Cathleen Nguyen and Gonzalo Torga
Two Johns Hopkins prostate cancer researchers found significant variability when they submitted the same patient samples to two different commercial fluid biopsies. Fluid biopsy is a new and inconclusive alternative to tumor tissue sequencing, and it is intended to specifically detect and follow tumor DNA circulating in patients ’blood. The results are used to help guide physicians to tailor the best treatment for patients at each point of their disease.
Gonzalo Torga, MD, is a postdoctoral research associate at the Brady Urological Institute at Johns Hopkins and Kenneth Pienta, MD, director of research for the Brady Institute and a prostate cancer researcher at Johns Hopkins Kimmel Cancer Center, reported their findings in the Dec. 14 issue of JAMA Oncology. The results, they say, raise the likelihood that cancer patients may be prescribed different treatments depending on which commercial provider is used for the liquid biopsy.
The two liquid biopsy panels compared the Guardant360, from Guardant Health, Inc., which followed at least part of the coding sequences of 73 genes, and the PlasmaSELECT panel from Personal Genome Diagnostics, which follows the coding features of 64 genes.
Both laboratories are licensed by the Clinical Transformation Laboratory, known as CLIA, and are accredited by the College of American Pathologists, and are reported to have high sensitivity (in this case, the ability to correctly detect mutations when they occur) and high accuracy (the ability to correctly report as negative in the absence of mutations).
The two companies differ in which genes, and regions within each gene, are covered. Only 25 of the 40 patients in the study had even one genetic change reported within the overlapping genetic sequences covered by the same companies.
Even when companies analyze DNA from the same blood obtained, their results are rarely consistent. When comparing the internal outcomes of the overlapping genetic sequences, the outcomes from the same companies were perfectly matched for all mutations reported in only 7.5 percent (3 of 40 patients) of the case. In 15 percent of patients (6 of 40), similar outcomes coincide with at least one of the reported mutations. In 40 percent (16 of 40) of the patients, no mutations were reported that were possibly covered by the same panels seen by the same companies.
Torga and Pienta’s research was aimed at finding the most highly commercial lab to test patient samples from their metastatic prostate cancer clinic. “We wanted to make the best choice for our patients, so we started submitting samples to the same area at the same time to compare the results,” Torga said.
Patients may respond similarly to cancer medications depending on how differently their gene changes are in their tumors. Oncologists use the results of a fluid biopsy to determine if a patient is responding to treatment and to monitor tumors to determine if they are progressing and may need new or additional. therapy. They can also use the results to select patients for clinical trials, if a specific change is detected in the DNA of their tumor.
A small study was published earlier this year on JAMA Oncology compared mutations identified in tumor tissue samples and in fluid biopsy results from the same patients and found a similar lack of support.
Each commercial trial costs about $ 4,000, Torga notes, and patients often simply send their samples to a lab for testing. “Fluid biopsy is a promising technology, with a tremendous potential to affect our ability to treat patients, but it’s a new technology that may require a lot of time and experience to improve,” he said. Torga. “We don’t know from these studies which laboratory panel is better, but we can say that certification for laboratories needs to improve.”
Torga and Pienta told them they wanted to make sure oncologists were aware of misunderstandings so they wouldn’t be blind to trusting the fluid biopsy results.
The PlasmaSELECT test used in the study described in this publication is owned by Personal Genome Diagnostics (PGDx). Johns Hopkins University has licensed PGDx technologies and, as a result, has a financial interest in the company. This arrangement was reviewed and approved by Johns Hopkins University in accordance with conflict of interest policies.