Multiscale deep learning framework captures systemic immune features in lymph nodes predictive of triple negative breast cancer outcome in large‐scale studies

Abstract

The suggestion that the systemic immune response in lymph nodes (LNs) conveys prognostic value for triple-negative breast cancer (TNBC) patients has not previously been investigated in large cohorts. We used a deep learning (DL) framework to quantify morphological features in haematoxylin and eosin-stained LNs on digitised whole slide images. From 345 breast cancer patients, 5,228 axillary LNs, cancer-free and involved, were assessed. Generalisable multiscale DL frameworks were developed to capture and quantify germinal centres (GCs) and sinuses. Cox regression proportional hazard models tested the association between smuLymphNet-captured GC and sinus quantifications and distant metastasis-free survival (DMFS). smuLymphNet achieved a Dice coefficient of 0.86 and 0.74 for capturing GCs and sinuses, respectively, and was comparable to an interpathologist Dice coefficient of 0.66 (GC) and 0.60 (sinus). smuLymphNet-captured sinuses were increased in LNs harbouring GCs (p < 0.001). smuLymphNet-captured GCs retained clinical relevance in LN-positive TNBC patients whose cancer-free LNs had on average ≥2 GCs, had longer DMFS (hazard ratio [HR] = 0.28, p = 0.02) and extended GCs' prognostic value to LN-negative TNBC patients (HR = 0.14, p = 0.002). Enlarged smuLymphNet-captured sinuses in involved LNs were associated with superior DMFS in LN-positive TNBC patients in a cohort from Guy's Hospital (multivariate HR = 0.39, p = 0.039) and with distant recurrence-free survival in 95 LN-positive TNBC patients of the Dutch-N4plus trial (HR = 0.44, p = 0.024). Heuristic scoring of subcapsular sinuses in LNs of LN-positive Tianjin TNBC patients (n = 85) cross-validated the association of enlarged sinuses with shorter DMFS (involved LNs: HR = 0.33, p = 0.029 and cancer-free LNs: HR = 0.21 p = 0.01). Morphological LN features reflective of cancer-associated responses are robustly quantifiable by smuLymphNet. Our findings further strengthen the value of assessment of LN properties beyond the detection of metastatic deposits for prognostication of TNBC patients.

Computational methods for metastasis detection in lymph nodes and characterization of the metastasis-free lymph node microarchitecture: A systematic-narrative hybrid review

“…An early adoption of machine learning techniques was used in a study by Niemisto et al 25 In 2005, where K-means clustering algorithm was used to segment LNs based on a simplified three-color clustering scheme. Verghese et al 26 applied a modified Otsu thresholding method to localize LNs in the whole slide breast cancer images. In the study by Wang et al, 27 a U-Net autoencoder was trained using 1x magnification regional gastric cancer LN images to localize the LNs within the slide.…”

“…The authors raised concerns about the choice of the optimal magnification for the model, illustrating this with tumor cell detection differences at 20x and 40x magnification. Verghese et al 26 trained multi-scale U-Net autoencoders with atrous convolution layers for LN sinus and germinal center detection in H&E-stained breast cancer slides. The authors trained the model using WSI crops at 2.5x, 5x and 10x magnifications, concluding that a mixture of different slide magnifications during training improved the final U-Net model performance.…”

“…Among patch classifiers, ResNet models were utilized most frequently (15 studies), 27 , 46 , 47 , 48 , 49 , 50 , 51 , 52 , 53 , 54 , 55 , 56 , 72 , 84 , 98 followed by MIL ( n =9), 31 , 41 , 77 , 78 , 79 , 80 , 81 , 82 , 99 VGG models ( n =8), 36 , 37 , 44 , 46 , 51 , 84 , 100 , 101 and others as shown in Fig. 4 B. Amongst pixel-level classification methods, U-Net was the most commonly used (10 studies) 26 , 27 , 30 , 59 , 60 , 61 , 62 , 71 , 96 , 102 followed by DeepLabv3 ( n =7), 28 , 29 , 62 , 63 , 64 , 65 , 66 Mask-RCNN ( n =1), 16 and proprietary Visiopharm AI metastasis model 90 as can be seen in Fig. 4 B.…”

Pre-treatment peripheral blood immunophenotyping and response to neoadjuvant chemotherapy in operable breast cancer

“…The presence of high levels of TILs and of tumor-derived immune-related gene expression are associated with improved prognosis and therapeutic response, particularly in triple-negative and HER2 + breast cancer [ 1 – 6 , 23 ]. In addition, morphological immune features identified in regional lymph nodes are also prognostic in TNBC [ 24 , 25 ]. Based on the hypothesis that tumor-triggered immune responses can be detected not only in the tumor microenvironment and lymph nodes but also in the peripheral blood, this study utilized CyTOF to evaluate the circulating immune cell repertoire of patients with operable breast cancer before initiation of NAC and potential associations with response to NAC.…”

Spatial analyses of immune cell infiltration in cancer: current methods and future directions: A report of the International Immuno‐Oncology Biomarker Working Group on Breast Cancer

“…Superpixels is another method that segments tumors into raster cells according to similarities in adjacent pixel colors or other features [17,18]. Finally, approaches can be implemented that analyze tumors on different scales (multiscaling) or to optimize raster cell size as a hyperparameter, as it is effectively changing the resolution and accuracy of the statistical modeling [19].…”