Introduction

Gastroesophageal cancers (encompassing adenocarcinomas and squamous cell carcinomas of the esophagus and stomach, including gastroesophageal junction tumors) have seen more treatment advances in the past two years than in the preceding decade. New data have reshuffled standards of care in both the resectable and metastatic settings, introduced a new class of biomarker-driven therapies, and raised pressing questions about which patients truly benefit from immunotherapy.

This review covers the most clinically significant recent advances: the shift from chemoradiation to perioperative chemotherapy in resectable esophageal adenocarcinoma, the emergence of novel HER2-targeted strategies, the expanding role of claudin 18.2 as a therapeutic target, updated biomarker-driven treatment selection in squamous cell carcinoma, and the evolving question of PD-L1 testing in perioperative and metastatic settings alike.

Esophageal Adenocarcinoma - Resectable Disease: The ESOPEC Trial Changes the Standard

For many years, the treatment of resectable esophageal and GEJ adenocarcinoma followed a chemoradiation-first approach based on the CROSS trial: preoperative carboplatin/paclitaxel with concurrent radiation, followed by surgery and adjuvant nivolumab for those with residual disease (per CheckMate 577 data). Updated CheckMate 577 overall survival data presented at ASCO 2025 showed a median OS of 45.5 months with nivolumab versus 33.5 months with placebo in PD-L1 CPS ≥1 patients (HR 0.79), confirming the long-term benefit of adjuvant nivolumab in this setting.

That perioperative paradigm has nonetheless shifted for adenocarcinoma specifically. The ESOPEC trial randomized patients with resectable esophageal adenocarcinoma to either perioperative FLOT chemotherapy (the gastric cancer approach, based on the FLOT4 trial) or chemoradiation per the CROSS regimen, without the adjuvant nivolumab component, which had not yet been established at the time of trial design. ESOPEC demonstrated superiority of perioperative FLOT over CROSS for both overall survival and progression-free survival. Pathologic complete response rates were 16–17% for FLOT versus 10% for the chemoradiation arm, which was lower than the 23% historically reported with CROSS, a discrepancy that may reflect lower-than-current radiation doses used in the trial's chemoradiation arm, in addition to lower treatment completion rate.

Despite these caveats, the data were compelling enough to shift NCCN guidelines toward perioperative chemotherapy as the preferred approach for fit, surgical candidates with resectable esophageal adenocarcinoma.

The Matterhorn Trial: Adding Immunotherapy to FLOT

For resectable gastric and GEJ adenocarcinoma, the Matterhorn trial addressed a critical question: does adding perioperative durvalumab to FLOT improve outcomes? Results were published in the New England Journal of Medicine in 2025 and updated at ESMO 2025. The trial demonstrated a positive primary endpoint of event-free survival. Two-year event-free survival improved from approximately 60% to nearly 70%, and pathologic complete response rate increased substantially, from 7% to nearly 20%, with no new safety signals beyond the expected immune-related adverse events (23% overall; 7% grade 3 or 4). Overall survival data, presented at ESMO 2025 by Tabernero et al., were also positive. FDA approval of durvalumab for resectable gastric and GEJ adenocarcinoma followed.

The PD-L1 question in the perioperative setting is where the complexity lies. The Matterhorn authors presented data suggesting benefit regardless of PD-L1 status, but this interpretation warrants scrutiny. PD-L1-negative patients (TPS <1%) comprised roughly 10% of the trial population, approximately forty patients per arm. Reconstructed Kaplan-Meier curves for this subgroup show confidence intervals that overlap completely. The trial was not powered to answer the question of PD-L1-negative benefit, and a pattern has emerged across this disease area: immunotherapy is approved broadly, then subsequently restricted to PD-L1-positive patients as evidence matures. This has already occurred in the metastatic setting with both pembrolizumab and nivolumab in first-line gastric cancer, and with nivolumab in squamous cell esophageal cancer. Two negative perioperative immunotherapy trials in gastric cancer are also instructive: KEYNOTE-585 (perioperative pembrolizumab plus chemotherapy) was negative, as was ATTRACTION-5 (adjuvant nivolumab after D2 gastrectomy).

For PD-L1-negative patients in the perioperative setting, shared decision-making is appropriate, weighing the absence of demonstrated benefit against the real toxicity burden of a full year of durvalumab, including colitis and other immune-mediated adverse events.

It is also worth noting that practice around FLOT cycle sequencing has evolved. The Matterhorn regimen uses four cycles preoperatively and four postoperatively, but postoperative completion rates are typically around 50%. There is increasing institutional experience with frontloading chemotherapy (six cycles preoperatively and two postoperatively), which appears to yield higher pathologic response rates, though event-free and overall survival data are similar to the standard four-and-four approach. The D-FLOT trial, presented at ESMO 2025 by Lorenzen et al., also demonstrated improved OS over standard FLOT plus surgery, adding to the evidence base for combining immunotherapy with FLOT in the perioperative setting.

Metastatic Disease: Biomarker Testing Is Non-Negotiable

Four biomarkers form the minimum testing framework for designing first-line therapy in metastatic gastric and GEJ adenocarcinoma: mismatch repair (MMR) status, HER2, PD-L1 combined positive score (CPS), and claudin 18.2.

HER2-Positive Disease: A Rapidly Evolving Landscape

The current standard of care for HER2-positive (IHC 3+ or IHC 2+/FISH-amplified) metastatic gastric and GEJ adenocarcinoma is pembrolizumab plus trastuzumab plus chemotherapy, based on KEYNOTE-811. The FDA in November 2023 restricted this indication to PD-L1-positive (CPS ≥1) tumors. Median overall survival in the PD-L1 CPS ≥1 population is 20.0 months with pembrolizumab versus 15.7 months with placebo (HR 0.81), a benchmark worth keeping in mind when evaluating emerging data.

Zanidatamab is a bispecific antibody binding two distinct extracellular subdomains (II and IV) of HER2 in a trans configuration, enabling HER2 receptor clustering, enhanced internalization, and immune-mediated cytotoxicity. This mechanism is functionally distinct from the trastuzumab-plus-pertuzumab combination, which targets the same two receptors but was negative in the JACOB trial in gastric cancer. The HERIZON-GEA-01 trial, presented by Elimova et al. at ASCO GI 2026, randomized HER2-positive metastatic gastric/GEJ patients to three arms: standard chemotherapy plus trastuzumab (the comparator arm, notably lacking immunotherapy), chemotherapy plus zanidatamab plus tislelizumab (a PD-1 inhibitor), or chemotherapy plus zanidatamab alone.

Despite the suboptimal comparator arm, the results were positive: the triplet achieved a PFS of 12.4 months versus a shorter duration with chemotherapy plus trastuzumab. Overall survival was also improved in both zanidatamab-containing arms. The triplet arm showed a particularly notable difference in duration of response: 27 months versus approximately 14 months with the doublet. The primary toxicity concern is diarrhea, including high rates of grade 3 diarrhea even with prophylactic antidiarrheal therapy; this needs to be treated proactively and discussed with patients before starting treatment. The zanidatamab-containing regimen is expected to become a new standard of care, possibly within this year.

Claudin 18.2-Positive, HER2-Negative Disease

Claudin 18.2 is a tight junction protein validated as a therapeutic target in approximately 35–40% of gastric and GEJ adenocarcinomas. Zolbetuximab, an anti-claudin 18.2 antibody, is now established for claudin 18.2-positive, HER2-negative metastatic gastric and GEJ cancer based on the SPOTLIGHT and GLOW trials. In a pooled analysis of both trials, zolbetuximab plus chemotherapy improved mPFS (9.2 vs 8.2 months) and mOS (16.4 vs 13.7 months) compared with chemotherapy alone. The primary toxicity is nausea and vomiting, manageable but clinically significant, particularly in the first two infusion cycles.

In May 2025, the FDA also restricted pembrolizumab and nivolumab indications in HER2-negative gastric cancer to PD-L1-positive tumors, reflecting the consistent finding across trials that PD-L1-negative patients do not derive meaningful benefit from the addition of immunotherapy.

The claudin 18.2 and PD-L1 co-positive scenario is a common clinical question: when a tumor is both claudin 18.2-positive and PD-L1-positive, which pathway should be targeted? There is no phase III randomized answer. The general principle is that higher PD-L1 expression (CPS ≥5, or more conservatively ≥10) tends to favor immunotherapy, based on consistent subgroup data across trials. At lower CPS values, zolbetuximab or immunotherapy are both guideline-consistent options; individual patient factors inform the choice.

The ILUSTRO trial (phase II, non-randomized), presented by Shitara et al. at ASCO GI 2026, explored the combination of chemotherapy plus nivolumab plus zolbetuximab in claudin 18.2-positive (≥50% of tumor cells), HER2-negative patients. In the CLDN18.2-high subgroup (≥75% expression), the objective response rate was 68.1%. PFS in the claudin 18.2-positive/PD-L1 CPS >1 biomarker-positive cohort was 23.6 months, higher than the overall survival figures from either SPOTLIGHT/GLOW or KEYNOTE-811. This is promising, and a confirmatory phase III trial, LUCERNA (NCT06901531), is now open. Antibody-drug conjugates targeting claudin 18.2 in combination with PD-L1 inhibition are also in development.

Squamous Cell Carcinoma: PD-L1 Restriction and a New Second-Line Option

In squamous cell esophageal cancer, first-line treatment is now chemotherapy plus immunotherapy, established by multiple phase III trials. The FDA's Oncologic Drug Advisory Committee voted 11 to 1 in September 2024 against the use of checkpoint inhibitors in patients with PD-L1 expression less than 1, based on pooled analyses of the pivotal approval trials demonstrating that PD-L1-negative patients do not derive meaningful benefit; adding immunotherapy to this subgroup adds toxicity without a survival advantage.

In the second-line setting, a significant unmet need has long existed: standard second-line chemotherapy achieves an objective response rate of less than 10% and a median OS of less than six months in patients who have progressed on first-line chemoimmunotherapy. The PANKU-Esophagus01 trial, presented by Lu Z et al. at ASCO 2026 (Abstract 17141), addressed this gap. The trial randomized 488 patients with recurrent or metastatic squamous cell esophageal cancer who had progressed on first-line PD-1/PD-L1 inhibitor plus platinum-based chemotherapy to izalontamab brengitecan (iza-bren) versus investigator's choice chemotherapy. Iza-bren is a first-in-class EGFR×HER3 bispecific antibody-drug conjugate conjugated to a topoisomerase I inhibitor via a cleavable linker.

Both dual primary endpoints were met. Median PFS was 4.17 months with iza-bren versus 1.97 months with chemotherapy (HR 0.50; p<0.0001). Median OS was 9.79 months with iza-bren versus 7.20 months with chemotherapy (HR 0.64; p=0.0004). Objective response rate was 35.3% with iza-bren versus 13.1% with chemotherapy. The benefit was consistent across prespecified subgroups. Grade ≥3 treatment-related adverse events occurred in 85.1% of iza-bren patients versus 60.2% with chemotherapy, though treatment discontinuation rates were low (2.0% vs 3.3%). This trial is likely to establish iza-bren as a new second-line standard of care in China; validation in Western populations will be needed.

For Patients

Gastroesophageal cancers are among the more molecularly complex GI malignancies, and treatment decisions now depend heavily on biomarker testing. If you or a loved one has been diagnosed with stomach or esophageal cancer, asking about comprehensive biomarker testing (including HER2, PD-L1, claudin 18.2, and mismatch repair status) is one of the most important steps you can take. These results determine which treatments are most likely to work.

For patients with resectable disease, the sequencing and type of chemotherapy given before surgery have evolved significantly. Seeking care at a high-volume center with a multidisciplinary team that includes a GI surgeon experienced in minimally invasive gastrectomy, a medical oncologist, and a radiation oncologist will allow the full range of treatment options to be considered.

Clinical trials remain an important option at multiple points in treatment. Enrollment in trials evaluating next-generation HER2 agents, claudin 18.2-targeting combinations, and novel antibody-drug conjugates can provide access to emerging therapies while contributing to the evidence base.

Key Takeaways

  • The ESOPEC trial established perioperative FLOT as superior to CROSS-based chemoradiation for resectable esophageal adenocarcinoma in fit surgical candidates; NCCN guidelines now reflect this.

  • Updated CheckMate 577 OS data confirm long-term benefit of adjuvant nivolumab after chemoradiation in PD-L1 CPS ≥1 esophageal cancer (median OS 45.5 vs 33.5 months).

  • The Matterhorn trial (NEJM 2025) showed perioperative durvalumab plus FLOT improves EFS and pCR in gastric/GEJ adenocarcinoma; the benefit in PD-L1-negative patients remains unestablished; KEYNOTE-585 and ATTRACTION-5 were both negative in overlapping settings.

  • Minimum biomarker testing for first-line metastatic gastric/GEJ adenocarcinoma: MMR, HER2, PD-L1 (CPS), and claudin 18.2.

  • Zanidatamab plus tislelizumab plus chemotherapy (HERIZON-GEA-01) showed improvements in PFS, OS, and duration of response over trastuzumab plus chemotherapy in HER2-positive disease; proactive diarrhea management is essential.

  • In claudin 18.2-positive/PD-L1-positive tumors, immunotherapy is generally preferred at CPS ≥5–10; the ILUSTRO trial combining nivolumab, zolbetuximab, and chemotherapy showed an ORR of 68.1% in CLDN18.2-high patients; phase III LUCERNA is ongoing.

  • FDA restricted immunotherapy in PD-L1-negative squamous esophageal cancer following an 11-1 ODAC vote in September 2024; iza-bren (PANKU-Esophagus01, ASCO 2026) demonstrated PFS and OS benefit in the second-line post-immunotherapy setting (HR 0.50 and 0.64, respectively).

  • Universal upfront DPYD/UGT1A1 testing before fluoropyrimidine-based therapy is increasingly standard; coordination between surgical and medical oncology teams is key.

References

  1. Al-Batran SE, et al. Perioperative chemotherapy with fluorouracil plus leucovorin, oxaliplatin, and docetaxel versus fluorouracil or capecitabine plus cisplatin and epirubicin for locally advanced, resectable gastric or gastro-oesophageal junction adenocarcinoma (FLOT4). Lancet. 2019;393(10184):1948–1957.

  2. Hoeppner J, et al. (2024). Prospective randomized multicenter phase III trial comparing perioperative chemotherapy (FLOT protocol) to neoadjuvant chemoradiation (CROSS protocol) in patients with adenocarcinoma of the esophagus (ESOPEC trial). Journal of Clinical Oncology, 42(16_suppl), LBA1-LBA1.DOI: 10.1200/JCO.2024.42.17_suppl.LBA1

  3. Shapiro J, et al. Neoadjuvant chemoradiotherapy plus surgery versus surgery alone for oesophageal or junctional cancer (CROSS). Lancet Oncol. 2015;16(9):1090–1098.

  4. Kelly RJ, et al. Adjuvant nivolumab in resected esophageal or gastroesophageal junction cancer (CheckMate 577). N Engl J Med. 2021;384(13):1191–1203.

  5. Kelly RJ, et al. CheckMate 577 updated OS. ASCO 2025. Abstract 4000.

  6. Janjigian YY, Al-Batran SE, Wainberg ZA, et al. Perioperative Durvalumab in Gastric and Gastroesophageal Junction Cancer. N Engl J Med. 2025;393(3):217-230. doi:10.1056/NEJMoa2503701

  7. Tabernero J, et al. Matterhorn final OS. ESMO 2025.

  8. Lorenzen S, et al. D-FLOT improves OS vs SOC FLOT and surgery. ESMO 2025.

  9. Janjigian YY, et al. KEYNOTE-585: perioperative pembrolizumab plus chemotherapy in gastric/GEJ cancer. J Clin Oncol. 2024;42(suppl 3):abstr 247.

  10. Kang YK, et al. ATTRACTION-5: adjuvant nivolumab after D2 gastrectomy. Lancet Gastroenterol Hepatol. 2024;9(8):705–717.

  11. Janjigian YY, et al. KEYNOTE-811: pembrolizumab plus trastuzumab and chemotherapy for HER2-positive gastric/GEJ cancer. Nature. 2021;595(7867):587–592.

  12. Janjigian YY, Kawazoe A, Bai Y, et al. Pembrolizumab plus trastuzumab and chemotherapy for HER2-positive gastric or gastro-oesophageal junction adenocarcinoma: interim analyses from the phase 3 KEYNOTE-811 randomised placebo-controlled trial. Lancet. 2023;402(10418):2197-2208. doi:10.1016/S0140-6736(23)02033-0

  13. Tabernero J, et al. JACOB: trastuzumab plus pertuzumab plus chemotherapy in HER2-positive gastric cancer. Lancet Oncol. 2018;19(10):1372–1384.

  14. Elimova E, et al. HERIZON-GEA-01: zanidatamab plus tislelizumab plus chemotherapy in HER2-positive gastric/GEJ cancer. ASCO GI 2026.

  15. Shitara K, et al. Zolbetuximab plus mFOLFOX6 in CLDN18.2-positive, HER2-negative gastric/GEJ adenocarcinoma (SPOTLIGHT). Lancet. 2023;401(10389):1655–1668.

  16. Shah MA, et al. Zolbetuximab plus CAPOX in CLDN18.2-positive gastric/GEJ adenocarcinoma (GLOW). Nat Med. 2023;29(8):2133–2141.

  17. Shitara K, et al. ILUSTRO cohort 4B: zolbetuximab plus nivolumab plus chemotherapy in CLDN18.2-positive gastric/GEJ cancer. ASCO GI 2026.

  18. LUCERNA phase III trial. NCT06901531.

  19. Lu Z, et al. PANKU-Esophagus01: izalontamab brengitecan versus chemotherapy in second-line recurrent/metastatic esophageal squamous cell carcinoma. ASCO 2026. Abstract 17141.

  20. Meeting of the Oncologic Drug Advisory Committee. FDA. September 26, 2024. https://tinyurl.com/3j2aufse

  21. Lemery SJ. FDA restricts pembrolizumab in PD-L1-low gastric cancers. BLA 125514/S-182. FDA. May 22, 2025.

  22. Lemery SJ. FDA restricts nivolumab in PD-L1-low gastric cancers. BLA 125554/S-133. FDA. May 23, 2025.

  23. Patel AK, Sethi NS, Park H. Gastric Cancer: A Review. JAMA. 2026;335(5):439–450. doi:10.1001/jama.2025.20034