Top 5 Papers
#1
Source: The Lancet | Authors: Janjigian YY, … Tabernero J — 147 centres, 20 countries | Published: September 9, 2026
Score: 17/20 — Base: The Lancet (9) + Phase III RCT (+3) + new standard of care (+3) + overall survival benefit (+2) = 17. No colleague engagement bonus: the X sweep confirmed no qualifying KOL posts on this paper in the window.
MATTERHORN randomised 948 patients with resectable gastric or gastro-oesophageal junction adenocarcinoma to perioperative durvalumab plus FLOT or placebo plus FLOT across 147 centres in 20 countries. Overall survival favoured durvalumab with an HR of 0.78 (95% CI 0.63–0.96, p=0.021), clearing the prespecified significance threshold of p<0.0499. Treatment-related deaths were uncommon and near-balanced at 6 (1%) versus 2 (<1%). The authors conclude that durvalumab plus FLOT is a new standard treatment option. Worth reading the endpoint structure carefully: overall survival was a key secondary endpoint, with event-free survival the previously reported primary.
Post angle: This digest covered NCCN adding durvalumab + FLOT as category 1 back in April. This is the survival data that was standing behind that decision. #GIOnc #GastricCancer
#2
Source: ESMO Open | Authors: Coutzac C, Zaanan A, … de la Fouchardière C — French multicentre (NCT04795661) | Published: September 9, 2026
Score: 9/20 — Base: ESMO Open (6, row added 2026-09-10) + Phase II (+2) + biomarker-guided/precision (+1) = 9. No colleague engagement bonus confirmed.
The oesophagogastric cohort of the IMHOTEP umbrella trial treated 41 patients with localised dMMR/MSI adenocarcinoma with perioperative pembrolizumab, 36 evaluable for efficacy. All 27 patients who went to surgery achieved an R0 resection, and 4 of 27 (14.8%) reached the primary endpoint of pathological complete response. At a median 25.8 months, 25 of 27 (93%) were free of progression, recurrence or death, and grade ≥3 treatment-related adverse events occurred in 22%. The authors' conclusion is deliberately restrained — feasible and safe, with the duration and nature of neoadjuvant immunotherapy still to be optimised.
Post angle: Set this against the IMHOTEP colorectal cohort this digest led with in June: pCR 52.7% there, 14.8% here. Same drug, same design, same trial. #GIOnc #ImmunOnc
#3
Source: JAMA Surgery | Authors: El Khoury S, … Shen JP — MD Anderson, external validation at MSKCC | Published: September 9, 2026
Score: 9/20 — Base: JAMA Surgery (7, row added 2026-09-10) + retrospective/real-world (+1) + biomarker-guided/precision (+1) = 9. No colleague engagement bonus confirmed.
This retrospective cohort followed 439 patients with stage I–III appendiceal adenocarcinoma at MD Anderson for a median 62.6 months, with external validation in 128 stage II patients from MSKCC. Relapse occurred in 9.4% of surgical patients — 6% of stage II and 19.5% of stage III — and carried a steep survival penalty (5-year OS 95.7% without relapse versus 77.2% with, HR 5.50, 95% CI 3.07–9.83, p<0.001). Histology and molecular features drove risk: mucinous HR 5.60 and enteric-type HR 6.60 versus goblet-cell, TP53 in goblet-cell HR 6.93, GNAS in non-goblet HR 17.0. Adjuvant chemotherapy was not associated with improved recurrence-free or overall survival on multivariable analysis (RFS HR 0.98, p=0.90). Note the univariate signal pointed the other way, which is confounding by indication — the multivariable null is the honest reading, not evidence of harm.
Post angle: The colorectal risk-factor playbook does not transfer here: poor differentiation, perforation, LVI and PNI predicted nothing. Histology and GNAS/TP53 did the work. #GIOnc
#4
Source: Gut | Authors: Gut, September 2026 | Published: September 7, 2026
Score: 8/20 — Base: Gut (7, row added 2026-09-10) + biomarker-guided/precision (+1) = 8. Mechanistic study, so no trial study-type bonus applies. No colleague engagement bonus confirmed.
This study identifies an S1P–TREM2–NRF2 axis, acting through GPX4 and SLC7A11, that renders tumour-associated neutrophils in hepatocellular carcinoma resistant to ferroptosis. Deleting TREM2 or pharmacologically inhibiting S1P synthesis restored CD8+ T-cell function and improved anti-PD-1 efficacy in HCC models. In patient samples, TREM2+ PMN-MDSCs correlated with SPHK1/2 expression, T-cell dysfunction and poorer prognosis. The work is preclinical, but the S1P synthesis arm points at an existing and already-drugged pathway.
Post angle: A druggable explanation for why HCC neutrophils survive where they ought to die — and why that keeps checkpoint blockade underpowered. #GIOnc #HCC
#5
Source: Nature Communications | Authors: Liao Y, … Zhao L — Southern Medical University and Peking Union Medical College | Published: August 10, 2026
Score: 7/20 — Base: Nature Communications (6) + biomarker-guided/precision (+1) = 7. Spatial/single-cell translational study with in-vivo validation; no trial study-type bonus. No colleague engagement bonus confirmed.
Combining spatial transcriptomics and single-cell RNA sequencing on human gastric cancer specimens with in-vivo functional work, this study maps an immune niche specific to HER2-positive disease. C1Q+ macrophages engage CD8+ T cells through Galectin-9 (LGALS9), which is downregulated in HER2-positive tumours; knocking LGALS9 down enhanced CD8+ T-cell function and suppressed tumour growth in a CD8-dependent manner, amplified by anti-PD-1. SFRP2+ cancer-associated fibroblasts drove CD8+ exhaustion via CXCL12–CXCR4, reversible by CXCL12 knockdown. HER2-positive tumours were enriched for C1Q+ macrophages and CD8+ Tem cells and depleted of SFRP2+ CAFs. The predictive framing is a proposal from descriptive and mouse data — no patient cohort was treated on the basis of it.
Post angle: A mechanistic case for why anti-HER2 plus checkpoint blockade should work in a subset — proposed, not yet validated in patients. #GIOnc #GastricCancer
Additional Papers of Interest
-
Nature Communications — H. pylori-induced gland hyperplasia runs through YAP-dependent fetal-like transcription and loss of epithelial BMP signalling, via an epithelial-to-immune-to-stromal cascade; stromal IL-1 receptor deletion abolishes it
-
ESMO Open — 92 trials and 659,904 participants over 45 years; 50% met their primary endpoint, oesophagogastric 58.3%, with success odds rising under pharmaceutical sponsorship and falling for weaker pretrial evidence
-
Clinical Cancer Research — paired 89Zr immunoPET and 225Ac alpha therapy on the codrituzumab backbone; a single 9.3 kBq dose gave excellent local control with several complete pathologic responses. Entirely preclinical
-
Nature Communications — mutant beta-catenin alone is a negative fitness event in healthy liver, becoming adaptive and tumorigenic only as ageing or chronic injury erodes tissue integrity, via AKT-NRF2
Back to all digests