Results of a randomized phase II/III study comparing perioperative adriamycin plus ifosfamide and gemcitabine plus docetaxel for high-grade soft tissue sarcomas: Japan Clinical Oncology Group study JCOG1306

11504 Background: Our previous phase II study for high-grade soft tissue sarcomas (STS), JCOG0304, suggested long-term favorable effects of perioperative adriamycin plus ifosfamide (AI) on survival of STS patients. We have also reported in 2015 ASCO Annual Meeting that a phase II/III trial to confirm the non-inferiority of perioperative gemcitabine plus docetaxel (GD) to AI for high-grade STS (JCOG1306) had been started. We herein report the results of JCOG1306 at the early termination by the preplanned second interim analysis. Methods: Patients with operable, FNCLCC grade 2/3 STS primary tumor (T2bN0M0 or anyTN1M0, AJCC 7th edition) or first local recurrent tumor in the extremities or trunk were randomized to AI or GD. Chemotherapy consisted of adriamycin 60 mg/m2 plus ifosfamide 10 g/m2 for AI or gemcitabine 1,800 mg/m2 plus docetaxel 70 mg/m2 for GD. The treatments were repeated for 3 courses preoperatively and 2 courses postoperatively in a 3-week interval. The primary endpoint in phase III part was overall survival (OS). Planned sample size was 140 with a one-sided alpha of 0.1, power of 0.7 and a non-inferiority margin of 8% at 3-year OS, assuming 3-year OS of AI to be 85% and that of GD as 87%. The patient accrual has started in February 2014 and finished in September 2018. Results: A total of 143 patients were enrolled and included in the efficacy analysis. Seventy and 73 patients were assigned to AI and GD, respectively. At the second interim-analysis on December 2019, the estimated 2-year OS was 94.3 % (95% confidence interval (CI) 83.4–98.1) in AI and 91.6 % (80.9–96.4) in GD (hazard ratio (HR) 2.55, 95% CI 0.67–9.78). The estimated 2-year progression-free survival was 81.9 % (95% CI: 69.5–89.7) in AI and 64.0 % (51.1–74.4) in GD (HR 2.32, 95% CI 1.22–4.39). There were no treatment-related deaths in both groups. The most common Grade 3 or higher adverse events in AI were neutropenia (88.4%), anemia (49.3%), and febrile neutropenia (36.2%), whereas those in GD were neutropenia (79.5%), febrile neutropenia (17.8%), and alanine aminotransferase (9.6%). Based on the result of this analysis, the Data Monitoring and Safety Committee of JCOG recommended terminating the study since the point estimate of HR was above the pre-specified allowable HR of 1.61. Conclusions: Although the toxicities were modest in GD, non-inferiority of GD to AI could not be confirmed. In the perioperative chemotherapy for high-grade STS in the extremities and trunk, AI remains the standard regimen. Clinical trial information: UMIN000013175 .