In patients with locally advanced esophageal cancer, preoperative chemotherapy or chemoradiotherapy has been shown to improve outcome with respect to survival. Patients who respond to induction therapy have a significantly improved survival, compared with patients who do not respond to the therapy. However, surrogate markers that predict response or prognosis-especially early in the course of therapy-are not available in clinical routine. In patients with esophageal cancer, PET with the glucose analog (18)F-FDG can be used for assessing response to therapy. Therapy response can be assessed with (18)F-FDG PET and (18)F-FDG PET/CT late, that is, after completion of therapy, and early in the course of therapy. In adenocarcinomas of the esophagogastric junction, (18)F-FDG has been established and validated in several studies as a surrogate marker that allows prediction of response and prognosis, whereas in other studies (18)F-FDG PET was not predictive of response and prognosis. The MUNICON study was an initial unicenter trial showing that a PET-guided treatment algorithm was feasible in patients with adenocarcinomas of the esophagogastric junction. The results of this study are important toward individualization of multimodal treatment. The use of (18)F-FDG PET and PET/CT for therapy monitoring in esophageal cancer is the subject of intense discussion, underlining the need for randomized multicenter studies. From a methodologic point of view, the most important issue in therapy monitoring using (18)F-FDG PET and PET/CT is the standardization of patient preparation, data acquisition and processing, and data interpretation, especially for prospective randomized multicenter studies. In conclusion, single-center studies investigating response assessment in patients with esophageal cancer have provided promising results. In the future, prospective randomized multicenter trials will have to be performed and research will address new imaging probes and innovative therapy regimens.