Background: Gastric cancer (GC) is the second cause of cancer related death in the world. It may develop by progression from its precancerous condition, called gastric atrophy (GA) due to gastritis. The aim of this study was to evaluate the accuracy of serum levels of pepsinogens (Pg) and gastrin-17 (G17) as non-invasive methods to discriminate GA or GC (GA/GC) patients. Materials and Methods: Subjects referred to gastrointestinal clinics of Golestan province of Iran during 2010 and 2011 were invited to participate. Serum levels of PgI, PgII and G17 were measured using a GastroPanel kit. Based on the pathological examination of endoscopic biopsy samples, subjects were classified into four groups: normal, non-atrophic gastritis, GA, and GC. Receiver operating curve (ROC) analysis was used to determine cut-off values. Indices of validity were calculated for serum markers. Results: Study groups were normal individuals (n=74), non-atrophic gastritis (n=90), GA (n=31) and GC patients (n=30). The best cut-off points for PgI, PgI/II ratio, G17 and HP were $80{mu}g/L$, 10, 6 pmol/L, and 20 EIU, respectively. PgI could differentiate GA/GC with high accuracy (AUC=0.83; 95%CI: 0.76-0.89). The accuracy of a combination of PgI and PgI/II ratio for detecting GA/GC was also relatively high (AUC=0.78; 95%CI: 0.70-0.86). Conclusions: Our findings suggested PgI alone as well as a combination of PgI and PgI/II ratio are valid markers to differentiate GA/GC. Therefore, Pgs may be considered in conducting GC screening programs in high-risk areas.