NEW YORK – Whole-genome sequencing may be a practical method for finding markers to guide treatment in metastatic cancer patients, new research suggests.
At the European Society for Medical Oncology's Virtual Congress on Saturday, pathologist Kim Monkhorst from the Netherlands Cancer Institute shared interim results for the first 800 of the 1,200 patients who will be profiled through the "Whole-genome sequencing implementation and standard of care diagnostics for every cancer patient," or WIDE, study.
Patients enrolled in WIDE were required to have metastatic disease and a solid tumor type. Using tumor biopsies and blood samples, the team attempted to perform both whole-genome sequencing and standard-of-care diagnostics for each patient, generating reports that were reviewed by experts and multidisciplinary teams, for more complex cases. Based on the tumor biomarkers uncovered with each testing approach, the researchers then determined whether the diagnostic method affected the biomarker-based treatment options available.
The study's primary endpoints were feasibility of whole-genome sequencing, particularly with a fast enough turnaround time, and concordance with molecular findings from standard test types, Monkhorst explained. Cost effectiveness and clinical value features such as increased treatment availability or improved differential diagnoses were tracked as secondary endpoints.
He called whole-genome sequencing "an attractive solution" to some of the problems that pathologists currently face in the clinic, since it provides a far more complete look at tumor genomic features and tumor mutational signatures than available gene panels or other targeted tests.
"Pathologists increasingly use genomic information to solve their differential diagnoses," he added. "Even in cancers of unknown primary, you can solve the diagnoses often [with genomic data]."
Standard clinical genetic testing tends to be tumor type-dependent, he noted, and if a suitable biomarker is not found during the first round of testing, it may lead to a series of additional tests that are time-consuming and potentially limited by the amount of biopsy material available.
"This poses a challenge for us pathologists, because we want to get the best diagnostics for our patients," Monkhorst said. "We continuously keep implementing new biomarkers, new platforms, and new [next-generation sequencing] panels — and when we're done, we have to start all over again because there is a new set of biomarkers."
In contrast, having a genome sequence available for a patient's tumor at the time of diagnosis makes it possible to go back and re-evaluate the sequence as new biomarkers are identified, which would ideally help to get the individual on an appropriate treatment more quickly.
Monkhorst highlighted the wealth of genomic biomarker-targeted oncology drugs that are on the market, or potentially coming to the market, including those identified through the Drug Rediscovery Protocol (DRUP) in the Netherlands.
"This protocol allows physicians to treat their patients with off-label drugs if we can identify such a biomarker in their tumors," he explained, noting that there are already more than 160 treatment cohorts being considered through DRUP.
Still, Monkhorst and other members of the Netherlands Cancer Institute team wanted to take a look at the feasibility of using whole-genome sequencing routinely in cancer patients — from the practicality of obtaining sufficient high-quality tumor DNA from fresh-frozen biopsy samples to the timeline for getting genomic sequence data back for each patient.
"You need to optimize and adjust your standard [diagnostic] pipelines and routine clinical care to be able to get fresh-frozen DNA and still do your normal diagnostics," he said, explaining that samples should contain more than 50 nanograms of DNA and be comprised of more than 20 percent tumor cells to yield informative sequences.
To that end, the investigators came up with "fresh-frozen, fine fixed, MMC-embedded," or 4FME, method for processing biopsy samples, producing slides that pathologists mark for microdissection before sending the remaining sample on for longer-term formalin fixing. The 4FME approach is used unless there is a very small amount of biopsy tissue available for a patient, Monkhorst said, or unless the patient's pathologic diagnosis is uncertain.
Starting with 867 biopsy samples from 800 metastatic cancer patients, for example, the team was able to generate whole-genome sequencing data for 566 patients and 66 percent of the available biopsies, after accounting for missing or low-quality samples and DNA preparations.
Although the turnaround time between biopsy collection and return of a whole-genome sequence-based report jumped when the team transitioned to a new pipeline last fall, Monkhorst explained, it is now around 10 business days — well within the 12 day deadline the researchers set for themselves when designing the WIDE study.
"We are very confident that we will reach this [turnaround time] endpoint," he said.
On the clinical validation side, meanwhile, the whole-genome sequencing method identified 645 informative markers in all, including 620 biomarkers found with both genomic analyses and standard of care testing. Whole-genome sequencing produced 19 false-positive or false-negative calls, compared to six false-positives or -negatives found by standard testing.
The team has adjusted its analytical pipeline to continue improving concordance, since many of the alterations initially missed by whole-genome sequencing were present in the data but not reported. So far, 52 patients are receiving treatments based on actionable mutations found in their tumor sequence data, while 285 patients had actionable mutations that would make them eligible for clinical trials underway in the Netherlands but were not receiving targeted treatment.
"Our study shows that whole-genome sequencing performs as well as standard of care diagnostics," Monkhorst said during a panel discussion on cancer sequencing following his presentation. "It has a lot of extras, which we can use in the clinic, but also for research, it's very important for the development of [new] biomarkers."
Nuria Lopez-Bigas of the Barcelona Institute of Research in Biomedicine, who reviewed the study at the meeting, added that these and other whole-genome sequencing studies in cancer are part of a "virtuous cycle" that provides clinical data for one patient, while producing data to help interpret and treat other cancers in the future.