Background: ROC patients (pts), heavily pretreated with IV chemotherapy (CT) are a heterogenous group, median OS 3-9 mo. Oral MCy is well tolerated, avoids IV CT and has been shown to have clinical benefit. MCy has anti-angiogenic properties, inhibiting growth of tumour, endothelial and host vasculature. Augmentation with bevacizumab showed encouraging results in single arm studies. METRO-BIBF evaluates a novel combination of MCy with either placebo (P) or nintedanib (N)- oral inhibitor VEG/PDG/FGFR tyrosine kinases. Methods: Eligible pts received ≥2 lines of CT; platinum-resistant/intolerant, ECOG PS 0-2, life-expectancy of > 6 weeks. MCy (100mg/d) was given continuously with either N or P, to progression / toxicity. No prior TKI permitted; pts stratified for prior bevacizumab. Priamry aim to increase median OS by 2 mo. PFS, toxicity and QoL also measured. Results: 117 pts were randomised 08/14 – 10/16, median age 64, 87% HGS ROC, 39% pts ≥ 5 lines prior CT. Dose of N/P was reduced from 200 to 150 mg bd after planned review of N/P/MCy in first 46 pts. 96 patients have died. N had no effect on OS (HR 1.08, p = 0.72). Table shows OS/PFS by arm. Combining MCy pts from both arms, median OS was 6.4 mo and 26 patients (23%) took MCy for ≥ 6 mo. Pts without prior bev treatment (69%) stayed longer on MCy than pts with prior bev (31%) (mean diff: 53 days, p < 0.01). Grade 3-4 toxicity as expected in 64.4% pts on N/MCy vs. 54.5% pts on P/MCy; more N/MCy pts had neutropenia and diarrhoea. Two treatment-related deaths: 1 bowel perforation, possibly disease-progression (N), 1 liver failure (P). QoL did not differ between arms. Conclusions: Adding N to MCy did not improve OS/PFS. But in this large study of MCy in heavily pre-treated ROC, almost one quarter remained on therapy for > 6 months, suggesting either more indolent disease and/or MCy has longer-term cytostatic or immunological benefits requiring further investigation. Clinical trial information: NCT01610869