Intermountain Healthcare Researchers Launch Major Three-Year Genomics Breast Cancer Study

Cancer researchers at Intermountain Medical Center and the Intermountain Healthcare Precision Genomics Program in Salt Lake City are launching a three-year study to determine if a blood test that looks for DNA from a cancer tumor can be used to complement mammography to improve the way breast cancer is diagnosed.

The goal of this new genomics study is to show whether screening patients for the presence of circulating tumor DNA, known as ctDNA, can successfully detect breast cancer using a blood draw.

Breast cancer is the second-leading cause of cancer deaths in women, behind only lung cancer, with an estimated 40,610 deaths each year from the disease. Nearly 253,000 new cases of invasive breast cancer are diagnosed each year, along with about 60,000 non-invasive, early-stage cases, according to the American Cancer Society.

The Intermountain study is unique in that researchers will also help develop a specific test to check for ctDNA, and will have access to both mammography results and the DNA blood test results, which will allow a direct comparison of the “liquid-based biopsy” to be made.

The idea behind the science is simple, though researchers say the execution is not yet proven: Little pieces of DNA that come from dying cells end up in the peripheral blood stream, including circulating tumor cells. The goal of researchers is to use those markers to identify breast cancer, perhaps even before mammography can detect it, said Lincoln Nadauld, MD, PhD, co-lead investigator of the study and executive director of the Intermountain Healthcare Precision Genomics Program.

“As a tumor is growing, some of the cells will die and their DNA will end up in the peripheral blood stream,” Dr. Nadauld said. “We’re able to distinguish DNA from cancer vs. DNA from normal cells. The idea is to leverage DNA to see if we can detect that it comes from a tumor.”

In the study, patients with known breast cancer will be compared with those in a screening group.

“We don’t know what we’ll see yet,” said Brett Parkinson, MD, co-lead investigator of the study, who is also imaging director and medical director of the Intermountain Medical Center Breast Care Center in Murray. “We might find those who have breast cancer will have a negative blood test and learn it’s not a good screening tool.”

Even a successful blood test isn’t expected to replace mammography outright. If it detects the circulating tumor DNA, imaging would be needed to find the tumor. But it could help eliminate unneeded biopsies, Dr. Parkinson added.

Dr. Nadauld said cancers have mutations in their DNA that aren’t always unique.

“Sometimes those are the same whether it’s a breast cancer or a colon cancer. If we do create a blood test, it’s possible it would detect mutant DNA, but it might look so similar it would be hard to tell what kind of cancer it came from,” he said. “That’s part of what this trial is going to accomplish. We want to determine the signature for early breast cancer.”

If successful, a liquid biopsy might also be used to monitor a breast cancer survivor for recurrence, Dr. Nadauld said. It might even lead to development of similar tests for different types of cancer. But that would be a challenge for the future.

“We want to approach this with laser-like focus,” he said. “It’s needed to help us diagnose breast cancer. We need to detect it earlier, when it’s curable.”

Breast cancer survival depends largely on finding the disease early —and mammography is the only screening exam that’s been shown by multiple randomized clinical trials to reduce the mortality rate for breast cancer. Since 1991, the death rate from breast cancer is down 38 percent, largely because mammography screening tests lead to early detection.

Although mammography finds most breast cancers, it may not detect malignancy in women who have dense breast tissue, especially premenopausal women, or those under 50.

“We pick up most breast cancer in women with average breast density,” said Dr. Parkinson. “When breast tissue is denser, we can miss up to 30 percent of breast cancers.”

Mammography also has a false-positive or call-back rate of 10 percent, which may subject women to additional imaging and emotional duress. Plus, a mammogram can be uncomfortable, since breast tissue is compressed for imaging, which also exposes a woman to a small amount of radiation. Mammography may also be inconvenient, often requiring women to take time off work, he noted.

For those, and perhaps other reasons, mammography screening rates in the United States are low. In Utah, only about 65 percent of eligible women are screened, despite Intermountain Healthcare’s recommendations that women over 40 undergo yearly screening mammography. All major medical and advocacy organizations agree that screening every year after a woman is 40 saves more lives. About 20 percent of breast cancers occur in women under 50.

Dr. Nadauld said the unusual confluence of three factors weigh in Intermountain’s favor on this quest, starting with access to a lot of patients in one place who are getting mammograms, which are the gold standard screening test for breast cancer. Second, the researchers have access to the results of those mammograms; they know if the results were positive or negative. The third major factor is Intermountain’s genomic technology capability.

“This is the big conversation right now in all of oncology — the use of liquid biopsy to determine how to screen for breast cancer, a woman’s risk of recurrence, and how to monitor their treatment,” Dr. Nadauld said.

The study is being made possible by a generous donation from the Beesley Family Foundation.

For Normal Heart Function, Look Beyond The Genes

Researchers have shown that when parts of a genome known as enhancers are missing, the heart works abnormally, a finding that bolsters the importance of DNA segments once considered “junk” because they do not code for specific proteins.

The team, led by scientists at the Department of Energy’s Lawrence Berkeley National Laboratory (Berkeley Lab), examined the role of two heart enhancers in the mouse genome, showing that the loss of either one resulted in symptoms that resemble human cardiomyopathy, a disease in which the heart muscle often becomes enlarged or rigid. In humans, the disease often leads to heart failure.

The findings appear in a study to be published Oct. 5 in the journal Nature Communications.

In that same paper, the researchers provided a comprehensive genome-wide map of more than 80,000 enhancers considered relevant to human heart development and function. The two heart enhancers that they tested were the mouse equivalent of enhancers chosen from among that catalog.

“The cardiac changes that we observed in knockout mice lacking these enhancers highlight the role of noncoding sequences in processes that are important in human disease,” said study co-senior author Axel Visel, senior staff scientist and one of three lead researchers at the Mammalian Functional Genomics Laboratory, part of Berkeley Lab’s Environmental Genomics and Systems Biology (EGSB) Division. “Identifying and interpreting sequence changes affecting noncoding sequences is increasingly a challenge in human genetics. The genome-wide catalog of heart enhancers provided through this study will facilitate the interpretation of human genetic data sets.”

Study lead author Diane Dickel, project scientist, and co-senior author Len Pennacchio, senior staff scientist, both work with Visel at Berkeley Lab’s Mammalian Functional Genomics Laboratory.

DNA Dark Matter

When scientists sequenced the human genome, they discovered that less than 5 percent of our DNA were genes that actually coded for protein sequences. The biological functions of the noncoding portions of the genome were unclear.

Over the past fifteen years, however, there has been a growing appreciation for the importance of these noncoding regions, thanks in large part to the efforts of individual labs and, more recently, large international efforts such as the Encyclopedia of DNA Elements (ENCODE) project.

What became clear from this work is that there are many elements of the genome, including enhancers, that are involved in regulating gene expression, even though they do not encode for proteins directly.

This realization meant that there were vast sections of the genome that needed to be explored and understood. Dickel noted that there are about 20,000 genes in the mouse genome, and in many cases, scientists have a fairly good understanding of what will happen if any one of them is disabled. In contrast, there are 80,000 candidate heart enhancers in the human genome, and it is still unclear how important they are for human development.

“In genetic studies, the way you establish whether a gene is important is you delete it from the genome and see what happens,” said Dickel. “In many cases, there are genes that, if disabled, make it difficult for the organism to survive. For enhancers, it’s less known what the consequences are if they are damaged or missing. To use a car analogy, if we took the battery out of a car, it wouldn’t start. That’s a critical component. A missing or damaged enhancer could be essential like a battery, or more similar to a missing passenger seat in the car. It’s not as nice, but it’s still possible to drive the car.”

Mapping and Testing the Enhancers

To assess the function of heart enhancers, the researchers first compiled a single road map to guide them. They used results from a technology called ChIP-seq (chromatin immunoprecipitation sequencing) to identify the likely heart enhancers in the human genome.

The researchers say this map will become an important tool as advances in genomics usher in a new era of personalized medicine.

“This compendium of human heart enhancers will be a valuable resource for many disease researchers who have begun adopting whole genome sequencing of patients to look for disease-causing mutations in both the coding and noncoding portion of the genome,” said Dickel.

Using the map, the researchers picked two enhancers located near genes associated with human heart disease. They then determined their equivalent enhancers on the mouse genome and disabled them in mice.

They compared the mice with the disabled enhancers with control mice that had no mutation and saw very large changes in gene expression in the test mice.

Echocardiograms used to image the hearts from the two groups of mice confirmed that the heart tissue of mice with a disabled enhancer was pumping with less power than normal, consistent with the signs of human cardiomyopathy.

“Prior to this work, no study had looked at what happens to heart function as a result of knocking out the heart enhancers in the genome,” said Dickel. “What was surprising to me was that outwardly, the knockout mice seemed fine. If you just looked at them, you wouldn’t necessarily see anything wrong.”

With so many enhancers to test, the map could help scientists prioritize which ones to assess in animal studies and in disease research, the researchers said.

Spotting Dna Repair Genes Gone Awry

Researchers led by Ludwig Cancer Research scientist Richard Kolodner have developed a new technique for sussing out the genes responsible for helping repair DNA damage that, if left unchecked, can lead to certain cancers.

Genome instability suppressing (GIS) genes play an important role in correcting DNA damage involving the improper copying or reshuffling of large sections of chromosomes. Called gross chromosomal rearrangements, or GCRs, these structural errors can disrupt gene order or even result in an abnormal number of chromosomes.

“Mutated GIS genes have long been suspected of playing a role in the development of many types of cancers, but identifying them has been difficult due in large part to a lack of comprehensive GCR tests, or assays, in mammalian systems,” said Christopher Putnam, an associate investigator at the Ludwig Institute for Cancer Research, San Diego and one of the first authors of the study.

In the current issue of the journal Nature Communications, Putnam, Kolodner and their colleagues describe a novel two-pronged approach that combines methods from genetics and bioinformatics to identify GIS genes—first in yeast and then in humans.

“This is one of the first large-scale studies to integrate these two methods,” said co-senior author Sandro de Souza, a professor of Bioinformatics at the Federal University of Rio Grande do Norte’s Brain Institute in Brazil, who received Ludwig support for this study.

In the first step, the scientists used assays and technologies developed by Kolodner—who is director of the Ludwig Institute for Cancer Research, San Diego and a distinguished professor of Cellular and Molecular Medicine at the University of California, San Diego, School of Medicine—and his lab to screen thousands of mutant yeast strains for genes that suppress GCRs. They identified 182 GIS genes, 98 of which had not been described before. “Ours is probably one of the most comprehensive lists of GIS genes in yeast to date,” Putnam said.

The team also uncovered more than 400 previously unknown cooperating Genome Instability Suppressing genes (cGIS) genes, which only affect genome stability when combined with other mutations. “Before our experiment, only a few dozen cGIS genes were known. Now we know of hundreds,” said Putnam, who is also an adjunct assistant professor of Medicine at UC San Diego School of Medicine.

“These results have highlighted the complex genetic network that maintains genome integrity in normal cells,” said first author Anjana Srivatsan, a postdoctoral fellow in Kolodner’s lab at the Ludwig San Diego Branch and UC San Diego School of Medicine.

To determine how many of the yeast GIS genes had human counterparts implicated in cancers, the researchers searched The Cancer Genome Atlas (TCGA)—a compilation of genomic data from thousands of patients—for such human gene homologues.

They also supplemented their candidate list with human genes that are not found in yeast but that participate in the same pathways and protein complexes as the yeast GIS genes. “We didn’t want to just look for the human equivalent of yeast GIS genes because there are human GIS genes that don’t have yeast homologs,” Putnam said.

Three cancers in the TCGA were selected for screening: ovarian cancer, colorectal cancer and acute myeloid leukemia. The scientists hypothesized that a greater number of GIS gene defects should be implicated in ovarian cancer and colorectal cancer because these two cancers tend to involve numerous large-scale rearrangements of the genome. Leukemia served as an important control because it is a cancer with little genome instability, and thus should not involve any GIS gene defects.

As expected, the team found that 93% of ovarian cancers and 66% of colorectal cancers had genetic defects affecting one or more of the predicted GIS genes, whereas acute myeloid leukemia did not appear to have defects involving GIS genes.

The researchers are already screening more than a dozen other human cancers in the TCGA for GIS gene defects. “Understanding this process allows us to think more about how carcinogenesis proceeds and it might give us insights into defects that could be therapeutically actionable in the future,” said Putnam.