The ESMO 2026 Congress, held in Madrid from October 23–27, presented important new evidence across gastroesophageal cancer, including advances in HER2- and CLDN18.2-directed therapies, immunotherapy, biomarker-guided treatment, and chemotherapy optimization. Several phase III studies reported significant survival improvements, while others provided evidence for shorter treatment schedules, alternative regimens, or emerging therapeutic strategies. Here are ten notable updates from the congress.
HERIZON-GEA-01: Zanidatamab in 1L HER2-positive GEA
The phase III HERIZON-GEA-01 trial (LBA79), presented by Kohei Shitara, evaluated zanidatamab plus chemotherapy, with or without tislelizumab, versus trastuzumab plus chemotherapy in previously untreated HER2-positive advanced or metastatic gastroesophageal adenocarcinoma.
At a median follow-up of 35.3 months, zanidatamab plus chemotherapy significantly improved overall survival (OS) compared with trastuzumab plus chemotherapy, with median OS of 22.9 versus 19.6 months (HR 0.76; 95% CI 0.62–0.93; p=0.0088). Median OS was 27.0 months with zanidatamab plus tislelizumab and chemotherapy (HR 0.65; 95% CI 0.52–0.80 versus trastuzumab plus chemotherapy). However, the direct comparison between the two zanidatamab-containing regimens did not demonstrate a statistically significant additional OS benefit from tislelizumab (HR 0.81; 95% CI 0.65–1.00; p=0.0502).
Median progression-free survival (PFS) was 12.7, 12.4, and 8.3 months with the triplet, zanidatamab plus chemotherapy, and trastuzumab plus chemotherapy, respectively.
Key takeaway: The updated results strengthen the evidence supporting zanidatamab-based first-line treatment in HER2-positive advanced gastroesophageal adenocarcinoma, although the additional OS benefit of tislelizumab over zanidatamab plus chemotherapy alone was not statistically established.
DEWI: Dostarlimab and Watch-and-Wait in Localized dMMR/MSI-H Disease
The phase II GERCOR-GONO DEWI study evaluated dostarlimab followed by a watch-and-wait strategy in patients with localized mismatch repair-deficient (dMMR)/microsatellite instability-high (MSI-H) gastroesophageal adenocarcinoma. The primary endpoint was clinical complete response (cCR) at one year without surgery. In the planned first interim analysis involving 24 evaluable patients, 15 (62.5%) achieved one-year cCR without surgery.
No relapse or death occurred among these patients at a median follow-up of 16.2 months. Grade 3–4 dostarlimab-related adverse events occurred in 21% of patients. Among patients who underwent surgery, postoperative complications included anastomotic fistulas, and one postoperative death was reported.
Key takeaway: The findings provide preliminary evidence supporting further evaluation of dostarlimab-based organ preservation in localized dMMR/MSI-H gastroesophageal adenocarcinoma. The small interim analysis population and limited follow-up warrant cautious interpretation.
KC-WISE: Anbenitamab After Trastuzumab Progression
The phase III KC-WISE trial (LBA80), presented by Ruixing Zhang, evaluated anbenitamab plus chemotherapy versus placebo plus chemotherapy in patients with HER2-positive advanced gastric or gastroesophageal junction (GEJ) cancer following progression on trastuzumab-containing therapy.
At the second interim OS analysis, median OS was 19.6 versus 9.6 months (HR 0.39; 95% CI 0.28–0.54; p<0.0001), demonstrating a statistically significant survival improvement with anbenitamab.
Independent review committee-assessed median PFS was 7.1 versus 2.7 months, while the objective response rate (ORR) was 56.0% versus 14.0%. Grade ≥3 treatment-emergent adverse events (TEAEs) occurred in 74.2% versus 54.5% of patients, with higher rates of hematologic adverse events in the anbenitamab group.
Key takeaway: Anbenitamab plus chemotherapy demonstrated a significant OS improvement and higher response rate following trastuzumab progression, although these benefits were accompanied by a higher incidence of severe TEAEs.
LOMAC: Four Versus Six Months of Adjuvant XELOX
The phase III LOMAC trial (LBA81), presented by Dazhi Xu, evaluated whether four months of adjuvant capecitabine plus oxaliplatin (XELOX) was noninferior to six months in patients with stage II–IIIB gastric cancer following curative D2 gastrectomy.
The trial included 1,023 patients. At a median follow-up of 52.8 months, three-year disease-free survival (DFS) was 69.2% versus 70.8% (HR 1.01; 95% CI 0.82–1.25; one-sided p=0.006 for noninferiority), meeting the primary endpoint. The shorter regimen was associated with fewer grade ≥3 adverse events (8.4% vs 15.0%) and higher treatment completion (83.6% vs 60.4%). Grade ≥3 peripheral sensory neuropathy occurred in 0.4% versus 5.9%.
Key takeaway: Four months of adjuvant XELOX demonstrated noninferior DFS compared with six months, with reduced severe toxicity and improved treatment completion. The findings support consideration of shorter adjuvant treatment in appropriately selected patients after D2 gastrectomy.
CLARITY-Gastric 01: Sonesitatug vedotin in CLDN18.2-positive GEC
The phase III CLARITY-Gastric01 trial (LBA7), presented by Rui-Hua Xu, evaluated the CLDN18.2-directed antibody–drug conjugate (ADC) sonesitatug vedotin versus investigator’s choice of therapy in patients with previously treated, unresectable locally advanced or metastatic gastroesophageal carcinoma. The study required CLDN18.2 membranous staining at any intensity in ≥25% of tumor cells.
The trial had two primary endpoints: overall survival (OS) in the third-line-or-later population and blinded independent central review (BICR)-assessed progression-free survival (PFS) in the overall second-line-or-later intention-to-treat (ITT) population. In the third-line-or-later population, sonesitatug vedotin significantly improved median OS to 10.0 versus 5.0 months(HR 0.47; 95% CI 0.36–0.62; p<0.000001), meeting the primary OS endpoint.
In the overall ITT population, median OS, a key secondary endpoint, was 10.6 versus 6.4 months (HR 0.58; 95% CI 0.47–0.73; p=0.000004). However, the co-primary BICR-assessed PFS endpoint was not met, with median PFS of 4.2 versus 3.8 months (HR 0.77; 95% CI 0.62–0.96; p=0.053193). Grade ≥3 treatment-related adverse events occurred in 36.4% versus 44.8% of patients. Treatment-related deaths were reported in 1.1% versus 0.8%, respectively.
Key takeaway: Sonesitatug vedotin demonstrated a significant OS benefit in previously treated CLDN18.2-positive gastroesophageal cancer, including patients with CLDN18.2 expression in ≥25% of tumor cells. However, the co-primary PFS endpoint was not met, and the findings should be interpreted in the context of the study population and safety profile.
G-HOPE-001: Arcotatug Tavatecan in Previously Treated Gastric/GEJ Cancer
The phase III G-HOPE-001 trial (LBA77), presented by Kohei Shitara, evaluated arcotatug tavatecan, a CLDN18.2-directed ADC, versus physician’s choice of chemotherapy in patients with CLDN18.2-positive, HER2-negative advanced gastric or GEJ adenocarcinoma who had received two to four prior treatment regimens.
The trial met both primary endpoints of OS and PFS. At the second prespecified interim analysis, median OS was 10.2 versus 6.9 months (HR 0.57; 95% CI 0.45–0.73; p<0.0001), while median PFS was 7.1 versus 3.1 months (HR 0.45; 95% CI 0.35–0.58). Confirmed ORR was 37.4% versus 7.7%. Grade ≥3 treatment-related adverse events occurred in 47.4% versus 55.9% of patients. One treatment-related fatal adverse event, reported as sudden death of unknown cause, occurred in the arcotatug tavatecan group.
Key takeaway: Arcotatug tavatecan significantly improved both OS and PFS in previously treated CLDN18.2-positive gastric/GEJ adenocarcinoma, providing further phase III evidence for CLDN18.2-directed ADCs in this setting.
KEYNOTE-975: Pembrolizumab With Definitive Chemoradiotherapy
The phase III KEYNOTE-975 trial (2875O), presented by Manish Shah, evaluated pembrolizumab versus placebo added to definitive chemoradiotherapy (dCRT) in patients with previously untreated, unresectable esophageal or GEJ cancer.
At the second interim analysis, pembrolizumab did not significantly improve event-free survival (EFS) in patients with PD-L1 combined positive score (CPS) ≥10. Median EFS was 15.2 versus 14.7 months (HR 0.97; 95% CI 0.77–1.23; p=0.3994). Median OS was 32.2 versus 35.0 months (HR 1.00; 95% CI 0.77–1.30).
Because the prespecified EFS endpoint was not met, subsequent formal statistical testing of other EFS populations and OS was not performed. Grade ≥3 adverse events occurred in all patients in both groups, while adverse events leading to treatment discontinuation were more frequent with pembrolizumab (34% vs 27%). The study was stopped following the negative efficacy analysis.
Key takeaway: KEYNOTE-975 did not demonstrate an EFS benefit from adding pembrolizumab to dCRT. These findings do not support routine incorporation of the tested regimen for unresectable esophageal or GEJ cancer.
RAMIRIS: FOLFIRI Plus Ramucirumab After Prior Taxane Treatment
The phase III RAMIRIS trial (2878RO), presented by Magdalena Scheck, compared FOLFIRI plus ramucirumab with paclitaxel plus ramucirumab as second-line treatment in taxane-pretreated advanced gastroesophageal adenocarcinoma.
Neither co-primary endpoint was met. Median OS was 8.7 versus 7.5 months (HR 0.95; 95.8% CI 0.74–1.22; one-sided p=0.33), while ORR was 25% versus 23% (p=0.29). However, the secondary PFS endpoint favored FOLFIRI plus ramucirumab, with median PFS of 5.5 versus 3.7 months (HR 0.71; 95.8% CI 0.56–0.90; one-sided p=0.002). FOLFIRI plus ramucirumab was associated with more gastrointestinal adverse events but less peripheral sensory neuropathy.
Key takeaway: Although RAMIRIS did not meet its co-primary endpoints, FOLFIRI plus ramucirumab demonstrated longer PFS and may represent an alternative for selected taxane-pretreated patients, particularly those with pre-existing neuropathy.
TQB2102: A HER2-Directed ADC Combination in First-Line Treatment
A phase II study presented by Miaozhen Qiu evaluated TQB2102, a HER2 bispecific ADC, combined with the anti-PD-L1 antibody benmelstobart (TQB2450), with or without capecitabine, as first-line treatment for HER2-positive advanced gastric or GEJ cancer.
The study included 86 patients across three treatment cohorts receiving different TQB2102 doses and combinations. The ORR was 69.6% with TQB2102 plus benmelstobart and 84.2% across the triplet cohorts receiving additional capecitabine. In the 6 mg/kg TQB2102 triplet cohort, ORR reached 95.8% (23/24), with a confirmed ORR of 91.7%. Median PFS was 12.68 months in the doublet cohort, while PFS data for the triplet cohorts remained immature. These findings arise from a phase II cohort study and do not establish a comparative survival advantage between regimens.
Key takeaway: TQB2102-based combinations demonstrated encouraging antitumor activity in first-line HER2-positive gastric/GEJ cancer. Larger randomized studies are needed to establish efficacy, safety, and the optimal combination regimen.
FORTITUDE-101/102: FGFR2 Amplification as a Potential Predictive Biomarker
An exploratory biomarker analysis of the phase III FORTITUDE-101 and FORTITUDE-102 trials (2883RO), presented by Zev A. Wainberg, examined outcomes with the FGFR2b-targeted antibody bemarituzumab according to FGFR2 amplification status.
In patients with FGFR2-amplified tumors, median OS numerically favored bemarituzumab in both trials. In FORTITUDE-101, median OS was 19.3 versus 8.0 months (HR 0.50; 95% CI 0.22–1.12). In FORTITUDE-102, median OS was 16.6 versus 8.0 months (HR 0.67; 95% CI 0.34–1.32). However, confidence intervals crossed 1 in both exploratory subgroup analyses, and the findings did not establish FGFR2 amplification as a validated predictive biomarker.
Key takeaway: Exploratory findings suggest that FGFR2 amplification may be associated with greater benefit from bemarituzumab in FGFR2b-overexpressing gastroesophageal cancer. However, its predictive value remains unvalidated, and prospective studies are needed.
What ESMO 2026 means for GEC
ESMO 2026 provided important evidence for both new therapeutic approaches and optimization of established treatments in gastroesophageal cancer.
Among the most notable findings, HERIZON-GEA-01 and KC-WISE demonstrated significant survival improvements with HER2-directed regimens in different treatment settings. CLARITY-Gastric01 and G-HOPE-001 provided phase III evidence supporting CLDN18.2-targeted ADCs, although their populations, biomarker criteria, and primary endpoint outcomes differed.
In resected gastric cancer, LOMAC demonstrated that a shorter course of adjuvant XELOX could maintain DFS outcomes while reducing treatment-related toxicity. Meanwhile, KEYNOTE-975 and RAMIRIS highlighted the importance of interpreting negative primary endpoints alongside secondary efficacy and safety findings.
Early-phase and exploratory studies, including DEWI, TQB2102, and the FORTITUDE biomarker analyses, provided additional evidence for strategies requiring further clinical evaluation.
Together, these results demonstrate continued progress toward biomarker-selected treatment and more individualized therapeutic approaches in gastroesophageal cancer. Their clinical application will depend on the populations studied, endpoints achieved, safety profiles, and regulatory availability.
The full abstracts of the studies discussed are available through the official ESMO Congress 2026 programme.
Written by Tatev Barseghyan, MD