Thanks to the genomic revolution begun by the Human Genome Project, we now hold the secrets of cancer’s DNA in our hands. But genomics alone does not offer enough pieces to complete the puzzle of personalized medicine for cancer. New research published Wednesday in the journal Nature demonstrates the power of filling in those missing pieces of cancer biology ― proteins.
The first large-scale study of its kind in breast cancer, the research demonstrates how proteomics, or studying all the proteins in a given cell, can reveal which of the many mutations in a tumor are actually driving a cancer’s development and suggest new targets for personalized cancer-killing drugs.
“Not all of those mutations at the DNA level are translated forward to the protein level where they actually affect the cancer cell,” said Dr. Amanda (Mandy) Paulovich, one of the study’s leaders and an oncologist and cancer geneticist at Fred Hutchinson Cancer Research Center. “So what proteomics does is add another layer of clarity as to which of those DNA changes might be functionally important and worth targeting for new treatments in patients.”
The research team is part of a large consortium funded by the National Cancer Institute, called the Clinical Proteomic Tumor Analysis Consortium, or CPTAC, which is dedicated to accelerating cancer research through proteomics research and technology development. The breast cancer team was co-led by Paulovich, Dr. Matthew Ellis of Baylor College of Medicine and Dr. Steven Carr of the Broad Institute.
“By using proteomics integrated with genomics, we hope to understand at a mechanistic level the basis of cancer progression, metastasis and resistance to therapy,” said Carr, director of the Broad’s Proteomics Platform, about his ultimate goals for cancer proteomics. “In addition, we hope to identify new, aberrantly activated proteins, especially kinases and other enzymes … that are potential new therapeutic targets.”
Proteins carry out most of the cells’ activities and are critical components in the structures of our organs and tissues. If you take a medication for any reason, chances are that drug acts on a protein to do its job. And although changes to DNA are ultimately at cancer’s root, it’s proteins that directly drive cancerous cells’ activity.
Speaking broadly, your DNA codes for your RNA, which codes for your proteins. According to that simple view, knowing coding sequences in an organism’s DNA should tell you about its proteins as well. But reality is much more complicated; vast molecular networks in cells control which genes get transcribed into RNA as well as which of the RNA sequences are translated into proteins, and the activity of the proteins themselves.
In short: Genetic sequences alone just give you a small peek at a cell’s ultimate biology. And that’s a problem when you’re designing a cancer therapy targeting a specific aspect of cellular biology.
“A lot of effort and resources over the past decade have gone into characterizing human cancers at the level of genomics ― meaning DNA and RNA characterization. But very little effort has gone into characterizing the proteome,” Paulovich said. “Our genes are the blueprints for making our proteins, but it’s the proteins that ultimately carry out the functions of our cells, and it’s the proteins that are the targets of our drugs.”
The new breast cancer paper published by Paulovich and her colleagues is one of a set of large pilot projects the consortium is undertaking in three cancer types ― breast, colorectal and ovarian cancers. In this set of projects, the researchers are adding a layer of proteomic data to tumors that have been genomically characterized in the National Institute of Health’s The Cancer Genome Atlas, or TCGA, initiative to learn how proteomic data can contribute additional insights not offered by studies of the cancers’ genomes.
“The Cancer Genome Atlas project has informed many aspects of cancer research including identifying potential new, previously unrecognized targets for drug therapy and defining new ways to classify and identify cancer subtypes,” study co-leader Carr said. “But what effect the bewildering variety of genomic mutations have on a patient’s outcome or their response or resistance to therapy remains largely unknown. Proteins are the workhorses of the cell and the end products of the genome. A natural next step in understanding cancer is to probe the proteome of tumors to identify which proteins are changing and how.”
In this study, the researchers employed a powerful, specialized technology called mass spectrometry to characterize the proteomes of 77 breast tumors in TCGA. Using mass spectrometry, scientists can identify proteins in a sample based on the mass and electrical charge of the proteins’ fragments. This methodology is a far more powerful way of characterizing proteins than the traditional methods employed by most research labs. The researchers quantified more than 12,000 proteins in the tumors, plus 33,000 sites on the proteins that are modified as the proteome transmits cell signals — modifications that are not detectable with genomics.
This study was designed to give hypothesis-generating (rather than definitive) findings, but it is clear from this and the team’s other pilot studies that “with a few exceptions, genomic profiles are not accurate predictors of protein levels or activities,” Paulovich said.
The study also suggested proteins whose activity in cancer was previously underappreciated that could be targets for new drugs.
“The evidence is unequivocal now that proteomics can fill in missing biology missed by genomics,” Paulovich said.
The CPTAC team is currently validating their results from this research to gain certainty on the specific proteomic pathways activated in these types of tumors. They are also expanding their research to encompass new tumor types. All of these studies will use tumor samples that are collected using proteomics-optimized methods and that are linked to detailed clinical information, allowing the investigators to see how changes in the proteome may affect a patient’s treatment outcomes.
This work is an “extremely important and necessary” step toward the development of fully personalized cancer care, said Fred Hutch breast cancer oncologist Dr. Gary Lyman, who was not involved in this study. Lyman recently helped to create recommendations for the development and implementation of molecular tests to guide personalized medicine.
Currently, particular subsets of breast cancer patients ― for example, the minority with an overproduction of a growth-promoting protein called HER2 ― have an effective, targeted therapy available to them. Lyman’s hope is that these studies by the CPTAC collaborators will lay the groundwork for the eventual development of additional targeted therapies that benefit more patients.
The development of HER2-targeting therapy “was a game changer. It took a disease that was absolutely the worst breast cancer type to get, to what many would argue — myself included — is now the best kind of have,” Lyman said. “If we could just do the same with these other intrinsic subtypes of breast cancer ― by identifying targets, getting good assays for those targets, and developing therapies … to target those vulnerable targets ― I’m very confident over the next several years we’re going to revolutionize [the care of] the vast majority of patients with breast cancer, and I think other tumors.”
Toward this goal, CPTAC’s next phase will provide Paulovich and other proteomics experts with the opportunity to integrate their expertise into the clinical development of new therapies, to better understand and predict tumor responses to drugs. The plan is for proteomics experts to partner with scientists who are leading National Cancer Institute-funded clinical trials of new cancer treatments.
“We hope that the proteomics can complement genomics and improve our ability to select the right drugs for the right patients,” Paulovich said.
Join the conversation. Talk about this story on Facebook.
Susan Keown is a staff writer at Fred Hutchinson Cancer Research Center. Before joining Fred Hutch in 2014, Susan wrote about health and research topics for a variety of research institutions, including the National Institutes of Health and the Centers for Disease Control and Prevention.