Gastric and gastro-oesophageal junction adenocarcinoma

Gastric and gastro-oesophageal junction (GEJ) adenocarcinoma is a heterogeneous disease spanning intestinal and diffuse (signet-ring) histological subtypes and a spectrum from proximal junctional tumours to distal gastric tumours. GLOBOCAN 2024 estimates rank stomach cancer fifth worldwide for both incidence and mortality, with 980,286 new cases and 641,554 deaths a year. In the United States, SEER projects 31,510 new cases and 10,740 deaths for 2026, and five-year relative survival is 39.8% overall, but that figure conceals an enormous stage gradient: 78.1% for localised disease, which is 32% of cases, 39.0% for regional disease at 23%, and 8.1% for distant disease at 35%. Most Western patients present with regional or distant disease, which is why the field has organised itself around biomarker-directed systemic therapy rather than around early detection.

Treatment splits by resectability and, in the metastatic setting, by biomarker. Resectable gastric or GEJ adenocarcinoma is treated with perioperative FLOT chemotherapy, which the FLOT4 trial showed improves overall survival over ECF/ECX (median 50 versus 35 months), combined with D2 lymphadenectomy at surgery; the phase 3 MATTERHORN trial has since shown that adding durvalumab to FLOT improves event-free survival further and NCCN restricts the category-1 recommendation for that combination to PD-L1-positive tumours. Resectable oesophageal or junctional (Siewert I-II) adenocarcinoma has long been treated with neoadjuvant chemoradiotherapy per the CROSS trial, which gave a median overall survival of 49.4 versus 24.0 months and a 13 percentage point absolute ten-year survival benefit, but the newer ESOPEC trial reports perioperative FLOT outperforming CROSS in this population (median overall survival 66 versus 37 months), so practice is shifting. In the metastatic setting, first-line therapy is chosen by testing for HER2 (trastuzumab, with pembrolizumab added if PD-L1 CPS is at least 1), PD-L1 CPS in HER2-negative disease (nivolumab or pembrolizumab added to chemotherapy), CLDN18.2 positivity in HER2-negative disease (zolbetuximab), and mismatch-repair status, with pembrolizumab recommended second line for MSI-H/dMMR disease; trastuzumab deruxtecan is approved second line after trastuzumab progression.

Nuclear medicine's role is limited and this is the central problem the field is trying to solve. FDG PET/CT is an adjunct to CT rather than a replacement for staging laparoscopy, and NCCN guideline text records FDG-PET/CT accuracy of 68% in preoperative staging against 47% for FDG-PET alone and 53% for CT alone, while a Dutch prospective cohort found FDG-PET/CT sensitivity of only 33% for distant metastases against a specificity of 97%, and staging laparoscopy sensitivity of 82% for peritoneal disease against 78% specificity. FDG uptake is characteristically low in diffuse and signet-ring histology, a common subtype in gastric cancer, so the modality is weakest exactly where peritoneal spread most needs to be excluded; PET sensitivity for peritoneal carcinomatosis has been reported at only about 50%. This gap between what biomarker testing can select for systemically and what imaging can see anatomically, particularly for peritoneal disease, is the opening that new imaging targets such as CLDN18.2 and FAPI tracers are being explored to fill.