Abstract Background Neoadjuvant Chemoradiotherapy (nCRT) for locally advanced oesophageal cancer improves survival. Currently, there is equipoise in the survival benefit of nCRT versus neoadjuvant chemotherapy. Excessive radiotherapy to the gastric conduit prior to surgery has long been a concern. Recent studies have linked radiation doses to the gastric fundus around 30 Gy to an increased risk of anastomotic leak. This study reviews a simple process aiming to reduce radiation dose to the gastric fundus. Methods This is a single-centre, retrospective analysis of 63 adult patients treated with nCRT (CROSS protocol) then oesophagectomy with a gastric conduit reconstruction. The Gastric Fill (GF) arm were given 500 mL of water prior to radiotherapy planning and treatment. The control group were consecutive patients prior to the implementation of the GF protocol. The fundus was mapped on radiotherapy software and planned to treat it like other organs at risk. Radiation dose to the fundus was recorded and compared between the two groups. Results The GF arm included 25 patients, with 38 patients in the control arm. Patients drank 350–800 mL of water prior to planning and treatment. There were no reported aspiration events. Radiotherapy techniques included VMAT and 3DCRT. All GF patients were treated with VMAT. There was a higher proportion of mid-oesophageal tumours in the control group (21% compared with 4%). The mean volume of gastric fundus in the GF group (mean = 215 cc, SD 144) was higher than the control group (mean = 155 cc, SD 100). The proportion of gastric fundus receiving >30 Gy was lower in the GF group (mean = 10.9% SD = 16.0) compared with the control group (mean = 19.1% SD = 24.8). Conclusion Analysis did not show statistically significant difference between the groups, however, there was a trend suggesting radiation to the gastric fundus was reduced in the GF group. This study shows the feasibility of a simple and low risk intervention in reducing the radiation to the future gastric conduit. Further review with a larger cohort, and impact on anastomotic complications would be of interest for future review.