Introduction

Esophageal cancer is a common, aggressive malignancy with high global incidence and mortality. According to GLOBOCAN estimates, over 510,000 new cases and more than 450,000 deaths occurred in 2022, making it the seventh leading cause of cancer deaths worldwide.1 In the US, over 22,000 new cases and 16,000 deaths are projected for 2025.2 Survival remains dismal with only 5% of patients with metastatic disease surviving five years.2 Clinical presentation often includes symptoms of esophageal luminal narrowing such as dysphagia, weight loss, odynophagia, and regurgitation.3 Physical findings of advanced disease may include lymphadenopathy, hoarseness from laryngeal nerve involvement, pleural effusions, ascites, hepatomegaly, bony tenderness, or upper gastrointestinal (GI) bleeding.4 Esophageal cancer includes two main histologic subtypes: esophageal adenocarcinoma (EAC), more common in Western countries and linked to Barrett’s esophagus (BE);5 and esophageal squamous cell carcinoma (ESCC), more common in Asia and associated with tobacco and alcohol use.6 Anatomically, esophageal cancers are classified as upper, middle, or lower.7 Distal tumors near the gastroesophageal junction (GEJ) are further categorized by Siewert classification.8 Type I and II Siewert tumors, located within 5 cm proximal or 2 cm distal to the GEJ, are treated as esophageal cancers based on American Joint Committee on Cancer9 and National Comprehensive Cancer Network10 guidelines. The anatomic and histologic variability of esophageal cancers contributes to the complexity of the disease.

Tumor heterogeneity drives treatment resistance and contributes to variable long-term therapeutic responses.11–14 Molecular studies14,15 have shown that ESCC and EAC are distinct at the genomic and transcriptomic levels.16 ESCCs display CCND1, TP63, SOX2 amplifications, and KDM6A deletions. Mid and distal EACs are often enriched for TP53 mutations, ERBB2 and VEGFA amplifications, and can be classified as the chromosomal instability (CIN) subtype. Further, distally into the stomach, additional molecular subtypes appear, including Epstein-Barr virus (EBV), microsatellite instability (MSI), and genomic stability (GS).16–18 These findings underscore the genomic diversity of esophageal cancers while also highlighting the molecular similarities between esophageal, GEJ, and gastric adenocarcinomas. Such overlap is reflected in clinical practice, where data from gastric cancer trials are often extrapolated to guide treatment of EAC and GEJ adenocarcinomas.

The emergence of predictive and prognostic biomarkers has enabled practical subclassification of esophageal cancers and personalized treatment. All newly diagnosed patients should be tested for programmed death-ligand 1 (PD-L1) and microsatellite instability (MSI) or mismatch repair (MMR) status, while patients with advanced disease should also be tested for HER2/neu and claudin-18.2 (CLDN18.2).19 This review focuses on biomarker-driven systemic therapies for esophageal cancers, with emphasis on EAC given its rising incidence in Western populations. We include phase II–III clinical trials evaluating biomarker-guided systemic therapies, primarily in advanced metastatic disease, detailing key efficacy and safety outcomes, Table 1. Here, we also highlight emerging biomarker-driven treatments supported by clinical efficacy data such as antibody-drug conjugates, novel monoclonal antibodies, and cellular therapies, Table 2. Finally, we examine key knowledge gaps that must be addressed to improve future esophageal cancer treatment, Figure 1. Our work is timely given the changing epidemiology,20,21 emergence of new targeted therapies,22 and improved molecular understanding of these diseases at unprecedented resolution12,23,24 that will continue to herald treatment advances.

Figure 1 Gaps in knowledge and future directions. Schematic reflecting key gaps in knowledge and areas of research for esophageal cancer. Highlighting: 1) elucidation of tumor heterogeneity with multi-omic analyses at bulk and single-cell spatially-resolved resolutions, 2) improving biomarkers of treatment efficacy, 3) targeting novel tumor intrinsic vulnerabilities, 4) overcoming emerging resistance mechanisms to new treatments, and 5) developing novel cancer prevention strategies. Created with https://BioRender.com.

Table 1 Current Treatments for Esophageal Cancers

Table 2 Future Treatments for Esophageal Cancers

Current Therapeutic Landscape for Esophageal Cancers HER2-Overexpressing Esophageal Adenocarcinomas

HER2/neu, a member of the human epidermal growth factor receptor family associated with tumor proliferation, apoptosis, adhesion, migration, and differentiation, is overexpressed in 10% to 30% of gastroesophageal adenocarcinomas.75,76 HER2-positivity is defined as immunohistochemistry (IHC) 3+ or fluorescence in situ hybridization (FISH) positive (HER2:CEP17 ratio ≥2).25 The ToGA trial established the efficacy of adding trastuzumab, a monoclonal antibody against HER2, to chemotherapy in HER2-expressing advanced gastric or GEJ adenocarcinomas.77 Trastuzumab plus chemotherapy improved median overall survival (OS) from 11.1 to 13.8 months and progression-free survival (PFS) from 5.5 to 6.7 months. The KEYNOTE-811 trial showed that adding pembrolizumab, a programmed cell death-1 (PD-1) inhibitor, to trastuzumab and chemotherapy further improved OS in advanced gastric or GEJ adenocarcinoma patients with PD-L1 combined positive score (CPS) ≥1 (20.1 vs 15.7 months).28,29 For second-line and beyond treatment, trastuzumab deruxtecan, a HER2 antibody–drug conjugate (ADC), improved survival in gastroesophageal adenocarcinomas.27,78,79 Specifically, the DESTINY-Gastric04 trial showed a median OS benefit over ramucirumab plus paclitaxel (14.7 vs 11.4 months).27 Notably, interstitial lung disease and pneumonitis occurred in 15% of patients treated with trastuzumab deruxtecan.80

PD-L1-Positive Esophageal Cancers

Immune checkpoint inhibitors targeting the immune inhibitory PD-1 receptor81 have become frontline treatment for gastroesophageal cancers. Efficacy depends on PD-L1 expression, assessed by IHC-based tumor proportion score (TPS) or CPS.82–84 For advanced ESCCs, the utility of immunotherapy has been shown in the CheckMate 648 trial.30 This phase III trial compared nivolumab (an anti–PD-1 antibody) plus chemotherapy, nivolumab plus ipilimumab (an inhibitor of another immunologic synapse termed cytotoxic T lymphocyte antigen 4 [CTLA-4]), and chemotherapy alone. In patients with PD-L1 ≥1, OS improved in both immunotherapy arms. Nivolumab plus chemotherapy extended OS to 15.4 months versus 9.1 months with chemotherapy alone; dual immunotherapy yielded 13.7 months. For EACs (and gastric/GEJ cancers), the CheckMate 649 trial31,85,86 evaluated nivolumab plus chemotherapy versus chemotherapy alone, demonstrating improved OS (14.4 vs 11.1 months) and PFS (7.7 vs 6.0 months) among patients with PD-L1 CPS ≥5.

Additional PD-1 inhibitors have also shown efficacy in esophageal cancers. The KEYNOTE-590 trial evaluated pembrolizumab plus chemotherapy in advanced ESCCs (73%) and EACs.32 In patients with PD-L1 CPS ≥10, pembrolizumab improved OS (13.5 vs 9.4 months) and PFS (6.3 vs 5.8 months). KEYNOTE-859 showed similar benefits in PD-L1 CPS ≥10 gastric and GEJ adenocarcinomas, with OS of 15.7 vs 11.8 months and PFS of 8.1 vs 5.6 months.33 Tislelizumab, a PD-1 inhibitor with reduced Fc gamma receptor binding, improved survival in ESCC when combined with chemotherapy in the RATIONALE-306 trial.34 Patients receiving tislelizumab and chemotherapy achieved longer OS compared with chemotherapy alone (17 vs 10 months) in those with PD-L1 CPS ≥1, as well as by Tumor Area Positivity (TAP) score, an alternative measure of PD-L1 expression.87 In the RATIONALE-305 trial, tislelizumab plus chemotherapy also showed efficacy in frontline treatment of gastric and GEJ adenocarcinomas.35 Its side effect profile is similar to other PD-1 inhibitors.34,35

While these trials collectively establish PD-1/PD-L1 blockade as a key treatment strategy, direct comparisons across studies are limited by differences in trial design including backbone chemotherapy regimen, PD-L1 scoring systems, histologic compositions and regional patient populations. To this point, multiple studies have attempted to better correlate PD-L1 levels and immunotherapy efficacy, and to elucidate potential discrepancies seen between studies.88,89 Addition of immunotherapy to chemotherapy is likely to benefit patients with tumors that have high PD-L1 expression (CPS ≥10) and is unlikely benefit patients with low/negative expression (CPS < 1). The data is less clear for patients with intermediate PD-L1 expression levels. Although a subset appears to derive clinical benefit—particularly among those with squamous cell carcinoma—better predictive biomarkers are needed to identify these patients.

CLDN18.2-Positive Esophageal Adenocarcinomas

CLDN18.2, a splice variant of a tight junction protein in normal gastric epithelium, is frequently overexpressed in subsets of gastroesophageal adenocarcinomas.22 Zolbetuximab, a monoclonal antibody against CLDN18.2, promotes antibody-dependent cellular cytotoxicity (ADCC) and complement-dependent cytotoxicity.90 CLDN18.2 positivity is defined by immunohistochemistry as strong (≥2+) membrane staining in ≥75% of adenocarcinoma cells.91 The SPOTLIGHT trial enrolled patients with CLDN18.2-positive, HER2-negative gastric or GEJ adenocarcinomas and compared zolbetuximab plus chemotherapy to chemotherapy alone. Zolbetuximab improved OS (18.23 vs 15.54 months) and PFS (10.61 vs 8.67 months).38 The GLOW trial similarly assessed zolbetuximab with chemotherapy, showing improved OS (14.39 vs 12.16 months) and PFS (8.21 vs 6.80 months) over chemotherapy alone.39 Common adverse events (AEs) included nausea (82%), vomiting (67%), and decreased appetite (47%).92

Deficient MMR/MSI-High Esophageal Cancers

Deficient mismatch repair (dMMR) or microsatellite instability-high (MSI-H) gastroesophageal cancers have impaired DNA repair, leading to high mutation rates and genomic instability.93,94 They represent about 10% of gastroesophageal cancers.16,17,95,96 Immune checkpoint inhibitors have shown benefit in this subset based on retrospective data from phase III trials. In CheckMate 649, subgroup analysis of 44 dMMR patients showed improved OS with nivolumab plus chemotherapy vs chemotherapy alone (38 vs 12 months).31 Similarly, in KEYNOTE-859, 39 MSI-H patients had improved outcomes with pembrolizumab plus chemotherapy.33 Immunotherapy alone may also be effective. In CheckMate 649, nivolumab plus ipilimumab improved OS vs chemotherapy in 21 MSI-H patients (OS not reached vs 10 months).31 Pembrolizumab monotherapy also showed durable benefit in dMMR esophagogastric cancer based on exploratory analysis of the KEYNOTE-062, −059, and −061 trials.40

Angiogenesis-Targeted Therapies

Targeting the vascular endothelial growth factor (VEGF) and its receptor (VEGFR) is an important second-line approach for advanced gastroesophageal cancers. Ramucirumab, a monoclonal antibody against VEGFR-2, has shown efficacy as monotherapy and combined with chemotherapy. The REGARD trial randomized previously treated advanced gastric or GEJ adenocarcinoma patients to ramucirumab or placebo.43 Ramucirumab improved OS (5.2 vs 3.8 months) and PFS (2.1 vs 1.3 months). The RAINBOW trial assessed ramucirumab plus paclitaxel vs paclitaxel alone and showed improved OS (9.6 vs 7.4 months) and PFS (4.4 vs 2.9 months).44 However, targeting this pathway has not shown efficacy in the frontline setting,97,98 and current treatments are not biomarker driven.99,100

Locally Advanced Disease

While our focus has been on advanced metastatic EACs, recent shifts in treatment for locally advanced esophageal cancers are noteworthy. Previously, these cancers (EACs and ESCCs) were treated with preoperative chemoradiation using carboplatin and paclitaxel (the CROSS regimen).101 Peri-operative FLOT chemotherapy (5-FU, leucovorin, oxaliplatin, and docetaxel), the standard for gastric and GEJ adenocarcinomas,102 has recently shown superiority over CROSS for esophageal adenocarcinomas.103 The ESOPEC trial demonstrated improved OS and PFS with peri-operative chemotherapy compared to neoadjuvant chemoradiation.

Checkpoint inhibitors have also shown utility in potentially curative, locally advanced esophageal cancers. Nivolumab improved disease-free survival compared to observation (22.4 vs 11.0 months) when given after neoadjuvant chemoradiotherapy in patients with pathologic residual disease at surgery.104 More recently, durvalumab, a PD-L1 inhibitor, showed benefit in the perioperative setting. The MATTERHORN trial evaluated durvalumab with FLOT chemotherapy in resectable gastric and GEJ adenocarcinomas, showing improved two-year event-free survival (67.4% vs 58.5%) and higher pathologic complete response (pCR) rates (19.2% vs 7.2%).36 Subgroup analyses from both studies suggest benefit may be greater in PD-L1-expressing tumors.36,105 In contrast, pembrolizumab (KEYNOTE-585)106,107 and nivolumab (ATTRACTION-5)108 added to chemotherapy have not shown clinical efficacy, and the DANTE trial is evaluating atezolizumab, another PD-L1 antibody, in this setting.37 Notably, the subgroup in KEYNOTE-585 receiving FLOT (rather than cisplatin-based doublet) chemotherapy with pembrolizumab had outcomes similar to those in MATTERHORN.106,107

Neoadjuvant or peri-operative immunotherapy is being actively studied for localized dMMR/MSI-H esophageal cancers, with emerging evidence of clinical benefit. In KEYNOTE-585, adding pembrolizumab to peri-operative FLOT chemotherapy led to a 38.9% pCR rate.106,107 Interim results from the DANTE trial showed a 63% pCR rate in the MSI-H subgroup with atezolizumab plus FLOT.37 Immunotherapy alone is also under investigation, in the ongoing NEONIPIGA trial,41 peri-operative nivolumab and ipilimumab treatment resulted in a 58.6% pCR rate. Similarly, the INFINITY trial reported a 60% pCR rate with the combination of tremelimumab (CTLA-4 antibody) and durvalumab.42 These preliminary results, drawn from small cohorts, require confirmation in phase III trials before routine use in clinical practice.

Promising Future Therapies for Esophageal Cancers

Despite recent advances, prognosis in advanced esophageal cancer remains poor, and new effective therapies are needed. Strategies under investigation include adoptive cell therapies, ADCs, and novel approaches to targeting the TME, including new checkpoint inhibitors. Our discussion will focus on modalities with efficacy data, Table 2. While two-drug chemotherapy regimens have been the standard, three-drug regimens like FOLFIRINOX (5-FU, leucovorin, irinotecan, oxaliplatin) have shown clinical efficacy.109 In HER2-negative patients, FOLFIRINOX treatment achieved objective response rate (ORR) of 61%, PFS of 8.4 months, and OS of 15.5 months; in HER2-positive patients, adding trastuzumab improved outcomes (ORR 85%, PFS 13.8 months, OS 19.6 months). We will continue by discussing therapies by biomarker and mechanism of action.

HER2-Overexpressing Esophageal Adenocarcinomas

New approaches to blocking HER2 signaling are emerging for esophageal cancers. These include dual HER2 domain targeting bispecific antibodies such as zandidatamab.110 In a phase II trial, zandidatamab plus chemotherapy displayed an ORR of 76.2%, PFS of 12.5 months, and OS of 36.5 months.45 KN026 another HER2 bispecific antibody targeting two distinct HER2 domains,111 demonstrated an ORR of 56% in HER2 high expressors (IHC 3+ or 2+/FISH-positive) in a phase II trial for pre-treated gastric and GEJ adenocarcinomas.46

Another HER2 blockade strategy involves novel monoclonal antibodies. HLX22 is a new anti-HER2 antibody that binds a different epitope than trastuzumab.112 It was evaluated with a trastuzumab biosimilar, HLX02, and chemotherapy in a phase II trial.47 Fifty-three first-line metastatic HER2-positive gastric/GEJ patients were randomized to HLX22 plus HLX02 and chemotherapy, or placebo plus HLX02 and chemotherapy. HLX22 improved PFS to 15.1 versus 8.2 months. A phase III trial is underway to assess HLX22 with trastuzumab and chemotherapy with and without pembrolizumab.113

Given the emergence of trastuzumab deruxtecan, additional HER2-targeting ADCs are in development. Interest is also growing to target HER2-low gastroesophageal cancers (IHC 2+/FISH-negative or IHC 1+), which account for up to 25% of cases.114 The DESTINY-Gastric 01 trial analysis of HER2-low cohorts showed ORRs of 26.3% (IHC 2+/FISH-negative) and 9.5% (IHC 1+) with PFS of 4.4 months and 2.8 months, and OS of 7.8 months and 8.5 months, respectively.48 Disitamab vedotin (DV, RC48) is a HER2 ADC with a monomethyl auristatin E payload.115 In a phase I trial, DV plus anti-PD-1 toripalimab showed an ORR of 43%, PFS of 6.2 months, and OS of 16.8 months.51 Responses were seen in HER2-positive and HER2-low patients. Common grade 3 AEs were neutropenia and leukopenia. A phase II trial has tested combinations of DV, toripalimab, chemotherapy, and trastuzumab.52 For HER2-positive patients, DV with toripalimab and trastuzumab showed ORR of 82.4%, and for HER2-low patients, DV with toripalimab and chemotherapy showed ORR of 70.8% (76.9% with dose reduced chemotherapy). ARX788, another novel HER2 ADC utilizing a tubulin inhibitor payload showed ORR of 38%, PFS of 4.1 months, and OS of 10.7 months in heavily pretreated HER2-positive patients.49 Common AEs included dry eye, thrombocytopenia, and elevated liver enzymes.116

Cellular therapies targeting HER2 are emerging for gastroesophageal cancers. HER2-directed chimeric antigen receptor (CAR) T cells have shown preclinical promise,117 but still limited real clinical efficacy.118 HER2 vaccines are also in development. HER-Vaxx (IMU-131), a B-cell HER2 targeting vaccine showed improved OS of 13.9 versus 8.3 months when combined with chemotherapy in a phase II study.50

CLDN18.2-Positive Esophageal Adenocarcinomas

The approval of zolbetuximab for CLDN18.2 expressing gastroesophageal cancers has spurred additional interest in targeting this protein.119 Osemitamab (TST001), a CLDN18.2 antibody with higher binding affinity and ADCC than zolbetuximab, showed promising results combined with nivolumab and chemotherapy in the TranStar102 phase I/II study resulting in ORR of 68%, PFS of 16.6 months, and OS of 20.4 months.53 ASKB589, another CLDN18.2 antibody also showed early clinical efficacy combined with a PD-1 inhibitor and chemotherapy for CLDN18.2 expressing gastric/GEJ adenocarcinomas (ORR of 73.5%, 9-month PFS of 58.1%, and 12-month OS of 77.1%).54 CLDN18.2-targeting ADCs are also in development. CMG901 in the KYM901 trial showed efficacy for treatment refractory gastroesophageal cancers with CLDN18.2 membrane staining intensity of at least 2+ in at least 5% of cancer cells with ORR of 29%, PFS of 3.7 months, and of OS 10.1 months.55 Most responses (97%) occurred in tumors with ≥20% of cells showing expression. Common AEs were anemia, hypoalbuminemia, weight loss, nausea and vomiting. IBI343, another CLDN18.2 ADC, showed a similar ORR of 29% and PFS of 5.5 months in a phase I trial for treatment refractory CLDN18.2 high expressing gastric/GEJ adenocarcinoma patients.56 Cellular therapies against CLDN18.2 are emerging as well. Satricabtagene autoleucel (satricel, CT041), an autologous CAR-T cell therapy targeting CLDN18.2,120 improved PFS (3.25 vs 1.77 months) and OS (7.92 vs 5.49 months) versus physician’s choice therapy in the phase II CT041-ST-01 trial.57 Grade ≥3 cytopenias occurred in 99% and cytokine release syndrome in 95% of satricel-treated patients.

FGFR2-Overexpressing Esophageal Adenocarcinomas

Fibroblast growth factor receptor (FGFR) signalings are frequently altered in cancers.121FGFR2 amplifications occur in 4–7% of gastric cancers122 and FGF2b overexpression is seen in 4–30% of gastroesophageal cancers.123,124 Bemarituzumab, an FGFR2b monoclonal antibody was assessed with chemotherapy in gastroesophageal cancer patients in the phase II FIGHT study.58 Eligibility was based on FGFR2b overexpression (IHC 2+/3+ in >0% of tumor cells) or FGFR2 amplification detected by circulating tumor DNA (ctDNA) sequencing. Bemarituzumab and chemotherapy improved PFS (9.5 vs 7.4 months) and OS (19.2 vs 13.5 months) compared with chemotherapy with greater benefit in tumors overexpressing FGFR2b in ≥10% of cells.59 Corneal adverse events were commonly seen in the bemarituzumab treated arm (77.4%). The ongoing phase III FORTITUDE-101 trial has met its interim OS endpoint.60 The benefit is therefore limited to selected patients with FGFR2b overexpression and ocular toxicity requires careful management.

TROP2-Overexpressing Esophageal Cancers

Trophoblast cell surface antigen 2 (TROP2), a calcium-transducing transmembrane protein, is overexpressed in 50–60% of gastric cancers125,126 with similar rates in EACs127 and ESCCs.128 Sacituzumab tirumotecan (SKB264/MK2870) is a TROP2-directed ADC with a topoisomerase I payload.129 Preliminary results from the phase II KL264-01 trial demonstrated early efficacy in treatment-refractory gastric and GEJ adenocarcinomas.61 The ORR was 22.0%; in patients with ≥3 prior treatments, PFS was 3.7 months and OS was 7.6 months. Patients with medium/low TROP2 expression by immunohistochemistry showed an ORR of 21.7%, while high expressing patients had an ORR of 25.0%. Common AEs were anemia, decreased white blood cells, and neutropenia.

Targeting the Tumor Microenvironment PD-1/PD-L1 and Beyond

The emergence of PD-1 inhibitors (nivolumab, pembrolizumab, tislelizumab) and PD-L1 inhibitors (durvalumab) have prompted continued interest in targeting the TME in gastroesophageal cancers, especially ESCC. In the ESCORT-1st trial, camrelizumab, a PD-1 monoclonal antibody, plus chemotherapy improved OS (15.3 vs 12.0 months) and PFS (6.9 vs 5.6 months) over chemotherapy for metastatic ESCC patients with benefits seen regardless of PD-L1 expression.62 Most common immune-related AE was reactive capillary endothelial proliferation. Similar outcomes were observed for additional PD-1 inhibitors when combined with chemotherapy compared to chemotherapy alone: in the JUPITER-06 trial with toripalimab (OS 17.0 vs 11.0 months),63 and in the ORIENT-15 trial with sintilimab (OS 16.7 vs 12.5 months).64 Finally, the GEMSTONE-304 trial showed that the PD-L1 inhibitor, sugemalimab, improved OS to 15.3 from 11.5 months when added to chemotherapy in treatment-naïve advanced ESCC patients.65

Beyond PD-1/PD-L1, the tumor antigen, B7-H3 (CD276) has emerged as a novel target.130 It is overexpressed in many cancers, including 45% of ESCCs,131 and implicated in tumor progression and immune evasion.132 YL201, a B7-H3-targeting ADC,133 showed ORR of 25% and PFS of 4.8 months in a phase 1/1b trial of heavily pretreated ESCC patients.66 Common grade ≥3 AEs were neutropenia, leukopenia, and anemia. Phase III trials assessing this ADC have begun for other tumor types, but not yet for esophageal cancers.

CTLA-4 is another immune checkpoint of interest,134 particularly in combination with PD-1 blockade as shown in CheckMate 648 and 649 (discussed above). Cadonilimab a tetravalent anti-PD-1 and CTLA-4 bispecific antibody that is Fc-null for improved safety135 was evaluated in the Chinese-centered COMPASSION-15 trial in untreated gastric/GEJ adenocarcinoma patients.67 Cadonilimab plus chemotherapy improved OS to 14.1 vs 11.1 months compared to chemotherapy. In PD-L1 CPS ≥5 patients, OS was further improved to 15.3 compared to 10.9 months for chemotherapy alone. Common AEs were cytopenias and grade 1/2 hypothyroidism. Efficacy of cadonilimab internationally and compared to chemotherapy and PD-1 blockade will need to be addressed.

Lymphocyte-activation gene 3 (LAG-3) is an inhibitory receptor often co-expressed with PD-1 on T-cells.136 Dual inhibition of LAG-3 and PD-1 has been found preclinically to have anti-tumor activity,137 and LAG-3 has been found to be expressed in 40–50% of untreated gastroesophageal cancers.138 In the RELATIVITY-60 study, combining relatlimab, a LAG-3 blocking antibody, with nivolumab and chemotherapy showed no OS benefit in patients with ≥1% LAG-3 expression.68 Exploratory analyses showed a possible efficacy signal in patients with ≥5% LAG-3 expression and in patients with ≥1% LAG-3 expression combined with PD-L1 expression positivity (≥1 and <5). In neoadjuvant settings, relatlimab plus nivolumab with chemoradiation yielded 21.4% pCR and 93.8% 2-year OS vs 40% pCR and 75% 2-year OS with nivolumab alone.69 Favezelimab, another anti-LAG-3 antibody, when combined with pembrolizumab in pre-treated advanced gastric cancer patients showed modest activity (ORR 11.3%) with better outcomes in PD-L1-positive patients.70 These results indicate gastroesophageal cancer patients may benefit from dual LAG-3 and PD-1 blockade, but better biomarkers are needed to select for these subsets of patients.

The lymphocyte inhibitor receptor, T cell immunoglobulin and ITIM domain (TIGIT), has also emerged as a new target to restore and augment anti-tumor responses combined with PD-1 blockade.139 Rilvegostomig (AXD2936), a bispecific TIGIT and PD-1 antibody,140 showed promising results in the GEMINI-Gastric phase II trial.71 The combination of rilvegostomig with chemotherapy achieved an ORR of 67.5% (81.3% for CPS ≥5) in metastatic gastric/GEJ adenocarcinomas. Common AEs were nausea, neutropenia, and decreased appetite. Domvanalimab (AB154), a Fc-silent anti-TIGIT antibody, has been combined with zimberelimab (anti-PD-1 antibody) and chemotherapy in the EDGE-Gastric study.72 ORR was 59% and PFS was 12.9 months; in patients with high PD-L1 expression (≥5% TAP score) ORR was 69% with PFS of 13.8 months.141 Neutropenia, nausea, and anemia were the most common treatment-related AEs. The phase III STAR-221 trial is comparing this regimen to nivolumab with chemotherapy.142 The TME includes immune targets beyond lymphocytes. Dickkopf-related protein (DKK1) is a secreted protein that controls tumor growth, is overexpressed in 50–70% of gastroesophageal adenocarcinomas,143,144 and suppresses antitumor immune responses through regulation of tumor associated macrophages.145 DKN-01, a DKK1 blocking antibody, was assessed in the phase II DisTinGuish trial with tislelizumab and chemotherapy.73 The combination achieved ORR 73%, PFS of 11.3 months and OS of 19.5 months. DKK1 expression was assessed using an RNAscope chromogenic in situ hybridization (CISH) assay, and a quantitate H-score was determined with ≥35 defined as DKK1-high. In patients with DKK1-high expression, ORR increased to 90%. Common AEs were fatigue, diarrhea, and neutropenia. A randomized trial is ongoing assessing the efficacy of adding DKN-01 to PD-1 blockade and chemotherapy.

During gastrointestinal cancer progression and treatment resistance, dysregulation of inflammatory toll-like receptor and interleukin-1 receptor activates the pro-survival and immunosuppressive NF-κB pathway.146,147 Interleukin-1 Receptor Associated Kinase-4 (IRAK4) is a key regulator of this pathway and required for carcinogenesis.148 Emavusertib (CA-4948), a small molecule IRAK4 inhibitor originally developed for hematologic malignancies,149 is currently being evaluated in combination with chemotherapy for patients with untreated advanced gastroesophageal cancers.74

Gaps in Knowledge and Future Directions Elucidating Tumor Heterogeneity

The difficulty in treating gastroesophageal cancers remains intratumoral and TME heterogeneity. Variability in biomarkers like HER2 and PD-L1 across tumor regions, over time, and after therapy complicates treatment;150–152 overcoming this requires integrated higher-resolution analyses. The National Cancer Institute’s Clinical Proteomic Tumor Analysis Consortium (CPTAC) is advancing proteogenomic characterization using multi-omics to link genomic mutations with the transcriptome, epigenome, microbiome and proteome.153 Beyond “bulk” analysis, single-cell and spatial sequencing now apply multi-omics at cellular resolution.154 Tracking esophageal cancer at single-cell resolution could personalize treatment. Targeted RNA-sequencing marks a shift toward clinical use of additional “omics” approaches that may provide clinically meaningful information beyond genomic profiling.155

Improving Biomarkers

To find more clinically significant subtypes, better predictive and prognostic biomarkers are needed. Liquid biopsies and ctDNA, extracellular DNA from cancer cells,156 offer real-time cancer detection and monitoring. CtDNA quantification may enable early detection and predict recurrence postoperatively.157–159 It has shown strong concordance with tissue profiling in advanced gastroesophageal cancers.160 Furthermore, longitudinal ctDNA monitoring during treatment is emerging as a way to personalize therapy sequencing.26,161–164 Advanced technologies have enhanced the sensitivity of liquid biopsy “omics”,165,166 but clinical trials are needed to define their applicability in clinical practice.

PD-L1 expression assessed via CPS (as well as TPS and TAP) remains the best surrogate for immunotherapy response in gastroesophageal cancers.88 However, IHC scoring is subjective and relies on a single marker to represent a complex immune environment. In addition, as discussed above, better predictive biomarkers are needed to identify patients who will benefit from immunotherapy among the patients with intermediate PD-L1 expression. Single-cell multi-omics have identified six distinct ecotypes (EC1-6) in gastroesophageal cancers167 with biological and clinical implications.168 EC3 and EC6, predominant in gastric cancers, comprise CD4/8 T, natural killer, and dendritic cells; and stromal and B cells, respectively. These ecotypes, particularly within the CIN group, further stratify survival, and clinically identifying ecotypes may improve selection for specific immunotherapy combinations.

Finally, improved biomarkers to guide, sequence and integrate current targeted therapies are also needed, especially for patients with HER2 negative, and dual CLDN18.2 and PD-L1 positive (CPS≥5) gastroesophageal adenocarcinomas. These double expressing tumors comprise almost half of all CLDN18.2 positive gastroesophageal adenocarcinomas.169 Cross trial comparisons suggest that specific subsets of patients may derive differential benefit as GEJ adenocarcinomas and non-Asian subgroups showed less clinical benefit.38,39 While the addition of zolbetuximab improved OS and PFS, there was no improvement in response rates when compared to immunotherapy trials.38,39 Multiple trials, Table 2, are underway testing the combination of targeting CLDN18.2 and immunotherapy including zolbetuximab and pembrolizumab170 that will lead to better biomarker driven guidance on optimal treatment selection.

Targeting Novel Tumor Intrinsic Vulnerabilities

Early TP53 mutations drive genomic and phenotypic heterogeneity in gastroesophageal cancers and are early events in BE to EAC progression and ESCC carcinogenesis.171–173 Targeting p53 loss-of-function is appealing but difficult due to mutation diversity.174 Eprenetapopt (APR-246) binds mutated p53 cysteine residues triggering cell death.175 Rezatapopt (PC14586) targets TP53 Y220C mutations to reactivate p53 function,176 while ELX-02, a small-molecule ribosomal selective glycoside, restores translation of TP53 in TP53 nonsense mutated cancers.177 Wildtype p53 epitopes can also activate cytotoxic T lymphocytes using dendritic vaccines to target p53 mutated tumors.178 Though mostly pre-clinical, early trials show promise for these strategies. For example, the PYNNACLE trial assessing rezatapopt in TP53 Y220C tumors has shown a 33.3% ORR.179

Efforts to target mutated p53—a protein once considered “undruggable”—have encouraged the pursuit of other undruggable proteins including KRAS, which is mutated (3%) or amplified (14%) in gastroesophageal cancers.180,181 Common mutations include G12D, G12V, and non-codon 12/13 variants.180,182 Preclinical data show KRAS amplified gastric cancer models respond to SHP2, MEK, and CDK4/6 inhibition.181,183 In addition, BI 3706674, a small molecule inhibitor of the KRAS mutations and amplifications, has shown activity in tumors with KRAS G12V and copy number gains >10.184 As technology advances, more protein targets are becoming druggable. One such emerging technology is proteolysis-targeting chimeras (PROTACs), which harness the cell’s own ubiquitin-proteasome system to tag specific proteins with polyubiquitin chains, marking them for targeted proteasomal degradation.185 While the challenges and applications of PROTACs are numbers,185–187 preclinical development in gastroesophageal cancers has focused on targets such as FGFR2,188 VEGFR-2,189 and the histone demethylase, KDM4.190

While p53 and KRAS are common targets in solid tumors, efforts are underway to identify gastroesophageal-specific targets. Glycosylation aberrancies are common and may offer novel therapeutic avenues.191 Mucin 1 (MUC1), α-fetoprotein (AFP) and carcinoembryonic antigen (CEA) have been assessed as glycoprotein targets for CAR-T cells.192 Glycoprotein biology also drives carcinogenesis. Carbohydrate antigen 19–9 (CA19-9) induces pancreatitis and promotes pancreatic cancer with KRAS mutations,193 and Mucin 6 (MUC6) loss induces gastric inflammation leading to cancer.194 Abnormal glycosylation also affects the TME, as most immune checkpoints are glycoproteins, suggesting potential impact on future immunotherapy strategies.195

Most EACs have TP53 mutations, ERBB2 and VEGFA amplifications, and are classified as CIN tumors.16 It has been recently recognized that CIN tumors harbor vulnerabilities that may be uniquely targeted.196 CIN cancers, due to aneuploidy, show mitotic spindle defects and are sensitive to kinesin motor inhibition.197 Inhibiting the kinesin motor, KIF18A, has shown preclinical efficacy198 and early trials with the oral KIF18A inhibitor, VLS-1488, in solid tumors including esophageal cancers are promising199 CIN tumors also activate the cGAS-STING immune pathway, and inhibiting this signaling may offer a novel treatment for chromosomally unstable esophageal cancers.200

Overcoming New Treatment Resistance

With advances in trastuzumab deruxtecan and zolbetuximab, understanding resistance mechanisms is critical. We will highlight a few areas, specifically, that will need to be addressed. The topic of immunotherapy resistance is beyond the scope of this review and has been well detailed.201–203 Resistance to HER2 blockade with trastuzumab is well understood in breast oncology,204 but strategies from breast cancer often do not translate to gastroesophageal cancers.205 Known breast cancer HER2 ADC resistance involves HER2 downregulation and altered trafficking, payload resistance, and activation of alternative signaling pathways, including PI3K/AKT/mTOR.206 Trastuzumab deruxtecan resistance may also be induced by changes in the tumor microenvironment and mutations in SLX4 a gene involved in DNA repair.207 Specific to gastroesophageal adenocarcinomas, CCNE1 amplifications that induce cell cycle dysregulation,208 human leukocyte antigens (HLAs) loss that may abrogate antibody-dependent cellular cytotoxicity, and upregulation of oxidative phosphorylation209 are reported mechanisms of trastuzumab deruxtecan resistance. Novel HER2 ADCs, including disitamab vedotin,210 and dual payload ADCs211 are being investigated to overcome resistance these mechanisms, and combination approaches targeting alternative signaling pathways or dual HER2 targeting may also improve outcomes. Similar studies are needed to uncover resistance to zolbetuximab with the best evidence showing CLND18.2 expression downregulation.212 Other suggested mechanisms may involve changes in the tumor microenvironment and compensatory growth pathways.213 Overall, a systematic understanding of resistance including target modulation, intracellular signaling alterations, tumor heterogeneity and payload-specific mechanisms are needed to optimize next-generation HER2 and CLDN18.2-guided therapies.

Preventing Esophageal Cancer Development

Finally, all esophageal adenocarcinomas develop from Barrett’s esophagus, making intestinal metaplasia a key target for prevention in western countries.214,215 While dysplastic BE is managed with endoscopic therapy, preventing progression of non-dysplastic BE is critical.216 The AspECT trial demonstrated proton pump inhibitors and aspirin reduce death and EAC/dysplasia,217 providing the best evidence for BE chemoprevention. Tamoxifen, a selective estrogen receptor modulator, has been associated with decreased upper GI intestinal metaplasias218 and altered gene expression in BE stroma.219 In mice, selumetinib, a MEK inhibitor, reversed Helicobacter pylori-induced gastric intestinal metaplasia.220 A phase I trial of trametinib, another MEK inhibitor, showed improvements in gastric oxyntic atrophy and intestinal metaplasia post-resection for early-stage gastric cancer patients.221 These therapeutic strategies that may treat BE and prevent EAC development warrant further study.

Conclusion

Biomarker-driven therapies have transformed the management of advanced esophageal adenocarcinoma, with agents such as trastuzumab deruxtecan and zolbetuximab demonstrating significant improvements in patient outcomes. PD-L1, HER2, CLDN18.2, and MSI/MMR status now guide personalized treatment selection, yet anatomic, histologic, and molecular heterogeneity in tumor biology and overlapping actionable biomarkers complicate optimal therapy sequencing. New integrative multi-omics approaches provide deeper insights into tumor-intrinsic and microenvironmental mechanisms, offering potential strategies to overcome resistance and refine therapeutic interventions. Continued efforts are needed to translate basic science into clinical practice, and unique features of gastroesophageal cancers including TP53 mutations, CIN molecular subtype, altered glycosylation, and intestinal metaplasia precursor lesions offer new exciting avenues for therapeutic intervention. Moving forward, the integration of new and more robust biomarkers with novel therapeutics will be essential to improve outcomes and guide precision-based management of esophageal cancers.

Funding

R.U.J. was supported by P30DK056338, P30DK052574, AGA2025-31-01 (AGA-Caroline Craig Augustyn & Damian Augustyn Award in Digestive Cancer), K08DK139376, and the BCM Chao Physician-Scientist Award.

Disclosure

Haeseong Park reports grants from Genetech, grants from Mirati/BMS, grants from Exelixis, grants from Tizona, grants from StrataPATH, grants from Bolt Biotherpeutics, grants from StrataPATH, grants from Chugai, grants from Huaota, grants from D3Bio, grants from Amgen, grants from Idience, grants from Incyte, grants from Pfizer, grants from Alterome, grants from Yuhan, personal fees from Merck, personal fees from Daiichi Sankyo, personal fees from Astellas, outside the submitted work. The authors report no other conflicts of interest in this work.

References

1. Bray F, Laversanne M, Sung H. et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2024;74(3):229–263. doi:10.3322/caac.21834

2. Siegel RL, Kratzer TB, Giaquinto AN, Sung H, Jemal A. Cancer statistics, 2025. CA Cancer J Clin. 2025;75(1):10–45. doi:10.3322/caac.21871

3. Puhr HC, Pablik E, Berghoff AS, et al. Viennese risk prediction score for advanced gastroesophageal carcinoma based on Alarm Symptoms (VAGAS score): characterisation of alarm symptoms in advanced gastro-oesophageal cancer and its correlation with outcome. ESMO Open. 2020;5(2). doi:10.1136/esmoopen-2019-000623

4. Jin RU, Navo K, Park H. Chapter 15: esophageal and Gastric Cancer. In: Ramaswamy G, Daniel M, Armin G, editors. The Washington Manual of Oncology. 4th ed. Wolters Kluwer; 2021.

5. Spechler SJ, Souza RF. Barrett’s esophagus. N Engl J Med. 2014;371(9):836–845. doi:10.1056/NEJMra1314704

6. Morgan E, Soerjomataram I, Rumgay H, et al. The global landscape of esophageal squamous cell carcinoma and esophageal adenocarcinoma incidence and mortality in 2020 and projections to 2040: new estimates from GLOBOCAN 2020. Gastroenterology. 2022;163(3):649–658e2. doi:10.1053/j.gastro.2022.05.054

7. Rice TW, Patil DT, Blackstone EH. 8th edition AJCC/UICC staging of cancers of the esophagus and esophagogastric junction: application to clinical practice. Ann Cardiothorac Surg. 2017;6(2):119–130. doi:10.21037/acs.2017.03.14

8. Siewert JR, Stein HJ. Carcinoma of the gastroesophageal junction - classification, pathology and extent of resection. Dis Esophagus. 1996;9(3):173–182. doi:10.1093/dote/9.3.173

9. Amin MB, American Joint Committee on Cancer, American Cancer Society. AJCC Cancer Staging Manual. 8th ed. American Joint Committee on Cancer, Springer; 2017:xvii,1024.

10. Ajani JA, D’Amico TA, Bentrem DJ, et al. Esophageal and esophagogastric junction cancers, Version 2.2023, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw. 2023;21(4):393–422. doi:10.6004/jnccn.2023.0019

11. Chao J, Bedell V, Lee J, et al. Association between spatial heterogeneity within nonmetastatic gastroesophageal adenocarcinomas and survival. JAMA Netw Open. 2020;3(4):e203652. doi:10.1001/jamanetworkopen.2020.3652

12. Dinh HQ, Pan F, Wang G, et al. Integrated single-cell transcriptome analysis reveals heterogeneity of esophageal squamous cell carcinoma microenvironment. Nat Commun. 2021;12(1):7335. doi:10.1038/s41467-021-27599-5

13. Li X, Francies HE, Secrier M, et al. Organoid cultures recapitulate esophageal adenocarcinoma heterogeneity providing a model for clonality studies and precision therapeutics. Nat Commun. 2018;9(1):2983. doi:10.1038/s41467-018-05190-9

14. Stachler MD, Jin RU. Molecular pathology of gastroesophageal cancer. Clin Lab Med. 2024;44(2):239–254. doi:10.1016/j.cll.2023.08.005

15. Zeng Y, Jin RU. Molecular pathogenesis, targeted therapies, and future perspectives for gastric cancer. Semin Cancer Biol. 2022;86(Pt 3):566–582. doi:10.1016/j.semcancer.2021.12.004

16. Cancer Genome Atlas Research N, Analysis Working Group: Asan U, Agency BCC, et al. Integrated genomic characterization of oesophageal carcinoma. Nature. 2017;541(7636):169–175. doi:10.1038/nature20805

17. Cancer Genome Atlas Research N. Comprehensive molecular characterization of gastric adenocarcinoma. Nature. 2014;513(7517):202–209. doi:10.1038/nature13480

18. Sohn BH, Hwang JE, Jang HJ, et al. Clinical significance of four molecular subtypes of gastric cancer identified by the cancer genome Atlas project. Clin Cancer Res. 2017;23(15):4441–4449. doi:10.1158/1078-0432.CCR-16-2211

19. Network NCC. NCCN clinical practice guidelines in oncology: esophageal and esophagogastric junction cancers. Available from: https://www.nccn.org/professionals/physician_gls/pdf/esophageal.pdf. Accessed June25, 2025.

20. Rodriguez GM, DePuy D, Aljehani M, et al. Trends in epidemiology of esophageal cancer in the US, 1975-2018. JAMA Netw Open. 2023;6(8):e2329497. doi:10.1001/jamanetworkopen.2023.29497

21. Thrift AP. Global burden and epidemiology of Barrett oesophagus and oesophageal cancer. Nat Rev Gastroenterol Hepatol. 2021;18(6):432–443. doi:10.1038/s41575-021-00419-3

22. Zeng Y, Lockhart AC, Jin RU. The preclinical discovery and development of zolbetuximab for the treatment of gastric cancer. Expert Opin Drug Discov. 2024;19(8):873–886. doi:10.1080/17460441.2024.2370332

23. Croft W, Evans RPT, Pearce H, Elshafie M, Griffiths EA, Moss P. The single cell transcriptional landscape of esophageal adenocarcinoma and its modulation by neoadjuvant chemotherapy. Mol Cancer. 2022;21(1):200. doi:10.1186/s12943-022-01666-x

24. Zhang X, Peng L, Luo Y, et al. Dissecting esophageal squamous-cell carcinoma ecosystem by single-cell transcriptomic analysis. Nat Commun. 2021;12(1):5291. doi:10.1038/s41467-021-25539-x

25. Ruschoff J, Hanna W, Bilous M, et al. HER2 testing in gastric cancer: a practical approach. Mod Pathol. 2012;25(5):637–650. doi:10.1038/modpathol.2011.198

26. Shitara K, Bang YJ, Iwasa S, et al. Trastuzumab deruxtecan in HER2-positive advanced gastric cancer: exploratory biomarker analysis of the randomized, phase 2 DESTINY-Gastric01 trial. Nat Med. 2024;30(7):1933–1942. doi:10.1038/s41591-024-02992-x

27. Shitara K, Van Cutsem E, Gumus M, et al. Trastuzumab deruxtecan or ramucirumab plus paclitaxel in gastric cancer. N Engl J Med. 2025. doi:10.1056/NEJMoa2503119

28. 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

29. Janjigian YY, Kawazoe A, Bai Y, et al. Pembrolizumab in HER2-positive gastric cancer. N Engl J Med. 2024;391(14):1360–1362. doi:10.1056/NEJMc2408121

30. Doki Y, Ajani JA, Kato K, et al. Nivolumab combination therapy in advanced esophageal squamous-cell carcinoma. N Engl J Med. 2022;386(5):449–462. doi:10.1056/NEJMoa2111380

31. Janjigian YY, Ajani JA, Moehler M, et al. First-line nivolumab plus chemotherapy for advanced gastric, gastroesophageal junction, and esophageal adenocarcinoma: 3-year follow-up of the phase III CheckMate 649 trial. J Clin Oncol. 2024;42(17):2012–2020. doi:10.1200/JCO.23.01601

32. Sun JM, Shen L, Shah MA, et al. Pembrolizumab plus chemotherapy versus chemotherapy alone for first-line treatment of advanced oesophageal cancer (KEYNOTE-590): a randomised, placebo-controlled, phase 3 study. Lancet. 2021;398(10302):759–771. doi:10.1016/S0140-6736(21)01234-4

33. Rha SY, Oh DY, Yanez P, et al. Pembrolizumab plus chemotherapy versus placebo plus chemotherapy for HER2-negative advanced gastric cancer (KEYNOTE-859): a multicentre, randomised, double-blind, phase 3 trial. Lancet Oncol. 2023;24(11):1181–1195. doi:10.1016/S1470-2045(23)00515-6

34. Xu J, Kato K, Raymond E, et al. Tislelizumab plus chemotherapy versus placebo plus chemotherapy as first-line treatment for advanced or metastatic oesophageal squamous cell carcinoma (RATIONALE-306): a global, randomised, placebo-controlled, phase 3 study. Lancet Oncol. 2023;24(5):483–495. doi:10.1016/S1470-2045(23)00108-0

35. Qiu MZ, Oh DY, Kato K, et al. Tislelizumab plus chemotherapy versus placebo plus chemotherapy as first line treatment for advanced gastric or gastro-oesophageal junction adenocarcinoma: RATIONALE-305 randomised, double blind, phase 3 trial. BMJ. 2024:385:e078876. doi:10.1136/bmj-2023-078876

36. Janjigian YY, Al-Batran SE, Wainberg ZA, et al. Perioperative durvalumab in gastric and gastroesophageal junction cancer. N Engl J Med. 2025. doi:10.1056/NEJMoa2503701

37. Lorenzen S, Gotze TO, Thuss-Patience P, et al. Perioperative atezolizumab plus fluorouracil, leucovorin, oxaliplatin, and docetaxel for resectable esophagogastric cancer: interim results from the randomized, multicenter, phase II/III DANTE/IKF-s633 trial. J Clin Oncol. 2024;42(4):410–420. doi:10.1200/JCO.23.00975

38. Shitara K, Lordick F, Bang YJ, et al. Zolbetuximab plus mFOLFOX6 in patients with CLDN18.2-positive, HER2-negative, untreated, locally advanced unresectable or metastatic gastric or gastro-oesophageal junction adenocarcinoma (SPOTLIGHT): a multicentre, randomised, double-blind, phase 3 trial. Lancet. 2023;401(10389):1655–1668. doi:10.1016/S0140-6736(23)00620-7