8 min
A biomarker test that can identify patients most likely to benefit from a drug raises the probability of success for a clinical trial and leads to a higher return on investment.1 Although the FDA has approved several targeted oncology drugs without a companion diagnostic test (CDx) in recent years (they required post-market assay development), a missing assay could negatively impact market uptake, patient safety, and access.2
However, developing a CDx to stratify patients often requires sponsors to transition from a research-grade assay to a clinical-grade one. This is a multidisciplinary resource management challenge, especially for emerging companies. Therefore, they must determine the most efficient, least burdensome regulatory path to minimize the time and resources needed.
We often work with companies preparing for pre-Investigational New Drug (pre-IND) meetings with the FDA who suspect they will need a CDx to select patients during development but have limited data.
Usually, we advise clients to keep discussions about the assay under the IND. Once Phase 1 or 2 trials demonstrate the need for a commercial CDx, they may need to file an Investigational Device Exemption (IDE) with the Center for Radiological Health and Devices (CRDH). Sometimes, an IDE may be required from the start.
From decades of advising CDx developers, we have refined five strategies for maximizing the benefits of IND meetings and streamlining the development process.
Sponsors need precise goals for each early interaction with the FDA. For example, if you plan to use a test to stratify patients in the Phase 1 trial, there are three critical goals for the pre-IND meeting.
First, ensure the FDA agrees you have done adequate analytical validation of your proposed test, especially around the cutoff. How are you defining who is biomarker positive and who is negative? Unless you have a firm agreement with regulators, you risk receiving a clinical hold letter in response to your IND.
Second, make sure the Agency concurs with your assessment of the risks of the testing device, whether it is a lab-developed test (LDT), a marketed in vitro diagnostic (IVD), or a new custom IVD. If the Agency deems the use of the LDT or IVD in your study as a “significant risk” per 21 CFR 812.3(m), they may require you to file an IDE application with CRDH. Then, the Center for Drug Evaluation Research (CDER) or the Center for Biologic Evaluation Research (CBER) will oversee the Phase 1 trial with CDRH.
Third, get detailed feedback on your Phase 1 study design, including the number and timing of tests and how the results will be used.
Recently, we helped a sponsor explain why the risks of their testing device were sufficiently low that the Phase 1 study did not need CRDH oversight. This required a deep understanding of the biomarker and its relationship to the disease, including how it predicts treatment response. We advised the sponsor to use publicly available databases of genetic variants and medical literature searches instead of original research, significantly reducing the regulatory burden.
The agency does not want to design your study. It prefers to offer targeted comments. So, it’s better to make proposals and ask for a response instead of asking open-ended questions in early-stage CDx regulatory meetings.
Once the agency’s advice has been recorded, if you do not follow it, you might need to explain why in a future meeting or interaction with the agency. The pre-IND meeting is just one of many milestones. It will not be your only interaction, so be careful only to share what the FDA needs to know at this stage, not necessarily more.
Think about the medicinal product-CDx co-development story as you progress. If you reveal that you don’t yet have a firm plan for some aspect of your program, you may invite the agency to increase your burden with onerous expectations.
Recently, a sponsor asked us for help preparing for a pre-IND meeting. They wanted their Phase 1 sites to perform an assay to explore whether they would need a CDx. They planned to ask the FDA how to proceed. However, based on our interactions with the FDA, we knew that site-to-site variability is a common problem for biomarker testing. One site may use next-generation sequencing (NGS) for molecular tests, while another may use polymerase chain reaction (PCR) technology. There may be varying levels of analytical validation of those assays. The FDA wants to know about your plan to qualify those sites. You should also establish minimum performance requirements to demonstrate adequate analytical validation and minimize site-to-site variability.
When a company does not have a fully detailed plan, it is still possible to present enough that the FDA will conclude that you have an acceptable plan for the stage of development. We helped the sponsor proactively address the FDA’s concerns with a plan that showed they had prioritized analytical validation and would generate adequate data for the next assay development stage. The FDA accepted their outline plan with few comments, saving them the need to meet with the agency again to resolve outstanding issues.
Sponsors can reduce the risk of having their plans changed by the FDA after the fact by presenting a high-level one first. For example, human factor (HF) studies mitigate risk by ensuring that assay labeling is understandable to users, such as healthcare providers or laboratory technicians, and adequately prevents the most common use errors, such as adding too much liquid when prepping samples or incubating them for too long.
Often, sponsors use specialized firms to design and conduct HF studies. This can be expensive and time-consuming. Rather than hiring them to design a study and having the FDA critique it after the fact, present the FDA with a high-level summary of the HF study plan. The plan should be rigorous, not raise any red flags, and draw a boundary around the issues to avoid eliciting tangential feedback or requirements. Then, let agency reviewers fill in the blanks with their recommendations. At the same time, gauge their responses and body language (if the meeting is in person or a videoconference) and note their questions.
Armed with that information, you can engage an HF firm and ask them to write a protocol based on the FDA feedback with a much lower risk of revisiting it.
Recently, we helped a large pharmaceutical company complete their IND filing on a tight deadline. They planned to use an in-house assay for their Phase 1 trial to select and stratify patients but did not believe they needed to file an IDE with CDRH. However, they knew the FDA might disagree based on how the assay will be used in the study. Meanwhile, they were facing a firm deadline for their Phase 1 first patient in (FPI), they could not wait for regulators to confirm their approach, and they did not have enough time to write an IDE application. During the FDA review of the IND, we helped them prepare a complete IDE application in case it was required. In the event, the FDA agreed with them and allowed the IND to proceed. But they were ready to respond if things had gone the other way.
Because timing is paramount to co-developing a CDx, investing in a backup plan often pays off.
Sponsors must conduct an exploratory study early to determine whether they need a biomarker. By Phase 2, ideally, they must know whether they need a CDx to stratify patients or whether they can test efficacy in an all-comers trial. By Phase 3, they need a validated, clinical-grade selection tool.
However, sometimes, the FDA can require a CDx unexpectedly. Recently, we helped a company on the brink of initiating a Phase 3 combination therapy trial adapt its development plan. The sponsor thought their product was effective in all comers for a given cancer, and they did not need to stratify patients by biomarker status. However, the FDA maintained that the course of treatment was different for biomarker-positive and -negative patients. By chance, there was a commercially available IVD kit for this biomarker, but it was only validated to identify biomarker-positive patients. We supported the sponsor’s discussions with the kit developer to validate it for biomarker-negative results. We then advised the client to submit a request for a Type C meeting with the FDA with the kit developer in attendance. Fortunately, the sponsor had banked patient samples from earlier studies and could adapt to the last-minute requirement.
Planning for unexpected developments and quickly adapting is more efficient and less burdensome than spending months (or years) recovering from setbacks. In this case, having a bank of samples helped our client recover and proceed swiftly to Phase 3.
Co-developing a precision oncology CDx can unnecessarily consume a lot of time and resources if a sponsor isn’t well prepared. At Parexel, our team of former regulators, including device regulators, helps clients prepare for meetings and interpret FDA feedback to streamline the development and approval processes.
Contributing Experts