Content area
Abstract
Objectives:In recent years, the tumor immune microenvironment [TIME] has become a central element of cancer research. The TIME significantly influences the expression of immune checkpoints as well as anti-tumor reaction and clinical survival. To date there are only few data available on the TIME in upper urinary tract carcinoma [UTUC], representing a minority (5-10%) of the urothelial carcinoma in general.
The aim of this study was the further characterization of the TIME in UTUCs. First, we investigated the association between infiltrating immune cells and clinical-pathological characteristics as well as their impact on survival. Second, we asked whether identification of different immune cell phenotypes associates with distinct clinical- pathological characteristics and survival rates.
Materials and Methods:A variety of immunohistochemical stainings (Cluster of Differentiation [CD] CD8, CD3, CD45, PD-1, PD-L1) was performed on a study cohort of 201 UTUC patients of all tumor stages from the centers in Erlangen and Marburg. Quantitative immunohistochemical analysis was performed using QuPath software for the CD8, CD3, CD45 and PD-1 stainings. PD-L1 quantification was performed using the established algorithms for PD-L1 analysis.Results were interpreted by descriptive statistical methods. Survival rates were estimated by the Kaplan-Meier method. The cohort of UTUCs was subdivided into different immune cell phenotypes by cluster analysis.
Results:We found that a higher tumor stage coincides with an increased rate of infiltrating immune cells. There was no significant difference in infiltrating immune cells between male and female UTUC patients. However, higher immune cell infiltrates were associated with high grade tumors, lymph node involvement and metastases. Survival analyses revealed a positive correlation between increased immune cell infiltration and reduced disease specific survival. We performed a subclassification of the UTUC cohort into four different immune cell phenotypes (`uninflamed ́, `inflamed low ́, `inflamed high ́, `evasion ́) in similarity to the classification in muscle invasive bladder cancer. The `uninflamed ́ phenotype with low immune cell infiltration was associated with the best survival rate. In contrast, the `inflamed high ́ and the `evasion ́ phenotype, 1 characterized by a high PD-L1 expression and mostly invasive tumor stage, were associated with the least favorable survival rates.
Conclusion:High numbers of infiltrating immune cells in UTUCs are associated with unfavorable survival. UTUCs can be subclassified into different immune cell phenotypes with the `inflamed high ́ and the `evasion ́ phenotype with high PD-L1 expression displaying a worse survival. These biomarkers might be useful for prognostication of the course of disease as well as prediction of response to immune therapy. Additionally, our findings might contribute to the search for an individualized anti-cancer therapy stratified by TIME characteristics for UTUC patients.





