BackgroundMelanoma, a highly malignant skin tumor, can develop systemic metastases during the early stage. Several studies of melanoma animal models indicate that curcumin, a natural plant extract, inhibits melanoma growth through various mechanisms. To evaluate the relationships among different experimental conditions, curcumin itself, its derivatives, and special formulations, it is necessary to conduct a systematic review and meta-analysis. PurposeThis meta-analysis aims to evaluate the potential of Curcumin as a drug for inhibiting the growth of melanoma and to determine the optimal dosage range and treatment duration for Curcumin administration. MethodsA systematic search of studies published from inception to December 2023 was conducted across six databases (PubMed, Web of Science, Embase, China National Knowledge Infrastructure, Wanfang Data, and VIP). Methodological quality was assessed using SYRCLE's RoB tool. Study heterogeneity was assessed using Cochran's Q test and I2 statistics. Publication bias risk was evaluated using a funnel plot. All analyses were performed using R (version 4.3.3). Additionally, three-dimensional effect analysis and machine learning techniques were utilized to determine the optimal dosage range and treatment duration for Curcumin administration. ResultsForty studies involving 989 animals were included. The results demonstrated that, relative to the control group, administration of Curcumin resulted in a significant reduction in tumor volume. [SMD=−3.44; 95 % CI (−4.25, −2.63); P<0.01; I2 = 79 %] and tumor weight [SMD=−1.93; 95 % CI (−2.41, −1.45); P<0.01; I2 = 75 %]. Additionally, Curcumin demonstrated a significant capacity to decrease the number of lung tumor nodules and microangiogenesis, as well as to extend survival time, in animal models. The results from three-dimensional effect analysis and machine learning emphasize that the optimal dosage range for Curcumin is 25–50 mg/kg, with an intervention duration of 10–20 days. ConclusionCurcumin can inhibit the growth of melanoma, and the dose-response relationship is not linear. However, further large-scale animal and clinical studies are required to confirm these conclusions.