Top 5 Papers
#1
Source: ASCO 2026 Annual Meeting (Oral Abstract 4008) / PR Newswire | Authors: Lu Z et al. — multicenter (China), ASCO 2026 GI oral session (June 1) | Published: May 22, 2026 (press release; oral June 1)
Score: 14/20 — Base 8 (ASCO oral) + Ph3 RCT (+3) + OS/PFS benefit (+2) + colleague engagement on X (+1) = 14
Phase 3 BL-B01D1-305 randomized 2L recurrent/metastatic esophageal squamous cell carcinoma (ESCC) to izalontamab brengitecan (iza-bren) — a first-in-class anti-EGFR/HER3 bispecific antibody-drug conjugate carrying a topoisomerase I inhibitor payload — versus physician's choice chemotherapy. Both co-primary endpoints (OS and PFS) were positive per the company's pre-ASCO disclosure, with detailed efficacy and safety data scheduled for oral presentation at the ASCO 2026 Annual Meeting GI session on June 1. If confirmed at the podium, this would be the first positive ADC Phase 3 in ESCC and a likely new 2L standard in a setting dominated by single-agent chemotherapy with sub-10-month survival.
Post angle: First positive ADC Phase 3 in 2L esophageal SCC. Iza-bren (anti-EGFR/HER3 bispecific ADC, topo-I payload) hit BOTH OS and PFS endpoints. Embargoed data drops at #ASCO26 June 1. If it holds, 2L ESCC chemo gets dethroned. #GIOnc #EsophagealCancer #ADC
#2
Source: JAMA Oncology | Authors: Yan C, Zhu Z, et al. — 9-center Chinese consortium | Published: May 21, 2026
Score: 12/20 — Base 7 (JAMA Oncology) + Ph3 RCT (+3) + survival benefit (+2) = 12
Phase 3 multicenter RCT in 222 patients with laparoscopy-confirmed gastric adenocarcinoma peritoneal metastases — historically a near-uniformly fatal scenario with median OS under 14 months on systemic chemotherapy alone. Adding intraperitoneal paclitaxel to IV paclitaxel + S-1 nearly doubled the chance of long-term benefit: median OS 19.4 vs 13.9 mo (HR 0.67, p=0.01), median PFS 11.2 vs 7.2 mo (HR 0.72). Grade 3/4 AEs were essentially unchanged (38.5% vs 41.9%) and there were no treatment-related deaths. This is the first Phase 3 to demonstrate an OS benefit for IP-directed therapy in gastric carcinomatosis and may rapidly become the new reference regimen for peritoneal-only metastatic disease at high-volume centers.
Post angle: DRAGON-01 (JAMA Oncol Ph3, n=222): IP + IV paclitaxel + S-1 vs IV paclitaxel + S-1 in gastric cancer peritoneal mets → mOS 19.4 vs 13.9 mo (HR 0.67, p=0.01), mPFS 11.2 vs 7.2 mo. No toxicity tradeoff. First Ph3 OS win for IP chemo in gastric carcinomatosis. #GIOnc #GastricCancer #Peritoneal
#3
Source: Gut | Authors: Cai R, Zhang J, Deng Y, et al. — Sun Yat-sen University Sixth Affiliated Hospital | Published: May 22, 2026
Score: 7/20 — Base 6 (Gut) + biomarker/precision-target mechanism (+1) = 7
PTEN loss occurs in 19-36% of MSS colorectal cancers and is associated with reduced CD8+ tumor-infiltrating lymphocytes and poor ICB outcomes — but the mechanism has been opaque. This work nails it: PTEN deficiency triggers KEAP1 degradation → NRF2 hyperactivation → selective autophagy of MHC class I, stripping tumor cells of antigen presentation. Pharmacologic NRF2 inhibition with ML385 restored surface MHC-I, reignited CD8+ infiltration, and resensitized PTEN-null MSS CRC models to anti-PD-1. A clean, mechanism-to-target story that explains a large slice of MSS-CRC ICB resistance and identifies a tractable druggable axis — NRF2 inhibitors are already in clinical development for other indications.
Post angle: Why MSS CRC doesn't respond to PD-1: PTEN loss → KEAP1 degradation → NRF2 hyperactivation → autophagic chewing-up of MHC-I. ML385 (NRF2 inhibitor) restores MHC-I and resensitizes to anti-PD-1 in PTEN-null MSS CRC models. Druggable axis in 19-36% of MSS tumors (Gut). #GIOnc #CRC #ImmunoOnc
#4
Source: Cancer Discovery | Authors: Elhossiny AM, Pasca di Magliano M, et al. — University of Michigan Rogel Cancer Center | Published: May 21, 2026
Score: 7/20 — Base 6 (Cancer Discovery) + translational/precision impact (+1) = 7
Most pancreatic intraepithelial neoplasias (PanINs) never become pancreatic cancer, but the field has lacked a spatial framework that explains why. Using exceptionally rare donor pancreas tissue containing PanINs in their native microenvironment, the Pasca di Magliano group built a side-by-side spatial atlas of PanIN vs invasive PDAC. The headline finding: PanIN epithelium sits on a continuum with PDAC epithelium, but the stroma is strikingly different. The leap from precancer to cancer requires profound geographical reorganization of myeloid cells and lymphocytes plus the emergence of a cancer-specific SMA+/LRRC15+/LEF1+ fibroblast population driven by WNT signaling. Absence of that stromal reprogramming is the dominant brake on PanIN-to-PDAC progression — pointing to the microenvironment, not the epithelium, as the most actionable interception target.
Post angle: Why most PanINs never become PDAC: it's the stroma, not the epithelium. Spatial atlas of human donor pancreas shows PanIN epithelium ≈ PDAC epithelium, but PDAC requires emergence of SMA+/LRRC15+/LEF1+ WNT-driven CAFs + remodeled myeloid/lymphoid geography. Interception target: the microenvironment (Cancer Discovery). #GIOnc #PDAC #Interception
#5
Source: BMC Cancer | Authors: Wang S, Qu M, et al. — Nanjing University of Chinese Medicine | Published: May 21, 2026
Score: 7/20 — Base 5 (BMC Cancer) + meta-analysis (+2) = 7
Meta-analysis of 6 randomized trials (n=1,460) attempting to settle the recurring dMMR/MSI-H mCRC question: PD-1 monotherapy vs dual checkpoint blockade vs chemo. PD-1 mono beat chemo on response (OR 1.52, p=0.04) with less grade ≥3 toxicity. Dual ICI (nivo+ipi) beat PD-1 mono on response (OR 1.77, p=0.001) but doubled severe-toxicity risk (OR 1.73, p=0.002) — and individual-trial OS HRs (0.74-1.03) never reached significance. The clinical takeaway is unchanged from CheckMate-8HW: dual ICI buys deeper responses but no clear survival edge over PD-1 monotherapy, so the tolerability-vs-depth tradeoff still has to be made patient-by-patient. Useful synthesis but no new data.
Post angle: dMMR/MSI-H mCRC meta-analysis (6 RCTs, n=1,460): dual ICI > PD-1 mono on ORR (OR 1.77) but doubles G≥3 toxicity (OR 1.73). Individual-trial OS HRs (0.74-1.03) all non-significant. CheckMate-8HW conclusion holds: depth ≠ survival here. Tolerability still tips most decisions (BMC Cancer). #GIOnc #CRC #ImmunoOnc
Additional Papers of Interest
-
ASCO 2026 Annual (Oral Abstract 4006) — Dual ICI + chemo did NOT improve outcomes over standard chemo in 1L HER2-neg GEA; an important negative trial that should curb the impulse to escalate beyond single-agent ICI + chemo in this setting.
-
ASCO 2026 Annual (Oral Abstract 4007) — Adds a prostaglandin E2/EP4 pathway block to standard ICI + chemo in 1L HER2-neg GEA; novel mechanism aimed at TAMs and Tregs in the gastric TME. Detailed efficacy at the podium June 1.
-
J Robot Surg (n=140) — Robotic natural-orifice specimen extraction matched robotic TME on oncologic short-term endpoints (lymph node yield, conversion, complications) while delivering faster GI recovery, less pain, and zero wound-site complications (vs 5.6%). Long-term oncologic data still pending.
-
Cancer Discovery News — Real-world signal that starting a GLP-1 RA after stage I-III diagnosis of CRC, HCC, breast, or lung cancer may reduce progression to stage IV. Hypothesis-generating only; experts not ready to prescribe outside RCTs, but worth tracking as the GLP-1/oncology question moves from observation to prospective design.
Back to all digests