Abstract

Background Preclinical data and results from non-randomised trials suggest that the multikinase inhibitor sorafenib might be an effective drug for the treatment of acute myeloid leukaemia. We investigated the efficacy and tolerability of sorafenib versus placebo in addition to standard chemotherapy in patients with acute myeloid leukaemia aged 60 years or younger. Methods This randomised, double-blind, placebo-controlled, phase 2 trial was done at 25 sites in Germany. We enrolled patients aged 18–60 years with newly diagnosed, previously untreated acute myeloid leukaemia who had a WHO clinical performance score 0–2, adequate renal and liver function, no cardiac comorbidities, and no recent trauma or operation. Patients were randomly assigned (1:1) to receive two cycles of induction therapy with daunorubicin (60 mg/m2 on days 3–5) plus cytarabine (100 mg/m2 on days 1–7), followed by three cycles of high-dose cytarabine consolidation therapy (3 g/m2 twice daily on days 1, 3, and 5) plus either sorafenib (400 mg twice daily) or placebo on days 10–19 of induction cycles 1 and 2, from day 8 of each consolidation, and as maintenance for 12 months. Allogeneic stem-cell transplantation was scheduled for all intermediate-risk patients with a sibling donor and for all high-risk patients with a matched donor in first remission. Computer-generated randomisation was done in blocks. The primary endpoint was event-free survival, with an event defined as either primary treatment failure or relapse or death, assessed in all randomised patients who received at least one dose of study treatment. We report the final analysis. This trial is registered with ClinicalTrials.gov, number NCT00893373, and the EU Clinical Trials Register (2008-004968-40). Findings Between March 27, 2009, and Nov 28, 2011, 276 patients were enrolled and randomised, of whom nine did not receive study medication. 267 patients were included in the primary analysis (placebo, n=133; sorafenib, n=134). With a median follow-up of 36 months (IQR 35·5–38·1), median event-free survival was 9 months (95% CI 4–15) in the placebo group versus 21 months (9–32) in the sorafenib group, corresponding to a 3-year event-free survival of 22% (95% CI 13–32) in the placebo group versus 40% (29–51) in the sorafenib group (hazard ratio [HR] 0·64, 95% CI; 0·45–0·91; p=0·013). The most common grade 3–4 adverse events in both groups were fever (71 [53%] in the placebo group vs 73 [54%] in the sorafenib group), infections (55 [41%] vs 46 [34%]), pneumonia (21 [16%] vs 20 [14%]), and pain (13 [10%] vs 15 [11%]). Grade 3 or worse adverse events that were significantly more common in the sorafenib group than the placebo group were fever (relative risk [RR] 1·54, 95% CI 1·04–2·28), diarrhoea (RR 7·89, 2·94–25·2), bleeding (RR 3·75, 1·5–10·0), cardiac events (RR 3·46, 1·15–11·8), hand-foot-skin reaction (only in sorafenib group), and rash (RR 4·06, 1·25–15·7). Interpretation In patients with acute myeloid leukaemia aged 60 years or younger, the addition of sorafenib to standard chemotherapy has antileukaemic efficacy but also increased toxicity. Our findings suggest that kinase inhibitors could be a useful addition to curative treatment for acute myeloid leukaemia. Overall survival after long-term follow-up and strategies to reduce toxicity are needed to determine the future role of sorafenib in treatment of this disease. Funding Bayer HealthCare.