EORTC 22911
After radical prostatectomy in men with pathological high-risk features, does immediate postoperative prostate-bed radiotherapy improve biochemical control compared with observation and treatment at biochemical progression?
Phase III randomized trial N=1,005 Median follow-up 10.6 years
What did the trial actually show?
| Outcome | Postoperative RT | Wait-and-see | Effect |
|---|---|---|---|
| 5-year biochemical progression-free survival | 74.0% | 52.6% | +21.4 percentage points; p<0.0001 |
| Long-term biochemical/clinical progression or death median follow-up 10.6 years | 39.4% had an event | 61.8% had an event | HR 0.49 (95% CI 0.41–0.59); p<0.0001 |
| Locoregional failure | 5% | 15% | HR 0.45 (95% CI 0.32–0.68); p<0.0001 |
| Clinical progression-free survival | Early benefit was not maintained long-term | HR 0.81 (95% CI 0.65–1.01); p=0.054 | |
| Distant failure | No significant improvement demonstrated | Not statistically significant | |
| Overall survival | No significant improvement demonstrated | No significant difference | |
Absolute effect
Locoregional control: cumulative locoregional failure was 5% with postoperative RT versus 15% with wait-and-see, an absolute reduction of 10 percentage points. This corresponds to an approximate NNT ≈ 10 (based on locoregional failure).
74.0% versus 52.6%.
NNT ≈ 5 to prevent one biochemical progression event by 5 years, using the reported absolute difference.
39.4% versus 61.8% had biochemical or clinical progression or died over median 10.6 years.
This corresponds to an approximate NNT ≈ 5, but the event proportions are not a fixed-time survival estimate and should not be presented as a 10-year risk.
10-year cumulative incidence of any late adverse effect: 70.8% versus 59.7% (p=0.001).
NNH ≈ 9 for one additional patient experiencing any late adverse effect, based on the reported 10-year cumulative incidence difference. This endpoint includes any grade and is not equivalent to severe toxicity.
Toxicity
| Endpoint | Result |
|---|---|
| Any late adverse effect, 10-year cumulative incidence | 70.8% RT vs 59.7% wait-and-see; p=0.001 |
| Grade ≥3 toxicity at 5 years | 4.2% RT vs 2.6% wait-and-see; p=0.073 |
| Grade 2–3 late effects | Significantly more frequent with postoperative RT; p=0.0005 |
| Grade 4 toxicity | None reported in the long-term trial report. |
The pivotal trial predates contemporary IMRT, image guidance, dose escalation and modern postoperative target/OAR techniques, so its toxicity profile should not be assumed to represent current RT practice.
Important subgroup signal
Pathology review suggested that positive surgical margin status was the strongest predictor of greater biochemical benefit. In the pathology-reviewed subgroup, the hazard ratio for biochemical disease-free survival was 0.38 (95% CI 0.26–0.54) with positive margins versus 0.88 (95% CI 0.53–1.46) with negative margins; the interaction was significant (p<0.01). This was an exploratory subgroup analysis and should not be interpreted as a randomized comparison restricted to margin-positive patients.
What is the clinical takeaway?
EORTC 22911 established that immediate postoperative prostate-bed RT substantially improves biochemical control and locoregional control after prostatectomy in men with adverse pathological features.
However, the long-term trial did not demonstrate an overall-survival or distant-metastasis benefit, and the early clinical-progression benefit was no longer statistically significant at 10.6 years. RT increased late adverse effects.
The trial therefore answers the historical question of whether adjuvant RT can improve biochemical/local control. It does not by itself answer today's timing question of adjuvant RT versus very-early salvage RT, because the control strategy allowed treatment after biochemical progression and differs from modern PSA-triggered early salvage approaches.
Why this trial matters in the evidence pathway
EORTC 22911 → establishes the efficacy of postoperative RT for pathological high-risk disease → later trials test whether the same disease-control benefit can be achieved by reserving RT for early PSA relapse.
That distinction is central when interpreting EORTC 22911 alongside modern timing trials such as RADICALS-RT and GETUG-AFU 17.
Primary / authoritative sources
Bolla et al. Lancet 2005 — primary 5-year report
Bolla et al. Lancet 2012 — long-term 10.6-year report
Pathology/subgroup analysis — margin status and treatment effect