Using data from nearly 18,000 people who were treated for cancer, scientists found that particular patterns of gene activity corresponded to patients’ survival odds — across a whole range of cancers, including brain, breast, colon and lung cancers.
What’s more, the particular mix of immune system cells within patients’ tumors also correlated with their prognosis, the investigators found.
The researchers said the findings, published online July 20 in Nature Medicine, could eventually be used to find new targets for cancer therapies — or to help predict patients’ chances of responding to some existing treatments.
The discoveries have been compiled in a database that’s available to other researchers, said senior study author Dr. Ash Alizadeh, an assistant professor at Stanford University’s School of Medicine, in California.
One hope, he said, is that the information can be translated into ways to predict a cancer patient’s outlook and help guide treatment choices.
Eventually, it might also prove useful in zeroing in on patients who are likely to respond to new cancer treatments that target the immune system. Those treatments, known collectively as immunotherapy, aim to boost the body’s innate cancer-fighting potential.
“The most exciting emerging therapies for cancer are immune-based,” Alizadeh said. “But so far, these therapies work well for certain cancers, but not for others. Or they work for a subset of patients, and not for others.”
Researchers are still trying to understand why, Alizadeh said.
At its essence, cancer is the uncontrolled growth and spread of abnormal cells. According to Alizadeh, many of the genes linked to cancer survival are involved in controlling cell growth and division.
One gene, called FOXM1, was associated with a poor prognosis, across a range of cancers, when its activity was high, the study found. In contrast, high activity in a gene called KLRB1 — which influences the immune response to cancer — was connected to better survival odds.
Overall, the researchers said, they were able to pinpoint the top 10 genes tied to a better prognosis, and the 10 most strongly connected to a poor outcome.
To get a better look at the immune cells tied to survival, the researchers “blended up” tumor samples, then took a detailed look at their “ingredients,” Alizadeh explained.
The researchers found some clear patterns: For example, people whose tumors had high levels of immune cells called plasma cells had a better prognosis; plasma cells secrete antibodies to fight foreign invaders.
On the other hand, when tumors had a high concentration of cells called neutrophils, the patient’s outlook tended to be worse.
Dr. Marcus Bosenberg, a researcher who was not involved in the work, said the study’s approach — looking at patterns across different cancers — can “give you insights you wouldn’t have when looking at individual cancers.”
However, it’s not clear yet how the information could be practically useful, according to Bosenberg, a co-leader of the Cancer Genetics and Genomics Program at Yale Cancer Center in New Haven, Conn.
“We don’t know how this would add to existing ways of estimating patients’ prognosis,” Bosenberg said.
The hunt for “prognostic markers” is a hot area in cancer research right now, Bosenberg added.
There are even some tests on the market that analyze certain sets of genes, to determine whether a cancer patient might benefit from a given treatment. An example is the Oncotype DX test, which can help predict whether a woman with early stage breast cancer has a high risk of recurrence: If she does, additional therapy — with chemo or radiation — could be considered.
Bosenberg agreed that immune-based therapies are the “most exciting” approaches in cancer research right now. And it’s possible, he said, that the immune cells that were key in this study could offer some type of “biomarker” that would predict patients’ likelihood of responding to certain immunotherapies.
But much more research is needed first, Bosenberg said.
The American Cancer Society has more on genes and cancer.
SOURCES: Ash Alizadeh, M.D., Ph.D., assistant professor, medicine, Stanford University School of Medicine, Stanford, Calif.; Marcus Bosenberg, M.D., Ph.D., co-leader, Cancer Genetics and Genomics Program, Yale Cancer Center, New Haven, Conn.; July 20, 2015, Nature Medicine, online