What CBER expects in an INTERACT briefing document

The INTERACT program (Initial Targeted Engagement for Regulatory Advice on CBER products) provides sponsors of novel biological products an early, informal exchange with the Center for Biologics Evaluation and Research in advance of the pre-IND meeting. For cell and gene therapy programs, particularly those originating in academic laboratories, it is frequently the Agency's first encounter with the product. The meeting is nonbinding and the Agency's comments are advisory, yet the written feedback shapes every subsequent decision concerning the nonclinical package, the control strategy, and the clinical plan. Whether the meeting is granted, and whether its outcome is actionable, is determined almost entirely by the briefing document.

Timing: after the product is defined, before the development plan is committed

INTERACT occupies a narrow window. The Agency expects a defined product: a selected construct or editor and guide, a delivery modality, a manufacturing concept, and preliminary proof of concept in a disease-relevant system. A request submitted before those elements are fixed will be judged premature, because no specific program exists on which to advise. Conversely, once definitive IND-enabling studies have been designed and contracted, the pre-IND meeting is the appropriate venue, and the Agency will direct the sponsor accordingly. CBER grants one INTERACT meeting per product; a second request for the same product is routinely denied.

The operative test is whether Agency feedback could still alter the design of the IND-enabling toxicology study, the proof-of-concept model, or the release-testing concept. If so, the program is appropriately staged. If those decisions are already committed, the window has closed.

The questions define the submission

The meeting request is granted on the basis of the questions it poses, and the review team is constituted to answer those questions. The Agency has stated explicitly that questions are not to be added or materially altered between the request and the meeting package. Consequently, the questions must be finalized before any other section is drafted; the remainder of the document exists to supply each question with the context required to answer it.

Three properties distinguish a question that elicits substantive guidance from one that elicits a restatement of published guidance.

• It asserts a position. "Does the Agency concur that patient-derived iPSC-RPE cultures constitute an acceptable basis for proof of concept in the absence of an animal model carrying the target variant?" elicits concurrence, non-concurrence, or concurrence with conditions. "What model does the Agency recommend?" elicits a restatement of the guidance.
• It is answerable at this stage. Questions concerning the design of definitive GLP toxicology studies, specific clinical protocol elements, or statistical analysis plans belong to the pre-IND meeting and will be deferred to that forum. INTERACT questions concern the development approach: the proof-of-concept strategy, the scope of the nonclinical package, the manufacturing and control concept, the treatment of novel excipients or starting materials, and the general structure of the first-in-human study.
• It is one of a limited number. The Agency requests that questions be confined to what can be addressed within the allotted meeting time. A package carrying a dozen questions receives a dozen abbreviated answers. Five or six rigorously constructed questions, ordered by consequence, is characteristic of a well-prepared package for a cell or gene therapy program with genuinely novel features.

Required content by section

Product description. State the product's identity, its manufacturing concept, its route and method of administration, and the attributes that distinguish it from precedent. Novelty is the premise of the meeting; a product that follows established precedent in every respect may be judged not to warrant INTERACT. For a genome-editing product, the description comprises the editor, the guide RNA, the delivery vehicle, the target cell population, and the intended edit, together with the preliminary editing-efficiency and specificity data that support them. For a cellular product, it comprises the cell source, the genetic modification, the manufacturing concept, and the formulation.

Chemistry, manufacturing, and controls. INTERACT is the appropriate forum for CMC questions that would otherwise emerge as deficiencies at IND: a novel lipid excipient without prior clinical exposure, a starting material of uncertain provenance or donor eligibility, a potency concept dependent on an assay not yet developed, or a process expected to change between the nonclinical and clinical batches. The section should present the manufacturing concept at the resolution of a process flow diagram, identify the critical quality attributes as currently understood, and set out the release-testing and stability approach as a proposal for Agency comment rather than a settled plan.

Pharmacology and toxicology. The section should present the proof-of-concept model and its rationale, the proposed approach to biodistribution and toxicology, and, for editing products, the off-target assessment strategy. Where no animal model of the disease exists, the document must make the affirmative case for the alternative: the relevance of the model system, the data obtained in it, and the limitations acknowledged. Noting the absence of a model and soliciting direction is not a substitute. Definitive study designs are neither expected at this stage nor appropriately presented as settled.

Clinical. The clinical section is deliberately high-level: the indication, the intended population, the natural history, the current standard of care, and the general structure of the first-in-human study. Dose-escalation schemes, eligibility criteria, and endpoint definitions are pre-IND matters. The Agency will engage at INTERACT on whether the target population and the development plan cohere with the product's mechanism and the available evidence, and, for rare diseases, whether the program is positioned to invoke the evidence pathways available to it.

Deficiencies that lead to denial or an unproductive meeting

• Presentation of alternatives. INTERACT is not a forum in which the sponsor presents options for the Agency to select among; it is a forum for feedback on a proposed program. A document that equivocates on its own development plan receives an equivocal reply.
• Subordination of the critical question. The question the sponsor is most reluctant to pose is typically the one most in need of an answer. Placing it first, supported by the strongest argument the data permit, is the only means of ensuring it receives adequate discussion.
• Volume in place of focus. The package is expected to be concise. A focused document of a few dozen pages, organized by question, serves the reviewer better than a comprehensive one that reproduces the laboratory's publication record.
• Scope outside the program's remit. Requests are denied when the feedback sought is not appropriate to INTERACT: the program is too early or too advanced, a prior INTERACT meeting has been held for the product, or the product has ample regulatory precedent. The cover letter should state why this product, at this stage, warrants the meeting.

Following the meeting

CBER provides written comments, and those comments are the durable product of the exercise. Each should be carried forward into the pre-IND package with an explicit account of its disposition: a study design revised, an assay added, or a position maintained with additional justification. Reviewers at pre-IND will have the INTERACT record before them; a package that demonstrates the earlier advice was incorporated is received differently from one that appears to have disregarded it.

Crawling Stone's role

Crawling Stone Consulting authors INTERACT and Pre-IND briefing documents and chairs the meetings on the sponsor's behalf. Most recently, Crawling Stone carried two LNP-delivered genome-editing programs in inherited retinal disease through INTERACT, posing questions on novel-excipient CMC strategy, off-target assessment, potency assurance, and the use of patient-derived cellular models as proof of concept in the absence of an animal model. Both meetings secured Agency alignment on the preclinical model, the nonclinical package, and the CMC development strategy. Sponsors approaching this window should settle their questions before drafting a page of the document. Engage Crawling Stone to define them.

This article is general in nature and does not constitute regulatory advice for any specific program. Agency procedures and expectations change; consult current FDA guidance and the CBER Office of Therapeutic Products INTERACT page.