Abstract
BACKGROUND: Pathological examination of lymph node metastasis (LNM) is crucial for treating pancreatic ductal adenocarcinoma (PDAC). Although the tumour stroma is correlated with prognosis in multiple solid tumors, its role in detecting LNM in PDAC is unclear. Thus, this study aimed to investigate the relationship of tumor-stroma ratio (TSR) with LNM, survival and mutational profile in PDAC. METHODS: In this multicenter retrospective study, we examined molecular and clinicopathologic features of 737 PDAC patients from 5 independent cohorts, including surgically resected and endoscopic ultrasound fine-needle aspiration (EUS-FNA) biopsy specimens. TSR was evaluated on hematoxylin and eosin-stained slides and classified as stroma-low (<50% stroma) or stroma-high (≥50% stroma). RESULTS: Compared to TSR-high cases, TSR-low cases were significantly associated with LNM (P < 0.001). TSR could accurately distinguish patients with and without LNM with an area under curve (AUC) of 0.749, with the sensitivity and specificity of 76.5% and 71.6%, respectively. This accuracy of TSR for identifying LNM was further increased by adding other factors including PD-L1 expression or pretreatment serum CA19-9 levels. TSR showed similar levels of accuracy in analysis of resected tumor specimens and EUS-FNA biopsies. Moreover, we found that TSR could also identify residual nodal involvement after neoadjuvant therapy (NAT) using pretreatment EUS-FNA biopsy samples. Heterogeneous genetic alterations were observed between TSR-low and TSR-high subgroups. TSR was identified as an independent predictor of LNM and worse disease-free survival. Major findings were all reproducible in validation, EUS-FNA biopsy, and pre-treatment NAT EUS-FNA biopsy cohorts. CONCLUSIONS: TSR served as a robust and reproducible biomarker that identifies patients at risk for LNM. TSR might be used to select treatment and management strategies for PDAC patients.