Background: Treatment of brain metastases (BM) remains a major concern among patients (pts) with metastatic renal cell cancer (mRCC). Previous retrospective reports suggest that TKI (especially sunitinib) can be safely used in patients with asymptomatic or previously treated BM. This prospective trial evaluates efficacy and safety in pts with non pretreated BM. Methods: mRCC pts with measurable and non-operable BM, no prior sunitinib, ECOG PS ≤2 and adequate organ function were included. Sunitinib was given at 50mg/day for 4 weeks cycles followed by 2 weeks off treatment until disease progression or unacceptable toxicity. The primary end point is objective response (OR) rate on central nervous system (CNS) after 2 cycles of treatment according to RECIST criteria. Using a Simon's optimal two-stage design with a OR rate of 35% as the minimum needed to warrant further investigation and an uninteresting rate of 15%, 28 evaluable patients were needed (15 in the first stage). Grade ≥ 3 neurologic toxicities were tracked down. Results: 17 pts were enrolled from April 2009 to January 2011. All but one had a clear cell histology. One patient died before treatment start because of a cerebral hemorrhage, 16 pts were evaluable after 2 cycles. Best responses on CNS were stabilisation of the disease in 5 pts and no OR was observed. Accrual was stopped after the first step of the trial. 20 grade ≥ 3 adverse events were observed in 12 pts. One toxic death was observed (peritonitis with gastric perforation). In addition, 3 other pts presented at least 1 grade ≥ 3 toxicity related to sunitinib (pulmonary embolism, neutropenia and hypertension). However, no neurological complication related to sunitinib was registered. Conclusions: Sunitinib treatment in pts with untreated brain metastases of mRCC cannot be recommended as a valid option, due to a limited efficacy on BM; notably no neurologic complication was observed under this treatment. Supported by Pfizer.