Abstract 2369: Spatial organization of the tumour immune microenvironment (TIME) in primary and metastatic melanoma is associated with patient outcome

Abstract Introduction The complex spatial cellular interactions of immune cell populations within the melanoma TIME are associated with patient prognosis and immunotherapy response. Novel multiplex immunofluorescence (mIHC) and spatial analysis techniques enable comprehensive investigation of these...

Full description

Saved in:
Bibliographic Details
Published inCancer research (Chicago, Ill.) Vol. 83; no. 7_Supplement; p. 2369
Main Authors Attrill, Grace Heloise, Shteinman, Eva R., Bai, Xinyu, Marsh-Wakefield, Felix, Quek, Camelia, Palendira, Umaimainthan, Vergara, Ismael A., Long, Georgina V., Scolyer, Richard A., Wilmott, James S.
Format Journal Article
LanguageEnglish
Published 04.04.2023
Online AccessGet full text

Cover

Loading…
More Information
Summary:Abstract Introduction The complex spatial cellular interactions of immune cell populations within the melanoma TIME are associated with patient prognosis and immunotherapy response. Novel multiplex immunofluorescence (mIHC) and spatial analysis techniques enable comprehensive investigation of these interactions. This study aimed to develop an automated spatial analysis workflow for large patient cohorts and to determine immunospatial patterns associated with outcome in patients (pts) with primary melanoma as well as pts with stage III melanoma treated with adjuvant immunotherapy. Methods Two cohorts with available clinical & mIHC data were studied: one of primary melanoma pts (primary cohort, n=46), and another of baseline resected stage III melanoma pts who received adjuvant anti-PD-1 (stage III cohort, n=119). Melanoma tissue was stained for CD8+ T cells and the markers CD103, PD-1 and CD39 in both cohorts and the primary cohort was additionally stained for NK cells, B cells, and Langerhans cells (LCs). mIHC data was investigated using the SPIAT spatial analysis R package. Immune-melanoma and immune-immune spatial relationships were quantified throughout the entire tissue and analysed using neighborhood analysis and cell colocalization metrics including entropy and mixing interaction scores. Results In the primary melanoma cohort, 11 neighborhood metaclusters (NMCs) were identified from 1617 neighborhoods. NMCs with high B cells, CD39+CD103+ CD8+ T cells, or LCs were associated with improved outcomes in this cohort. Neighborhood composition varied significantly based on tissue localisation, with B cells, LCs and NK cells significantly increased in marginal and stromal neighborhoods (p<0.0001). Over 100 B cell aggregate (BCA) neighborhoods were identified predominantly at the tumor margins and adjacent stroma. These BCAs were likely early TLSs based on their localisation and immune composition. BCA composition was further compared based on patient outcomes, with those harboring increased CD103-PD-1+ CD8+ T cells associated with poor outcome. In-depth pairwise analysis of immune phenotypes with melanoma cells found that increased B cell-melanoma and CD103+PD-1- CD8+ T cell-melanoma interaction was associated with improved outcomes in the primary cohort. In the stage III cohort, increased intratumoral immune diversity was associated with reduced melanoma recurrence (p=0.0004). Conclusions In developing a cohort-wide TIME spatial analytic approach and workflow, we identified multiple immunospatial associations with patient outcome in primary and stage III metastatic melanoma. Intratumoral immune diversity and immune cell neighborhoods indicate beneficial anti-tumor immune functions contributing towards improved patient outcomes. Citation Format: Grace Heloise Attrill, Eva R. Shteinman, Xinyu Bai, Felix Marsh-Wakefield, Camelia Quek, Umaimainthan Palendira, Ismael A. Vergara, Georgina V. Long, Richard A. Scolyer, James S. Wilmott. Spatial organization of the tumour immune microenvironment (TIME) in primary and metastatic melanoma is associated with patient outcome [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 2369.
ISSN:1538-7445
1538-7445
DOI:10.1158/1538-7445.AM2023-2369