**Background:** Extra-virgin olive oil (EVOO) and virgin olive oil (VOO) are high-value products with unique sensory and nutritional properties attributed to their phenolic and triterpenic compound profiles. Adulteration of olive oils with cheaper vegetable oils is a significant economic concern, necessitating reliable authentication methods. This study aimed to characterize the phenolic and triterpenic composition of EVOO, VOO, and other vegetable oils using targeted and untargeted UHPLC-HRMS analysis combined with chemometric tools to identify specific biomarkers for olive oil authentication.
**Methods:** A total of 57 oil samples were analyzed, including authentic EVOO from Italian producers (EVOO*, n=13), commercial EVOO (n=7), commercial VOO (n=20), one VOO from olive cakes, and other vegetable oils (sunflower n=5, grape seed n=2, hemp n=2, and one sample each of pumpkin, linseed, sesame, rape, walnut, palm, rice, almond, coconut, and soybean). Phenolic compounds were extracted using liquid-liquid extraction with methanol/water (80/20) and solid-phase extraction using NH2 SPE cartridges. Targeted and untargeted analyses were performed using UHPLC-ESI/HRMS with a Q Exactive Orbitrap mass spectrometer. Full scan data in negative mode covered m/z 75–1000 at 70,000 FWHM resolution. Quantification was based on external calibration curves (25–1750 μg/L for minor phenolics, 25–1000 μg/L for major compounds), with recoveries between 75% and 98%. PCA and heat map analysis were applied to both targeted quantitative data and untargeted semi-quantitative data.
**Key Results:** A total of 30 bioactive compounds were identified and quantified. Cinnamic acid was found at concentrations ten times higher in olive oils (EVOO: n.d.–5.08 µg/g, VOO: 0.02–4.83 µg/g) compared to sunflower oils (0.01–1.52 µg/g) and other vegetable oils (0.01–0.37 µg/g). p-Coumaric acid levels in EVOO (n.d.–0.54 µg/g) and VOO (0.01–0.71 µg/g) were significantly higher than in sunflower (0.01–0.09 µg/g) and other vegetable oils (0.01–0.18 µg/g). Apigenin (n.d.–6.49 µg/g) and pinocembrin (n.d.–0.38 µg/g) were characteristic of EVOO and VOO. Hydroxytyrosol was absent in all non-olive oils, with highest levels in EVOO* (0.01–24.58 µg/g), followed by commercial EVOO (0.01–10.72 µg/g) and VOO (n.d.–5.38 µg/g). Maslinic acid was significantly higher in olive oils (0.24–18.73 µg/g) than other oils (n.d.–4.53 µg/g). Oleuropein was quantified in olive oils (EVOO*: 0.81–18.81 µg/g, commercial EVOO: 0.30–21.81 µg/g, VOO: 0.81–33.0 µg/g) and surprisingly also in some commercial vegetable oils, suggesting possible supplementation with olive leaf extracts. PCA based on targeted compounds explained 39% of variance (PC1+PC2) and showed clear separation of olive oils from other oils, with cinnamic acid, p-coumaric acid, apigenin, pinocembrin, hydroxytyrosol, and maslinic acid as key discriminators. Untargeted PCA explained 62% of variance and also demonstrated clear discrimination. Heat map analysis confirmed two main clusters: olive oils (cluster C1 for targeted, C2 for untargeted) and seed/nut oils. Untargeted analysis additionally identified secoiridoids (elenolic acid, ligstroside, oleocanthal) and lignans (pinoresinol, 8-hydroxypinoresinol, acetoxypinoresinol) as olive oil biomarkers.
**Clinical Implications:** This study provides a robust analytical methodology for authenticating olive oils, which is critical for ensuring consumers receive genuine EVOO and VOO with their associated health benefits, including antioxidant, anti-inflammatory, and cardioprotective properties attributed to phenolic compounds like hydroxytyrosol and oleocanthal. The identified biomarkers can help detect adulteration and protect the economic value of olive oil production. The methodology could be extended to authenticate olive oils based on variety, geographical origin, or adulteration practices, supporting regulatory compliance with European labeling regulations for nutrition and health claims.