using direct evolution has been licensed by Adverum Biotechnologies for use in treating the wet form of age-related macular degeneration.

The renaissance of gene therapy was best illustrated in a 2019 cover story on Jim Wilson in Chemical & Engineering News on the twentieth anniversary of Jesse Gelsinger’s death. The cover headline said it all: “The redemption of James Wilson.”⁴³ His new operation, featuring a staff of two hundred working in multiple buildings at Penn, was more a production line than an academic lab. “Ten years ago, no one would touch Jim with a ten-foot pole. Now everyone is happy to work with Jim and gives him lots of money,” said one biotech CEO.⁴⁴ Indeed, Wilson was finally able to commercialize the production of AAV vectors in a company, RegenXbio, which went public on the sixteenth anniversary of Gelsinger’s death. The image of a sharply dressed Wilson and the company executives laughing as they were showered in confetti at the Nasdaq exchange dismayed Paul Gelsinger. “It really was all about the money,” he said.

Despite this remarkable turnaround, gene augmentation is not a perfect therapy. The faulty gene still lurks in the patient’s cells. More importantly, despite their excellent overall safety profile, Wilson and others have sounded the alarm about safety concerns using AAV vectors at higher doses.⁴⁵ In 2018, Wilson resigned as a scientific advisor to Solid Biosciences, because of concerns about toxicity linked to high AAV dosage. In June 2020, six months after its $3 billion acquisition by Japan’s Astellas Pharma, Audentes disclosed the deaths of two young boys with XLMTM receiving the highest dose of the AAV8 vector. The children died from sepsis, and while pre-existing liver conditions might have been a factor, the FDA halted the trial.⁴⁶ The XLMTM tragedies are a humbling reminder that nature still has a say in what we can and can’t do. Nicole Paulk, a gene therapy expert at UCSF, says we have to design viruses better so such extreme doses aren’t necessary. (The boys who died in the trial received around four quadrillion—a million billion—viruses each.) “As scientists and clinicians,” Paulk says, “we owe it to these boys to make sure this doesn’t happen again.”⁴⁷

So gene therapy’s renaissance is not yet complete. But that has not stopped the technology marching forward. What if we could build on the promise of the gene-editing technologies highlighted earlier and actually go into the cell to correct the corrupt code? What if we take our molecular scissors and repair some of the more than 75,000 mutations, deletions, and rearrangements that give rise to genetic diseases? What if, in the words of Chris Martin in fact, we could “fix you”—genetically speaking?

I. Gene therapies can be classified as in vivo or ex vivo. For in vivo, the therapy is administered directly into the patient’s body. For ex vivo, cells are removed from the body, treated in the lab, and then readministered.

II. The CRISPR Journal was launched in 2018, published by Mary Ann Liebert Inc., with Rodolphe Barrangou as chief editor.

III. There is good evidence to suggest that the SCD mutation has actually occurred spontaneously on four different occasions in different populations.

IV. Ingram’s given name was Werner Adolf Martin Immerwahr.

V. Scientists like to show off their warped sense of humor when it comes to naming genes in certain species (notably fruit flies) or, it turns out, newly discovered viruses. Examples (courtesy of Amanda Warr) include CaptnMurica, IceWarrior, Heffalump, PuppyEggo, BeeBee8, and Megatron. Rule #1 is literally: “Do not name your phage after Nicolas Cage.”