Virtual Library
Start Your Search
S.D. Martin
Author of
-
+
P2.02 - Biology/Pathology (ID 616)
- Event: WCLC 2017
- Type: Poster Session with Presenters Present
- Track: Biology/Pathology
- Presentations: 1
- Moderators:
- Coordinates: 10/17/2017, 09:30 - 16:00, Exhibit Hall (Hall B + C)
-
+
P2.02-038 - Imaging Platform for the Quantification of Cell-Cell Spatial Organization within the Tumour-Immune Microenvironment (ID 9605)
09:30 - 09:30 | Author(s): S.D. Martin
- Abstract
Background:
The contribution of the tumour-immune microenvironment to tumour progression and patient outcome has become increasingly evident. Newly developed genomic tools have enabled the study of immune cell composition from bulk tumour data. However, such tools (e.g. CIBERSORT) do not provide the key spatial information that is crucial to understand tumour-immune cell interactions. To this end, we have developed a multispectral imaging platform that improves upon traditional analysis methods of cell segmentation and cell density calculations by further quantifying nearest-neighbour interactions (cell-cell spatial relationships). We apply this technology to investigate tumour-immune cell spatial relationships and their clinical significance to discover novel biological insights.
Method:
Whole tissue sections from 20 lung adenocarcinomas were stained for CD3, CD8, and CD79a and counterstained with haematoxylin. Multispectral images were acquired for five fields of view and analyzed to quantify cell types. Regions of Interest (ROIs) were then identified for the characterization of intra-tumoural and dense inflammatory regions. Image files including ROIs were analyzed in order to quantify cell-cell spatial relationships. Non-random patterns of immune cell distributions were identified using the Monte Carlo re-sampling method (500 iterations). Immune cell counts, densities, spatial relationships, and significant immune cell distributions were associated with clinical features by two-group comparison (Kruskal-Wallis p<0.001).
Result:
Our analysis generated 234 image files for analysis, including ROIs. Each field of view contained an average of 16,400 cells. The densities of intra-tumoural CD3+CD8+ and CD3+ T cells were significantly lower in recurrent cases, agreeing with literature reports. Following Monte Carlo analysis, non-random cell-cell spatial proximities emerged that were not observed at a cell density level. For example, an increased proximity of CD3+ T cells and B cells was observed in never smokers, while a decreased proximity was observed in ever smokers.
Conclusion:
While immune cell densities are of clinical prognostic importance, their spatial organization within the tumour architecture is of functional importance (e.g. the inhibition of cytotoxic T cell activity by adjacent PD-L1 expressing cells). In addition to cell densities, our platform is capable of quantifying cell-cell spatial relationships, thereby providing further information for clinical associations and for the identification of novel prognostic interactions. This automated quantification could be used to complement visual diagnostics and improve prognostic interpretation of histology specimens.
