The diagnosis of esophageal adenocarcinoma (EAC) is currently made by endoscopy and confirmed with biopsy and histopathology, which are both invasive and expensive procedures. Despite increasing use of endoscopic surveillance for individuals in at-risk categories, this approach has failed to reduce rising EAC incidence. Alternative strategies such as blood-based biomarkers offer a different diagnostic pathway which might be used for disease detection, or for post-treatment surveillance. This study aimed to develop and clinically validate a novel serum glycoprotein biomarker blood test, named PromarkerEso, designed to discriminate EAC from negative controls. Serum glycoprotein concentrations were measured using a lectin-based magnetic bead array pulldown method, followed by multiple reaction monitoring mass spectrometry in 259 samples across 3 independent cohorts from Australia (2 cohorts) and the United States (1 cohort). An algorithm was developed using a panel of 4 glycoproteins combined with simple clinical factors (age, sex and body mass index) to categorise samples as low-, moderate- or high-risk of EAC. PromarkerEso demonstrated strong discrimination performance for EAC vs cancer-free controls, with an area under the curve (AUC) of 0.91 in the development cohort and AUC of 0.82 and 0.98 in the validation cohorts. The algorithm exhibited high sensitivity for EAC in the development cohort (98%), which was confirmed in validation (99.9% and 91%), with high specificity (88% in development cohort, and 86% and 99% in the validation cohorts). PromarkerEso effectively identified individuals without EAC (95% negative predictive value), and those with EAC (96% positive predictive value). The results highlight the potential of glycoprotein biomarkers as a tool for detecting esophageal adenocarcinoma and a step towards improved diagnosis of this challenging disease.