Serial imaging using [18F]Fluorodeoxyglucose positron emission tomography and histopathologic assessment in predicting survival in a population of surgically resectable distal oesophageal and gastric adenocarcinoma following neoadjuvant therapy

BACKGROUND AND OBJECTIVES: We retrospectively evaluated the value of PET/CT in predicting survival and histopathological tumour-response in patients with distal oesophageal and gastric adenocarcinoma following neoadjuvant treatment. METHODS: Twenty-one patients with resectable distal oesophageal adenocarcinoma and 14 with gastric adenocarcinoma between January 2002 and December 2011, who had undergone serial PET before and after neoadjuvant therapy followed by surgery, were enrolled. Maximum standard uptake value (SUVmax) and metabolic tumour volume were measured and correlated with tumour regression grade and survival. RESULTS: Histopathological tumour response (PR) is a stronger predictor of overall and disease-free survival compared to metabolic response. ∆%SUVmax ≥70% was the only PET metric that predicted PR (82.4% sensitivity, 61.5% specificity, p = 0.047). Histopathological non-responders had a higher risk of death (HR 8.461, p = 0.001) and recurrence (HR 6.385, p = 0.002) and similarly in metabolic non-responders for death (HR 2.956, p = 0.063) and recurrence (HR 3.614, p = 0.028). Ordinalised ∆%SUVmax showed a predictive trend for OS and DFS, but failed to achieve statistical significance. CONCLUSIONS: PR was a stronger predictor of survival than metabolic response. ∆%SUVmax ≥70% was the best biomarker on PET that predicted PR and survival in oesophageal and gastric adenocarcinoma. Ordinalisation of ∆%SUVmax was not helpful in predicting primary outcomes.