In Vitro Antiproliferative Activity of Stevia Phytoconstituents Against Cancer Cell Lines: A Systematic Review and Meta-Analysis

Background/Objectives: Stevia rebaudiana Bertoni, a perennial plant of the Asteraceae family, is widely used as a natural sweetener owing to its steviol glycosides, but it also contains phenolic compounds and diterpenes reported to exhibit antioxidant, anti-inflammatory and antiproliferative effects. Heterogeneous treatment protocols and the use of different cell viability assays, however, hinder reliable cross-study comparison of their antiproliferative potential. This systematic review and meta-analysis assesses the in vitro antiproliferative activity of stevia constituents and crude extracts across diverse cancer cell lines based on cell viability assay data. Methods: A comprehensive literature search was performed following PRISMA guidelines. Data were synthesized through meta-analysis and meta-regression using IC50 values as the effect size. Key variables included assay type (MTT, SRB), incubation time (8, 24, 48, 72 h), constituent class (phenols, glycosides, diterpenes, crude extracts) and cancer cell type. Results: Of the 250 articles initially identified, seven met eligibility criteria, yielding 39 effect estimate datasets. A meta-regression of IC50 values across MTT and SRB assays revealed no statistically significant differences at 24 h (p = 0.080) and 48 h (p = 0.456), suggesting that values from the two assays are broadly comparable. Antiproliferative potency increased with incubation time, with IC50 decreasing from 3481.20 μg/mL at 8 h to 125.22 μg/mL at 72 h. Among constituent classes, phenols showed the highest potency (IC50 = 97.15 μg/mL at 48 h), exceeding that of glycosides (229.73 μg/mL) and diterpenes (171.55 μg/mL). Neuronal cancer cells were the most sensitive (IC50 = 21.00 μg/mL at 48 h), markedly more than breast (58.90 μg/mL) and gynecological (238.35 μg/mL) cancer cells. Conclusions: Stevia constituents, and particularly phenols, demonstrate promising in vitro antiproliferative activity that is both dose- and time-dependent. Furthermore, the lack of a significant difference between MTT- and SRB-derived results supports their combination in meta-analytical syntheses of antiproliferative data.