Danny Funt

home

authors

danny-funt

Leo Celi, MD, is striving to advance precision medicine. Can he help transform healthcare?

 or it can occur over years, as dogma gradually breaks down in the face of data. For Leo Anthony Celi, MD, there was never a “eureka moment,” as he puts it, during his journey from studying medicine in the Philippines to becoming clinical research director at the Massachusetts Institute of Technology (MIT) Laboratory of Computational Physiology. Instead, it took rigorous training, along with extensive experience in clinical practice, before he recognized that the art of medicine can be woefully unscientific.

Celi, 51, says that realization can be “very disturbing.” He could have found it demoralizing too. Yet colleagues describe an investigator, professor, and physician with seemingly boundless enthusiasm. “Frankly, I don’t think I know anyone who is as energetic as he is,” says Peter Szolovits, PhD, who has worked at MIT for more than 30 years and is now head of its Clinical Decision Making Group, where he collaborates frequently with Celi. “I suspect he’ll continue to work like that when he’s 80.”

It might take that long before Celi’s vision for healthcare is even partially realized. The problem with the art of medicine is that it varies from doctor to doctor. As Szolovits explains, “What fraction of the clinical decisions you’ll make as a practicing doctor have had randomized clinical trials? If you’re an optimist, it’s 10% to 15%.” After more than 2 decades working in the intensive care unit (ICU), Celi knows this firsthand. “When it comes to what should I do for this 86-year-old gentleman in front of me with all these chronic conditions, it’s all guessing because there was no experiment that only enrolled 86-year-old men with his conditions,” he says. “That’s why we see so much variation in care.”

Hunting for the Heart of a Changing Community

“How can we make the practice of medicine more transparent and less likely to be affected by bias?” Celi asks. “To me, the only way to do that is by partnering with computers that can process a voluminous amount of data so much better than humans can.”

Virtually every industry is integrating data science, and practitioners in those fields often feel as though their expertise is under attack. It’s compelling to hear Celi speak on the fallibility of physicians in light of his credentials: Along with master’s degrees in public health from Harvard University and biomedical informatics from MIT, he completed specialized residency fellowships at the Cleveland Clinic, Stanford University, and Harvard University. “People are always paranoid that it’s going to be doctors versus computers,” Celi says. “If I were a patient, I would go to a doctor with a computer.”

Celi and his team foresee an optimal version of doctors with computers, dynamic clinical data mining. They imagine a system in which the “collective experience” of every possible patient on earth is aggregated and analyzed, as they’ve written. “This approach would interrogate data to suggest next-step options and weigh the risks and benefits of a treatment or test for a specific patient.” It would be, in short, “the Holy Grail of personalized medicine.”

That concept is made possible by electronic medical records (EMRs) and made impossible, for now, by the widespread reluctance to share patient data. One remarkable exception is being modeled by the lab at MIT where Celi works, using data from the ICU at Beth Israel Deaconess Medical Center in Boston, where he practices. The lab compiled elaborate data from more than 50,000 hospital admissions and has made the database accessible, for free, to thousands of investigators and students around the world. Granted, nobody can form conclusive insights by studying a decade’s worth of patient data from 1 unit at 1 hospital, but the database has been an amazing resource for exploring research questions and—more important&mdash;demonstrating how big data and machine learning could radically augment the wisdom behind treatment decisions.

From residents at Beth Israel Deaconess to professors at MIT, people who know Celi rave about his ability to facilitate collaboration among clinicians, data scientists, engineers, statisticians, and whoever else is needed for a project to succeed. In the ICU, he learned how the wellbeing of patients depends on a “collaborative culture,” he says. That’s also the only way forward to healthcare’s holy grail.

 was born in the Philippines, psychologists at the Oregon Research Institute were performing groundbreaking experiments on the shortcomings of human judgment. As it happens, the first of their tests observed medical professionals. When examining stomach x-rays for cancer, radiologists “tended to describe their thought processes as subtle and complex and difficult to model,” as Michael Lewis recounts in his latest book, 

. Yet computers using a simple algorithm did better at diagnosing cancer than did the doctors being studied. For the lead Oregon investigator, the results were “generally terrifying.”

In college and then medical school at the University of the Philippines in Manila, Celi certainly had no encounters with data science. His mother was a nurse, and his sisters and cousins also became nurses—in their country, a career in medicine was a guaranteed way out of poverty. When Celi entered medical school, he found “it was all memorization and embracing concepts that are handed to you.” His professors imparted what he later came to see as a naïve sense of confidence.

“In school examinations, there are always multiple choice and true or false, when in fact there are no correct answers,” Celi explains. “It depends on what patients value more: Do they value this probability of improvement from an intervention versus this probability of harm from a side effect? As doctors, we can’t help them navigate that because we ourselves are very uncomfortable with probabilities.”

Now Celi believes medical schools and residency programs “need to be transformed radically,” with data science incorporated into core curricula so that doctors can make better use of EMRs and feel comfortable working with investigators. (New York University School of Medicine is leading the way on this.) He and his lab partners at MIT outlined that recommendation in their 2016 paper Bridging the Health Data Divide. Clinicians are inclined to believe they know what’s best for patients, while engineers can find doctors stubbornly set in their ways. Data silos are notorious in healthcare, but expertise has been placed in silos, too.

After earning his medical degree in 1990, Celi followed the rest of his family to the United States, where he received training in internal medicine at Cleveland Clinic, in infectious diseases at Harvard, and then in critical care medicine at Stanford. In 1999, he joined a startup out of Johns Hopkins University called Visicu, where he worked with software and hardware engineers to design technology that would improve the efficiency of ICUs.

Around that time, Roger Mark, MD, PhD, was pushing a computer on a cart through the ICU at Beth Israel Deaconess Medical Center, plugging it into a monitor to record data with the permission of 1 patient at a time. In an effort to study arrhythmias, Mark recalls, “we’d ask the nurse, ‘Who had a bad night? Who’s unstable?’ Then we’d start recording, and of course the patient would be stable as a rock that day.”

Mark directs the Laboratory of Computational Physiology at MIT, and this primitive method of data collection was the origin of Medical Information Mart for Intensive Care, or MIMIC, the database with de-identified records from patients admitted to the hospital’s ICU from 2001 to 2012. (It was made possible after the institutional review board at Beth Israel approved of Mark’s gathering data without individual consent.) Mark recognized that each patient’s experience was, in essence, an experiment. Years later, people like Celi would be able to parse that data for all sorts of unanticipated findings.

In 2002, Celi accepted a faculty position at the University of Otago in New Zealand, where he developed safety protocols for the ICU while continuing to practice in the emergency department. A medical student there, Xaviour Walker, was struck by Celi’s “thirst for knowledge” and his  example as a “hugely generous and humble person. He always talked to nurses, techs, students, and administration staff to get to know them as people.” Celi encouraged Walker to continue studying in the United States. By 2007, Celi had decided that he, too, needed additional education—the art of medicine, as he’d been taught it, was failing patients. So he returned to Cambridge, Massachusetts.

As Celi began dual master’s programs at MIT and Harvard, the National Institutes of Health awarded Roger Mark’s lab a major grant to expand MIMIC. “I went back to school to try to improve the way we deliver care, and then suddenly the government granted funding to build the tools I was interested in building,” Celi says. “It was almost too good to be true.”

For his master’s thesis at MIT (with Peter Szolovits’ supervision), Celi used MIMIC to analyze 1400 patients admitted to the ICU with acute kidney injury and found that traditional models for predicting hospital mortality, which use prospective observational studies, were less effective than customized modeling would be using retrospective data. In conclusion, he noted that “the question remains whether clinicians will embrace this approach. Will we be able to convince them that information from a very large cohort of patients whose clinical course is stored in an electronic database might be more reliable than a composite of the patients they have encountered in the past whose clinical course may be imperfectly stored in their memory?”

Nearly a decade later, many clinicians remain unconvinced. Szolovits, for his part, maintains hope that the medical community will embrace data science. But, he adds, “I was optimistic in 1974, and it’s taken a while. I would take my optimism with a grain of salt.”

, Celi is involved in about 25 research projects. He continues to work as an internist in the Beth Israel Deaconess ICU for a week each month—as an investigator, he considers it crucial to remain in touch with the challenges facing clinicians. When he’s not at the hospital, Celi typically starts his day at 7 AM on conference calls with collaborators based all over the world. He works at his office on MIT’s campus from 9 AM to 6 PM and then returns home for more conference calls until 11 PM or so. Somehow he finds an hour every day for kickboxing. “Wherever we travel in the world,” explains one of his lab partners and kickboxing converts, Matthieu Komorowski, MD, “we bring the boxing gloves and kicking shields.”

On Friday mornings, Celi teaches a course in the Health Science and Technology program run jointly by MIT and Harvard Medical School. The course, Global Health Informatics, allows students to conduct research using MIMIC. The database includes, for example, progress notes from care providers; timestamped physiological measurements; continuous intravenous drip medications and fluid balances; laboratory test results; and data from the hospital’s 2 critical care information systems, Philips CareVue and iMDsoft MetaVision. All this data collection was possible without disturbing ICU workflow. Mark estimates that 6000 people around the world have completed the credentialing process to use MIMIC.

Kenneth Mukamal, MD, has worked at Beth Israel Deaconess since 1994 and was skeptical when Celi began using natural language processing to analyze doctors’ notes to find out what people were taking before they came to the hospital. Mukamal assigned a resident to go back and compare 100 charts to Celi’s method. “I’m not embarrassed to say that the algorithm was almost perfect. Leo was right, and I was wrong.”

Celi’s lab was designing “really ingenious ways of using what most people like me didn’t even consider data,” Mukamal says. With the natural language processing algorithm in hand, the team analyzed 11,490 patients in 2013 and documented how diuretics can dangerously lower magnesium levels. “That work had a direct impact on how physicians across the United States prescribe these pills,” Mukamal says. “I’m not sure even Leo has a sense of how important some of those papers were.”

At the beginning of the semester, Celi’s course invites anyone at the hospital to pitch ideas that the students will help investigate. Celi is like a “magnet,” as Mark puts it, for attracting diverse participation. For instance, 2 years ago a resident at Beth Israel Deaconess, Robert Stretch, MD, went to Celi with the idea of studying the effect of boarding—when patients are assigned to open beds in a different section of the ICU&mdash;on mortality. “Leo set me up with a machine-learning expert who also happened to have expertise in some sophisticated statistical methods,” Stretch says. “When I agreed that the statistical methodology Leo recommended made sense, he said, ‘Great, I’ve got exactly the right person for you,’ and within 48 hours we had the core of a research group for a publication.” Their research did, in fact, indicate that boarding critically ill patients increased mortality.

Marzyeh Ghassemi, PhD, was working on machine-learning algorithms in Peter Szolovits’ lab when Celi went to her with a research question: How do selective serotonin reuptake inhibitors affect in-hospital mortality? “When you have a collaboration between a clinician who has hunches about treatment, and a machine-learning person who wants to look at data in specific ways, that’s when you learn things more robustly,” Ghassemi says.

To draw meaningful insights, investigators will need access to data from tens of millions of patients, a lesson learned from genomics. At events where investigators explore MIMIC, Celi reminds participants that their findings “might be completely useless.”

One of his priorities right now is persuading hospitals around the world to build their own MIMICs, and not just for the ICU. “My main focus was just [EMRs],” Celi says, “but we’re realizing we need to have a comprehensive representation of the patient,” even beyond clinical data. As long as vendors see data as a commodity worth hoarding, or are overly paranoid about patient privacy, data science will remain stalled. Celi says federal regulation is necessary to compel sharing. “To me,” he explains, “the data really belong to the patients and the public.”

In 2007, Philips Healthcare acquired Visicu, the startup where Celi once worked, for $430 million. Using their connections, he and Mark were able to persuade Philips to release a 200,000-patient subset of its data archive for public use. This “eICU,” like MIMIC, is accessible to investigators and students for free.

The dearth of data has limited the ability to test new technology, “and that’s why people are still very skeptical of the true value of machine learning,” Celi says. Most of the progress with artificial intelligence in healthcare has involved image recognition, for which there are clean data sets available. “I could argue that those are low-hanging fruits,” Celi says. “Improved diagnosis is obviously valuable, but improved treatment is going to benefit so much more from machine learning.”

He says is, not could, but there’s no indication, yet, of when.

 degrees and continuing to practice on weekends in the ICU, Celi and another master’s student, Kenneth Paik, MD, launched a program called Sana, which brought together students in medicine, engineering, and business to improve healthcare in developing countries using new technology. Sana (which means “healthy” or “life” in several languages) had considerable success early on, but the challenge became how to sustain those efforts. Celi and his partners realized their greatest objective needed to be supporting people in those countries to build their own health information systems.

Those who believe in the urgent importance of data sharing can exude an evangelical’s passion, yet they continue to struggle to win over enough converts. Szolovits described a recent meeting with the head of Children’s Hospital Los Angeles. Over dinner, the administrator agreed to contribute his hospital’s data to MIMIC, according to Szolovits, which would have tripled MIMIC’s pediatric data. “But then he sobered up the next day and didn’t do it.”

“Everyone agrees this is the right thing to do,” Szolovits says, “but then they find a reason not to.”

Sana is now under the umbrella of MIT Critical Data, and Celi and a small team travel almost every month to host events—over the past 4 years, they’ve been to the United Kingdom, Ireland, Germany, Italy, France, Greece, Spain, Brazil, Argentina, Colombia, Mexico, Uganda, China, Taiwan, Thailand, Singapore, Australia, and the Philippines. “The best part of this job is meeting all these brilliant students from around the world,” Celi says. His team partners with local institutions to organize “datathons,” often using MIMIC, which aim to grow excitement around data science.

“Young people are more collaborative in the way they approach problem solving,” he says. “They will transform how we do things. If it weren’t for them, we probably would have burned out a long time ago.”

Danny Funt is a freelance journalist based in New York. He can be reached on Twitter 

An Innovative Way of Collecting DNA Samples Should Have Researchers Salivating

Interoperability During Disasters: Lessons from Tragedy

Articles

Nearly everyone extols the need to bust data silos. Meet the man who’s doing it.

Leo Celi and the  'Holy Grail of  Personalized Medicine'

Led by (from left to right) Chief Medical Officer Edward Conner, President and CEO Sandy Macrae, and Vice President of Technology Edward Rebar, Sangamo Therapeutics is looking to convert decades of research into a potentially transformative treatment for genetic diseases.

 investigators at Sangamo have slowly been stocking freezers at their Bay Area headquarters with pairs of zinc fingers, the proteins that regulate gene expression. At one point, their technology was considered the cutting edge of genomic medicine, and Sangamo’s scientists were regarded as the best in the world at building it. Then simpler and more accessible gene-editing techniques came along, such as CRISPR, and many people dismissed Sangamo—as though it manufactured steam engines in an age of internal combustion. Still, the company’s 40 or so investigators continued to toil away at mastering zinc fingers and adding to their library, 1 pair after another.

Sangamo’s president and CEO, Sandy Macrae, PhD, imagines an enormous checkerboard of all the different zinc finger combinations necessary to target particular nucleotides. After investing hundreds of millions of dollars to refine this technology, the company has a proprietary library that now includes about 3000 such pairs, allowing the rapid preparation of medicine that can manipulate an ultra-specific DNA sequence: Targeting efficiency has gone from 2% to above 99.5%, and treatment that once took several months to assemble can now be ready in fewer than 10 days. In short, Macrae explains, “we can now target any nucleotide in the genome.”

At last, Sangamo is starting to take more of its medicine out of the freezer. The results of these early trials could help initiate the transformation in precision treatment—with wide-ranging implications for the cost of care&mdash;that the medical world has been waiting for.

In the past 18 months, Sangamo has overhauled its leadership team and revitalized its mission, changing its name from Sangamo Biosciences to Sangamo Therapeutics and planning a corporate relocation from the East Bay to San Francisco’s biotech hub as it strives to cash in on years of research. When Macrae took over, he had photographs installed throughout the company’s headquarters of people with the rare diseases that these first clinical trials seek to treat. The point was to refocus on patients, and the results of that seem promising.

In November, doctors in Oakland used Sangamo’s zinc finger nucleases to treat a man with the metabolic disease mucopolysaccharidosis type II (MPS II), or Hunter syndrome, the first person to receive treatment that edits genes within the body. With enzyme replacement therapy, the current treatment option, patients with MPS II require weekly transfusions. The patient in Oakland received a single 3-hour transfusion. If it works, he will produce the vital enzymes for the rest of his life.

Some investigators at Sangamo have spent their entire careers pursuing that medicine. For them, “seeing this go literally from the bench to the bedside has really been a life’s work realized,” says Russell DeKelver, PhD, who began working as a research associate and is now the rare diseases team leader at Sangamo.

Once the preeminent developer of gene-based medicine, Sangamo was overshadowed by the emergence of competitors, and biotech investors grew impatient with gene editing in general because of slowness in bringing it to market. If current trials are successful, Sangamo executives say, the medicine could be extended with relative ease to up to 1000 diseases that are also caused by a single gene mutation—they’ve already received $70 million up front from Pfizer to pursue such a treatment. The ultimate goal, however, is to be able to cure any genetic disease, permanently, at its source. Recent progress provides an infusion of optimism for genomic medicine and is no small triumph for a company seeking to reestablish itself at the forefront of its field.

Near the New Year, a family that has 2 sons with hemophilia visited Sangamo headquarters to describe their experience with the disease and to learn about Sangamo’s treatment, which is also entering clinical trials. Macrae recalls a topic of discussion: “What does it mean to change someone who has hemophilia to someone who doesn’t have hemophilia, and how do they feel about that?”

It’s a peculiar way of describing medicine, indicative of the fact that this treatment approach aspires to be, in the most literal sense, life-changing.

 tends to assume an intense gravity, and deservingly so, even if that kind of medicine has been almost entirely hypothetical. In his acclaimed 2016 book The Gene, the oncologist Siddhartha Mukherjee, MD, asks, “And what if we learned to change our genetic code intentionally? If such technologies were available, who would control them, and who would ensure their safety? Who would be the masters, and who the victims, of this technology?”

It’s fitting, in a way, that his warning echoes a prayer before the Jewish Day of Atonement: “How many shall pass away, and how many shall be born? Who shall live, and who shall die?”

Yet when asked to foresee the distant prospects of this medicine, Paul Harmatz, MD, betrays a polite frustration with that focus. Harmatz is a pediatric gastroenterologist at University of California, San Francisco, Benioff Children’s Hospital in Oakland, and he is most concerned about patients with metabolic diseases whom he has been treating for decades. As a principal investigator of the phase 1/2 clinical trials for Sangamo’s in vivo gene-editing treatment of Hunter syndrome, Harmatz describes this moment—cautiously&mdash;as “the forefront of a new paradigm in medicine.”

One of Harmatz’s longtime patients, Brian Madeux, a 44-year-old from Arizona, lacked a gene that produces an enzyme responsible for breaking down toxic carbohydrates in cells throughout the body. One of about 100,000 people, primarily men, has this debilitating disease. Madeux has required 26 operations throughout his life for all sorts of ailments, the Associated Press reported. (However expensive a onetime gene-editing treatment may turn out to be, it waits to be seen how that compares with a lifetime of recurring care.)

Under Harmatz’s supervision, Madeux received a transfusion with Sangamo’s medicine packaged in adeno-associated virus (AAV) vectors, a virus engineered to avoid bodily harm. The virus infects the albumin locus in the liver, where zinc fingers attach to a precise DNA sequence. A highly optimized nuclease makes a double-strand break, and the therapeutic genes are inserted. Billions of copies of this corrective gene are delivered, modifying less than 1% of the body’s vast surplus of albumin. As Sangamo’s chief medical officer, Edward Conner, PhD, puts it, “It’s almost like having a permanent transfusion pump based in the liver.”

“This treatment isn’t actually treating anything,” Harmatz explains. “It’s setting up a factory in the liver to produce a lot of this enzyme over a long time period.” About 6 months after Madeux’s treatment, a liver biopsy will reveal whether the enzyme is being produced successfully.

Doctors could, in theory, use that same process to implant corrective genes for any monogenic disease that involves a protein deficiency. Clinical trials are under way for MPS I, MPS II, and hemophilia B, with trials for Fabry disease forthcoming. There are altruistic reasons for focusing on ultrarare diseases but also practical ones: For novel technologies in particular, it’s extremely difficult to get FDA approval for diseases with huge patient populations, and those trials require elaborate designs to demonstrate efficacy. The MPS II phase 1/2 trials, by contrast, will treat up to 12 patients. These rare diseases have significant unmet need, largely because pharmaceutical and biotech companies don’t see the potential return on investment.

“We’re just exponentially grateful to Sangamo, recognizing that MPS diseases are so small compared with diseases everybody knows about,” says Terri Klein, president and CEO of the National MPS Society. “Brian is very brave, not knowing what to expect. Someone has to be on that front line.”

Indeed, the cutting edge brings risk. In 1999, 18-year-old Jesse Gelsinger died from complications after receiving gene therapy in a clinical trial at the University of Pennsylvania. An FDA investigation found several areas of negligence. Genomic medicine has advanced dramatically since then, but the lessons from that first “biotech death” are not forgotten.

 was not so much a scientific breakthrough as it was a shrewd observation by its founder, Edward Lanphier. After graduating from Knox College with a degree in biochemistry, Lanphier entered the business side of biotech and by 1995 was in charge of building the “patent estate” of Somatix, a California-based gene-therapy company. He was captivated by research at 4 labs studying the impressive DNA-binding utility of zinc fingers, especially work out of the Medical Research Council in London led by the Nobel laureate Aaron Klug, PhD. Somatix wasn’t interested, but Lanphier was: He licensed intellectual property from those 4 labs and set up laboratory operations of his own.

In 1997, Lanphier moved his company from Colorado to Richmond, about 20 miles north of San Francisco, and after a couple of rounds of financing took it public in 2000. One of his first hires was Edward Rebar, then just a year removed from receiving his PhD at the Massachusetts Institute of Technology, where he’d worked on the first system for selecting zinc fingers to bind to new DNA sequences. Rebar has been the leading architect behind zinc finger optimization at Sangamo and is now its vice president of technology. Another pivotal early hire was Philip Gregory, PhD, who recalls Sangamo’s thriftiness at that stage: As Lanphier used to say, “we were throwing dimes around like manhole covers.”

Two months after Sangamo raised $48.8 million in its initial public offering (IPO), the former New York Times (NYT) biotech reporter Andrew Pollack offered this sobering check on the industry: “The sequencing of the human genome is often described as a triumph akin to landing on the moon. For investors, that might not be the most comforting analogy. Three decades after Apollo 11, no one has yet made a profit going to the moon.” Investors were apparently undeterred. Later that year, the Times quoted a joke going around Wall Street: “If you want to cause a buying stampede among investors, just insert the prefix ‘gen’ into a company’s name.”

Indeed, a year after its IPO, Sangamo acquired Gendaq in a transaction valued at $40 million. Wall Street may have sniggered, but Sangamo got the last laugh—that was almost certainly the most lucrative business move the company would ever make. Cofounded by Aaron Klug, Gendaq had been researching zinc fingers for 15 years. Sangamo added 16 scientists to its roster and acquired 22 patent applications, 2 issued patents, and $6 million in cash. “This consolidation of intellectual property and scientific assets catapults Sangamo into a unique position,” Klug said at the time. That leverage would go on to empower the company and embolden its critics.

Sangamo’s first decade was devoted to proving the efficacy of zinc fingers as a viable medicine, and Gregory led a team that sought to discover whether zinc finger nucleases could edit genes in human cells. After 2 years of research, their work led to a landmark 2005 paper in Nature revealing that site-specific gene modification with a nuclease was possible in a human T cell without suppressing anything else detectable in the genome.

“That was really the result everybody was waiting for,” Gregory says. “I have to say, we were very worried that we were getting tricked by the fantasy of the data. There had been a technology a few years earlier that had offered similar gene correction, but it turned out to be essentially an experimental artifact. So the supplementary data in our paper are just gigantic. We did every experiment imaginable to try to confirm that the data were real and not some artifact of the way we were measuring outcomes. 

“To date,” he points out, “that’s still the most rigorous evaluation, from a data perspective, of gene correction.”

Three months after that article was published, Sangamo was accused of controlling a “zinc finger nuclease monopoly” in a Nature article by Christopher Thomas Scott, a graduate student at Stanford who went on to direct the university’s Program on Stem Cells in Society. “[T]he secrets of a cutting-edge technology that could transform gene therapy lie hidden in the intellectual property vaults of a small biotech company.” The strategy makes financial sense, Scott wrote, but it could hinder advancement of the science and antagonize investigators, who might then pursue alternate technologies.

“When zinc finger nucleases were the only game in town, it rankled a number of people that Sangamo was keeping their technology close to the chest,” Dana Carroll, PhD, one of the leading scientists in the field, tells 

™. “They developed the technology within their own company to a very, very high standard. Then they pursued practical applications that I think are admirable. It’s hard to criticize a company too severely for protecting their investment in the platform that they were developing.” (Sangamo licenses a patent from the University of Utah for a technology Carroll developed.)

During the mid-2000s, Sangamo did well over 100 material transfer agreements to make its technology available to other labs. “What people don’t talk about when discussing our control of the IP was that it’s extraordinarily difficult to make the kind of high-quality zinc finger proteins that we were making,” says Lanphier, who retired from Sangamo last year. “So yes, there was a patent barrier to entry, but in reality there was a much greater technical barrier.” The threat, the company believed, was not only competition but also the potential of second-rate experiments to sully their product’s reputation.

Sangamo’s breakthrough with zinc finger nucleases in 2005 eventually yielded 4 treatment approaches: gene regulation, in which genes are knocked out or activated; genome editing, in which a new gene is delivered to the nucleus of a cell; gene therapy, in which DNA is cut and a new gene is inserted; and cell therapy, in which T cells or stem cells are edited outside the body. Macrae argues that conflating the company’s original and current zinc finger technology “is like comparing the Model T Ford to a Porsche. The modern car still has 4 wheels and an engine, but it’s a very different beast.”

And so, with this optimized technology in hand, Sangamo and investigators at the University of Pennsylvania launched clinical trials in 2009 for a treatment of HIV. Doctors removed T cells from 102 patients and used zinc fingers to disable their CCR5 gene, making it harder for the virus to take hold. Those edited T cells were allowed to multiply and then were transfused back into patients. NYT reported on one such patient who, a month after being treated, stopped taking his antiviral drugs for 12 weeks as part of the experiment: “As expected, the amount of HIV in his blood shot up. But then it fell back to an undetectable level just before the end of the 12-week period. The patient’s immune cell counts also shot up. ‘I felt like Superman,’ he said in an interview.”

Although the treatment “worked,” in a technical sense, Sangamo realized it wasn’t a sufficient approach and has since diverted its focus. “What we didn’t know was how many of the T cells needed to mutate, and was it just T cells? Was it stem cells?” Macrae explains. “Now we need to work out whether other people could take this forward for us and look for that ultimate cure for HIV. It’s not something we’re going to do, because we just don’t have the bandwidth.”

Wall Street, meanwhile, is less bullish on genomic medicine than when Sangamo went public. “Genomics as a field has ebbed and flowed in terms of investor enthusiasm,” says Gregory, who left Sangamo as chief scientific officer in 2015. “Around the sequencing of the human genome, there was all this enthusiasm about how it would change medicine forever. That, of course, took a long time to happen, and the investor community lost patience. It turns out things are very complicated.”

With about 175 employees and $90 million to $100 million in operating expenses, Sangamo is relatively small for its field, and so it must be, in Macrae’s words, “ruthlessly strategic.” When developing a treatment would be unreasonably expensive and beyond the biological expertise of in-house scientists, Sangamo has formed partnerships with other companies; for example, with Bioverativ for treatment of beta-thalassemia and sickle cell disease, with Shire for treatment of Huntington disease, and with Pfizer for treatment of hemophilia A. That Pfizer deal included an initial $70 million for Sangamo to search for candidate zinc finger proteins, the largest up-front payment for a gene therapy to date. 

In early January, Sangamo and Pfizer announced a collaboration to pursue treatment for a form of amyotrophic lateral sclerosis (ALS) that makes up about one-third of all ALS cases. Between its two deals with Pfizer, Sangamo is eligible to receive up to $625 million in milestone payments, along with royalties.

 used to go into meetings and have to explain the basic principles of gene editing. That’s no longer a problem. The downside is that when people nowadays hear “gene editing,” they’re likely to respond, “Oh, you mean CRISPR?” 

That presumption is hardly limited to lay people. Take Mukherjee’s book The Gene: It contains 495 pages on genetics, an extensive discussion of CRISPR, a footnote about another new gene-editing technology called TALEN, and not a word about Sangamo or its zinc finger proteins. 

“If you are in an academic research lab and you want to do genome editing,” Carroll says, “you’d be a lunatic if you didn’t adopt the CRISPR technology at least initially because it’s easy

When Macrae makes a similar point, saying, “If I were doing my postdoc again, I would happily use CRISPR to knock out a cell or treat a mouse,” it has a slightly condescending air—as though someone who made grand pianos were to say, “If I were learning to play again, I would happily use a keyboard.” 

Daniel Voytas, PhD, a specialist in the genomic engineering of plants at the University of Minnesota, cofounded the Zinc Finger Consortium, which offered open-source software for academic research. Later, he helped invent TALEN. “It was difficult to make zinc finger nucleases, and Sangamo was the best in the world at it,” Voytas says. “There was a tsunami of publications because everyone had access to new reagents, and Sangamo was sort of diluted by that.”

Sometimes an older technology has had more time to be perfected. Other times it has simply gone out of date. “There’s a perception that zinc fingers were slow and clumsy and large and expensive,” Macrae says. As for CRISPR, development was impeded by patent disputes among its 3 leading companies, and some say the technology is more susceptible to off-target mutations, which could potentially cause cancer. Crispr Therapeutics is seeking approval to pursue clinical trials this year for beta-thalassemia in Europe and sickle cell disease in the United States.

Perhaps the greatest obstacle for any gene therapy is delivering the medicine to the targeted part of the body. By that measure, Sangamo can be grateful that CRISPR and TALEN have dramatically expanded the search for an optimal delivery method.

“One of the things that Sangamo has in its favor is that zinc fingers are relatively small compared with the CRISPR-Cas9 protein and the TALENs,” Carroll says. “They can be packaged more easily into AAV vectors,” the virus that Sangamo is using to treat Hunter syndrome. “That’s not an insignificant thing.” In 2013, Sangamo acquired Ceregene, a private company developing the AAV vector platform, along with more than 120 issued, pending, or in-licensed patents.

“It’s all about delivery. Delivery will open up new diseases,” Macrae says. “That will be the future of the company.”

Sangamo’s ever-evolving zinc finger technology could catapult precision medicine. So why haven’t you heard about it?

Lost in the CRISPR Hype, a Gene-Editing Giant Is Fighting Back

DNAsimple CEO Olivier Noel (right) and COO Joel Cobble (left) talk business in a New York City restaurant earlier this month with Mark Cuban, who invested $200,000 in their company on this season of ABC’s “Shark Tank.” (Photo courtesy DNAsimple)

PHILADELPHIA — Last month, 2 Penn State doctoral candidates, Olivier Noel and Joel Cobble, invited friends over to watch a new episode of “Shark Tank.” The former lab partners are at the helm of DNAsimple, 

 born out of their frustration with the tedious process traditionally used to recruit donors for genetics studies. They gathered that night to watch Noel pitch their solution on ABC: In essence, give saliva to researchers, get $50.

It takes just a few minutes to register online with DNAsimple, which launched 2 years ago. If donors fit a researcher’s parameters, Noel’s company confirms their medical history and then mails them a kit. Swab inside the cheek, mail it back, and get paid. Participants earn $50 each time their DNA is used in a study.

More than 40,000 entrepreneurs apply each year to appear on “Shark Tank,” and that isn’t solely for the opportunity to solicit powerful investors; nearly as enticing is the chance to pitch your product to the millions of people 

 (and the millions more who will later stream the show or catch reruns on CNBC). Within 2 minutes of Noel appearing on screen, the viewing party at his and Cobble’s house had stopped watching the television—they’d become transfixed by their laptops, which showed sign-ups for DNAsimple “going berserk.” DNAsimple.org crashed before the most coveted Shark of the bunch, Mark Cuban, closed a deal to invest $200,000 in exchange for a 15% stake, valuing the grad students’ company at $1.3 million.

What caught the tech billionaire’s eye? “It’s the ultimate example of doing well by doing good,” Cuban told 

“By monetizing the sharing of one’s DNA in a manner similar to donating blood or plasma, they can help enable research projects that may otherwise never reach the critical mass needed. I love the opportunity and the potential societal impact.”

Even with the improved accessibility of genetic data, it can often take several years for researchers to collect a sufficiently large and diverse sample. Particularly when DNA donors are recruited at nearby hospitals, studies that seek to investigate rare diseases may never achieve the necessary sample to get off the ground.

DNAsimple’s donor registry has surpassed 130,000, up about 500% since Noel appeared on the show. Now the company’s goal is to hit 1 million sign-ups by the end of 2018, which would vastly expand its ability to partner with academic and for-profit researchers.

 at their office in downtown Philadelphia, Noel and Cobble described their optimism about the company’s future while hardly mentioning its financial upside. Their first objective remains their foremost: to dramatically facilitate genetics research, while generously compensating people for a surprisingly unobtrusive contribution.

“The infrastructure that you have to set up to recruit people into a study is a major undertaking,” explains Carolyn Hutter, PhD, acting director for the Division of Genome Sciences at the National Human Genome Research Institute. Although Hutter was previously unfamiliar with DNAsimple, she says, “If this model works, they really are providing a service that would enable researchers to, ultimately, save money.”

But for all the intuitive resonance of DNAsimple’s model, Noel’s success at securing an investment might have seemed improbable to TV viewers, considering what appeared to be 5 major strikes against him when he stepped into the Tank.

The Sharks usually insist that entrepreneurs be singularly focused on their startup. Noel, who’s now 29, created DNAsimple while also pursuing an MD-PhD full-time. “That was not advisable,” he jokes in hindsight. If he sounds overextended as a student-CEO, Noel’s backstory helped put the Sharks at ease.

He emigrated from Haiti after the devastating 2010 earthquake, joining family in New York City. There, Noel held up to 5 jobs at once in order to pay his way through Queens College, where he studied biomedical research and graduated with honors, earning a full scholarship to Penn State. His company’s launch was boosted by the prestigious Y Combinator fellowship in Silicon Valley, where Noel and 2 co-founders won a $12,000, equity-free grant. This year 

“30 Under 30” list for rising stars in science.

After Noel told his story on “Shark Tank,” the British billionaire Richard Branson, a guest investor that episode, interjected: “Can I make a non-political comment here? You are the exact reason that America should open up to immigrants.” And so the Sharks seemed satisfied with his commitment.

The Sharks rarely bet on companies that haven’t already generated significant revenue. To date, DNAsimple has delivered a full order of samples to about a dozen small-scale studies, so they’ve only paid donors a few hundred times. Researchers typically pay DNAsimple $155 per sample, and each kit costs $12.50 to manufacture. Compensation for a returned kit has increased over the company’s brief history from $30 to $40 to $50.

For a variety of reasons, but mainly to reduce overhead, DNAsimple doesn’t actually extract the DNA from saliva, nor does it store the genetic data. For potential donors who fear that their DNA results could later be used against them—say, to deny healthcare&mdash;DNAsimple protects participant anonymity at every step in the process. First, donors must consent to having their DNA analyzed for a particular study. Then, that study works with DNAsimple to confirm donors’ medical history. Donors return the saliva kit in a prepaid box, DNAsimple removes any trace of their names from the packages, and then delivers the samples to researchers. As a result, donors must provide a new sample each time they contribute to a study.

Early on, Noel’s team realized that it’s too risky to pay for a saliva contribution before an order is locked in, a move that could have grown its registry—and its debt. The emphasis has been on refining the process, especially by filling orders for studies of rare conditions, while building the database of people who’ve signed up to provide samples.

Many genetics studies evaluate thousands of samples—sometimes tens or hundreds of thousands&mdash;and therefore DNAsimple’s value to researchers is directly tied to the size of its donor registry (and its quality, since the diversity of participants can be as important as their sum). This creates a growth-stunting Catch-22: It takes a large registry to get money from studies, but it takes money from studies to grow the registry. Cuban must have realized that publicity from “Shark Tank” would provide a shot in the arm for sign-ups, and if the registry exceeds a certain threshold, partnership opportunities would increase exponentially.

The Sharks are reluctant to invest in products facing serious competition. DNAsimple is hardly the only service that collects saliva for DNA; around 2010, saliva replaced blood as scientists’ preferred source of DNA. One perk of using saliva is it’s clearly safe to mail—after all, as Noel points out, spit already seals envelopes and secures stamps.

Many genetics studies can get samples from biobanks. In 2016, the National Institutes for Health 

 the Mayo Clinic $142 million over 5 years in order to build a first-class biobank, part of President Obama’s Precision Medicine Initiative. Like DNAsimple, this Minnesota biorepository aims to enroll more than 1 million participants.

Sarah Nelson, a research scientist and PhD candidate in public health genetics at the University of Washington, described a common shortcoming of biobanks. “There’s always a concern with batch effects,” she says, “where you have genetic data cobbled together but without the medical information needed to answer your research question.”

Another alternative to recruiting donors from hospitals is partnering with companies that sell direct-to-consumer gene-testing services, such as 23andMe. In 2015, Pfizer announced 

 to conduct research using 23andMe’s database. A year later, the pharmaceutical giant had used 450,000 samples from that database for a study evaluating genetic indicators of depression. A week before the Pfizer deal, 23andMe announced a $60 million agreement with Genentech. 

reported that in 2014, 23andMe formed 14 such partnerships with researchers.

One advantage for a business like 23andMe is that it makes money on both ends: Customers pay to have their DNA analyzed and then aren’t compensated when the company sells their data for research. The 23andMe database now exceeds 3 million people, 

is told, and more than 80% of those customers consent to having their data aggregated for analysis by researchers.

Noel and Cobble point to several areas where DNAsimple claims a competitive edge: Saliva samples are purchased for each specific study, and exact demographics can be recruited with targeted advertising. The registry includes people from all over the US and Canada, with whom researchers can communicate directly (though still anonymously), requesting precise biographical and medical info as needed.

Although biobanks can supply samples on demand, they usually store a massive number of them, using maybe 2% of the collection at a given time. DNAsimple avoids that storage expense entirely.

The Sharks are wary of working with entrepreneurs who lack a money-first mindset. Early on, DNAsimple was approached by an academic who had spent a year looking for samples from just 1 or 2 people with a very rare disease. Noel and Cobble were thrilled to be able to supply that researcher with 6 samples. Did they hike the fee for each kit delivered, considering the obscurity of those samples compared to supplying saliva from 6 people with more conventional backgrounds?

They didn’t, and wouldn’t, for now. Noel and Cobble are no less proud of that deal. “Maybe a traditional business person would not appreciate that, but we get excited as researchers first,” Noel says.

The Sharks hardly ever give in to negotiations. Cuban, who’s notorious on the show for being ferociously uncompromising, offered Noel $200,000 for 20% of the company. When Noel asked if Cuban would go down to 15%, viewers probably expected Cuban to blanch.

Instead, he shrugged, said “OK,” and they had a deal.

The negotiation was taped this summer, and Noel and Cobble have found Cuban to be a surprisingly accessible partner. They’ve been awed, they explain, by his shrewd guidance on virtually any aspect of their business.

Although the company considers the diversity of its registry a major asset, Hutter, of the National Human Genome Research Institute, questioned whether the type of people who sell their saliva would represent a random sample. In any case, there are many genetics studies for which randomness isn’t necessary. DNAsimple is open to expanding overseas, where it would have to adapt to varying consent laws. As the registry grows, Noel says that basically any part of their operation could change.

The company’s office in Philadelphia happens to be next door to the Chemical Heritage Foundation Museum. It’s a fitting coincidence, given the sort of impact Noel’s team hopes to have. “Who’s ready to join me in saving mankind?” he said to close his pitch on “Shark Tank,” playing along with the show’s typical grandiosity. If nothing else, his company is trying to save scientists doing potentially life-saving work a considerable amount of time and money. It’s that simple.

How a genetic data company jumped from “Shark Tank” and into groundbreaking research projects.

More than 600 people were rushed to Las Vegas hospitals after the shooting rampage at a country music festival on October 1, some of them arriving in the backs of pickup trucks. “It was like we were in a war zone,” a trauma surgeon at University Medical Center of Southern Nevada told the 

. The injured were shot, trampled, or hurt jumping over fences at the late-night event, attended by many people from out of state. These patients, of course, were hardly in a position to supply the health records that can be vital for emergency treatment. 

Interoperability is perhaps most essential, and in some respects most inadequate, in emergency departments, especially under the pressure of large-scale disasters. The shooting in Las Vegas, the fires in Northern California, and the hurricanes in Houston, Puerto Rico, and throughout the Southeast underscore how crucial it is for emergency personnel to be able to access electronic health records and identify relevant information—namely medication history and chronic conditions&mdash;immediately. In the 10 days after Hurricane Harvey hit, Dave Fuhrmann, vice president of interoperability at Epic Systems, observed a 75% increase in the exchange of records between Houston and Dallas, to which many victims had been evacuated. 

Hospitals nationwide should take note: These events highlight the necessity for preparing information-sharing systems for disasters before they strike. They also contextualize the appeal of an emergency record-sharing tool that was employed for the first time this fall in California.

Evacuations across state lines and transfers to new providers pose additional challenges during these sorts of crises. Jitin Asnaani, executive director of CommonWell Health Alliance, a national group that offers cross-vendor interoperability service, described how the accessibility of health records “can make all the difference in saving a life or providing the appropriate care in a timely manner. 

“When people are displaced or away from their home,” he explained in an e-mail, “there is less of a chance their records will be available to the new providers they are seeing due to lack of nationwide interoperability between providers and health IT systems.” 

Electronic record sharing can also help families identify missing persons amid the pandemonium of a tragedy. That’s possible only if a hospital can identify a patient, however; one story out of Las Vegas described a 32-year-old from Los Angeles named Michelle Vo, who befriended a man at the concert hours before being shot in the chest and rushed in the bed of someone’s truck to a hospital, her phone and ID left behind in a purse. The recent acquaintance and Vo’s relatives spent hours calling every hospital they could find in the city trying to locate Vo, as the 

 reported, finally confirming that she had been receiving treatment at Sunrise Hospital, where she later died. 

Asked about Sunrise’s ability to access records for the 214 patients treated on that nightmarish night, Fran Jacques, the hospital’s vice president of marketing, said the urgency to provide lifesaving care was too great—and the scene too chaotic&mdash;to have worried about pulling up medical histories. 

About a week after the shooting, more than a dozen wildfires broke out across Northern California’s wine country and caused unprecedented devastation to the state. Two hospitals were evacuated, residents of nursing homes were relocated, and some patients affected by the fires were sent 50-plus miles to hospitals in Davis and Sacramento. 

Kaiser Permanente, an Epic member, safely evacuated 130 patients from its Santa Rosa Medical Center, while its other hospitals in the region remained operational. In a statement to 

, the Permanente Federation’s chief information officer, Patricia Conolly, MD, praised a feature Epic integrated last year called Happy Together, which consolidates a patient’s history at Epic and, in certain cases, non-Epic providers onto a single interface. That efficiency is especially useful for emergency departments receiving someone for the first time. Last month, Epic planned to release Share Everywhere, which will allow patients to log on to a portal and instantly share their health records with doctors. 

Back in 2005, floods from Hurricane Katrina destroyed hundreds of thousands of medical records that were stored in basement filing cabinets. Lee Stevens, with the Office of the National Coordinator for Health Information Technology, says the transition to electronic record keeping prevented any such destruction from natural disasters over the past few months. 

Use of electronic records in emergency departments increased 84% from 2006-2011, according to the CDC. But by 2015, 72% of respondents to a Black Book survey said they were dissatisfied with the interoperability of their department’s system. “Eighty-nine percent of emergency department leaders believe their hospitals rushed to purchase new electronic health records and ED systems between 2010 and 2013 for meaningful use dollars”—which were part of the Health Information Technology for Economic and Clinical Health Act’s interoperability incentives&mdash;“just to see productivity fall, liability rise, and connectivity stall,” Black Book reported. 

During a crisis, the secretary of health and human services can suspend certain Health Insurance Portability and Accountability Act regulations on electronic record sharing, and 72-hour waivers were granted to states in the Southeast after the hurricanes and to California during the fires. Stevens has not heard of any dramatic problems with record sharing for victims of the fires or hurricanes, partly because of where those crises occurred. “California is the most progressive state I’ve worked with on preparing for disasters,” he said. “Florida is also incredible. They don’t miss a beat when it comes to interoperability for storms.” 

Still, a variety of difficulties remain, which is why Stevens is enthusiastic about the progress of a government-funded technology called Patient Unified Lookup System for Emergencies (PULSE). “What is missing is the integration of emergency medical services to the health information exchange infrastructure,” explains Richard Cothren, the executive director of the California Association of Health Information Exchanges (CAHIE). With $3 million in grants over 3 years, CAHIE and other developers created a portal for disaster workers, mainly at field hospitals, to view a summary of a patient’s medical history aggregated from different interoperable systems. PULSE would activate after an emergency is declared and provide immediate, mobile access to this potentially lifesaving information.  

This summer, California simulated how PULSE would perform after a massive earthquake in the northern part of the state, and participants in the drill were largely successful at retrieving available records using the type of limited information that would be available to identify patients. “That’s a huge success,” Stevens said. “This is a design that we hope to see on a national scale. It meets the exact needs of every single one of the events we’ve just had.” In fact, PULSE was employed for the first time at emergency camps for victims of the fires in Napa Valley, and the system was activated on a limited basis but without problem, according to Cothren. 

Participants in the PULSE drill said the system could better simplify the information that medical personnel have to sort through with little time to spare. Cothren echoed that advice for hospital administrators, who can do more to prepare their systems to distill patient histories down to “the minimum necessary data” for emergency departments under the duress of a disaster. 

“A lot of what is going on today with EHRs and [health information exchange] systems is targeted at interoperability for the purposes of care coordination,” Cothren explained. “What we’re learning is that the needs during a disaster are different than during normal conditions. What we’re finding is that people who were rushed from their homes may just need their medication.” 

“There’s a lot of interest in these things when there’s a disaster,” Stevens cautioned, “but 2 weeks after the disaster, people go back to their lives.” Improving interoperability during emergencies, in other words, will require more preemptive engagement.  

What can hospitals learn from recent crises that have filled emergency rooms across the country?