Clinical Study Planner

Clinical development · planning tool

From hypothesis to powered efficacy study.

AIRCHILL already has measured mechanism and preclinical safety signals. The next clinical programme is designed to test whether those effects translate to patients and to estimate the magnitude in the right phenotype. This page turns that objective into explicit study-design assumptions that can be challenged: control event rate, expected absolute benefit, statistical power, attrition, recruitment speed, follow-up and cost. The output is planning arithmetic, not a protocol or regulatory decision.

Development logic

Three funded gates, and one decision we have not taken.

Gate A · bench, preclinical, human performance

Qualify the thermal dose before any patient.

The staged cold-air ladder below: bench mapping, large-animal confirmation at the −20 °C reference, then sequential human escalation −10 → −15 → −20 °C with an independent safety committee releasing each step.

Gate B · feasibility, randomised 1:1

Can the intended thermal dose be delivered safely — and can we randomise at the scene?

40–80 patients, randomised against standard care and deliberately not powered for clinical benefit. Randomising this early buys three things a later trial would otherwise borrow from the literature: our own control event rate, our own distribution of time to target, and proof that emergency consent, in-vehicle allocation and hospital handover actually work.

Gate C · confirmatory, adaptive

Patient-relevant endpoint, prespecified estimand, sample size re-estimated rather than assumed.

Start at 400, blinded interim, sample-size re-estimation against the control rate observed by then, upper bound fixed in advance. Enrolment is enriched: sites qualify where device start within 20 minutes of EMS arrival is demonstrably achievable, because that is the window our own evidence points at.

Not decided · Gate B2 dose and signal finding

A dedicated randomised stage would give a real effect estimate rather than a plausible one.

150–250 patients, randomised and multicentre, to select timing, target, duration and phenotype. It would push the confirmatory study out by roughly a year and needs additional funding. This decision is open and is shown here as an option, not as part of the plan.

Why the sample size carries a condition. At a 45 % control rate for good neurological outcome, 600 patients resolve an odds ratio of about 1.6 — roughly a quarter more effect than PRINCESS2 assumes for its 1,022-participant trial. That assumption is defensible in the enriched, early-cooled population and not in an unselected one, so the enrichment criterion is stated as part of the sample size rather than separately. The counter-examples stay in view: POLAR reached target at a median of 1.8 h and was negative; TTM2 was negative at 33 °C in 1,900 patients. Sources for this box: the 45 % control rate and the 1,022-participant plan are PRINCESS2’s own design assumptions; the enrichment case is the PRINCESS time-to-cooling subanalysis (Awad et al., Intensive Care Med 2020; shockable subgroup n = 114, OR 3.25, 95 % CI 1.06–9.97 — a propensity-matched subgroup, which is why it is a trial objective and not a finding) inside the PRINCESS trial, which missed its primary endpoint. The arithmetic is ours: two-sided α 0.05, 80 % power, two independent proportions — the same method the calculator below uses. It reproduces the PRINCESS2 plan to within 3 %, which is the only check available without a biometrician.

Regulatory frame. EU MDR Annex XV requires an appropriate clinical investigation plan, scientific validity and justified statistical methodology. ISO 14155:2026 covers good clinical practice for medical-device investigations. FDA recommends early interaction on pivotal device-study design; a life-supporting investigational ventilator study would need an explicit IDE / significant-risk assessment before US enrolment.

Airway-safety programme · cold-air dose escalation

Do not jump directly to the coldest technically achievable gas.

The literature does not support a single minimum safe inspired temperature. AIRCHILL should therefore qualify a multidimensional respiratory thermal dose and use a staged safety programme. The provisional −20 °C first-generation lower setpoint is a planning decision anchored to the existing six-hour porcine endotracheal evidence, not a human injury threshold.

COLD-DOSE	How do temperature, humidity, ventilation and duration interact in human airway response?	Dose justification, risk controls, proposed operating envelope, FDA Pre-Sub / Notified Body discussion.
Large-animal AIRCHILL	Does prolonged endotracheal delivery produce tissue, secretion, gas-exchange or mucociliary injury?	Preclinical safety package and justification to proceed to route-specific human testing.
AIRCHILL-FIRST	Can the qualified device deliver the intended respiratory thermal dose safely in humans?	MDR clinical investigation / CER, FDA IDE/Pre-Sub evidence, labeling and contraindication development.
Emergency feasibility	Does safety/performance remain acceptable in the real target workflow?	Intended-use validation, EMS human factors, transport performance and basis for later efficacy trial.

Pre-Sub / Notified-Body question to ask explicitly. “Is the proposed temperature–humidity–ventilation dose model, together with staged −10/−15/−20 °C human exposure and route-specific preclinical airway/mucociliary evidence, sufficient to support the proposed first-generation operating envelope, or are additional duration/repetition or vulnerable-airway cohorts required before labeling?”

Operational clinical package

The study strategy now has controlled working documents.

Full development Clinical Investigation Plans for cardiac arrest and stroke are now linked to a shared Schedule of Assessments/CRF data dictionary, DSMB Charter, Statistical Analysis Plan, screening/monitoring plan, safety-reporting plan, Investigator’s Brochure framework and EU/FDA submission map.

Open the AIRCHILL Clinical Investigation Package →

Protocol package · airway translation programme

The airway-safety programme now has two separate protocol synopses so controlled human cold-air physiology and direct AIRCHILL product evidence are not conflated. Winter-sport airway data remain background and mechanistic evidence rather than a separate AIRCHILL study programme.

COLD-DOSE — randomized crossover mapping of temperature × absolute humidity × ventilation × duration, including recovery.
AIRCHILL-FIRST — staged product-specific intubated human safety/performance programme with independent escalation review.

Planning sample sizes and exposure tiers are explicitly labelled ASSUMPTION until the primary endpoints, pilot variances, statistical simulations, risk-management package and regulatory/ethics pathway are finalized.

15 September 2026 · protocol-operations intelligence

STAR · PMID 42733931 · NCT04921189. The neutral phase-2 OHCA result does not provide an AIRCHILL treatment-effect estimate and does not trigger a change to current AIRCHILL power assumptions or primary endpoints. For protocol design, keep blinded patient-relevant neurological outcomes above NSE or other injury biomarkers in the outcome hierarchy; biomarkers remain mechanistic/supportive and require prespecified sampling windows, assay handling and interpretation.

NCT07818785 · EMS human factors. The newly posted 1,500-participant AP-versus-AL defibrillator-pad trial is adjacent workflow evidence, not cooling-efficacy evidence. Add/retain operational fields for defibrillator pad position (AP/AL), time to first rhythm analysis, time to first shock, and any treatment interruption attributable to AIRCHILL device set-up. Human-factors simulations should test AIRCHILL use with both pad configurations and verify that cooling preparation does not delay rhythm analysis or defibrillation.

Decision boundary. No change to AIRCHILL efficacy claims, regulatory status, primary endpoints or sample-size assumptions is made solely from these sources. COLD-DOSE + AIRCHILL-FIRST remain the active airway programme; ATHLETE-AIRWAY is not part of the programme.

Primary sources: PMID 42733931 · DOI 10.1016/j.eclinm.2026.104191 · NCT07818785.

IH-TTM comparator and dose benchmark · 17 September 2026

NCT07086703 is now recruiting 788 adults after in-hospital cardiac arrest. The active hypothermia arm targets 33±0.5 °C within 60 minutes after randomization, maintains that target to 28 hours, then rewams at 0.2 °C/h; if coma or sedation persists, normothermia at 36.5–37.7 °C is maintained through 72 hours. The primary endpoint is mRS 0–3 at 6 months. Supportive endpoints include NSE, NFL, mortality, MoCA, SF-36 and ICU complications including severe arrhythmias, nosocomial infection, VTE, ischemic and sedation complications.

AIRCHILL planning decision. Preserve treatment setting (IHCA vs OHCA), time-to-target, duration, rewarming slope and post-intervention fever control as separate variables. Keep 6-month patient-relevant neurological outcome above biomarkers in the endpoint hierarchy. Do not transfer an IH-TTM treatment effect to AIRCHILL power assumptions. The participating IH-TTM centres are a potential temperature-management network map for future investigator outreach only; no partnership or site suitability is implied.

Primary source: ClinicalTrials.gov NCT07086703, updated 17 September 2026.