Strategies for Accelerating Progress in Personalized Medicine

Last year around this time, I wrote a blog aimed toward skeptics of personalized medicine, citing the progress we have made in lung cancer as just one example of how well this approach works for treating diseases. As I have reflected upon the advances we have made in areas like cancer and diabetes and how personalized medicine now seems to be the cornerstone for drug discovery and development for many pharmaceutical companies, it has become clear to me that there are still a number of areas for continued investment and focus.

What can we do to continue to grow the field of personalized medicine, find new and exciting therapies, and maintain the momentum we have seen over the past few years?

I believe that developing a personalized medicine strategy is the answer, but this is far from simple.

It begins by first appreciating that one needs to study the diversity inherent in diseases. When a new medicine is introduced in human clinical trials, for example, it is important to describe and understand the diversity of human responses to the medicine and to determine ways to identify those who will benefit the most from the medicine.

While we would ideally like to have such a clear scientific understanding of a disease that we can predict in advance who the likely benefiting patients are (which can be done in some instances), in most cases the personalization of a medicine comes from the observations made in clinical studies of diverse populations.

In contrast to basic laboratory research, where experimental designs seek to minimize variation, the “noise” in clinical trials can actually provide us with important clues to help form the basis of a personalized medicine. Well-designed studies, therefore, should study both the variation of the response as well as the scientific basis for that variation. When done well this can uncover biologic marker that become diagnostic agents for patient selection.

In addition to designing more clinical trials with these two aspects included, I would propose that those of us working in the field of personalized medicine should continue to support efforts that:

• Fund basic research on the biologic basis of disease. Personalized medicine is based on fundamental scientific discoveries. Without these, there will be no improvement in our understanding of disease.
• Fund research in diagnostic testing. Companion diagnostics can be useful tools to help healthcare providers and their patients make more informed treatment decisions.
• Educate physicians and other healthcare providers about personalized medicine. Patients in specialized populations that may respond to a good therapy for their subgroup need to have access to the appropriate tests. This means making sure that those who can order the tests and use them as a part of standard care are aware that these tests and potential therapies exist.
• Reimburse for diagnostic testing. Personalized medicine drugs will not have any benefit if they are not used and a healthcare provider won’t know to use a potentially life-saving drug unless the correct diagnostic tests have been performed.

Together, we can maintain and even accelerate the progress we have already made in delivering the right drug to the right patient at the right dose, every time.

_____________________________________________________________________

Join Dr. Stephen Eck on November 6, 2013 at the 9th Annual Personalized Medicine Conference hosted by Partners HealthCare Center for Personalized Genetic Medicine, Harvard Medical School, and Harvard Business School. 

View Dr. Eck’s video interview “Personalized Medicine: How Progress Happens” online at the Personalized Medicine Coalition.

Randy Scott: Bringing Metcalfe’s Law to Genomic Medicine

Each year, the Personalized Medicine Coalition recognizes an individual whose contributions in science, business, and/or policy have helped advance the frontiers of personalized medicine. This year, the Leadership in Personalized Medicine Award was presented to Randy Scott, Ph.D., during the Harvard Personalized Medicine Conference on November 28, 2012 in Boston, Massachusetts. 

Receiving the 2012 Leadership in Personalized Medicine Award from the Personalized Medicine Coalition (PMC), Randy Scott reflected on his success as the founder and former CEO of Genomic Health, but also looked ahead to new opportunities with his latest venture, InVitae Corporation.

Scott received the award this week at the 8th Annual Personalized Medicine Conference at Harvard Medical School. “Randy has transformed our understanding of how medicine can be practiced by creating one of the most successful personalized medicine companies to date,” stated Brook Byers, a partner with Kleiner Perkins Caufield & Byers and a previous honoree. Past winners of the PMC Award include Janet Woodcock (FDA), Elizabeth Nabel (NIH), Ralph Synderman (Chancellor Emeritus, Duke University), and Leroy Hood (Institute for Systems Biology).

After a successful stint at Incyte, Scott founded Genomic Health in 2000 and led the firm for nine years, overseeing the development of the Oncotype Dx gene expression test for breast cancer. He modestly shared the credit with numerous colleagues. “My contribution was I probably did a good job of hiring a lot of people at Genomic Health who are way smarter than I was,” he said, naming in particular co-founder Joffre Baker, CMO Steven Shak, and CEO Kim Popovits.

As a graduate student in the early 1980s, Scott said he had been excited about biotech but worried he was too late. “All the exciting genes had been cloned! TPA, Factor VIII, human growth hormone, insulin,” he recalled thinking. Today, Scott said, “we’re on the precipice of incredible accelerating change in this field… Everything we’ve experienced to date pales in comparison to what we’re going to experience in the next 5-10 years.”

But he also shed some personal insight into the launch of his latest venture, InVitae Corporation. He said he is “unabashedly excited” about the future of personalized medicine. “Personalized medicine is really when disease happens to you—your friends or your family. Suddenly it’s no longer just an industry we’re working in but something so personal, so intense, and so emotional. We should never forget that.”

The Network Effect

Scott said reading Intel founder Andy Grove’s book Only the Paranoid Survive in the mid-90s, during his tenure at Incyte racing to identify human genes, was highly influential. In the book, Grove discussed the impact of Moore’s Law on the revolution in computing; Scott saw parallels with the biotech industry. “The way we were sequencing DNA [at the time] was so embarrassingly simple,” he said. Just as computing costs were plummeting, Scott reasoned it was inevitable that sequencing costs would also fall.

Perhaps more importantly was the concept of “the network effect.” Just as Metcalfe’s Law—the community value of a network is proportional to the square of the number of its users—drove change in the computing world, so too will it drive the future of biotechnology.

“Having a really cheap genome sequenced is really not very useful. We still see articles in The New York Times, ten years after the genome project, [saying] ‘so what?’ At some level, they’re horribly wrong, and at some level, they’re horribly right. We’ve not yet seen the network effect or the full implication of Moore’s Law.”

Scott said the community is still “1-2 years away from the inflection point” where the cost of sequencing reaches the point that will trigger “massive consumer demand.” The value of genome sequencing will be most strongly felt in the network effect. “How we connect that genomic information across millions and millions of individuals… Somebody can be sitting at a computer, link into the network, and find how a mutation and how it correlates with their patient and a patient somewhere else in the world.”

Scott said he was also a believer in what he called the “Law of Finite Genomes.” The human genome is like a complex finite puzzle with about 150,000 pieces (20,000 genes and 100,000 non-coding RNAs). “All common diseases are really rare diseases,” Scott said, with cancer a prime example. “Medicine goes from an infinite game to a finite game,” he said. By comparing lots of genomic information, we can begin to rule things out.

Patients, Patients, Patients 

Scott was inspired to launch Genomic Health when a close friend was diagnosed with colon cancer in 1999. For the first time, Scott was personally struck by the chasm between science/technology and medicine. “We’ve got to bridge the gap—bring the science into clinical practice,” he said.

The focus at Genomic Health, Scott said, was “patients, patients, patients.”

“I’m not sure we had a model other than this maniacal focus on patients that wouldn’t be denied,” he said. If we could really do the science right, the science would sell.” Genomic Health spent an enormous effort on clinical studies.

“Clinical data wins over physicians, and it is physicians that win over the payors,” Scott said. “The onus is on us as an industry to build the value proposition [for payors]… so physicians have to adopt those products. If physicians adopt, they will drive payers to cover.”

Scott left Genomic Health this year to launch InVitae, spurred by the impact of rare genetic diseases affecting members of his family.

In 2000, Scott’s nephew had a daughter with galactosemia. Fortunately, the disorder was diagnosed within 48 hours of birth, and her diet could be changed, otherwise there could have been “a dramatically different outcome.” In 2005, an adopted nephew collapsed on a tennis court and died from hypertrophic cardiomyopathy. Advanced screening could have saved his life, but nobody knew any family history of cardiac disease, he said.

Finally, one of his wife’s relatives had a young son who developed serious seizures at age 2 years. The infant is developmentally impaired and severely autistic. Earlier this year, Scott revealed that exome sequencing of the child and his parents revealed a single de novo point mutation as the putative cause of the disorder. This is unlikely to provide any tangible medical benefit, but “it gives a clue into potential causes of these disorders,” he said.

Ridiculous Goal 

Scott said his goal in launching InVitae was to bring the power of genetics into the real world of clinical practice. “We have a ridiculous goal,” he said. “We want to aggregate all of the world’s genetic tests into a single assay—for less than the cost of a single assay today!”

In other words, InVitae plans to collapse all Mendelian inherited traits into a single assay that can be performed “reproducibly, at high quality and at reasonable cost for the medical system. So instead of going into these diagnostic odysseys… every parent thinking about conceiving a child can know exactly what their carrier status is and what disease risks lie in their family.”

The initial assay will essentially be an elaborate gene panel, but Scott’s plan is eventually that this will lead into whole-genome sequencing (WGS). Scott believes that “within 10-20 years, everyone in any developed health care system will be able to be provided with a low-cost [WGS] analysis at birth… We’ll be talking about managing your genome over the course of your lifetime.”

As for the question of how to deal with the plethora of data, “that’s Metcalfe’s Law, the network effect,” said Scott. “Much of the data won’t be of value to the patient or physician ordering the test. But collectively, they will be massively valuable to the research community.”

We’re big fans of “Free the Data!” said Scott. The universe of clinical genetic data “won’t be a database held by one company or one academic institution, but you’ll see a massive movement over the course of the next decade to make data broadly available within the research community.” This will create a huge disruption in medicine, Scott predicted, a shift from phenotypically driven medicine to more of a genotype foundation as sequencing costs fall and the network builds.

“Everything will drive off the genotype and it will move very fast,” he said. “This is a given. To me, this is the investment thesis. This will be the place to be, the chance to help people suffering from rare diseases. At the end of the day, every disease is rare.”

InVitae is building a strong management team. The company recently merged with Locus Development, a start-up co-founded by Sean George and Michele Cargill, founding scientists at Navigenics. Steve Lincoln and Jill Hagenkord, both formerly with Complete Genomics, also joined the cause this year, as did Reece Hart, former manager of research computing and informatics at Genentech.

– This article was first published at the Bio·IT World website on November 30, 2012.

Sustaining Progress in Personalized Medicine

Last week, I had the opportunity to speak at the Harvard Personalized Medicine Conference in Boston, MA. No other conference on personalized medicine brings together the array of scientists, stakeholders, and experts that this event does. This year the conference drove home to me that the potential to improve patient care via personalized medicine is greater than ever – yet the scientific and clinical challenges remain daunting. It is more important than ever to sustain biomedical innovation, and to ensure that health policy is informed by the enormous opportunity, and complexity, of making continued progress in this field.

The event also underscored that biopharmaceutical research companies are deeply committed to advancing the science of personalized medicine and building it into their research and development strategies. It affirms findings of a report released by the Tufts Center for the Study of Drug Development in 2010 which found that 94% of biopharmaceutical companies surveyed are investing in personalized medicine and 100% are using biomarkers in the discovery stage to learn about compounds. This research has required large up-front investments in new research tools and training. But, as we have seen in the last year-and-a-half with FDA approval of new targeted therapies for lung cancer, melanoma, and cystic fibrosis, it is starting to bear fruit for patients.

I’m hopeful we’ll see more approvals in the months ahead. In the report from Tufts, companies reported that 12-50% of compounds being researched are personalized medicines and over the last five years, they have seen a roughly 75% increase in their investment in personalized medicines. The importance of personalized medicine was illustrated in the reauthorization of the Prescription Drug User Fee Act, which provides FDA with increased resources and staffing to advance the regulatory science in areas such as pharmacogenomics and biomarkers.

This progress, however, doesn’t happen in isolation. The Harvard Conference participants represented, and illustrated, the wide range of organizations and individuals from different sectors that make up the research ecosystem that drives progress in personalized medicine. As the science of personalized medicine advances, research partnerships and collaborations will be more important than ever. To sustain progress in personalized medicine, it is vitally important to ensure that policy and regulation do not erect barriers to these types of partnerships.

Biomedical innovations like personalized medicine will help address major unmet medical needs, and offer a solution to rising healthcare costs.  As we face continued pressure to contain healthcare costs, it is crucial to ensure that healthcare policy sustains the innovation ecosystem and incentivizes continued progress in personalized medicine.

Harvard Personalized Medicine Conference Showcases Progress in the Field

Great themes emerged at the Harvard Personalized Medicine Conference last week. This is the third year that I’ve attended and there’s a palpable sense of progress. Examples abound on how precise molecular profiling tools are now being used in clinical practice and the potential going forward is huge.

But this transformation is causing a number of new considerations to come into focus. Working in industry, we have to think about what do these opportunities mean for business? Where can we best add value; what parts of medicine are currently underserved? How can we bring industrial and operational strengths to the emerging new paradigm? And how is the regulatory environment likely to change…for therapies and for diagnostics?

The opportunities and the challenges are breathtaking. Contributors spoke of the opportunity for sequencing data integration into the Electronic Medical Record (EMR) and the subsequent effect on medical decision making. And, the emerging Personalized Medicine paradigm should provide opportunity for the sector to make the case of the value of innovation, both clinical and economic. But what will people make of a new generation of EMRs containing contextual individual information across a continuum of patient care…from predisposition, through screening, diagnosis to therapy selection and monitoring.

The audience-sourced data demonstrated how peoples’ acceptance of precision medicine has increased over the years in which the conference has been held. And although oncology still dominates, pharma representatives described case studies in cystic fibrosis and cardiology. Those same pharma contributors acknowledged the absolute necessity for powerful diagnostics to complement the delivery of personalized medicine.

On the first day of the conference, I moderated a session entitled: “Business Models for Genetic Information.” Thought leaders from Siemens, PerkinElmer and Oracle described their companies’ visions for the molecular and digital age.

We had interesting audience feedback. Over 50% of the respondents felt that DNA sequence will become a routine part of an individual’s medical record within the next 10 years. Also interesting was the fact that only about 10% felt that availability of science or technology was the biggest obstacle to the adoption of personalized medicine in the clinic.

We discussed how personalized medicine can become mainstream. This isn’t about enabling personalized medicine for the few, but using technology and scale to deliver to the many. Doing this right not only will benefit patients, but input at the conference from regulators, drug manufacturers, and payors, told us there are a lot of parties interested in the success of this new paradigm. 

Making sense of all this data is a real challenge – how to format; how to integrate? Some contributors spoke of a data-driven industry transformation as has happened in other industries – financial services, retail, and communications. The result has been efficiency and transparency and the appearance of new interested parties – such as those capable of integrating workflow and data. 

The need for workflow integration will continue to be important as well. There are many players out there making individual contributions – from platforms, to molecular markers, to infrastructure, ordering, fulfillment, billing, and reimbursement. But who should take on the task of integration – which is surely needed if the benefits and efficiencies of personalized medicine are to be realized?

Personalized Medicine Is Waiting on the Shelf

While there’s a tendency to paint the future of personalized medicine as sunny, there are a number of issues persistently clouding the horizon. Bob Langreth and I examine one of these concerns in a story you can read at Bloomberg entitled, “Life-Saving DNA Test Overlooked in Rise of Colon Cancer.” Provider education has long stood in the way of patient access to genetic tests for Lynch syndrome, an inherited cause of high cancer risk. While the testing has been available for more than 10 years, enjoys recommendations from clinical and public health groups, and has been shown to save lives in studies, many providers remain unaware of its availability and power. The test costs as little as $300 in families where the mutation has already been identified. People who test positive can take preventive action with frequent monitoring, perhaps having vulnerable tissue removed.  Yet in too many cases it remains on the shelf.

How are we going to get the benefit of full genomes when simple, straightforward tests like this go unused?

John Lauerman will be participating in the panel discussion, “Impact of Genome Sequencing and Health,” at Harvard’s 8th Annual Personalized Medicine Conference, November 28-29, 2012, in Boston, Massachusetts.

The 8th Annual Personalized Medicine Conference

On November 28th and 29th, Partners HealthCare, Harvard Medical School, and Harvard Business School will host the 8th Annual Personalized Medicine Conference in Boston. As has been customary since the conference began, the meeting is also the venue at which the Personalized Medicine Coalition presents its leadership award. This year, it will go to Randy Scott, Ph.D., a founder and former CEO of Genomic Health, Inc.

During the past seven years, we have seen many changes in personalized medicine. Much of the excitement about personalized medicine continues to be driven by the ever decreasing cost of sequencing DNA and RNA. Whole genome sequencing, estimated to cost a billion dollars or more at the beginning of the century has gone down to a few thousand dollars, and the so called $1,000 genome is not too far behind. All in all, the 20,000 or so genes of the human genome only add up to less than two percent of the genome. If the sequencing cost is directly proportional to the size of DNA that is being sequenced, the cost of sequencing all genes should one day be no more $100. We are not quite there yet, but the rapid decline in costs is fueling the evolution of personalized medicine.

There is also a rapid evolution in our understanding of the genes and genomes of many organisms, including humans. Understanding how a normal gene functions and how changes in it (mutations) alter the cellular function is increasing at a rapid pace. As a result, sequencing whole genomes, whole exomes (all of the coding portions of genes), and subsets of genes for medical purposes is becoming ever more common. The fruits of all of these exciting developments are evident in many specialties of medicine but especially in prenatal testing, diagnosis of newborn and pediatric disorders, and cancer.

This year’s conference will highlight the genomic advances in science and medicine, and how they are being applied to diagnose and treat many disorders. The rapid development of technology also raises many questions about the clinical utility of these advances; about the policy and regulatory issues around what, when, who, and how to regulate; and about how the costs are reimbursed.

Personalized medicine is not just an American issue. Many nations of the world are developing strategies to implement personalized medicine in their own countries, and a number of those efforts will be featured at the conference. Professor Richard Hamermesh of the Harvard Business School and Norman Selby of Perseus LLC developed a business case on companion diagnostics that promises to be thought provoking and, based on previous years’ experience, a highlight of the conference.

Please join us at the end of November and participate in the dialog about how personalized medicine is evolving and what its implications are for your health and that of all peoples of the world.

Register online at www.personalizedmedicineconference.org. Follow the conversation online at @HarvardPMConf and #PMConf.

Less May Be More

Personalized Medicine has come to be strongly associated with drug-diagnostic combinations (companion diagnostics).  While this is a very important aspect of personalized medicine, it is not the goal.  The goal of personalized medicine is to find the best possible care for each individual patient so as to maximize the likelihood of the individual achieving his/her personal life goals.  While this may seem obvious, this principle has become increasingly harder to identify in every day medical care.

A variety of different motivations has encouraged the healthcare enterprise to view illnesses, or even everyday complaints as the opportunity – if not obligation – to do something for the patient.  The better approach is to ask: What is the best thing one can do for a patient, including the possibility of doing nothing?

As we develop better drug-diagnostic combinations, this will become increasingly apparent.  Soon we will have the ability (through use of precise molecular diagnostics) to know with reasonable certainty whether there is little or no benefit for a particular individual to pursue a therapy which had historically been the standard of care.

But, are we truly ready to forgo a potential therapy even if such a decision is based on strong scientific data?  Or, will we continue to use a therapy on the off chance it might work?  Our desire to do something has to include the possibility that the best treatment might be no therapy, based on the risks to the patient, the likely benefits, and the life goals of the individual.  In the end, patients want their healthcare providers to help them make the best choice and personalized medicine can be a strong tool to do that.  This will provide the largest benefit for patients as they avoid the risks that come with all therapies when no benefit is realistically to be had.

Dr. Stephen Eck will be moderating a panel discussion on Wednesday, November 28, 2012 at the 8th Annual Personalized Medicine Conference in Boston, Massachusetts.  Join the discussion at #PMConf. 

Still Skeptical about Personalized Medicine?

Over a century ago Sir William Osler, M.D., stated: “Variability is the law of life and as no two faces are the same, so no two bodies are alike and no two individuals react alike and behave alike under the abnormal conditions which we know as disease.” Despite our deep and long-standing understanding of the heterogeneity of disease and the variations in response to treatment, we are slow to adopt the notion that despite its complexity, the heterogeneity of human illness is decipherable. Skepticisms that we can actually deliver on the promise of personalized medicine is understandable, since converting such variables as severity of illness, uncertain vulnerability to side effects, co-morbid conditions and our cultural environment, to name a few, to precise algorithms for care seem daunting.

So, why has it taken so long for medical science to unravel this heterogeneity and why should we be optimistic that personalized medicine will happen? In part, our current views stems from where we have previously focused our attention. “Bergkrankheit “(mountain sickness) was known in the 14th century as an affliction of metal ore miners in Europe. By the 20th century we knew this as lung cancer and attributed it to a variety of environmental exposures. In this century, we have further refined our description of this disease to specific aberrations in molecular pathways, which if not entirely causative, account for much of the disease biology.

By contrast, medicinal chemistry as a science started much later than clinical medicine and is now closing the gap between knowing precisely what causes an illness to precisely what to do about it to improve the outcome for individual patients. No doubt we have a long way to go, but the current pace of personalized medicines suggests that it is becoming everyday reality for many lung cancer patients.

Join Dr. Stephen Eck on November 28, 2012, when he moderates a panel discussion at the 8th Annual Personalized Medicine Conference hosted by Partners HealthCare Center for Personalized Genetic Medicine, Harvard Medical School, and Harvard Business School.