NEW YORK (GenomeWeb) – Leaders of the British Columbia Cancer Agency's ongoing Personalized OncoGenomics (POG) clinical trial have published data on the results from the first 100 patients recruited to the study, in which researchers are sequencing the whole genomes and transcriptomes of participants with advanced cancer to inform personalized treatment.
The report, published earlier this month in a new journal, Cold Spring Harbor Molecular Case Studies, showed that amongst the first patients recruited to the POG trial — 100 adults and six children between June 2012 and August 2014 — most were successfully sequenced and a majority of those had actionable results.
The researchers also reported that a significant number of those with actionable results — 34 of 55 cases — were actually able to be treated based on their genomic findings.
Janessa Laskin, the study's first author and a co-leader of the POG project, told GenomeWeb this week that these initial 106 cases can be looked at as a first phase of the POG program. The hope is that reflecting on the challenges and solutions implemented by the trial through this early phase can help inform the growing ranks of similar clinical sequencing programs and trials.
According to Laskin, the results also serve as a proof of principle for the POG project moving forward as it seeks to expand its efforts to sequence not just hundreds, but thousands of cancer patients.
"It was never our intention to stop at 100 patients," Laskin said. "I have ethics [board] approval for 5,000 patients in the next five years and that's what we want to do."
Earlier this year, leaders of the project announced they hoped to expand the POG trial cohort to 300 patients using $5 million in funding raised through a BC Cancer Foundation gala in 2014, and a total of $12.5 million committed by the agency for the POG project this year.
Laskin said this week that money will likely take the POG project through sequencing of close to 500 subjects.
"As of today, from the inception of the study we have consented 380 patients — 358 adults and 22 pediatric cases," she said. "We have completed sequencing on 292 of those … and with the BC Cancer Foundation pledge to raise about $12.5 million, that will take us to about 450 patients."
According to Laskin, the project's sequencing strategy hasn't changed substantially. POG combines deep sequencing of patients' whole cancer genomes, germline DNA, and RNA, and also incorporates a targeted panel that can return an initial set of limited results in a shorter timeframe.
One change to the project protocol, Laskin said, has been the adoption of an in-house targeted sequencing panel rather than the AmpliSeq Cancer panel that was initially employed by the researchers.
Overall, Laskin said, the POG researchers have found that regardless of the panel used, the results have not been very informative for patients in the trial.
Of 81 subjects that received panel results in the study, 73 percent carried mutations that were classed as informative but not actionable, the majority of which were mutations in TP53, the study authors wrote. In the remaining 27 percent of cases, no variants were detected using the AmpliSeq panel at all.
"It's interesting to see for each individual in the trial what comes out in the panel versus the whole genome," Laskin said, "because when we look at our panel data, we almost never act on that when we have all of the other information, particularly the transcriptome information."
This insight makes some of the more disappointing results coming out of other trials of molecularly-guided cancer therapy a little more understandable, she added.
For example, the SHIVA trial, recently published in The Lancet Oncology, found that the use of molecularly targeted agents outside their indications did not improve progression-free survival compared with a physician's choice of treatment in heavily pretreated patients with cancer.
"You see how disappointing that was that you can't seem to use just a panel and copy number information. It just didn't seem to have a huge impact," Laskin said. "But it kind of doesn't surprise me because we've seen numerous instances where we have something abnormal in the panel but it doesn't play out in the RNA."
Meanwhile, the POG project's broad sequencing approach has resulted in a fairly high yield of so-called actionable genomic findings, at least in the study's first 100 recruited subjects.
According to the authors, the project launched in July 2012 and enrolled one patient per month for the first five months, scaling up accrual after that. As of August 2014, 100 adult and six pediatric patients had been consented to the study, representing 30 different cancer types.
Though only 78 of the 106 total subjects could be successfully sequenced — reasons listed by the authors for why patients could not be sequenced included unavailable biopsy tissue, insufficient tumor content, patient withdrawals, and deaths — more than 50 percent of the successfully sequenced patients received what the study protocol deemed actionable results.
Even more importantly, among the 55 cases with potentially actionable findings, more than half were actually able to be treated based on those results. Among these 34 patients, clinicians also reported that 14 had clinical or radiographic improvement when treated according to their genomic status.
Eight additional patients in the program were eligible for molecularly informed therapy based on sequencing results, but did not receive that therapy for a variety of reasons, and another 13 had an actionable target identified but were either too unwell or died before POG-informed therapy could be offered to them.
In five cases a patient's diagnosis was either changed to a different primary tumor than originally diagnosed or a primary site was suggested in cases where it was not previously known, the authors reported.
Laskin and her colleagues also described in the study many of the POG's strategies for reporting sequencing results to participating physicians, including describing levels of evidence to support relationships between somatic alterations and candidate drug classes.
"We noted that clinicians were most comfortable when several abnormalities coalesced around a pathway, (e.g., evidence of both “activating” copy number changes and high gene expression levels), and only rarely used single somatic alterations to inform a treatment decision," the authors reported.
According to Laskin, the major bottleneck in treating patients based on the POG sequencing results has been drug accessibility.
"Nearly every patient has dysregulated pathways, and many of them rise to the surface as things we can target, but drug accessibility is what we struggle with most," she said.
"We as oncologists are moving very quickly to understand that a biological marker doesn't necessarily reside in an anatomic silo, but regulatory authorities are slower to pick that up," she added.
She said the group has a robust Phase I trials program already and is working to build that up, "but what we want to do is repurpose approved drugs, so we are also working on some pilots to try to address that."
In addition, Laskin said that the trial has also shifted slightly since the first 100 patients in an attempt to increase the chance of genomic data actually influencing treatment.
"When we realized that a Phase I population didn't stay well or live long enough to get into many trials we made the major change since July of 2014 that is not mentioned in the paper, which was to change the eligibility criteria," she said. "It's still metastatic incurable disease, but patients have to be either chemotherapy naive or after one line of chemo, so we catch people earlier and we have more time to use the data."
Though the data isn't in yet as to whether this has had an impact on the frequency of molecular therapies being used in these more recently recruited patients, Laskin said it looks initially like clinicians in the trial may have been able to treat more of these patients than they did among the first 100 participants.
As they move forward, Laskin and her colleagues are tracking not only patient sequencing results, and statistics on the frequency of molecularly guided treatment, but also digging more deeply into patients' outcomes on various therapies, to try to track treatment modalities that worked and those that didn't according to the deep sequencing data available in the trial.
We are also doing cost analysis and economic evaluations at the same time," Laskin added.
"What we believe is that five years from now, say with 5,000 patients, we will understand what we need in order to integrate genomics into standard cancer care that would be paid for either by insurance companies or by the government in Canada as it happens to be," she said, "but we will need [more] funding to get us to that point."