85 Spatial heterogeneity of tumor associated macrophages in the tumor immune microenvironment in ccRCC. (9th November 2020)

• Record Type: Journal Article
• Authors: Chakiryan, Nicholas; Kimmel, Gregory; Kim, Youngchul; Nguyen, Jonathan; Chahoud, Jad; Spiess, Philippe; Dhillon, Jasreman; Wang, Liang; Moran-Segura, Carlos; Mule, James; Altrock, Philipp; Manley, Brandon
• Is Part Of: Journal for immunotherapy of cancer. Volume 8(2020)Supplement 3
• Page Start: A54
• Page End: A54
• Publication Date: 2020-11-09
• Subjects: Cancer -- Immunotherapy -- Periodicals; Cancer -- Immunological aspects -- Periodicals; Tumors -- Immunological aspects -- Periodicals; Immunotherapy -- Periodicals; 616.99406105
• DOI: 10.1136/jitc-2020-SITC2020.0085
• Languages: English
• ISSNs: 2051-1426

Abstract

Abstract : Background: Tumor associated macrophages (TAM) stimulate tumor proliferation and facilitate immune escape via production of immunosuppressive cytokines. We hypothesize that non-random spatial clustering of TAMs within the tumor are associated with poor survival in ccRCC patients. Methods: Tumor specimens were obtained from 41 patients with metastatic ccRCC who received immunotherapy (IT). Sections from the tumor core underwent multiplex immunofluorescence staining for CD68, CD163, and CD206. Digital pathologic analysis was used to convert the digital images to spatial point pattern plots (PPP). Ripley's K function, the current standard metric for spatial heterogeneity, was utilized. Novel metrics were developed using a probability density function (PDF) for distances between cells, assuming that cells can be located anywhere with equal probability. Empirical histograms were generated from the PPPs. Deviation from the PDF demonstrates a non-random distribution. Deviations were quantified with the Kolmogorov-Smirnov (KS) test and Cramér-von Mises (CVM) criterion. Overall survival (OS) was assessed between groups stratified by the median value for each metric using Kaplan-Meier and log-rank analysis. Figure 1 A. Results: 75 slides were analyzed from the 41 patients. The three metrics for measuring spatial heterogeneity had moderate and statistically significant correlation with each other (Spearman's R: Ripley/KS=0.68, p<0.01; Ripley/CVM=0.54, p<0.01; KS/CVM=0.47, p<0.01; figure 1 B). Using CVM, increasingly non-random distribution of the Tumor-CD68+ cell relationship was associated with worse OS (p<0.01, figure 1 C), and increasingly non-random distribution of CD163+ cells suggested an association with worse OS without reaching statistical significance (p=0.06, figure 1 C). No statistically significant associations were identified using the KS or Ripley's K metrics. Conclusions: We describe CVM and KS as novel metrics for measuring spatial heterogeneity of immune cells. Increased spatial heterogeneity of CD68+ TAMs and tumor cells was associated with worse OS in patients with metastatic ccRCC who received IT. These findings corroborate prior reports of TAMs eliciting an immunosuppressive effect on the tumor-immune microenvironment, and demonstrate the novel finding of a clinically significant effect of TAM spatial clustering on OS.