The FDA just published a post on its “Voices” blog telegraphing to the industry that, as the title suggests, “Good Clinical Practices Are Not Optional.”
Four officials signed it: the acting directors of CDER and CBER, the director of CDRH, and the director of the Oncology Center of Excellence. Drugs, biologics, and devices, in one message, so everyone involved in clinical work should pay attention to this one.
Before we dig into it, this is a blog post, not a guidance document. The FDA says so directly, calling this the logical application of existing law rather than a new policy. It’s best framed as the FDA telling the industry where enforcement attention is moving and what a program should therefore look like from that vantage point.
If this sparks a need for GCP support in any dimension, get in touch with us to get paired with an SME from our bench. We support firms across basically any axis of clinical work, especially clinical quality operations staff augmentation and auditing.
And be sure to watch our conversation on GCP compliance and inspection readiness if you haven’t already:
The two core mechanisms behind the FDA’s warning
The FDA goes through two different mechanisms to reject data in its post, and the distinction matters for how you assess risk.
The first concerns validation. If the FDA cannot validate that data were collected in accordance with GCP, including at sites where its ability to perform in-person inspections is constrained or denied, FDA regulations allow the agency to refuse to accept that evidence in support of a marketing application. Note what triggers this: not a finding of bad data, but an inability to confirm good data. A site that blocks an inspection produces the same outcome as a site with problems, to give one example.
The second concerns falsified or otherwise invalid data, and the post specifically includes duplicated data. Where such data are identified during review or once a product is on the U.S. market, the FDA says it has the authority to exclude that data from consideration and, if the remaining evidence cannot support the marketing authorization, to deny, withhold, or rescind it. Read that sequence carefully! Elimination comes first, and the authorization consequence follows only if what remains cannot carry the application. Also note “rescind” and “once on the U.S. market.” This is not solely a pre-approval exposure!
For devices, the post frames the same principle through evidence standards: all evidence in a device submission must be credible, including clinical evidence, and the FDA relies on valid scientific evidence to determine whether there is reasonable assurance that a device is safe and effective.
The access problem the agency is encountering
Maybe the most concrete passage in the post is the FDA describing what it has run into around access. For some foreign inspections of clinical trial sites, the agency has said it's been denied access outright, or told that inspection would require signing agreements that restrict the scope or conduct of the inspection, or that attest to geopolitical principles unrelated to the inspection.
The FDA is pretty clear here: it will not sign those agreements! Which means the exposure at such a site has nothing to do with the quality of the science being done there. Good data at an inaccessible site is still data that the FDA may decline to accept.
We’ve heard auditors working in this space often describe the sponsor-side version of the same wall: being refused access to electronic medical records and source documents at some foreign sites unless they agreed to conditions no auditor could reasonably accept. That experience points to something useful, though. If your own monitors or auditors have hit access restrictions at a site, you already have evidence about what the FDA might (or likely will in some cases) encounter. That friction is a leading indicator sitting in your own files. If you’ve encountered it, address it.
The FDA’s framing of the obligation is actually worth quoting in your internal materials to make sure it accommodates:
Before the FDA opens the door to let a product reach the American market, sponsors and investigators must open their doors to the FDA.
That means access to sites and records, including source data and consent documentation.
And sponsors should treat an inability to inspect or access all relevant data as a material factor in regulatory strategy rather than an administrative footnote.
What changes operationally
To our reading, there are three actions in the post that could be cast as resourcing and process changes rather than statements of principle.
1. More foreign BIMO resources
The FDA says it’s adding resources for foreign Bioresearch Monitoring inspections, including inspecting more Phase 1 and early-stage trials, and expanding the scope of some inspection assignments to cover more trials occurring at a given facility. It’s also planning updates to the risk-based criteria it uses to select sites to better reflect compliance risks in specific countries or regions.
There are two very important implications there: