Top 5 Papers
#1
Source: Clinical Cancer Research | Authors: Wainberg ZA, Lee KW, ... Han SW; multicenter (US, South Korea, Germany), NCT (Leap Therapeutics) | Published: July 23, 2026
Score: 9/20 — Base 6 (Clinical Cancer Research) + Phase II RCT (+2) + biomarker-guided/precision (+1, prespecified plasma DKK1 signal) = 9
DeFianCe was a randomized phase II trial of sirexatamab, a first-in-class DKK1-neutralizing antibody, added to investigator's-choice FOLFIRI or mFOLFOX6 plus bevacizumab as second-line therapy for metastatic colorectal cancer (n=188). The study did not meet its primary endpoint: median PFS was 9.2 months with sirexatamab versus 8.3 months with control (HR 0.84, 95% CI 0.58–1.21, one-sided p=0.17), and ORR was 35.1% versus 26.6%. However, in prespecified exploratory analyses, patients with baseline plasma DKK1 above the median derived consistent benefit across PFS (HR 0.61), OS (HR 0.42), and ORR (38.0% vs 23.7%). Safety was comparable between arms. The authors conclude that DKK1-directed therapy may require biomarker-selected evaluation rather than all-comer development.
Post angle: A negative all-comer phase II with a genuinely interesting biomarker story — DKK1-high patients may still benefit. A case study in why we should design selected-population trials rather than bury a signal. #CRC #GIOnc #PrecisionMedicine
#2
Source: Gastric Cancer | Authors: Korpan M, ... Ilhan-Mutlu A (senior); multicenter (Medical University of Vienna + Austrian network) | Published: July 23, 2026
Score: 8/20 — Base 5 (Gastric Cancer, specialty journal) + real-world/retrospective (+1) + biomarker-guided PD-L1 CPS≥10 (+1) + colleague engagement (+1, surfaced via colleague discussion on X) = 8
This exploratory, retrospective, multicenter real-world study from Austria evaluated therapeutic strategies in 55 patients with resectable dMMR/MSI-H gastroesophageal adenocarcinoma — to the authors' knowledge the first to compare the full spectrum of chemotherapy, chemoimmunotherapy, immunotherapy, and upfront resection. Pathological complete response rates were 22% (2/9) with chemotherapy, 27% (4/15) with chemoimmunotherapy, and 100% (5/5) with doublet immune checkpoint inhibition. Among patients managed non-operatively, the clinical complete response rate was 37.5% in the immunotherapy subgroup. Complete responses (pCR/cCR) were significantly associated with PD-L1 CPS ≥10 (p=0.027). Although the numbers are small, the data lend real-world support to doublet ICI and to the emerging debate over non-operative management in this molecular subset.
Post angle: Doublet immunotherapy = 100% pCR in dMMR/MSI-H gastroesophageal cancer (small n, real-world). The 'can we skip surgery?' conversation in MSI-H upper GI is officially live. #GastricCancer #Immunotherapy #GIOnc
#3
Source: Cancer Communications | Authors: Wu F, ... Wang G (senior); multicenter (Hebei Medical University network, China) | Published: July 23, 2026
Score: 7/20 — Base 5 (Cancer Communications, not on tier table) + Phase II RCT (+2) = 7. pCR primary endpoint missed significance (p=0.052); MPR gain was significant but secondary, so no clinical-impact bonus.
Neo-STAR was a randomized phase II trial testing whether adding the PD-1 inhibitor tislelizumab to short-course radiotherapy-based total neoadjuvant therapy (SCRT followed by CAPOX) improves outcomes in locally advanced rectal cancer (n=111). The pathological complete response rate was 45.3% with tislelizumab versus 27.6% without (OR 2.17, p=0.052), narrowly missing statistical significance, while the major pathological response rate was significantly higher at 50.9% versus 31.0% (OR 2.31, p=0.033). Grade 3–4 adverse events were comparable between arms, with anemia most common. The trial adds to accumulating evidence that immunotherapy may deepen pathological responses even in mismatch-repair-proficient rectal cancer, though longer follow-up for survival and organ-preservation endpoints is awaited.
Post angle: Adding tislelizumab to short-course-RT TNT: pCR 45% vs 28% (just misses p<0.05) but MPR significantly better. Immunotherapy keeps knocking on the door of MMR-proficient rectal cancer. #CRC #RectalCancer #GIOnc
#4
Source: Journal of Hepatology | Authors: Ten Haaft BHEA, ... Erdmann JI (senior); 27 centers across 9 countries | Published: July 24, 2026
Score: 7/20 — Base 6 (Journal of Hepatology, premier hepatology journal) + retrospective/real-world (+1) = 7. Large multinational cohort (n=1,307) strengthens the prognostic tool.
This international, retrospective cohort study (27 centers across 9 countries, 1,307 resected patients, 2006–2022) developed the ABC system for preoperative risk stratification in perihilar cholangiocarcinoma, a cancer in which curative-intent surgery carries substantial mortality and high early-recurrence risk. Three independent prognostic factors — tumor size ≥25 mm (aHR 1.32), CA19-9 ≥500 U/mL (aHR 1.49), and WHO performance status ≥1 (aHR 1.35) — each contribute one point. Compared with the lowest-risk (ABC-0) group, patients with the maximal score had a 3.4-fold higher 90-day mortality (21% vs 6%), a 3.0-fold shorter median OS (13 vs 39 months), and a markedly lower 5-year OS (11% vs 35%). The authors argue clinicians should be reluctant to offer resection to ABC-3 patients, providing a simple, reproducible tool to refine surgical selection beyond anatomy alone.
Post angle: A simple preoperative ABC score (tumor ≥25 mm + CA19-9 ≥500 + PS ≥1) that separates 5-yr OS of 35% vs 11% in perihilar cholangiocarcinoma. Anatomy alone isn't enough — think twice before a high-risk Whipple. #Cholangiocarcinoma #GIOnc
#5
Source: Gut | Authors: Lee M, ... Chung MJ, Jang SI (senior); Korean National Health Insurance Service cohort | Published: July 24, 2026
Score: 7/20 — Base 6 (Gut) + retrospective/real-world (+1) = 7. Very large nationwide cohort (n=402,663) over 15 years.
This 15-year nationwide Korean cohort study (402,663 individuals, 2005–2019) examined how the severity and trajectory of new-onset diabetes (NOD) relate to pancreatic cancer risk, moving beyond the simple presence or absence of diabetes. Pancreatic cancer risk rose stepwise with baseline antidiabetic treatment intensity, with adjusted HRs of 3.01, 4.32, and 5.60 for no antidiabetics, oral agents, and insulin, respectively, versus people without diabetes. Rapid treatment escalation within 6 months conferred further risk; drug-naïve individuals who initiated therapy within 6 months had a higher risk (aHR 6.50) than those stably managed on oral agents (aHR 3.67). These associations were most pronounced for cancers diagnosed within 3 years of NOD onset, suggesting that both initial severity and early aggravation of new-onset diabetes could help sharpen pancreatic-cancer surveillance strategies.
Post angle: It's not just 'new diabetes' — it's severe or rapidly-worsening new diabetes that flags pancreatic cancer risk (adjusted HR up to 6.50), especially in the first 3 years. A pragmatic early-detection signal from 402,663 people. #PancreaticCancer #GIOnc #EarlyDetection
Additional Papers of Interest
-
J Gastrointest Cancer (RCT, n=202) — chemo-first TNT raised pCR to 55.4% vs 42.6% (p=0.023) vs radiation-first sequencing, with less thrombocytopenia; OS not yet different
-
J Gastrointest Cancer (meta-analysis, 7 cohorts, 1,803 patients) — no OS (HR 0.90) or DFS (HR 0.98) difference; minimally invasive gave less blood loss and shorter stay at the cost of longer operative time
-
Ann Surg Oncol (expert debate) — two gastric-cancer experts argue both sides of omitting surgery after neoadjuvant immunotherapy, framed around dMMR/MSI-H pCR rates near 60% vs ~20% in MMR-intact tumors
-
Radiother Oncol (709 patients, 36 UK sites) — dose-escalated ACT5 patients retained bowel-function deficits at 6 months, informing the toxicity cost of intensifying anal-cancer chemoradiation
-
Eur J Cancer (phase II, n=49) — first-line lenvatinib + nivolumab missed its prespecified ORR target (32% vs 43%), though median OS reached 26.6 months; a cautionary IO-combo readout in HCC
-
J Gastrointest Cancer (case report) — mFOLFOX6 + zolbetuximab induced a ypT0N0 pCR and R0 conversion in initially unresectable CLDN18.2-strong gastric cancer with peritoneal disease
Back to all digests