Postoperative Chemoradiation for High-Risk Head and Neck Cancer

Bernier VS Cooper

Comparing EORTC #22931 and RTOG #9501

 

Background

 

  • Locally advanced HNSCC: poor outcomes with surgery and postoperative RT alone
  •  
  • Hypothesis: Adding concurrent chemotherapy may improve outcomes
  • Two large randomized trials: EORTC #22931 and RTOG #9501
  • Both published in the same issue of NEJM

Europe Vs the USA

Study Designs

EORTC #22931 RTOG #9501
Patients 334 459
Primary endpoint Progression-free survival Locoregional control
Radiotherapy 66 Gy in 33 fractions 60-66 Gy in 30-33 fractions
Chemotherapy Cisplatin 100 mg/m2 on days 1, 22, and 43 Cisplatin 100 mg/m2 on days 1, 22, and 43

Patient Eligibility

EORTC #22931 RTOG #9501
  • Surgical margins ≤5 mm
  • Extracapsular extension (ECE)
  • pT3/T4 (except pT3N0 larynx)
  • pN2/N3
  • Perineural invasion
  • Vascular embolism
  • Level IV/V lymph nodes (oral cavity/oropharynx primaries)
  • Two or more positive lymph nodes
  • Extracapsular extension (ECE)
  • Microscopically involved surgical margins

Results - Locoregional Control

 

  • EORTC #22931:
    5-year estimate: 82% (CRT) vs 69% (RT), p=0.007

 

  • RTOG #9501:
    2-year estimate: 82% (CRT) vs 72% (RT), p=0.01
    5-year estimate: 84% (CRT) vs 74% (RT)
    10-year estimate: 88% (CRT) vs 71% (RT)

Results - PFS vs DFS

 

  • EORTC #22931: 5-year:
    • 47% (CRT)
    • 36% (RT) p=0.04

 

  • RTOG #9501: 3-year:
    • 47% (CRT)
    • 38% (RT), p=0.04

Results - Overall Survival

EORTC #22931

  • 3-year estimate: 65% (CRT) vs 49% (RT)
    5-year estimate: 53% (CRT) vs 40% (RT), p=0.02

RTOG #9501:

  • 2-year estimate: 63% (CRT) vs 57% (RT), p=0.19
    5-year estimate: 29% (CRT) vs 27% (RT), p=0.31
  • Subset with positive margins or ECE (RTOG #9501):
    10-year estimate: 27% (CRT) vs 20% (RT), p=0.07

Results - Overall Survival

EORTC #22931

  • 3-year estimate: 65% (CRT) vs 49% (RT)
    5-year estimate: 53% (CRT) vs 40% (RT), p=0.02

RTOG #9501:

  • 2-year estimate: 63% (CRT) vs 57% (RT), p=0.19
    5-year estimate: 29% (CRT) vs 27% (RT), p=0.31
  • Subset with positive margins or ECE (RTOG #9501):
    10-year estimate: 27% (CRT) vs 20% (RT), p=0.07

Toxicity

Toxicity EORTC #22931 RTOG #9501
Acute Grade 3-4 41% (CRT) vs 21% (RT) 77% (CRT) vs 34% (RT)
Late Grade 3-4 21% (CRT) vs 17% (RT) No significant difference

 

  • Increased acute toxicity with chemoradiation
  • Similar late toxicity profiles

Comparative Analysis - Risk Factors

 

  • ECE and/or microscopically involved surgical margins were the most significant prognostic factors

  • Patients with these risk factors benefited most from chemoradiation in both trials

  • Patients with ≥2 positive lymph nodes without ECE did not seem to benefit significantly from chemoradiation

  • No interaction between number of lymph nodes and benefit with chemo (DM p=0.8, OS p=0.161, DFS p=0.45)

Conclusions

 

  • Postoperative chemoradiation improves outcomes in high-risk HNSCC

  • Greatest benefit in patients with ECE and/or positive margins

  • Consider chemoradiation for patients with stage III-IV disease, perineural invasion, vascular embolism, or level IV-V nodes (oral cavity/oropharynx)

  • No OS benefit if only intermediate risk factors present (p=0.33 in EORTC and p=0.78 in RTOG)

Implications for Practice

 

  • Chemoradiation should be standard for patients with ECE and/or positive margins
  • Careful patient selection based on risk factors is crucial
  • Balance potential benefits with increased toxicity
  • Multidisciplinary approach to treatment decision-making
  • Consider updated analysis (Lu 2022) showing no benefit with chemo for rising nodal count in OS or DM

Limitations and Future Directions

 

  • Retrospective subgroup analysis
  • Differences in eligibility criteria between trials
  • Need for prospective validation of risk stratification
  • Ongoing research on de-escalation strategies for lower-risk patients
  • Investigation of novel systemic therapies (e.g., immunotherapy) in the adjuvant setting
  • Need for updated analysis of intermediate risk factors with modern statistical approaches

Wait but what about lymph nodes..... ??

Impact of Nodal Metastasis Count on Outcomes

  • Secondary analysis of RTOG 9501, RTOG 0234, and EORTC 22931 (947 patients)
  • Increasing number of positive lymph nodes associated with worse outcomes
  • Strongest association up to 5 positive nodes
  • The association of +LN on outcomes was strongest up to 5 +LNs, with each metastatic LN being associated with an independent additional 19% increased risk of death

Impact of Nodal Metastasis Count

Should I give chemo for >5+ lymph nodes?

No

This secondary analysis  it was not designed to test the benefit of chemotherapy based on nodal count alone.

No significant interaction:

The study found no statistically significant interaction between nodal count and the effect of systemic therapy on outcomes

(OS p=0.161, DFS p=0.45, DM p=0.802, LRR p=0.07).