GI Oncology Weekly — Issue 6

This week’s calls — 5 readouts · 3 to act on · ~11 min read. Tap any row to jump.

TALENTOP Resection after response to atezolizumab–bevacizumab nearly doubles time to treatment failure, at a real toxicity cost Practice-shaping STAR-TREC Long-course chemoradiotherapy leaves far more patients TME-free at 12 months than short-course radiotherapy Practice-shaping ESMO Express Update — daraxonrasib ESMO writes daraxonrasib into metastatic pancreatic cancer through its out-of-cycle update mechanism Practice-shaping Paclitaxel oral solution An oral paclitaxel beats the injection on overall survival with no PFS difference — a tolerability story Worth watching SKYSCRAPER-07 Adding TIGIT blockade bought nothing; the monotherapy arm underneath the failed primary looks genuinely active 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 Liver · advanced HCC with macrovascular invasion

TALENTOP

The first randomised answer to a question that has been argued on case series for a decade: when a patient with macrovascular invasion responds to first-line atezolizumab–bevacizumab, should you operate? TALENTOP says yes, and the effect size is substantial — time to treatment failure 20.4 versus 11.8 months. Two things belong in the same breath as that number. The endpoint is time to treatment failure, not survival. And the surgery arm carried nearly double the grade 3–4 toxicity plus two treatment-related deaths. This changes the conversation you have with a good responder; it does not make the operation automatic.

Go deeper

489 treatment-naive patients with macrovascular invasion and no extrahepatic spread received three cycles of atezolizumab plus bevacizumab; the 201 who achieved a response or stable disease and were judged resectable were randomised 1:1 to liver resection or continued maintenance. Time to treatment failure was 20.4 versus 11.8 months (HR 0.60, 95% CI 0.39–0.91; p=0.015) at a median follow-up of 18.4 months. The cost is real and belongs in the same sentence as the benefit: grade 3–4 treatment-related adverse events ran 39% in the surgery arm versus 21% with maintenance, with two treatment-related deaths after resection (abnormal liver function and liver failure). Two further caveats before this changes anyone's practice — the endpoint is time to treatment failure rather than survival, and all 24 sites are in China, so generalisability to populations with predominantly non-HBV aetiology is untested. A companion Lancet editorial in the same issue asks whether this opens a new era for conversion surgery; the honest answer today is that it opens the question. Full daily entry →

Survival over time · schematic — TTF*

Liver resectionMaintenance therapy

· Time to treatment failure 20.4 vs 11.8 mo (HR 0.60, 95% CI 0.39–0.91, p=0.015)· median follow-up 18.4 mo· 489 enrolled in induction, 201 responders randomised 1:1· Grade 3–4 treatment-related AEs 39% (surgery) vs 21% (maintenance)· Two treatment-related deaths after resection

HR 0.6 ↓ 40% risk of an event 95% CI 0.39–0.91 · Δ +8.6 mo median

The Lancet — Sun H-C, … Fan J (senior), 24 hospitals, China; Zhongshan Hospital, Fudan University

Practice-shaping Rectal · response-adapted organ preservation

STAR-TREC

If organ preservation is the goal, the schedule you start on is the decision that matters, and STAR-TREC is now the best randomised evidence for making it. Among patients pursuing organ preservation, long-course chemoradiotherapy left 78.5% still free of total mesorectal excision at 12 months versus 60.6% with short-course radiotherapy — an 18-point gap in how many people keep their rectum. Serious gastrointestinal toxicity stayed low in both radiotherapy arms and was highest with upfront surgery. The honest limit: the prespecified primary endpoint sits at 30 months, so this is the schedule to build a watch-and-wait pathway on today, not a closed question.

· 12-mo TME-free survival 78.5% (long-course chemoradiotherapy) vs 60.6% (short-course radiotherapy)· hazard of requiring TME 1.90 (95% CI 1.29–2.81) with short-course· Grade 3–4 GI serious adverse events 2% LCCRT vs 4% SCRT vs 8% primary TME· n=409 modified ITT across 37 sites in 5 European countries· Prespecified primary endpoint at 30 months

12-month TME-free survival among patients pursuing organ preservation · 0–100% scale

78.5%Long-course chemoradiotherapy

60.6%Short-course radiotherapy

050100%

Go deeper

STAR-TREC randomised 409 patients (modified ITT: 163 long-course chemoradiotherapy, 168 short-course radiotherapy, 78 primary TME) across 37 sites in five European countries to test which radiotherapy schedule better supports response-adapted organ preservation in early- and intermediate-stage rectal cancer. At 12 months, TME-free survival among patients pursuing organ preservation was 78.5% with long-course chemoradiotherapy versus 60.6% with short-course radiotherapy (hazard ratio for requiring TME 1.90, 95% CI 1.29–2.81) — a substantial separation in the proportion of patients still keeping their rectum. Grade 3–4 gastrointestinal serious adverse events were low and similar in both radiotherapy arms (2% LCCRT, 4% SCRT) and highest with upfront surgery (8% primary TME). Importantly, the prespecified primary endpoint sits at 30 months, so this is a strong interim signal about which schedule to build a watch-and-wait pathway on, not yet the final verdict. Full daily entry →

Lancet Oncology — Bach SP, de Wilt JHW (senior) and the STAR-TREC Collaborative, 37 sites across 5 European countries

Practice-shaping Pancreas · RAS-mutant metastatic adenocarcinoma

ESMO Express Update — daraxonrasib

A guideline moving at the speed of the data it responds to. ESMO used its Express Update mechanism — reserved for evidence that should not wait for a full guideline revision — to write daraxonrasib into metastatic pancreatic cancer, months rather than years after the RASolute 302 plenary. The underlying trial is the reason: median OS 13.2 versus 6.7 months against investigator's-choice chemotherapy, HR 0.40. One caveat about this citation specifically — PubMed carries no abstract for the Express Update and the full text was not retrievable, so the exact recommendation wording and eligible population are not reproduced here.

· Evidence base RASolute 302: median OS 13.2 vs 6.7 mo, HR 0.40, p<0.0001 (NEJM; ASCO 2026 plenary LBA5)· First Phase 3 randomised trial of a RAS(ON) multi-selective inhibitor in any tumour· Indexed by PubMed as a Practice Guideline (PMID 42617733)· Authors: T. Conroy and M. Ducreux

Go deeper

ESMO has issued an Express Update — its out-of-cycle mechanism for data that should not wait for the next full guideline revision — writing daraxonrasib into the treatment of metastatic pancreatic cancer. The update is authored by Thierry Conroy and Michel Ducreux, and is indexed by PubMed as a Practice Guideline with the keywords KRAS inhibitors, daraxonrasib, metastatic pancreatic cancer, molecular profiling, and systemic treatment. The evidence base is RASolute 302, the first Phase 3 randomised trial of a RAS(ON) multi-selective inhibitor in any tumour, which reported median OS 13.2 versus 6.7 months against investigator's-choice chemotherapy, HR 0.40, p<0.0001. Note that PubMed carries no abstract for this Express Update and the full text was not retrievable, so the precise wording and strength of the recommendation, and the exact eligible population, are not reproduced here; the citation, journal, authorship and subject are verified from the PubMed record. Full daily entry →

Annals of Oncology — ESMO Clinical Practice Guideline (Express Update); T. Conroy (Institut de Cancérologie de Lorraine) and M. Ducreux (Gustave Roussy)

Worth watching Gastric · second line, advanced disease

Paclitaxel oral solution

A reformulation trial that turned into a survival trial, which is not how these usually go. Oral paclitaxel matched injection on progression-free survival and beat it on overall survival — 9.13 versus 6.54 months. That shape deserves scrutiny rather than a headline: a 2.6-month OS gain with no PFS difference is not what greater tumour control looks like. The most plausible explanation is tolerability and time on treatment, and the safety data point the same way. Two limits belong beside the result: every site is in China, and the sponsor has three co-authors on the paper.

· Overall survival 9.13 vs 6.54 mo (HR 0.770, 95.5% CI 0.635–0.934, p=0.006)· PFS non-inferior 3.02 vs 2.89 mo (HR 0.894, 95% CI 0.719–1.112, p=0.311)· n=536 randomised 1:1, ~45 centres in China· Less peripheral neuropathy, hypersensitivity, alopecia and musculoskeletal toxicity with the oral solution· Treatment-related fatal events 1.5% vs 1.1%

HR 0.77 ↓ 23% risk of death 95% CI 0.635–0.934 · Δ +2.6 mo median

Survival over time · schematic — OS*

Paclitaxel oral solutionPaclitaxel injection

Go deeper

536 patients with unresectable, recurrent or metastatic gastric cancer progressing after fluoropyrimidine-based first-line therapy were randomised 1:1 to paclitaxel oral solution (200 mg/m² twice daily on days 1, 8 and 15 of a 28-day cycle) or standard paclitaxel injection (175 mg/m² on day 1 of a 21-day cycle), with dual primary endpoints of blinded independent-review PFS and OS. The oral formulation was non-inferior on PFS — 3.02 versus 2.89 months, HR 0.894 (95% CI 0.719–1.112; P=0.311) — and superior on overall survival, 9.13 versus 6.54 months, HR 0.770 (95.5% CI 0.635–0.934; P=0.006). That combination deserves scrutiny rather than a headline: a 2.6-month OS gain without a PFS difference is not the usual shape of a cytotoxic benefit, and the most plausible explanations are tolerability and time on treatment rather than greater tumour control. The safety data point the same way, with less peripheral neuropathy, fewer hypersensitivity reactions, less alopecia and fewer musculoskeletal disorders; treatment-related fatal events were rare in both arms (1.5% versus 1.1%). Two limits belong beside the result: every site is in China, and the sponsor, Haihe Biopharma, has three co-authors on the paper. Full daily entry →

ESMO Open — Li J, Huang M, Deng T, … Qin S (senior), ~45 centres in China; Haihe Biopharma

On the radar Oesophageal · unresectable SCC after definitive chemoradiotherapy

SKYSCRAPER-07

A 760-patient trial that failed its primary endpoint and is more instructive for it. Adding tiragolumab to atezolizumab after definitive chemoradiotherapy did nothing. Underneath that failure, atezolizumab alone versus placebo produced an overall survival hazard ratio of 0.69. Because the testing hierarchy failed at the first gate, those monotherapy numbers are descriptive and cannot support a licence or a guideline on their own — and I would not present them as if they could. But consolidation PD-L1 blockade in a setting with very little to offer is a serious signal that deserves its own confirmatory trial rather than a dismissive headline.

· Primary comparison (atezolizumab + tiragolumab vs placebo) MISSED: PFS HR 0.82 (95% CI 0.65–1.03, p=0.0947)· OS HR 0.91 (95% CI 0.70–1.18, p=0.4772)· Descriptive monotherapy comparison (atezolizumab vs placebo): OS HR 0.69 (95% CI 0.52–0.91, p=0.0085), PFS HR 0.74 (95% CI 0.58–0.93)· n=760 randomised 1:1:1 across 166 centres in 28 countries

Go deeper

Patients with stage II-IVA unresectable oesophageal squamous cell carcinoma who had completed definitive chemoradiotherapy were randomised 1:1:1 to atezolizumab plus the anti-TIGIT antibody tiragolumab, atezolizumab plus placebo, or placebo. The primary comparison — the doublet versus placebo — missed: investigator-assessed PFS HR 0.82 (95% CI 0.65–1.03; p=0.0947) and OS HR 0.91 (95% CI 0.70–1.18; p=0.4772). Underneath it, atezolizumab alone versus placebo produced OS HR 0.69 (95% CI 0.52–0.91; p=0.0085) and PFS HR 0.74 (95% CI 0.58–0.93). Because the testing hierarchy failed at the first gate, those monotherapy numbers are descriptive and cannot support a licence or a guideline on their own. Read plainly, the trial says two things at once: adding TIGIT blockade bought nothing, and single-agent PD-L1 blockade after chemoradiation looks genuinely active in a setting with few options. The second finding needs its own confirmatory trial, and this one cannot substitute for it. Full daily entry →

Annals of Oncology — Xu R-H (Sun Yat-sen), … Chau I (senior, Royal Marsden), 166 centres across 28 countries

* 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

The take — the order of operations

Not one trial this week asked whether a therapy works. Every one of them asked where it sits in the sequence — and that is a harder question, because the answer changes depending on what came before it.

The two surgical trials are the clearest illustration, and they point in opposite directions in the same week. TALENTOP adds an operation back: in advanced hepatocellular carcinoma with macrovascular invasion, patients who respond to first-line atezolizumab–bevacizumab do better with resection than with continued maintenance, 20.4 versus 11.8 months to treatment failure. STAR-TREC takes an operation away: in early- and intermediate-stage rectal cancer, the right radiotherapy schedule leaves 78.5% of patients free of total mesorectal excision at 12 months rather than 60.6%. Neither trial contradicts the other. Both are asking when local therapy earns its place relative to what preceded it, and both answer that the systemic or radiotherapy phase determines the answer.

The costs are not symmetric, and it matters that we say so. TALENTOP's benefit came with grade 3–4 toxicity of 39% versus 21% and two treatment-related deaths — a benefit a patient has to be well enough to collect. STAR-TREC's advantage came with less serious gastrointestinal toxicity than upfront surgery, not more. When you are adding an intervention you owe the patient the harm column; when you are removing one, the harm column is the argument.

SKYSCRAPER-07 is the week's discipline check. The doublet failed, the monotherapy comparison underneath it returned an overall survival hazard ratio of 0.69, and because the hierarchy broke at the first gate that second number cannot carry a licence. It is tempting to quote the 0.69 and move on. The correct read is that a failed trial has produced a strong hypothesis, and a hypothesis is what needs funding — not a practice change.

The ESMO Express Update is the sequencing question answered at the level of the field rather than the patient: RASolute 302 read out, and the guideline moved in under three months rather than waiting for a revision cycle. That is the mechanism working. Worth noting for honesty's sake that we could not retrieve the full text, so the exact recommendation wording is not something this issue reproduces.

The common thread:

A trial that changes sequencing is not a smaller result than one that adds a drug; it is usually a harder one to run.
When you add an intervention, publish the harm column beside the benefit — TALENTOP does this well.
A significant comparison sitting below a failed primary is a hypothesis, not a finding. Say which one you are quoting.
An overall survival gain with no progression-free survival difference is a tolerability story until proven otherwise.
Interim results answer 'what should I do on Monday', not 'what is true'. STAR-TREC's real endpoint is eighteen months away.

Regulatory · what to watch

Recent Regulatory Approvals

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

FDA APPROVAL

Zanidatamab-hrii (Ziihera) with and without tislelizumab-jsgr (Tevimbra), plus chemotherapy — first-line HER2-positive gastroesophageal adenocarcinoma

Approved August 25, the same day this issue went out. Two regimens, and the label splits them by HER2 intensity: zanidatamab + tislelizumab + fluoropyrimidine/platinum chemotherapy for IHC 3+ or IHC 2+/ISH+ disease with no PD-L1 requirement, and zanidatamab + chemotherapy alone restricted to IHC 3+. In HERIZON-GEA-01 the triplet raised median overall survival to 26.4 months versus 19.2 with trastuzumab + chemotherapy (HR 0.72, 95% CI 0.57–0.90, p=0.0043) and median PFS to 12.4 versus 8.1 months (HR 0.63, 0.51–0.78, p<0.0001). The immunotherapy-free doublet improved PFS but not interim OS, with the effect concentrated in IHC 3+ tumours (median PFS 14.2 vs 7.6 months, HR 0.55) — which is where the FDA drew the line. Two Roche/Ventana companion diagnostics were approved alongside it, so HER2 scoring now selects the regimen rather than merely establishing eligibility. Full breakdown in the August 26 daily edition.

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