First-line programmed death-1 blockade plus chemotherapy in low programmed cell death ligand 1 esophageal squamous cell carcinoma: beyond a single biomarker

The study by Wu et al., “Clinical benefit of first-line programmed death-1 antibody plus chemotherapy in low programmed cell death ligand 1-expressing esophageal squamous cell carcinoma”, tackles a central and highly practical question in advanced esophageal squamous cell carcinoma (ESCC): should first-line programmed death-1 (PD-1) inhibitor plus chemotherapy be offered to patients whose tumors show low programmed cell death ligand 1 (PD-L1) expression (1)? Using patient-level data from the phase III JUPITER-06 trial and a meta-analysis of five randomized phase III trials (JUPITER-06, KEYNOTE-590, CheckMate 648, ESCORT-1st, ORIENT-15) (2-6), Wu et al. show that adding a PD-1 antibody to platinum-based chemotherapy improves overall survival (OS), progression-free survival (PFS) and objective response rate (ORR) even in tumors with low PD-L1 [tumor proportion score (TPS) <1% or combined positive score (CPS) <10] (1). In JUPITER-06, toripalimab plus paclitaxel/cisplatin significantly prolonged PFS and OS versus chemotherapy alone, and this benefit was preserved when patients were reclassified by TPS (≥1% vs. <1%) in addition to CPS (1). In their meta-analysis restricted to low PD-L1, pooled hazard ratios favored PD-1 inhibitor plus chemotherapy over chemotherapy alone for OS, PFS and ORR (1). The treatment effect in low PD-L1 cohorts was smaller than in PD-L1-high disease but not abolished, supporting the notion that PD-L1 behaves as a quantitative modifier of benefit rather than a binary predictive marker.

The conclusions of Wu et al. appear, at first glance, to diverge from the deliberations of the US Food and Drug Administration (FDA)’s Oncologic Drugs Advisory Committee (ODAC), which in 2024 evaluated class-wide data for PD-1 inhibitors in first-line unresectable or metastatic ESCC (7). In an integrated analysis of KEYNOTE-590, CheckMate 648 and RATIONALE-306 (2,3,8), the FDA briefing document emphasized that most of the OS benefit with anti-PD-1 therapy accrued in patients with higher PD-L1 expression (for example CPS or TPS ≥10), with intermediate benefit in PD-L1 1 to <10 and marginal or numerically unfavorable outcomes in PD-L1 <1. ODAC therefore concluded that the overall risk-benefit profile was not clearly favorable in ESCC with PD-L1 <1% (7). These observations are broadly consistent with the meta-analysis by Yap et al., who reconstructed survival curves from multiple ESCC trials and reported that the OS benefit of immune checkpoint inhibitors (ICIs) was largely confined to PD-L1-high subgroups, with no clear survival advantage in PD-L1-low tumors (9). Together, these datasets support an “enrichment” role for PD-L1 in ESCC: higher expression identifies patients more likely to derive meaningful benefit, whereas the magnitude—and even the presence—of benefit in PD-L1-low disease remains uncertain.

Several quantitative syntheses support a coherent model. Using reconstructed PD-L1 subgroups from phase III gastroesophageal cancer trials, Lee et al. showed that first-line ICI-containing regimens clearly improved OS when CPS ≥1, with progressively greater benefit at higher CPS and little or no benefit when CPS <1; in the ESCC subset, the largest relative benefit was seen in PD-L1 ≥10, intermediate benefit in PD-L1 1 to <10, and attenuated or absent benefit in PD-L1 <1 (10). Chen et al., in an ESCC-specific network meta-analysis, similarly demonstrated that PD-1 inhibitor plus chemotherapy improved OS and PFS in PD-L1-positive patients, with camrelizumab plus chemotherapy ranking highly in PD-L1 ≥10 and nivolumab-, pembrolizumab-, and sintilimab-based regimens also conferring benefit in PD-L1-positive groups (11). The narrative review by Zafar et al. in esophageal cancer more broadly supports PD-L1 as an enrichment biomarker—higher expression identifying patients more likely to benefit—while cautioning that PD-L1 should not be used as an absolute gatekeeper (12). Taken together with the patient-level analysis by Wu et al. (1), these data support a consistent model in ESCC: PD-L1 is a continuous quantitative modifier of benefit from PD-1-based therapy, but its predictive power becomes weak and statistically uncertain in the PD-L1 <1% extreme. The apparent divergence between Wu et al. and ODAC largely reflects differing tolerance for this uncertainty and different priorities—clinical versus regulatory—when interpreting modest, imprecisely estimated effects in small PD-L1-negative subgroups (1,7). Importantly, interpretation of treatment effects within PD-L1-low subgroups warrants explicit consideration of statistical uncertainty. Such strata are typically characterized by smaller sample sizes and reduced event counts, inevitably leading to wider confidence intervals and less stable hazard ratio estimates. Under these conditions, apparent variability of effect sizes across trials is statistically expected and should not be interpreted as evidence of biological inconsistency. Moreover, failure to achieve statistical significance does not necessarily indicate absence of treatment benefit, particularly when point estimates remain directionally favorable. This distinction is well recognized in the statistical literature on subgroup analyses and heterogeneity of treatment effects, where limited power and sampling variability can substantially influence inference (13,14). Between-trial heterogeneity—including differences in patient populations, PD-L1 assay methodologies, chemotherapy backbones, and regional characteristics—further complicates deterministic interpretation. Collectively, these considerations support a probabilistic rather than dichotomous interpretation of outcomes in PD-L1-low ESCC.

In JUPITER-06, PD-L1 was assessed centrally with the JS311 antibody, which shows good concordance with other therapeutic clones (5). Blueprint studies in lung cancer have demonstrated that, with careful calibration, PD-L1 assays can be harmonized at broad cutoff points (15). Nevertheless, these efforts also highlight the structural weaknesses of PD-L1 as a solitary biomarker: strong spatial heterogeneity between primary and metastatic sites and within tumors; temporal changes under therapy; scoring differences between TPS and CPS; and restricted biological scope, capturing only one axis of the cancer-immune interaction (16). Wu et al. therefore appropriately call for multiomic biomarker development in ESCC (1). PD-L1 negativity on a single biopsy does not equate to immunological ignorance, nor does PD-L1 positivity guarantee durable response—particularly when PD-1 inhibitors are combined with chemotherapy (17). Conceptually, PD-L1 expression is more appropriately viewed as a continuous biological variable rather than a binary classifier. Rigid cutoff thresholds may impose artificial distinctions on a graded and dynamic feature, potentially obscuring treatment-effect modulation. This limitation is well recognized in statistical methodology, where dichotomization of continuous variables reduces power and may distort clinical interpretation (18). Within this framework, lower PD-L1 expression is best understood as modifying—not negating—the probability of benefit from PD-1 blockade. Such a probabilistic interpretation has direct relevance for clinical decision-making, particularly in patient counseling and shared decision-making.

The observation that PD-1 inhibitor plus chemotherapy can benefit PD-L1-low ESCC is biologically plausible. Platinum-taxane chemotherapy can induce immunogenic cell death, releasing damage-associated molecular patterns and tumor neoantigens that promote dendritic cell maturation and CD8⁺ T-cell priming (19). Cytotoxic therapy may also transiently reduce regulatory T cells and myeloid-derived suppressor cells, shifting the balance toward antitumor immunity (19). These effects can convert immunologically “cold” tumors into more inflamed lesions, thereby attenuating the predictive value of baseline PD-L1. In addition, immune escape mechanisms beyond PD-1/PD-L1—such as human leukocyte antigen (HLA) class I loss, β2-microglobulin mutations, interferon signaling defects and TGF-β-driven stromal exclusion—can modulate response independently of PD-L1 (16,17,19-21). Some of these may be partially reversible or modifiable with chemotherapy through effects on vasculature, stroma and immune cell trafficking, potentially unmasking sensitivity to PD-1 blockade even in tumors classified as PD-L1-low (22). These considerations caution against equating PD-L1 negativity with the absence of targetable antitumor immunity, particularly in the context of combination therapy.

Given the limited predictive precision of PD-L1 alone, composite biomarker strategies are needed. Tumor mutation burden (TMB) and mutational signatures may capture tumor immunogenicity more comprehensively than PD-L1 (23,24) and help define APOBEC-driven ESCC subsets with heightened immunogenic potential (16,20,21). Detailed profiling of the immune microenvironment based on CD8+ T-cell density, spatial distribution and stromal architecture can distinguish “inflamed”, “immune-excluded” and “immune-desert” phenotypes, with inflamed phenotypes generally associated with better ICI outcomes and often providing information beyond PD-L1 expression (16,21). Genomic determinants of immune escape, including HLA class I loss of heterozygosity and alterations in antigen-presentation or interferon-signaling pathways, may identify tumors intrinsically resistant to ICIs (19,20). Dynamic biomarkers such as early clearance or major reduction of circulating tumor DNA (ctDNA) after initiation of chemo-immunotherapy can quantify depth of response and emerging resistance in ways that static PD-L1 cannot (16,25). Ultimately, a multi-parameter score integrating PD-L1, TMB, immune-microenvironment features, immune-escape genomics and ctDNA kinetics—potentially derived using machine-learning approaches—may better predict benefit from first-line chemo-immunotherapy in ESCC than PD-L1 alone (16).

For daily practice, the key question is whether PD-1 inhibitor plus chemotherapy should be routinely withheld from advanced ESCC patients with low PD-L1. Wu et al., together with multiple meta-analyses, suggest that the answer is no. First, PD-1 inhibitor plus chemotherapy confers clinically relevant benefit in ESCC across PD-L1 strata, including TPS <1% and CPS <10 (1). Second, PD-L1 is prognostic and modestly predictive—higher expression generally associates with greater benefit—but low expression does not reliably identify non-responders. Third, in health systems with constrained resources, PD-L1 may help prioritize patients expected to derive the largest benefit, yet should not be used as an absolute exclusion criterion in otherwise fit patients. At the same time, ODAC’s cautionary stance underscores that regulatory decisions must account for the uncertainty in small PD-L1 <1 cohorts, the possibility of minimal benefit at this extreme, and the class-wide safety and cost profile (7). From a regulatory perspective, this uncertainty may be considered insufficient to support an unrestricted indication. Clinicians, particularly in regions with high ESCC burden and limited alternatives, may reasonably weigh the totality of evidence differently and continue to offer chemo-immunotherapy to selected PD-L1-low patients after shared decision-making (12,16,26).

The work by Wu et al. thus moves the field beyond a binary view of PD-L1 in ESCC and provides a strong rationale for continuing to offer first-line PD-1 inhibitor plus chemotherapy irrespective of PD-L1 expression, while at the same time highlighting the urgent need for robust, composite biomarkers that better capture the multidimensional nature of immunotherapy sensitivity in this disease.

Acknowledgments

None.

Footnote

Provenance and Peer Review: This article was commissioned by the editorial office, Chinese Clinical Oncology. The article has undergone external peer review.

Peer Review File: Available at https://cco.amegroups.com/article/view/10.21037/cco-2025-1-168/prf

Funding: None.

Conflicts of Interest: The author has completed the ICMJE uniform disclosure form (available at https://cco.amegroups.com/article/view/10.21037/cco-2025-1-168/coif). S.K. has received grants from Ono Pharmaceutical Co., Ltd., MSD Co., Ltd., Nobelpharma, Janssen Pharmaceutical K.K., Eli Lilly Co., Ltd., Chugai Pharmaceutical Co., Ltd., Daiichi-Sankyo Co., Ltd., AstraZeneca Co., Ltd., Abbvie, Kyowa Kirin, Astellas Pharmaceutical Co., Ltd., Novartis Co., Ltd., and Bayer; and honoraria from Ono Pharmaceutical Co., Ltd., Taiho Pharmaceutical Co., Ltd., MSD Co., Ltd., Daiichi-Sankyo Co., Ltd., Merck Biopharma Co., Ltd., Bristol-Myers Squibb Co., Ltd., Eli Lilly Co., Ltd., Chugai Pharmaceutical Co., Ltd., Bayer, Eisai, Novartis Co., Ltd., and Astellas Pharmaceutical Co., Ltd. The author has no other conflicts of interest to declare.

Ethical Statement: The author is accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Wu HX, Pan YQ, He Y, et al. Clinical Benefit of First-Line Programmed Death-1 Antibody Plus Chemotherapy in Low Programmed Cell Death Ligand 1-Expressing Esophageal Squamous Cell Carcinoma: A Post Hoc Analysis of JUPITER-06 and Meta-Analysis. J Clin Oncol 2023;41:1735-46. [Crossref] [PubMed]
2. 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;398:759-71. [Crossref] [PubMed]
3. Doki Y, Ajani JA, Kato K, et al. Nivolumab Combination Therapy in Advanced Esophageal Squamous-Cell Carcinoma. N Engl J Med 2022;386:449-62. [Crossref] [PubMed]
4. Luo H, Lu J, Bai Y, et al. Effect of Camrelizumab vs Placebo Added to Chemotherapy on Survival and Progression-Free Survival in Patients With Advanced or Metastatic Esophageal Squamous Cell Carcinoma: The ESCORT-1st Randomized Clinical Trial. JAMA 2021;326:916-25. [Crossref] [PubMed]
5. Wang ZX, Cui C, Yao J, et al. Toripalimab plus chemotherapy in treatment-naïve, advanced esophageal squamous cell carcinoma (JUPITER-06): A multi-center phase 3 trial. Cancer Cell 2022;40:277-288.e3. [Crossref] [PubMed]
6. Lu Z, Wang J, Shu Y, et al. Sintilimab versus placebo in combination with chemotherapy as first line treatment for locally advanced or metastatic oesophageal squamous cell carcinoma (ORIENT-15): multicentre, randomised, double blind, phase 3 trial. BMJ 2022;377:e068714. [Crossref] [PubMed]
7. US Food and Drug Administration. Oncologic Drugs Advisory Committee Meeting: PD-L1 expression and immune checkpoint inhibitors for first-line esophageal cancer. FDA Briefing Document. September 26, 2024. Available online: https://www.fda.gov/media/182143/download
8. 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;24:483-95. [Crossref] [PubMed]
9. Yap DWT, Leone AG, Wong NZH, et al. Effectiveness of Immune Checkpoint Inhibitors in Patients With Advanced Esophageal Squamous Cell Carcinoma: A Meta-analysis Including Low PD-L1 Subgroups. JAMA Oncol 2023;9:215-24. [Crossref] [PubMed]
10. Lee CK, Park S, Lee Y, et al. Efficacy of the first-line immune checkpoint inhibitor plus chemotherapy for gastroesophageal cancer: A meta-analysis of phase III trials including unreported PD-L1 subgroups. Cancer Lett 2025;623:217718. [Crossref] [PubMed]
11. Chen W, Cao K, Zhang L, et al. Efficacy and safety evaluation of frontline immunotherapy combinations in advanced esophageal squamous cell carcinoma: a network meta-analysis highlighting the value of PD-L1 expression positivity scores. Front Immunol 2024;15:1414753. [Crossref] [PubMed]
12. Zafar S, Shehzadi R, Dawood H, et al. Current evidence of PD-1 and PD-L1 immune checkpoint inhibitors for esophageal cancer: an updated meta-analysis and synthesis of ongoing clinical trials. Ther Adv Med Oncol 2024;16:17588359231221339. [Crossref] [PubMed]
13. Kent DM, Steyerberg E, van Klaveren D. Personalized evidence based medicine: predictive approaches to heterogeneous treatment effects. BMJ 2018;363:k4245. [Crossref] [PubMed]
14. Riley RD, Higgins JP, Deeks JJ. Interpretation of random effects meta-analyses. BMJ 2011;342:d549. [Crossref] [PubMed]
15. Tsao MS, Kerr KM, Kockx M, et al. PD-L1 Immunohistochemistry Comparability Study in Real-Life Clinical Samples: Results of Blueprint Phase 2 Project. J Thorac Oncol 2018;13:1302-11. [Crossref] [PubMed]
16. Wang X, Wang P, Huang X, et al. Biomarkers for immunotherapy in esophageal cancer. Front Immunol 2023;14:1117523. [Crossref] [PubMed]
17. Havel JJ, Chowell D, Chan TA. The evolving landscape of biomarkers for checkpoint inhibitor immunotherapy. Nat Rev Cancer 2019;19:133-50. [Crossref] [PubMed]
18. Royston P, Altman DG, Sauerbrei W. Dichotomizing continuous predictors in multiple regression: a bad idea. Stat Med 2006;25:127-41. [Crossref] [PubMed]
19. Sharma P, Hu-Lieskovan S, Wargo JA, et al. Primary, Adaptive, and Acquired Resistance to Cancer Immunotherapy. Cell 2017;168:707-23. [Crossref] [PubMed]
20. Deng J, Chen H, Zhou D, et al. Comparative genomic analysis of esophageal squamous cell carcinoma between Asian and Caucasian patient populations. Nat Commun 2017;8:1533. [Crossref] [PubMed]
21. Baba Y, Nomoto D, Okadome K, et al. Tumor immune microenvironment and immune checkpoint inhibitors in esophageal squamous cell carcinoma. Cancer Sci 2020;111:3132-41. [Crossref] [PubMed]
22. Galluzzi L, Humeau J, Buqué A, et al. Immunostimulation with chemotherapy in the era of immune checkpoint inhibitors. Nat Rev Clin Oncol 2020;17:725-41. [Crossref] [PubMed]
23. Samstein RM, Lee CH, Shoushtari AN, et al. Tumor mutational load predicts survival after immunotherapy across multiple cancer types. Nat Genet 2019;51:202-6. [Crossref] [PubMed]
24. Cristescu R, Mogg R, Ayers M, et al. Pan-tumor genomic biomarkers for PD-1 checkpoint blockade-based immunotherapy. Science 2018;362:eaar3593. [Crossref] [PubMed]
25. Guo Y, Qu W, Liu Y, et al. Circulating Tumor DNA Dynamic Changes in Esophageal Squamous Cell Carcinoma Receiving Immunochemotherapy. Thorac Cancer 2025;16:e70084. [Crossref] [PubMed]
26. Rogers JE, Ajani JA. The role of ramucirumab and pembrolizumab combination in patients with advanced non-small cell lung cancer, gastroesophageal adenocarcinoma, or urothelial carcinoma. Chin Clin Oncol 2021;10:30. [Crossref] [PubMed]