GI Oncology Weekly — Issue 11

This week’s calls — 5 readouts · 1 to act on · ~11 min read. Tap any row to jump.

ASCOT Final 5-year analysis confirms adjuvant S-1 after resection for biliary tract cancer, 64% vs 52% overall survival Practice-shaping Matched targeted therapy after chemo-IO failure Matched targeted therapy after first-line chemoimmunotherapy failure doubled survival — but only 32% of eligible patients received it Worth watching RELIGHT A PD-L1/TGF-β bispecific nearly halves death risk in first-line HER2-negative gastric cancer — against chemotherapy alone Worth watching HERBOT Trastuzumab plus nivolumab added to gemcitabine–cisplatin meets its primary endpoint in first-line HER2-positive biliary tract cancer Worth watching ASAC Adjuvant aspirin fails after curative-intent treatment of colorectal liver metastases, with a safety signal On the radar

This week · 5 readouts

The Verdicts

Every trial that matters this week, distilled to a single call — act on it, watch it, or keep it on the radar.

Practice-shaping Biliary tract · adjuvant therapy

ASCOT

The adjuvant question in biliary tract cancer has had thin and conflicting evidence for a decade, and this is the most durable answer yet: a 12-percentage-point absolute gain in 5-year overall survival that has held through the prespecified final analysis at a median 61.5 months. Relapse-free survival is the softer signal — the unstratified hazard ratio crosses unity — so the honest reading is that the survival benefit is real and the disease-control mechanism behind it is less clean than the OS curve suggests. The open question is generalisability: S-1 pharmacokinetics differ outside Japan, and this was a Japanese trial.

Go deeper

The prespecified final 5-year analysis of JCOG's ASCOT trial confirms adjuvant S-1 as a substantive intervention after curative resection for biliary tract cancer. Among 440 patients randomised (S-1 218, observation 222) with a median follow-up of 61.5 months, 5-year overall survival was 64% with S-1 versus 52% with observation — HR 0.72 (95% CI 0.55–0.95), one-sided p=.01 — corresponding to a median OS of 8.2 versus 5.9 years. Relapse-free survival favoured S-1 less decisively: the stratified hazard ratio was 0.76 (95% CI 0.59–0.98), while the unstratified estimate of 0.80 (95% CI 0.62–1.03) crosses unity. Treatment effects were generally consistent across prespecified subgroups. A significant 12-point absolute OS gain at five years is a large effect in a disease whose adjuvant evidence base has been thin; the open question is how well it generalises outside Japan, where S-1 pharmacokinetics differ. Full daily entry →

Survival over time · schematic — OS*

Adjuvant S-1Observation

· 5-yr OS 64% vs 52% (HR 0.72, 95% CI 0.55–0.95, one-sided p=.01)· median OS 8.2 vs 5.9 years· RFS stratified HR 0.76 (0.59–0.98), unstratified 0.80 (0.62–1.03)· n=440 (S-1 218, observation 222)· median follow-up 61.5 mo

HR 0.72 ↓ 28% risk of death 95% CI 0.55–0.95 · Δ +27.6 mo median 5-yr OS 64% vs 52%

Journal of Clinical Oncology · randomised phase 3, prespecified final 5-year analysis

Worth watching Biliary tract · precision oncology access

Matched targeted therapy after chemo-IO failure

This is a retrospective study, so the hazard ratio is not the point and should not be read as one. The point is the denominator. Across 1,358 patients treated with first-line cisplatin, gemcitabine and durvalumab, molecular testing found an ESCAT tier-I actionable alteration in 238 — and fewer than a third of those patients ever received a matched agent. Where matched therapy was given, survival was substantially longer, and the effect was concentrated in second line with nothing demonstrable by third. Both halves of that sentence argue for the same change in practice: profile before first-line progression, not after it.

· Post-progression matched vs non-targeted: mOS 24.8 vs 16.6 mo (HR 0.36, 95% CI 0.20–0.63, p=0.0004)· adjusted HR 0.35 (0.22–0.56), p<0.0001· second line (n=127): OS HR 0.50 (0.30–0.84), PFS 5.9 vs 3.2 mo, ORR 27.1% vs 9.1%· no significant benefit in third line· ESCAT tier-I in 238/1,072 tested (22.2%), only 76/238 (32.0%) treated· n=1,358 across 55 centres in 12 countries

HR 0.36 ↓ 64% risk of death 95% CI 0.20–0.63 · Δ +8.2 mo median

Survival over time · schematic — OS*

Matched targetedNon-targeted

Go deeper

This retrospective study pooled 1,358 patients with advanced biliary tract cancer treated with first-line cisplatin, gemcitabine and durvalumab across 55 centres in 12 countries. Molecular testing was performed in 1,072 patients and identified ESCAT tier-I actionable alterations in 238 (22.2%) — but only 76 of those 238 (32.0%) ever received a matched targeted agent. Among the 125 patients who received any active treatment after first-line progression, matched targeted therapy was associated with median OS of 24.8 vs 16.6 months (HR 0.36, 95% CI 0.20–0.63, p=0.0004), with an adjusted HR of 0.35 (95% CI 0.22–0.56, p<0.0001). The benefit was concentrated in second line (n=127: OS HR 0.50, 95% CI 0.30–0.84, p=0.0009; PFS 5.9 vs 3.2 months, HR 0.60, 95% CI 0.38–0.95, p=0.028; ORR 27.1% vs 9.1%, p=0.010), with no significant advantage demonstrated in third line. The design is retrospective and the authors themselves frame the results as practice-informing and hypothesis-generating rather than definitive. Full daily entry →

Journal of Hepatology · retrospective, 55 centres / 12 countries, Cox + IPTW + landmark analyses

Worth watching Gastric / GEJ · first-line immunotherapy

RELIGHT

A stratified hazard ratio of 0.53 for overall survival in first-line gastric cancer is the kind of number that would normally settle a question. This one does not, and the authors say so themselves: the protocol closed before PD-1 inhibitor plus chemotherapy became approved standard of care in China, so the comparator is chemotherapy alone — a regimen no patient in this position would now be offered. What the trial establishes is that dual PD-L1/TGF-β blockade is active, at essentially unchanged grade ≥3 toxicity. What it does not establish is superiority over the regimen it will actually have to beat. That head-to-head is the trial that matters next.

· PD-L1 CPS≥5: mOS 16.8 vs 10.4 mo (stratified HR 0.53, 95% CI 0.40–0.68, one-sided p<0.0001)· ITT: mOS 15.8 vs 11.2 mo (HR 0.66, 0.53–0.81)· PFS 7.6 vs 5.5 mo in CPS≥5 (HR 0.52, 0.42–0.66)· grade ≥3 TRAEs 62.6% vs 59.0%· n=731

HR 0.53 ↓ 47% risk of death 95% CI 0.40–0.68 · Δ +6.4 mo median

Survival over time · schematic — OS*

Retlirafusp alfa + CAPOXPlacebo + CAPOX

Go deeper

RELIGHT randomized 731 previously untreated patients with unresectable, locally advanced or metastatic HER2-negative gastric or gastroesophageal junction adenocarcinoma 1:1 to retlirafusp alfa — an anti-PD-L1/TGF-β bispecific antibody — or placebo every 3 weeks, on top of capecitabine and oxaliplatin. In the PD-L1 CPS≥5 population median OS was 16.8 vs 10.4 months (stratified HR 0.53, 95% CI 0.40–0.68, one-sided p<0.0001), and in the intention-to-treat set 15.8 vs 11.2 months (stratified HR 0.66, 95% CI 0.53–0.81, one-sided p<0.0001). PFS moved the same way: 7.6 vs 5.5 months in CPS≥5 (HR 0.52, 95% CI 0.42–0.66) and 7.0 vs 5.5 months in ITT (HR 0.57, 95% CI 0.48–0.69). Grade ≥3 treatment-related adverse events were essentially flat at 62.6% (228/364) vs 59.0% (216/366). The asterisk is the control arm, and the authors say so themselves: the protocol was finalized before PD-1 inhibitor plus chemotherapy became approved standard of care in China, so this is a bispecific beating chemotherapy alone — not beating the regimen a patient would actually be offered today. Full daily entry →

Journal of Clinical Oncology · randomised, double-blind phase 3

Worth watching Biliary tract · HER2-directed therapy

HERBOT

A 55% response rate and 10.6-month median PFS in first-line HER2-positive biliary tract cancer is a genuinely encouraging result, with the usual phase 1b/2 caveats — 40 patients, single arm, no comparator. Toxicity is not trivial: grade ≥3 neutropenia in well over half. The finding most likely to outlive the trial is the preplanned AI whole-slide analysis showing that a higher proportion of HER2 3+ tumour cells tracked with greater benefit. Quantifying HER2 heterogeneity rather than scoring it as a yes/no may be what actually selects these patients — and that idea is testable in every HER2-directed programme, not just this one.

· ORR 55% (95% CI 38.5–70.7), 1 complete + 21 partial responses· DCR 95% (83.5–99.4)· median DoR 12.6 mo (5.7–NR)· median PFS 10.6 mo (7.8–17.4)· median OS not reached at 17.0 mo follow-up· grade ≥3 neutropenia 57.5%, anaemia 30.0%, thrombocytopenia 22.5%· n=40, single arm

Go deeper

HERBOT, a multi-institutional phase 1b/2 trial from the Korean Cancer Study Group, added trastuzumab and nivolumab to gemcitabine–cisplatin as first-line therapy in 40 patients with HER2-positive unresectable biliary tract cancer. The trial met its primary endpoint with an objective response rate of 55% (95% CI 38.5–70.7), comprising one complete and 21 partial responses, and a disease control rate of 95% (95% CI 83.5–99.4). Median duration of response was 12.6 months (95% CI 5.7–NR) and median progression-free survival 10.6 months (95% CI 7.8–17.4); at a median follow-up of 17.0 months median overall survival was not reached, and two patients (5.0%) proceeded to curative-intent conversion surgery. Toxicity was not trivial — grade ≥3 neutropenia in 57.5%, anaemia in 30.0% and thrombocytopenia in 22.5%. A preplanned AI whole-slide analysis found that a higher proportion of HER2 3+ tumour cells was associated with greater benefit, which is arguably the more generalisable finding: quantifying HER2 heterogeneity, rather than scoring it dichotomously, may be what selects these patients properly. Full daily entry →

Nature Medicine · phase 1b/2, single-arm, Korean Cancer Study Group

On the radar Colorectal · adjuvant aspirin

ASAC

An instructive negative, and the most useful trial of the week for what it rules out. Aspirin 160 mg daily after curative-intent treatment of colorectal liver metastases did not improve disease-free survival, and the secondary overall survival result ran the wrong way alongside a real excess of serious adverse events. Because the trial missed its primary endpoint, that OS signal should be read as a safety flag rather than proof of harm — but it is not a result to wave through. Read against ALASCCA, which halved recurrence at the identical dose, the discriminating variable is selection: ALASCCA enrolled PI3K-pathway-altered localized disease, ASAC an unselected post-metastasectomy population. Aspirin is not a general-purpose adjuvant.

· Primary endpoint not met — DFS HR 1.08 (95.44% CI 0.85–1.38, p=0.58) across 269 events· median DFS 1.13 vs 1.40 years· 36-mo OS 76% vs 85%, HR for death 1.64 (95% CI 1.01–2.65, p=0.045), a nominally significant secondary endpoint in a trial that missed its primary· serious adverse events 17/217 (8%) vs 4/211 (2%)· n=466 randomised, 428 in the full analysis set

Go deeper

ASAC randomised 466 patients (428 in the full analysis set: aspirin 217, placebo 211) to aspirin 160 mg daily or placebo for up to three years after curative-intent treatment of colorectal cancer liver metastases, across 14 centres in Norway, Sweden and Denmark. The primary endpoint was not met: disease-free survival HR was 1.08 (95.44% CI 0.85–1.38, p=0.58) across 269 events, with median DFS of 1.13 years on aspirin versus 1.40 years on placebo. More concerning, 36-month overall survival was 76% with aspirin versus 85% with placebo, HR for death 1.64 (95% CI 1.01–2.65, p=0.045), and serious adverse events occurred in 17/217 (8%) on aspirin versus 4/211 (2%) on placebo, including excess myocardial infarction, epistaxis, duodenal ulcer and cerebral haemorrhage. The OS result is a nominally significant secondary endpoint in a trial that missed its primary, so it should be read as a safety signal rather than proof of harm. Read against ALASCCA — which this digest led with on April 21 and which halved recurrence with the same 160 mg daily dose — the discriminating variable is patient selection: ALASCCA enrolled PI3K-pathway-altered localized disease, ASAC an unselected population after liver-metastasis resection. Full daily entry →

Lancet Gastroenterology & Hepatology · randomised, placebo-controlled phase 3

* Survival curves are illustrative — modeled from each trial’s reported median and hazard ratio (assuming exponential survival), not the published Kaplan–Meier curves.

The take

Four of this week's stories are biliary tract cancer, which almost never happens, and read together they describe the whole arc of a field maturing — plus the thing that arc keeps leaving behind.

ASCOT gives the disease a durable adjuvant standard on mature five-year data. Lirafugratinib becomes the third approved FGFR2 inhibitor in cholangiocarcinoma. HERBOT adds a credible HER2-directed first-line signal. On any ordinary reading, that is a very good week for a cancer that spent a decade with almost nothing.

Then the fourth paper reports what happens to those advances in practice. Across 1,358 patients on modern first-line chemoimmunotherapy, molecular testing found an ESCAT tier-I actionable alteration in 238 — and 76 of them received a matched drug. Not 76 percent. Seventy-six patients. Fewer than one in three.

The two facts sit badly together, and the pairing is the story:

The bottleneck in biliary cancer is no longer only the absence of active drugs. For a meaningful minority of patients it is the interval between progression and death, which is too short to absorb a molecular profiling turnaround that was started too late.
The benefit in that study was concentrated in second line and undetectable by third, which means the cost of profiling late is not a delay — it is the whole effect.
And a third approved FGFR2 inhibitor only helps the patients whose FGFR2 fusion was actually found.

None of this argues against the week's three positive trials. It argues that the operational question — who gets tested, when, and whether the result arrives while it can still change something — now carries as much survival as the next agent does. Profile at diagnosis, not at progression.

Regulatory · what to watch

Recent Regulatory Approvals

Regulatory milestones from the week, with the practice-relevant detail.

FDA APPROVAL

Lirafugratinib (LYRFIGTU) for previously treated FGFR2 fusion or rearrangement-positive cholangiocarcinoma

Approved September 23, 2026 for adults with previously treated unresectable, locally advanced or metastatic cholangiocarcinoma harbouring an FGFR2 gene fusion or other rearrangement, four days ahead of its PDUFA date. Approval rests on the REFOCUS efficacy population of 116 FGFR-inhibitor-naïve patients: confirmed ORR 46% (95% CI 36–55), median duration of response 11.8 months (95% CI 7.5–13.0), dosed at 70 mg orally once daily. This is not a first-in-class event — pemigatinib and futibatinib are already approved here and infigratinib was withdrawn — so the practical question has shifted from whether to inhibit FGFR2 to how to sequence the agents that do, and what to offer once acquired FGFR2 resistance emerges.

GI Oncology Weekly · curated by Dr. Allan Pereira, Moffitt Cancer Center. Educational summaries for practising oncologists — not medical advice; verify against primary sources before acting.

All weekly issues