Generalised myasthenia gravis is a chronic, unpredictable, and debilitating autoimmune disease. New treatments for this disease are needed because conventional therapies have limitations, such as side-effects (eg, increased infection risk) or inadequate control of symptoms. Rozanolixizumab is a neonatal Fc receptor blocker that might provide a novel therapeutic option for myasthenia gravis. We aimed to assess the safety and efficacy of rozanolixizumab for generalised myasthenia gravis. MycarinG is a randomised, double-blind, placebo-controlled, adaptive phase 3 study done at 81 outpatient centres and hospitals in Asia, Europe, and North America. We enrolled patients (aged ≥18 years) with acetylcholine receptor (AChR) or muscle-specific kinase (MuSK) autoantibody-positive generalised myasthenia gravis (Myasthenia Gravis Foundation of America class II-IVa), a Myasthenia Gravis Activities of Daily Living (MG-ADL) score of at least 3 (non-ocular symptoms), and a quantitative myasthenia gravis score of at least 11. Patients were randomly assigned (1:1:1) to receive subcutaneous infusions once a week for 6 weeks of either rozanolixizumab 7 mg/kg, rozanolixizumab 10 mg/kg, or placebo. Randomisation was stratified by AChR and MuSK autoantibody status. Investigators, patients, and people assessing outcomes were masked to random assignments. The primary efficacy endpoint was change from baseline to day 43 in MG-ADL score, assessed in the intention-to-treat population. Treatment-emergent adverse events (TEAEs) were assessed in all randomly assigned patients who received at least one dose of study drug. This trial is registered with ClinicalTrials.gov (NCT03971422) and EudraCT (2019-000968-18); an open-label extension study has been completed (NCT04124965; EudraCT 2019-000969-21) and another is underway (NCT04650854; EudraCT 2020-003230-20). Between June 3, 2019, and June 30, 2021, 300 patients were assessed for eligibility, of whom 200 were enrolled. 66 (33%) were randomly assigned to rozanolixizumab 7 mg/kg, 67 (34%) to rozanolixizumab 10 mg/kg, and 67 (34%) to placebo. Reductions in MG-ADL score from baseline to day 43 were greater in the rozanolixizumab 7 mg/kg group (least-squares mean change -3·37 [SE 0·49]) and in the rozanolixizumab 10 mg/kg group (-3·40 [0·49]) than with placebo (-0·78 [0·49]; for 7 mg/kg, least-squares mean difference -2·59 [95% CI -4·09 to -1·25], p<0·0001; for 10 mg/kg, -2·62 [-3·99 to -1·16], p<0·0001). TEAEs were experienced by 52 (81%) of 64 patients treated with rozanolixizumab 7 mg/kg, 57 (83%) of 69 treated with rozanolixizumab 10 mg/kg, and 45 (67%) of 67 treated with placebo. The most frequent TEAEs were headache (29 [45%] patients in the rozanolixizumab 7 mg/kg group, 26 [38%] in the rozanolixizumab 10 mg/kg group, and 13 [19%] in the placebo group), diarrhoea (16 [25%], 11 [16%], and nine [13%]), and pyrexia (eight [13%], 14 [20%], and one [1%]). Five (8%) patients in the rozanolixizumab 7 mg/kg group, seven (10%) in the rozanolixizumab 10 mg/kg group, and six (9%) in the placebo group had a serious TEAE. No deaths occurred. Rozanolixizumab showed clinically meaningful improvements in patient-reported and investigator-assessed outcomes in patients with generalised myasthenia gravis, for both 7 mg/kg and 10 mg/kg doses. Both doses were generally well tolerated. These findings support the mechanism of action of neonatal Fc receptor inhibition in generalised myasthenia gravis. Rozanolixizumab represents a potential additional treatment option for patients with generalised myasthenia gravis. UCB Pharma.