Optimized tumour infiltrating lymphocyte assessment for triple negative breast cancer prognostics

Abstract

The tumour microenvironment has been shown to be a valuable source of prognostic information for different cancer types. This holds in particular for triple negative breast cancer (TNBC), a breast cancer subtype for which currently no prognostic biomarkers are established. Although different methods to assess tumour infiltrating lymphocytes (TILs) have been published, it remains unclear which method (marker, region) yields the most optimal prognostic information. In addition, to date, no objective TILs assessment methods are available.For this proof of concept study, a subset of our previously described TNBC cohort (n ¼ 94) was stained for CD3, CD8 and FOXP3 using multiplex immunohistochemistry and subsequently imaged by a multispectral imaging system. Advanced whole-slide image analysis algorithms, including convolutional neural networks (CNN) were used to register unmixed multispectral images and corresponding H&E sections, to segment the different tissue compartments (tumour, stroma) and to detect all individual positive lymphocytes. Densities of positive lymphocytes were analysed in different regions within the tumour and its neighbouring environment and correlated to relapse free survival (RFS) and overall survival (OS).We found that for all TILs markers the presence of a high density of positive cells correlated with an improved survival. None of the TILs markers was superior to the others. The results of TILs assessment in the various regions did not show marked differences between each other.The negative correlation between TILs and survival in our cohort are in line with previous studies. Our results provide directions for optimizing TILs assessment methodology.

Prognostic Value of Pretreatment Neutrophil-to-Lymphocyte Ratio in HER2-Positive Metastatic Breast Cancer

“…Some studies have shown that high NLR is significantly correlated with poor prognosis in triple-negative breast cancer (TNBC) but not in luminal A-like, luminal B-like, or HER2-enriched subtypes [ 8 ]. The predictive prognostic value of TILs has been established for TNBC [ 31 ], but whether peripheral blood neutrophils or lymphocytes have similar value in TNBC is not known. Noh et al [ 13 ] found a significant prognostic value of NLR in luminal subtype breast cancer (ER+ or PR+, HER2−), whereas Yao et al [ 16 ] found that NLR has a significant prognostic value regardless of luminal type or TNBC.…”

Automated Quantification of sTIL Density with H&E-Based Digital Image Analysis Has Prognostic Potential in Triple-Negative Breast Cancers

“…Compared to prior attempts to apply image analysis for computational assessment of sTIL, such as patch- [ 26 ], object- [ 28 , 29 ], or segmentation-based methods [ 27 , 48 ], our study incorporates all parts of the TIL-WG guideline; from discriminating tissue from glass, and excluding necrotic regions and inflammation related to the non-invasive epithelium, such normal glands and DCIS/LCIS. A recent study [ 33 ] investigated several aspects of computational TIL assessment for prognosis in TNBC. To find the optimal compartment (margin, tumor-associated stroma, etc.…”

“…Most studies of computational TILs have employed patch- or object detection-based approaches [ 26 , 27 , 28 , 29 ] with manual region outlining as part of the pipeline [ 30 ]. Some of these also used multiplexed immunofluorescence (mIF) [ 31 ] or immunohistochemistry (IHC) [ 32 , 33 ] to classify cells as lymphocytes. All existing studies proposing H&E-based algorithms rely on only manual H&E ground truth annotations to train their model even though the manual human limitations have shown inconsistencies in this task [ 14 ].…”

Artificial intelligence in breast cancer histopathology

“…This said, current development is still limited by the scarcity of publicly annotated TIL‐associated datasets. TIL detection was also attempted on IHC 53,54 and multispectral IHC, 55 facilitating the assessment of in‐tumour immunology. Deep learning has become an indispensable research tool and sheds light on the prognostic implications, 56 immune response pathway, and activated genes 57 in TIL‐rich cancer.…”