This story has been corrected to reflect that the PanCan initiative cited is the Know Your Tumor initiative, not the Know Your Cancer intiative.
NEW YORK (GenomeWeb) – A consortium of researchers affiliated with the Pancreatic Cancer Action Network's (PanCAN) Know Your Tumor Initiative have found that molecularly tailored treatments could improve patient outcomes.
In a study published last week in Clinical Cancer Research, the team found that using patient genomic and proteomic profiles to personalize pancreatic cancer therapy increased progression-free survival, demonstrating the potential of such approaches, said Michael Pishvaian, first author on the study and a medical oncologist at Georgetown University's Lombardi Comprehensive Cancer Center.
Pishvaian is also the chief medical officer of personalized medicine firm Perthera, which has partnered with PanCAN on the Know Your Tumor Initiative, arranging the patient molecular profiles and tumor board reviews of each patient's data to determine potential targeted treatment options. Emanuel Petricoin, a Perthera co-founder and co-director of the Center for Applied Proteomics and Molecular Medicine at George Mason University, was senior author on the CCR study.
Perthera was launched in 2013 by Petricoin and venture capitalist Dendy Young with the aim of providing cancer patients, particularly in the community setting, the sort of molecular profiling and expert interpretation typically available at major cancer centers.
The company does not perform the assays itself but helps physicians access genomic, proteomic, and phosphoproteomic profiling services from outside providers, serving as an intermediary handling the many steps involved in obtaining and sending out patient tumor samples to molecular profiling firms and returning assay results.
In the CCR study, the firm arranged for testing on 640 pancreatic cancer patients from 287 practices in 44 states, including both academic (112) and community centers (175). Of the 640 patient samples, 96 percent were sufficient for next-generation sequencing analysis, which was performed by Foundation Medicine or PGDx, and 91 percent were sufficient for immunohistochemistry testing, which was done by Caris Life Sciences or NeoGenomics.
Additionally, 20 samples were sent to phosphoproteomics profiling firm Theranostics Health (which Petricoin also co-founded), where the phosphorylation levels of 21 signaling proteins were tested to assess whether irregular protein kinase activity was present. Also, Caris or Foundation Medicine tested 241 samples for evidence of microsatellite instability.
The median time for report delivery was 30 days, with 90 percent of patient reports delivered within 51 days.
Of the patients profiled, half had actionable genomic alterations, while 27 percent had genomic alterations the authors described as highly actionable. On the protein side, 5 percent had proteomic alterations deemed actionable. Patients with highly actionable alterations who received therapy targeting those alterations had a median progression-free survival of 4.1 months, compared to 2.8 months for patients without highly actionable alterations, and 1.9 months for patients with highly actionable alterations but who did not receive a targeted therapy based on those alterations.
Petricoin suggested that the study contains several pieces of good news for proponents of personalized cancer therapies, among them the fact that it demonstrated the feasibility of performing extensive molecular profiling on patients across a wide range of geographies and in the community setting.
Additionally, he said, the study showed "that a significant proportion of these patients have highly actionable findings… and we were able to show statistically that when they are matched [to therapy] they appear to have increased [progression-free] survival."
At the same time, the study also points to some of the hurdles still facing this sort of molecularly guided approach, as the researchers did not find a statistically significant improvement in overall survival for patients receiving matched therapy. Pishvaian said he believed that data on additional subjects would ultimately demonstrate such an improvement.
"As we are gathering more patients and collecting more data, the statistical significance is increasing, and we anticipate that eventually we'll get to a large enough sample set that it will show a meaningful improvement in overall survival, as well," he said.
The relatively small number of subjects in the matched treatment cohort points to another challenge. Of the 640 patients for whom Perthera generated a report, follow-up information was not available for 200; 111 passed away before starting new treatment; 173 remained on the treatment they were using prior to receiving their report; and 156 began a new treatment regimen.
Of the 156 who started a new treatment, 126 chose a treatment presented by the report. Of those 126, 35 had highly actionable alterations. However, less than half of those patients (17) chose therapies targeted to those alterations, while the other 18 patients received either standard of care or molecularly targeted therapies that did not match the alterations identified in the Perthera report.
This, Pishvaian said, indicates the even when patients are good candidates for targeted treatments, they often don't pursue this route.
"The reasons for that, I think, are worthy of being explored," he said. "Whether it's sort of reluctance to follow these molecular tests, or if it's just [a simple matter of] patient access — an inability to access off-label therapies [or] to access clinical trials."
Pishvaian said that in a previous analysis four years ago by Perthera of why eligible patients didn't receive therapies recommended by the company's reports, it found that physician reluctance was a major factor. He said, though, that this physician reluctance has dropped significantly since that time and is no longer the primary issue.
"I think the primary cause now is just access to therapies," he said. "The patients either are unable to get access to off-label therapy because they can't get insurance to approve it, or they can't get the pharma company who makes the drug to give it to them for a reasonable cost, or they don't live close enough to the clinical trial sites that might have access to those therapies."
Petricoin added that, particularly in the case of pancreatic cancer, which is typically quite aggressive, a large proportion of patients might not have the performance status needed to participate in a clinical trial.
"There's a constellation of issues" preventing patients from receiving targeted therapies, he said.
The study is ongoing and aims ultimately to collect outcomes data on around 1,000 pancreatic cancer patients, Pishvaian said, adding that such a dataset would enable research into a wide range of "scientific and clinically oriented questions that we could start to ask around prognostic variables, around predictive biomarkers."
"The bigger vision really is being able to come up with a structured database with molecular-based profiling outcomes, treatment histories, clinical-epidemiological and chart-extracted [patient] data," Petricoin said. "We're following these patients longitudinally, and that really is creating an unprecedented sandbox, and it's growing."
Pishvaian said that while the aggressiveness and lack of treatment options make pancreatic cancer a particularly challenging disease for such a study, he believed the results suggest the potential of molecularly targeted treatment approaches to other cancers.
He added that the findings were "very consistent with a couple of other key publications that have come out in the last few years that have basically told the same story, albeit, not in pancreatic cancer exclusively, where patients with actionable mutations who go on match therapy have better outcome than those who don't."
One recent example of such a finding was presented at the American Society of Clinical Oncology meeting last month by researchers involved in the University of Texas MD Anderson Cancer Center's "Initiative for Molecular Profiling in Advanced Cancer Therapy" (IMPACT) study. Looking at 3,743 cancer patients, the researchers found that median progression-free survival time in the molecularly targeted treatment group was four months compared to 2.8 months in the non-matched group, while overall survival was a median of 9.3 months in the matched group relative to 7.3 months in non-matched patients.