Abstract 2872: Image analysis algorithms for whole-slide counting, regional assignment, and subtype classification of tumor-associated macrophages (TAMs).

Abstract

Abstract Tumor-associated macrophages (TAMs) regulate cancer progression on multiple levels, including angiogenesis, treatment response, and metastasis. Additionally, multiple studies have demonstrated a significant association between clinical outcome and TAM abundance, M1 vs. M2 subtype, and intratumoral distribution into stromal or epithelial compartments. However, manual counting of TAMs and assignment into tumor stroma or epithelial nests is laborious, inaccurate, subjective, and does not provide a statistically optimal sample size. These pitfalls may be limiting its utility as a diagnostic tool or an endpoint in drug efficacy studies. Using tissue sections from human breast cancers as a model, we have developed image analysis algorithms for the rapid automated counting and compartmental assignment of total TAMs (CD68+) and TAMs of the more strongly tumor-promoting M2 subtype (CD163+). Using whole slide scanning and proprietary image analysis tools called Feature Analysis on Consectutive Tissue Sections (FACTS), we have developed an approach to not only tally but also map the location of CD68+ and CD163+ macrophages in serial tissue sections, align those sections, and provide information about regional distribution of these macrophages. These outcomes can also be aligned with other serial sections containing relevant biomarkers associated with macrophage infiltration such as hypoxia or angiongenesis, using our proprietary multivariate correlation analysis tool Multivariatemap. Additionally, we have begun investigations into the association of these regional macrophage densities with clinical outcome using patient data available for the model set. These approaches represent a powerful toolkit for elucidating significant quantitative associations between TAMs and tumor vasculature, hypoxia, or other relevant features or markers, for the determination of prognostic or drug efficacy endpoints. Citation Format: Brian E. Laffin, Joseph Krueger, Mohamed Salama. Image analysis algorithms for whole-slide counting, regional assignment, and subtype classification of tumor-associated macrophages (TAMs). [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 2872. doi:10.1158/1538-7445.AM2013-2872