EORTC Probabilistic Risk Model
Tumor Characteristics
Awaiting Data
Complete all six tumor characteristics to generate EORTC probability estimates.
Guidelines & Evidence
Verified
Last Review: 2026
When to Use
When to Use the EORTC Risk Tables
Predicting recurrence risk after TURBT for newly diagnosed NMIBC (Ta, T1, or CIS) to guide surveillance frequency (cystoscopy every 3 vs 6 vs 12 months)
Predicting progression risk to muscle-invasive bladder cancer (MIBC, T2+) to guide adjuvant intravesical therapy (BCG vs chemotherapy) vs early radical cystectomy
Stratifying patients in clinical trials of NMIBC (entry criteria based on EORTC risk groups)
Shared decision-making with patients about treatment intensity (e.g., high-risk patient may choose early cystectomy rather than repeated BCG)
Determining need for restaging TURBT at 4–6 weeks (recommended for high-risk EORTC progression score ≥ 8)
Counseling patients about compliance (high-risk recurrence predicts need for multiple cystoscopies and intravesical treatments)
Post-BCG failure management: EORTC risk table progression score helps decide between repeat BCG (low progression risk) vs early cystectomy (high progression risk)
EORTC vs CUETO (Spanish) vs EAU Risk Groups — Comparative Performance
| Risk Tool | Variables Included | Derivation Cohort (N) | C-Index for Recurrence | C-Index for Progression | Advantages | Disadvantages |
|---|---|---|---|---|---|---|
| EORTC (2006) | Tumor number (3 levels), tumor size (<3 vs ≥3 cm), prior recurrence rate (3 levels), T category (Ta vs T1), CIS (yes/no), Grade (G1–3) | 2,596 (7 EORTC trials, all received TURBT + intravesical chemotherapy (not BCG) or TURBT alone) | 0.61–0.64 | 0.71–0.74 | Most widely used, validated in multiple external cohorts (Asia, Europe, US), simple to calculate, 1-year and 5-year predictions, helps select adjuvant BCG vs chemotherapy | Overestimates recurrence risk (especially in modern cohorts with better TURBT quality, photodynamic diagnosis), does not include BCG maintenance in derivation (all chemotherapy), Grade based on WHO 1973 (not 2004/2016), no TVC (variant histology) |
| CUETO (Spanish, 2009) | Tumor number (single vs multiple), tumor size (<3 vs ≥3 cm), prior recurrence (yes/no), T category (Ta vs T1), CIS (yes/no), Grade (G1–3), age (<70 vs ≥70) | 1,062 (CUETO trials, all received TURBT + BCG maintenance) | 0.61–0.63 | 0.68–0.71 | Derived in BCG-treated patients (more relevant to modern practice), includes age (older age predicts worse progression), better calibration for BCG-treated cohorts | Less widely used, no risk tables for no-BCG patients, still uses WHO 1973 Grade, not updated for WHO 2004/2016 |
| EAU Risk Groups (2024 update) | Combines EORTC and CUETO plus additional variables (TVC, lymphovascular invasion, prostatic urethra involvement, completeness of TURBT) | Expert consensus (not derived from single cohort) | Not applicable (guideline, not predictive model) | Not applicable | Simpler group assignment (low, intermediate, high, very high), incorporates modern histology (WHO 2004/2016 Grade Group, TVC, LVI), guides BCG use and cystectomy timing | Subjective (not point-based), less granular than EORTC, not validated as a predictive model (consensus-based) |
Modern Validation — EORTC Overestimates Recurrence but Still Valid for Progression
Multiple external validations (2010–2024, n > 15,000 patients) show EORTC tables overestimate recurrence risk by 10–30% in modern cohorts (due to better TURBT quality, photodynamic diagnosis, immediate post-op intravesical chemotherapy, and improved risk-stratified treatment). However, progression prediction remains accurate (c-index 0.70–0.74). Clinical implication: For recurrence, use EORTC to identify relative risk (low vs high) but not absolute percentages; for progression, EORTC absolute percentages are reliable. Example: EORTC predicts 5-year recurrence 60% for high-risk patient; actual with modern TURBT + immediate chemo + BCG may be 40–50%, but still high-risk relative to low-risk patient with 20% actual. Do NOT deny patient BCG because EORTC recurrence overestimated; BCG reduces recurrence by 50% regardless.
Last Comprehensive Review: 2026