The Washington University Genome Center
, started in 1993, has always been a leader in innovative genome sequencing, from primitive bacteria to the complex human blueprint. The center helped lead the way in the Human Genome Project and currently is a principal participant in both the Cancer Genome Atlas (TCGA) project and the 1000 Genomes Project. Yet the greatest recognition came in late 2008 when the center became the first to sequence and decode the complete DNA of a human cancer genome, tracing acute myelogenous leukemia (AML) back to its genetic roots.
A genome is a blueprint or map of an individual person, but it’s in a language we don’t yet understand. It contains our genetic makeup, or about 3 billion chemical base pairs, which code for everything from hair color and height to how specific organs function. Genetic errors, or mutations, are known to multiply and accumulate in normal cells, leading to an alteration that can eventually result in cancer.
“We want to know why a difference, or even similarities, exist,” says Richard K. Wilson, PhD, director of the Genome Center. “One genome may differ from another minutely, but that difference can have huge implications in how the body functions. Leukemia was the first opportunity to look at both the cancer and healthy cells from the same person and determine exactly what differences were present.”
Past: The impossible
A decade ago, Wilson and Timothy Ley, MD, Washington University hematologist and the Alan A. and Edith L. Wolff Professor of Medicine, knew cancer was a good target for sequencing and started planning. Around two years ago, they started the first project.
“We were able to be the first because we were ready, we had the DNA samples in our database, we have the technology, and we were just waiting for it all to fit together,” Ley says. “It was a perfect storm so to speak, but an intentional one.”
Due to its nature, the researchers selected AML as the best place to start. The disease consists of multiple tumor cells, and because it is a marrow-based cancer, there’s no need to remove a tumor mass. It does not completely destroy the DNA, leaving comparable information for study. And, with an estimated 13,000 cases diagnosed in the United States each year and 8,800 patients dying from the disease, it is a high significance to patients and scientists alike.
The importance of completing the entire genome lies in the complexity of human DNA and the mystery that still surrounds cancer. A genome contains both the genes and non-coding sequences of DNA.
“If there’s a one-in-a-million variance from one genome to the next, you wouldn’t know if it’s relevant unless you did the entire genome and compared it to another genome from that same host.” Funding was also an issue with the initial project. A few years ago, one complete sequencing would have cost $100 million, and it still didn’t lend enough understanding for a significant analysis. Through gifts from philanthropist Alvin Siteman, as well as the Barnes-Jewish Hospital Foundation and others, the AML venture got off the ground.
The cost has since decreased significantly, and sequencing is becoming both more informative and affordable for multiple projects.
Present: Breaking the barrier
In short, scientists get a patient’s DNA through bone marrow samples and chop it up into little segments. “It’s partly a biochemistry and partly a statistics problem,” Wilson says. “Take a 10,000 piece jigsaw puzzle, throw it onto the floor and now put it back together, but some of the pieces may be missing or changed.”
The DNA fragments are then placed on small glass slides that hold up to 150 million base pairs of different DNA fragments each, and using florescent nucleotides, computers illuminate and photograph the data. They then take the data – DNA sequences from both the tumor and normal cells – and compare the two to discover the changes that have arisen in the tumor genome.
Some differences will be one in a thousand and aren’t noteworthy. But when a mutation changes the amino acid sequence of a protein gene, it may signal something significant. Biochemistry is a big part of the process, but analyzing and making sense of the information is more. Washington University scientists had to make their own software to help decipher the sequence, called the analysis pipeline, which was a trial-and-error process in itself.
The importance of the first project was breaking the barrier. “We didn’t even know if it could be done,” says Ley. “We feared a massive number of changes from the cancerous genome, or what we call passenger mutations. Researchers once thought all cancer genomes were unstable. Now with our finding of so few differences, we know this is not true.”
The complete genome revealed many layers of sequences never before seen or analyzed. In the process of analyzing, the data also found 10 mutations, only two of which had been previously linked to AML. “So we had to ask ourselves, ‘Are the others relevant?’” Wilson says. “And the answer is yes; it just underscores how much we are still learning.” It also proved the value of complete sequencing versus analyzing small sections of the genome.
Although the first complete genome project took 10 years to complete, the second went much faster with more comprehensive results. The team recently finished the second AML genome in January, and the third started subsequently.
It is still expensive to sequence whole genomes, so many projects focus on the regions already known to hint at significant clues. The paradigm is shifting, and the Siteman Cancer Center and Washington University are right in the middle of it.
Future: Personalized medicine
Before finding a cure, physicians must find out how to better treat and manage cancer. That future, however, is not far off. “By the time we have 100 plus genomes completed in the next few years, we will find the markers,” Wilson says. “We will be able to tell how a patient will react to a medication, not just how a cell reacts. We’ll be able to say, ‘You have gene X, this is the likely outcome and this is what we’ll give to treat it.’ And it will happen sooner than you can imagine.”
Wilson also believes medicine will get to the point where a cancer patient’s initial workup will include comprehensive DNA analysis. Researchers’ work to understand the genetics of cancer, however, will be an ongoing process.
“We will eventually solve the rules of cancer: why it develops, how to respond, what is a good therapy for a particular individual, is it likely to relapse,” Ley says. “Yet the fundamentals of today will need to be done forever since biology is always changing. Unfortunately, one side effect to more knowledge is that distinct cancer classifications will increase because we’ll be able to specify among previously unknown subtypes. Once we know thousands, we’ll only have a glimpse of understanding.”
A Personal Touch
The first step in sequencing acute myelogenous leukemia (AML) genomes involves obtaining tissue samples from patients with the disease. That work is completed by research patient coordinator Sharon Heath, who began working at the Siteman Cancer Center in March 2000. She has had cancer strike on both sides of her family, which motivated her to join in the fight against the disease.
Since 2002, Heath has been dedicated solely to the AML sequencing project. She makes her rounds daily, reviewing new files from campus. If a suitable patient is identified, Heath then initiates the groundwork for patient consent and subsequent bone marrow donation which contains the necessary DNA for sequencing.
She also makes sure every sample is anonymous, then delivered to the tissue procurement facility while still reviewing the patient’s medical history and following it until the end. “I’ve witnessed marrow
biopsies, held hands and talked to patients through their procedures,” Heath says.
To date, she has acquired over 1,000 DNA samples.
At any given time, there may be approximately 130 patients being followed. “I love to hear when patients are doing well. I am comforted in knowing that their important donation of tissue could change this job in the near future.”
The Future is Now
The idea of personalized medicine is already becoming a reality with some forms of cancer due to partial genome sequencing at the Siteman Cancer Center.
The genetics of chronic myeloid leukemia (CML, a slower-growing cancer that may take years to progress, unlike acute leukemia) are now well-understood, and certain forms of the disease are being treated with specialized tyrosine kinase inhibitors (TKI). “We knew the cause 20 years ago so we were able to develop a
drug for that cause,” Ley says.
In glioblastoma, one of the most frequent and aggressive types of primary brain tumors, the standard treatment seems to make the symptoms worse in some patients. Through sequencing, scientists can tell that in people with one gene mutation, the treatment works well. But with other mutations, the treatment reacts poorly and does make patients sicker. If doctors know this ahead of time, other treatments may
In lung cancer, TKI chemotherapy has an effect in a small number of patients, but it is significant enough that the drug was approved by the FDA. Experts are now able to tell if it will work or not by looking at one gene.
Matt Ellis, MD, PhD, the Anheuser-Busch Endowed Professor in Medical Oncology, is making progress in developing personalized treatments for breast cancer.