Clinical Development
Clinical development is where the product meets people, under the most heavily proceduralised evidence-generation regime in regulated industry. For a medicine the arc runs from first-in-human and dose-finding through the pivotal trials that will carry the marketing application; for a device, clinical investigations under the EU MDR or an FDA pathway generate the clinical evidence its conformity assessment or submission requires. The governing discipline is GCP — ICH E6(R3) — resting on three commitments: the rights, safety, and well-being of participants come first; the trial is scientifically sound and described in a clear protocol; and the data are credible, with every change traceable. Authorisation to begin flows through an IND under 21 CFR 312 in the US or a clinical trial application under Regulation (EU) 536/2014 in Europe, with informed consent and independent ethics review — 21 CFR Parts 50 and 56 — as non-negotiable preconditions.
What this page does not claim
Phases are an organising device, not a regulatory mandate — real programmes overlap them, and a medical device, a food product and a small-molecule drug traverse them differently. Each page says where the differences bite.
What happens in this phase
Clinical development is where the product meets people, under the most heavily proceduralised evidence-generation regime in regulated industry. For a medicine the arc runs from first-in-human and dose-finding through the pivotal trials that will carry the marketing application; for a device, clinical investigations under the EU MDR or an FDA pathway generate the clinical evidence its conformity assessment or submission requires. The governing discipline is GCP — ICH E6(R3) — resting on three commitments: the rights, safety, and well-being of participants come first; the trial is scientifically sound and described in a clear protocol; and the data are credible, with every change traceable. Authorisation to begin flows through an IND under 21 CFR 312 in the US or a clinical trial application under Regulation (EU) 536/2014 in Europe, with informed consent and independent ethics review — 21 CFR Parts 50 and 56 — as non-negotiable preconditions.
The operational machinery is substantial. Sites are selected, qualified, and initiated; investigational product is manufactured under IMP GMP (in the EU, Regulation 2017/1569) and supplied blinded through randomisation and trial-supply systems; subject data accumulates in the clinical database under edit checks and query management; monitors verify conduct against protocol and source; safety events are collected, assessed, and expedited where required. E6(R3) pushes sponsors toward quality by design and risk-proportionate oversight: identify the critical-to-quality factors, set quality tolerance limits, and concentrate monitoring where risk to participants or data reliability is real, rather than spreading effort evenly.
Quality's stake spans two layers. Within each trial, it is the sponsor's quality-management obligation: vendor and site oversight, serious-breach recognition and reporting, trial-master-file health, and computerised-system control proportionate to what the system carries. Across the programme, it is coherence: the pivotal evidence must be generated, cleaned, locked, analysed per the statistical plan under ICH E9, and reported in study reports whose every number is traceable to the database. A trial whose conduct cannot be reconstructed from its records — whatever its results — has failed at the only thing GCP exists to guarantee.
- Regulatory authorisation to proceed — an effective IND, an authorised CTA, or the approvals a device clinical investigation requires — plus ethics-committee approval at every site.
- A nonclinical package supporting the proposed exposure, and investigational product manufactured under the applicable IMP GMP with retained batch documentation.
- An approved protocol with defined critical-to-quality factors, a statistical rationale, and a monitoring plan proportionate to risk.
- Validated or fit-for-purpose computerised systems — clinical database, randomisation, outcome capture — with defined data governance and source designations.
- Database lock achieved after documented cleaning, reconciliation, and query closure — with any unlock controlled and justified.
- Clinical study reports (ICH E3 structure) whose results trace to the locked data and the pre-specified analysis.
- The trial master file complete and contemporaneous, evidencing conduct, oversight, and every essential decision.
- Safety data reconciled between the clinical database and the safety system, with all expedited-reporting obligations discharged.
- A pivotal evidence package sufficient, in quality and quantity, to carry the intended submission.
SPEQ synthesis. Phase boundaries and gate criteria are an organising device for planning and review, not a regulatory mandate. Real programmes overlap phases and re-enter them; treat these as the questions worth answering, not a compliance checklist.
- Operate the sponsor's trial-level quality management: risk assessment, quality tolerance limits, and the escalation path when a signal fires.
- Oversee CROs and vendors as delegated activity with retained accountability — auditable, evidenced, and reconciled, not assumed.
- Own serious-breach and misconduct handling: recognition, investigation, reporting to authorities where required, and corrective action.
- Audit sites, systems, and the TMF against the protocol and GCP, independently of the monitoring function.
- Govern computerised-system compliance across the trial estate — validation proportionate to risk, access control, and audit-trail integrity.
- Approved protocol, amendments, and investigator's brochure versions
- Informed-consent documents and the consent records for every participant
- The trial master file — the filed essential records of conduct and oversight
- Locked clinical database with full audit trail, plus the query and reconciliation record
- Clinical study reports and the statistical analysis outputs behind them
- Safety reporting records — expedited reports and periodic safety updates for the trial
- Oversight by dashboard: CRO metrics reviewed, underlying records never sampled, until an inspection reconstructs what oversight should have seen.
- TMF assembled retrospectively before an inspection — the filing-date pattern itself proving that oversight was not operating in real time.
- Monitoring effort spread evenly across sites regardless of risk, missing the anomalous site a central statistical review would have flagged.
- Safety reconciliation deferred to database lock, surfacing SAE discrepancies at the moment they are most expensive to resolve.
- Protocol deviations logged but never trended, so a systemic site problem is discovered by the regulator rather than the sponsor.
Derived from the 8 standards SPEQ maps to this phase, across 4 regulatory bodies: ICH, FDA, EMA, EC.
FREQUENTLY ASKED
What changed with ICH E6(R3)?
E6(R3) rebuilt the guideline around principles rather than prescriptions. It embeds quality by design — identify the factors critical to participant safety and data reliability at the design stage, and set quality tolerance limits around them — and makes risk-proportionate oversight the expected posture rather than a permitted exception. It reframed the fixed "essential documents" list as essential records defined by what they let an inspector verify, and it addresses the modern trial estate: decentralised elements, electronic source, and computerised systems governed by fitness for purpose. The direction of travel is judgement over ritual, with the evidence burden unchanged.
Who is responsible when a CRO runs the trial?
The sponsor, without dilution. GCP is explicit that a sponsor may transfer trial-related duties to a CRO but retains ultimate responsibility for the quality and integrity of the trial data. That makes oversight a designed activity, not a sentiment: documented delegation, access to the CRO's systems and records, sampling of underlying evidence rather than reliance on status reports, reconciliation between sponsor and CRO files, and an escalation path with teeth. Inspectors routinely test the arrangement by asking the sponsor to produce, from its own oversight records, what it knew and when.
Do medical-device trials follow the same rules as drug trials?
The ethical spine is identical — informed consent, independent ethics review, participant safety first — but the framework differs. Device clinical investigations in Europe run under the EU MDR's own clinical-investigation provisions rather than the Clinical Trials Regulation, and the evidence question differs too: a device trial serves a clinical evaluation that also weighs equivalence and post-market data, and investigations are often smaller, unblinded where blinding is physically impossible, and iterated across design changes. The GCP habit of mind — protocol discipline, data traceability, documented oversight — transfers wholesale even where the citations change.