Metastatic prostate cancer presents an appealing target for precision oncology, according to new work from scientists at Fred Hutchinson Cancer Research Center and the University of Washington. In a study the researchers showed that though metastases from different patients varied widely in their genetic characteristics, metastases within a single patient were remarkably similar. These results suggest not merely that patients with metastatic prostate cancer may benefit from treatment tailored to their particular tumors, but also that a single biopsy may provide enough information to oncologists to guide such therapy.
“If you look in multiple metastases within a given patient, they’re actually very, very similar,” said Dr. Pete Nelson, the study’s lead author and a prostate cancer researcher at Fred Hutch and oncologist at Seattle Cancer Care Alliance, the Hutch's treatment arm. Unlike initial tumors in the prostate, in which molecular characteristics can vary by location, the tumor cells that break free and travel to distant locations within the body, forming metastases, appear to share characteristics. “They’re not identical, but they are very similar… We can feel generally confident, at least with prostate cancer, that if you did sample a single tumor, you could make clinical decisions based on what you find.”
An unmet need
More than 27,000 men die from metastatic prostate cancer each year in the U.S. according to the American Cancer Society. Men with metastatic prostate cancer are treated with drugs that suppress their testosterone — a therapy that works very well, until it doesn’t. Metastatic prostate cancer almost always will develop ways to circumvent the loss of testosterone, recurring and progressing even more aggressively. Despite this phenomenon, this standard of care — also known as androgen deprivation or androgen suppression therapy — has remained almost unchanged for half a century, and patients urgently need new therapies.
A deeper understanding of the changes that metastatic prostate cancer can undergo to resist treatment has opened up several potential new therapeutic avenues, but researchers don’t yet know which method will work for which patient.
Successful precision oncology, in which each patient’s therapy is tailored based on the molecular alterations of their cancer, requires both that patients differ dramatically from each other — which bars a one-size-fits-all approach. It also relies on relative uniformity among the tumors within each patient — so the treatment any individual receives is effective against every one of their tumors.
Researchers have been gathering evidence that cells within the original, or primary, tumor, which can take decades to develop, can exhibit different molecular features than their neighbors. But scientists had little information about whether this also holds true for the cells that have left their home tumor and taken up residence elsewhere, such as in the bones or the liver.
'Essence of precision medicine'
To compare the molecular characteristics of different tumors, Nelson and co-authors Dr. Akash Kumar of the UW and Ilsa Coleman, a researcher working in Nelson's lab, studied metastatic tumors from 176 men who had died of prostate cancer and previously agreed to donate their tumors posthumously to research. The team examined various types of mutations and alterations in how strongly particular genes were dialed up or down.
“We found tremendous variation across patients, which is sort of the essence of precision medicine: not every tumor from every person looks the same … so you would want to use a particular therapy in one person and a different therapy in another,” said Nelson.
But tumors within individuals were quite similar, he said, “so you could be confident that if you selected a treatment based on what the composition of tumor A was, that the rest of the tumors should respond.”
Avenues of exploration
In addition to demonstrating that prostate cancer may present just the characteristics needed for precision medicine strategies, Nelson’s team identified a few potential leads for specific drug regimens.
Cells respond to testosterone via a molecule called the androgen receptor, or AR, which turns on a suite of genes known as the “AR signature.” Though all the men in the study had undergone testosterone-suppressing therapy, more than two-thirds of the men had tumors with a high AR signature. Surprisingly, tumor cells with this specific signature appeared to grow more slowly than tumor cells from other men in which these same genes were active at lower levels. This may be a hint that men whose metastases exhibit a very high AR signature could be candidates for a new strategy currently being tested in a Phase 2 clinical trial at SCCA: alternating high doses of testosterone with suppressive therapy. The hope is that the burst of testosterone will push tumor cells to stop dividing, said Nelson, though he cautioned that it’s too soon to know how well the strategy would work.
The team also found that in the tumor cells that multiplied more quickly, genes that encode DNA-repair proteins were more likely to be active. These critical proteins keep our DNA from accumulating mutations, and defective versions of DNA-repair enzyme genes such as BRCA1 and BRCA2 have been linked to cancer.
Nelson previously collaborated on a study showing that almost 25 percent of men with metastatic prostate cancer had tumors with mutations in DNA-repair genes. Drugs aimed at treating other types of tumors that lack the ability to repair DNA are already in the clinic, including PARP inhibitors for ovarian cancer. A recent clinical trial of one PARP inhibitor showed dramatic responses in men whose prostate tumors had DNA-repair defects.
“Our [new] study showed that if you biopsied one tumor in that man, and you showed [the cells] had a DNA repair defect, it would be present in the other tumors from that man. And so you could be pretty confident to try that type of drug,” said Nelson, though he noted that PARP inhibitors are not yet approved for patients with prostate cancer.
However, drugs currently available for prostate cancer may also be good candidates for tailored use against some of these types of tumors. Nelson and colleagues at UW published acase report in December of 2015 describing three patients with metastatic prostate cancer who all had dramatic responses to a platinum-based drug — a drug also known to be effective against tumors with defects in their ability to repair DNA and used extensively in ovarian cancer treatment.
“And that’s a drug we can use today because it is approved for prostate cancer,” said Nelson.
Oncologists already have tools to analyze successful biopsies of their patients’ tumors. “We have an assay today called UW-OncoPlex that we can order to assess the mutation status of a man’s tumor,” said Nelson. “In many situations, this is now a very straightforward procedure and very useful information can be obtained to guide therapy.”
However, it’s not always easy to get the necessary information from biopsies, Nelson said. He noted that prostate cancer often metastasizes to the bone, and this hard substance can make it difficult to remove tumor cells to analyze. Consequently, information about tumor cells can only be gleaned from about 65 percent of bone biopsy samples.
But tumors often shed cells and snippets of DNA into the bloodstream. Nelson and his colleagues are working to develop a much less invasive, blood-based test to detect those free-floating cells or molecules in order to one day replace biopsies of metastases.
The researchers will also need to explore how a patient’s disease might adapt to a therapy custom-selected based on a biopsy result. The blood-based test that Nelson’s team is working on would make this project easier, since repeated biopsies can cause tissue damage to patients. Their newly published analysis “doesn’t tell you what will happen over time, under the selective pressure of your treatment. And those are the next studies that need to happen,” said Nelson.