precisionFDA: Why It Matters

Screen Shot 2015-12-15 at 9.32.57 AMI hadn’t intended to write about precisionFDA going live – this post by Dr. Taha Kass-Hout and Elaine Johanson of the FDA provides a terrific summary, and this post by Angela Anderson of DNAnexus offers valuable additional context. However, I found myself today so excited by this project and what it represents that I can’t resist offering a few additional thoughts about what makes this initiative so special.

First, it addresses an important problem in the field: the analytic validity of NGS tests. The ready availability of relatively inexpensive sequencing has enabled us to contemplate diagnostic sequencing at a scale that would have been difficult to imagine even a decade ago. At the same time, the drive to apply sequencing in different clinical contexts raises a critically important question: do I trust this test? A key starting point for clinical interpretation of DNA data is to agree on the sequence itself. If your procedure and analysis reports that a particular sequence in a DNA sample is “GATCGATC” and my procedure and analysis of the same DNA says the sequence is “GATTGATC,” then we’ve got a problem. precisionFDA will allow users to compare approaches, figure out what’s working, and determine where refinements might be needed.

Second, precisionFDA represents a novel and forward-thinking approach to regulation. Rather than envisioning governmental regulators as the folks who will define and then impose a specific set of performance standards, precisionFDA instead sees the government as providing the platform that will enable the NGS community to evolve the standards on their own — organically and transparently.

Finally, the ability to design, refine, and deploy this platform in such a rapid and agile fashion reflects in part the value of well-conceptualized public-private partnerships, in this case between the FDA and DNAnexus. By intentionally leveraging the skills and capabilities of a company like ours, the FDA was able to implement and realize their exciting and ambitious vision.

The ultimate success of the precisionFDA platform will of course depend upon how well it serves the community it is intended to support. However, it’s hard to think of a more auspicious beginning, and my hope would be that success here will encourage more leaders to evaluate the potential of public/private partnerships to deploy platforms that leverage the power of a distributed innovation community to address important shared challenges.

Precision Medicine: Seeking Impact, Needing Champions

This weekend, I had the opportunity to present a short talk, entitled “Precision Medicine 2.0: Seeking Impact, Needing Champions” at the Stanford Medicine X conference. I thought I’d share a short digest of the talk; the slides are here.

(Disclosure/reminder – also highlighted during my presentation: I’m Chief Medical Officer of DNAnexus, a cloud genomics company intensively involved with precision medicine; while this talk reflects my views, and not necessarily the views of DNAnexus, my experiences at DNAnexus have obviously informed my perspective. My relationship with DNAnexus should be kept in mind by the reader.)

Genomics 1.0

To understand where we are now, it’s useful to consider the excitement, promise, and extravagant expectations around the conceptualization and execution of the original human genome project, what you might call Genomics 1.0. Early proponents like Sydney Brenner stated in 1982 that with a complete DNA sequence and a big enough computer, “you could compute the organism,” while sequencing pioneer Wally Gilbert suggested in 1991 that with the complete genome in hand, “we’ll know what it is to be human” (see here for reference).

In a memorable response to this apotheosis of reductionism, Harvard geneticist Richard Lewontin penned a memorable essay in the New York Review of Books in 1992, suggesting proponents might be fetishizing DNA. His piece helpfully included this OED definition of fetish: “An inanimate object worshipped by savages on account of its supposed inherent magical powers, or as being animated by a spirit.”

Our tendency to imbue emerging technologies with almost mystical powers isn’t limited to DNA, of course, and might be applied equally well to “big data,” “design,” and “mobile.”

[Denny Ausiello and I offered our reaction to the completion of the sequencing phase of the first human genome announced by Celera in April 2000, in an essay for the New York Times “Week in Review” here.]

Following the completion of the human genome project, many in medicine felt a palpable sense of disappointment and disillusionment – where were the cures that genomics had promised? “We fought cancer…and cancer won” asserted a thoughtful 2008 Sharon Begley Newsweek article from this period. “Advances elusive in drive to cure cancer,” Gina Kolata of the New York Times detailed in 2009, as part of a memorable series, “The Forty Years War.”

Genomics 2.0

Flash forward to the present, what might be called Genomics 2.0, defined by two important features: next generation sequencing (NGS), which has dramatically reduced the cost and increased the speed of sequencing, and cloud computing, which provides a way to manage the deluge of data that’s now emerging. Interestingly, in testimony before a Congressional committee last summer, genome pioneer (and Human Longevity Institute founder) Craig Venter said that the most important changes occurring in genomics technology involve computing, and added, “The cloud is the entire future of this field.”

The Precision Medicine Initiative, announced at the White House in January, captured much of the contemporary hopes and vision around the potential of the science, excitement not only around genomics but also around the ability to leverage digital health to provide greater visibility into patient-participants, and enable deeper connection with them.

The often breathless excitement around the promise of precision medicine has generated a pointed response from skeptics (see herehere, here, and here), who ask – quite reasonably and appropriately – whether the science has been oversold.

Needed: Tangible Impact

The only way to escape the expectation/disillusionment cycle is for genomics and precision medicine to demonstrate palpable utility, tangible impact. We’ll be there when sequencing is embraced not because it’s interesting or edgy, but because it’s so obviously useful that once you’ve seen the benefits you can’t imagine not availing yourself of it.

The obvious analogy is Uber, though the point is not to become the “Uber of X” (as helpful as that would be to your buzzword bingo card). Uber is immediately appealing. You don’t routinely reach for Uber to express your solidarity with the gig economy, or to project hipster cred. Rather, to use Uber once is to understand it’s utility, and recognize what an elegant solution it is for a problem you might not have even known you had. The challenge for precision medicine is to be as intuitively and immediately useful as Uber.

Early Signs Of Progress

While genomics may not quite be at this point yet, I think we’re seeing many early signs of progress. Examples include:

  • Reproductive Health – Arguably the most tangible impact of precision medicine to date has been in the area of reproductive health, particularly carrier screening and non-invasive prenatal testing; see my recent discussion here.
  • Oncology – I am moved by the progress represented by the story of Corey Wood, a young woman diagnosed with lung cancer, and discovered, via a Foundation Medicine test, to have a mutation for which a category of targeted therapies was available.
  • Mendelian Disease – Consider the inspiring story of Bill Elder Jr., a patient (and medical student) with cystic fibrosis caused by a relatively unusual mutation, for which a targeted therapy (Vertex’s ivacaftor) is available.
  • Infectious Disease – You may have seen the the New York Times feature on 14-year-old Joshua Osborn, whose life-threatening occult infection was identified by a genetic testing approach pioneered by UCSF’s Charles Chiu; Osborn was then treated for the detected pathogen, and he recovered completely. Chiu recently received funding from the California Precision Medicine Initiative to expand this approach (disclosure: DNAnexus is involved in the effort).
  • Drug Development – In a concept I’ve previously described as “happy genetics,” the identification of “human knockouts” such as Sharlayne Tracy contributed to the development of anti-PCSK9 drugs, and inspired broader efforts to integrate genotype and phenotype to drive both drug discovery and healthcare research. The canonical example here may be the Regeneron/Geisinger partnership, described here by the New York Times. (Disclosure: the DNAnexus platform is foundationally involved in this effort.)
  • Undiagnosed Genetic Disease – The story of Bertrand Might, as reported so eloquently by Seth Mnookin in the New Yorker last summer, highlights both the potential of DNA sequencing to identify causative mutations and end diagnostic odysseys, but also the lack of existing medical resources to identify and bring together rare disease patients. To find other families, Bertrand’s father Matt needed to intentionally deploy traditional social media, which he did extremely effectively.

Four Areas To Focus Efforts

Screen Shot 2015-09-28 at 7.54.45 PM

To accelerate progress, we can consider four broad categories:

  • More Push – Precision medicine requires significantly more data. This will be accelerated by the continued reduction in the cost of sequencing, and by the rise of engaged patients, as it seems unlikely most healthcare organizations will make it easy to access data within their control any time soon (see here, and summarizing a recent Medicine X discussion, here).
  • More Pull – The sequencing that’s done needs to be more meaningful, along the ACCE framework highlighted by Muin Khoury and others at the CDC (I’ve described this framework here and here). Specifically, we need improvements in analytic validity (I’m excited by the precisionFDA pilot platform [disclosure: DNAnexus was contracted by the FDA to develop this]); in clinical validity (a key focus of ClinGen/ClinVar, under the inspired leadership of Heidi Rehm (of Partners Healthcare and the Broad) and others; and, perhaps most importantly, clinical utility – the ability to tie the results of genetic tests to meaningful clinical outcomes.
  • Basic Research – It’s essential to recognize the importance of basic research – research conducted for its own sake, rather than in search of any particular downstream use. A great current example is CRISPR DNA editing technology, which was discovered in the context of basic research trying to understand how bacteria defend themselves from viruses. A vibrant NIH is critical to basic research – as is the ability collaborate better and increase the rate of knowledge turns, a point Josh Sommer made so eloquently at the first Sage Bionetworks Congress.
  • Translational Research – Science generally doesn’t find it’s way into application without deliberate effort. Turning a promising Nature paper into a robust therapy you can confidently administer to a patient is distinctly non-trivial. There are critical roles to be played by engaged patient-participants, front-line providers, and entrepreneurs, whom I see as vitally important drivers of medical translation (here, here).

Impassioned Champions

As we contemplate the future of genomics, and of precision medicine, it’s easy to get caught up in the technologies, in all the amazing gadgets and devices that we hear about every day. Yet, we must remember that while the promise of precision medicine may reflect advances in technology, realization of this potential will require impassioned champions.

The need for champions has been made by Flowers and Melmon, in the context of pharmaceutical drug discovery, and by the late Judah Folkman, who said medical progress is driven by “inquisitive physicians” and “inquisitive researchers.” In a sense, this is all just a restatement of Margaret Mead, “Never doubt that a small group of committed citizens can change the world; indeed, it’s the only thing that ever has.”

Precision medicine is especially fortunate to have already attracted so many inspiring champions – research champions such as CRISPR pioneers Emmanuelle Charpentier and Jennifer Doudna, patient champions such as the Might family, data champions such as Regina Holliday, clinical champions such as Alice Shaw of MGH, regulatory champions such as the FDA’s Taha Kass-Hout, collaboration champions such as Sage Bionetwork’s Stephen Friend and John Wilbanks, translational champions such as Susan Desmond-Hellmann, formerly at Genentech and UCSF, and now at the Gates Foundation.

And, hopefully, precision medicine will be fortunate to attract and inspire future champions, people like you.

Dr. Shaywitz is Chief Medical Officer of DNAnexus. He also holds an adjunct appointment as Visiting Scientist in the Department of Biomedical Informatics at Harvard Medical School, and is co-host, with Lisa Suennen, of “Tech Tonics,” a podcast “focused on the people and passion at the intersection of technology and health,” available on iTunes (here).

Using Genomics To Improve Patient Care: DNAnexus to Support Both Projects Selected by California Precision Medicine Initiative

dna-art2As Chief Medical Officer at DNAnexus, I’ve come to see my remit in very simple terms: I represent the voice of the patient-participant.

At least, that’s my aspiration. It’s been tremendously heartening to see the centrality of the patient-participant increasingly acknowledged (see here, here), especially in the context of the President’s Precision Medicine Initiative, for example (see here).

As a physician, there are few things more clarifying than the absolute commitment to do your best for the patient in front of you. Even as a resident, I was struck by how many seemingly intractable turf battles could be resolved simply by pausing to ask, “what’s in the best interest of this patient?”

When I set up the PASTEUR translational research training program with Denny Ausiello at Harvard about fifteen years ago, our focus was squarely on the patient; PASTEUR’s core philosophy – embedded in its motto – was: “Our patients as partners in discovery” (see here.)

Today — especially here in Silicon Valley – we are awash in data, and in folks who love, and have remarkable expertise collecting data, working with data, analyzing data.

The challenge, it has long seemed to me, is ensuring that we maintain our focus, and acknowledge data milestones (33 petaflops! One million genomes!) but reserve our greatest respect and joy for accomplishments that impact people – the conclusion of a diagnostic odyssey, the successful treatment of a patient – or even better, the prevention of disease in a high-risk participant.

Not only should technology be measured by its impact on people, but technology should also strive to leverage and bring out the best in the people hoping to use it – which in healthcare means researchers, providers, and participants.

During my training, I was especially moved by the words of pioneering pediatric surgeon Judah Folkman (see here), who preached the importance of inquisitive physicians and inquisitive researchers (and presumably, inquisitive participants) in driving medical progress.

This perspective perfectly captures how I view the DNAnexus platform – not as a magical solution (biology is still hard), but rather as a secure, cloud-based collaboration and analytics tool that can enable inquisitive clinicians, researchers (and increasingly, I hope, participants) to ask thoughtful questions of massive amounts of data, to collaborate easily, and to generate and share new insights that help patients and participants.

In this context, we feel especially honored to be supporting the two demonstration projects that were just selected by the California Initiative to Advance Precision Medicine (CIAPM), a public-private effort launched by Governor Edmund Brown, led by Dr. Atul Butte, and hosted by UCSF (see UCSF’s press release here). Both projects aim to yield quick results for patients.

The first project, led by Charles Chiu, an associate professor of laboratory medicine at UCSF, uses genetic sequencing to identify potentially treatable infections in acutely ill patients; Chiu’s approach was utilized, perhaps most famously, to save the life of Joshua Osborn, a 14-year old boy who had been placed in a medically-induced coma after falling critically ill and experiencing brain swelling for unknown reasons. Genetic testing, using a methodology developed by Chiu and colleagues, identified the causative pathogen, which led to the successful treatment of Joshua and his subsequent recovery.

The current project aims to expand the reach of this approach, initially to additional hospitals within the UC system, and eventually, I hope, to clinics across the nation (and globe). DNAnexus is committed to cloud-enabling the analytic pipeline, ultimately allowing samples and sequencing from a range of sites to be analyzed and securely shared using a common platform. Structurally, this approach of distributed sequencing and centralized analytics – which enables rapid scaling — is very similar to what DNAnexus has also enabled at Natera (see here).

Paul De Kruif’s classic Microbe Hunters (see here) played a critical role in spurring my earliest interest in the excitement and promise of medical research; consequently, it’s been especially gratifying to see the DNAnexus platform applied to accelerate the understanding of infectious disease. The platform has also supported the Ebola research of pioneering Broad Institute investigator Pardis Sabeti and her colleagues (work also supported by the Broad, Illumina and the CDC); see here for a summary, and here for the June 2015 paper in Cell.

The second project, led by David Hausler, a professor of biomedical engineering at UCSC and Ted Goldstein, a research associate at UCSC (and a former VP at Apple), seeks to identify precision treatments for pediatric cancer patients in California based on the genetic characteristics of their disease. This work builds on the UCSC Treehouse Childhood Cancer Project, which enables the combined analysis of cancer datasets.

DNAnexus is currently supporting oncology analysis pipelines at UCSF and Intermountain. I am particularly excited to contribute to the analysis and sharing of oncology data because the opportunity and the need seem especially great. It’s essential that the culture of medical research – including (some might say particularly in) oncology — evolve from one of hoarding data to one that recognizes the urgency – certainly from the patient’s perspective – of sharing data, and catalyzing knowledge turns (to borrow Andy Grove’s term), and accelerating the rate of progress.   We hope that our platform – and the analysis, collaboration, and insight it enables – can contribute impactfully to this process.

As excited as we are to be an industry partner of both teams selected by the CIAPM, we are, as always, focused on execution, and can’t wait to roll up our sleeves and get to work here.

We understand that for both demonstration projects, the ultimate measure of success is not the amount of DNA analyzed, or the amount of computation brought to bear. It’s also not the number of collaborators assembled nor the translational insights generated — though both are important intermediate markers.

Rather, our success metric is the difference we are able to make in the lives of patients and families. Our ambition is the relentless and unwavering pursuit of this goal. Our conscience is all those who continue to count on us and on science to fulfill our solemn commitment.

Dr. Shaywitz is Chief Medical Officer of DNAnexus. He also holds an adjunct appointment as Visiting Scientist in the Department of Biomedical Informatics at Harvard Medical School, and is co-host, with Lisa Suennen, of “Tech Tonics,” a podcast “focused on the people and passion at the intersection of technology and health,” available on iTunes (here).