Drug developers are asking the US Food and Drug Administration (FDA) for more details on definitions, methodologies, and use cases for employing circulating tumor DNA (ctDNA) as a biomarker or an early endpoint in early-stage solid tumor clinical trials.
FDA issued draft guidance on the topic in May 2022, including ctDNA for patient selection based on molecular alteration, ctDNA molecular residual disease for patient enrichment and ctDNA as a measure of response. The guidance also addressed assay and investigational device considerations.
Definitions, methodologies
In public comments on the draft guidance, GSK asked the agency for more details on assay validation considerations, specific definitions for terms like baseline, pre-treatment, and post-treatment and for more information on what studies are needed to establish concordance between ctDNA and tumor assays for use in patient selection/enrichment. Specifically, GSK sought clarification on whether FDA would require the use of prospective study samples for concordance studies of ctDNA and tumor assays, or if retrospective studies in clinical samples would be acceptable.
GSK urged FDA to allow clinical trials to enroll patients using only a validated ctDNA assay, rather than performing reflex testing in tissue, which would be “time consuming, burdensome, and in some cases invasive.” GSK also noted that ctDNA provides a real-time status on tumor progression, rather than an archival biopsy, which could be months or years old.
Genentech called on the agency to provide clarity around potential use cases for ctDNA in early-stage clinical studies, including specific methodological approaches that could be applicable to single or multiple use cases, and more details on common methodological approaches and evidence requirements that could be applicable to any use of ctDNA in an early-stage clinical study.
Genentech also asked for more specific guidance from FDA on how to overcome challenges with ctDNA assays, such as the variability across assays. “For example, are there preferred approaches (e.g. panel-based approaches) or minimum requirements that can be established that better enable comparison across assays?” Genentech asked in comments to the agency.
Friends of Cancer Research asked FDA to further define the clinical metrics for reporting ctDNA as an early endpoint. For instance, should the ctDNA endpoint be based on clearance in ctDNA (detectable or undetectable) or as percent change from baseline?
“The use of these metrics may differ depending on the indication (e.g., biology of the cancer, tumor stage, drug class, administration schedule, timing of measurement, definitive treatment, and medical necessity) and therefore should be defined within the context of the use case,” the group commented. “Statistical considerations, including baseline measurement versus change from baseline and aligned measurement outputs across assays (e.g., VAF vs mtm/mL) would also be helpful to highlight.”
Role of tumor tissue
Both Labcorp and the Blood Profiling Atlas in Cancer (BLOODPAC) Consortium praised the agency for outlining the potential regulatory and clinical uses for ctDNA as a biomarker in the early-stage solid tumor drug development but questioned the idea within the guidance that assessing tumor tissue is a true measure of patient metastasis. Specifically, Labcorp asked FDA to revise the recommendation that when no variants are detected in ctDNA, tumor testing may need to be performed to confirm a negative result. The company suggested that the sensitivity of the ctDNA assay in detecting variants of clinical interest be evaluated using Limits of Detection studies with either clinically or functionally characterized contrived plasma samples.
“We recommend the agency consider ctDNA assays as complementary to tissue-based assessments, each with their own unique advantages and disadvantages depending on the clinical scenario,” Labcorp commented.
Reference materials
The draft guidance also calls for the development of a set of reference materials that would allow for comparability across multiple molecular residual disease (MRD) assays but does not specify who would develop the materials. Labcorp urged FDA to enlist qualified standards organizations in the creation of reference materials to assist in standardization of ctDNA assays, rather than leave it to individual diagnostic developers.
“Rather than put the onus on the diagnostic developer, standards organizations such as NIST and WHO should be engaged to develop appropriate reference materials that could be distributed to diagnostic developers to aid in analytical validation of new tests,” Labcorp wrote. “Collaborative groups, such as the BLOODPAC, could be engaged to perform comparability studies using proposed reference materials. However, the certification, storage, and disbursement of those materials should be handled by a qualified standards organization.”
