The regulatory function defined: what investigator sites must do, and who is responsible

Before you build, you survey

Consider what happens at a site running 18 active studies when the institutional review board approval for one protocol lapses by a single day. The investigator cannot enroll the next participant who was scheduled for screening that morning. But the cascade does not stop there. The coordinator who was preparing informed consent materials for that participant now has nothing to consent to. The data manager expecting screening data has a gap. The sponsor's enrollment timeline slips. And the investigator -- who signed the FDA Form 1572 attesting to IRB oversight -- is now technically in noncompliance with 21 CFR 56.103, regardless of whether anyone at the site noticed the lapse.

One missed renewal date. Six downstream failures. This is not a hypothetical constructed to alarm you -- it is an unremarkable Tuesday at a busy research site that lacks a regulatory system.

If you are stepping into a regulatory coordinator role, your instinct will be to start doing things: filing submissions, organizing binders, tracking deadlines. That instinct is understandable, and it is premature. Before you can build an effective regulatory operation, you must first understand what the operation is for -- the complete set of regulatory obligations that falls on an investigator site, how those obligations depend on one another, and where the boundary sits between what the investigator must do personally and what can be operationally delegated. That is the work of this lesson.

The obligation landscape: six functional categories

ICH E6(R3) Annex 1, Sections 2.1 through 2.13, enumerates the investigator's responsibilities across more than 60 discrete requirements. If you read them in order -- qualification, resources, responsibilities, communication, compliance, termination, safety, consent, records, reports -- the sheer volume can feel arbitrary, as though each section were an isolated rule dropped onto the site from a great height.

But these are not isolated rules. They are the components of a regulatory system. And the first step toward operating that system intelligently is to reorganize these obligations by function -- by what they actually require the site to do -- rather than by the section numbers that happen to contain them.

I find it useful to think of the site's regulatory obligations as falling into six functional categories. This is not the only way to organize them, and ICH did not design these sections with this taxonomy in mind. But in over three decades of teaching investigators and site staff, this categorization reflects how obligations actually generate work at the site level -- which is ultimately what matters when you are the person responsible for making sure the work gets done.

What the categories reveal

The table above is worth studying carefully, because the categorization tells you something that reading the sections sequentially does not: the regulatory function is not primarily about paperwork. It is about maintaining six interconnected operational capacities simultaneously, across every active study, for the entire duration of each trial.

Qualification is not a one-time checkbox. Section 2.2.2 requires sufficient qualified staff for the foreseen duration of the trial. When a sub-investigator leaves the site mid-study, the qualification obligation re-activates: the delegation log must be updated, the replacement must be trained, and the IRB/IEC must be notified if required by local policy. A regulatory coordinator who thinks of qualification as "something we handled at study start-up" will miss this entirely.

is, in my view, the most underappreciated category. Section 2.3.1 establishes a principle that shapes everything else the site does: the investigator may delegate activities, but retains and must maintain proportionate to the importance of the data and the risks to participant safety. This is not bureaucratic language. It means that every delegated task -- every consent discussion conducted by a coordinator, every adverse event assessed by a sub-investigator, every regulatory submission prepared by an RC -- exists within an accountability structure that traces back to one person. Understanding this structure is essential to understanding what the RC role is and what it is not.

Cascading dependencies: why the system metaphor matters

The six categories are not independent departments that happen to coexist at the same site. They are operationally dependent on one another in ways that create cascading failure chains. When one domain fails, it does not simply create a problem in that domain -- it forces stops or degradations in others.

This is the single most important insight for a regulatory coordinator to internalize. The regulatory function is a system, and systems fail in systemic ways.

Three dependency chains illustrate this principle with particular clarity.

IRB/IEC approval is the master dependency. Per Section 2.4.2, the site must have documented IRB/IEC approval before initiating a trial. But this is not merely a start-up requirement. The approval must remain current throughout the trial's duration. If continuing review approval lapses -- even by a single day -- the site cannot enroll new participants, cannot obtain new consents, and under many institutional policies, cannot administer investigational product to already-enrolled participants. The consent domain, the safety domain, and the records domain all depend on the IRB/IEC communication domain being current. One missed renewal date halts an entire study.

Qualification failures propagate through delegation. If a sub-investigator's medical license expires, or if a coordinator's required GCP training lapses, the delegation under Section 2.3.1 becomes unsupported. That person can no longer perform delegated trial activities until the qualification gap is resolved. If that person was the only staff member trained to obtain informed consent for a particular study, the consent process stops. If that person was responsible for assessing adverse events, the safety reporting chain weakens. The qualification domain feeds the oversight domain, which feeds every operational activity at the site.

The nondelegable boundary: where the investigator's personal responsibility ends and the RC's operational domain begins

Section 2.3.1 is one of the most consequential passages in all of E6(R3) for understanding the regulatory coordinator role. It establishes two principles simultaneously: the investigator may delegate trial-related activities, and the investigator retains ultimate responsibility regardless of delegation. These principles are not in tension. They define a boundary.

On one side of that boundary sits a set of responsibilities that the investigator must fulfill personally -- responsibilities that cannot be delegated because they require the investigator's professional judgment, medical expertise, or personal attestation. On the other side sits the vast operational infrastructure required to those responsibilities: the tracking, the filing, the monitoring, the coordinating, the system-building that makes the investigator's personal responsibilities possible to discharge across a portfolio of concurrent studies.

The nondelegable/delegable boundary is not a limitation on the RC's role. It is the definition of the RC's role. Everything on the delegable side of that boundary -- the submission tracking, the document management, the training oversight, the reporting infrastructure, the deadline monitoring, the inspection readiness -- is operational work that the investigator cannot realistically perform personally across a portfolio of studies. Someone must build and maintain the systems that make the investigator's nondelegable responsibilities possible to fulfill. That someone is the regulatory coordinator.

And this framing matters for another reason that I think is often overlooked: it protects both parties. The investigator cannot later claim that regulatory failures were "the RC's responsibility" -- because Section 2.3.1 makes clear that the investigator retains ultimate responsibility regardless of delegation. And the RC cannot be placed in the position of making medical or clinical judgments that exceed their qualification -- because the nondelegable boundary reserves those decisions for the investigator. The line is protective in both directions.

The obligation landscape as a system

Step back now and consider the full picture. An investigator site conducting clinical trials under ICH E6(R3) must simultaneously maintain:

• Qualified, trained personnel with documented delegations for every active study
• Active IRB/IEC approval with continuous communication across initial review, continuing review, amendments, safety reports, and status updates
• Protocol compliance with documented deviations and immediate hazard procedures
• A functioning safety reporting infrastructure with immediate SAE notification and investigator causality assessment

Key takeaways

The investigator site's regulatory obligations under ICH E6(R3) are not a collection of independent requirements. They form an interconnected system organized around six functional capacities: qualification, oversight and delegation, IRB/IEC communication, records and data integrity, safety and medical care, and informed consent. These capacities depend on one another -- a failure in one domain cascades into enforced stops in others.

The investigator retains ultimate responsibility for all trial conduct under Section 2.3.1, but may delegate operational activities to qualified staff with proportionate oversight. The regulatory coordinator works on the delegable side of this boundary, building and maintaining the operational infrastructure that makes it possible for the investigator to fulfill nondelegable responsibilities across a multi-study portfolio.

Understanding this complete obligation landscape -- as a system, not a checklist -- is the prerequisite for everything that follows in this course.