Across 63 Phase II gene therapy trials, the median interval from initial EU CTIS submission to first authorization was 104 days (SD 57.2). Country-level CTIS Part II review was substantially more variable: the median was 201.5 days (SD 240.7) across 114 country assessments. Orphan trials reached a first authorization faster, yet their later country Part II assessments were much more likely to exceed 90 days.
The trial-level EU end-to-end interval had a median of 104 days and an SD of 57.2 days. Half of trials fell between 33 and 123 days, while the observed range was 1 to 293 days. This metric belongs to the overall CTIS submission and is not assigned to individual countries.
The median country Part II interval was 97.5 days longer than the median first-authorization interval, and its SD was 4.2 times larger. EU launch planning therefore needs separate assumptions for first authorization and later country completion.
Thirteen of 63 trials (20.6%) reached a first CTIS authorization within 30 days, 23 of 63 (36.5%) within 60 days, and 28 of 63 (44.4%) within 90 days. By 120 days, the cumulative share increased to 46 of 63 trials (73.0%).
A 60-day assumption covered only about one third of Phase II gene therapy submissions. A 120-day planning window captured 73.0%, while 180 days captured 95.2%.
Across 114 country assessments, 39 (34.2%) concluded within 60 days and 47 (41.2%) within 90 days. At the other end, 34 assessments (29.8%) exceeded 365 days. The distribution therefore did not resemble the more compact first-authorization timeline.
Country completion risk remained material well beyond three months: 68 of 114 Part II assessments (59.6%) lasted at least 90 days, and only 80 of 114 (70.2%) concluded within one year.
Among countries with at least five assessments, Belgium had a median Part II interval of 26.5 days, the Netherlands 31 days, and Poland 31 days. Spain had the highest median at 286 days, followed by Germany at 253.5 days and Italy at 238 days.
| Country | n | Median days | SD | ≤60 days | ≥90 days |
|---|---|---|---|---|---|
| Bulgaria | 1 | 15 | 0.0 | 100.0% | 0.0% |
| Belgium | 8 | 26.5 | 282.9 | 62.5% | 37.5% |
| Netherlands | 13 | 31 | 251.3 | 69.2% | 30.8% |
| Poland | 6 | 31 | 238.1 | 66.7% | 33.3% |
| Hungary | 3 | 59 | 18.3 | 66.7% | 0.0% |
| Romania | 1 | 63 | 0.0 | 0.0% | 0.0% |
| Norway | 1 | 89 | 0.0 | 0.0% | 0.0% |
| Ireland | 2 | 144 | 119.0 | 50.0% | 50.0% |
| Slovenia | 1 | 154 | 0.0 | 0.0% | 100.0% |
| France | 20 | 225.5 | 222.7 | 25.0% | 70.0% |
| Italy | 19 | 238 | 224.9 | 10.5% | 78.9% |
| Germany | 14 | 253.5 | 213.6 | 28.6% | 71.4% |
| Austria | 2 | 264.5 | 183.5 | 0.0% | 50.0% |
| Sweden | 4 | 271 | 256.4 | 25.0% | 75.0% |
| Spain | 19 | 286 | 250.6 | 26.3% | 73.7% |
Country medians alone understate risk where SD is large. Belgium, the Netherlands and Poland combined low medians with occasional very long assessments, while France, Italy, Germany and Spain each had at least 70.0% of assessments lasting 90 days or more.
Orphan trials reached first EU authorization in a median 44 days versus 115 days for non-orphan trials. However, their country Part II median was 448 days versus 81 days, and 34 of 39 orphan country assessments (87.2%) lasted at least 90 days.
The factors associated with a rapid first authorization were not necessarily associated with rapid multinational completion. The orphan result is the clearest example and supports separate forecasting for the first authorized member state and the remaining Part II pathway.
Systemic intravenous gene therapy assessments had a median Part II duration of 430 days and an 87.8% rate of at least 90 days. Other local or mixed routes had a median of 30 days and a 19.0% rate. Open-label, combination and CRO-supported assessments also showed higher delay rates.
| Factor comparison | Median days | ≥60 days | ≥90 days |
|---|---|---|---|
| Systemic intravenous vs other local or mixed | 430 vs 30 | 90.2% vs 28.6% | 87.8% vs 19.0% |
| Orphan vs non-orphan | 448 vs 81 | 89.7% vs 54.7% | 87.2% vs 45.3% |
| Open label vs blinded | 242 vs 54 | 75.3% vs 45.5% | 69.1% vs 36.4% |
| Combination vs single treatment | 276 vs 98 | 80.0% vs 60.8% | 77.1% vs 51.9% |
| CRO present vs no CRO | 280 vs 84.5 | 73.6% vs 54.8% | 68.1% vs 45.2% |
| Randomised vs non-randomised | 31 vs 256 | 21.7% vs 78.0% | 13.0% vs 71.4% |
The largest delay signals clustered around systemic delivery, orphan development, open-label designs and operationally complex submissions. Randomised studies were an exception, with substantially shorter Part II medians in this cohort.
For months with at least six country assessments, March submissions had a median of 27 days and November submissions 50 days. April submissions had a median of 614.5 days and May submissions 445 days. The 90-day delay rate ranged from 11.1% in November to 91.7% in April.
| Month | n | Median days | ≤60 days | ≥90 days |
|---|---|---|---|---|
| March | 8 | 27 | 87.5% | 12.5% |
| November | 9 | 50 | 77.8% | 11.1% |
| July | 15 | 149 | 40.0% | 53.3% |
| October | 11 | 154 | 36.4% | 63.6% |
| June | 12 | 215 | 16.7% | 66.7% |
| August | 14 | 210.5 | 35.7% | 64.3% |
| May | 11 | 445 | 9.1% | 81.8% |
| April | 12 | 614.5 | 8.3% | 91.7% |
Submission month was more informative for country Part II timing than the number of countries or sites. March and November formed the fastest clusters, while April and May carried the highest long-delay rates.
The number of countries had almost no monotonic relationship with first authorization (Spearman ρ=0.07) or Part II duration (ρ=−0.06). Total site count was similarly weak (ρ=0.07 for first authorization and ρ=−0.12 for Part II). In contrast, recruitment-window length correlated moderately with Part II duration (ρ=0.37), and primary-endpoint count correlated at ρ=0.30.
Geographic scale alone was not the main source of review variation. Submission timing, country mix, route and design characteristics provided stronger descriptive signals than simply adding countries or sites.