Healthcare Analytics, Population Health Management, Healthcare Big Data

FEATURES

How Precision Medicine Will Shift from Research to Clinical Care

Precision medicine is an exciting area of innovation, but there's a long way to go before research meets quality clinical care.

Source: Thinkstock

President Obama’s widely hailed unveiling of the Precision Medicine Initiative in 2015 may have brought the term to the forefront of the nation’s consciousness, but “personalized medicine” in and of itself is far from a new concept.

In fact, it has always been the foundation of all medical activity, from the ancient days of charms, potions, and poultices to the astounding delicacy and accuracy of robotics used in modern micro-surgery.

Healthcare practitioners have always focused on discovering the most innovative and effective ways to solve pressing problems for the patient on the table, according to his or her individual needs, although the tools and techniques of the 21st century differ vastly from the curious methodologies of the first physicians.

Genomics are starting to replace guessing games for clinicians targeting cancers, chronic conditions, and neurodegenerative diseases.  Providers are reaching for big data as a primary diagnostic and decision-making tool.  Patients are increasingly being spared painful or unnecessary experiments caused by a lack of actionable information and knowledge of the latest literature at the point of care.

Personalized medicine is suddenly starting to live up to its name, spurred on by the digitalization of the healthcare system, the plummeting costs of genomic sequences, and a rapidly growing body of knowledge that redefines how we view the very nature of the origins and development of deadly diseases.  

“Precision medicine has always been moving fast, but over the last two years or so, it has really accelerated into the clinical space,” says Jonathan Sheldon, Global VP of Oracle Health Sciences. 

“Even just five or six years ago, it was more of a research experiment focused on trying to understand the molecular basis of disease.  But it has become much more about how you impact clinical care and clinical decisions.”

The acceleration of precision medicine holds a great deal of hope for patients with rare cancers, inherited conditions, and other diseases, but the shift from research to reality is not without its challenges.  From basic data interoperability roadblocks to privacy and security concerns to the inability to recruit the right patients into clinical trials, the nation’s nascent focus on the clinical side of precision medicine is fraught with difficulties.

“When the genome was first sequenced, there was a great hope that it was going to give us laser-like precision to understand a wide array of problems,” said David Delaney, MD, Chief Medical Officer of SAP.  “The thought was that we would be able to sort people into buckets based on the absence or presence of a gene or mutation.  But the challenge, of course, is so much more complex than that.”

“There are so many factors that will cause expression or non-expression of a gene.  And when you add in factors around a patient’s environment, it turns out that there are so many different pieces to this.  The hope right now is that once we get more of the puzzle pieces out there, we will be able to start making the links that will eventually lead to major breakthroughs in things like cancer.”

Redefining the cancer landscape

Cancer is, of course, a major focus for most organizations working towards integrating precision medicine into their treatment protocols.  With more than 1.6 million new cases estimated to arise in 2016, and close to 600,000 deaths predicted for American patients within the same time period, the burdens of cancer are both social and financial in nature.

Thanks to the tireless and often unsung efforts of investigators and patient advocates working for decades to advance the health system’s understanding of how to treat this common threat, as well as new initiatives like the Cancer Moonshot 2020, researchers are starting to develop a new perspective on how to effectively attack malignancies.

“We are starting to learn more about developing specific treatments for cancers based on the origin of the illness, rather than just diagnosing something based on a few tests and assuming that every patient arrives at the same end state for similar reasons,” said Delaney.

“Oftentimes when we use history as a guide, we’re incorrectly parsing diseases that have a similar symptom cluster but very different causes.  That is often the explanation for why a drug may work very well for one patient, not so well for another, and produce an adverse effect in a third.”

Researchers are starting to be able to identify the genetic roots of certain cancers, but the acknowledgement that “cancer” is not a single disease is central to being able to treat it effectively, added Marcin Imielinski, MD, PhD, a Core Member and Assistant Investigator at the New York Genome Center.

Members of the Cancer Moonshot 2020 leadership board announce the initiative on January 12, 2016
Members of the Cancer Moonshot 2020 leadership board announce the initiative on January 12, 2016

Source: Cancer Moonshot 2020 / Youtube

“Ultimately, cancer is not a single disease,” said Imielinski, who is also an Assistant Professor at Weill Cornell Medical Medical College.  “It’s a constellation of different diseases that you can subdivide based on organ type or tissue pathology, but you can also divide it on the basis of their genetic changes.”

“If we do that, we can start considering each of those almost as a separate disease, and we can find right the drug for that specific disease.”

But sequencing tumors one by one will not help researchers move quickly enough to generate the large-scale insights necessary to produce blockbuster breakthroughs for the millions of patients who may benefit from targeted approaches to their treatments, he added.

“We need to do larger panels or exomes or genomes to define those subgroups.  These are the options we’re pursuing.  But until that happens, we’re not going to generate the critical mass of data that can kick start this whole process.”

Funding and focus from the Precision Medicine Initiative

The Precision Medicine Initiative has the issue of scale firmly in its sights.  Its cornerstone project, the PMI Cohort, aims to gather genomic and background data from at least one million patients in order to give big data researchers something to chew on.

“The Precision Medicine Initiative Cohort Program will be a broad, powerful resource for researchers working on a variety of important health questions,” said Dr. Francis S. Collins, MD, PhD, Director of the NIH in late 2015 when announcing the roadmap for the program.

“Equally important, the Precision Medicine Initiative Cohort Program will change the way we do research. Participants will be partners in research, not subjects, and will have access to a wide range of study results. What we’re doing with the Precision Medicine Initiative cohort is intersecting in a synergistic way with other fundamental changes in medicine and research to empower Americans to live healthier lives.”

The PMI Cohort won’t just gather an unprecedented amount of data in one place, to be shared across the research community.  It will allow researchers to explore a more accurate portrait of patients who experience a variety of comorbidities, conditions, socioeconomic challenges, and lifestyles, something not always accessible from strictly controlled clinical trials.

“The PMI needs accurate representation of people in general, not just the gung-ho fitness buffs or quantified self enthusiasts who have been most willing to contribute their data,” Delaney said.  “On the other side, you get people with rare diseases or cancers who want to contribute to research because it might help their own situation.”

"What we’re doing with the Precision Medicine Initiative cohort is intersecting in a synergistic way with other fundamental changes in medicine and research to empower Americans to live healthier lives.”

“But what you don’t get is a lot of the people in the middle.  They might not have a pressing need to participate in a clinical trial, but their data is very valuable for big data analytics.  It’s hard to recruit them, because they’re the ones hearing this drumbeat of breaches and hacks, and they may have significant anxiety about what happens with their data.  That’s something that the PMI is going to have to address if they want to get this information.”

“There will need to be a lot of learning going on, especially around privacy, interoperability, and data liquidity.  It’s going to force us to deal with a lot of these mechanics as we bring all of these issues together.”

The PMI is also bringing together resources and expertise from organizations across the care continuum, including federal agencies like the NIH, FDA and Department of Veterans Affairs, providers such as the Mayo Clinic, Intermountain, and Cedars-Sinai, as well as advocacy and professional groups including the American Medical Association, HIMSS, Get My Data, and the Patient-Empowered Precision Medicine Alliance.

The Mayo Clinic will take on the challenge of building infrastructure for the PMI Cohort
The Mayo Clinic will take on the challenge of building infrastructure for the PMI Cohort

Source: Wikipedia

“It’s a very positive development for the industry,” Delaney said.  “It’s a good starting place. I really like the collaborative attitude that the Initiative is fostering across the public and private sectors to leverage the strength of each segment of the industry.”

However, collaboration has never been the easiest task for the healthcare system.  Stymied by technical obstacles including a lack of health data interoperability, and wary of anything that might sap patients and profits from organizations operating within tight margins, learning to share information and resources is a tough ask for many participants.

Despite believing that the PMI is “essential” for moving forward with personalized care, Imielinski believes that finances will quickly become of the biggest concerns.

“The big roadblock to the community is the fact that genetic sequencing is not routinely incorporated yet in clinical care for most cancers,” he explained.  “And in cases where it is, academic medical centers or patients are generally paying out of pocket for it.”

The cost of genomic sequencing has dropped to about the same price as a CAT scan or an MRI, he pointed out.  These tests are currently reimbursed at standard rates whether or not they contribute positively to patient care, yet there is no standardized methodology for reimbursing genomic testing yet.

In order for genomics to become a routine part of clinical care, that needs to change.

“There are a lot of academic medical centers that are already performing these tests, and there are commercial labs that can interpret a pretty small set of targets that have already been associated with a clinical feature or targeted response,” Imielinski said. 

The PMI is likely to bring more awareness to this issue, and may even encourage payers to see the value in relatively inexpensive sequencing for patients who might benefit from a targeted approach to their condition.

“For those tests to be reimbursed, I think it’s just a matter of the insurance companies deciding that it’s worthwhile,” he said.  "But the PMI, on the whole, is really an example of how in the future, we should be trying to use all the data we have and match patients to trials based on their genomic features and other features as well.”

Turning the data we have into the data we need

The PMI Cohort will not just collect reports on genomic sequencing, but will also include patient data from electronic health records, as well as information on lifestyle choices, environmental exposure histories, and biological samples including blood, saliva, or urine.

Adding these data points to the mix will help researchers delve into the complicated origin of cancers, much of which is unknown.

“There’s a huge amount of data around the patient that needs to be collected that isn’t actually genomic, but is absolutely critical in interpreting the genomics,” Sheldon explained.  “The genomics are largely meaningless without that critical context, so it’s great that the PMI is looking at both sides of the coin, as it were, from its earliest stages.”

 "No organization alone can tackle this.  No one has the size of data set for it.  Sharing big science is the future of the field.”

For providers and other research organizations to join in, however, increased data sharing will become essential, he added.

“If you want to be successful, you have to share data. Everyone knows you need to do it, but the willingness to share genomic data is somewhat more sporadic than it needs to be.  No organization alone can tackle this.  No one has the size of data set for it.  Sharing big science is the future of the field.”

Few organizations have the capability to integrate genomic data into the clinical workflow, either, Sheldon continued.  Electronic health records are simply not designed to serve up the results of genomic testing to providers in the consult room – especially because no one is quite sure yet which parts of those results have real clinical relevance.

“A lot of those variants that are present in a whole genome have no clinical importance – at least as far as we can tell today,” said Sheldon.  “That might change very quickly, but right now it would simply be an overload of information that is not useful.  You wouldn’t want to pile all of that into the EHR.  It wouldn’t make sense.”

“So first we have to define what that clinically actionable dataset is, and then we can think about how to bring it into the EHR.  That needs to be more than dropping a PDF in there, which is sort of the way things are done at the moment.  It needs to be integrated in a way that can fire off various rules and actions within the EHR so that the technology can deliver real decision support.”

Bridging the gap between theory, planning, and practice

The official Precision Medicine Initiative is still in its planning phases, as the NIH and FDA start choosing their development partners and chart out the best strategies for fostering a cooperative research environment.  It is intended to be a long-term project that may reveal surprising and unanticipated results somewhere down the line.

In the meantime, stakeholders from across the industry, including big ticket providers like Kaiser Permanente and Geisinger Health System, research instutitions like the New York Genome Center and the Neuroblastoma and Medulloblastoma Translational Research Consortium and technology vendors such as IBM, Oracle, Dell, and SAP, are also working on their own precision medicine competencies.

"We have a combination of just doing what we’ve learned to do most of the time, and then not doing what we know we should do for the rest of the time.” 

These parallel efforts, many of which are likely to intertwine with the PMI over the next few years, are showing that the healthcare system may not always need a big federal push to convene, collaborate, and succeed.

That being said, the standardization, best practices and leadership that is likely to result from the guidance of the NIH and its colleagues could have even broader benefits for the use of big data analytics and personalized medicine in everyday care.

“On average, 80 percent of what physicians do is based on how they learned to practice medicine from their particular mentors and curriculums,” Delaney said.  “It’s not really driven by great data.  There has been this apprenticeship style of learning passed down between generations, and conventional wisdom says that this strategy or that treatment is likely to work.”

“Only about 20 percent of what physicians do is based on data right now.  And if you look at that 20 percent, studies have shown that providers adhere to evidence-based medicine guidelines only about half of the time. So we have a combination of just doing what we’ve learned to do most of the time, and then not doing what we know we should do for the rest of the time.” 

This “weird combination” of over-treating some patients and under-treating others is perpetuated by a lack of good data and meaningful feedback, he added.  The Precision Medicine Initiative, and the lessons learned from associated private projects, will help to change that.

“Precision medicine is very focused on taking that 20 percent of medicine, the evidence-based part, and getting it to be even more precise,” he said.  “But there’s so much work to be done on the other 80 percent of how we practice.  We need to do a better job of understanding, based on outcomes, what we should be doing to treat the majority of patients.”

Precision medicine may be bringing optimism and excitement to the cutting edge of clinical care, but Delaney reminds providers that even the most advanced new therapies must be built on a solid foundation of methodical implementation. 

“We’re still struggling with the basics of data standards, interoperability, workflows, and patient engagement,” he said.  “It’s important not to forget that backdrop to this conversation.”

Much work will still need to be done in terms of technology, process, and infrastructure, agreed Imielinski.

“A lot of academic medical centers are already equipped with the sequencing facilities that are up and running,” he said.  “If the reimbursement landscape changes, I think sequencing could be routine within the next two years, easily – at least at those top academic medical centers.  And that would increase the incentive for other organizations to follow suit.”

Finding that incentive should be too difficult, Sheldon added, if payers and providers start to view precision medicine as a complement to the other major trend in healthcare: population health management.

“They kind of seem like they’re at odds, but in fact they’re highly synergistic,” Sheldon asserted.   “Precision medicine gives us the molecular tools to dissect diseases with enough granularity that you actually can manage the risk of a very detailed and focused patient cohort.”

“In the next few years, we’re going to start seeing those two trends work more closely together, and it will produce benefits for both.”

Population health management may be where precision medicine brings the broadest rewards: not necessarily in the dissection of the origins of an extremely rare tumor, but in the fight against diabetes, Alzheimer’s, and cardiovascular disease. 

Both are equally valuable to the individual patient, of course, and there is a great hope among stakeholders that both will eventually become problems of the past as the industry’s knowledge and capabilities advance.

It will require a team effort, however, as White House Chief Data Scientist DJ Patil said in a blog post last year, and success will not come without a willingness to commit to solving the technical, financial, and process issues that could impede innovation.

“We need all sectors to work together,” wrote Patil and Stephanie Devaney, Project Manager of the Precision Medicine Initiative last August.

“We need people to actively engage in research and voluntarily choose to share their data with responsible researchers who are working to understand health and disease. We need healthcare providers to share their insight and help translate new findings into better care. And we need a strong, secure, and nimble infrastructure for health data that protects privacy, ensures security and facilitates new research models.”

“With support from patients, research participants, researchers, providers, and private sector innovators, we can make precision medicine a reality.  We need your creativity, on-the-ground experience, and enthusiasm to realize the promise of delivering individually tailored treatments to patients.”

X

Join 25,000 of your peers

Register for free to get access to all our articles, webcasts, white papers and exclusive interviews.

Our privacy policy

no, thanks