Two September developments deserve to be read together by a biotech choosing its nonclinical programme or a CRO developing alternative testing methods. They do different things, and that distinction should survive every summary written about them.
On 22 September, FDA published a direct final rule updating nonclinical testing terminology. Earlier in the month, EMA opened a voluntary route for discussing data from new approach methodologies, or NAMs, outside a marketing-authorisation application. Neither announcement establishes that a particular method is sufficient for a particular development decision. FDA rule · EMA pilot
That leaves a substantial task for the nonclinical writer: stating the precise claim the sponsor wants the evidence to support.
What changed, and what a programme can infer
FDA's rule replaces specified references to animal tests or studies with "nonclinical" tests or studies, adds a definition of those terms and makes related amendments. FDA describes these as "noncontroversial changes in terminology that are not expected to affect industry practice," and the rule "adds no new requirements." It is scheduled to take effect on 4 February 2027. Comments are due by 7 December 2026, and FDA will withdraw the rule if it receives timely significant adverse comments. It should not appear in a development plan as an exemption from animal studies. Federal Register notice
EMA's pilot offers independent, non-binding feedback on the regulatory readiness of a method. Companies, CROs, method developers and academic laboratories can apply within a defined scope, and eligible applicants are invited to submit a full briefing package. Applications are expected to stay open until September 2027. Feedback may point towards further scientific or qualification advice, so participation should be described accurately when the programme later appears in a partner presentation or a submission.
EMA's pilot information document describes the process as non-decisional: taking part does not qualify a method. It asks the briefing package to set out the proposed context of use, the endpoint and biological mechanism, the method's characterisation or validation, and its limitations, and to say whether the method will stand alone, form part of a battery or contribute to a weight-of-evidence approach. EMA pilot information document, page 7
The team therefore needs to identify which unresolved decision an agency discussion could help with. Without that question, preparing a package can turn into a long description of the technology.
Start with the sentence the evidence is meant to support
Imagine a CRO with an in vitro assay intended to inform a defined safety question. This is an illustrative case, not a description of a method accepted under either initiative.
An assay report explains the procedure and presents reproducible results. The sponsor's submission has to connect those results to the proposed use: which endpoint was measured, how it relates to the concern in this programme, under what conditions that relationship holds and what remains uncertain.
We would begin the drafting discussion by asking the scientific lead to write the proposed conclusion in one sentence, and then ask which result supports each part of it. If the sentence reaches beyond what was measured, the team has found a scientific question to resolve before anyone produces a polished briefing document.
A careful draft can also show where two people mean different things by the same phrase. "Supports safety assessment" might mean useful alongside other evidence to one author and sufficient to replace a study to another. Those positions carry different evidentiary burdens, and the final text has to say which one the sponsor is advancing.
The CRO report and the sponsor's argument
For a specialist CRO, the handoff to the sponsor is worth agreeing explicitly: who explains the method's limitations, who assesses its relevance to the sponsor's product and who integrates the findings with the rest of the nonclinical programme.
The writer can connect those contributions and spot unsupported steps in the argument. The scientific owners still have to decide whether the inference is justified, and a writer handed a report and a deadline without access to them is in a difficult position. It also helps to settle early whether the commission covers only the assay report or the briefing package and any follow-up questions as well, so the work can be staffed and priced accordingly.
Keep the limitations when the text travels
Our recommendation is to maintain one agreed account of what the method supports and where it stops. Later documents can shorten it, but they should keep the qualifications that matter to the conclusion.
Suppose a briefing package limits a proposed use to a defined range of conditions. If the next programme summary drops that limitation, the organisation has quietly widened its claim without generating any new evidence. The inconsistency can start with an innocent attempt to shorten a paragraph, which is one reason we keep saying the document was never the source of truth.
For a biotech deciding whether to invest in an alternative method, a useful early deliverable is a short, scientifically reviewed statement of the intended claim, the support available and the questions still open. It gives management something concrete to assess before committing to the longer package, and it gives the regulatory writer a clear argument to build.