Diagnostic performance of whole-body diffusion-weighted imaging for staging esophageal cancer: a comparative study...

Accurate preoperative staging is crucial for selecting treatment strategies and assessing prognosis in esophageal cancer. Positron emission tomography-computed tomography (PET-CT) is currently recognized as an essential staging modality, whereas the value of whole-body diffusion-weighted imaging (WB-DWI), a radiation-free functional magnetic resonance imaging (MRI) technique in esophageal cancer staging remains unclear. The objective of the study was to conduct a head-to-head comparison of the diagnostic accuracy of WB-DWI versus PET-CT in the initial diagnosis of esophageal cancer patients for local T staging, regional lymph node (N) assessment, and distant metastasis (M) detection.

This study collected data from 96 patients with esophageal cancer who underwent pathological diagnosis and imaging evaluation at Haian People's Hospital between January 2023 and January 2025. Propensity Score Matching (PSM) was employed to balance confounding factors between groups, ultimately forming matched cohorts for analysis: the WB-DWI group (n = 32) and the PET-CT group (n = 32). Using postoperative pathology (PTNM) and/or 6-month follow-up clinical diagnosis as the gold standard, the following metrics were evaluated: T staging accuracy, N staging sensitivity/specificity, distant metastasis (M staging) detection rate, overall staging accuracy, image signal-to-noise ratio (SNR), examination duration, and inter-observer agreement (Kappa value). Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were calculated for both methods. Diagnostic performance was analyzed using ROC curves, and the area under the curve (AUC) was compared via DeLong's test.

WB-DWI and PET-CT demonstrated comparable overall staging accuracy for esophageal cancer (84.4% vs. 87.5%, p > 0.05). For N staging, PET-CT demonstrated higher sensitivity (90.6% vs. 78.1%, p < 0.05), while both methods showed identical specificity (87.5%). In detecting distant metastases, PET-CT exhibited marginally higher detection rates and accuracy than WB-DWI, though differences were not statistically significant. WB-DWI demonstrated lower accuracy for T staging (71.9% vs. 84.4%, p < 0.05). WB-DWI images exhibited significantly lower SNR than PET-CT (p < 0.01), but the examination was shorter in duration and showed good inter-observer agreement (Kappa > 0.60, p < 0.001).

WB-DWI demonstrates comparable overall efficacy to PET-CT in esophageal cancer staging. Although its sensitivity and efficiency are slightly lower, its radiation-free nature and high specificity make it a viable supplement or alternative to PET-CT for specific patient populations. These findings provide evidence for precise clinical selection.