By Deborah Borfitz
August 25, 2020 | At day one of the Next Generation Dx Summit, Samir Hanash with MD Anderson Cancer Center highlighted “unprecedented opportunities” for blood-based biomarkers in cancer detection, including the identification of high-risk individuals at the bottom of the eligibility pyramid for low-dose CT lung cancer screening who develop advanced stage disease despite being less than a 20 pack year smoker or having never smoked at all. One of the more astonishing discoveries is that a single biomarker—pro-surfactant protein B—can significantly improve smoking-based risk assessment. It subsequently became part of a four-protein panel that similarly boosted the lung cancer detection rate relative to clinical grade assays when tested on multiple big-study cohorts. A high-performance, pan-cancer screening tool will need to be built one cancer type at a time, he added, meaning it is still a long way from the clinic.
The regulatory environment for companion diagnostics (CDx) was the theme of a presentation by Merck’s Eunice Lee, complicated by the reality that they less often involve one test, drug and indication but “all possible combinations.” While their development ideally happens concurrently with a drug approval, that’s not often the case. Best practice is for drug and diagnostic developers to pair up early and have a strategy that considers the policies of global regulatory bodies as well as each stage in the process from development through test access and adoption.
As pointed out by Reena Philip, with the FDA’s Center for Drug Evaluation and Research (CDER), the agency clearly sees more education and pre-planning is needed around CDx. Her focus was on clinical validation, a realm where companies tend to have too few test-positive and test-negative cases in their bridging studies to demonstrate efficacy, potentially causing delays in approval timelines. Lessons were drawn from recent approvals of a molecular test for tissue-agnostic tumor mutational burden and a liquid biopsy test that leverages next-generation sequencing.
The FDA’s Julie Schneider provided an update on selected precision oncology initiatives of the agency, including the Oncology Center of Excellence that works closely with medical oncologists in CDER’s Office of Oncologic Diseases whose job includes working on approvals of drug/diagnostic pairs. She homed in on guidance around low-frequency subsets, where patients with the targeted rare variant may not even be represented in trials, and in vitro diagnostic risk, comparing the regular and streamlined approval process and when each is appropriate. She also reviewed the status of the public-private Medical Device Innovation Consortium, now in its second phase where cell lines of defined variants are being produced and characterized.
The risks and benefits of home-based diagnostic tests was a hot topic of conversation among presenters and panelists, as was the impact of COVID-19 on that calculation. Elliot Cowan with Partners in Diagnostics LLC immediately pivoted to the rise in the FDA’s Emergency Use Authorizations (and tips for getting one) and the agency’s just-out template for molecular and antigen tests for non-lab use. John Hopkins’ Charlotte Gaydos recited positive experiences and high user satisfaction with its I Want The Kit laboratory for sexually transmitted infection testing involving at-home specimen collection, noting that the pandemic has helped hike demand by over 500%. The problem, she says, is that the FDA is not regulating the high volume of pricey tests available via the internet, and no company has ever done a study looking at the home collection of specimens by patients. But the FDA is starting to crack down on the online purveyors, reports Lawrence Worden with IVD Logix. Perhaps laboratory accreditation groups (e.g., COLA or CAP) could work hand in hand with telemedicine providers, suggested John Hopkins’ Jeanne Mumford, who additionally favored the “test to break it” approach to more full-proof product development.
Rounding out the day was a lively panel discussion about precision medicine and companion diagnostics, which included CDER’s Wendy Rubinstein weighing in on the role of real-world data. Once the newly launched COVID Shield initiative has solved the lab data disharmony issue for SARS-CoV-2, she said, the issue will be solved for “the rest of medicine.” Janssen’s Mark Curran believes consortia may be the only way to come up with large enough cohorts to move from “tantalizing biology” to clinical usefulness with biomarkers that are highly predictive, accurate and replicable. For people with autoimmune diseases whose current treatment is impacting their immune status and there is no clear baseline for dose, Gemab’s Neeraj Adya said composite biomarkers may require a steroid signature to properly monitor patients’ status. Early engagement with clinicians as well as regulators is critical in overcoming potential barriers to entry, added Jason Christiansen with Boundless Bio, echoing a series of voices stressing the importance of upfront strategic planning.