Background
Long-term follow-up (final analysis) of JCOG1008, a Japanese multicenter, open-label, phase II/III noninferiority randomized trial in patients with postoperative high-risk locally advanced squamous cell carcinoma of the head and neck (LA-SCCHN). N=261. Tests whether weekly cisplatin chemoradiotherapy is noninferior in overall survival to the standard 3-weekly cisplatin regimen after surgical resection.
Results
Interventions and follow up: Arm A: Weekly cisplatin 40 mg/m² IV once weekly ×7 cycles with concurrent RT (n=129)
Arm B: 3-weekly cisplatin 100 mg/m² IV once every 3 weeks ×3 cycles with concurrent RT (n=132)
Primary endpoint: Overall survival (noninferiority margin HR<1.32)
mFollow up: 5.6 years (final analysis; interim analysis at 2.2 years)
Results: 5-yr OS: 71.2% (weekly) vs 58.7% (3-weekly), HR 0.76, 95% CI 0.52–1.12, confirming noninferiority (margin HR<1.32)
5-yr RFS: 64.3% vs 53.0%, numerically favoring weekly
5-yr local RFS: 68.8% vs 57.2%, numerically favoring weekly
Arm B: 3-weekly cisplatin 100 mg/m² IV once every 3 weeks ×3 cycles with concurrent RT (n=132)
Primary endpoint: Overall survival (noninferiority margin HR<1.32)
mFollow up: 5.6 years (final analysis; interim analysis at 2.2 years)
Results: 5-yr OS: 71.2% (weekly) vs 58.7% (3-weekly), HR 0.76, 95% CI 0.52–1.12, confirming noninferiority (margin HR<1.32)
5-yr RFS: 64.3% vs 53.0%, numerically favoring weekly
5-yr local RFS: 68.8% vs 57.2%, numerically favoring weekly
Adverse events
Grade ≥3 neutropenia (acute, index analysis): 35% (weekly) vs 49% (3-weekly)
Grade ≥3 infection: 7% vs 12%
Renal / hearing impairment: less frequent with weekly cisplatin
Treatment-related death: 2 [1.6%] (weekly) vs 0 (3-weekly)
Late adverse events (5-yr follow-up): no clinically meaningful difference (≤10%) between arms; no new safety signal
Grade ≥3 infection: 7% vs 12%
Renal / hearing impairment: less frequent with weekly cisplatin
Treatment-related death: 2 [1.6%] (weekly) vs 0 (3-weekly)
Late adverse events (5-yr follow-up): no clinically meaningful difference (≤10%) between arms; no new safety signal
Conclusions
Mature 5-year follow-up of JCOG1008 confirms noninferior long-term overall survival with weekly cisplatin chemoradiotherapy versus standard 3-weekly cisplatin in postoperative high-risk LA-SCCHN, with numerically improved relapse-free survival and a more favorable acute toxicity profile. These results reinforce weekly cisplatin as an evidence-based, better-tolerated alternative in the adjuvant setting.
Key Limitations
Noninferiority design with a wide margin (HR<1.32); the 5-yr OS point estimate favors weekly cisplatin (HR 0.76) but the 95% CI (0.52–1.12) crosses 1, so a true survival benefit versus true equivalence cannot be distinguished. All-Japanese, single-country cooperative-group population (JCOG); dosing intensity, comorbidity profile, and supportive care patterns may not generalize directly to Western cohorts, where higher cisplatin dose-density is more common. Cumulative weekly-cisplatin dose (40 mg/m²×7=280 mg/m²) is lower than the cumulative 3-weekly dose (100 mg/m²×3=300 mg/m²), so the relative contribution of lower cumulative dose versus schedule to the toxicity difference is not isolated. This final analysis is a confirmatory long-term update of the noninferiority result already reported at interim analysis in 2022, which already informed practice; it does not represent a new practice change.
Clinical Context
NCCN and ASCO guidance already lists weekly cisplatin (40 mg/m²) as an acceptable alternative to 3-weekly cisplatin (100 mg/m²) with concurrent radiotherapy, a position informed substantially by the 2022 JCOG1008 interim analysis alongside other randomized trials of dosing schedule in both definitive and postoperative settings. This mature 5-year update strengthens confidence in the durability of that noninferiority specifically in the postoperative/adjuvant setting, where 3-weekly cisplatin (per EORTC 22931/RTOG 9501) has historically been standard. Weekly cisplatin remains a reasonable, better-tolerated option, particularly for patients less likely to tolerate high-dose 3-weekly cisplatin, without displacing 3-weekly cisplatin as the preferred regimen where tolerated.