IMPORTANCE Most patients with tuberous sclerosis complex (TSC), an autosomal-dominant disorder that is caused by the constitutive activation of mammalian target of rapamycin, experience disfigurement caused by skin lesions involving facial angiofibromas. Many have been left untreated because of a lack of therapeutic options that are less invasive than surgery or laser treatment. OBJECTIVE To confirm the efficacy and safety of sirolimus gel, 0.2%, for treatment of patients with angiofibromas and/or skin lesions. DESIGN, SETTING, AND PATIENTS Multicenter, randomized clinical trial at 9 centers in Japan from December 2015 to October 2016 including 62 children and adults with TSC. INTERVENTIONS Patients who developed angiofibromas were randomly assigned, in a 1:1 ratio, to receive sirolimus gel, 0.2%, or placebo, each applied topically twice daily for 12 weeks. MAIN OUTCOMES AND MEASURES The primary end point was composite improvement in the size and color of angiofibromas in photographs at week 12 of treatment. It was assessed by an independent review committee comprising 3 blinded dermatologists who categorized patient results into the following 6 categories: "markedly improved," "improved," "slightly improved," "unchanged," "slightly aggravated," and "aggravated." RESULTS Sixty-two patients (27 pediatric and 35 adult; 34 [55%] female; mean [SD] age, 22.5 [11.9] years) were enrolled and randomly assigned to receive sirolimus gel, 0.2% (30 patients), or placebo (32 patients). The response rates of angiofibromas at weeks 4, 8, and 12 of treatment were 0 each in the placebo group in contrast to 20% (95% CI, 8%-39%; P = .01), 43% (95% CI, 26%-63%; P < .001), and 60% (95% CI, 41%-77%; P < .001), respectively, in the sirolimus group. None of the 31 assessable patients in the placebo group were rated improved or better, and 26 of them (84%) were rated unchanged. In contrast, 5 (17%) and 13 (43%) patients in the sirolimus group were rated markedly improved and improved, respectively. Adverse events were mild to moderate and were observed in 27 (90%) and 22 (69%) patients in the sirolimus and placebo groups, respectively; however, none of the trial participants discontinued treatment. Acute pancreatitis developed as a serious adverse event in 1 patient in the sirolimus group, and the patient recovered soon after hospitalization without discontinuing treatment. CONCLUSIONS AND RELEVANCE Sirolimus gel, 0.2%, demonstrated a significant clinical benefit for patients with TSC involving angiofibromas, thus providing a promising therapeutic modality.