Cancer cells are able to spread from a nascent tumor much earlier than scientists long thought and are more adept than later emigrants at forming potentially lethal metastases at distant sites such as the brain and bones, researchers reported in two studies published Wednesday.
The discovery offers the first molecular explanation of how that early spread occurs, hinting at why early detection and treatment often fail to prevent cancer deaths: “Early” is still too late.
Why it matters:
Cancer dogma holds that after cells acquire genetic mutations that make them malignant, they proliferate out of control, eventually forming a tumor. Cells that make up the tumor accumulate additional mutations, some of which enable them to move into the bloodstream, reach distant organs, and set up a malignant colony. Such metastases are responsible for more than 90 percent of cancer deaths.
But the two new studies of breast cancer suggest that model is wrong — or, at least, that it doesn’t describe all cancers. The findings “challenge everything we thought we knew about how cancer spreads and forms metastases,” said Dr. Julio Aguirre-Ghiso of the Icahn School of Medicine at Mount Sinai in New York, who led one of the new studies, published in Nature.
The discovery might shed light on some of the most frustrating mysteries of cancer.
For instance, why doesn’t early detection, such as with mammography or even more advanced methods such as breast MRI, save more lives? The metastases-forming, super-early departees leave when the tumor is microscopic. By the time a tumor is detectable, they’ve already — perhaps for years — been seeding potentially fatal metastases at distant sites, said Aguirre-Ghiso.
The new findings also might explain why targeted drugs, which are viewed as the greatest successes in precision medicine, don’t keep most patients alive much longer than standard chemotherapy. These drugs disable the molecular mechanisms that spur proliferation, which doctors identify by biopsying the tumor. But if cancer cells leave before there’s any tumor detectable enough to biopsy, attacking the original tumor is the proverbial locking the barn door after the horse has fled. By then, cells that left have had lots of time to evolve their own distinct mutations to proliferate at a distant organ.
You’ll want to know:
Earlier research had hinted that some cancer cells set out for distant organs long before a tumor is detectable. A 2008 mouse study, led by Christoph Klein of Germany’s University of Regensburg, who also led one of the new studies, found that cancer cells reached the bone marrow months before breast tumors formed. A 2011 study of 30 women with “non-invasive” breast cancer found that eight actually had cancer cells in the bone marrow. In about 8 percent of non-invasive breast cancers, a 2015 study reported, metastases develop — even though “non-invasive” means the malignant cells can’t enter the bloodstream and travel to vital organs.
The two new studies are the first to identify the molecular mechanisms by which cancer cells spread even before a tumor forms. They also offer a long-sought explanation for why some 5 percent of cancer patients have metastases but no original tumor, called “cancer of unknown primary”: Malignant cells formed metastases even before the original tumor was detectable.
The new studies were done in lab mice. The study at Regensburg found that in mice given the human breast cancer gene HER2, the hormone progesterone triggers the migration of cancer cells almost as soon as they form — that is, before a tumor is detectable. And these early émigrés are better at forming metastases than cancer cells that depart the tumor later.
At Mount Sinai, Aguirre-Ghiso found that a gene called p38 acts as a brake on the departure of cancer cells from still-forming breast tumors. When p38 is silenced and HER2 is activated, the combination awakens molecules that, eventually, mobilize cancer cells into the bloodstream and on to the lungs and bones.
But keep in mind:
It remains to be seen whether this early departure of cancer cells to seed metastases occurs in cancers other than those of the breast. Odds are, however, that it does in at least some, especially melanoma and pancreatic cancer.
What they’re saying:
In a commentary he coauthored in Nature, cancer biologist Cyrus Ghajar of Fred Hutchinson Cancer Research Center in Seattle praised the “compelling data” in the studies. One important implication for women using hormone replacement therapy, he said, involves progesterone’s ability to trigger metastasis: “This adds further evidence that post-menopausal administration of progesterone is a bad idea, as it could theoretically fuel dissemination from any breast [malignancies] that emerge later in life.”
“There’s no question this is interesting work,” said cancer biologist Tyler Jacks of the Massachusetts Institute of Technology, director of the Koch Institute for Integrative Cancer Research. By undermining “the canonical view that metastasis occurs very late in carcinogenesis,” he said, it raises questions about early detection and about matching targeted therapies to mutations in the original tumor. Metastases usually have similar but not identical mutations, he said, “so the question is, how similar is similar enough?”
The bottom line:
“The vast majority of patients [diagnosed with cancer] already have disseminated cells,” said Aguirre-Ghiso. “That’s a negative. But the positive is that [these] cells haven’t activated their proliferation programs, so there is an opportunity to intervene,” by somehow attacking their weak spots — something that will require whole new avenues of basic research.
Given this, perhaps there should be a lot more emphasis on prevention. This requires something almost unheard of in the US: the Precautionary Principle. Regarding prophylaxis, there should be a lot more research on subclinical dental infections, and whether creating havens for anaerobic microbes is really a good idea. It may help to look at triggers in addition to mutations themselves.
I agree totally with your opinion.
How does this reconcile with the very high survival rates when breast cancer is found and treated early? It must be a bit more complex.
It would also be good to genome profile cells from multiple sites simultaneously…
The high x-year survival rates might be partly due to overdiagnosis and lead time bias.
Can you provide any research information about clear cell ovarian cancer, particularly reoccurrence rates?
Great! You must read also “Molecular oncology” (2016), edited by Vasef and Auerbach;-)
Great summary of complex results!
These findings confirm the rising criticism about precision medicine: Just sequencing the primary tumor and targeting it with “smart drugs” is too naive.
This also reinforces another view of cancer causation: It is not about mutations (eg due to genotoxic agents) but about awakening these sleeping monsters – which is a natural REGULATORY event that could be triggered by non-mutagenic toxins, and notably hormone mimetic chemicals (BPA etc) that act on preexisting signaling pathways.
Very insightful comment. For those of us who treat patients and watch wide-eyed as you scientists unravel these normal pathways gone rogue, it is like taking apart a Russian doll. There seems to be an infinite number of dolls inside each newly uncovered doll.
Comments are closed.