The Investigator and the Ethics Committee

Thirteen months after approval, a registered letter arrives from the ethics committee. The trial's annual safety report is overdue; the committee has heard nothing from the study since the day it said yes; and the letter cites, politely, the committee's power to suspend or revoke the authorisation it granted. Nobody at the site knew a report was due. Everyone thought approval was the end of the conversation.

This module is about the conversation the approval letter actually starts. Module 5 showed you the review that produces an approval; what you have not yet seen is the running relationship behind it — what must reach the committee while the trial runs, and what must reach it when the trial ends. That traffic is not courtesy. It is what keeps the committee's protection of your participants real after the day it read the dossier.

The module runs in trial order: the committee's continuing mandate; the public registration that brackets the trial; changes to a running trial; deviations from the protocol; the safety reports the committee must see; and the end of the trial — on deadline and in public.

Learning Objectives

After this module, you can:

• Explain why ethics-committee approval starts a supervised relationship — and what the committee can do mid-trial if safety is at risk
• Apply the registration and results-publication duties: which registry, whose job, by when
• Run a change through the committee — decide what is substantial, get approval before implementation, and use the immediate-hazard exception correctly
• Handle protocol deviations: document all of them, explain and prevent recurrence of the important ones, report the hazard-driven ones
• Deliver the committee's reporting calendar on deadline — SUSARs, the annual safety report, and the trial's lifecycle notifications

The Committee Doesn't Leave the Room

The approval letter reads like an ending: reviewed, weighed, authorised. In the guideline's own architecture it is a beginning. GCP's third principle puts independent review on both sides of that letter — the trial must run in compliance with the protocol the committee approved, and the committee keeps reviewing the trial while it runs. ICH E6(R3)

How often it re-reviews is not fixed: the committee conducts continuing review at intervals appropriate to the risk your participants face, which is why a first-in-human trial hears from its committee more often than a low-risk comparison of licensed products. ICH E6(R3)

Swiss law then gives the continuing mandate teeth.

One structural note before the duties themselves: Swiss law lets the sponsor assume the investigator's notification and reporting obligations toward the committee, if the application documents say so — the arrangement Module 5 flagged when the sponsor submits the dossier. KlinV One look at the application documents settles who holds the pen; the events, dates, and records the reports are built from are yours either way.

Registered Before It Begins, Public When It Ends

Module 2 told the story of results that stayed hidden until people were harmed — and registration is the safeguard that story produced. A trial entered in a public registry before it starts has nowhere to hide a disappointing answer; the entry is a promise, made in advance, that there will be one.

Swiss primary law states the duty in one sentence: authorised clinical trials must be recorded in a public registry. HRA

The ordinance fills in the mechanics. The sponsor must register the authorised trial in a primary registry recognised by the World Health Organization or in the registry of the U.S. National Library of Medicine, and enter the Swiss dataset in the cantonal information system, from which the federal portal makes it public. Registration must happen before the trial is conducted, and within six months of approval — and the sponsor answers for the entry being accurate, complete, and kept up to date. KlinV

The ethical layer says the same thing with an earlier clock: the Declaration of Helsinki requires registration in a publicly accessible database before recruitment of the first participant, Declaration of Helsinki and GCP's ninth principle expects timely public registration and the public posting of results. ICH E6(R3)

The promise comes due at the other end of the trial: within one year of completion or premature termination, the sponsor must enter a summary of the results in the registry — including a lay summary, in plain language, in the Swiss system. KlinV

Registration is the sponsor's pen, but not something a site can ignore: your protocol, consent documents, and dossier carry the registration number, and an inspector will compare the registry's dates against your enrolment log. Selective reporting itself — choosing which true results to publish — is Module 4's ground.

Changing an Approved Trial: Substantial Amendments

Tuesday morning, the sponsor's study newsletter announces protocol version 4.0 — "final; please implement": two follow-up visits merged, one blood draw dropped. The coordinator updates the visit calendar the same day; the new schedule is friendlier for participants, and "final" sounds final. Three weeks later the monitor asks for the ethics committee's approval letter for version 4.0. There isn't one. It is still under review — and every visit conducted on the new calendar since Tuesday has run outside the approved protocol.

"Final" from the sponsor is not "approved" from the committee. The version that binds your site is the one the committee said yes to, and it changes only when the committee says yes again.

Substantial means the change could affect participants' safety, health, or rights — or it touches the trial's scientific spine: design, method, endpoints, statistical analysis. Changes of trial site, sponsor, coordinating investigator, or the investigator responsible at a trial site are on the same list — and "investigator" there means the site's responsible lead, the role Module 3 taught, not every sub-investigator on the delegation log. KlinV

Approval attaches to the change, not the whole trial: you submit the application documents the modification actually affects, with the reasons for it — not a fresh submission of the dossier. KlinV The protocol-craft side of amendments — writing them, versioning them, keeping the document coherent — is Module 15's.

Changes below that line — non-substantial modifications — do not wait for approval: they are notified to the committee in the annual safety report, which the safety section of this module comes to. KlinV

GCP runs the same two-track logic and adds the paperwork you will actually see: no changes to the protocol without prior documented approval except to eliminate an immediate hazard; revised consent materials approved by the committee before use; an updated Investigator's Brochure (the sponsor's safety dossier) supplied to the committee during the trial. ICH E6(R3)

And the Tuesday mistake? It gets unwound the day it is found: the calendar goes back to version 3.0, the sponsor is told, and every visit run on the unapproved schedule is documented as a deviation — the next section's machinery — with the committee's letter, when it comes, marking the date version 4.0 may actually start.

When the Trial Departs from the Plan: Deviations

No trial runs exactly as written. A protocol deviation is any departure, in the trial's actual conduct, from the protocol the committee approved — a visit outside its window, a test skipped, an eligibility criterion misapplied. It is neither a crime nor invisible — what decides its fate is what you do next.

One boundary first, because Module 8 set it up: a planned adaptation written into the protocol under Swiss risk-adapted GCP is not a deviation at all — it was reviewed and approved with everything else. KlinV A deviation is unplanned by definition.

The GCP architecture has three tiers. Document all of them — every deviation, however minor, goes in the record. The sponsor defines a subset as important deviations — those that could significantly affect the trial's data or a participant's rights, safety, or well-being — and for these you also explain what happened and put measures in place so it does not recur. ICH E6(R3)

For the ordinary case — the missed window, the skipped test — handling looks like this: the log records what happened, when, why, and your assessment of the impact; no committee notification follows; the monitor reviews the log at the next visit. If the same deviation keeps recurring, the pattern surfaces in the annual report's account of the trial's progress — and in questions about the site's processes.

The third tier is the only deliberate one: the investigator may deviate without waiting for approval solely to eliminate an immediate hazard to participants. Then the sequence is fixed — inform the sponsor promptly, and report the hazard, the change you implemented, and any proposed protocol amendment to the ethics committee. ICH E6(R3) Swiss law puts a number on the committee's copy: immediate safety and protective measures are notified to the committee, with the circumstances that forced them, within 7 days — 2 days in trials of the device-type products Swiss law groups under the Therapeutic Products Act. KlinV

The Safety Reports the Committee Must See

Most of what the committee learns about your running trial, it learns on a calendar. This section is that calendar — what lands on the committee's desk, from whom, by when. The definitions behind the words — what makes an event serious, how causality is assessed, what "unexpected" means against the product's known risk profile — and the parallel reporting lines to the sponsor and Swissmedic are all Module 13's; one gloss carries you through this section: a SUSAR is a suspected unexpected serious adverse reaction — a serious event, suspected to be caused by the product, that is not part of its known risk profile.

Everything starts at the site, on two legs. Serious adverse events (SAEs) and SUSARs go from you to the sponsor within 24 hours of your becoming aware KlinV — that 24-hour clock, and the sponsor's onward machinery, is Module 13's leg. Everything in the table below is the other one: what goes to the committee, and when.

*By law these duties sit with the investigator; in most industry-sponsored trials the application documents assign them to the sponsor (KlinV Art. 44a) — check yours once. The deadlines never change.

Every deadline in the table is measured from awareness, not from convenience — the clock starts when the site knows, which is why your documentation of when you knew is part of the safety system and not an administrative nicety. In Category B and C trials the same events also flow to Swissmedic, and that leg rests on the sponsor.

The annual safety report is the calendar's steady heartbeat, and the one the registered letter in this module's opening was about. Once a year, the committee receives the year's events and reactions in one list, your assessment of their severity, causality, and what they mean for participant safety — and an account of the trial's general progress. The ordinance fixes the cadence, not the anchor: in practice the clock runs from the trial's Swiss approval date — the same anchor Swissmedic's guidance gives the sponsor's parallel report. Swissmedic SAE Guidance It also carries the non-substantial modifications the amendments section parked here. GCP left the frequency of trial-status reporting to local rules; this report is Switzerland's answer. KlinV ICH E6(R3)

Ending It Properly — and in Public

Trials end more ways than one, and the same article of Swiss law tracks all of them — it is, fittingly, the article that also opened the file. The investigator notifies the committee of the first visit of the first participant in Switzerland within 30 days; the bracket that opens there closes with a completion notification. KlinV In between and at the end, the deadlines are fixed:

• Completion in Switzerland — within 30 days
• Global completion of a multinational trial — within 90 days
• Premature termination, interruption, or resumption — within 15 days, with reasons; an interruption lasting more than two years counts as a premature termination
• Summary final report to the committee — within one year of completion or premature termination, unless the protocol provides longer

Who stopped the trial decides who must tell whom, and GCP's matrix is worth knowing before you need it — participants are informed and cared for first, whoever pulled the plug:

• The investigator stops without prior sponsor agreement — institution, sponsor, committee, and authorities all get a detailed explanation
• The sponsor stops the trial — committee and authorities are promptly informed
• The committee pulls its approval — the investigator informs the institution and promptly notifies the sponsor

At completion, the committee receives a summary of the trial's outcome. ICH E6(R3)

Two duties outlive the last visit, and this module has met both: the results entry in the public registry within a year — the registration section's promise coming due — and the trial's records, which enter their long archive; retention and filing are Module 11's ground.

Module Summary

The approval letter opens a correspondence. The committee keeps reviewing what it approved and holds the power to suspend it; the registry holds the trial's public promise from before the first participant to a year past the last; every change waits for the committee's yes unless a participant's safety cannot wait; every deviation is documented and the hazard-driven ones are reported; and the safety calendar — SUSARs in 7 or 15 days, measures in 7, the annual report every year, the lifecycle notifications in 30, 90, 15 — is what keeps the committee able to do its job.

You can now:

• Explain why ethics-committee approval starts a supervised relationship — and what the committee can do mid-trial if safety is at risk
• Apply the registration and results-publication duties: which registry, whose job, by when
• Run a change through the committee — decide what is substantial, get approval before implementation, and use the immediate-hazard exception correctly
• Handle protocol deviations: document all of them, explain and prevent recurrence of the important ones, report the hazard-driven ones
• Deliver the committee's reporting calendar on deadline — SUSARs, the annual safety report, and the trial's lifecycle notifications

Which is why, at a well-run site, the registered letter from this module's opening never gets written. The committee that receives your reports on their deadlines has nothing to demand — it already knows, which is the entire point of the correspondence.