As it faces off against the body, cancer cheats. New research investigates one of its more audacious tricks: stashing cancer-driving genes in circular pieces of DNA outside chromosomes.
A trio of new studies with funding from Cancer Grand Challengesexternal link, opens in a new tab reveals the unfair advantage this extrachromosomal DNA, or ecDNA, grants tumors. Physician-scientist Howard Y. Chang, a Howard Hughes Medical Institute Investigator, co-led two of the studies along with Paul Mischelexternal link, opens in a new tab, a fellow Stanford University colleague who served as a lead on all three papersexternal link, opens in a new tab published on November 6, 2024.
“There are rules normal cells follow. One of the ways cancer wins is that it doesn’t play by the same rules,” Chang says. “What we are doing is figuring out what extra moves it has in its repertoire.”
Normal human cells generally keep their DNA in chromosomes. But that’s hardly the only way ecDNA disregards norms. Research by Chang and Mischel’s team delineates how it violates one of the fundamental principles that govern inheritance to aid tumors’ growth and make them more resilient. Together with international collaborators, they show the scope of ecDNA’s contribution to cancer, especially to the most pernicious cases. The group also reports a strategy for developing therapies that target ecDNA in order to outmaneuver the tumors that contain it.
All three papers are the work of Team eDyNAmiC – led by Mischel – and its international collaborators to study ecDNA from a variety of perspectives. Through Cancer Grand Challenges, Team eDyNAmiC is funded by Cancer Research UK and the US National Cancer Institute, with support to Cancer Research UK from Emerson Collective and The Kamini and Vindi Banga Family Trust.
‘Pound for pound’
On its own, ecDNA is nothing new. Scientists in the 1960s first documented its existence as small dots of DNA outside chromosomes, which contain a cell’s DNA packaged with proteins. Until relatively recently, scientists took it for a rare and likely insignificant feature of cancer, occurring in perhaps 1.4 percent of tumors.
That perception began to change roughly a decade ago, when Mischel and his team traced the numerous copies of cancer-driving genes, or oncogenes, often found in tumors to ecDNA – not chromosomes, where they were assumed to reside.
Chang joined Mischel to study ecDNA after his analysis of cancer cells turned up a massive signal indicating that their chromatin, the DNA-protein material in chromosomes, was open and so available for gene expression. He subsequently determined that ecDNA was responsible. For every copy of a gene it possesses, “pound for pound, ecDNA is producing more genetic output,” he says.
This, the researchers have determined, is central to how ecDNA molecules contribute to cancer: Their DNA is transcribed — or turned into mRNA, an early step in gene expression — four times more than the same sequences on chromosomes.
They gain this advantage, in part, through cooperation. In research described in 2021external link, opens in a new tab, Chang, working with Mischel and others, found that ecDNAs come together to form hubs. While in close proximity, distinct ecDNA molecules interact, with expression-enhancing elements on one elevating the activity of oncogenes on separate ecDNAs.
Under the normal rules of inheritance, this toxic synergy shouldn’t persist. Yet it does. New researchexternal link, opens in a new tab by Chang and Mischel’s team shows how.
Breaking the law
When cells divide, they duplicate their chromosomes and then divvy them up evenly among their offspring. According to one of the foundational principles of modern genetics — Gregor Mendel’s third law, that of independent assortment — genes on the same chromosome are inherited together, while those on different chromosomes are distributed independently of one another.
“There are very few laws in biology, but this is one of them,” Chang says.
At the outset, the team knew that ecDNA wouldn’t behave just like chromosomes. These molecules lack centromeres, structures used to distribute chromosomes among daughter cells. So, it’s not entirely surprising that the inheritance of ecDNA is much more chaotic, even including scenarios in which offspring received more copies of an oncogene than their parent possessed.
In their experiments, Chang and Mischel’s team uncovered another, more predictable dynamic. By tagging distinct, oncogene-carrying ecDNAs in cancer cells and then watching through a microscope as the cells divided, they found that offspring tended to inherit ecDNAs together.
“When you have two different DNA molecules with a Mendelian type of inheritance, you expect a lot of cells with one molecule or the other,” Chang says. “But what we found was that most cells have both ecDNA species. So, if you have more of ecDNA flavor A, you also have more flavor B.”
On one level this result made sense: Together, the cooperating ecDNAs aid cancer cells. By keeping them together, the next generation perpetuates this benefit. The problem is that this kind of coordinated inheritance violates Mendel’s third law.
Next, Chang and Mischel’s team set out to determine what was driving this unconventional inheritance. Genes A and B could, in theory, occur together because those that receive only A, for example, face some sort of debilitating disadvantage that leads them to die off. However, computational modeling experiments favored another scenario: that ecDNA molecules somehow travel together.
Through further work, Chang and Mischel’s group found that, by inhibiting transcription, they could break up this dynamic so that the ecDNA molecules became randomly distributed among the daughter cells. This discovery makes intuitive sense: As part of their supercharged transcription, ecDNAs form physical connections between enhancers and oncogenes that could anchor them together through cell division.
“It’s quite novel, the idea that the ecDNA molecules in the cancer cells are not randomly inherited, but that there is an inherent characteristic related to transcription that helps cells inherit ecDNA together,” says Maite Huarteexternal link, opens in a new tab, a molecular biologist at the Centre for Applied Medical Research at the University of Navarra in Spain, who was not involved in the ecDNA studies.
By breaking numerous rules that govern normal cells and tissue, tumors create opportunities for themselves to escape the immune system and escape treatment. ecDNA’s coordinated inheritance feeds this dynamic by increasing the diversity of cells within the tumor, and so giving it more options for adapting under pressure, according to Huarte.
“This is a feature that accelerates the evolution of the tumor, which can be quite impactful for the malignancy,” she says.
A ‘formidable problem’
Cancer appears to employ this advantage in far more than 1.4 percent of tumors, according to an analysisexternal link, opens in a new tab in the trio of papers published on November 6. Along with Mischel, Team eDyNAmiC member Mariam Jamal-Hanjaniexternal link, opens in a new tab at University College London (UCL) Cancer Institute and Charles Swantonexternal link, opens in a new tab at the Francis Crick Institute and UCL led this work.
The group found ecDNA in just over 17 percent of whole genomes from 39 types of tumors, from nearly 15,000 patients. These circular molecules appeared in all stages of cancer, starting with their emergence. However, they were more likely to turn up in advanced tumors, including those that had metastasized to new sites within the body. Likewise, the researchers determined they were more prevalent in patients who had previously received cancer treatments, and those who survived for less time.
This study also alluded to the diversity of ways ecDNA propels cancer. In addition to oncogenes, cancer’s bread and butter, the researchers identified other malignancy-promoting elements on these molecules. For example, the team frequently found ecDNAs bearing only expression-enhancing elements, while others carried genes involved in regulating the immune system.
ecDNA, is a “formidable problem,” that contributes to poor outcomes among patients, this group writes, echoing the results of prior researchexternal link, opens in a new tab.
“There are clearly very different routes towards the development of cancer, and those routes matter in terms of the biology of disease,” says Mischel, a co-author of this study. “When the route to cancer does not involve Mendelian genetics, it changes the game.”
From advantage to liability
Knowing what they do so far, he and Chang identified a way to turn ecDNA’s essential advantage — supercharged transcription — against cancer. They test this approach in the third paperexternal link, opens in a new tab.
The underlying strategy is simple: Exploit a natural conflict between the cells’ imperative to transcribe their DNA and to duplicate it. The machinery for both processes moves along the DNA; collisions it can cause cells to die.
The team focused on an enzyme called CHK1 that can prevent such fatal conflicts by holding replication in check. In experiments, they found that inhibiting CHK1 selectively destroyed ecDNA-containing cancer cells. A CHK1 inhibitor was most effective, however, when used with another drug, infigratinib. This approach appeared to handicap tumors by suppressing their growth and preventing their cells from acquiring resistance to the drug through their ecDNA.
A biotech company that Mischel and Chang founded, Boundless Bio, is developing ecDNA-targeting therapies, including one focused on CHK1. The company is now recruiting patients with advanced solid tumors to participate in a clinical trialexternal link, opens in a new tab testing a different CHK1 inhibitor in combination with targeted therapies.
Much more to do
Team eDyNAmiC has many more studies of ecDNA in the works. Another, more preliminary project is underway in team member and HHMI Investigator Harmit Malik’s lab at the Fred Hutchinson Cancer Center, where postdoc Grant King is studying a unique variety of ecDNA found not in cancer, but in fungi.
While ecDNA does not occur in normal human cells, a particular molecule called the two-micron plasmid is common among yeast. Unlike the cancer molecules whose content varies, the plasmid always encodes four genes, which aid its propagation. At least as far as scientists can tell, it appears to provide at best limited benefits to the yeast, yet somehow the two have established a long-term relationship.
“This ecDNA has managed to co-evolve with its host such that it’s been able to propagate itself for millions and millions of years,” he says.
King wants to understand how the yeast cells maintain a stable co-existence with their plasmid passenger. This research may reveal vulnerabilities that could, one day, be exploited to fight ecDNA-containing cancer cells.
“This basic biology approach will, hopefully, produce information that’s relevant outside of yeast,” he says.
Other team members and HHMI Investigators Zhijian (James) Chen, at The University of Texas Southwestern Medical Center, and Michelle Monje, at Stanford, are investigating different questions about these molecules and their role in cancer.
As a cancer scientist herself, Huarte is looking forward to what’s to come.
“It is very exciting for us, the researchers, because we see there is so much more that we can learn and so many new opportunities for treating cancer,” she says. Huarte notes, however, that the recent discoveries about ecDNA also point to a humbling truth. “Even though we have thousands of studies of tumor genomes over the decades, there is so much that we still don’t understand.”
###
