Summary Background Worldwide, many patients with HER2-positive early stage breast cancer do not receive trastuzumab—the standard adjuvant treatment. We investigated the efficacy and safety of adjuvant lapatinib for patients with trastuzumab-naive HER2-positive early-stage breast cancer, started at any time after diagnosis. Methods This study was a placebo-controlled, multicentre, randomised phase 3 trial. Women outpatients from 405 centres in 33 countries with HER2-positive early-breast cancer who had previously received adjuvant chemotherapy but not trastuzumab were randomly assigned (1:1) to receive daily lapatinib (1500 mg) or daily placebo for 12 months. Randomisation was done with a computer-generated sequence, stratified by time since diagnosis, lymph node involvement at diagnosis, and tumour hormone-receptor status. Investigators, site staff, and patients were masked to treatment assignment. The primary endpoint was disease-free survival in the intention-to-treat population. This study is registered with ClinicalTrials.gov , number NCT00374322. Findings Between August, 2006, and May, 2008, 3161 women were enrolled and 3147 were assigned to lapatinib (n=1571) or placebo (n=1576). After a median follow-up of 47·4 months (range 0·4–60·0) in the lapatinib group and 48·3 (0·7–61·3) in the placebo group, 210 (13%) disease-free survival events had occurred in the lapatinib group versus 264 (17%) in the placebo group (hazard ratio HR 0·83, 95% CI 0·70–1·00; p=0·053). Central review of HER2 status showed that only 2490 (79%) of the randomised women were HER2-positive. 157 (13%) of 1230 confirmed HER2-positive patients in the lapatinib group and in 208 (17%) of 1260 in the placebo group had a disease-free survival event (HR 0·82, 95% 0·67–1·00; p=0·04). Serious adverse events occurred in 99 (6%) of 1573 patients taking lapatinib and 77 (5%) of 1574 patients taking placebo, with higher incidences of grade 3–4 diarrhoea (97 6% vs nine <1%), rash (72 5% vs three <1%), and hepatobiliary disorders (36 2% vs one <1%). Interpretation Our data show that there was no significant difference in disease-free survival between groups when analysed in the intention-to-treat population. However, exploratory analyses restricted to patients who had HER2-positive disease confirmed by central fluorescence in-situ hybridisation review suggested marginal benefit with lapatinib in terms of disease-free survival. Thus lapatinib might be an option for women with HER2-positive breast cancer who do not or cannot receive adjuvant trastuzumab. Funding GlaxoSmithKline.