Who Decides When The Comparator Changes?

By Angeliki Kotsianti, MD, Ph.D.

Comparator sourcing is usually run as a procurement task. Several of the decisions inside it affect participants directly, and they are often made without a clinician involved. Most clinical supply managers will recognize this situation. A Phase 3 study is about two-thirds enrolled when the comparator, bought from a competitor through a sourcing vendor, starts to run low in its original source country. The vendor finds stock in a second country. It arrives with its certificate of analysis, goes through release and relabeling, and is loaded into the IRT in time. Nobody on the study team would call this a failure, and on the supply metrics it isn't one.

What usually doesn't happen at that point is a conversation about whether the second country's product is the same medicine as the first, whether participants already randomized should be switched to it, and whose call that is. Sometimes those questions come back months later as a protocol amendment.

In the two-part series Hedley Rees and I wrote for Clinical Supply Leader in August, we argued that a supply problem becomes a clinical one once it reaches a site.^1,2 Comparators are a particularly clear example, because the sponsor doesn't make the product and has limited control over where it comes from.

How Comparators Are Bought And What Those Decisions Mean For Participants

Most sponsors buy comparators through a specialist sourcing vendor or a CRO offering the same service. Which country the product comes from depends on practical things: price, how much stock is available and for how long, whether the manufacturer is willing to supply a competitor, and whether the product comes with documentation showing where it has been. Those are sensible criteria for a purchase. The difficulty is that some of them have clinical consequences the purchasing process isn't designed to weigh.

Source market is the most obvious. A medicine sold under the same nonproprietary name can differ between countries in its excipients, strengths, presentation and approved dosing, and sometimes in its labeled indications. The choice of market therefore also determines which version of the product the investigational drug is compared against. Regulators pay attention to this. In U.S. biosimilar development, for example, the expectation is generally a comparison with the U.S.-licensed reference product, and using a product sourced elsewhere means providing bridging data.³ In the EU, a comparator authorized outside the EEA can be used in some biosimilar studies, but only with data bridging it to the EU-authorized reference product.⁴

It gets harder once the study is running. If resupply comes from a second market, some participants will start on one version and finish on another. That may be perfectly acceptable, but it is a judgment about those participants and about the comparison, and in my experience it is rarely put in front of a clinician. There is also a question for the participants themselves. If the pen, the device, or the instructions change, do they need to be told, retrained, or asked to consent again? The supply plan can't answer that.

Blinding And Shelf Life

Blinding adds another layer. Over-encapsulating a tablet can change how it dissolves, the backfill excipient may matter for some patients, and the repackaged product needs its own stability data. Sponsors generally do this comparability work well. What is less clear in many organizations is who signs off that the evidence is enough. That sign-off is really a decision about whether the control arm still behaves like the medicine patients would receive in practice.

Shelf life then works against long studies. Comparators bought on the commercial market have usually used up part of their shelf life before they arrive, and repackaging can shorten it further. Every expiry means resupply, and with each resupply the lot, the source, or occasionally the presentation may change.

Supply teams plan for all of this carefully. Whether each change is acceptable for participants already on treatment tends to be addressed only when someone happens to raise it.

When The Comparator Isn't Available

Comparators are marketed medicines, so they are exposed to the same shortages, reformulations, and market withdrawals as anything else on a pharmacy shelf. When that happens during a study, the sponsor has to decide whether participants continue on an alternative source, pause treatment or switch products, and whether it still makes sense to randomize new patients to that arm. Those are clinical and scientific decisions, even though they start as supply events.

Earlier in my career, I worked on allocation decisions for marketed medicines during drug shortages. The stock data were usually available early. What slowed decisions down was that the clinical criteria for making them had not been agreed beforehand. A single trial is a much smaller setting, but the lesson carries over, and trial participants have an additional expectation: when they consented, they were told their treatment had been planned.

Comparator Changes And Protocol Amendments

Amendments have become routine. The most recent Tufts CSDD benchmark found that 76% of Phase 1 to 4 protocols had at least one amendment, with an average of 3.3 per protocol and around 260 days to implement each.⁵ About three-quarters were judged unavoidable, mainly because of regulatory requests and changes in study strategy. The remainder were at least partly within the sponsor's control.

Rachel Grabenhofer recently asked on this site whether the operational data clinical supply teams collect could help catch the assumptions behind those avoidable amendments earlier.⁶ I think comparators are one of the better places to try. Changing a comparator's source, presentation, or product can require an amendment or a regulatory modification, and these changes are more predictable than most amendment triggers. Supply teams already hold records of how often comparators were re-sourced, ran short, or expired during past studies. They also know what a change costs administratively: in the EU, moving to a new comparator source will often require a substantial modification, while in the U.S. the same change is handled through an amendment to the IND.⁷

That history only helps if someone at the design stage is responsible for acting on it, including agreeing what will happen to participants if the risk materializes.

Where The Decision should sit

A workable test is this: a sourcing decision becomes a protocol decision once it changes what a participant receives, when they receive it, or what the comparison can tell us. Beyond that point it is part of the sponsor's responsibility for participant safety and the reliability of the data.

ICH E6(R3) asks sponsors to identify the factors critical to trial quality during design and to manage risks to them proportionately.⁸ The comparator clearly qualifies. In practice, sourcing strategy is often settled after the protocol is finalized, by a different group and under commercial pressure, and many protocols say nothing about which sources, versions, or substitutions are allowed.

Clinical supply leaders are usually the first to spot these risks. They generally don't have the authority to decide what those risks mean for participants, and the people who do are often not involved until something has already gone wrong.

Four Questions To Answer Before The Comparator Is Bought

The study team, with a clinician accountable for participants taking part, should be able to answer these, drawing on what the organization's past studies show.

1. Which market's product are we buying, and does it match what the protocol, the investigators, and the regulators expect?
2. If blinding changes the product, what evidence shows it still performs like the marketed medicine, and who has decided that evidence is sufficient?
3. Will any participant move between sources, lots, or presentations during treatment? If so, has a clinician agreed that this is acceptable, and what will participants be told?
4. If the comparator runs short, is withdrawn, or is superseded, who decides what happens to participants on that arm, using what clinical criteria, and where is the decision recorded?

The fourth question overlaps with one we raised in the August series. I have repeated it deliberately, because for comparators it is the one most likely to be left open. A treatment decision taken under supply pressure, with no named decision maker and no record, is difficult to justify afterward to an ethics committee, a regulator, or the participant.

What Supply Leaders Can Do Now

Most of this can be done within existing roles.

• Pull together what happened with comparators on similar past studies, including re-sourcing, shortages, expiry-driven resupply, and any amendments that followed, and bring it to protocol design discussions.
• Review the comparator sourcing plan with the medical lead or medical monitor before contracts are signed, and go through the decisions in it that affect participants.
• Ask for permitted sources, versions, and substitutions to be written into the protocol or the clinical supply plan.
• Agree in advance who decides what happens to participants if the comparator becomes unavailable, what criteria they will use, and where the decision will be documented.

These steps are mostly about timing. If the clinical questions about a comparator are settled before it is bought, a change during the study can be handled under an agreed plan rather than worked out under pressure and possibly without an amendment at all.

Go back to the Phase 3 study at the start. The second country's stock still arrives on time. This time, though, the possibility was anticipated when the protocol was written, a named clinician has already agreed that participants can move to it, and the decision is on record. The supply metrics look the same as before. The difference is that if anyone later asks who decided, and on what basis, there is an answer.

References:

1. Kotsianti A, Rees H. How To Build Clinical Supply Into Drug Development From The Start. Clinical Supply Leader, August 2026.
2. Kotsianti A, Rees H. When Clinical Supply Becomes A Clinical Decision. Clinical Supply Leader, August 2026.
3. U.S. Food and Drug Administration. Scientific Considerations in Demonstrating Biosimilarity to a Reference Product: Guidance for Industry. 2015.
4. European Medicines Agency, CHMP. Guideline on similar biological medicinal products (CHMP/437/04 Rev 1). 2014.
5. Tufts Center for the Study of Drug Development. New Benchmarks on Protocol Amendment Practices, Trends and their Impact on Clinical Trial Performance. Therapeutic Innovation & Regulatory Science. 2024. doi:10.1007/s43441-024-00622-9.
6. Grabenhofer R. Could Smarter Use of Clinical Supply Data Help Avoid Certain Protocol Amendments? Clinical Supply Leader, September 3, 2026. https://www.clinicalsupplyleader.com/doc/could-smarter-use-of-clinical-supply-data-help-avoid-certain-protocol-amendments-0001
7. Regulation (EU) No 536/2014 of the European Parliament and of the Council on clinical trials on medicinal products for human use; US Code of Federal Regulations, 21 CFR Part 312 (Investigational New Drug Application).
8. International Council for Harmonisation. ICH E6(R3) Guideline for Good Clinical Practice. January 2025.

About The Author:

Angeliki Kotsianti, MD, Ph.D., is a physician-scientist who spent 19 years at Pfizer, where she held final medical judgment on benefit-risk decisions across the product life cycle, including manufacturing, reformulation, recall, and supply resilience. She is a consultant with PharmaFlow Ltd.