The great equalizer

In February, UCI announced the launch of the Institute for Precision Health. To veteran researchers like Suzanne Sandmeyer, professor of biological chemistry, it was a dream come true.

“When I started what’s now the Genomics Research and Technology Hub in 1999, the long-range goal was always that our research would translate into better patient care,” says Sandmeyer, director of the facility. Over the years, she and colleagues realized phenomenal leaps in technology and capabilities, but still their work involved primarily academic studies.

“Fast-forward to now,” she says. “What excites me about the precision medicine movement here at UCI is that all this data that we’re empowered to collect with the new high-throughput instruments can be turned toward clinical trials and, ultimately, improve patient treatment.” The Institute for Precision Health, to her, is the culmination of many years of hoping and working toward an objective.

While certainly many in the UCI orbit join Sandmeyer in her excitement about IPH, others find the whole thing a bit of a head-scratcher.

“It’s not uncommon for me to hear ‘Now what exactly is precision health, and why do we need IPH?’” says Dr. Alpesh N. Amin, the Thomas and Mary Cesario Endowed Chair in Medicine and co-director and medical director of IPH. “For a lot of people, the idea of precision health is still new.”

Which isn’t surprising. Even precision health trailblazers have had to parse what exactly it is.

What is precision health?

In the 2015 State of the Union address, President Barack Obama announced the Precision Medicine Initiative. At the time, precision medicine was described as a new approach to disease prevention and treatment that considers individual differences in people’s genes, environments and lifestyles. The White House earmarked money in the 2016 budget to develop the National Institutes of Health’s All of Us project, a national data collection effort that aims to enroll 1 million or more volunteers who serve as diverse sources of information. Today, UCI is the biggest enroller of All of Us participants in Southern California.

To delve more into what precision medicine is, though, first understand that it fits within the broader concept of data-driven decision science (the process of making decisions based on actual data rather than intuition or observation alone). Retail, banking, education, logistics – there probably isn’t an industry today that hasn’t been touched by the process. Think back to the 2011 film “Moneyball.” That was a story about baseball using data-driven decision science to find undervalued players. Healthcare has also embraced it. Electronic health records, real-time alerting, supply chain management and improved prescription management are all part of data-driven decision science and big data.

But knowing this still doesn’t get to the essence of what precision medicine, or precision health, aims to be.

Precision medicine is an endeavor – or perhaps, as researchers argue in a 2017 article in the European Respiratory Journal, a process. Patient data – far more data and different types of data than ever before – is used alongside the power of computer algorithms, predictive modeling and artificial intelligence to help clinicians and patients in making individualized treatment and health decisions. And then, once these decisions are made, outcomes are fed back into the system to help better inform the next decision and all subsequent patients. This means that precision health’s aim is to use the power of big data to create a healthier individual and thereby a healthier society. That’s the broad concept, at least.

“It’s common to hear people talk about precision medicine being about both the multitudes and then really the individual,” Amin says. “That may be confusing messagewise, but it’s accurate.”

The new ‘evidence-based’

Haven’t medical decisions long been data-driven? The first successful randomized, controlled clinical trial, in 1946, gave rise to modern medicine’s evidence-based approach. So yes, we’ve been ostensibly data-driven for quite a while. But because of progress in many areas – everything from biostatisticians pioneering extremely complex decision-science methodology to improvements in cloud storage and computing power – we’re now able to be next-level data-driven. Advancements in artificial intelligence are the game-changer.

The newest form of AI, deep-learning neural networks, has completely revolutionized the way machine-learning algorithms learn and think, says Dr. Peter Chang, co-director of UCI’s Center for Artificial Intelligence in Diagnostic Medicine and assistant professor-in-residence of radiological sciences. Older forms of AI required a human to carefully go through a list of patterns, rules and assumptions and manually program the human experience into a computer.

“Modern forms of AI, however, allow computers to extract patterns and make inferences without a priori human assumptions,” says Chang, who directs IPH’s AI research program. “For example, if I wanted to teach the algorithm how to play the game of chess, I could simply explain the rules of chess and allow two AIs to play against each other.”

This represents a major leg up on human thinking. “Interestingly, this strategy has allowed modern AI systems to learn new information that may have been previously unrecognized by even human experts,” Chang says. “With video games, oftentimes we may think that the AI is intentionally losing, only to realize at the very end that the computer has come back and beat the human by a small but consistent margin every single time. For healthcare, the implication, of course, is that an AI may be allowed to discover patterns without the biases of flawed human assumptions or explicit programming – that’s really where the power lies.”

A growing focus on patient-centered health services research has also helped foster a climate ripe for precision health. The research has sometimes revealed a divide between what patients value and what clinicians value. Furthermore, research has also acknowledged that personal experiences – sometimes having to do with systemic barriers like bias and discrimination – can alter, for example, a medicine’s effect or a patient’s perception of the value of care.

The machinations and methodologies behind precision health aim to consider all of this as data, as messy as it may sound. So when data is being crunched by precision medicine processes, it might include clinical information (biomarkers, mortality, etc.), patient-reported impacts (e.g., function, mood, symptoms), treatment-related attributes (mode of administration, dose frequency, adverse events, etc.), use of resources (e.g., hospitalizations) and/or societal effects (ability to work, caregiver burden, productivity, etc.).

Bernadette Boden-Albala, professor and director of UCI’s Program in Public Health and founding dean of the planned School of Population and Public Health, warns people to not be surprised, however, if it turns out that we should be collecting other data or looking at it in different ways.

“We just don’t yet know what we don’t know,” she says. “But I can imagine that with all the new abilities to look at data in all the new ways, we’ll discover that we’re sometimes actually missing key data and need to go back and collect that to reach more meaningful conclusions.”

Omics Approaches Aren’t the Whole

Getting people’s biological blueprints – genetic sequencing, heritable modifications of DNA (epigenetics) and omics, which include sequence RNA and cellular building blocks determined by DNA and RNA (transcriptomics), proteins (proteomics), and profile metabolites (metabolomics) – into health records would be a huge precision health achievement.

In fact, researchers and clinicians at UCI are already doing formidable work using data like genetics and omics in precision medicine-type approaches.

Clinicians in UCI’s Chao Family Comprehensive Cancer Center often utilize data-driven methods to determine whether a given drug could be effective in a specific individual, says Leslie Thompson, co-director of the Institute for Precision Health and Donald Bren Professor and Chancellor’s Professor of psychiatry & human behavior as well as neurobiology and behavior.

For example, breast cancer patients who test positive for a protein called human epidermal growth factor receptor 2 receive treatments that specifically target HER2. The treatments are now so effective that the prognosis for HER2-positive breast cancer has improved drastically.

These types of precision health capabilities – particularly what they could mean to clinical trials – are primary motivators for scientists such as Thompson.

“I’ve dedicated my career to studying neurodegenerative diseases like Huntington’s and ALS,” she says. “With so many of these diseases, including the more common Alzheimer’s and Parkinson’s, there are no treatments available that change their course, and so many clinical trials have failed to show benefit to patients.”

She says this is because we haven’t been able to fully understand many diseases in individuals.

Precision health should allow for more sophisticated approaches to diseases in subgroups of patients – approaches that incorporate their genetics, environment and other health factors so that clinicians can better define, understand and treat diseases. The potential impact on clinical trials could be revolutionary.

Traditional clinical trials have favored patient homogeneity to find treatments that appear to work for a broad swath of people. In fact, patient heterogeneity – difference – is often seen as a research nuisance to avoid.

Precision medicine approaches, however, promise to change this. Sandmeyer explains: “Right now, if you have a clinical trial where a certain drug only helped 10 percent of your study participants, that drug would likely fail the trial. But the drug could’ve succeeded in people who were biologically receptive to it. We need to get genomic data into the clinical record so information like that can be utilized. We’ll make considerable advancements when we know who is likely to be biologically receptive to a certain treatment and who is not.”

Having and using omics and other data, however, shouldn’t be confused as the whole of precision medicine, she says: “We haven’t delivered on precision medicine until all the omics information and everything else gets into the medical record, gets into clinical trials and gets delivered back in the form of more equitable healthcare to the people we serve. When that happens, that’s truly precision medicine.”

How will the process work at UCI?

The Institute for Precision Health falls under the direction of Steve Goldstein, vice chancellor for health affairs and professor of pediatrics, who describes the effort as “heralding a future of tailored care that places the patient at the center and in control.”

IPH was launched with leaders in eight different areas across UCI (see page 23) to unite the campus and support precision health as a focused, collaborative effort. In fact, the data engine of the process has been named the Collaboratory for Health and Wellness.

Tom Andriola, UCI vice chancellor for information, technology and data, as well as an IPH advisor, calls the collaboratory a “dream” foundation. “It’s a powerful networked landscape,” Andriola says. “Researchers will be able to put together a whole variety of forms of structured and unstructured data and information in one place.” He says this will mean de-identified health information from medical records, including text and images, but it might also include genomic, environmental, demographic and other data that contribute to health and well-being. Equally important, the collaboratory will support precision health’s new and evolving methodologies for analyzing data.

Through the collaboratory, UCI hopes to serve as a trusted leader and safe hub for precision health care and research, making information and capabilities widely available while also tightly guarding patient privacy – all under the auspices of precision health’s overriding aim to deliver high-quality, equitable care and support lifelong wellness.

Growing out of this commitment, IPH has paved the way for UCI to step up as the academic lead of a public-private collaboration called the global health ecosystem, an informal gathering of complimentary organizations focusing on practice redesign, deployable equity and informed policy. Through the global health ecosystem, UCI is working with collaborators, such as Mitre Corp., on cloud-based health and medical solutions for the public good.

IPH leadership plans to conduct outreach to discuss capabilities and potential projects with UCI researchers and clinicians, as well as industry.

“Advances are scalable and open-access, so they improve healthcare and wellness in Orange County and far beyond,” Goldstein says. “Success will be measured as improved patient outcomes, cost-effectiveness and equity.”

IPH intends to eventually settle into a brick-and-mortar campus location where education for clinicians, community events and outreach can take place.

As the campus and wider community become more acquainted with IPH, Amin, who helped found and develop UCI Health’s successful hospitalist program back in 1998, predicts that precision medicine will someday be the norm for research and medicine. It will have that much impact.

“But for right now,” he says, “we have somewhat of a startup mentality in that we’re building our plane and starting to fly it at the same time. This is why it’s an exciting time to take part in IPH. It’s the start of a new era in healthcare, and it’s a privilege to help get something like this off the ground.”

Chang agrees – and recommends that everyone breathe. “AI and precision health are exciting new areas of research, but I’d urge everyone to stay grounded and be patient. There are a lot of unknowns and a lot to explore and understand, so a balanced perspective is needed to truly make strides translating these technologies in ways that, ultimately, will help researchers, clinicians and patients.”

Creating Big Data: UCI and All of US

Since 2017, volunteers have signed up to have their medical records; gene profiles, metabolites (chemical makeup), and microorganisms in and on the body; environmental and lifestyle data; and even information from personal devices like Fitbit used as part of the National Institutes of Health’s All of Us project. All of Us hopes to collect health data on more than 1 million people. Most of it will come from traditionally understudied, diverse populations – better reflecting our increasingly varied population. The NIH has partnered with leading experts in privacy, bioethics, civil liberties and technology. All research data is de-identified. This means that names, addresses and other identifying information are removed. UCI is the largest enroller of All of Us participants in Southern California. To take part or learn more, go to https://allofus.health.uci.edu.

UCI Institute for Precision Health: Directors and Leads

UCI leaders say the vision for IPH had long been in the works, but the COVID-19 pandemic was a beta test of sorts, demonstrating how precision medicine approaches could efficiently address medical needs. In early 2020, UCI clinicians, biomedical and computer scientists, and public health experts collaborated to create an AI-driven tool to assess COVID patients. The app-based tool, the COVID Vulnerability Index, showed that a data-driven approach could help yield the best outcomes for individual and community health. On the heels of that success, Steve Goldstein, vice chancellor for health affairs – who has said that a significant way to advance society is through health – engaged partners to create IPH. On campus, IPH comprises eight programs focused on three goals: redesigning health practice to improve care and decrease costs, deploying solutions to achieve health equity, and empowering effective health policy.

• The statistics, machine learning and artificial intelligence program develops novel statistical methodology to integrate and analyze health records, molecular data and observational clinical outcomes. It’s led by Daniel Gillen, Chancellor’s Professor and chair of statistics, and Zhaoxia Yu, professor of statistics.
• The applied AI research program translates machine learning methods into deployable solutions addressing clinical problems and matching the cost of care to its value. It’s led by Peter Chang, assistant professor-in-residence of radiological sciences.
• The applied analytics and AI program brings novel solutions to improve health and well-being to ambulatory and inpatient settings. It’s led by Daniel Chow, assistant professor-in-residence of radiological sciences.
• The precision omics program generates and translates genomic, proteomic and metabolomic research results into clinical applications. It’s led by Suzanne Sandmeyer, professor of biological chemistry and director of the Genomics Research and Technology Hub, and Leslie Thompson, Donald Bren Professor and Chancellor’s Professor of psychiatry & human behavior as well as neurobiology and behavior.
• The Collaboratory for Health and Wellness is an ecosystem that fosters collaboration across disciplines and organizations through the integration of health-related data sources. It’s led by Kai Zheng, professor of informatics, and was launched by Tom Andriola, vice chancellor for information, technology and data.
• The deployable equity program engages community stakeholders and health equity groups to create solutions that narrow the disparities gap in the health and wellbeing of underserved and at-risk populations. It’s led by Dan Cooper, professor of pediatrics and director of the UCI Institute for Clinical and Translational Science, and Bernadette Boden-Albala, professor and director of the Program in Public Health and founding dean of the planned School of Population and Public Health.
• The end-to-end data infrastructure makes available real-world data in order to positively impact clinical outcomes, quality, research and operations. It’s led by Alpesh N. Amin, the Thomas and Mary Cesario Endowed Chair in Medicine, and David Merrill, director of enterprise data and analytics at UCI Health.
• For the education and training program, leadership from each of the areas plans to bring courses, seminars and other educational opportunities in statistics, machine learning-artificial intelligence, omics and bioinformatics to practitioners and students.