The Patient as Collaborator: How Personalized Medicine is Giving Back to the Patient

While the advancement of personalized medicine hinges on collaboration among all stakeholders from researchers to industry to clinicians to policymakers, the ultimate stakeholders in personalized medicine efforts are the patients themselves.  In Total Cancer Care™, patients are not only the ultimate beneficiary, but also the major contributor to the effort.  More than 90% of patients who are invited to participate in the Total Cancer Care™ Protocol accept this offer.  This high participation rate is primarily a reflection of the intrinsic altruistic nature of patients and their desire to contribute to the solution through research.    We formed a Patient Advocacy and Ethics Council to assist us in developing and implementing the Total Cancer Care™ Protocol.  We asked this group, “What can we give back to the patient, not just those patients who may develop recurrent disease, but also those who may be cured by initial therapy?” (Approximately 55% are long term survivors.)  Without hesitation, the council told us that all patients desire to have access to their own information in a usable and understandable format.

To that end, we developed a Patient Portal to the data warehouse which provides patients with their own medical histories, data and other important information.  Under the leadership of Mark Hulse (formerly of Partners Healthcare) and Dr. David Fenstermacher, we began this effort at Moffitt in October 2009, and we are gradually extending the portal access to all patients.   Our goal is to extend this service and resource not only for patients at Moffitt but for all patients at all consortium sites.  Much work needs to be done in this area and requires a much improved “real-time” information system.  We also are developing the system to not only be a repository of patient’s personal health records, but a portal where the patients can use the information to make informed decisions.  Again, working with the Institute of Human and  Machine  Cognition (IHMC) we are developing virtual learning technology and applying a process called C-map tools, originally developed at IHMC, to assist patients and physicians to navigate the Internet resources to ultimately meet patient’s needs.

In summary, these are very exciting times.  I truly believe we are at the threshold of translating and just as importantly, DELIVERING on the promise of personalized medicine.  We hope that our effort in developing Total Cancer Care™ will be a major part of the foundation of what some day will be considered common place—a healthcare system and technologies that are organized in such a way that every patient’s needs are identified and inform an individualized approach to meet their needs.  Primary stakeholders in developing personalized medicine, including researchers, clinicians, industry, policymakers, and patients themselves must come together to organize the framework and environment to promote personalized medicine.  It is unrealistic to expect any one stakeholder to collect the resources needed to create a rapid learning information system that will be required to capture data, leverage and enhance informatics needed for analysis, and communicate new knowledge to all the stakeholders involved in developing a better healthcare system built on the foundation of personalized medicine.  Teams comprised of broad expertise across the healthcare and research spectrums, and an unprecedented effort by all will be required to exploit the advantages of the necessary team science approach.  To complement the scientific infrastructure and technology that has already been developed, additional resources will be required including expanding information systems to community hospitals and physician practices, such as electronic medical records, biomedical informatics applications, and information technology professionals.  Ultimately, by developing evidence-based healthcare systems, we will improve the quality of healthcare by identifying best options for patients based on their personal traits and characteristics; such is the promise of personalized medicine.

The Role of Comparative Effectiveness Research in Total Cancer Care™

In my previous entry, I discussed how the launch of the Total Cancer Care™ initiative at Moffitt Cancer Center nearly eight years ago led to the development of one of the largest prospective observational studies in the world.  Through the enrollment of more than 60,000 patients and collection and genetic profiling of tens of thousands of tumors, Total Cancer Care™ collaborators have generated a vast information system to be leveraged as a clinical decision tool, and as a means of quality performance and comparative effectiveness research (CER).

One of the stated aims of Total Cancer Care™ is to raise the standard of care for all patients by integrating new technologies in an evidence-based approach to maximize benefits and reduce costs.  Although we developed this aim over seven years ago, I believe it is completely consistent with the current definition of comparative effectiveness being used by AHRQ and other policymakers. 

As I mentioned in my previous entry, strategic partnerships are an essential component to achieving the goals of Total Cancer Care™, and this is clearly demonstrated in our efforts in CER. Dr. David Fenstermacher and colleagues from Moffitt as well as the Institute of Human and  Machine  Cognition (IHMC), in Pensacola, Fla., are collaborating on a major NIH/NCI grant to enhance the Total Cancer Care™ infrastructure to support CER by expanding data management resources, integrating  automated data extraction methodologies (including natural language processing technology a particular area of expertise for IHMC), and creating user interfaces to data for researchers, clinicians and even patients. 

A major focus of our current efforts in CER is to determine the information and technology gaps in the CER infrastructure for data capture and data sharing.  Ultimately, it will be important to involve the community at large who are enrolling patients in the Total Cancer Care™ Protocol so that they can use the Total Cancer Care™ data warehouse as a decision tool based on evidence generated by the study itself.  The importance of the community network cannot be over emphasized both for populating the Total Cancer Care™ biorepository and database, and the ultimate utilization of the information and evidence generated for delivering the right treatment for the right patient.

To enhance Moffit’ts ability to establish this large research initiative the cancer center formed a wholly owned for-profit company, M2Gen, in 2006.  Merck and Co., Inc., through a Merck affiliate, signed on as our ”Founding Collaborator”. This experience has taught us how to service a global healthcare client and produce measurable scientific insights to accelerate drug candidates through translational medicine advances.

“Partnering for Cures” to Advance Personalized Medicine

I think we all would agree that finding cures and improved treatment options for cancer are a moral imperative. They will have a dramatic impact not only for those fighting the disease, but also for the families, friends, healthcare providers, and other caretakers that support them in their battle.   Personalized molecular medicine provides a promising path forward in cancer care, but accelerating this research requires the brightest minds, great laboratories, cross-disciplinary collaboration, rich software tools and LOTS of relevant, annotated, real-time data. Today, we are missing the “LOTS of data” piece, because our health information technology (HIT) and consent systems are not effectively connected for either the improvement of care or the acceleration of research.

Estimates in the U.S. indicate that more than 1.5 million will be diagnosed with – and more than a half million people will die of – cancer in 2010. And, as of 2007, 11.7 million Americans were living with the disease. Of those 11.7 million cancer survivors, it’s estimated that only 5% are enrolled in clinical trials, and only 15% are being treated at major research centers – which means more than 9 million people with cancer are not part of formalized research.  This is a highly motivated community, many of whom would welcome the chance to participate in research that could help their children, or their children’s children, receive more effective treatments if they suffer from the disease.

In partnership with the National Cancer Institute (NCI) and SAIC, we have built a prototype to demonstrate that we can solve this problem now. Earlier this week at the Partnering for Cures conference in NYC, Ken Buetow, Ph.D., Director at the Center for Biomedical Informatics and Information Technology at the NCI and Dr. Jon Handler, from Microsoft’s Health Solutions Group, presented a jointly developed prototype that showcases the potential for information technology to accelerate personalized healthcare research and improve clinical care.  Dr. Buetow talks here about the information challenges faced by researchers, providers and patients, and looks at the potential for technology to drive meaningful transformation in support of these stakeholders’ needs.

The prototype uses Microsoft HealthVault and the Patient Outcomes Data Service (PODS) created by the NCI to collect provider and patient-generated data on cancer diagnoses, treatments and outcomes. Since PODS and HealthVault are easily accessible outside research centers, the prototype highlights ways to engage a broader set of clinicians and patients in research – making it easier to reach those 9 million people who are not currently represented in research studies. In addition, gathering regular reports from patients on their experience with cancer treatments – for example, tracking daily pain levels, sleep patterns and mood – can provide researchers and clinicians with a richer set of data for understanding the impact of cancer treatments, particularly among certain patient sub-types and populations.

Using Microsoft Amalga, this data can be made anonymous and aggregated with data available in other research databases to create a disease registry that enables more complete analyses of the efficacy of cancer treatments.  Providers and patients have the opportunity to not only contribute their own information to benefit others; they can also view trended data from across similar patient populations, enabling shared decision-making around diagnoses and treatment plans.  

While the prototype we built focused on cancer research, there is potential to use HIT to further the personalization of treatments for other diseases  – Parkinson’s, Multiple Sclerosis, and Alzheimer’s – and begin to see how we can use the power of technology to create closer connections and valuable feedback loops across providers, patients and researchers.  Ultimately, we hope this will translate to people arriving at critical insights more quickly and partnering with each other to not only improve the care of the individual patient, but also to find cures for cancer and other devastating diseases.

Peter Neupert is a Corporate Vice President in Microsoft Health Solutions Group.

Reimbursement of Personalized Medicine Diagnostics: What Path Forward?

By their nature, diagnostic tests play a central role in the personalization of medicine: one can only better characterize a disease process, or predict who might respond well or poorly to a treatment, by measuring some biological characteristic of the patient. In fact, the explosion of human genetic information and advances in diagnostic technology platforms over the past decade have at last permitted real progress in personalized medicine.

This scientific progress is yielding notable successes. For example, use of certain therapies for colorectal cancer is now linked to testing for the KRAS gene mutation based on evidence developed by the therapeutics manufacturers.

However, despite the dramatic growth in the number and power of advanced diagnostic technologies, the increasing clinical and economic value of the tests derived from them, and the therapeutic power of the drugs whose use can and should be guided by those tests, the reimbursement system for diagnostic tests has not evolved to accommodate the development and adoption of personalized medicine diagnostics (PMDs). 

Reimbursement for diagnostic tests is grounded in decisions made and systems developed decades ago.  It reflects outmoded patterns of health care delivery and increasingly antiquated payment mechanisms.  It relies on timelines that progress far more slowly than the pace of PMD development, evidence standards that are ill-suited to the clinical and economic realities of PMD development, and payment methodologies that reflect neither the purpose nor the clinical and economic value of PMDs.

Consequently, many PMDs are not reimbursed appropriately. Such inappropriate reimbursement inevitably leads to inadequate access. If there is uncertainty about the ability to recoup the cost of developing or performing a PMD test, then the laboratory will not offer it, if indeed anyone even invests in its development. If the physician must provide an elaborate justification of medical necessity, then the test will not be ordered. If the patient is told that Medicare is unlikely to reimburse, then the test will be refused. In all cases, the patient will be denied the benefits of personalized treatment – the right medicine, at the right time, for the right indication.

What, then, can be done to improve this situation? Because appropriate reimbursement results from a complex interplay of codes, coverage decisions, policies, and payment amounts, addressing any of these aspects individually is inadequate. Instead, obtaining the full benefits of PMDs will require a fundamentally different reimbursement paradigm. This paradigm will require a system of PMD coding capable of distinguishing among PMDs with subtly different technical or clinical characteristics. It will require a payment methodology capable of delivering payment commensurate with the clinical utility and health economic value provided by PMDs. And it will require a coverage process and evidence requirements compatible with the realities of the clinical uses and market sizes of PMDs.

How we move from the current state to this new paradigm is not yet clear. We may need to carve PMDs out of the current reimbursement system.  Novel statistical methods for projecting clinical utility on the basis of small clinical data sets may need to be developed. Ultimately, legislation may be required to address shortcomings in Medicare’s reimbursement of PMDs.

What is clear is this: companies developing PMDs face significant reimbursement hurdles, particularly in cases where the PMD is not developed jointly with a companion therapeutic. If we want more PMDs, the required investment in product and clinical development must be recoverable through appropriate reimbursement, evidence requirements – whether at FDA or CMS – must be appropriate to the PMD being assessed, and reimbursement uncertainty must be reduced. It is critical that we address these issues, and soon, if we are to see personalized medicine continue to advance, and enjoy its benefits.

For a more in depth analysis of the reimbursement landscape facing PMDs, I invite you to read “The Adverse Impact of the US Reimbursement System on the Development and Adoption of Personalized Medicine Diagnostics” available for download on the Personalized Medicine Coalition website.