ESMO 2026 upper GI cancer round-up: Five important advances in upper GI cancer

New Phase III findings across gastric, gastroesophageal, biliary tract and pancreatic cancers explore the next generation of targeted therapies, immunotherapy combinations and treatment strategies, with important implications for the management of advanced disease.

The European Society for Medical Oncology (ESMO) Congress 2026 brought several important updates in gastrointestinal oncology, with new data addressing some of the key challenges in treating upper gastrointestinal (GI) malignancies. Among the studies attracting attention are HERIZON-GEA-01 and G-HOPE-001 in gastric and gastroesophageal junction (GEJ) cancer, HARMONi-GI1 in biliary tract cancer and RASolute 302 in pancreatic cancer. Meanwhile, KEYNOTE-975 provides important evidence on the role of immunotherapy alongside definitive chemoradiotherapy in esophageal and GEJ cancer. Here, we examine five abstracts selected for their potential clinical relevance and contribution to the evolving upper GI treatment landscape.

Dr Ken Kato (National Cancer Center Hospital, Tokyo, Japan) shares his selection of key abstracts, examining the latest findings and their potential implications for clinical practice.

HARMONi-GI1: Ivonescimab combination improves survival in first-line advanced biliary tract cancer

Abstract: LBA8 | ClinicalTrials.gov: NCT06591520 | Phase: III

HARMONi-GI1 compared ivonescimab, a bispecific antibody targeting PD-1 and VEGF, plus gemcitabine and cisplatin against durvalumab plus the same chemotherapy regimen as first-line treatment for unresectable locally advanced or metastatic biliary tract cancer (BTC). A total of 682 patients were randomized, with a median follow-up of 14.3 months. The primary endpoint was met, with median OS significantly improved with ivonescimab plus chemotherapy compared with durvalumab plus chemotherapy (14.9 vs 12.1 months; HR 0.75; 95% CI, 0.61–0.92; P=0.0031). Median PFS was also longer (8.4 vs 5.8 months; HR 0.61; 95% CI, 0.51–0.72; P<0.0001), while ORR was 53.8% versus 32.5% (P<0.0001). Treatment-related adverse events were generally well tolerated, with hematological toxicity among the most common TEAEs.

Key clinical takeaway: HARMONi-GI1 provides a direct comparison between two immunotherapy-based first-line regimens and demonstrated improvements in OS, PFS and ORR with ivonescimab plus chemotherapy. The findings raise the possibility of an alternative first-line treatment approach for advanced BTC, although the study was conducted in China and further evidence will be important when considering the applicability of the results across different populations.

HERIZON-GEA-01: Improved long-term survival with zanidatamab plus tislelizumab and the challenge of toxicity management

Abstract: LBA79 | ClinicalTrials.gov: NCT05152147 | Phase: III

Chemotherapy plus trastuzumab and pembrolizumab is already established as the standard of care for HER2-positive gastric cancer with PD-L1 CPS ≥1. Therefore, the clinical impact of HERIZON-GEA-01 has been questioned because its control arm did not include an immune checkpoint inhibitor (ICI). Updated results with longer follow-up demonstrated further improvement in OS. Compared with the initial ASCO-GI presentation, the OS hazard ratios improved from 0.80 to 0.76 for zanidatamab and from 0.72 to 0.65 for zanidatamab plus tislelizumab, both versus standard treatment. Although the direct comparison between the two experimental arms did not reach statistical significance (HR 0.81), the results suggest additional survival benefit from tislelizumab, potentially reflecting a long-term tail effect. However, toxicity remains a major concern, particularly diarrhea. Grade ≥3 diarrhea occurred in approximately 20% of patients, even in the highly selected population enrolled in clinical trials. Its incidence and severity could be greater in routine clinical practice, emphasizing the importance of effective supportive care and patient education. Historically, the introduction of novel anticancer therapies has repeatedly brought new toxicity challenges, including hypertension and thromboembolism with bevacizumab, immune-related adverse events with ICIs, and cytokine release syndrome and ICANS with T-cell engagers. Although these toxicities have not been completely overcome, accumulating clinical experience and advances in supportive care have improved their management.

Key clinical takeaway: Ultimately, maximizing the clinical benefit of novel therapies requires not only improved antitumor efficacy, but also effective management of treatment-related toxicities.

RASolute 302: Daraxonrasib demonstrates consistent survival benefit across RAS subgroups in metastatic pancreatic cancer

Abstract: LBA82 | ClinicalTrials.gov: NCT06625320 | Phase: III

RASolute 302 evaluated daraxonrasib, an oral RAS(ON) multi-selective inhibitor, against standard-of-care chemotherapy in previously treated metastatic pancreatic ductal adenocarcinoma (mPDAC). The primary analysis demonstrated improvements in OS and PFS with daraxonrasib in the overall population, including patients with and without an identified tumor RAS mutation. The additional subgroup analyses presented at ESMO examined OS according to tumor RAS mutation status, chemotherapy regimen and the disease setting of prior treatment. Daraxonrasib demonstrated a consistent OS treatment effect across the preselected chemotherapy subgroups, with HRs ranging from 0.32–0.46. An OS benefit was also observed across the reported RAS mutation subgroups, including KRAS G12D (HR 0.49; 95% CI, 0.33–0.74) and G12V (HR 0.32; 95% CI, 0.19–0.54). However, the smaller G12R, Q61 and no-identified-RAS-mutation subgroups had wider confidence intervals.

Key clinical takeaway: These findings provide additional evidence supporting the potential of RAS(ON) inhibition in previously treated mPDAC. The observed consistency across several mutation subgroups and chemotherapy comparators is particularly relevant given the limited treatment options in this setting. However, the smaller subgroup sizes and associated confidence intervals should be considered when interpreting these results.

G-HOPE-001: CLDN18.2-targeted ADC improves progression-free survival in previously treated gastric cancer

Abstract: LBA77 | ClinicalTrials.gov: NCT06238843 | Phase: III

Several CLDN18.2-targeted antibody–drug conjugates (ADCs) are currently under development, including sonesitatug vedotin (Sone-Ve), presented in the Presidential Session, and arcotatug tavatecan. These agents differ in their mechanisms of action: Sone-Ve carries the microtubule inhibitor MMAE, whereas arcotatug tavatecan delivers a topoisomerase I inhibitor and has silenced Fc-mediated antibody-dependent cellular cytotoxicity (ADCC). The G-HOPE-001 trial evaluated arcotatug tavatecan in patients with gastric cancer receiving third-line or later treatment, whereas CLARITY-Gastric01 included patients receiving second- or third-line therapy. Both trials required reassessment of CLDN18.2 expression after prior zolbetuximab treatment. G-HOPE-001 demonstrated remarkable efficacy, with hazard ratios of approximately 0.50 for both progression-free and overall survival and an objective response rate of 37%. These findings highlight the substantial antitumor activity of arcotatug tavatecan in heavily pretreated patients. Regarding safety, gastrointestinal adverse events related to on-target CLDN18.2 activity remain a concern. Interestingly, silencing Fc-mediated ADCC may reduce gastric mucosal toxicity, including gastritis, compared with Sone-Ve. However, cross-trial comparisons should be interpreted cautiously.
Importantly, differences in patient selection may also influence efficacy. G-HOPE-001 required CLDN18.2 expression in ≥75% of tumor cells, whereas CLARITY-Gastric01 included patients with lower expression levels. Therefore, differences in response rates cannot be attributed solely to drug activity.

Key clinical takeaway: Overall, CLDN18.2-targeted ADCs represent promising new treatment options for advanced gastric cancer. The key challenge will be optimizing treatment selection and sequencing between agents with different payloads, ADCC activities, toxicity profiles, and biomarker requirements. Further studies are needed to identify which patients benefit most from each ADC and determine their optimal positioning in gastric cancer treatment.

KEYNOTE-975: Pembrolizumab does not improve event-free survival with definitive chemoradiotherapy in esophageal and GEJ cancer

Abstract: 2875O | ClinicalTrials.gov: NCT04210115 | Phase: III

Definitive chemoradiotherapy (dCRT) is a curative treatment option for patients with esophageal cancer who are unfit for surgery or prefer organ preservation. However, its efficacy remains suboptimal, prompting investigations into combining dCRT with immune checkpoint inhibitors (ICIs). The SKYSCRAPER-07 trial suggested improved survival outcomes with sequential atezolizumab following dCRT. In contrast, recent phase III trials evaluating concurrent ICIs have yielded disappointing results. KEYNOTE-975 failed to demonstrate significant improvements in event-free survival (HR 0.97 in patients with PD-L1 CPS ≥10) or overall survival, while increasing grade ≥3 adverse events. Similarly, RATIONALE-311 showed no meaningful benefit from concurrent tislelizumab (PFS HR 0.92; OS HR 1.10). The ARION trial also failed to demonstrate additional efficacy. Similar findings have been reported in other malignancies. In non-small-cell lung cancer, consolidation durvalumab improved survival in PACIFIC, whereas concurrent administration failed in PACIFIC-2. Negative results have also been observed in head and neck cancer. One potential explanation is that irradiation of tumor-draining lymph nodes disrupts the CCR7–CCL19/CCL21 pathway, impairing T-cell trafficking and antitumor immunity. The ongoing CINSREC trial is investigating a reduced elective nodal irradiation dose combined with tislelizumab. Results from the KUNLUN trial evaluating concurrent durvalumab followed by maintenance therapy are also awaited.

Key clinical takeaway: Overall, current evidence does not support adding concurrent ICIs to dCRT in esophageal cancer. Sequential administration appears more promising, highlighting the importance of optimizing immunotherapy timing, duration, and radiation fields.

This content has been developed independently by Touch Medical Media for touchONCOLOGY. It is not affiliated with the European Society for Medical Oncology (ESMO). Views expressed are the speaker’s own and do not necessarily reflect the views of Touch Medical Media. This article was created by the editorial team utilizing AI as an editorial tool (Claude.AI and ChatGPT 5.5). The content was developed and edited by human editors. No funding was received in the publication of this article.

Cite: ESMO 2026 upper GI cancer round-up: Five important advances in upper GI cancer. touchONCOLOGY. 4th November 2026.

Editor: Sophie Nickelson