Oral immunotherapy is effective at inducing desensitisation to allergens and induces sustained unresponsiveness (ie, clinical remission) in a subset of patients, but causes frequent reactions. We aimed to investigate whether addition of a probiotic adjuvant improved the efficacy or safety of peanut oral immunotherapy. PPOIT-003, a multicentre, randomised, phase 2b trial, was conducted in three tertiary hospitals in Australia (Adelaide [SA], Melbourne [VIC], and Perth [WA]) in children aged 1-10 years, weighing more than 7 kg, with peanut allergy confirmed by a double-blind placebo-controlled food challenge (cumulative 4950 mg dose of peanut protein) and positive peanut skin prick test (≥3 mm) or peanut-specific IgE (≥0·35 kU/L). Children were randomly assigned (2:2:1) to receive probiotic and peanut oral immunotherapy (PPOIT), placebo probiotic and peanut oral immunotherapy (OIT), or placebo probiotic and placebo OIT (placebo) for 18 months, and were followed up until 12 months after completion of treatment. Oral immunotherapy consisted of increasing doses of peanut protein (commercially available food-grade 12% defatted peanut flour [50% peanut protein]) until a 2000 mg daily maintenance dose was reached. The probiotic adjuvant was a daily dose of 2 × 1010 colony-forming units of the probiotic Lactobacillus rhamnosus ATCC 53103. Placebo immunotherapy comprised maltodextrin, brown food colouring, and peanut essence, and placebo probiotic was maltodextrin. Dual primary outcomes were 8-week sustained unresponsiveness, defined as no reaction to a cumulative dose of 4950 mg peanut protein at treatment completion and 8 weeks after treatment completion, in the PPOIT versus placebo groups and the PPOIT versus OIT groups, analysed by intention to treat. Safety endpoints were adverse events during the treatment phase, and peanut ingestion and reactions in the 12-month post-treatment period. This study is registered with the Australian New Zealand Clinical Trials Registry, 12616000322437. Between July 4, 2016, and Sept 21, 2020, 201 participants were enrolled and included in the intention-to-treat analysis. 36 (46%) of 79 children in the PPOIT group and 42 (51%) of 83 children in the OIT group achieved sustained unresponsiveness compared with two (5%) of 39 children in the placebo group (risk difference 40·44% [95% CI 27·46 to 53·42] for PPOIT vs placebo, p<0·0001), with no difference between PPOIT and OIT (-5·03% [-20·40 to 10·34], p=0·52). Treatment-related adverse events were reported in 72 (91%) of 79 children in the PPOIT group, 73 (88%) of 83 children in the OIT group, and 28 (72%) of 39 children in the placebo group. Exposure-adjusted incidence of adverse events was 10·58 in the PPOIT group, 11·36 in the OIT, and 2·09 in the placebo group (ratio 0·92 [95% CI 0·85 to 0·99] for PPOIT vs OIT, p=0·042; 4·98 [4·11-6·03] for PPOIT vs placebo, p<0·0001; 5·42 [4·48-6·56] for OIT vs placebo, p<0·0001), with differences seen primarily in gastrointestinal symptoms and in children aged 1-5 years. During the 12-month post-treatment period, 60 (85%) of 71 participants in the PPOIT group, 60 (86%) of 70 participants in the OIT group, and six (18%) of 34 participants in the placebo group were eating peanut; rescue epinephrine use was infrequent (two [3%] of 71 in the PPOIT group, four [6%] of 70 in the OIT group, and none in the placebo group). Both PPOIT and OIT were effective at inducing sustained unresponsiveness. Addition of a probiotic did not improve efficacy of OIT, but might offer a safety benefit compared with OIT alone, particularly in preschool children. National Health and Medical Research Council Australia and Prota Therapeutics.