ARO 04-01 finds no overall survival advantage for cisplatin over mitomycin-based chemoradiotherapy

In 364 patients with advanced head and neck cancer, cisplatin/5-FU did not significantly improve 10-year survival over mitomycin/5-FU.

KEY POINTS

  • ARO 04-01/AHMO (DRKS00040575) was a multicenter randomized phase III superiority trial conducted in Germany between 2004 and 2008. It enrolled 364 patients with unresectable stage IVA–IVB squamous cell carcinoma of the oropharynx (58.5%) or hypopharynx (41.5%), classified according to UICC TNM sixth edition. Patients were randomized equally to cisplatin/5-fluorouracil or mitomycin C/5-fluorouracil with identical hyperfractionated accelerated radiotherapy. Median follow-up was 87 months, and overall survival was the primary endpoint.
  • Both treatment arms received 72 Gy in 45 fractions over six weeks. Radiotherapy began with 2 Gy once daily for 15 fractions, followed by 1.4 Gy twice daily with at least six hours between fractions. The cisplatin arm received 30 mg/m² weekly for up to six cycles, while the mitomycin arm received 10 mg/m² on days 5 and 36. Both groups received continuous-infusion 5-FU at 600 mg/m²/day on days 1–5. Treatment used sequential shrinking fields with 3D conformal RT or, during later years, IMRT.
  • Cisplatin/5-FU did not significantly improve the primary overall survival endpoint. Ten-year OS was 23.3% with cisplatin/5-FU versus 18.0% with mitomycin/5-FU, corresponding to an absolute difference of 5.3 percentage points. The treatment hazard ratio was 0.89 (95% CI 0.69–1.15; log-rank p=0.418). Although the observed survival rate numerically favored cisplatin, the superiority hypothesis was not demonstrated. Importantly, failure to establish superiority does not prove equivalence or noninferiority.
  • Exploratory competing-risk analyses favored cisplatin for distant disease control. Five-year cumulative distant metastasis incidence was 25.9% with cisplatin versus 36.1% with mitomycin, corresponding to a subdistribution HR of 0.66 (95% CI 0.45–0.96; p=0.031). Cumulative progression incidence was 45.6% versus 56.3% (sHR 0.73, 95% CI 0.55–0.98; p=0.037). However, these competing-risk analyses were performed post hoc, and conventional Kaplan–Meier comparisons were not significant. Sensitivity analyses counting death as an event also showed no significant differences.
  • Locoregional control was similar between treatment strategies. At ten years, local control was 61.5% versus 64.1%, regional control 75.3% versus 72.5%, and locoregional control 54.2% versus 53.5% for cisplatin/5-FU and mitomycin/5-FU, respectively. None of these comparisons reached statistical significance. Long-term differences between oropharyngeal and hypopharyngeal subgroups were also not significant, although the trial predates routine HPV-based classification.
  • Cisplatin was associated with greater renal toxicity and a higher recorded adverse-event burden. Mean serum creatinine during treatment was 0.87 ± 0.28 mg/dL with cisplatin versus 0.76 ± 0.18 mg/dL with mitomycin (p<0.001). Recorded adverse or serious adverse events occurred in 24.2% versus 15.4% (p=0.048), respectively. Acute grade 3 mucositis, dysphagia, and dermatitis were common in both arms, without significant treatment-group differences. Grade ≥2 thrombocytopenia occurred in 6.7% versus 2.8%, respectively, and did not significantly favor either regimen.
  • Treatment completion was high, but the historical protocol limits current applicability. Radiotherapy was completed within the planned treatment window in 95% of cisplatin patients and 96% of mitomycin patients. In the cisplatin arm, 72% received at least five of six weekly doses, while 86% of mitomycin patients received both planned doses. The trial used a maximum cisplatin exposure of 180 mg/m², a historical accelerated RT regimen, and staging without routine HPV or modern molecular characterization. Consequently, its results cannot establish that mitomycin/5-FU is equivalent to contemporary cisplatin-based IMRT regimens.

CLINICAL TAKEAWAY

Mitomycin C/5-FU achieved similar observed long-term overall survival to cisplatin/5-FU when combined with historical hyperfractionated accelerated radiotherapy, while cisplatin produced more renal toxicity and exploratory improvements in distant disease control. These findings provide randomized evidence supporting further consideration of mitomycin-based treatment when cisplatin is unsuitable, but do not establish noninferiority or justify replacing cisplatin in eligible patients receiving contemporary chemoradiotherapy.

SOURCE

International Journal of Radiation Oncology, Biology, Physics

Browse more research Suggest a correction