The purpose of this work was the optimization of the extraction from spent coffee grounds, specifically 100% Arabica coffee blends, using a desirability approach. Spent coffees were recovered after the preparation of the espresso coffee under the typical conditions used in coffee bars with a professional machine. Spent coffee was subjected to different extraction procedures in water: by changing the extraction temperature (60, 80, or 100 °C) and the solvent extraction volume (10, 20, 30 mL for 1 g of coffee) and by maintaining constant the extraction time (30 min). The ranges of the process parameters, as well as the solvent to be used, were established by running preliminary experiments not reported here. The variables of interest for the experimental screening design were the content of caffeine, trigonelline, and nicotinic acid, quantitatively determined from regression lines of standard solutions of known concentrations by a validated HPLC-VWD method. Since solvent extraction volumes and temperatures were revealed to be the most significant process variables, for the optimization of the extraction process, an approach based on Response Surface Methodology (RSM) was considered. In particular, a Box-Wilson Central Composite Design, commonly named central composite design (CCD), was used to find the optimal conditions of the extraction process. Moreover, the desirability approach was then applied to maximize the extraction efficiency by searching the optimal values (or at least the best compromise solution) for all three response variables simultaneously. Successively, the best extract, obtained in a volume of 20 mL of water at an extraction temperature of 80 °C, was analyzed for total phenol content (TPC) through the Folin–Ciocalteu assay, and the antioxidant capacities (AC) through the trolox equivalent (TE) antioxidant capacity (DPPH), ferric-ion reducing antioxidant parameter (FRAP), and radical cation scavenging activity and reducing power (ABTS). The TPC and the AC for spent coffee were high and comparable to the results obtained in previous similar studies. Then, the extract was evaluated by inductively coupled plasma mass spectrometry (ICP–MS), revealing that potassium was the most abundant element, followed by phosphorus, magnesium, calcium, sodium, and sulfur, while very low content in heavy metals was observed. Preliminary in vitro assays in keratinocyte HaCaT cells were carried out to assess the safety, in terms of cytotoxicity of spent coffee, and results showed that cell viability depends on the extract concentration: cell viability is unmodified up to a concentration of 0.3 mg/mL, over which it becomes cytotoxic for the cells. Spent coffee extract at 0.03 and 0.3 mg/mL showed the ability to reduce intracellular reactive oxygen species formation induced by hydrogen peroxide in HaCaT cells, suggesting its antioxidant activity at intracellular levels.