Metabolic nodal response as a prognostic marker after neoadjuvant therapy for oesophageal cancer

Background The ability to predict recurrence and survival after neoadjuvant chemotherapy (NAC) and surgery for oesophageal cancer remains elusive. This study evaluated the role of [18F]fluorodeoxyglucose (FDG) PET–CT in assessing tumour and nodal response as a prognostic marker. Methods This was a single-centre UK cohort study. From 2006 to 2014, patients with oesophageal cancer staged with PET–CT before NAC, and restaged by CT or PET–CT before resection, were included. Pathological tumour response was evaluated using Mandard regression grades. Metabolic tumour and nodal responses (mTR and mNR respectively) were quantified using absolute and threshold reductions. Results Among 294 included patients, mTR and mNR independently predicted prognosis before surgery. After surgery, mNR (but not mTR), pathological tumour response, resection margin status and pathological node category predicted prognosis. Patients with FDG-avid nodal disease after NAC were at high risk of recurrence/death at 1 and 2 years (43 and 71 per cent respectively; P = 0·030 and P = 0·025 versus patients without avid nodes), and had a worse prognosis than patients with non-avid nodal metastases: hazard ratio 4·19 (95 per cent c.i. 1·87 to 9·40) and 2·11 (1·12 to 3·97) respectively versus patients without nodal metastases. Considering mTR and mNR response separately improved prognostication. Conclusion mNR is a novel prognostic factor, independent of conventional N status. Primary and nodal tumours may respond discordantly and patients with FDG-avid nodes after NAC have a poor prognosis.