TOPLINE

In an analysis of patients with resectable oesophagogastric adenocarcinoma from two large trials, those with mismatch repair-deficient (MMRd) tumours treated with perioperative chemotherapy had significantly better survival than those with MMR-proficient (MMRp) tumours, and chemotherapy did not erase that advantage.

METHODOLOGY

• Researchers conducted a retrospective analysis of patients with resectable oesophagogastric adenocarcinoma enrolled in the UK MRC OE05 and ST03 trials, which evaluated different perioperative chemotherapy regimens in patients randomly assigned between January 2005 and March 2015.
• A total of 1424 patients were included: 736 from the OE05 trial (oesophageal or junctional adenocarcinoma) and 688 from the ST03 trial (gastric or junctional adenocarcinoma); by tumour location, 45.6% were in the oesophagus, 26.1% were in the gastroesophageal junction, and 28.2% were in the stomach.
• MMR status was assessed using brightfield single-protein immunohistochemistry in 497 patients and multiplex immunofluorescence in 927 patients, evaluating the expression of MLH1, MSH2, MSH6, and PMS2 proteins in formalin-fixed, paraffin-embedded tumour tissue samples.
• Patients received one of four treatment regimens: neoadjuvant cisplatin and fluorouracil (26.8%); neoadjuvant epirubicin, cisplatin, and capecitabine (24.9%); perioperative epirubicin, cisplatin, and capecitabine (24.7%); or perioperative epirubicin, cisplatin, and capecitabine plus bevacizumab (23.6%).
• The primary outcome was overall survival (OS); secondary outcomes included progression-free survival (PFS) and postprogression survival, with a median follow-up duration of 64 months from randomisation.

TAKEAWAY

• MMRd was identified in 78 (5.5%) tumours, with 80.8% showing MLH1 loss alone or combined with PMS2 loss, 3.8% showing isolated PMS2 loss, and 15.4% showing MSH2 and/or MSH6 loss.
• Patients with MMRd vs MMRp tumours had significantly improved OS (median, not reached vs 28.0 months; P = .0009; 5-year rates were 50.8% vs 34.3%) and PFS (median, 56.7 vs 21.4 months; P = .0032; 5-year rates were 47.8% vs 31.6%).
• In multivariable analysis adjusting for preoperative factors, MMRd remained independently associated with improved OS (hazard ratio [HR], 0.60; P = .007) and PFS (HR, 0.64; P = .011).
• In the subgroup of patients who had surgery, MMRd was linked to longer OS (HR, 0.48) and PFS (HR, 0.55); however, no survival difference by MMR status was seen among patients who did not undergo resection. No significant difference in postprogression survival was observed by MMR status (log-rank P = .31).

IN PRACTICE

"Overall, our findings do not support the hypothesis that peri‑operative chemotherapy is detrimental in patients with MMRd oesophagogastric cancer," the authors of the study wrote.

SOURCE

The study was led by Panagiotis Ntellas, Department of Gastrointestinal Medical Oncology, Royal Marsden Hospital, London and Surrey, England. It was published online on August 25, 2026, in the British Journal of Cancer.

LIMITATIONS

A surgery-only comparison group was not included, so the study could not determine whether MMR status predicted the benefit from chemotherapy. Several subgroup analyses were underpowered, especially the small group with isolated PMS2 loss. Because the participants came from clinical trials, they were generally healthier and younger than patients seen in everyday practice, which may have limited generalisability.

DISCLOSURES

This study did not receive any specific funding. Some authors reported receiving grants or honoraria or having other ties with various pharmaceutical companies, including BMS, MSD, Roche, AstraZeneca, and Johnson & Johnson. Full disclosures are reported in the original article.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.