PET-directed combined modality therapy for gastroesophageal junction cancer: Results of the multicentre prospective MEMORI trial of the German Cancer Consortium (DKTK)

Background: Positron emission tomography (PET) may differentiate responding and non-responding tumours early in the treatment of locally advanced gastroesophageal junction adenocarcinomas. Early PET non-responders (P-NR) after induction CTX might benefit from changing to chemoradiation (CRT). Methods: Patients underwent baseline ¹⁸F-FDG PET followed by 1 cycle of CTX. PET was repeated at day 14–21 and responders (P-R), defined as ≥35% decrease in SUV_mean from baseline, continued with CTX. P-NR switched to CRT (CROSS). Patients underwent surgery 4–6 weeks post-CTX/CRT. The primary objective was an improvement in R0 resection rates in P-NR above a proportion of 70%. Results: In total, 160 patients with resectable gastroesophageal junction adenocarcinomas were prospectively investigated by PET scanning. Eighty-five patients (53%) were excluded. Seventy-five eligible patients were enrolled in the study. Based on PET criteria, 50 (67.6%)/24 (32.4%) were P-R and P-NR, respectively. Resection was performed on 46 responders, including one patient who withdrew the ICF, and 22 non-responders (per-protocol population). R0 resection rates were 95.6% (43/45) for P-R and 86.4% (19/22) for P-NR. No treatment related deaths occurred. With a median follow-up time of 24.5 months, estimated 18 months DFS was 75.4%/64.2% for P-R/P-NR, respectively. The estimated 18 months OS was 95.5% for P-R and 68.2% for P-NR. Conclusion: The primary endpoint of the study to increase the R0 resection rate in metabolic NR was not met. PET response after induction CTX is prognostic for outcome with a prolonged OS and DFS in PET responders. Trial registration: NCT 2014-000860-16.