📬 GI Cancer Intelligence Digest — Week 2026-W35

ESMO added daraxonrasib to second-line metastatic pancreatic cancer, TALENTOP showed liver resection after atezolizumab/bevacizumab prolonged time to treatment failure in advanced HCC, and the ASCO Living Guideline for gastroesophageal cancer moved to biomarker-directed first-line therapy.

⚡ TL;DR

📄 Publications

Atezolizumab with or without tiragolumab in unresectable esophageal squamous cell carcinoma following definitive concurrent chemoradiotherapy (SKYSCRAPER-07): a randomised, phase III study. — 760 patients were randomised to atezolizumab plus tiragolumab (n=257), atezolizumab plus placebo (n=250), or placebo (n=253)
Liver resection after atezolizumab and bevacizumab versus maintenance therapy for locally advanced hepatocellular carcinoma (TALENTOP): a multicentre, open-label, randomised, phase 3 trial. — 489 patients entered the induction phase and 201 were randomised to surgery (n=101) or maintenance therapy (n=100)

🐦 Twitter / 🦋 Bluesky

Genome-wide in vivo CRISPR screens identify P2RY8 and GNAS as brakes on human T cells in solid tumors, including pancreatic and gastroesophageal models — The study reports the first genome-wide in vivo screens for genetic modifications to enhance human T cell therapies for solid tumors, published in Nature
ESMO Express Update adds daraxonrasib to second-line metastatic pancreatic cancer — The pan-RAS(ON) inhibitor daraxonrasib is included in second-line therapy for metastatic pancreatic cancer in the ESMO express guideline update
TALENTOP phase III: liver resection after atezolizumab plus bevacizumab extends time to treatment failure in advanced HCC with macrovascular invasion — In TALENTOP, 201 patients who responded to atezolizumab plus bevacizumab were randomised to resection followed by 12 months of atezo/bev or to maintenance atezo/bev
Phase III Chinese trial: oral paclitaxel beats IV paclitaxel on overall survival in second-line advanced gastric cancer — In 536 patients with gastric cancer progressing after fluoropyrimidine-based therapy, oral paclitaxel gave OS 9.13 vs 6.54 months (HR 0.77, P=0.006) and PFS 3.02 vs 2.89 months (HR 0.89), with ORR 13.6% vs 9.8%
NCDB analysis of 3,676 cT2N0 gastric cancers questions whether upfront surgery is adequate — In an NCDB analysis of 3,676 patients with cT2N0 gastric adenocarcinoma, 61.3% received no chemotherapy, 19.4% perioperative chemotherapy and 19.2% surgery followed by adjuvant chemotherapy

Full details below ↓

📄 What's Being Published

Manuscripts:

Immunotherapy and Targeted Therapy for Advanced Gastroesophageal Cancer: ASCO Living Guideline, Version 2026.1.2. — Shah Manish A et al., J Clin Oncol

Living guidelines are developed for selected topic areas with rapidly evolving evidence that drives frequent change in recommended clinical practice, updated on a regular schedule by a standing expert panel that systematically reviews the health literature on a continuous basis. The 2026.1.2 version addresses first-line management of advanced gastroesophageal cancer, incorporating biomarker-directed checkpoint inhibitor use.

Clinical bottom line: For first-line HER2-positive advanced gastroesophageal cancer, the updated ASCO Living Guideline ties the addition of a checkpoint inhibitor to PD-L1 status, so PD-L1, MSI/dMMR, CLDN18.2, and HER2 must all be tested up front.

Read full article →

Atezolizumab with or without tiragolumab in unresectable esophageal squamous cell carcinoma following definitive concurrent chemoradiotherapy (SKYSCRAPER-07): a randomised, phase III study. — Xu R-H et al., Ann Oncol

760 patients were randomised to atezolizumab plus tiragolumab (n=257), atezolizumab plus placebo (n=250), or placebo (n=253). At a median survival follow-up of 25.0 months, atezolizumab plus tiragolumab versus placebo did not statistically improve INV-PFS (HR 0.82, 95% CI 0.65–1.03; p=0.0947) or OS (HR 0.91, 95% CI 0.70–1.18), whereas atezolizumab plus placebo versus placebo significantly improved OS (HR 0.69, 95% CI 0.52–0.91) and INV-PFS (HR 0.74, 95% CI 0.58–0.93). Treatment-related adverse events occurred in 74.8%, 65.2%, and 55.4% of patients across the three arms respectively, with treatment-related deaths in 1.2%, 0.8%, and 1.6%.

Clinical bottom line: Adding tiragolumab to atezolizumab after definitive chemoradiotherapy in unresectable esophageal squamous cell carcinoma failed the primary endpoint and added toxicity, while atezolizumab alone is the arm that delivered the survival gain.

Read full article →

Liver resection after atezolizumab and bevacizumab versus maintenance therapy for locally advanced hepatocellular carcinoma (TALENTOP): a multicentre, open-label, randomised, phase 3 trial. — Sun Hui-Chuan et al., Lancet

489 patients entered the induction phase and 201 were randomised to surgery (n=101) or maintenance therapy (n=100). After a median follow-up of 18.4 months, median time to treatment failure was 20.4 months in the surgery group versus 11.8 months in the maintenance therapy group (HR 0.60, 95% CI 0.39–0.91; p=0.015). Grade 3 or 4 treatment-related adverse events occurred in 32 (39%) of 83 surgery patients versus 21 (21%) of 100 maintenance patients, and two treatment-related deaths occurred in the surgery group.

Clinical bottom line: After response or stabilisation on atezolizumab plus bevacizumab, resectability should be re-assessed by a multidisciplinary team in advanced HCC with macrovascular invasion, because resection nearly doubled time to treatment failure at the cost of substantially more grade 3–4 toxicity and two treatment-related deaths.

Read full article →

Ultrasensitive circulating tumor DNA analysis improves detection of molecular residual disease in locally advanced esophageal cancer. — Rhoades Justin et al., Clin Cancer Res

331 plasma samples from 36 patients were collected preoperatively and postoperatively. With the Oncodetect V2 test, 9 of 14 patients (64%) with recurrence tested ctDNA-positive at their first postoperative sample, and all 14 patients tested ctDNA-positive for at least one postoperative sample prior to recurrence. Notably, ctDNA was detected below the limit of detection of first-generation assays in 59% of all samples in which ctDNA was detected.

Clinical bottom line: Molecular residual disease testing in locally advanced esophageal cancer requires part-per-million sensitivity, because most detectable ctDNA in this cohort sat below the floor of first-generation tumor-informed assays.

Read full article →

🗂 Every paper considered this week, including the ones that didn’t make it →

🗣️ What People Were Discussing This Week

How these are chosen: filtered for clinical relevance, grouped by the new development discussed, then ordered by engagement (bookmarks and reposts).

Genome-wide in vivo CRISPR screens identify P2RY8 and GNAS as brakes on human T cells in solid tumors, including pancreatic and gastroesophageal models

Reported in Nature, these represent the first genome-wide in vivo screens for genetic modifications to enhance human T cell therapies for solid tumors. P2RY8 appeared to limit T cell entry into tumors while GNAS transmitted suppressive signals; in one melanoma model, all 6 mice treated with GNAS-edited CAR T cells achieved durable tumor clearance. The approach also improved tumor control across pancreatic, lung, uterine sarcoma, and gastroesophageal preclinical models. The central question for specialists is whether multiplexed edits such as P2RY8 plus GNAS knockout can translate these in vivo screen findings into cell therapy that works in pancreatic and gastroesophageal cancer.

Discussed by: @EladSharonMD, @PatrickHwuMD

→ View representative tweet

ESMO Express Update adds daraxonrasib to second-line metastatic pancreatic cancer

The ESMO Clinical Practice Guideline Express Update on pancreatic cancer now includes the pan-RAS(ON) inhibitor daraxonrasib for second-line treatment of metastatic pancreatic cancer. In RASG12-mutant, ECOG 0–1 patients after chemotherapy, daraxonrasib is recommended [I,A], with OS 13.2 vs 6.6 months (HR 0.40), PFS 7.3 vs 3.5 months (HR 0.45), and ORR 33.2% vs 11.8%. With a [I,A] second-line recommendation now in place for RAS-mutant disease, oncologists are debating whether RAS genotyping should move to the time of diagnosis rather than at progression.

Discussed by: @ArndtVogel, @OncoAlert, @EladSharonMD, @myESMO, @Dr_ElvinaA

→ View representative tweet

TALENTOP phase III: liver resection after atezolizumab plus bevacizumab extends time to treatment failure in advanced HCC with macrovascular invasion

In TALENTOP, 201 patients who responded to atezolizumab plus bevacizumab were randomised to resection followed by 12 months of atezolizumab/bevacizumab or to maintenance atezolizumab/bevacizumab. The primary endpoint—time to treatment failure—was 20.4 vs 11.8 months (HR 0.60, 95% CI 0.39–0.91; p=0.015) in favor of resection, with R0 resection achieved in 93% and pathological complete response in 28% of resected patients. However, grade 3–4 treatment-related adverse events were 39% vs 21% in the surgery versus maintenance arms. Clinicians framed the results as support for neoadjuvant and response-adapted treatment concepts in HCC, while the debate centers on whether multidisciplinary re-assessment of resectability should become routine after response, given the toxicity differential.

Discussed by: @ArndtVogel, @OncoAlert, @ronanhsieh

→ View representative tweet

Phase III Chinese trial: oral paclitaxel beats IV paclitaxel on overall survival in second-line advanced gastric cancer

In 536 patients with gastric cancer progressing after fluoropyrimidine-based therapy, oral paclitaxel produced OS of 9.13 vs 6.54 months (HR 0.77, P=0.006), PFS of 3.02 vs 2.89 months (HR 0.89, meeting noninferiority), and ORR of 13.6% vs 9.8%, with less neuropathy, alopecia, and hypersensitivity but more diarrhea and treatment interruptions compared with IV paclitaxel. A noted caveat is that the comparator was q3-weekly paclitaxel rather than the contemporary ramucirumab plus weekly paclitaxel standard in many regions, and the study was conducted entirely in Chinese patients. The open question is whether a survival edge over q3-weekly IV paclitaxel is sufficient to change second-line practice where ramucirumab plus weekly paclitaxel remains standard.

Discussed by: @OncoAlert, @OncoAlert

→ View representative tweet

NCDB analysis of 3,676 cT2N0 gastric cancers questions whether upfront surgery is adequate

In an NCDB analysis of 3,676 patients with cT2N0 gastric adenocarcinoma, 61.3% received no chemotherapy, 19.4% perioperative chemotherapy, and 19.2% surgery followed by adjuvant chemotherapy; 28.8% of patients undergoing upfront surgery were pathologically upstaged. Perioperative chemotherapy versus no chemotherapy yielded HR 0.64, adjuvant chemotherapy versus no chemotherapy HR 0.68, and adjuvant versus perioperative chemotherapy HR 0.79 (p=0.015), though the signal favoring surgery followed by adjuvant chemotherapy was characterized as hypothesis-generating rather than practice-changing given retrospective selection bias. A related meta-analysis of pathological response after neoadjuvant chemotherapy in resected gastroesophageal adenocarcinoma found partial responders benefited from adjuvant chemotherapy (OS HR 0.64; RFS HR 0.71) while complete and minimal responders did not. Given that nearly a third of cT2N0 gastric cancers are upstaged at surgery, clinicians are examining whether perioperative chemotherapy should become the default rather than the minority approach.

Discussed by: @OncoAlert, @OncoAlert

→ View representative tweet

Satellite lesions versus intrahepatic metastases in multifocal intrahepatic cholangiocarcinoma: genomic profiling shows both are clonally derived from the primary

Genomic profiling demonstrated that satellite lesions and intrahepatic metastases in multifocal intrahepatic cholangiocarcinoma are clonally derived from the primary tumor, representing metastatic progression associated with poor long-term survival. In a parallel prospective observational cohort of advanced biliary tract cancer treated with nab-paclitaxel plus gemcitabine-cisplatin, activity was considered interesting but predictive biomarkers remain unidentified. If satellite lesions represent clonal metastases rather than local extension, the question for surgeons and oncologists is whether resection criteria for multifocal intrahepatic cholangiocarcinoma should be tightened accordingly.

Discussed by: @ArndtVogel, @ArndtVogel

→ View representative tweet

NRG GI-011 Amendment 2.0 loosens eligibility for locally advanced pancreatic cancer

Amendment 2.0 of the NRG GI-011 (LAP-100) trial in locally advanced pancreatic cancer liberalizes post-induction chemotherapy CA19-9 criteria, allows tumor treating fields, and permits up to 8 months of induction chemotherapy, with the stated aim of facilitating accrual. The update was described as more flexible and pragmatic. The open question is whether permitting up to 8 months of induction chemotherapy and tumor treating fields will improve accrual without diluting the trial question.

Discussed by: @NiuSanford, @NiuSanford

→ View representative tweet

Consensus statement on incretin mimetics in neuroendocrine neoplasms published in Cancer

A consensus statement published in Cancer provides guidance on incretin mimetics in neuroendocrine neoplasms, balancing cardiometabolic benefit against uncertain oncologic risk. The statement both provides expert advice for managing this challenge in clinic and highlights the need for robust prospective evidence to guide future use of these agents. Clinicians face the practical question of how to weigh cardiometabolic gains against uncertain oncologic risk when prescribing GLP-1-based agents to patients with neuroendocrine neoplasms.

Discussed by: @OncBrothers, @OncoAlert

→ View representative tweet

PANTHEIA-SEOM: systemic inflammation response index stratifies prognosis and chemotherapy benefit in 672 metastatic pancreatic cancers

In 672 patients with metastatic pancreatic adenocarcinoma across 30 centers, a systemic inflammation response index (SIRI) above 2.3 was associated with median OS of 9.2 vs 14.0 months, median PFS of 3.9 vs 5.9 months, and ORR of 25% vs 37%, independently identifying a poorer-prognosis, lower-response phenotype with greater liver tumor burden and cachexia. The central debate is whether a routine blood-derived inflammation index such as SIRI is ready to stratify first-line chemotherapy decisions in metastatic pancreatic cancer.

Discussed by: @OncoAlert, @OncoAlert, @OncoAlert

→ View representative tweet

🔥 Top Ranked Tweets

OncoAlert @OncoAlert

RT @JournalCancer: 📗🔎 Our latest #CancerScope covers the landmark RASolute 302 study presented at #ASCO26. Compared with chemotherapy, treatment with daraxonrasib more than doubled OS and PFS for patients with previously treated metastatic pancreatic cancer.

https://t.co/Jc3fvXCoUv

@OncoAlert https://t.co/8Po6sTfeq2

🔖 1 ❤️ 4 🔁 4 👁 859