How decoding your genes might unlock the future of health care.
—Photo courtesy of iStock
There’s nothing more frustrating than feeling like the treatment or medication your doctor prescribed isn’t working. Maybe you’re getting all the side effects with none of the benefits, or maybe it’s the sixth therapy you’ve tried to no avail. Whatever the reason, you’re left wondering if there’s a better way.
Good news: There is.
Or at least there will be, according to the Center for Biomedical Informatics and Personalized Medicine, a partnership between UCHealth, Children’s Hospital Colorado, and the University of Colorado School of Medicine. Established in 2014, the center pulls together disciplines and institutions across the University of Colorado Anschutz Medical Campus to uncover advancements in genetics that can improve diagnoses and develop medical solutions that are tailored to individuals—all of which rejects the one-size-fits-all mentality so prevalent in treating diseases today.
The burgeoning field of personalized medicine is peppered with technical terminology that can make the science feel anything but personal. Yet the implications are significant—so much so that President Barack Obama carved out $215 million in his 2016 budget request to solicit voluntary genetic data from the public and further drive this type of research. We talked to key players in Colorado’s movement to learn how it works, what it looks like, who’s behind it, who has access, and what the future holds for finally making your health all about you.
PAGE 6: What's Next? 
Under The Microscope
Behind the scenes at a University of Colorado lab where it’s safe to say, “Pay attention to me.”
Dr. Dara Aisner holds up a thumbnail-size tissue sample—asymmetrical and discolored at the edges—under the fluorescent lab lights. It’s preserved in a specialty wax. “There,” she says. “You can see the tumor.”
Aisner is leaning over a counter in the Colorado Molecular Correlates (CMOCO) Laboratory on the University of Colorado Anschutz Medical Campus. Truthfully, the sample looks like any other mottled slab of flesh to an untrained eye, but that’s OK; it’s what the researchers can see that’s important. Technicians and scientists clad in lab coats are conducting microdissections of tumors, staring into microscopes, and transferring DNA under a sterile glass hood for testing.
It’s all part of a process spearheaded in part by Aisner, co-director of the lab and an assistant professor of pathology at CU, to look for molecular “biomarkers,” or genetic mutations in already-diagnosed tumors. Mutations wreak havoc on the body because they force some cellular processes into overdrive—which, in the case of cancer, means the cells won’t stop dividing. Finding those mutations—via the lab’s complicated protocols and what’s called next-generation testing—ideally allows physicians to prescribe targeted therapies, which attack the mutation to tell it to stop interfering with normal cellular activity. “The molecular process has changed the way we do things,” Aisner says. “It’s a labor-intensive process with a whole string of people. What we do requires an enormous amount of technical expertise.”
Bottom line: Those lab minutiae are part of a transformative line of thinking that alters how health-care providers go about treating diseases. “We’re not trying to answer today’s questions,” Aisner says. “We’re trying to answer tomorrow’s.”
—Image courtesy of UCHealth
A Case Study
When Janet Freeman-Daily learned at age 55 that her nagging cough was actually lung cancer, she was baffled. She didn’t smoke or live or work with smokers. Her diagnosis seemed like a fluke. “When you first hear the diagnosis of cancer, it’s like your ears stop working,” says the Seattle-based former aerospace engineer. “You’ve got so many new terms coming at you, and you’re just trying to come out of the fog.”
In the months that followed, she faced endless appointments, radiation, chemotherapy, handfuls of pain pills, infections, and a new fast-growing tumor on her collarbone that rendered her a metastatic lung cancer patient. Doctors predicted she had two years to live.
Today, at 59, Freeman-Daily has crushed those odds. Here’s why: Someone in her online lung cancer forum suggested she look into a gene mutation, called ROS1, that was connected to the disease. Her local cancer center didn’t test for it, nor had it heard of any clinical trials. However, a fortuitous trip in 2012 to visit family in Denver led Freeman-Daily to the discovery that the University of Colorado Cancer Center had just starting testing for ROS1. She was positive and enrolled in CU’s clinical trial for crizotinib, a drug that targets the ROS1 mutation with fewer side effects than chemo.
Crizotinib does not cure lung cancer permanently. But it does attack the mutation to keep the disease at bay. Freeman-Daily’s initial eight-week scan showed her most recent cancer nodules were gone; now, nearly three years into her treatment, she still flies from Seattle to Denver at regular intervals for her trial care, and she is still NED (cancer lingo for “No Evidence of Disease”). Thanks to ongoing genetic research at CU, Freeman-Daily’s treatment plan was far different than the typical blast-it-with-chemo routine that’s still common outside larger research hospitals. “Precision medicine, to me, means that I’ve gone from having a two-year life expectancy to having a fairly normal life, with some tolerable side effects,” she says. “I’ve got more time with my family to enjoy life.”
—Image courtesy of Gilles Frydman of Smart Patients
Status Quo Shake-Up
Conventional wisdom suggests patients suffering from the same condition should be treated with the same therapy: Breast cancer patients should get Medicine A; colon cancer patients should get Medicine B. But science now tells us the reason the efficacy of one-size-fits-all drugs varies by patient is because each person has unique DNA that responds differently to the prevailing treatments.
The shift to personalized medicine is possible because many drugs in research and development today successfully zero in on the genetic variation that causes one person’s condition instead of treating the condition the same way across a range of genetically diverse individuals. “Perhaps we should stop thinking about cancers as breast cancer or lung cancer,” says University of Colorado medical oncologist Dr. Robert Doebele, “but more as genetic cancers.” Put another way, even though Person A and Person B have different types of tumors, those tumors may be caused by the same genetic disturbance, or biomarker. The new strategy is to put Persons A and B in a single clinical trial that targets the biomarker they have in common instead of two separate trials focused on the separate tumors. “It argues for a new way of thinking about clinical trials,” Doebele says. “Personalized medicine is about matching the right drug with the right person.”
—Infographic by Peter Hoey; Source: Bayer Healthcare pharmaceuticals
Breaking it Down
Discovering—and analyzing—the biological secrets of your DNA.
How we got here scientists crossed a major milestone in the personalized medicine movement in 2003 with the completion of the Human Genome Project, an international scientific collaboration to identify and sequence all three billion pairs of DNA nucleotides (which ultimately make up genes) in the human body. This became the genetic blueprint for how the human body functions. Since then, researchers have been using it as a baseline to identify abnormal genes; as they continue to discover more about the genome, the conversation about disease has expanded from “What’s causing this?” to “Who is at risk for this?” to “How do we develop a way to treat this more effectively for individuals?”
Answering those questions is the mission of the new Center for Biomedical Informatics and Personalized Medicine in Aurora. At the heart of bioinformatics is big data, which is another way to describe the unfathomable amounts of information we have the ability to collect from large populations. One of the biggest stumbling blocks, experts say, is figuring out how to store and analyze the reams of data once we get them. To tackle this challenge, Kathleen Barnes, Ph.D., joined the faculty of the University of Colorado School of Medicine in spring 2015 as the director of the new center. Her team of clinical geneticists and research scientists—from pathologists to immunologists to oncologists—built a system that relies on the latest technology to organize and analyze patients’ medical and genetic information.
What Happens to my Personal Data?
1. When patients visit a UCHealth facility, their medical records—such as height, weight, past treatments, allergies, and known diseases—are logged into the electronic medical records system.
2. From there, UCHealth’s new data warehousing technology can pull and store millions of data points from these records and anonymize them by removing names and identifying information.
3. During the same visit, patients can choose to donate a blood sample to UCHealth’s new biobank, which is slated to be operational this month with a capacity of 500,000 new patients a year.
4. Scientists can analyze DNA and use a special tool, called a single nucleotide polymorphism (SNP or “snip”) chip, to detect gene variations in the DNA samples. Most patients are willing to contribute, Barnes says, even if it’s just in the name of research. “But we’re learning that if you discover a mutation,” she says, “it’s key to be able to go back to that person”—in cases when the sample reveals a gene variant that could affect lifestyle or health-care choices or aid in related research. “Here, there’s an opt-in consent form to ask patients if we can go back to them in the future.”
5. This leaves two sets of data: a data warehouse full of observable characteristics and health conditions from medical records and a set of genetic information, including gene variations, or “typos,” from the biobank and SNP analyses. The challenge, says Barnes, is bridging those two data universes to draw conclusions. That’s where UCHealth’s new biocomputing center comes in. Using high-speed data analysis, the facility allows researchers to make connections between genetic blips and certain diseases—which eventually drives the development of new drugs to target those blips. The center also lets researchers predict responses to certain medications and practice more efficient health care. “For example, we already know that some people have a mutation that can predict how well they might respond to an inhaler if they have asthma,” Barnes says. “If we had a systematic way to screen for this mutation—and all of the others that predict responses to common asthma therapies—children diagnosed with asthma wouldn’t have to go through a series of trials and errors to find the best combination of drugs to manage their illnesses.”
—Infographic by Peter Hoey
Widespread applications are coming—eventually.
Oncologists have been dabbling in personalized medicine for a long time. In the mid-1990s, for instance, the National Cancer Institute found that a BRCA gene mutation gives carriers a 45 to 65 percent chance of developing breast cancer by age 70. Women now have the option to get tested and make preventive choices.
But what about other diseases? Less is known outside of oncology, and the quest to figure out other applications is part of what’s driving the field. Dr. David Schwartz, the chair of medicine at the University of Colorado Hospital, has made several genetic discoveries pertaining to pulmonary fibrosis, which means he can now look for variants in certain genes, like the MUC5B gene associated with the disease. “Now we know there are people who are much more susceptible to developing pulmonary fibrosis based on their family histories and genetics,” Schwartz says. Researchers aren’t yet ready to make treatment recommendations for the disease, but, Schwartz says, the genetic findings should be used in clinical trials.
The basic idea applies to other diseases such as heart failure (see “Cardiomyopathy” at right), ALS (see “ALS” on page 37), endocrine disorders, muscle diseases, and conditions like dementia and Parkinson’s. The process, though, is complicated, says UCH medical geneticist Dr. Matthew Taylor. For example, there are many types of diabetes—only some of which are linked to mutations—and there aren’t yet drugs and trials developed for every mutation identified. “For the majority of individuals with diabetes, we don’t have answers,” Taylor says. “But we’re getting better and better at classifying diseases properly—at identifying the genetic mistakes causing them.”
—Photo courtesy of iStock
A Case Study
Alfredo Hernandez ran marathons, cycled, wrestled competitively, and ate well. By all accounts, the now-retired biochemist based in Albuquerque, New Mexico, was the picture of fitness.
At age 39, his life changed. Fatigue, dizziness, and blackouts led to an arrhythmia (irregular heart beat) diagnosis followed by a cardiac ablation—a procedure to correct abnormal electrical circuits in the heart—a pacemaker, and, later, beta blockers (medication to reduce blood pressure). Still, his heart couldn’t sustain a normal rate. “The cardiologist just said, ‘I don’t know what’s causing your heart failure,’ ” Hernandez says. “I was pretty desperate.”
Then Albuquerque electrophysiologist Dr. Michelle Khoo—previously a professor at the University of Colorado—recommended genetic testing. Sure enough, Hernandez proved to be a carrier of the LMNA mutation, which affects the lamin A/C proteins in the heart. (Lamin A/C give cells stability and strength.) “These mutations are known to progress into conditions called dilated cardiomyopathies,” Hernandez says. “Now I knew I was at risk for sudden cardiac death.”
Khoo took the results to geneticist Dr. Matthew Taylor at UCHealth, who had already published research on the LMNA mutation. Together, they determined that a pacemaker—Hernandez was on his second—should be complemented by an implantable cardioverter-defibrillator (ICD), which uses electrical pulses to help control life-threatening arrhythmias. If Hernandez was in danger of sudden cardiac arrest, the device would shock him back to life—and it did, on several occasions. The ICD wasn’t a permanent fix, but it slowed a grim trajectory, buying Hernandez time to join Taylor’s clinical trial of medication for the mutation and keeping him alive long enough to get a heart transplant in March 2015. Had his team not discovered the mutation and given him an ICD, odds are he wouldn’t have lasted that long.
More than half of Hernandez’s large extended family, including his daughter, have since tested positive for the LMNA mutation, which means early symptoms can be managed. “If you know what mutation you’re dealing with,” Hernandez says, “that valuable piece of information could change your treatment plan. It’s about having the right physicians who are willing to take the time to figure it out and not just accept protocol treatment.”
—Photo courtesy of iStock
Personalized health care might be more readily available, but affordability depends on a range of factors.
Linked In five years ago, front range hospitals were pretty much standalone institutions. If you needed a lab test or the expertise of a specialist not available at your in-town facility, you faced what could amount to hours of round-trip travel time, a new set of records, more data entry, and insurance hassles. The system—or lack thereof—wasn’t set up for the brave new world of personalized care.
That’s part of the reason the UCHealth system was established in 2012. UCHealth is a network of newly partnered facilities along the Front Range that includes the University of Colorado Hospital Anschutz Medical Campus in Aurora, Poudre Valley Hospital in Fort Collins, the Medical Center of the Rockies in Loveland, and the Memorial Hospital Central and Memorial Hospital North, both in Colorado Springs. Later this year or early in 2017, UCHealth will also open a new $170 million facility in Longmont.
Ultimately, the network is a way to bring patients in rural areas—say, residents just across the border in southern Wyoming or west of Colorado Springs in the mountains—the same testing services, medical experts, and research (like the CMOCO Lab for molecular tumor testing) the main hub in Aurora offers. Providers will have resources from Fort Collins to Colorado Springs, and patients will have more options. “The most important thing from a patient’s perspective is that the platform is uniform throughout the hospital system,” says Elizabeth B. Concordia, UCHealth president and CEO. “Every facility uses the same electronic health record software. We have the same protocols, the same pharmacies. It’s certainly satisfying to a patient when you come to a provider and we know you.”
Shelling Out even with geographic access, some patients face a different hurdle to full-fledged personalized medicine: money. Experts at UCHealth readily admit there’s no black and white when it comes to what sort of genetic testing is covered by the spectrum of insurance plans out there. Because experts know the most about breast and lung cancers when it comes to genetics, it’s easier to establish clear coverage policies. For rare diseases, there may be no policy because of a lack of genetic data or in-depth studies. (Insurance companies usually decide whether to cover a treatment after it has gained FDA approval, which is dependent on extensive research.) “The challenge is that we’re just at the tip of the iceberg,” Concordia says. “There are so many genetic markers we’re going to identify over the next five to seven years; right now, there’s more that we don’t know than we do know.” Translation: It’s ultimately a case-by-case basis.
Pay to Play
How much does it cost to customize my medicine?
$100 to $2,000+ = Price range of testing for most genetic mutations; the cost depends on the complexity of the test and the lab that conducts it.
100 = Percentage of the cost of the genetic test for breast cancer that is typically covered by insurance as long as the patient meets certain criteria, such as having a family history of the BRCA mutation or a close blood relative who was diagnosed with breast cancer before the age of 40.
$1,000 to $1,500 = Estimated cost of a genetic test for a condition like Huntington’s disease; insurance companies often do not have policies in place for less common conditions.
$0 = Average patient cost to participate in a clinical trial (less travel expenses). Trials are typically sponsored by a pharmaceutical company or a health institution.
NOTE: In 2008, new DNA sequencing technology was introduced, which prompted the sharp downward trend in cost.
—Infographic by Peter Hoey; source: National Human Genome Research Institute
The Future is (Not Quite) Here
We asked the experts where personalized medicine is heading. Here’s what they told us.
A Case Study
Privacy fears are one thing, but Kelsey Skattum of Colorado Springs carries a different weight. The U.S. Air Force medic, 26, and her husband have a one-year-old son. They’d like to have more children, but it’s not that simple. Last March, Skattum’s father died from amyotrophic lateral sclerosis, or ALS, at the age of 51. The progressive disease attacks nerve cells in the brain and spinal cord and eventually renders patients unable to move, speak, eat, or breathe.
According to the University of Colorado’s Adult Medical Genetics Program, Skattum has a 50 percent chance of carrying the genetic mutation that caused her father’s disease. As such, she decided to undergo genetic testing. At press time, her results were pending; however, a positive result would mean she has a 70 percent chance of developing ALS symptoms by the time she’s 50—and that her son has a 50 percent chance of having the mutation as well.
“I’d rather know now than wait till I start developing symptoms,” Skattum says. “If there’s something I want to do, I feel like I should do it now. If I do have the mutation, I don’t want to have another child because I don’t want to pass on the disease.” It’s a heartbreaking choice for anyone, but especially for a family that wants to grow.
—Photo courtesy of iStock
The Privacy Question
Your DNA may hold information you want to keep secret.
Let’s say you donate dna to a biobank for research purposes. They’ll strip your identity from the sample, and no one except authorized medical officials can access the bank. The data will never be linked back to you, the hospital assures you.
But is this a gray area? Genetic information isn’t the only set of mass data with penetrable security measures. (Remember when Chinese hackers got access to security clearance info on millions of federal employees and contractors in 2015?) Your original DNA sample, tagged with your info, probably lives somewhere. What sorts of secrets could it reveal if leaked? “Strictly speaking, your DNA sequence is the ultimate personal identifier,” says Dr. John Reilly, dean of the University of Colorado School of Medicine, “much more so than your social security number. It’s what makes it such a powerful tool in forensics.”
Even if your genetic code were made public, you’d have some protection: In 2008, Congress passed the Genetic Information Nondiscrimination Act, which prevents health insurers from canceling, denying, changing, or refusing to renew coverage or premiums based on a genetic predisposition for specific diseases. (The law doesn’t apply to life or disability insurance.) It also bars employers from considering genetic information when making personnel decisions. “Our society is not perfect,” Dr. David Schwartz says, “but the advances we make are a function of us having to debate these tough issues around our identities.”
—Photo courtesy of iStock