BACKGROUNDNon-dystrophic myotonias are skeletal muscle channelopathies caused by ion channel dysfunction. Symptom onset is frequently in the first decade of life, causing disability in a young cohort. Although there is no cure, symptomatic treatments exist. Previous trials provide evidence of the efficacy of mexiletine. More recently, lamotrigine has been shown to be effective. Both treatments have different profiles, including pharmacokinetics and adverse events. This trial aimed to investigate whether lamotrigine is non-inferior to mexiletine to directly inform clinical practice.METHODSWe did a randomised, double-blind, crossover, non-inferiority, phase 3 trial at the National Hospital for Neurology and Neurosurgery (London, UK). Participants (aged ≥18 years) who had genetically confirmed symptomatic non-dystrophic myotonia were randomly assigned (1:1), by means of a block randomisation schedule created by a computer program, to receive either mexiletine for 8 weeks followed by lamotrigine for 8 weeks, or lamotrigine followed by mexiletine, with a 7-day washout period in between. Investigators and participants were masked to treatment allocation. The primary outcome measure was the mean interactive voice response (IVR) diary stiffness score (0-9 scale) over the participant's final 2 weeks of diary reporting in each treatment period. Non-inferiority was assessed using a mixed-effects model with a predefined margin of 0·5 and included all randomly assigned participants who contributed at least 7 days of IVR-diary data in either treatment period. The trial is registered at ClinicalTrials.gov, NCT05017155, and EudraCT, 2020-003375-17.FINDINGSBetween Aug 1, 2021, and Dec 12, 2022, of 60 participants were screened (24 females and 36 males) and randomly assigned between Aug 1, 2021 and Dec 12, 2022, to either the mexiletine-lamotrigine sequence (n=30) or the lamotrigine-mexiletine sequence (n=30). 53 participants contributed data to the primary analysis. The mean IVR stiffness score after treatment with mexiletine was 2·54 (95% CI 1·98 to 3·10) versus 2·77 (2·21 to 3·32) with lamotrigine (mean mexiletine-lamotrigine difference -0·23 [95% CI -0·63 to 0·17]). The most common adverse event with both treatments was indigestion-reflux (eight participants, 208 participant-days receiving mexiletine; seven participants, 130 participant-days receiving lamotrigine). No serious adverse events were reported.INTERPRETATIONWe were unable to conclude that lamotrigine is non-inferior to mexiletine; however, improvements in all outcome measures from baseline were similar between lamotrigine and mexiletine. Lamotrigine is an important treatment consideration in non-dystrophic myotonias alongside mexiletine; we propose a treatment algorithm to guide clinical practice.FUNDINGNeuromuscular Study Group, Jon Moulton Charity Trust, UCLH BRC Fast Track Grant.