A first-in-human (FIH) study is the first clinical trial in which a new investigational product is given to people. It is usually a small, closely monitored Phase 1 study that increases the dose step by step to learn about safety, tolerability and how the body handles the product.
Free sandbox · No credit card · 21 CFR Part 11 aligned
First-in-human study
noun · clinical research
The first trial in which an investigational product is administered to humans, typically a Phase 1 dose-escalation study focused on safety, tolerability and pharmacokinetics.
Related terms
In short
Explained
A first-in-human study is defined by what is being given, not by who receives it. Before one can start, the sponsor has non-clinical (laboratory and animal) data on pharmacology, toxicology and manufacturing quality, and a regulator or ethics committee has agreed that the risk to the first participants is acceptable. The study then answers a narrow set of questions: is the product tolerated at increasing doses, what adverse events appear, and what does the body do to the product (absorption, distribution, metabolism, excretion)?
FIH studies are almost always Phase 1, but not every Phase 1 study is first-in-human. A later Phase 1 study of a food-effect, a drug interaction or a new formulation uses a product that has already been given to people. The terms single ascending dose (SAD) and multiple ascending dose (MAD) describe common FIH designs: SAD gives one dose per participant at rising levels across cohorts; MAD gives repeated dosing for days or weeks at rising levels. Many protocols combine them, and some add a food-effect or patient cohort. For how these designs fit the phases, see phase 1 data collection.
The starting dose is derived from non-clinical findings, with safety margins. After the 2006 TGN1412 trial, in which healthy volunteers had severe reactions at a dose derived from animal data, EMA issued guidance (first in 2007, revised in 2017) on identifying and mitigating risks in first-in-human and early trials. It emphasises the MABEL approach for products whose mode of action, target or limited animal-model relevance raises concern that serious reactions could occur, and it expects the first dose to be given sequentially with an adequate observation period before further participants are dosed. Check the current text of the guideline that applies to your product and jurisdiction.
Escalation is rule-based. The protocol states cohort size, the observation window before the next dose level, what counts as a dose-limiting toxicity or a stopping criterion, and who reviews the data (often a safety review committee or the medical monitor). Those rules are only as good as the data behind them, so safety labs, ECGs, vital signs and adverse events must be entered and reviewed quickly. Grading adverse events consistently matters; the CTCAE grade lookup is a quick reference.
| Assessment | Screen | Day -1 | Day 1 | Day 2 | Day 8 | Follow-up |
|---|---|---|---|---|---|---|
| Eligibility check | ||||||
| Dose administration | ||||||
| Vital signs and ECG | ||||||
| Safety labs | ||||||
| PK blood samples | ||||||
| Adverse events |
Demo data only. Real schedules come from the protocol.
Running one
Participants are checked against tight inclusion and exclusion criteria before any dosing, and consent is documented first (informed consent).
One or two participants receive the product first. The team observes them for the period the protocol specifies before anyone else in the cohort is dosed.
If nothing meets a stopping rule, the remaining participants are dosed, with active and placebo allocation set by randomization when the design is blinded.
Adverse events, labs, ECGs and drug-level data are reviewed against the escalation rules.
The safety review group decides whether to move to the next dose level, repeat the level, or stop. The decision and the data behind it need to be traceable.
Set up cohorts, safety forms and edit checks in the free sandbox, enter dummy participants and see how the audit trail records every change. No credit card.
Data
Small cohorts and rapid decisions put unusual pressure on data. Safety values need range checks that raise queries as soon as a value looks wrong. Repeating records, such as concomitant medications or serial ECG readings, need their own rows. Calculated values such as QTcF should not be worked out by hand. Cohort membership, dose level and dosing time must be unambiguous. Read-only access for the safety review group keeps them looking at the same data as the site.
Teams planning an FIH study also tend to need blinded and unblinded roles that are really separated (blinded role management), an exportable audit trail, and datasets that can go to a statistician without rework (SDTM export). The protocol itself is the starting point; see how to write a clinical trial protocol.
In Capture
Capture is a self-serve EDC for Phase 1 through Phase 3. Build the visit schedule and safety forms, add edit checks that raise auto-queries when a site enters an out-of-range value, use built-in calculators such as QTcF, and keep a field-level audit trail with the old value, new value and reason for every change. You can test the whole build in the free sandbox before any participant is enrolled.
Edit checks / auto-queries
2Type
Range High
Operator
Greater than
Value
180
Priority: High
Type
Range Low
Operator
Less than
Value
80
Priority: Normal
Query raised automatically
Value 192 violates limit (180). Please verify.
It is the first clinical trial in which a new investigational product is given to people. It is usually a small Phase 1 dose-escalation study that looks at safety, tolerability and pharmacokinetics.
Not exactly. A first-in-human study is nearly always Phase 1, but many Phase 1 studies, such as drug interaction or formulation studies, use products already given to people and are not first-in-human.
It depends on the product. Many early studies use healthy volunteers, while products with a risk profile unsuitable for healthy people, such as many cancer treatments, are first given to patients.
Dosing one or a few participants first and observing them for a defined period before the rest of the cohort receives the product, so that an unexpected serious reaction does not affect everyone at once.
Minimal anticipated biological effect level: the lowest dose expected to produce any biological effect in humans. EMA guidance on early trials of higher-risk products uses it when choosing the starting dose.
Single ascending dose studies give each participant one dose at rising levels across cohorts. Multiple ascending dose studies give repeated doses over days or weeks at rising levels.
Keep exploring
Free sandbox with every feature. Enter sample cohorts and test your safety checks. No credit card.