BackgroundThe rate for upper gastrointestinal (UGI) cancers, and especially the rate for gastric cancer, varies from country to country and from region to region within country. In Turkey, the incidence of gastric cancer varies widely among regions due to the diverse cultures and accompanying food habits of the population. This study aimed to determine the endoscopic frequency of esophageal and gastric cancers and the usefulness of alarm symptoms in diagnosing gastric cancer in subjects undergoing endoscopy in northeastern Turkey.MethodsThis retrospective study was based on hospital records, reviewing the records of patients who had underdone esophago-gastro-duodenal (EGD) video endoscopy at two general hospitals in Erzurum. From July 2010 to January 2013, 25,037 patients from Erzurum underwent EGD procedures under either intravenous sedation or local anesthesia. Classifications of UGI cancer, based on location of the tumor, were defined as esophageal, cardia, cardia and fundus or corpus or all of these, corpus, corpus and antrum, and antrum. Metastasis was studied in 659 patients diagnosed with cancer.ResultsThroughout the study, 1,007 biopsy samples were reported as malignant tumors (719 gastric, 276 esophageal, and 12 duodenal cancers). The study considered the gastric cancer population under age 46, which included 99 (13.8 %) patients. Among them, the distribution of UGI cancer by age was as follows: under age 26 years, 9 patients (0.9 %); age 26–35 years, 30 patients (3 %); and age 36–45 years, 60 patients (6 %). Overall, 298 patients (298/1007, 29.6 %) presented with localized disease, and 361 patients (35.9 %) had distant metastases at the time of diagnosis.ConclusionsWe determined that UGI cancer patients are observed more frequently in northeastern Turkey than in western Turkey, Europe, and the USA. We believe that alarm symptoms and endoscopic scanning programs require new, region-specific criteria to diagnose UGI cancers in this region. For the patient groups with these different characteristics, we believe that new scanning, follow-up, and treatment strategies are needed that take into consideration differences in the histopathology of the tumors, their localization, and the patients’ ages.Trial registrationThere is registration number. This study is “retrospective study”. This study is “retrospectively registered”.