PurposeEORTC-1506-STBSG was a prospective, multicentric, randomised, open-label phase 2 trial to assess the efficacy and safety of second-line nintedanib versus ifosfamide in patients with advanced, inoperable metastatic soft tissue sarcoma (STS). The primary end-point was progression-free survival. Patients/methodsPatients with a variety of STS subtypes were randomised 1:1 to nintedanib (200 mg b.i.d. p.o. until disease progression) or ifosfamide (3 g/m2 i.v. days 1–3, every 21 days for ≤6 cycles). A Korn design was applied aiming to detect an improvement in median progression-free survival (mPFS) from 3 to 4.5 months (HR = 0.667). An interim look was incorporated to stop the trial for futility if <19 of the first 36 patients treated with nintedanib were progression-free at week 12. ResultsAt the interim analysis, among the first 36 eligible and evaluable patients randomised for nintedanib, only 13 (36%) were progression-free at week 12. The trial was closed for further accrual as per protocol. In total, 80 patients were randomised (40 per treatment group). The mPFS was 2.5 months (95% CI: 1.5–3.4) for nintedanib and 4.4 months (95% CI: 2.9–6.7) on ifosfamide (adjusted HR = 1.56 [80% CI: 1.14–2.13], p = 0.070). The median overall survival was 13.7 months (95% CI: 9.4–23.4) on nintedanib and 24.1 months (95% CI: 10.9–NE) on ifosfamide (adjusted HR = 1.65 [95%CI:0.89–3.06], p = 0.111). The clinical benefit rate for nintedanib and ifosfamide was 50% versus 62.5% (p = 0.368), respectively. Common treatment-related adverse events (all grades) were diarrhoea (35.9% of patients), fatigue (25.6%) and nausea (20.5%) for nintedanib; and fatigue (52.6%), nausea (44.7%) and vomiting, anorexia and alopecia (28.9% each) for ifosfamide. ConclusionThe trial was stopped for futility. The activity of nintedanib did not warrant further exploration in non-selected, advanced STSs.