Top 5 Papers
#1
Source: European Medicines Agency (CHMP), via Revolution Medicines press release | Authors: EMA Committee for Medicinal Products for Human Use; applicant Revolution Medicines | Published: October 15, 2026
Score: 13/20 — Regulatory action (base 8) + Phase III RCT evidence base, RASolute 302 (+3) + overall survival benefit (+2) = 13. Scored as a marketing-authorization recommendation, like the 08-27 FDA approval, not as the procedural phased-review milestone of 07-09, which deliberately did not re-count the trial. The impact bonus is +2 rather than the +3 used on 08-27 because the ESMO Express Update had already named daraxonrasib a standard in August; this opinion does not create the standard, it carries it into the EU. No colleague-engagement bonus: one GI oncology KOL post was reported by the X sweep but could not be confirmed at source.
On October 15 the EMA's Committee for Medicinal Products for Human Use adopted a positive opinion recommending marketing authorization of daraxonrasib (RASONQUE), an oral RAS(ON) multi-selective inhibitor, as monotherapy for adults with metastatic pancreatic adenocarcinoma who have received at least one prior systemic therapy. The opinion rests on the Phase 3 RASolute 302 trial (NEJM), which met all primary and key secondary endpoints in both the RAS G12-mutant and the overall intent-to-treat populations. In the ITT population, daraxonrasib reduced the risk of death by 60% versus investigator's-choice chemotherapy (HR 0.40, 95% CI 0.30–0.53; p<0.0001), with median OS of 13.2 vs 6.7 months and median PFS of 7.2 vs 3.6 months (HR 0.49). A later ad hoc OS analysis gave 12.4 vs 6.7 months (HR 0.49), which is why secondary coverage quotes two different medians. The most common adverse reactions were diarrhoea (63.1%), acneiform dermatitis (62.4%) and stomatitis (55.5%). The review ran under the EMA's Cancer Medicines Pathfinder project. The European Commission now has to adopt the decision; the US approval came on August 26.
Post angle: Seven weeks after the FDA, the EU committee has recommended daraxonrasib for second-line metastatic PDAC. Europe is on its way to the same RAS-targeted option for pretreated patients, and the 12.4 vs 13.2 month confusion is just the ad hoc vs primary analysis. #GIOnc #PDAC #PancreaticCancer #RAS #EMA
#2
Source: JCO Precision Oncology | Authors: Diamond-Rossi SA, Rothe M, Garrett-Mayer E, et al.; Schilsky RL (ASCO TAPUR Study) | Published: October 16, 2026
Score: 9/20 — JCO Precision Oncology (base 6, consistent with 08-09 and 10-03) + Phase II basket trial (+2) + biomarker-guided BRAF V600E selection (+1) = 9. No survival-benefit bonus: single-arm, with disease control as the primary endpoint. Biliary tract cancer is one of the two cohorts reported; the thyroid cohort is not GI and is mentioned only for context.
TAPUR is ASCO's phase II basket trial of commercially available targeted drugs matched to genomic alterations. This report covers two cohorts treated with cobimetinib plus vemurafenib for BRAF V600E-mutant cancers: thyroid (N=22, 16 evaluable) and biliary tract cancer (N=18, 15 evaluable). Patients had measurable disease and no standard treatment options left. In the BTC cohort, the disease-control rate (objective response or stable disease lasting at least 16 weeks) was 67% (one-sided 90% CI 47–100) and the objective response rate was 53% (95% CI 27–79). That rejects the 15% null disease-control rate (p<0.0001). Toxicity was substantial: 53% of all patients had at least one treatment-related grade 3–4 or serious adverse event. Both cohorts closed before reaching their target size at the manufacturer's request, so the inference rests on the protocol's prespecified thresholds.
Post angle: BRAF V600E is a small slice of biliary cancer, and ASCO's new BTC living guideline lists it among the actionable second-line targets. TAPUR shows a second BRAF/MEK pair, cobimetinib + vemurafenib, is active there too: ORR 53% in 15 evaluable patients. Dabrafenib + trametinib already holds a tumour-agnostic US approval for BRAF V600E, so the value here is another option, not a new indication. #GIOnc #BTC #Cholangiocarcinoma #BRAF #PrecisionOncology
#3
Source: Clinical Cancer Research | Authors: Cheng K, Zhang C, Liu J, et al.; Cao D (West China Hospital, multicentre) | Published: October 16, 2026
Score: 8/20 — Clinical Cancer Research (base 6) + Phase II (+2) = 8. No survival-benefit bonus: single-arm with no internal comparator (consistent with the 06-11 and 10-03 treatment of single-arm CCR trials). No biomarker bonus: ctDNA and tumour-microenvironment findings are exploratory correlates, not a selection criterion.
This multicentre, single-arm phase II trial (ChiCTR2300073237) gave 46 patients with previously untreated metastatic pancreatic ductal adenocarcinoma the anti-PD-1 antibody serplulimab plus gemcitabine and nab-paclitaxel in 21-day cycles, with stereotactic body radiotherapy to the primary or a metastatic site during cycle 2. The primary endpoint, 6-month PFS, reached 78.3% (95% CI 67.2–91.1%) against a prespecified target of 60%. Median PFS was 8.5 months (95% CI 7.3–12.0), median OS 14.0 months (95% CI 12.0–not reached), and the objective response rate 73.9% (95% CI 58.9–85.7%). There were no treatment-related deaths and no unexpected safety signals. Patients who were ctDNA-negative at baseline survived longer, and baseline FAP-positive cancer-associated fibroblast and T-cell proportions in tumour tissue tracked with outcome.
Post angle: An ORR of 73.9% in first-line metastatic PDAC is a striking number for any regimen. For context, gem/nab-paclitaxel alone gave a median OS of 8.5 months in MPACT and NALIRIFOX 11.1 months in NAPOLI-3. ICSBR reports 14.0, but it is 46 patients, one country and no control arm. Worth a randomised trial, not a practice change. #GIOnc #PDAC #PancreaticCancer #Immunotherapy #SBRT
#4
Source: Journal of Viral Hepatitis | Authors: Kudaravalli S, Yan T, Tseng CH, et al.; Nguyen MH (international collaboration) | Published: October 14, 2026
Score: 7/20 — Journal of Viral Hepatitis (not in the table; Other = 5, no prior edition has scored it) + meta-analysis (+2) = 7. No survival-benefit bonus: the OS difference comes from pooled observational studies, so it is an association that confounding by indication could explain, not a treatment effect.
Nucleos(t)ide antivirals are recommended for viraemic HBV-related cirrhosis, but in US guidelines HCC itself is not a criterion for starting antiviral treatment. This PRISMA systematic review pooled 78 studies (68,243 patients with HBV-related HCC, published 2006–2024) using random-effects models. The pooled antiviral treatment rate was 53.1% (52 studies, 41,345 patients). Only 17.2% had started treatment before their cancer was diagnosed. Treated patients had higher 3-year OS (80.4% vs 59.2%) and 5-year OS (72.1% vs 46.8%, both p<0.01) than untreated patients. The difference held in inactive HCC (78.0% vs 56.4%) but not significantly in active HCC (48.2% vs 22.6%, p=0.14). Antiviral therapy stayed independently associated with better 5-year OS in multivariable meta-regression, but HCC recurrence rates did not differ.
Post angle: Half of patients with HBV-related HCC never get an antiviral, and fewer than 1 in 5 were on one before the cancer appeared. The survival gap in these observational data is large (5-yr OS 72% vs 47%). It is not causal proof, but a cheap oral drug with this signal should be the default at an HCC diagnosis. #GIOnc #HCC #HepatitisB #LiverCancer
#5
Source: Clinical Colorectal Cancer | Authors: Valagni G, Pereira RA, Lenz G, Barcellos G, de Castria TB | Published: September 25, 2026 (online; indexed in PubMed October 2026)
Score: 7/20 — Clinical Colorectal Cancer (base 5, consistent with 10-13) + meta-analysis (+2) = 7. No survival-benefit bonus under scoring rule 4: the pooled HR loses significance when its only randomised trial is excluded, so the new pooled estimate adds no independent evidence of benefit beyond TransMet itself.
This PRISMA systematic review (PROSPERO CRD42025124542) identified 17 studies of liver transplantation for permanently unresectable, liver-confined colorectal metastases, three of them comparative and pooled for OS. The pooled estimate favoured transplantation over the best alternative therapy (HR 0.45, 95% CI 0.28–0.72; I²=3%). The authors say plainly that it is not robust: the TransMet randomised trial carries 63% of the weight, and without it the effect is no longer significant (HR 0.64, 95% CI 0.30–1.35). Comparators also varied, from chemotherapy alone to ALPPS. In narrative synthesis, 5-year OS reached 83% with strict biological selection. Recurrence after transplant was common (72–81.5%) but often treatable with curative intent. Conference-abstract referral data suggest only about 10% of patients evaluated actually reach transplant.
Post angle: Transplant for unresectable colorectal liver metastases is real but narrow. The meta-analysis gives HR 0.45, then honestly shows it is mostly TransMet: drop that trial and significance goes. Most transplanted patients recur, often treatably, and about 1 in 10 evaluated patients gets there. Select hard and refer early. #GIOnc #CRC #LiverTransplant #ColorectalCancer
Additional Papers of Interest
-
International Journal of Epidemiology — individual-level data on about 2 million adults and 8,561 gastric adenocarcinomas. Processed meat showed only weak evidence of association overall (HR 1.15, 95% CI 1.00–1.32, for ≥45 vs <3 g/day), confined to current smokers and absent in Asian cohorts; egg intake was positively associated; no association for unprocessed red meat, poultry or seafood.
-
Cell Host & Microbe — in MSS CRC mouse models, oral Akkermansia blunts macrophage efferocytosis via TLR2/NF-κB, triggering cGAS-STING and type I interferon and sensitising tumours to anti-PD-1. A phase 1 trial (NCT06865521) of the probiotic plus a PD-1 antibody in pretreated metastatic MSS CRC was safe, with an objective response rate of 19%.
-
Endoscopy — in the Swedish SCREESCO screening trial (9,982 colonoscopy-arm and 5,875 FIT-arm colonoscopies), 5-year post-colonoscopy CRC incidence was 0.16% with low vs 0.06% with high proximal serrated polyp detection rate (<10% vs ≥10%). Even among high-ADR endoscopists, each percentage-point rise in PSPDR lowered PCCRC risk (HR 0.90, 95% CI 0.82–1.00). Event numbers were small.
-
PLOS Medicine — among 461,682 US adults followed to 2018 (10,075 colorectal adenocarcinomas), total ultra-processed food intake was not associated with CRC (HR 0.97, 95% CI 0.91–1.03, top vs bottom quintile). Whole grain, fibre, dairy and calcium from UPF sources were inversely associated, and processed meat from UPF sources positively associated.
-
Diseases of the Colon and Rectum — in 5,032 resected stage I–III colon cancers at Asan Medical Center (656 early-onset matched 1:3), early onset was not independently prognostic (adjusted HR 0.85, 95% CI 0.66–1.10), yet younger patients received adjuvant chemotherapy more often, including low-risk stage II (37.7% vs 14.3%, p<0.001).
-
European Journal of Surgical Oncology — 376 patients with stage IV gastric cancer on chemotherapy across 36 Japanese institutions in 2022: resection rates 12.7% (age ≥75) vs 20.4% (<75), with no surgical mortality and 2-year OS of 87.5% vs 82.2% among those resected. Postoperative pneumonia was much more common in the elderly (25.0% vs 2.0%). Retrospective and highly selected.
Back to all digests