The default assumption at most first-time sponsors is that a clinical trial requires a CRO, and that the only question is which one. For a Phase 1 study, that assumption is worth testing before it costs you anything.
What a Phase 1 unit already does
An experienced Phase 1 unit is not simply a site. For a single-center first-in-human study it typically provides the protocol execution, the clinical staff, the pharmacy and dosing infrastructure, the safety monitoring in-house, and often the PK sample handling. Many run their own quality systems and have been inspected.
If your study is single-site, standard design, and run at a unit that does this routinely, a full-service CRO may be coordinating a small number of vendors on your behalf, at full-service overhead.
What still has to be owned
The work does not disappear. It relocates to you.
Sponsor oversight. Under E6(R3) this is explicitly non-delegable, and it exists whether or not you have hired anyone to help with it.
Vendor coordination. Central lab, bioanalytical, IRT, EDC, medical writing — someone has to manage the interfaces.
Regulatory submission and maintenance. IND or CTA, safety reporting, annual reports.
Data management and biostatistics, unless the unit provides them to a standard you have verified.
The TMF, which is yours, and which is where first-time sponsors most often accumulate silent debt.
When a CRO earns its cost at Phase 1
Multi-site studies. Once you have more than two or three sites, coordination and monitoring load rises sharply and a CRO's infrastructure starts paying for itself.
Complex designs. Adaptive dose escalation, multiple cohorts running in parallel, complex PK sampling, or anything requiring real-time decision-making across parties.
Multi-country studies, where regulatory and site management complexity multiplies quickly.
Patient studies rather than healthy volunteer studies, where recruitment becomes a genuine operational risk rather than a scheduling exercise.
And where the sponsor has no operational capacity at all — no one who has run a trial, no quality system, no intention of hiring. The CRO is then buying you a functioning operation, and that is a legitimate reason.
The middle option sponsors overlook
Between "full-service CRO" and "do it yourself" sits a shape that suits a lot of Phase 1 programs: run the study at a strong unit, contract the specialist vendors directly, and put senior operational oversight over the whole thing on a part-time basis.
That gives you the judgment without the full-service margin, and it keeps direct relationships with vendors you will use again in Phase 2. It requires someone experienced enough to know what to watch, which is the actual constraint.
How to decide, in practice
Three questions settle it more reliably than any rule about phase.
How many parties have to be coordinated? One unit and two vendors is a manageable span for an experienced operations lead. One unit and seven vendors is not, and neither is four sites.
What happens if something goes wrong at two in the morning? If your answer depends on a named person with the authority and experience to act, you have enough. If it depends on someone reading an SOP for the first time, you do not.
Who owns the TMF and the oversight record from day one? If nobody can answer immediately, that gap costs more than the CRO fee eventually.
The mistake that costs the most
It is not choosing wrong between these options. It is choosing a full-service CRO for a simple Phase 1 study, discovering at Phase 2 that you have built no internal capability and no vendor relationships, and arriving at the study that actually matters with no operational foundation of your own.
Phase 1 is the cheapest place to learn how you want to run trials. Spending it entirely inside someone else's operating model is a decision worth making deliberately rather than by default.
If you are weighing this for a first-in-human study, the five-minute readiness check will show you where the gaps are.
