Small-intestinal neuroendocrine tumors (SI-NETs) frequently give rise to mesenteric masses with fibrotic stroma, yet the primary-tumor features that signal mesenteric involvement are not well defined. We performed a single-centre retrospective study of 70 resected, well-differentiated SI-NETs, dichotomized into a primary-only cohort without mesenteric lesions (n = 21) and a mesenteric-lesion cohort with at least one mesenteric mass (n = 49), using the endpoint “any mesenteric lesion”. Clinicopathologic variables, including tumor size, grade, multiplicity, vascular invasion, and perineural invasion, were recorded. The compartment-specific collagen proportionate area in the primary bowel wall was quantified using digital picrosirius red (PSR) analysis. Patients with and without mesenteric lesions were similar in age, sex, grade, proliferation indices, and tumor multiplicity. Primary tumors were slightly but significantly larger in patients with mesenteric lesions (median 2.0 cm versus 1.5 cm; area under the receiver operating characteristic (ROC) curve 0.66). Vascular invasion and perineural invasion in the primary tumor were strongly enriched in the mesenteric lesion cohort, with odds ratios of 10.2 and 8.5, respectively. By contrast, mucosal/submucosal and serosal/subserosal collagen quantification revealed highly overlapping distributions between cohorts and only weak discrimination for mesenteric involvement. These findings indicate that primary-tumor size, vascular invasion, and perineural invasion are key histological correlates of mesenteric involvement in resected SI-NETs, whereas primary bowel wall collagen quantification is best regarded as a descriptive stromal feature rather than a stand-alone classifier of mesenteric involvement.
Tumor size and vascular and perineural invasion predict mesenteric involvement in small-intestinal neuroendocrine tumors
Vanoli, Alessandro;Bortolotto, Chandra;D'Este, Edoardo;Checchin, Filippo;
2026-01-01
Abstract
Small-intestinal neuroendocrine tumors (SI-NETs) frequently give rise to mesenteric masses with fibrotic stroma, yet the primary-tumor features that signal mesenteric involvement are not well defined. We performed a single-centre retrospective study of 70 resected, well-differentiated SI-NETs, dichotomized into a primary-only cohort without mesenteric lesions (n = 21) and a mesenteric-lesion cohort with at least one mesenteric mass (n = 49), using the endpoint “any mesenteric lesion”. Clinicopathologic variables, including tumor size, grade, multiplicity, vascular invasion, and perineural invasion, were recorded. The compartment-specific collagen proportionate area in the primary bowel wall was quantified using digital picrosirius red (PSR) analysis. Patients with and without mesenteric lesions were similar in age, sex, grade, proliferation indices, and tumor multiplicity. Primary tumors were slightly but significantly larger in patients with mesenteric lesions (median 2.0 cm versus 1.5 cm; area under the receiver operating characteristic (ROC) curve 0.66). Vascular invasion and perineural invasion in the primary tumor were strongly enriched in the mesenteric lesion cohort, with odds ratios of 10.2 and 8.5, respectively. By contrast, mucosal/submucosal and serosal/subserosal collagen quantification revealed highly overlapping distributions between cohorts and only weak discrimination for mesenteric involvement. These findings indicate that primary-tumor size, vascular invasion, and perineural invasion are key histological correlates of mesenteric involvement in resected SI-NETs, whereas primary bowel wall collagen quantification is best regarded as a descriptive stromal feature rather than a stand-alone classifier of mesenteric involvement.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


