Four out of five metastatic colorectal cancers have already spread before they are even diagnosed, according to new research out today, led by researchers at Stanford University.
The study published today in Nature Genetics counters a widely-held belief that metastasis (the spread of cancer from an original site to other parts of the body) normally happens after the original tumor is quite large and when many patients have already been diagnosed. The new work suggests that in metastatic colorectal cancers, this spread has happened before the tumor is larger than a poppy seed, far too small to be symptomatic or to be detected by regular screening.
“This finding was quite surprising,” said Christina Curtis, PhD, assistant professor of medicine and of genetics at Stanford and leader of the study. “In the majority of metastatic colorectal cancer patients analyzed in this study, the cancer cells had already spread and begun to grow long before the primary tumor was clinically detectable.”
The researchers looked at patterns of genetic mutations in the colorectal tumors of 21 people and compared these to the mutations found in their metastatic cancers from the liver or brain. They then used this information to construct an evolutionary timeline of events, tracking back to figure out when exactly the cell that formed the metastatic tumor had split off from the original colorectal tumor. To their surprise, in 17 patients, they found that the cell had broken off of the main tumor very early in development when it was tiny and undetectable.
“Metastasis isn’t the final stage event in a genetically advanced tumor that has been assumed for so long. The cells that formed the metastasis were more closely related to the ancestors of the primary tumor than its present-day relatives,” said Curtis.
It has previously been assumed that metastatic tumor cells evolve over time, gradually accumulating genetic faults called mutations which allow them to escape the original location of the tumor, enter the bloodstream and plant themselves somewhere else-in metastatic colorectal cancer this is commonly the liver, lungs and more rarely-the brain. However, the new work indicates that this is not the case, with the metastatic cancers seemingly having many of the same mutations as the original tumor they came from.
To learn more about this, the researchers used historical data from almost 3,000 patients with colorectal cancer, some with metastases and some without to see if they could use their findings to identify mutated genes which were predictive of spread.
“We found that specific combinations of mutations were highly predictive of metastasis. For example, mutations in a gene called PTPRT, in combination with mutations in classic colorectal cancer driver genes, were almost exclusively found in patients with metastatic cancers,” said Curtis.
PTPRT mutations have been found in up to a quarter of colorectal cancers, and although not everyone with a PTPRT mutation will go on to get metastatic disease, the new research indicates that these patients have a much higher likelihood of developing metastases.
When PTPRT is mutated and lost, this increases the activity of another protein called STAT3, which promotes survival of cells. Although none are FDA-approved currently, there are over a dozen ongoing clinical trialslooking at STAT3-inhibitors in various different types of cancer, suggesting that there may in the future be a way to target these often hard-to-treat metastatic colorectal cancers.
“I found it a very interesting study, rather brilliantly conceived with all of this computer modelling. One of the main findings is that not all colorectal cancers are created the same, that there is a subtype which is very likely to metastasize early. This doesn’t mitigate the lifesaving value of screening, but it does say that screening alone is not adequate to do everything possible to reduce mortality rates from colorectal cancer,” said Richard C. Wender, MD, Chief Cancer Control Officer at the American Cancer Society.
Although it is exceptionally hard to detect early stage cancers with conventional screening methods, new liquid biopsy techniques may in the future be able to offer a quicker way to pick up these cancers before they are symptomatic and even detect whether metastatic disease is present at the time of diagnosis. This, however, is like looking for a needle in a haystack when tumors are small, but biomarkers like a mutated PTPRT gene might well allow this emerging technique to hone in
“Can we put the clock back further and might it be possible to catch cancer and malignancy earlier, before the patient presents with symptoms? The goal is to identify patients with highly aggressive disease and if we know patients are at high risk of distant relapse – we can stratify their treatment early on,” said Curtis.
However, at the moment STAT3 inhibitors have not been proven to work in metastatic colorectal cancer, although two clinical trials are ongoing to evaluate this. Considering that metastatic disease is often resistant to conventional chemotherapies, will catching these tumors earlier at the moment really help patients?
“It is highly likely that throwing more cytotoxic drugs at these cancers earlier is not going to transform outcomes for these patients, it may extend life, but with the compromise of being exposed to more cytotoxic drugs. For some patients theres meaningful benefit but it hasn’t been transformative compared to some targeted therapies,” said Wender, mentioning that targeted therapies such as imatinib for chronic myeloid leukemia and herceptin for women with HER+ breast cancer have resulted in major leaps in survival for people with these cancers.
The new research certainly lays important groundwork for future investigations that will hopefully improve outcomes for metastatic colorectal cancer patients. Curtis has plans to further this work to look at why these cancers sometimes spread to the liver, brain or other sites in the body, with the view to tailoring treatments earlier and giving patients a better chance of survival.
“Why do some cells colonize a foreign site? Why is colonization to the liver is more common than to the brain. Would we need different targeted therapies to metastases in the brain and liver? So far, we don’t know this, but we want to further investigate this, what makes cells adapt better to different microenvironments,” said Curtis.
Earlier this year, Curtis co-led a published study showing that breast cancer is 11 distinctly different diseases, each with a different prognosis and likelihood of relapsing, sometimes several years after the original disease is diagnosed. These large genomics projects teasing out the evolution of cancers are beginning to impact screening, diagnosis and treatment plans, but is this work in colorectal and breast cancers likely to be relevant to all other cancer types?
“I don’t necessarily think other cancer types have the same patterns. We want to understand how often early dissemination of cancer cells happen and how much we need to invest in early detection. Lung cancer historically has been one where we detect cancer too late, this is an obvious one where we want to understand the combinations of mutations that predict an early relapse.We are looking at this pan-cancer,” said Curtis.