Cyclophosphamide Noninferior to Ifosfamide as Part of Consolidation Treatment in Standard-Risk Ewing Sarcoma


Key Points

  • Cyclophosphamide was noninferior in event-free survival vs ifosfamide when combined with vincristine/dactinomycin.
  • Some heterogeneity of treatment effect by sex was observed.
  • Cyclophosphamide was associated with more hematologic toxicity and less renal tubular function impairment.

In a European phase III noninferiority trial (Euro-EWING99-R1) reported in the Journal of Clinical Oncology, Le Deley et al found that cyclophosphamide was noninferior in event-free survival vs ifosfamide in combination with vincristine/dactinomycin as consolidation therapy in patients with standard-risk Ewing sarcoma. The investigators, however, stated that some uncertainty regarding noninferiority remained, due to the wide noninferiority margin in the trial, outcome in the per-protocol population, and potential sex differences in treatment effect.

Study Details

In the trial, conducted between February 2000 and March 2010, 856 patients with localized tumors with either a good histologic response to chemotherapy (> 10% cells) or small tumors (< 200 mL) resected at diagnosis or treated with radiotherapy alone received six courses of vincristine, ifosfamide, doxorubicin, and etoposide and one course of vincristine/dactinomycin plus ifosfamide. Patients were then randomly assigned to seven vincristine/dactinomycin plus cyclophosphamide courses with 1.5 g/m2 of cyclophosphamide (n = 431) or seven vincristine/dactinomycin plus ifosfamide courses with 6 g/m2 of ifosfamide (n = 425). The noninferiority limit was set at −8.5% for 3-year event-free survival, corresponding to a hazard ratio (HR) of 1.43, in the intent-to-treat population.

The cyclophosphamide and ifosfamide groups were generally balanced for age (87% and 88% < 25 years), sex (60% and 59% male), estimated tumor volume (< 200 mL in 73% and 70%), tumor type (osseous lesion with or without soft-tissue component in 84% and 86%), tumor site (eg, lower extremity in 35% in both, chest in 24% and 19%, spine in 10% and 10%), metastatic disease (no in 100% in both), local treatment (surgery after vincristine, ifosfamide, doxorubicin, and etoposide with or without radiotherapy in 75% in both; initial surgery of primary with or without radiotherapy in 15% and 14%; local radiotherapy before surgery in 2% in both; definitive radiotherapy in 8% in both), and histologic response (eg, no tumor in 45% and 43%, < 1% tumor cells in 5% and 4%, 1%–4% tumor cells in 8% and 10%, 5%–9% tumor cells in 14% and 16%, good response–not otherwise specified in 2% in each).

3-Year Event-Free Survival

Median follow-up was 5.9 years and > 3 years in 86% of patients. Three-year event-free survival was 75.4% with cyclophosphamide vs 78.2% with ifosfamide; the difference of −2.8% and hazard ratio of 1.12 (91.4% confidence interval [CI] = −7.8% to 2.2%) met the noninferiority criterion. Hazard ratios were 1.0 (95% CI = 0.69–1.46) for local recurrence, 1.30 (95% CI = 0.88–1.92) for secondary metastases without local recurrence, and 0.69 (95% CI = 0.23–2.07) for secondary malignancy.

There was no heterogeneity of treatment effect according to age group (P = .41), type of local treatment (P = .89), or cooperative group (P = .82); a marginal interaction by sex was observed (P = .083), with ifosfamide favored in men (HR = 1.34, 95% CI = 0.96–1.86) and cyclophosphamide favored in women (HR = 0.83, 95% CI = 0.54–1.28). On per-protocol analysis, the hazard ratio was 1.22, with an upper CI limit beyond the allowable margin (91.4% CI = 0.96–1.54).

Three-year overall survival was 85.9% vs 85.5% (HR = 1.09, 91.4% CI = 0.84–1.42; HR = 1.20, 91.4% CI = 0.91–1.57) in per-protocol analysis).


The frequency of hematologic toxicity ≥ grade 4 was higher in the vincristine/dactinomycin plus cyclophosphamide group (84% vs 79%, P = .03), particularly thrombocytopenia (45% vs 35%, P = .004). Infection ≥ grade 2 occurred in 43% and 42%. Among 208 and 229 evaluable patients, ≥ grade 2 impairment of renal tubular function occurred in 16% vs 31% (P < .001). Major treatment modifications, usually due to toxicity, were significantly less frequent with cyclophosphamide (< 1% vs 7%).

The investigators concluded, “Cyclophosphamide may be able to replace ifosfamide in consolidation treatment of standard-risk Ewing sarcoma. However, some uncertainty surrounding the noninferiority of [vincristine/dactinomycin plus cyclophosphamide] compared with [vincristine/dactinomycin plus ifosfamide] remains at this stage. [An] ongoing comparative evaluation of long-term renal and gonadal toxicity [will be] crucial to decisions regarding future patients.”

Marie-Cécile Le Deley, MD, PhD, of Gustave Roussy Institute, is the corresponding author for the Journal of Clinical Oncology article.

The study was supported by the Association Enfants et Santé, Société Française de Lutte Contre les Cancers et les Leucémies de l’Enfant et de l’Adolescent, Unicancer, Cancer Research United Kingdom, Deutsche Krebshilfe, and Bundesministerium für Bildung und Forschung. The study authors reported no potential conflicts of interest.

The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.