Moneyball Medicine

Though the p-value "determines everything we do in  drug development or medical research," says Dr. Ulo Palm , it may be one of the most misunderstood and misused quantities in experimental science—drug discovery included. At its core, the p-value shows the probability that an observed effect was due to random chance. In other words, if a drug seems to outperforms a placebo with an associated p-value of 0.05, there's only a 5 percent chance that the study was wrong and that the drug is, in fact, no better than the placebo. A p-value of 0.05 is the accepted threshold for validity in most scientific research, even though it's an arbitrary standard set nearly a century ago by statistician Sir Ronald Fisher. "People don't often realize that this p-value of 5 percent was pulled out of thin air," Dr. Palm says. "If Sir Ronald Fisher had had six fingers, we would all be using a p-value of 6 percent."The issue, Palm says, is that an arbitrary dividing line of 0.05 leads journal publishers (and paper authors themselve

Building on his March 2020 interview with Jake Glanville, the founding partner and CEO of South San Francisco-based computational antibody engineering startup Distributed Bio, Harry speaks with three company scientists in the trenches: JP Buerckert, director of computational immunology, and Shahrad Daraekia and Jack Wang, both senior scientists. Together they're working on projects such as engineering existing human antibodies to the SARS virus so that they'll also work against the novel coronavirus, SARS-CoV2.The company's special sauce lies in its computational algorithms for analyzing antibody gene sequences and generating billions of new candidate antibodies against different pathogens. "We have a very strong wet lab team that is generating data for us and then we have a very strong data team that is sorting through these data" to help scientists decide which antibody leads to move forward with, Buerckert explains.Please rate and review MoneyBall Medicine on Apple Podcasts! Here's how to do that from an i

If you've seen the recent Netflix docu-series "Pandemic," about efforts to check previous viral outbreaks, you've seen former Pfizer scientist Jacob Glanville in action. The inventor, entrepreneur, and Ph.D. immunologist capitalized on the advent of cloud computing to provide vaccine and drug developers with high-throughput genomic sequencing of antibodies in humans and other species. He calls it "using the ability to look deep into these maelstroms of antibodies to try to understand why vaccines fail to hit conserved epitopes [where antibodies attach to antigens] on influenza or HIV, or how to better produce an antibody medicine." Revenue from the service allowed the startup to grow without outside capital. Today the company is developing a universal flu vaccine for pigs and humans. Glanville says we'll know by April whether existing anti-malarial, anti-HIV or anti-Ebola antivirals work against the COVID-19 coronavirus. A vaccine will take far longer to develop, he says. Meanwhile, Distributed Bio is using i

Schrödinger makes software that models the physics of atomic-scale interactions to predict the chemical properties of candidate drug molecules, helping its customers speed up drug discovery. A decade ago, Farid tells Harry, the company faced the chicken-and-egg challenge of convincing customers that its computational platform works, so that they would scale up their commitment, so that they could gather evidence it was working. Close collaborations with customers like Nimbus Therapeutics helped it improve the software and surmount that challenge. "In order to really take it to the next level and make a difference, it was necessary to use the software as customers ourselves," Farid says. "You get real-time feedback, honest feedback. You can imagine how much we learned from that."Please rate and review MoneyBall Medicine on Apple Podcasts! Here's how to do that from an iPhone, iPad, or iPod touch:• Launch the “Podcasts” app on your device. If you can’t find this app, swipe all the way to the left on your home s

Rapid sequencing of viral genomes is giving physicians and epidemiologists new ways to identify, track, and potentially slow outbreaks of viral infections such as the novel Wuhan coronavirus. That means high-throughput genome sequencing—which had predominantly been a research tool—is taking its place as a front-line weapon in the fight to prevent pandemics, says Febbo, a medical oncologist. "Last year, 40 percent of our consumables in sequencing were for clinical testing, and we see the clinical testing increasing at a pace that's faster than research testing," he says.Whole-genome viral sequencing, as a supplement to more traditional PCR-based testing for RNA sequences, can not only reveal exactly which virus is afflicting a given patient, but can reveal where a virus originated and how it is evolving to evade vaccines or other interventions.  "The fact that the WHO heard of the first cases [of the Wuhan coronavirus] at the end of December, and the New England Journal published the full genome on January 24,

Doctors helping couples conceive through in-vitro fertilization typically must screen multiple fertilized embryos to select one embryo for implantation—but the process is fraught with risk and subjectivity. from In 2018 Gilboa and her colleagues Daniel Seidman and Eyal Schiff co-founded AIVF, an Israel-based startup developing decision support tools that use deep learning and computer vision to lower the risk by identifying the most promising embryos for intrauterine implantation.The company's technology takes the place of old-fashioned visual evaluation of embryos by humans, instead of capturing time-lapse video of embryos from the moment of conception to the fifth day after conception, at multiple focal planes. "It's an obscene amount of data," Gilboa says. "Instead of looking at the embryo once a day under the microscope, we have tons of images to annotate and look for the biological features that we know are correlated with success."Proprietary machine learning algorithms use the video data, together with

Tom Davenport knows analytics, big data, and AI—he teaches executive courses on the subject at Babson College, Harvard Business School, the Harvard School of Public Health, and the MIT Sloan School of Management, and is widely known for his books on analytics and AI in business, Competing on Analytics (2007), Only Humans Need Apply (2016), and The AI Advantage (2018).Davenport notes that a number of life science startups are attempting to use machine learning, big data, and AI to reinvent drug discovery (a subject thoroughly covered in previous episodes of MoneyBall Medicine). But in other areas, progress has barely begun. A few startups are trying to bring machine learning into the world of providers and payers, to offer insight-based recommendations about care gaps and treatment. And a few researchers are studying the use of deep learning for pattern recognition in radiology and pathology imaging. But substantive advances are years away.On the clinical side, Davenport says, "The biggest changes are in the i

Milind Kamkolkar joined Cellarity in January 2019 to help the company to prove that it is now possible to "encode a cell" digitally—to use big data, deep learning, and other methods to model many different interconnected networks of molecular interactions.  "The whole idea...is really only feasible now," he says. "What changed over the last number of years is the ability to compute at scale."The promise of Cellarity's computational models, Kamkolkar says, is that they look broadly at cell behavior, rather than taking a reductionist approach. "If you could see the forest and the trees, what does that look like?" he says. "Really taking into account all of these networks that exist not only at the molecular level, not only at the cellular level, but also at the tissue level, and being able to look at all of it at once. You could argue it sound quite preposterous, but I love the ambition."Kamkolkar joined Cellarity from Sanofi, where he was the industry's first enterprise chief data officer, driving the transfor

Dr. Alan Copperman is director of the Division of Reproductive Endocrinology and Infertility and Vice Chairman of the Department of Obstetrics, Gynecology, and Reproductive Science at the Mount Sinai Health System. He's also a clinical professor of Obstetrics, Gynecology, and Reproductive Science at the Icahn School of Medicine at Mount Sinai; medical director of Reproductive Medicine Associates of New York, one of the world's leading IVF centers; chief medical officer at Semaphore Genomics, a health intelligence company; and medical director at Progyny, a benefits management company.Copperman tells Harry that data first came into his practice in a major way at RMA, which needed to "learn about what the best way is to take care of patients to optimize their success rates. We fell back on that term that you use, 'MoneyBall Medicine,' because we want to have the best embryologists, the best egg-retrieving doctors, the best embryo-transferring doctors. We want to put a team on the field that optimizes the succes

Gini Desphande says she likes to think of "AI" as augmented intelligence rather than artificial intelligence: a system of human plus machine intelligence that can speed up drug development and cut R&D costs and failure rates in clinical trials. AI "really isn't at the point where it's automatable," she says. "We still need a lot of human intelligence to be coupled with this technology, to determine what are the questions you want to ask and to evaluate all the targets that come out, to say 'Do these make sense?'"NuMedii's specialty is analyzing bulk tissue to isolate gene sequences in single cells that can point to new drug targets and drug candidates for diseases such as idiopathic pulmonary fibrosis. "The AI component helps us look at new targets that are not obvious to the human eye," she says. "It enables us to find network-level connections between diseases of interest and targets that are relevant for that disease. We can look at which nodes are coming into play and which ones should be manipulated

Dr. Chris Boone, vice president and lead for global medical epidemiology and big data analysis at Pfizer, is a health futurist, social entrepreneurs, executive, professor, patient advocate, and self-proclaimed "data hippie." He says he long aimed to be CEO of a health system, but eventually embraced his "true self" as a student of informatics, business intelligence, and big data analytics. "I come into the world of pharma not as a conventional or traditional pharma guy but as someone who cut his teeth in the provider world," he says. "It's just something that came naturally to me. There was always an intellectual curiosity about how we can do things better, and how we could ultimately disrupt the way that we currently treat patients, and ultimately transform the system for the betterment of patients."In the pharma business, he believes that big data analytics can disrupt clinical research and development and ultimately the commercialization of therapies for patients. He's an advocate for the use of real-world

Harry talks with Kevin Tabb, MD, the CEO and president of Beth Israel Lahey Health, the product of Lahey Health's merger this spring with Beth Israel Deaconess Medical Center (BIDMC) and several other hospitals in the Boston region. How does Dr. Tabb manage change inside a growing organization that—by his own admission—has to build and implement new tools, processes and the actionable data it needs to evolve beyond the fee-for-service era. Dr. Tabb was CEO of BIDMC before the merger, and previously served as chief medical officer at Stanford Hospital & Clinics in Stanford, CA, as well as head of the clinical data service division at GE Healthcare IT. Raised in Berkeley, CA, he emigrated to Israel at the age of 18, served in the Israel Defense Forces, studied medicine at Hebrew University's Hadassah Medical School, and served as a resident in internal medicine at Hadassah Hospital. Tabb says the most significant challenge for healthcare leaders is "figuring out how to calibrate the pace of change,&quo

Harry talks this week with Salesforce's vice president of strategic research, Peter Coffee. The computer-industry veteran and former tech columist says that in the era of 1) outcomes-based payments for medical care, 2) an aging patient base, and 3) ubiquitous sensors and continuous data collection, there's a huge opportunity—and financial incentive—for healthcare providers to employ technology platforms that improve the client experience. Might Salesforce end up marketing such a platform? Coffee says it's logical for the company, best known for its cloud-based customer relationship management software, to think about offering hospitals or medical service providers a configurable, CRM-style system for managing patient intake, consultations, recurring exam schedules, transportation to clinics, and the like. Coffee says Traditional healthcare organizations didn't have the insights or incentives to think about improving long-term wellness or keeping their customers (patients) happy—just the opposite, in fact. &q

As one of the researchers involved in the 70-year-long Framingham Heart Study, Rhoda Au is in a unique position to investigate whether changes in speech patterns in middle-aged people could prefigure the onset of Alzheimer’s disease later in life, and whether early detection might give patients more time to take preventative measures. She’s been part of the Framingham study since 1990, and she’s applying voice analysis software to 9,000 digital audio recordings of neuropsychological exams of Framingham patients to see whether there were telltale biomarkers in the speech of patients who went on to develop dementia. Au is a professor of anatomy and neurobiology at Boston University, a professor of epidemiology at Boston University School of Public Health, a senior fellow at the Institute for Health Systems Innovation and Policy at BU’s Questrom School of Business, and the Framingham study’s director of neuropsychology. How to rate MoneyBall Medicine on iTunes with an iPhone, iPad, or iPod touch: • Launch the “P

Kathryn Teng, MD, is division chief of internal medicine and community medicine at MetroHealth, one of three major healthcare systems serving Cleveland and the rest of Cuyahoga County in Ohio. She believes that healthcare costs are out of control in part because too many patients go directly to specialists about issues that their primary care physician or nurses could and should handle. But figuring out how many primary care doctors a big healthcare system like MetroHealth needs, and where they should be placed, is a data, analytics, and management problem. When she arrived at MetroHealth in 2015, Teng set out to collect data points to help with decisions across what she calls the “four quadrants” of population health: access to care, patient experience, provider and caregiver experience, and lower costs. “The real joy in this job,” Teng says, “is really around…trying to achieve the vision of population health, which is how do we provide the right care for the right patients by the right team members, and in

Many of the processes carried out in traditional chemistry labs searching for new drugs or drug targets can be sped up through factory-style automation—and in fact, “combinatorial chemistry” was a big boost for the field. But Alán Aspuru-Guzik, a theoretical chemist in the departments of chemistry and computer science at the University of Toronto, says “the transition to autonomy is what we really want.” Think of a “self-driving chemical lab” that uses big data, AI, and robotics to explore chemical space through a cycle of synthesis, characterization, and testing: that’s what happening both at Aspuru-Guzik’s Cambridge, MA-based startup Kebotix, in cooperation with commercial partners, and at his lab in Toronto, where he holds the Canada 150 Research Chair in Theoretical Chemistry. “We’re trying to put together the molecular Lego pieces, with a finite set of reactions and fragments,” Aspuru-Guzik says. “The art of being successful is not getting lost in an infinite forest of possibilities.” How to rate MoneyBa

Dr. Jennifer Carter says it was watching friends and family members stricken with cancer struggle navigate the complexities of the healthcare system in the early 2000s that inspired her to start a company in the area of precision medicine. At that time, the development of targeted therapies for cancers with specific genetic markers was already offering new hope to patients, but it was also creating new challenges for doctors and patients, who had to digest, manage, and interpret unprecedented amounts of data. The vision of her company N-of-One, she says, was around "how do you create something that could cut across all the different stakeholders and create the knowledge necessary that connected physicians and patients with cutting edge diagnosic and treatment strategies in a way that made it understandable and accessible." That ended up being "a very good strategy for physicians, patients, and the company," Carter says—an observation confirmed by QIAGEN's acquisition of N-of-One in January

Adjusting to a more collaborative style may take doctors some time, says Dr. Mark Boguski, but if they stop confining themselves to disciplinary boundaries, they'll be able to see connections between different areas of medicine that aren't taught in medical schools. Boguski draws on examples from oncology, where he says doctors are gradually being retrained to think in terms of disease pathways instead of discreet organ systems. Dr. Boguski is the chief medical officer of Liberty Biosecurity and founder of the Precision Medicine Network. He's a member of the U.S. National Academy of Medicine and a fellow of the College of American Pathologists and the American College of Medical Informatics. He's served on the faculties of the U.S. National Institutes of Health, the Johns Hopkins University School of Medicine, and Harvard Medical School, and as an executive in the biotech and pharmaceutical industries. He is the former vice president and global head of genome and protein sciences at Novartis, and a graduate

Harry's guest Sandy Aronson argues that AI and apps are not the solution for better healthcare; more effective care workflows enabled by AI and apps are the solution. Aronson is the executive director of information technology at Partners HealthCare Personalized Medicine. His team develops the IT infrastructure needed to support genetic-based personalized medicine in both patient-based and laboratory settings. This episode is the second in a two-part series on getting AI, machine learning, and analytics working in the healthcare provider setting, recorded as part of the AI World conference produced by Cambridge Innovation Institute in Boston in December 2018. How to rate MoneyBall Medicine on iTunes with an iPhone, iPad, or iPod touch: Launch the "Podcasts" app on your device. If you can't find this app, swipe all the way to the left on your home screen until you're on the Search page. Tap the search field at th top and type in "Podcasts." Apple's Podcasts app should show up in the se

What if we could use machine learning to train software to read CT scans of patients with intracranial hemorrhaging? Time to diagnosis could be doubled, potentially saving lives. This week Harry discusses such questions with Dr. Aalpen Patel, a physician-engineer who chairs Geisinger's department of radiology and directs is 3D imaging and printing laboratory. This episode is the first in a two-part series on getting AI, machine learning, and analytics working in the healthcare provider setting, recorded as part of the AI World conference produced by Cambridge Innovation Institute in Boston in December 2018. You can read a full transcript of this episode and browse all of our other episodes at glorikian.com/podcast. How to rate MoneyBall Medicine on Apple Podcasts with an iPhone, iPad, or iPod touch: Launch the "Podcasts" app on your device. If you can't find this app, swipe all the way to the left on your home screen until you're on the Search page. Tap the search field at th top and type in &