CD163-positive macrophage enrichment is associated with mismatch repair deficiency and adverse pathologic features in early-stage endometrioid endometrial carcinoma: a retrospective cohort study.

Background: Mismatch repair deficiency defines an immunologically distinct subgroup of endometrioid endometrial carcinoma, but its relationship with CD163-positive macrophage infiltration in early-stage disease remains incompletely characterized. We evaluated the association between MMR status and quantitative CD163-positive TAM infiltration, together with secondary clinicopathologic and exploratory DFS analyses. Methods: This single-center retrospective cohort included patients with surgically treated FIGO 2009 stage I-II endometrioid endometrial carcinoma and available formalin-fixed paraffin-embedded tumor tissue. Immunohistochemistry for MLH1, PMS2, MSH2, MSH6, and CD163 defined MMR status and quantified macrophage density. CD163-positive TAM density was analyzed quantitatively as the principal measure, with a cohort-median low/high categorization retained for secondary phenotypic analyses. Median regression evaluated the association between MMR status and continuous CD163-positive TAM density, whereas logistic regression was used for the secondary binary TAM analysis and the composite adverse pathologic endpoint. Exploratory DFS analyses included single-marker and combined Cox models, formal MMR-by-TAM interaction testing, and comparison of models with and without continuous CD163-positive TAM density. Results: Among 254 patients assessed, 236 were included. MMR protein expression was retained in 172 tumors (72.9%), whereas 64 (27.1%) were classified as dMMR. CD163-positive TAM density was higher in dMMR than pMMR tumors (median, 76 vs 47 cells/HPF; P<0.001), and high TAM infiltration was more frequent in dMMR tumors (44/64, 68.8% vs 71/172, 41.3%; P<0.001). After adjustment for age, FIGO stage, tumor grade, tumor size, and deep myometrial invasion, dMMR remained associated with high macrophage infiltration (adjusted OR, 2.27; 95% CI, 1.19-4.18; P = 0.008). High macrophage infiltration was associated with grade 3 histology, deep myometrial invasion, LVSI, and the composite adverse pathologic feature; the association with the composite endpoint persisted after adjustment (adjusted OR, 1.78; 95% CI, 1.01-3.14; P = 0.047). During follow-up, 26 DFS events occurred. In exploratory Cox analyses, dMMR was associated with DFS (adjusted HR, 2.76; 95% CI, 1.25-6.10; P = 0.012), whereas continuous CD163-positive TAM density was not (adjusted HR per 10 cells/HPF, 1.04; 95% CI, 0.97-1.12; P = 0.281), and addition of CD163-positive TAM density did not significantly improve model fit beyond MMR status (likelihood-ratio P = 0.279). The dMMR/high-TAM phenotype retained an exploratory association with DFS in the conventional model (adjusted HR, 2.21; 95% CI, 1.02-4.81; P = 0.044), but the association was attenuated after additional adjustment for adjuvant therapy (HR, 2.04; 95% CI, 0.93-4.52; P = 0.075). Conclusions: IHC-defined dMMR was associated with increased CD163-positive macrophage infiltration, and higher TAM infiltration was associated with adverse pathologic features. DFS findings were exploratory and did not establish incremental prognostic value of CD163-positive TAM infiltration beyond MMR status; larger externally validated cohorts are required to assess prognostic relevance.