A month following surgery for thyroid cancer, a Hartford Hospital patient’s tumor grew to 10 inches. The case was presented to the hospital’s tumor board, which involved 30 doctors from different specialties.
The gene mutation found to be controlling the patient’s tumor growth was already well-established as a driver of melanoma, the deadliest form of skin cancer, says Dr. Sope Olugbile, medical oncologist at Hartford HealthCare. Chemotherapy wouldn’t work fast enough against the aggressive tumor. Tumor board members recommended a targeted therapy already treating patients with melanoma. “Without that genetic information, we wouldn’t have been able to come up with that therapy,” he says. The treatment saved the patient’s life, so far. “Our goal is to use more of the genetic information to drive the treatment of cancer patients.”
This type of personalized care, known as precision medicine and its subset, genomic medicine, has been offered for years at world-renowned cancer-treatment hospitals such as Memorial Sloan Kettering Cancer Center in New York, Dana-Farber Cancer Institute in Boston and University of Texas MD Anderson Cancer Center in Houston. It’s now the standard of care in Connecticut’s Hartford HealthCare Cancer Institute, UConn Health Center in Farmington, Connecticut Children’s Medical Center in Hartford and Smilow Cancer Center at Yale New Haven Health. “Cancer therapy has become precision therapy,” says Dr. Roy Herbst, professor of medicinal oncology and pharmacology, and chief of medical oncology at Yale Cancer Center and Smilow Cancer Hospital.
While it’s most commonly used with cancer patients, precision medicine is also making inroads into other areas of health care including the treatment of some cardiac patients. It’s also being studied and used on a limited basis to treat those with rare diseases. In the U.S., newborns are screened with a blood test for hearing loss and heart defects. If detected and treated early, this can prevent death and disability in some cases. For some doctors and researchers, precision medicine holds the promise of effective targeted diseases and chronic conditions, and, even more revolutionary, the chance to prevent illness before it arises. The race is on to gather as much data as possible in order to increase understanding of the connection between genes and overall health; here in Connecticut, Yale’s Center for Genetic Health last fall launched its “Generations” project to collect DNA from 100,000 volunteers (see sidebar below).
Precision medicine involves the study of human genes, called the genome. The human genome contains 23 pairs of chromosomes within all human cells, and each chromosome contains hundreds to thousands of genes. Using high-level computing and mathematics, genomics researchers analyze massive amounts of DNA-sequence data to find variations or mutations that affect health, disease or response to drugs, according to an online description by The Jackson Laboratory for Genomic Medicine in Farmington.
Researchers can sequence an entire tumor to look for markers or abnormalities that can be treated with a targeted medication that attacks that mutation, unlike traditional chemotherapy that kills healthy cells along with cancer cells, says Herbst, also associate director for translational science at the Yale School of Medicine.
These days, when Yale’s precision medicine tumor board meets weekly, they don’t focus on where the tumor began, he says. They look at what errors occurred in the DNA of the tumor, because once they know what’s driving the tumor, they can treat it. For example, lung cancer is the most common cancer in the world. When a nonsmoker gets lung cancer, doctors sequence the tumor’s DNA to see if it contains one of eight genes known to mutate.
Each cancer cell has about 18,000 to 20,000 genes, and there are some cancers where just one of those genes is directing the growth of the cancer, Olugbile says. “We call that the driver gene. The other 17,999 are just following the lead of that driver gene,” he says. “That means if we tag just that one gene with the medication then we can actually shut down the growth of the entire cancer.”
Traditional chemotherapy can only be given for 4-6 months because of the side effects, while targeted oral medications have very few side effects and patients remain on them for an average of two years, Olugbile says.
In the past five years, genetic testing has become standard of care for some cancers — specifically colon, lung and melanoma — because those types of cancers tend to have genetic mutations that have been known to respond to therapy, says Sara Patterson, manager of clinical analytics and curation at Jackson Labs, which works with UConn and Yale researchers. But targeted therapy is not a cure-all, and researchers are still a long way from using precision medicine to treat all cancer patients. Even if cancers have the same genomic change and mutation, there’s no guarantee they will all respond to the same therapy, she says. Overall, precision medicine is only effective at stopping the spread of cancer in an average of 20 percent of cancer patients treated, Olugbile says, with variations by cancer. Sometimes the cancer returns because the tumor changes to resist the therapy, Patterson adds.
As doctors and researchers do more genomic sequencing, the data pool will grow and so will knowledge of what medications work most effectively against various tumor types. “The more information we gather, the better we’ll know how to treat specific patients,” Patterson says.
Reimbursement from insurance companies can be a challenge. If precision treatment for a particular type of cancer hasn’t been approved by the insurance industry, it’s difficult to get reimbursed for genomic testing, says Sue Mockus, director of product innovation and strategic commercialization at Jackson Labs. It’s a catch-22. Even though a patient with pancreatic cancer could benefit from a targeted therapy, unless that patient is part of a clinical trial that would pay for the genomic testing, the patient would have to pay out of pocket, the annual cost of which can run into the hundreds of thousands of dollars. “If you do have a mutation identified and your physician wants to give you the medication off label, you have to fight with the insurance company,” Mockus says.
Experts have suggested a value-based approach to precision medicine, reports the International Journal of Public Health. This means policy decisions about reimbursement and investment in research and development will factor in how long patients’ lives are prolonged and the quality of those lives, the Journal reports.
Oncologists also offer cancer patients immunotherapy, another form of personalized medicine, Patterson says. They’re using diagnostic tests on tumors, independent of genomic sequencing, to determine if their tumor profiles make them a good immunotherapy candidate. Immunotherapy is approved for multiple tumor types, as long as they have certain markers, she says.
Former President Jimmy Carter became cancer free after receiving radiation and immunotherapy to treat the melanoma that had spread to his brain and liver. While immunotherapy can cure cancer for some, it’s only effective about 20 percent of the time, Olugbile says. It varies a bit by cancer, with some cancers having a higher success rate, he adds.
Through a collaboration with Memorial Sloan Kettering, Hartford HealthCare’s Advanced Disease Clinic was scheduled to open this spring to give patients even more options, he says. If targeted therapies and immunotherapies don’t work or are not a match for patients, doctors will look for suitable clinical trials that offer potential treatments, Olugbile says. “Our goal is to create awareness on two fronts, one is among the doctors. Yes, we are available to help if patients have gone through standard of care who didn’t respond,” he says. It’s also an option for patients who want to be treated with precision medicine closer to home. “The goal is to make it available so they don’t have to go to New York or Boston,” he says. “It’s right here in Hartford and hopefully at other cancer centers over time.”
From Yale, Herbst leads a clinical trial through the National Cancer Institute where he and his team are trying to match the right patient to the right drug. “Every tumor is getting sequenced. That’s accelerating the field. The sequencing techniques have gotten cheaper and faster, so we can analyze them at the point of care,” Herbst says. “This is why clinical trials are so important. What’s a clinical trial today is standard of care tomorrow.”
In a study published in the journal Science Translational Medicine, a multi-institutional research team including a Connecticut doctor developed an advanced method to analyze existing data from thousands of clinical trials, comparing which genes FDA-approved drugs work against to the genes active in pediatric brain tumor patients. This sped up the lengthy process of developing cancer drugs.
Dr. Ching Lau, head of the oncology-hematology division at Connecticut Children’s Medical Center and the pediatric oncology-hematology department at UConn School of Medicine, is accessing the World Community Grid, an IBM-funded program that allows researchers worldwide to perform tens of thousands of virtual experiments. “Instead of screening thousands and thousands of compounds to try to find a potential drug, we found we could use genomics data already available and do a more systems-approach analysis to figure out the predominant pathways driving the tumor cells,” Lau, professor at The Jackson Laboratory, says in an email. “Then we asked if there were any existing FDA-approved drugs that could potentially modulate those pathways.”
The researchers identified eight drugs that could potentially fight medulloblastoma (MB) tumors, the most common malignant brain tumor in children. One of the drugs showed an increased survival rate in mice with MB tumors, and a clinical trial is being pursued.
Personalized medicine and heart disease
Precision medicine’s applications have expanded beyond cancer care. At first, much heart disease research relied on a genetic analysis of whether someone was predisposed to a disease. Thanks to a growing database of patient information that is shared worldwide, researchers can mine huge data sets with hundreds of thousands of cases for patterns and abnormalities that lead to discoveries, says Beth Taylor, associate professor of kinesiology at UConn and director of exercise physiology research in cardiology at Hartford Hospital. Researchers and clinicians know that about half the people who have heart attacks don’t have the typical risk factors such as high blood pressure, obesity and diabetes. To determine why physically active people with healthy diets have heart attacks, researchers are using precision medicine to comb through large studies to find small predictors, Taylor says. “Often the influence of any one factor is hard to detect unless you have a big sample size,” she says.
The National Institutes of Health requires grant recipients to share their data to a national registry so that researchers have access to big data, she says. (Personal information such as date of birth, name and address are removed from files used for research studies.)
“When we first began to really measure genetic variations, it was believed that was going to be the big hope in treatment,” Taylor says. But genes are complex and environmental factors modify genetics for multiple generations.
“For the first time ever, we’ve got wide-scale computing ability to analyze huge data points. This can better allow us to predict disease progression and optimize treatment,” she says. “Many of us would say that this concept of big data is as or more important than genetic risk. Genetic risks are not the whole picture.”
For the first time ever, we’ve got wide-scale computing ability to analyze huge data points. This can better allow us to predict disease progression and optimize treatment.
Progress with diabetes
Precision medicine is not widely used in the treatment of diabetes in the U.S., except when it comes to a rare form of diabetes called neonatal diabetes mellitus. While type 1 and type 2 diabetes are controlled by two or more genes and additional genetic factors, neonatal diabetes mellitus involves a single gene and develops in babies under 6 months old.
Through genetic testing of babies with elevated blood sugar levels, researchers learned that about half the patients have gene mutations that respond well to a pill used to treat type 2 diabetes and they don’t need to be on insulin for the rest of their lives like type 1 diabetics, says Karel Erion, director of research stewardship and communications for the American Diabetes Association.
When infants show signs of type 1 diabetes at Yale New Haven Children’s Hospital or Connecticut Children’s Medical Center, they are automatically tested for neonatal diabetes, hospital doctors say.
An example of precision medicine as a predictor of disease is the TrialNet database, which uses genetic testing to determine whether the relatives of those with type 1 diabetes have two or more of the five diabetes-related autoantibodies (proteins produced by the immune system directed against the person’s own proteins) linked to increased risk of developing type 1 diabetes. Type 1 diabetics must take insulin for the rest of their lives to survive, and there’s no known way to prevent the autoimmune disease. Type 1 diabetes, formerly called juvenile diabetes, typically strikes children and adolescents, causing the pancreas to stop producing insulin, a hormone needed to process sugar, or glucose, from food. Type 2 diabetes was formerly known as adult-onset diabetes, but the disorder is being seen in more children, thought to be the result of a rise in childhood obesity. Screening identifies the early stages of the disease years before any symptoms appear, according to the TrialNet website.
In a study published in the New England Journal of Medicine, researchers from the TrialNet Study Group, led by Yale University’s Dr. Kevan Herold, found that an experimental medication delayed the onset of type 1 diabetes in high-risk participants by two years compared to the control group. The disease was diagnosed in 43 percent of the participants who received the medication, teplizumab, and 72 percent of those who received the placebo.
Alzheimer’s disease and dementia
Only 1 to 3 percent of the 5 million people living with Alzheimer’s disease have a genetic mutation that leads to what’s called genetic or familial Alzheimer’s. But one in three older adults will eventually develop some form of dementia, says Rebecca Edelmayer, the Alzheimer’s Association director of scientific engagement.
Like other diseases that strike large segments of the population, researchers rely on big data to learn about Alzheimer’s and which genes play a role in who gets it. Researchers have learned that there are several risk factors that contribute to dementia, she says. Specifically, the presence of heart disease, high blood pressure, diabetes, social and cognitive isolation, poor nutrition and the level of education, can contribute to cognitive decline, she says.
Scientists from around the world share research data and draw from data in the Global Alzheimer’s Association Interactive Network, she says. “The field has made some dramatic advances in understanding of how genetics play a role and how other underlying diseases play a role,” Edelmayer says. “We need to give doctors evidence-based recommendations.”