Background
Phase III randomized trial (PARADIGM). N=145 (planned ~400; closed early for slow accrual, 16 US sites, Aug 2004–Dec 2008). Untreated non-metastatic locally advanced HNSCC — unresectable, low surgical curability, or organ-preservation candidates (stage III/IV T3–T4 or N2–N3). Question: does TPF induction followed by CRT improve OS vs CRT alone?
Interventions and follow up
Arm A: Induction TPF (docetaxel 60 mg/m² D1 + cisplatin 60 mg/m² D1 + fluorouracil 600 mg/m²/day D1–5 q3wk × 3 cycles) → concurrent CRT (docetaxel or carboplatin + accelerated boost RT 72 Gy)
Arm B: CRT alone (cisplatin 100 mg/m² × 2 cycles + accelerated boost RT 72 Gy)
Primary endpoint: OS
mFollow up: 49 month
Arm B: CRT alone (cisplatin 100 mg/m² × 2 cycles + accelerated boost RT 72 Gy)
Primary endpoint: OS
mFollow up: 49 month
Results
3-year OS: 73% (IC+CRT) vs 78% (CRT alone), HR 1.09 (95% CI 0.59–2.03), P=.77 — not significant
Febrile neutropenia during IC: 16 patients (IC arm) vs 1 patient (CRT arm)
Febrile neutropenia during IC: 16 patients (IC arm) vs 1 patient (CRT arm)
Adverse events
Induction phase: febrile neutropenia 16 events (IC arm) vs 1 (control)
CRT phase: AEs similar between arms
Both arms: manageable but substantial grade 3–4 mucositis and dysphagia typical of cisplatin-based CRT
CRT phase: AEs similar between arms
Both arms: manageable but substantial grade 3–4 mucositis and dysphagia typical of cisplatin-based CRT
Conclusions
Adding TPF induction before CRT did not improve OS vs CRT alone in locally advanced HNSCC (HR 1.09 favoring CRT alone, non-significant). Terminated early for slow accrual, leaving it severely underpowered; a true difference could not be excluded. IC before CRT cannot be considered standard of care.
Key Limitations
Severely underpowered — 145 enrolled vs estimated 400+; a large true benefit could have been missed. Early closure biases toward null. Survival in both arms better than predicted, suggesting favorable case selection. HPV status not assessed — HPV+ oropharyngeal patients likely biased outcomes in both arms. The IC→CRT concept has consistently failed when CRT is the comparator (PARADIGM, DeCIDE).
Clinical Context
PARADIGM and DeCIDE established that induction TPF before CRT does not provide OS benefit over CRT alone. Induction is occasionally used in selected cases (impending airway compromise, very bulky disease, distant-failure risk in M0 N3 disease) but is not a guideline-endorsed standard. ASCO and ESMO consider induction chemotherapy optional, with CRT alone the preferred definitive treatment for most LA-HNSCC.