NEW YORK (360Dx) – As its first clinical test moves toward commercial launch this year in partnership with Genomic Health, Epic Sciences is ramping up development efforts for assays in new indications, and provided a first look at one, a PD-L1 assay, in a study published last month with collaborators at Yale University, the University of Southern California, and the University of California, San Diego.
The study, which appeared online in Cancer Epidemiology, Biomarkers & Prevention, demonstrated that the presence of PD-L1 protein on circulating cells from newly diagnosed lung cancer patients was prognostic of poor survival on standard-of-care therapies.
A number of assays have been developed for assessing PD-L1 expression from tissue in order to better predict which patients may benefit from the new class of immuno-oncology drugs.
For example, Roche received CE-label expansion last week for its Ventana PD-L1 (SP263) assay to identify untreated and previously treated patients with metastatic non-small cell lung cancer that are eligible for Merck's Keytruda (pembrolizumab) immunotherapy.
Increasingly, companies are also working on noninvasive tests that can be performed using blood or other body fluid samples. Biocept, for example, launched a blood-based assay for PD-L1 last year.
Although recent reports have highlighted that oncologists are embracing immunotherapies over drugs targeted on specific genomic or molecular alterations driving a tumor, only 20 percent to 30 percent of patients respond to these immune-mediating treatments.
"If immunotherapeutics were very affordable, we'd probably let the patient show us whether or not the agent works rather than invest in prediction," Dan Boffa, the first author of the study and an associate professor of surgery at Yale, said this week.
"But the reality is that there is both financial cost and a cost to the patient if you are potentially delaying an effective treatment, so we do need these tests," he added.
Ryan Dittamore, Epic's vice president of translational research and clinical affairs agreed. "When a patient has metastatic lung cancer, and the only way you can potentially cure them is a PD-L1 inhibitor, there has to be a very good reason not to treat them," he said.
"The problem with [current] tissue tests is that they are limited," he added, both by clinical issues with taking biopsies from critically ill patients, and by the heterogeneity and evolution of cancer, that can lead to false negatives if older tissue samples are used.
"It's our hope that the cells we are measuring are a true representation of the entirety of the disease, or the tumor, in the current moment," Dittamore said.
Epic's platform stands apart from many other liquid biopsy approaches because it captures and arrays cells in a stringently unbiased manner.
According to Boffa, while other technologies may be more appropriate for certain clinical questions, the goal of examining circulating cells for PD-L1 expression was something he and his colleagues felt the Epic platform, which they had been working with for some time, was particularly well suited.
In the study, Epic and its Yale collaborators tested retrospective samples from 112 newly diagnosed lung cancer patients who had been followed for up to four years.
When they compared clinical outcomes to PD-L1 expression in patients' CTCs, the researchers found that the biomarker was clearly associated with worse overall survival regardless of clinical staging.
Interestingly, because of the qualitative analysis that Epic's platform allows, the investigators were able to define two separate populations of PD-L1 expressing cells — one that was cytokeratin-positive, and one that was cytokeratin-negative, suggesting that liquid biopsy CTC methods that capture only CK+ cells might miss a proportion of PD-L1 positivity.
In the case of tissue-based PD-L1 testing, the diagnostics community has been grappling with the issue of discordance between various available antibodies for IHC staining.
Boffa said that he and his colleagues picked the antibody they used in the lung cancer study with Epic before this research came out, and got lucky, having selected one of the three that appear to behave more concordantly.
Though the ultimate goal is for Epic's CTC approach to enter the clinic as a test to predict patient response to immunotherapies, the current study doesn't yet directly support that.
Instead, it provides an interim validation step, demonstrating that Epic's CTC-based PD-L1 assessment does correspond to known clinical patient features, in this case, poorer response to non-immunotherapy treatment.
Both Dittamore and Boffa agreed that the essential clinical validity and utility data have yet to be established, but both said that first establishing a baseline of validity for the biomarker itself is a critical first step.
Dittamore said that though the Yale study represents its most mature PD-L1 study so far, it has ongoing projects using the same PD-L1 CTC analysis in a variety of different cancer types.
These efforts are expected to help the company establish clinical validity and utility for prediction of immunotherapy response both in cancers where relevant drugs are already approved, and in indications where there is interest in expanding these labels.
In addition to PD-L1 testing, Epic is also actively working to develop CTC-based methods for use in the context of PARP inhibitors, Ryan said.
Meanwhile, the company's first clinical assay — which measures expression of the AR-V7 splice variant in CTC nuclei in order to guide treatment for men with prostate cancer — is expected to be launched commercially in collaboration with Genomic Health later this year.