Clinical Trial Protocol

Clinical trial protocols determine regulatory approval and market acceptance. A well-designed protocol generates clear evidence that regulators trust and clinicians can readily adopt. A flawed protocol wastes millions, delays approval by years, and potentially undermines your device's commercial viability (regardless of technical merit). The difference between protocols that succeed and those that fail reflects whether founders understood regulatory expectations, designed thorough endpoints, and anticipated operational challenges before enrolment begins.

Endpoint errors

The most common protocol failure occurs if your endpoint selection doesn't match regulatory requirements. Founders often choose endpoints that seem clinically relevant but fail to satisfy regulatory standards for approval. This misalignment stems from inadequate regulatory dialogue or overconfidence in clinical logic.

Primary endpoints have to address the specific claims you intend to make. For instance, if you're claiming your device reduces complications, your primary endpoint must measure complication rates using definitions that regulators recognise. Or if you claim improved efficacy, you need to demonstrate superiority against established comparators using validated outcome measures. Regulatory bodies publish guidance documents that specify acceptable endpoints for device categories, including FDA guidance for cardiovascular devices, MHRA recommendations for surgical instruments, and European consensus statements for diagnostic equipment. Ignoring these frameworks invites rejection, regardless of how compelling your clinical data appears.

Secondary endpoints matter more than founders typically recognise. Whilst primary endpoints drive approval decisions, secondary endpoints often determine reimbursement coverage and clinical adoption. Health economic endpoints demonstrating cost-effectiveness, quality of life measures showing patient benefit, and procedural efficiency metrics proving workflow advantages all influence commercial success beyond regulatory approval.

Trial protocols that founders submit often reflect what they hope regulators will accept rather than what guidance documents actually require. They choose endpoints that seem reasonable without confirming regulatory acceptability. This optimism can be costly when regulators reject primary endpoints after enrollment is complete. Invest time early, engaging regulatory bodies about endpoint appropriateness.

_{Professor Martin Cowie. Professor of Cardiology at Imperial College London}

Statistical rigour with powerful analysis

Underpowered studies represent the second most common protocol failure. Founders underestimate the sample sizes required to demonstrate statistical significance for their claimed benefits. This underestimation stems from optimistic effect size assumptions, inadequate understanding of statistical power, or budget constraints that force compromises undermining study validity. Power analysis must precede protocol finalisation. Specify your primary endpoint, estimate the effect size you expect to observe, set your acceptable Type I and Type II error rates, and calculate the required sample size. Standard medical device trials are typically powered for 80% to 90% probability of detecting clinically meaningful differences (with 95% confidence).

Effective size estimation requires an honest assessment that’s based on existing literature, pilot data, or competitor performance. If existing devices reduce complications by 30% and you claim a 50% reduction, justify this assumption with preliminary evidence. If you lack such evidence, power your study conservatively, assuming smaller effect sizes.

Sample size calculations need to account for anticipated dropout rates, protocol violations, and loss to follow-up. If your calculation suggests 100 patients and you anticipate 15% dropout, enrol 118 patients to ensure adequate evaluable subjects. Document all assumptions explicitly in your protocol. Regulators scrutinise power calculations intensely and reject protocols with unjustified sample sizes.

Defining cohorts with precision

Vague inclusion and exclusion criteria undermine study validity by introducing heterogeneity that obscures treatment effects or allows enrolment of inappropriate patients whose outcomes don't reflect the target population performance. Inclusion criteria define your target population precisely using objective, measurable parameters. Rather than ‘patients with coronary artery disease,’ specify ‘patients with angiographically confirmed stenosis of 70% or greater in at least one major epicardial vessel.’ Rather than ‘moderate disease severity,’ define this severity using validated clinical scales with specific threshold scores.

Exclusion criteria require equal precision. So, identify the conditions or characteristics that might confound results, create safety concerns, or prevent proper device use. Common exclusions include significant comorbidities affecting outcomes independent of device performance, concurrent medications interfering with endpoint assessment, and anatomical variations precluding device placement. Balance restrictive criteria that enhance internal validity against broad criteria that improve generalisability and enrolment speed. Overly restrictive protocols enrol slowly and produce results applicable only to narrow patient populations. Overly broad protocols serve only to introduce confounding variables that obscure treatment effects.

Standardised procedures through protocols

Missing or inadequate standard operating procedures represent a frequently overlooked protocol weakness too. Protocols often specify what data to collect without adequately describing how procedures should be performed, how devices should be handled, or how personnel should be trained. Device handling procedures require detailed documentation: Specify your storage conditions, preparation steps, quality checks before use, implantation or application techniques, and post-procedure device management. If any technique variations affect the outcomes, make sure to standardise these techniques across sites with detailed procedural descriptions, training requirements, and competency verification.

Personnel training and certification standards need to appear explicitly in your protocols. Define required qualifications for investigators, proceduralists, and assessors. Specify training programmes that all personnel must complete before performing protocol procedures. Describe competency assessment methods, ensuring consistent skill levels across sites. Your data collection procedures need similar standardisation. Define how measurements should be obtained, what equipment should be used, how calibration should be verified, and how raw data should be recorded. Specify your timing requirements for assessments, the acceptable windows around scheduled timepoints, and your procedure for handling any missed assessments.

Mitigating bias through design

Bias undermines study credibility regardless of how rigorously other protocol elements are designed. Unblinded studies where investigators and patients know treatment assignments invite conscious or unconscious bias in outcome assessment. Protocols must incorporate bias mitigation strategies appropriate to study design and endpoint characteristics.

Blinding provides the strongest bias protection when feasible. Double-blind designs, where neither the investigators nor the patients know the treatment assignments, eliminate most bias sources. Open-label designs, where all parties know assignments, require especially rigorous endpoint definitions and independent assessment to maintain credibility.

Medical device trials struggle with blinding more than pharmaceutical trials because you cannot hide a surgical procedure or an implanted device. When blinding fails, protocol design must compensate through objective endpoints, independent adjudication, and rigorous source documentation. Subjective endpoints in unblinded device trials invite justified regulatory scepticism.

_{Professor Nicholas Freemantle. Professor of Clinical Epidemiology and Biostatistics. University College London}

Piloting protocols before full enrolment

The most valuable protocol development step involves piloting with small subject numbers before full study launch. Pilot phases with three to five subjects or animals expose operational flaws, unrealistic timeline assumptions, problematic procedures, and data collection challenges that protocol review cannot identify.
Budget your time and resources for protocol pilots. The weeks invested in piloting save months or years by identifying fatal flaws before expensive full enrolment begins. Pilots also generate preliminary outcome data that refine sample size calculations and validate assumed effect sizes.

Clinical trial protocol workshop

Clinical trial protocols determine whether your development programme generates evidence that regulators accept and clinicians trust. We’ll help you align endpoints with regulatory expectations through pre-submission dialogue. Create thorough power studies with honest effect size assumptions and adequate sample sizes. Define cohorts precisely using objective, measurable criteria. Standardise procedures through detailed protocols and training requirements. Mitigate bias through blinding, independent adjudication, or objective endpoints. When you pilot protocols with small subject numbers before full enrolment, these elements will transform protocols from administrative requirements into strategic tools that de-risk development and help accelerate your market access.

Waypoint checklist

These elements are often overlooked in clinical trial protocol:

• Endpoint errors fail to match regulatory expectations
• Weak stats with no power analysis or justification
• Vague cohorts with poor patient/animal selection criteria
• Missing SOPs without device handling/training standards
• Bias risks mean a lack of blinding/independent review