# Real-World Outcomes and Biomarker Analysis Based on Routine Clinical, Laboratory, and Pathologic Parameters in Metastatic or Unresectable Esophageal Cancer Treated with First-Line Anti-PD-1 Plus Fluoropyrimidine and Platinum

## TL;DR

This study shows that combining anti-PD-1 therapy with chemotherapy works well in real-world settings for advanced esophageal cancer, with a new scoring system to predict outcomes.

## Contribution

A new prognostic scoring system using ECOG, CRP, and PD-L1 CPS improves outcome prediction in real-world ESCC treatment.

## Key findings

- The combination therapy had a 48.3% response rate and 77.0% disease control rate in real-world ESCC patients.
- A point-based risk score using ECOG, CRP, and PD-L1 CPS effectively stratified patients into four survival risk groups.
- Stratifying PD-L1 CPS into three levels showed a graded association with treatment outcomes, suggesting the need for more detailed PD-L1 evaluation.

## Abstract

First-line anti-programmed death-1 (PD-1) plus chemotherapy is the current standard for advanced esophageal squamous cell carcinoma, but real-world data remain limited. In this retrospective analysis, combination therapy demonstrated survival outcomes comparable to pivotal trials despite the inclusion of patients with poor performance status. A point-based prognostic score using Eastern Cooperative Oncology Group (ECOG), C-reactive protein (CRP), and programmed death-ligand 1 (PD-L1) combined positive score (CPS) stratified survival risk, and refined PD-L1 CPS grouping revealed a potential graded association between CPS and treatment outcomes, suggesting value in more nuanced PD-L1 assessment in clinical practice.

Background/Objectives: The combination of anti-programmed death-1 (PD-1) inhibitors and chemotherapy is the standard first-line treatment for unresectable or metastatic esophageal squamous cell carcinoma (ESCC). However, real-world data remain limited, particularly regarding prognostic biomarkers. Methods: This multi-institutional retrospective study analyzed patients with metastatic or unresectable ESCC who received first-line pembrolizumab or nivolumab plus fluoropyrimidine and platinum-based chemotherapy. Treatment regimens mirrored those in KEYNOTE-590 and CheckMate 648. Efficacy, safety, and prognostic factors were assessed. Prognostic factors were identified using multivariable Cox regression, and a point-based risk scoring system was developed. Results: Among 87 patients, the objective response rate was 48.3%, and the disease control rate was 77.0%. Median progression-free survival (PFS) was 5.6 months (95% CI, 4.5–8.7), and the median overall survival (OS) was 13.1 months (95% CI, 10.6–not reached). Grade 3–4 treatment-related adverse events occurred in 51.7% of patients. Eastern Cooperative Oncology Group (ECOG) performance status ≥ 2, elevated C-reactive protein, and lower programmed death-ligand 1 (PD-L1) combined positive score (CPS) were independently associated with worse PFS and OS. A prognostic risk score ranging from 0 to 5 based on these factors stratified patients into four prognostic groups with distinct survival outcomes. Median PFS ranged from not reached in the low-risk group to 2.1 months in the high-risk group. Stratifying PD-L1 CPS into three levels (<10, 10–49, ≥50) revealed a graded association between CPS and treatment outcomes, supporting the need for more nuanced PD-L1 evaluation beyond binary classification. Conclusions: First-line anti-PD-1 therapy combined with chemotherapy demonstrated favorable real-world outcomes in ESCC. The proposed prognostic scoring system may help personalize treatment strategies.