- High-risk triggers (guideline-concordant):
- Positive margin (R+) or close margin (institutional cutoffs commonly < 1 to 5 mm)
- Extranodal extension (ENE / ECS +) in any positive node
- Rationale:
- These were the features driving benefit from adding concurrent cisplatin to postoperative RT in the pivotal randomized trials and follow-ups PMC+2PMC+2
- Pivotal trials:
- EORTC 22931 (Bernier, NEJM 2004):
- Design:
- Post-op RT alone (66 Gy) vs RT + cisplatin 100 mg / m² q3wk ×3
- 5-yr outcomes (KM estimates):
- OS 53% vs 40%
- PFS 47% vs 36%:
- Both favoring CRT
- Reported hazard ratios (RT + Cisplatin vs RT):
- OS HR ≈ 0.70–0.75
- PFS HR 0.75:
- Benefit across most high-risk features
- Takeaway:
- Clear KM separation for DFS / OS with combined therapy New England Journal of Medicine+2PubMed+2
- Design:
- RTOG 9501 (Cooper, NEJM 2004; 10-yr update 2012):
- Design:
- Post-op RT alone (60 Gy/6 wk) vs RT + cisplatin 100 mg / m² on days 1, 22, 43
- Overall cohort (10-yr KM):
- LRF 28.8% vs 22.3% (p=0.10)
- DFS 19.1% vs 20.1% (p=0.25)
- OS 27.0% vs 29.1% (p=0.31)
- No significant advantage in the unselected population
- Key subset (pre-specified high-risk: R+ and / or ENE+):
- LRF 33.1% (RT) vs 21.0% (CRT), p=0.02
- DFS 12.3% vs 18.4%, p=0.05
- OS 19.6% vs 27.1%, p=0.07 (trend)
- KM curves:
- Markedly diverge in this subgroup, establishing R+ / ENE+ as the clearest indication for cisplatin-RT PubMed+1
- Design:
- Cross-trial comparative / pooled insight:
- Bernier et al., Head & Neck 2005 compared EORTC 22931 and RTOG 9501:
- Concluded the greatest benefit from CRT accrues to patients with:
- ENE+ and / or positive margins PubMed+1
- Concluded the greatest benefit from CRT accrues to patients with:
- Updated combined analysis (2025):
- Again supports an OS benefit for CRT across the combined cohorts:
- While noting competing non-cancer mortality:
- Still, ENE and / or R+ remain the most reproducible risk features prompting CR PubMed+1
- While noting competing non-cancer mortality:
- Again supports an OS benefit for CRT across the combined cohorts:
- Bernier et al., Head & Neck 2005 compared EORTC 22931 and RTOG 9501:
- EORTC 22931 (Bernier, NEJM 2004):
- How to apply at tumor board:
- Offer adjuvant cisplatin-RT when any of the following are present::
- Positive margin:
- R1:
- Re-resection preferred when feasible, otherwise CRT
- R1:
- Close margin:
- Where institutional policy treats as high risk:
- Commonly < 1 to 5 mm, site-dependent
- Where institutional policy treats as high risk:
- ENE / ECS+ in a lymph node (any extent) PMC
- Intermediate-risk (e.g., PNI, LVI, pT3, pN2 without ENE, multiple nodes but ENE-negative):
- RT alone remains standard:
- CRT generally not indicated absent R+ / ENE+ ACS Publications
- RT alone remains standard:
- Positive margin:
- Cisplatin fitness:
- If contraindicated (CrCl < 60 mL/min, grade ≥ 2 SNHL, neuropathy, poor PS):
- Use RT alone or alternative systemic partner per site-specific guidance:
- But the randomized survival gain post-op is with cisplatin ACS Publications
- Use RT alone or alternative systemic partner per site-specific guidance:
- If contraindicated (CrCl < 60 mL/min, grade ≥ 2 SNHL, neuropathy, poor PS):
- Offer adjuvant cisplatin-RT when any of the following are present::
- Quick “exam-pearl” summaries:
- EORTC 22931:
- KM curves separate for PFS / OS:
- CRT beats RT in high-risk patients overall New England Journal of Medicine
- KM curves separate for PFS / OS:
- RTOG 9501 (10-yr):
- EORTC 22931:
- Bottom line:
- After resection of LA-HNSCC:
- Adjuvant cisplatin-RT is indicated for ENE+ and positive (or institutionally “close”) margin:
- The exact groups where the Kaplan–Meier curves in RTOG 9501 and EORTC 22931 show the clearest advantage for adding chemotherapy to RT
- Adjuvant cisplatin-RT is indicated for ENE+ and positive (or institutionally “close”) margin:
- After resection of LA-HNSCC:

