© iQoncept - Fotolia.com
In the history of cancer research, no finding has had more far-reaching impact than the discovery of cancer-causing genes, or oncogenes. These are normal cellular genes whose genetic code has become scrambled, leading to abnormal cell behavior. The discovery of oncogenes helped to solidify the notion that cancer is a genetic disease, and opened the door to targeted treatment.
The first oncogene, called Src, was identified in the mid-1970s by two scientists who later earned a Nobel Prize for their work. Since then, many oncogenes have been discovered and the modern field of cancer research essentially revolves around understanding what these genes do and targeting them with drugs. But how exactly do oncogenes cause cancer?
Ten years ago, if you had asked a cancer biologist to describe what an oncogene did, she’d have had no trouble listing a familiar set of traits that describe the wayward behavior of cancerous cells: increased proliferation, resistance to cell death, and a tendency to metastasize, or spread. One thing that would not have been on the list is anything having to do with the immune system. Indeed, nothing on the list would have had much to do with anything happening outside the cancer cell itself.
In recent years, with the discovery that the immune system can be harnessed to recognize and fight cancer, the way that researchers think about the causes of cancer is undergoing a profound shift. No longer is the focus just on the changes that occur inside the cancer cell to alter that cell’s behavior. Increasingly, scientists are looking at ways that cancer cells alter their surrounding environment—including the immune cells in their midst—with the result that certain fundamental cancer concepts are getting a substantial reboot.
Redrawing the Diagrams
When Dafna Bar-Sagi, Ph.D., lectures to students about her area of expertise—the oncogene called RAS—she often begins with a slide that illustrates what we know about the protein made from this gene. RAS is a protein that sits just inside the cell membrane of cells and acts as a kind of molecular telegraph switch, sending signals to a downstream chain of proteins that converge on the nucleus to alter gene transcription.
“All the arrows are pointing down into the cell, and there’s absolutely nothing pointing outside of the cell toward whatever it is that is out there—including the immune cells,” says Bar-Sagi, a professor of biochemistry and molecular pharmacology at NYU School of Medicine.
That’s mostly a legacy of how oncogenes were discovered and studied for many years afterward. Like nearly all oncogenes, RAS was discovered first as a gene contained in a virus known to cause cancer—in this case, one causing sarcoma in rats. Viruses have picked up these genes from the cells they infect, and the genes have become incorporated into the viral genome. When the virus infects a new host, it inserts the oncogene into the host’s DNA.
But viruses are not the only—or even the most common—source of oncogenes; their precursors are naturally found in cells, where they control normal cell behavior. These genes only become oncogenic, or cancer-causing, when the DNA sequence is changed through the process of mutation.
Definitive proof that RAS was an oncogene came from experiments conducted in 1982 by Robert Weinberg at MIT, who cut out the RAS gene from a human cancer cell and transferred it into a mouse skin cell. The transfer of this one altered gene was enough to turn the skin cell cancerous. Weinberg was also able to pinpoint the precise genetic change that converted the normal RAS gene into the cancer-causing oncogene: a single nucleotide base substitution was all it took.
These important discoveries kicked off a decades-long search to understand what RAS was doing in the cell. And, indeed, much of what we now know about the process of “signal transduction” in cells comes from studies of RAS. (RAS is what’s known as a GTPase, a kind of molecular switch; in the mutated version of RAS, the switch is permanently stuck in the “on” position.)
But importantly, almost all of this research was conducted on cells growing in a dish, rather than on cells growing inside a living organism. It turns out that the main advantage of cell culture—its relative simplicity—is also its major downside when it comes to understanding precisely what mutant RAS proteins are doing inside the body.
Bar-Sagi began to realize that our understanding of RAS was inadequate when, in 2004, she discovered that tumor cells with mutant RAS produce abundant quantities of an immune molecule—a cytokine called interleukin-8. “We knew then that RAS has the capacity to produce signals that will affect the immune system,” Bar-Sagi says.
“The immune inflammatory response is just unbelievable.” —Dafna Bar-Sagi
From there, the quest was on to understand more fully what RAS was doing outside the cancer cell. In 2012, Bar-Sagi and a CRI-funded post-doc in her lab, Yuliya Pylayeva-Gupta, Ph.D. (pictured left), performed an experiment in which they put cells containing mutant RAS into a normal mouse pancreas. RAS mutations are found in nearly 100% of pancreatic tumors, so the team wanted to find out what this mutant RAS was doing. What they found was that the cells with mutant RAS caused a vigorous inflammatory response; immune cells poured into the area.
“The immune inflammatory response is just unbelievable,” says Bar-Sagi. “It’s really very, very striking.”
That told the team that RAS is producing something that is attracting the immune cells. They eventually hit upon another cytokine, called GM-CSF, as the culprit. GM-CSF stands for granulocyte-macrophage colony stimulating factor.
“What happens when the cancer cells start secreting this cytokine is that there is a mobilization of myeloid derived suppressor cells toward the tumor,” explains Pylayeva-Gupta, now a professor at the University of North Carolina School of Medicine. Myeloid derived suppressor cells, or MDSCs as they are called for short, are immune cells that serve a regulatory role, tamping down the responses of the immune system’s attack cells, the killer T cells.
This observation led the team to investigate what happens when you get rid of GM-CSF in the tumor. They found that this causes a decrease in the number of MDSCs, an increase in T cell responses to the tumor, and a significant reduction in tumor cell growth. The team published these results in Cancer Cell in 2012.
Here was a good example of an oncogene leading directly to strong suppression of the immune response. And since more than 90% of pancreatic tumors have a RAS mutation, this may help to explain why pancreatic cancer has so far been resistant to existing immunotherapies—and why it remains such a deadly killer, with one of the lowest 5-year survival rates of any cancer.
Based on these and similar results obtained from other groups, scientists are now trying to figure out ways to overcome this dominant immunosuppression, so that an immune response against pancreatic cancer can finally take off.
Responders and Non-responders
Compared to pancreatic cancer, certain other types of cancer are much more responsive to existing immunotherapies. Approximately 22% of melanoma patients treated with the immunotherapy drug ipilimumab, for example, experience durable remissions and are effectively cured of their disease. Patients who receive another immune drug called nivolumab, or a combination of these drugs, have even higher response rates. That said, not everyone who receives these drugs responds to them, raising the question of why. Recent research from CRI scientists suggests that oncogenes may be partly to blame.
When pathologists look at tumor samples from melanoma patients being treated with immunotherapies, they find that the tumors fall into two general categories that correlate with treatment response. Tumors with lots of T cells in and around the tumor—called inflamed tumors—tend to be found in patients who are having a good response to treatment. Tumors with few immune cells—called non-inflamed—are found in patients who are not responding.
Stefani Spranger, Ph.D., a CRI-funded post-doc at the University of Chicago, wanted to understand the molecular basis for this difference. As a starting point, she first compared the expression of genes between the two groups of patients and found that the difference between inflamed and non-inflamed tumors was linked to an underlying genetic difference in the production of an oncogene called beta-catenin. Tumors that make beta-catenin lack T cells.
Spranger next wanted to unravel the mechanisms of this association, so she turned to a mouse model of melanoma in which she could control exactly which genes were added into the mix. “When we added on the beta-catenin, the T cells were basically absent,” she says.
In addition to T cells, the tumors also lacked dendritic cells, which are important for starting immune responses. This latter finding helps to explain why T cells are not found in the tumors—they’re not being activated.
“The dendritic cells are actually the pacemaker of the entire reaction.” —Stefani Spranger
Intriguingly, when the team put the dendritic cells back in, the T cells returned as well. “The dendritic cells are actually the pacemaker of the entire reaction,” Spranger says.
The team recently published their results in the journal Nature, and they are now looking at whether beta-catenin is driving exclusion of immune cells in other tumor types.
Pumping the Immune Brake
Of all the various immunotherapies in use, none has garnered as much attention as checkpoint inhibitors targeting the PD-1/PD-L1 pathway. PD-1 is a braking molecule, or checkpoint, found on T cells. When activated by its binding partner, PD-L1, found on some tumors and immune cells, this braking molecule shuts down the immune response.
PD-1 checkpoint inhibitors have been remarkably effective in treating several types of cancer—including melanoma, lung, kidney, and bladder cancer. Results from clinical trials suggest that patients with higher levels of PD-L1 in their tumors may respond better to PD-1/PD-L1 checkpoint inhibitors than those with lower levels of PD-L1. Which raises the question: what is driving the expression of PD-L1 in tumors in the first place?
A 2013 paper published in the journal Cancer Discovery by CRI Scientific Advisory Council associate director Glenn Dranoff, M.D., argues that—for some types of cancer, at least—mutations in another oncogene, called epidermal growth factor receptor (EGFR), may be involved. EGFR mutations were already known to be associated with certain types of lung cancer, which also makes them good target for EGFR-inhibitors, like the drug erlotinib (Tarceva®). What Dranoff and colleagues found in laboratory studies was that EGFR mutations cause lung cancer tumors to make more PD-L1, which effectively protects them from immune attack. This link between EGFR mutations and PD-L1 expression in tumors may be the clearest link yet between oncogenes and immune evasion.
The result has important therapeutic implications. As Robert Vonderheide, M.D., D.Phil., of the University of Pennsylvania and CRI’s Scientific Advisory Council noted in an accompanying commentary, with this knowledge, it becomes imperative to integrate analyses of genetic markers into immunotherapy treatment considerations. Combining EGFR-targeted therapies with immunotherapies could be an effective way to treat these patients.
In addition to EGFR, RAS, and beta-catenin, several other common oncogenes have recently been linked to immune dysfunction, including MYC, PTEN, and BCR-ABL. All told, these represent some of the major oncogenic drivers of cancer. The implication? No longer can cancer be viewed simply as the result of altered cell division; the immune system is deeply implicated, too.
“Oncogenes will always be oncogenes,” says Vonderheide, “but their immune face is just emerging.”
Putting the ‘Cancer’ in Cancer Immunology
For many years, the only effective cancer treatments were ones being offered by traditional oncologists—chemotherapy, radiation, and targeted therapies. These weapons all target cancer cells specifically and emerged out of basic research being conducted by cancer biologists.
Cancer immunologists—scientists who study the relationship between the immune system and cancer—had long suggested that there was a better way to treat the disease. But up until recently, they did not have much in the way of effective treatments to offer in support of their claims, and were therefore easily dismissed.
Now that effective immunotherapies have been FDA approved to treat a variety of cancers, traditional cancer biologists can no longer ignore the findings of cancer immunologists. But there is still a lag in the flow of information between the two groups.
“A lot of cancer biologists still think of the tumor cell in a bubble by itself,” says Spranger. “But the tumor cell has to escape the immune system before it can cause a tumor. Shutting down the immune system is half of the full story.”
What we’re seeing with these recent studies may be a kind of rapprochement between the two fields. Cancer biologists are learning to see the immune system as a crucial part of cancer development and spread, and cancer immunologists are learning how to integrate the genetic knowledge obtained by cancer biologists into their studies.
The time is right for this integration, says Penn’s Vonderheide. “We need, as a field, to put the ‘cancer’ into cancer immunology. No better place than the oncogene.”