Comprehensive approach was applied to optimize the SC-CO2 extraction conditions of Salvia officinalis L. (sage) leaves. Different extraction parameters (pressure, temperature, and CO2 flow rate) were investigated on: the extraction yield; the extracts chemical profile (determined by GC-MS); optimal extraction conditions by response surface methodology (RSM) for targeted most abundant components (oxygenated monoterpenes, alpha-humulene, viridiflorol and manool). The pressure exhibited the most significant influence on the yield (0.242-7.361%) and significantly influenced the amount of 1,8-cineole, alpha-/beta-thujone and camphor (within the range 15 or 20 MPa). On the contrast, the highest amount of alpha-humulene, viridiflorol and manool was obtained at the lowest pressure (10 MPa). The results showed statistically significant (p < 0.05) values for the regression models for all responses and non-significant lack of fit (p = 0.068-0.339), highlighting that the second order polynomial model can be effectively used for prediction of targeted compounds of sage SC-CO2 extracts.