Q&A: Developing and Labeling of In-vitro Companion Diagnostic Devices for Oncology Products

April 18, 2019 Becky Hurt

Authored By: Barry Sall, RAC, FRAPS, Principal Consultant, Parexel® Consulting


On December 6, 2018 FDA released a guidance, Developing and Labeling In vitro Companion Diagnostic Devices for a Specific Group or Class of Oncology Therapeutic Products, that describes approaches for companion diagnostics indication for use can refer to a specific group or class of oncology therapeutic, rather than one specific drug.  The guidance describes how companion diagnostic (CDx) developers may be able to include broader language in for intended use statements, when supported by scientific evidence.  This policy modification recognizes that the labeling for a given therapeutic product may refer to a particular companion diagnostic, but other companion diagnostics, previously approved or approved after the therapeutic product approval may also be used to determine patient eligibility for treatment with that therapeutic agent.  A companion diagnostic with a broader intended use provides the physician with a wider range of therapeutic choices, in a timelier manner, than if more than one assay must be performed to obtain the eligibility information.


1. How does FDA define a group or class of therapeutic products?

A class or group of therapeutic products are two or more products with the same indication(s) for use, including the same disease and the same mutation.  Some therapeutic products within a group may also have other indications for use, related to other diseases, but a grouping or class considers one indication.  Although not specifically mentioned in the guidance document, FDA’s recent approval of oncology drugs for patients with tumors expressing specific genetic biomarkers, irrespective of the tissue of origin, may also be used to prepare a rationale for a group or class.

2. The guidance includes an example listing four assays that identify tumors with EGFR exon 19 deletions or exon 21 (L858R) substitution mutations.  Can any assay that detects these deletions and mutations utilize the group or class wording in their indication for use statement?

Use of the broader group or class wording may not be possible in all cases.  Assays used to identify patients eligible for a particular therapeutic agent class must all generate equivalent data when the same sample is analyzed.  When applicable, cutoff values must also be equivalent.  All technical factors including, the biomarker target, detection technology, sample matrix, sample extraction and preparation must also be equivalent.

3. When a drug developer is planning a Phase III clinical trial utilizing a CDx assay to identify subjects likely to benefit from the investigational drug, can a CDx assay for that analyte that has already been approved as a CDx for other drugs, be used in this Phase III trial?

Several factors must be considered in order to determine if the commercial assay can be used for this clinical trial.  First, as discussed in Question 1, the commercialized CDx assay should be carefully evaluated to determine if it is detecting the same analyte as specified in the Phase III clinical trial.  Next, one must determine if results of the commercial assay are consistent with the goals of the Phase III trial.  This is especially important if another (investigational) assay was used to select subjects for previous trials.  In such cases, it will likely be necessary to conduct a bridging study to demonstrate equivalent results for both assays prior to initiation of the trial.  Ideally, the CDx assay can be selected during Phase II and bridging studies can be avoided.  Sponsors should consider discussing the use of multiple CDx assays with FDA before initiating such a trial.

4. If an assay developer already has a PMA CDx approved assay with an indication for use that references specific drugs, is a new field trial necessary to utilize the broader specific group or class wording?

In some cases de novo data may be necessary.  In other cases, journal articles and internal laboratory studies may be sufficient to demonstrate that the commercial assay applies to a specific group or class of therapeutic agents.  When assessing if such a label change is possible, technical as well as regulatory points should be considered.  For example, while two assays may detect the same biomarker, they may utilize different sample matrices, different sample extraction methods or employ different cutoff values.  In these cases, the assays could identify different patient populations eligible for treatment with the drug. In some circumstances, bridging data may be necessary to demonstrate equivalent performance, while a rationale supported by scientific evidence may be sufficient in other cases. Sponsors should consider discussing the use of broader specific group or class wording with FDA before initiating a trial utilizing a marketed assay.

5. Can an in vitro diagnostic company develop a CDx assay with an intended use that mentions a group or class of therapeutic products that are already commercialized?

Yes.  The FoundationOne assay and the Oncomine Dx Target Test are both examples of assays that list multiple drugs in their indication for use statements.  Marketing applications for these types of assays must include data demonstrating acceptable assay performance.

6. Does the addition of specific group or class wording always require the preparation and approval of a new PMA?

If a diagnostic developer currently has an approved PMA for a companion diagnostic assay, a PMA supplement can be submitted, including the necessary data, to add the specific group or class wording.  If the initial assay was cleared via a 510(k) PreMarket Notification, then a new 510(k) must be submitted, to add the specific group or class wording.  It should be noted that FDA considers nearly all CDx assays as high-risk Class III medical devices that require PMA approval.  A small subset of CDx assays are Class II moderate risk medical devices that may be cleared via the simpler and more flexible 510(k) PreMarket Notification process.

7.  The guidance document mentions that clinical validity of the companion diagnostic must be demonstrated.  It also states that developers of new CDx assays for the same analyte(s) can generally leverage information in a previously approved or cleared marketing application. Does this mean that no new testing data is necessary?

It may be possible to reference data included in previous submissions.  This approach is particularly useful when the developer of a new CDx assay already has approval for a similar assay or a business relationship with the owner of a commercialized CDx assay.  One assay developer cannot reference data owned by another assay developer unless they have authorization from the original developer.

8. The guidance document states that, “FDA acknowledges that such an approach may require collaboration with therapeutic product developers…”.  If an in vitro diagnostic company requests, but is unable to obtain such cooperation, what alternatives exist?

The therapeutic product developers own their data and can distribute it as they see fit.  In some cases, they may be reluctant to share proprietary data with assay developers.  One alternative is for the therapeutic product developer to submit those data to FDA in the form of a device master file and authorize the assay developer to reference it.


The use of companion diagnostics is still a relatively recent phenomenon.  Adding the specific group or class wording to indication for use statements will provide clinicians with greater flexibility and convenience. This draft guidance is one of the initial steps to implement that policy.  Stakeholder comments as well as experience implementing this process for various companion diagnostics will provide additional inputs so this process can be further refined.  Diagnostic assay developers planning to include this the specific group or class wording in their labeling should carefully consider the process described in this guidance and consult with appropriate FDA staff during their regulatory planning process.  Therapeutic product developers may also be able to use this policy to add additional CDx assays to their drug label.

Developing and Labeling In vitro Companion Diagnostic Devices for a Specific Group or Class of Oncology Therapeutic Products, Guidance for Industry, CBER, CDER, CDRH, December 2018,  https://www.fda.gov/ucm/groups/fdagov-public/@fdagov-afda-gen/documents/document/ucm627805.pdf

Many thanks to Angela Qu, MD, PhD, Senior Director, Biomarker and Genomic Medicine for her insightful comments during the preparation of this article.

Previous Article
New post-market Surveillance requirements (PMS) for medical device manufacturers
New post-market Surveillance requirements (PMS) for medical device manufacturers

Root Cause: Why does Regulatory Approval Not Always Equate to Commercial Success?
Root Cause: Why does Regulatory Approval Not Always Equate to Commercial Success?