On February 25, the Food and Drug Administration (FDA) will hold a public workshop on Next Generation Sequencing (NGS)-Based Oncology Panels, a highly anticipated program given the recent announcement of a “Cancer Moon Shot” spearheaded by Vice President Biden. FDA recently released a discussion paper for the workshop that hints at the role cloud-based computing and “big data” could play in FDA’s regulatory framework for laboratory-developed tests (LDTs).

In December, FDA launched precisionFDA, a “secure, cloud-based platform where participants can access and share datasets, analysis pipelines, and bioinformatics tools” (previously described on this blog.) The agency has published white papers on the analytical and clinical validation of NGS testing that explore “the potential use of databases as sources of clinical evidence in support of regulatory submissions” that could be used for NGS as well as for “any genetic test” (that is, any of the roughly 60,000 LDTs in use today.)

The new discussion paper describes a statement for the “intended use of NGS-based oncology panels” that is “intended to be a starting point for discussion.” The statement includes three parts:

  1. A statement that the [device name] is a qualitative in vitro diagnostic test that uses high throughput parallel sequencing technology intended to detect sequence variations using the [instrument name]. The [device name] is indicated as an aid in characterizing sequence variations in [xx genes] on [DNA and/or RNA] isolated from [specimen type] specimens.
  2. “Table 1” which would list genetic variants and the associated targeted therapies for which the “device is also indicated as a companion diagnostic to aid in selecting oncology patients for treatment” in accordance with the approved labeling of the therapy. Any test results not listed on Table 1 are “only indicated for use in patients who have already been considered for all appropriate therapies.”
  3. “Table 2,” which would list additional genetic variants for which “analytical performance has been established.” Importantly, Table 2 would not list any associated targeted therapies, and would caveat that the device is “not indicated to be used for standalone diagnostic purposes, screening, monitoring, risk assessment, or prognosis.”

In essence, this three-part system for labeling the intended uses of an NGS oncology panel circumvents the concern that NGS is “indication-blind.” The precisionFDA cloud-based platform would play a central role in laboratories’ demonstration of both clinical and analytical validity, as it permits the comparison of sequencing results to reference standards and calculation of the test’s sensitivity and specificity.