Across 257 European Phase III immunology trials, the median CTIS end-to-end timeline was 105 days (SD 40.6), while country-specific CTIS Part II took a median 99.5 days (SD 255.8). Only 48.3% of Part II reviews concluded within 90 days, and country medians ranged from 21.5 days in Latvia to 173 days in Bulgaria. Submission month was the clearest timing signal; autumn submissions had a 125-day end-to-end median versus 88 days in spring.
End-to-end timing—from initial CTIS submission to first authorization—is a trial-level measure and is not attributable to individual countries. Country comparisons therefore use only CTIS Part II, measured from the earliest country Part II submission to the latest country decision or authorization.
The Part II median resembles the overall end-to-end median, but its SD is more than six times larger, indicating that country-level timing is the main source of long-tail operational risk.
Cumulative approval rates show a sharp end-to-end concentration around 105–120 days, while Part II remains widely dispersed beyond the median.
A 120-day end-to-end planning assumption covered 80.9% of trials. The same buffer covered only 53.0% of country Part II reviews, and 27.0% remained unresolved beyond one year.
Among countries with at least 10 observations, Slovakia had the shortest median at 29.5 days, followed by Austria and Croatia at 35 days. Bulgaria was longest at 173 days, followed by Portugal at 161 and Poland at 144 days.
Country selection changes the expected Part II median by more than fivefold. High SDs across most countries show that a low median does not eliminate long-tail risk, so country-specific buffers remain necessary.
Submission timing was the clearest independent signal. CRO involvement and larger country footprints also marked slower trials, while sample size and total site count showed little direct correlation.
Calendar timing dominates the observed end-to-end signal. Operational complexity matters, but neither target sample size (ρ=−0.02) nor total site count (ρ=0.10) independently explained much of the variation.
Overall, 58.0% of country Part II reviews lasted at least 60 days and 51.7% lasted at least 90 days. Month of Part II submission, country, disease area, and modality showed the largest differences.
The largest actionable Part II differences were country and submission month. Trial attributes such as paediatric status, orphan designation, randomisation, and modality were secondary signals and may partly reflect document and review complexity.
Use one trial-level end-to-end forecast, then layer country-specific Part II assumptions. A single EU-wide Part II average obscures both the fivefold country-median spread and the substantial long tail.