Teaming up on metastatic prostate cancer

“The pervading wisdom is that there’s no clear role for doing clinical genomics in prostate cancer,” said Eliezer Van Allen, a research associate at the Broad Institute, instructor of medicine at the Dana-Farber Cancer Institute, and assistant professor at Harvard Medical School. “That’s because when we look at the data, we just don’t see much to target.” But that wisdom is based on investigations of primary prostate cancer. Van Allen and his colleagues believe there is something fundamentally—and thus genomically—different about metastatic prostate cancer. A paper from Van Allen and colleagues released earlier this week in Cell bares out that belief.

Van Allen is a member of an international consortium of scientists, funded by a grant from Stand Up to Cancer, that’s trying to answer some longstanding questions about metastatic prostate cancer. In the first major publication since the group’s inception, Van Allen and his colleagues—including institute member Levi Garraway, who serves as one of the consoritum's principal investigators—collected and sequenced tissue biopsies from 150 patients with metastatic prostate cancer and found its genomic landscape to be much different than that of the primary form of the disease.

Previous genomic studies of primary prostate cancer have revealed alterations that may be useful for diagnosis, but the therapeutic impact of these findings has been limited. In contrast, Van Allen and his colleagues found that nearly 90 percent of metastatic cancer patients harbor potentially actionable variations. For example, the team found that genes in the DNA repair pathway—such as BRCA2 and ATM—were mutated in over 20 percent of the cohort, and that many of these patients had both somatic and germline alterations in these genes. Patients with these alterations may selectively respond to emerging DNA repair oriented therapies, as was the case for a few patients in this study. 

Another important contribution of the research is the team’s method of obtaining high quality sequencing data from bone—the tissue to which prostate cancer most commonly metastasizes and which is notoriously difficult to biopsy. The team, which includes clinicians at more than a dozen institutions across three nations, used a state-of-the-art drill to safely and effectively obtain good bone tumor samples, Van Allen explained. By sequencing the samples, researchers at the Broad and University of Michigan were able to identify a high degree of clinically relevant, theoretically actionable genetic alterations.

Metastatic prostate cancer is dubbed “castration resistant,” meaning that if the source of androgen hormone—the tumor’s “primary fuel,” according to Van Allen—is removed, the cancer persists. The standard of care for primary prostate cancer is to perform “chemical castration” through various drug cocktails. But these drugs don’t work well for metastatic cancer. Despite that fact, most of the mutations Van Allen and his colleagues found still involved androgen receptor signaling, indicating the development of drug resistance in the metastatic disease. “Part of the longer term goal is to get biopsy samples from these patients at the time of resistance and try to figure out what’s happening at the genetic level to come up with a more rational combination of drugs,” Van Allen said.

In addition to analyzing DNA, the team also looked at alterations to RNA, a macromolecule that helps translate DNA sequence into proteins.

“There were many patients for whom the clinically actionable event happened at the RNA level,” said Van Allen. “So just doing DNA-based sequencing is insufficient—much of the action is happening at the RNA level so having a clinical platform for prostate cancer that can do both is probably going to be needed.”

The new research brings metastatic prostate cancer into the land of precision medicine. In this paradigm-changing approach to patient care, clinicians can examine a cancer’s genotype and select drugs more likely to be effective. “Rather than give all patients a nonspecific cytotoxic drug with side effects, we can tailor a therapy to a specific patient,” Van Allen explained. Now clinicians have a handful of genes that they can target when attempting to treat metastatic prostate cancer patients individually. But there’s still much work to be done.

“With these first 150 patients, we’ve really only scratched the surface of what we can learn,” Van Allen said. He and his colleagues plan to continue collecting bone biopsies and sequencing their genomes to home in on more specific questions about biomarkers of response and drug resistance.

The team hopes its work will serve as a model for metastatic cancer research in general. The fact that there is no systematic method for collecting metastatic biopsies—patients already have confirmed cancer, so diagnostic biopsies aren’t necessary—makes studies like this one difficult. But Stand Up to Cancer and similar consortium-based initiatives are helping. Without significant numbers of biopsied bone collected by clinicians across the consortium, the team would not have been able to identify relevant genetic alterations.

“Rather than have different institutions work on the exact same project many different ways in a micro scale, this is about having them all work together on a macro scale and actually be able to achieve something that no one group could do by themselves,” said Van Allen.

Paper cited: Robinson, D., Van Allen, E.M., et al. Integrative Clinical Genomics of Advanced Prostate Cancer, Cell (2015). Doi: 10.1016/j.cell.2015.05.001

For more information on the study, read the Dana Farber press release.