BACKGROUNDLymphangioleiomyomatosis is an ultra-rare disease mainly affecting women of childbearing age. The MILES trial showed the efficacy of sirolimus, an mTOR inhibitor, in stabilising lung function in patients with lymphangioleiomyomatosis. Drug toxicity and development of resistance are potential limitations of therapy with sirolimus. Nintedanib is a multikinase inhibitor that inhibits PDGFR, which is active in human and murine lymphangioleiomyomatosis lesions. We aimed to investigate the activity and safety of nintedanib in patients with lymphangioleiomyomatosis.METHODSThis phase 2, open-label, single-arm study was conducted at MultiMedica IRCCS, a national referral university centre for rare pulmonary diseases in Milan, Italy. Eligible participants were aged 18 years and older and had sporadic or tuberous sclerosis complex-associated lymphangioleiomyomatosis with progressive pulmonary function decline in the past year despite treatment with sirolimus or treatment naive. Patients received nintedanib 150 mg orally twice per day, with a possible reduction to 100 mg twice per day in case of side-effects or hepatoxicity, for 12 months, followed by a period of 12 additional months without study treatment. The primary endpoint was the change in FEV1 (FEV1 slope in L) over 12 months. This study is registered with ClinicalTrials.gov, NCT03062943.FINDINGSFrom Oct 14, 2016, to Dec 13, 2019, 35 female patients (mean age 50 years [SD 11]) entered the study, 30 of whom were eligible and received nintedanib. After 12 months, 22 patients completed the treatment, 19 of whom also completed the 12 months of follow-up. FEV1 remained stable after one year of treatment (predicted mean difference 0·001 L [95% CI -0·063 to 0·066]; p=0·97). During the 12 months off treatment, a slight decline in FEV1 was observed (predicted mean difference -0·076 L [95% CI -0·149 to -0·004]; p=0·040). The most frequent adverse events were nausea (15 [50%] patients), diarrhoea (eight [26%]), and abdominal pain (two [7%]). No serious adverse events were observed during the treatment period.INTERPRETATIONOur findings suggest that nintedanib did not improve FEV1, but that the treatment was generally well tolerated. These results might support nintedanib as a second-line therapy in patients not controlled by standard treatment with mTOR inhibitors. Further investigation, such as a non-inferiority trial comparing nintedanib and sirolimus could help to better clarify the role of this drug as a potential alternative treatment.FUNDINGBoehringer-Ingelheim.