INTRODUCTION: Advanced esophageal cancer benefits from immunotherapy. For squamous cell carcinoma (SCC), allocation for immune checkpoint inhibitors (ICI) is based on PD-L1 positivity assessed by the tumor proportion score (TPS) and combined positive score (CPS). It is unclear whether patients without an indication for ICI based on TPS are eligible when CPS is considered, or vice versa. METHODS: A systematic literature search in Medline, Embase, Web of Science, and Cochrane was conducted to identify studies evaluating PD-L1 expression using TPS and CPS in esophageal SCC. PD-L1 discordant cases were defined as TPS<1 and CPS≥10, or TPS≥1% and CPS<10. When paired patient-level TPS/CPS classifications were unavailable, possible discordance ranges were estimated from marginal positivity rates. RESULTS: Eight studies comprising a total of 1,252 patients were selected. Two studies reported the percentage of discordant TPS/CPS cases, with 30.2% (16 out of 53) showing discordance. The minimum possible discordance among the six studies without paired patient-level data was estimated at 5.7% (69/1199), ranging from 2.4% to 14.4% across studies. The maximum possible discordance estimates ranged from 43.2% to 87.3%, reflecting the uncertainty introduced by the absence of paired patient-level data. DISCUSSION: This review suggests that the current TPS≥1 and CPS≥10 thresholds are not interchangeable for determining treatment eligibility in ESCC. These discrepancies may significantly impact treatment eligibility, although the clinical benefit of ICI in TPS/CPS-discordant patients remains unknown. CONCLUSION: Understanding TPS/CPS concordance is crucial for optimizing patient selection for ICI in esophageal SCC. Sequential assessment, starting with TPS and followed by CPS when TPS is negative, may represent a pragmatic eligibility-based testing strategy to reduce the risk of overlooking potentially eligible patients.