Key Ideas

Perioperative immunotherapy has become the new standard of care for resectable, gastric and gastroesophageal junction (GEJ) adenocarcinoma, with both the MATTERHORN and NeoSUMMIT-01 trials showing meaningful improvements in pathologic response and long-term survival regardless of PD-L1 status. Separately, in metastatic HER2-positive gastroesophageal adenocarcinoma, the bispecific antibody zanidatamab is emerging as a potential new option that may extend survival beyond what trastuzumab-based regimens have achieved.

Setting the Stage: A Landmark Year for Upper GI Cancer

Several major trials reported in 2025 reshaped the treatment landscape for resectable esophageal and gastric cancers. The ESOPEC trial, a long-awaited comparison of perioperative chemotherapy (the FLOT regimen) versus neoadjuvant chemoradiation (the CROSS regimen) in resectable esophageal adenocarcinoma, showed a significant improvement in both progression-free and overall survival favoring perioperative FLOT. The same year, the global phase III MATTERHORN trial showed statistically significant improvement in event-free survival with the addition of durvalumab to perioperative FLOT chemotherapy in resectable gastric and GEJ adenocarcinoma. The addition of immunotherapy more than doubled the pathologic complete response rate, and a subsequent final analysis confirmed a significant overall survival benefit as well (hazard ratio 0.78). Perioperative chemotherapy with the addition of immunotherapy particularly for PD-L1-positive patients is the new standard of care and a category 1 NCCN recommendation.

NeoSUMMIT-01: Long-Term Benefit with Toripalimab

The NeoSUMMIT-01 trial, a prospective randomized phase 2 study conducted at two sites in China, evaluated adding the PD-1 inhibitor toripalimab to perioperative platinum-doublet chemotherapy in treatment-naive, locally advanced gastric or GEJ adenocarcinoma. Of 108 randomized patients, the majority had gastric cancer with about a third having GEJ tumors, and the population skewed toward advanced disease (mostly T4a, all node-positive).

The trial met its primary endpoint, initially published in Nature Medicine in 2024, with pathologic complete or near-complete response improving from 7% with chemotherapy alone to 22% with toripalimab added. Updated three-year follow-up now shows this pathologic benefit translated into durable survival gains: three-year event-free survival was 75% with toripalimab versus 56% with chemotherapy alone, and three-year overall survival was 81% versus 72% (hazard ratio 0.45). Immune-related adverse events occurred in 13% of patients, including one case of grade 3 myocarditis, alongside a modest increase in cytopenias and transaminitis; treatment discontinuation due to adverse events remained low.

Comparing NeoSUMMIT-01 with MATTERHORN is instructive. NeoSUMMIT-01 enrolled a smaller, more advanced population (all T3/T4, node-positive) using a platinum-doublet backbone, reflecting the historical East Asian standard of upfront D2 gastrectomy followed by adjuvant doublet chemotherapy for early-stage disease, with perioperative chemotherapy reserved for T4 or node-positive disease. MATTERHORN, by contrast, enrolled any resectable stage II-IVA patient using the FLOT backbone, the long-standing Western standard. Despite these differences in population, chemotherapy backbone, and geography, both trials showed more than a doubling of pathologic complete response with immunotherapy added, along with statistically significant PFS and OS improvement, and both showed benefit regardless of PD-L1 status, though the number of PD-L1-negative patients in each trial remains too limited to draw firm conclusions about that specific subgroup. Current NCCN guidance reflects this nuance: perioperative durvalumab is a category 1 recommendation for PD-L1-positive patients and category 2B for PD-L1-negative patients.

Several open questions remain across both trials: how best to select patients for perioperative immune checkpoint inhibition (TAP versus CPS score, PD-L1 threshold for treatment), the optimal duration of immunotherapy (one year in the durvalumab trials on which MATTERHORN's guideline recommendation is based, versus up to six months in NeoSUMMIT-01), whether complete responders truly need adjuvant chemotherapy given how poorly many patients tolerate chemotherapy after gastrectomy, and whether ctDNA could eventually help answer that question. Longer follow-up and subgroup analyses across these trials should help clarify some of these points.

HERIZON-GEA-01: A New Option for HER2-Positive Metastatic Disease

Roughly 20% of patients with metastatic gastroesophageal adenocarcinoma are HER2-positive (IHC 3+ or IHC 2+ with FISH positivity). Current standard of care is chemotherapy with trastuzumab, with the addition of pembrolizumab for PD-L1-positive patients, which is based on data from the ToGA and KEYNOTE-811 trials. While this represents a major improvement over chemotherapy alone, median progression-free survival remains less than one year and median overall survival around 20 months, leaving substantial room for improvement.

Zanidatamab is a HER2-directed bispecific antibody that binds two separate HER2 extracellular domains, forcing receptor clustering that increases internalization and degradation while blocking downstream signaling; it also carries immune-mediated cytotoxic effects, including complement-mediated cytotoxicity. The phase 3 HERIZON-GEA-01 trial randomized treatment-naive patients with HER2-positive locally advanced or metastatic gastroesophageal adenocarcinoma to trastuzumab plus chemotherapy (standard of care, without immunotherapy), zanidatamab plus chemotherapy, or zanidatamab plus the PD-1 inhibitor tislelizumab plus chemotherapy.

Results published in the New England Journal of Medicine showed statistically significant improvement in both progression-free and overall survival with zanidatamab added. Median PFS was 12.4 months in both zanidatamab-containing arms versus 8.1 months with trastuzumab plus chemotherapy. Median overall survival was 26.4 months with the triplet (zanidatamab plus tislelizumab plus chemotherapy), 24.4 months with the zanidatamab doublet, and 19.2 months with trastuzumab plus chemotherapy; the triplet's overall survival improvement reached statistical significance, while the doublet showed only a trend (p=0.06). Benefit with the triplet held regardless of PD-L1 expression.

Diarrhea was the most common adverse event, occurring in roughly 80% of patients, typically within the first two weeks; mandatory antidiarrheal prophylaxis was required during cycle 1, and most cases were grade 1-2 and resolved within about two weeks. Other notable grade 3 events included hypokalemia, cytopenias, and nausea. Left ventricular dysfunction occurred in about 9% of patients on the triplet versus 4% with trastuzumab, and immune-related adverse events, most commonly rash, occurred in 38% of triplet-treated patients, though treatment discontinuation rates remained low overall. In August 2026, the FDA approved zanidatamab and tislelizumab in combination with chemotherapy for first-line treatment of HER2+ unresectable or metastatic gastric, gastroesophageal or esophageal adenocarcinoma.

Questions remain regarding patient selection for the doublet versus the triplet zanidatamab arms. PFS was identical between both experimental arms, and the overall survival difference was about two months, raising a fair question about whether tislelizumab is truly adding benefit given the added cost of combining multiple biologic agents. That said, duration of response was meaningfully longer with the triplet (21 months versus 14 months with the doublet), and only the triplet reached statistical significance for overall survival. At this time, the FDA has approved the doublet regimen in patients with HER2 IHC 3+ tumors based on exploratory analyses however longer follow-up and, ideally, further analysis comparing the two experimental arms directly will hopefully provide more guidance. It's also worth noting this trial did not include U.S. sites, though it did enroll a broad international population across roughly 30 countries, and it did not include patients with HER2-low disease, an open area for future study.

Looking Ahead: Resistance and the Trials to Watch

As more HER2-targeted agents enter clinical use, re-biopsy at progression becomes increasingly important: as many as 60% of HER2-positive patients lose HER2 expression after trastuzumab exposure, and emerging resistance mechanisms, including PI3K pathway mutations and MET amplification, are increasingly recognized. Reassessing HER2 status at metastatic sites before selecting next-line therapy should be standard practice going forward.

Several other trials worth watching in upper GI oncology include CRITICS-2 (evaluating different neoadjuvant approaches in gastric cancer), ILUSTRO and CLARITY-Gastric 01 (trials targeting claudin 18.2 positive disease in the unresectable/metastatic setting), ATTRACTION-6 (a negative trial of first-line ipilimumab and nivolumab plus chemotherapy in unresectable gastroesophageal adenocarcinoma), and ONO-4578 (a prostaglandin receptor antagonist combined with nivolumab and chemotherapy, which showed positive phase 2 results and is now advancing to phase 3).

For Patients

Treatment for stomach and esophageal cancers that can be surgically removed is changing. New research shows that adding immunotherapy, medication that helps the immune system attack cancer, to standard chemotherapy before and after surgery meaningfully improves how the cancer responds and how long patients live afterward. Most people tolerate the combination about as well as chemotherapy by itself, but immunotherapy adds a risk of immune-related side effects that require close monitoring. This approach is becoming a new standard for many patients with these cancers.

For stomach and esophageal cancers that have already spread and test positive for a protein called HER2, a new bispecific antibody drug is showing promise for extending survival beyond what has previously been possible. This drug, zanidatamab, represents meaningful progress for patients who have had limited options.

One important message from this research: a cancer's genetic and protein makeup can change over treatment, which is why re-testing a tumor when the cancer progresses, rather than relying on older results, is becoming standard practice. Anyone being treated for stomach or esophageal cancer should ask their oncologist whether their tumor has been tested for these markers and whether newer combinations might apply.

Key Takeaways

  • MATTERHORN and NeoSUMMIT-01 both showed perioperative immunotherapy more than doubles pathologic complete response and significantly improves survival in resectable gastric/GEJ adenocarcinoma, establishing perioperative chemo-immunotherapy as the new standard of care.
  • NeoSUMMIT-01's 3-year update showed event-free survival of 75% with toripalimab added versus 56% with chemotherapy alone, and overall survival of 81% versus 72% (hazard ratio 0.45).
  • Optimal patient selection for perioperative immunotherapy (PD-L1 scoring method and cutoff) and optimal treatment duration remain open questions across both trials.
  • HERIZON-GEA-01 showed zanidatamab plus chemotherapy, with or without tislelizumab, significantly improved PFS and OS over trastuzumab plus chemotherapy in HER2-positive metastatic gastroesophageal adenocarcinoma (OS 26.4 vs. 19.2 months with the triplet).
  • Diarrhea occurred in roughly 80% of zanidatamab-treated patients and requires mandatory prophylaxis.
  • Up to 60% of HER2-positive tumors lose HER2 expression after trastuzumab exposure, making re-biopsy at disease progression important before selecting next-line HER2-targeted therapy.

References

  1. Hoeppner J, Brunner T, Schmoor C, et al. Perioperative Chemotherapy or Preoperative Chemoradiotherapy in Esophageal Cancer (ESOPEC) New England Journal of Medicine, 2025.
  2. Janjigian YY, Al-Batran SE, Wainberg ZA, et al. Perioperative Durvalumab in Gastric and Gastroesophageal Junction Cancer (MATTERHORN). New England Journal of Medicine, 2025; final overall survival analysis, European Society for Medical Oncology, 2025.
  3. Nie R-C, Jin Y, Liang C-C et al. Perioperative Toripalimab Plus Chemotherapy versus Chemotherapy Alone in Locally Advanced Gastric or Gastroesophageal Junction Cancer: 3-Year Follow-Up of NEOSUMMIT-01 Trial. Journal of Clinical Oncology, 2026.
  4. Yuan S-Q, Nie R-C, Jin Y et al. Perioperative toripalimab and chemotherapy in locally advanced gastric or gastro-esophageal junction cancer: a randomized phase 2 trial. Nature Medicine, 2024.
  5. Shitara K, Elimova E, Liu T, et al. Zanidatamab with and without Tislelizumab in HER2-Positive Gastroesophageal Cancer (HERIZON-GEA-01). New England Journal of Medicine, 2026.