Background: Currently there is no universally accepted method to accurately delineate the gross tumor volume (GTV) of primary esophageal cancer in patients undergoing radiotherapy. This prospective study aims to determine the impact of PET/CT on radiotherapy planning and outcomes in patients with localized esophageal cancer. Methods: 54 patients were recruited between June 2003 - May 2008. All underwent PET/CT scanning in the radiotherapy treatment position and received treatment planned using the PET/CT dataset. Of these, 13 (24.1%) had metastatic disease detected on PET and 3 patients had no radical radiotherapy, while another 3 patients had missing planning PET/CT data (excluded from planning component analysis). GTV was defined separately on PET/CT (GTV-PET) and CT (GTV-CT) data sets. A corresponding planning target volume (PTV) was generated for each patient. Volumetric and spatial analysis quantified the proportion of FDG-avid disease not included in CT-based volumes. Clinical data was collected for 38 patients treated radically to determine locoregional control and overall survival rates. Results: Mean age was 67 years (range:32 - 88). Median follow up was 4 years (range:2.7 – 6.8). FDG-avid disease would have been excluded from GTV-CT in 29 patients (79%) with a mean volume of 17% (range:1-100%). In 5 patients, FDG-avid disease would have been completely excluded from the PTV-CT (median volume missed = 6%, range:2-92%). For 8 patients, less than 95% of PTV-PET would have received at least 95% of prescription dose based on the CT-based plan. GTV-CT underestimated the cranial and caudal extent of FDG-avid tumor in 14 (36%) and 10 (26%) patients respectively. There were no significant differences in radiation doses to the lungs and liver. 5-year overall survival and locoregional failure free survival were 24%, and 42% respectively. Conclusions: PET/CT prevented futile radiotherapy for 1 in 4 patients and avoided geographic misses without significant impact on normal tissues in apparently localized esophageal cancer. However, survival remains suboptimal and indicates the need for further improvement in planning and therapeutic paradigms.