Esophageal cancer is one of the most lethal of all neoplasms. During the last two decades, there have been significant changes in the epidemiology and treatment of esophageal cancer. The incidence of adenocarcinoma is increasing whereas that of squamous cancer is decreasing. Surgery, the mainstay of treatment of esophageal cancer, has been used with neoadjuvant chemoradiotherapy to improve prognosis in patients with localized disease. Accurate staging is essential for selection of the best mode of therapy and to predict prognosis. In addition, with widespread use of neoadjuvant therapy, accurate assessment of response to therapy has become very important because responders have better a prognosis than nonresponders. Anatomical imaging methods, such as computed tomography and endoscopic ultrasonography, that are commonly used to evaluate esophageal cancer have shortcomings in demonstrating the true extent of disease and in assessing or predicting response to therapy. Positron emission tomography (PET) with 2-[(18)F]fluoro-2-deoxy-d-glucose (FDG) has been shown to be a useful adjunct to anatomical imaging methods. For initial staging of esophageal cancer, the combination of PET and endoscopic ultrasonography with fine-needle aspiration biopsy has been suggested to be the most effective strategy. For restaging and monitoring response to therapy, FDG-PET has been shown to be superior to conventional imaging. The incidence of gastric cancer is decreasing worldwide, but it is also a highly lethal cancer. Similar to esophageal cancer, noninvasive staging of this cancer is unsatisfactory. Approximately one-third of the patients thought to have limited disease and to be candidates for surgery by conventional staging methods, are found to have advanced disease at surgery. Only a few published studies have evaluated gastric cancer with FDG-PET. These studies suggest that FDG-PET may be useful in evaluating gastric cancers of intestinal type and nonmucinous tumors.