Lymphocyte density determined by computational pathology validated as a predictor of response to neoadjuvant chemotherapy in breast cancer: secondary analysis of the ARTemis trial

The expression patterns of different cell types and their interactions in the tumor microenvironment are predictive of breast cancer patient response to neoadjuvant chemotherapy

The tumor microenvironment (TME) is a complex ecosystem of diverse cell types whose interactions govern tumor growth and clinical outcome. While the TME's impact on immunotherapy has been extensively studied, its role in chemotherapy response remains less explored. To address this, we developed DECODEM (DEcoupling Cell-type-specific Outcomes using DEconvolution and Machine learning), a generic computational framework leveraging cellular deconvolution of bulk transcriptomics to associate the gene expression of individual cell types in the TME with clinical response. Employing DECODEM to analyze the gene expression of breast cancer (BC) patients treated with neoadjuvant chemotherapy, we find that the gene expression of specific immune cells (myeloid, plasmablasts, B-cells) and stromal cells (endothelial, normal epithelial, CAFs) are highly predictive of chemotherapy response, going beyond that of the malignant cells. These findings are further tested and validated in a single-cell cohort of triple negative breast cancer. To investigate the possible role of immune cell-cell interactions (CCIs) in mediating chemotherapy response, we extended DECODEM to DECODEMi to identify such CCIs, validated in single-cell data. Our findings highlight the importance of active pre-treatment immune infiltration for chemotherapy success. The tools developed here are made publicly available and are applicable for studying the role of the TME in mediating response from readily available bulk tumor expression in a wide range of cancer treatments and indications.

Spatial Biology of Breast Cancer

Spatial findings have shaped on our understanding of breast cancer. In this review, we discuss how spatial methods, including spatial transcriptomics and proteomics and the resultant understanding of spatial relationships, have contributed to concepts regarding cancer progression and treatment. In addition to discussing traditional approaches, we examine how emerging multiplex imaging technologies have contributed to the field and how they might influence future research.

Effect of tumor-infiltrating immune cells (mast cells, neutrophils and lymphocytes) on neoadjuvant chemotherapy response in breast carcinomas

INTRODUCTION

Despite screening, the incidence of breast cancer is increasing worldwide. Neoadjuvant chemotherapy (NAC) response is one of the most important parameters taken into consideration in surgery, optimal adjuvant chemotherapy planning and prognosis prediction. Research on predictive markers for the response to NAC is still ongoing. In our study, we investigated the relationship between tumor-infiltrating neutrophils/mast cells/lymphocytes and NAC response in breast carcinomas.

MATERIAL AND METHOD

Study included 117 patients who were diagnosed with invasive breast carcinoma using core needle biopsy. In these biopsies tumor-infiltrating neutrophils/mast cells/lymphocytes were evaluated and Miller Payne Score was used for NAC response.

RESULT

53 patients exhibited high TILs, 36 had high TINs, and 46 showed high TIMs. While pathological complete response was 27 % in all patients, it was 38 % in high TINs patients, 35 % in high TILs patients, and 28 % in high TIMs patients. High TIMs were observed to be statistically associated with survival. TILs, TINs, nuclear grade, ER, PR and HER2 expression, Ki-67 proliferation index were found to be associated with the Miller - Payne score. In multivariate analysis, TINs, nuclear grade, pathological stage, and molecular subtype were found to be independent risk factors for treatment response.

CONCLUSION

TINs have better prognostic value to predict neoadjuvant treatment than TILs. High TIMs are associated with increased overall survival. The inclusion of TINs in NAC response and TIMs in overall survival in pathology reports and treatment planning is promising in breast carcinomas as they are simple to use and reproducible markers.

Exploration of the relationship between tumor-infiltrating lymphocyte score and histological grade in breast cancer

BACKGROUND

The histological grade is an important factor in the prognosis of invasive breast cancer and is vital to accurately identify the histological grade and reclassify of Grade2 status in breast cancer patients.

METHODS

In this study, data were collected from 556 invasive breast cancer patients, and then randomly divided into training cohort (n = 335) and validation cohort (n = 221). All patients were divided into actual low risk group (Grade1) and high risk group (Grade2/3) based on traditional histological grade, and tumor-infiltrating lymphocyte score (TILs-score) obtained from multiphoton images, and the TILs assessment method proposed by International Immuno-Oncology Biomarker Working Group (TILs-WG) were also used to differentiate between high risk group and low risk group of histological grade in patients with invasive breast cancer. Furthermore, TILs-score was used to reclassify Grade2 (G2) into G2 /Low risk and G2/High risk. The coefficients for each TILs in the training cohort were retrieved using ridge regression and TILs-score was created based on the coefficients of the three kinds of TILs.

RESULTS

Statistical analysis shows that TILs-score is significantly correlated with histological grade, and is an independent predictor of histological grade (odds ratio [OR], 2.548; 95%CI, 1.648-3.941; P < 0.0001), but TILs-WG is not an independent predictive factor for grade (P > 0.05 in the univariate analysis). Moreover, the risk of G2/High risk group is higher than that of G2/Low risk group, and the survival rate of patients with G2/Low risk is similar to that of Grade1, while the survival rate of patients with G2/High risk is even worse than that of patients with G3.

CONCLUSION

Our results suggest that TILs-score can be used to predict the histological grade of breast cancer and potentially to guide the therapeutic management of breast cancer patients.

Deep learning-based predictive model for pathological complete response to neoadjuvant chemotherapy in breast cancer from biopsy pathological images: a multicenter study

Introduction: Early predictive pathological complete response (pCR) is beneficial for optimizing neoadjuvant chemotherapy (NAC) strategies for breast cancer. The hematoxylin and eosin (HE)-stained slices of biopsy tissues contain a large amount of information on tumor epithelial cells and stromal. The fusion of pathological image features and clinicopathological features is expected to build a model to predict pCR of NAC in breast cancer. Methods: We retrospectively collected a total of 440 breast cancer patients from three hospitals who underwent NAC. HE-stained slices of biopsy tissues were scanned to form whole-slide images (WSIs), and pathological images of representative regions of interest (ROI) of each WSI were selected at different magnifications. Based on several different deep learning models, we propose a novel feature extraction method on pathological images with different magnifications. Further, fused with clinicopathological features, a multimodal breast cancer NAC pCR prediction model based on a support vector machine (SVM) classifier was developed and validated with two additional validation cohorts (VCs). Results: Through experimental validation of several different deep learning models, we found that the breast cancer pCR prediction model based on the SVM classifier, which uses the VGG16 model for feature extraction of pathological images at ×20 magnification, has the best prediction efficacy. The area under the curve (AUC) of deep learning pathological model (DPM) were 0.79, 0.73, and 0.71 for TC, VC1, and VC2, respectively, all of which exceeded 0.70. The AUCs of clinical model (CM), a clinical prediction model established by using clinicopathological features, were 0.79 for TC, 0.73 for VC1, and 0.71 for VC2, respectively. The multimodal deep learning clinicopathological model (DPCM) established by fusing pathological images and clinicopathological features improved the AUC of TC from 0.79 to 0.84. The AUC of VC2 improved from 0.71 to 0.78. Conclusion: Our study reveals that pathological images of HE-stained slices of pre-NAC biopsy tissues can be used to build a pCR prediction model. Combining pathological images and clinicopathological features can further enhance the predictive efficacy of the model.

Combining the tumor-stroma ratio with tumor-infiltrating lymphocytes improves the prediction of pathological complete response in breast cancer patients

PURPOSE

The tumor-stroma ratio (TSR) is a common histological parameter that measures stromal abundance and is prognostic in breast cancer (BC). However, more evidence is needed on the predictive value of the TSR for the pathological complete response (pCR) to neoadjuvant chemotherapy (NAC). The purpose of this study was to determine the importance of the TSR in predicting pCR in NAC settings.

METHOD

We evaluated the TSR on pretreatment biopsies of 912 BC patients from four independent Chinese hospitals and investigated the potential value of the TSR for predicting pCR. Meanwhile, stromal tumor-infiltrating lymphocytes (sTILs) were assessed, and we evaluated the predictive value of the combination of sTILs and TSR (TSRILs).

RESULTS

Patients with low stroma showed a higher pCR rate than those with high stroma among the four independent hospitals, and in multivariate analysis, the TSR was proven to be an independent predictor for pCR to NAC with an odds ratio of 1.945 (95% CI 1.230-3.075, P = 0.004). Moreover, we found that TSRILs could improve the area under the curve (AUC) for predicting pCR from 0.750 to 0.785 (P = 0.039); especially in HER2-negative BCs, the inclusion of TSRILs increased the AUC from 0.801 to 0.835 in the discovery dataset (P = 0.048) and 0.734 to 0.801 in the validation dataset (P = 0.003).

CONCLUSION

TSR and sTILs can be easily measured in pathological routines and provide predictive information without additional cost; with more evidence from clinical trials, TSRILs could be a candidate to better stratify patients in NAC settings.

Advances in the application of computational pathology in diagnosis, immunomicroenvironment recognition, and immunotherapy evaluation of breast cancer: a narrative review

BACKGROUND

Breast cancer (BC) is a prevalent and highly lethal malignancy affecting women worldwide. Immunotherapy has emerged as a promising therapeutic strategy for BC, offering potential improvements in patient survival. Neoadjuvant therapy (NAT) has also gained significant clinical traction. With the advancement of computer technology, Artificial Intelligence (AI) has been increasingly applied in pathology research, expanding and redefining the scope of the field. This narrative review aims to provide a comprehensive overview of the current literature on the application of computational pathology in BC, specifically focusing on diagnosis, immune microenvironment recognition, and the evaluation of immunotherapy and NAT response.

METHODS

A thorough examination of relevant literature was conducted, focusing on studies investigating the role of computational pathology in BC diagnosis, immune microenvironment recognition, and immunotherapy and NAT assessment.

RESULTS

The application of computational pathology has shown significant potential in BC management. AI-based techniques enable improved diagnosis and classification of BC subtypes, enhance the identification and characterization of the immune microenvironment, and facilitate the evaluation of immunotherapy and NAT response. However, challenges related to data quality, standardization, and algorithm development still need to be addressed.

CONCLUSION

The integration of computational pathology and AI has transformative implications for BC patient care. By leveraging AI-based technologies, clinicians can make more informed decisions in diagnosis, treatment planning, and therapeutic response assessment. Future research should focus on refining AI algorithms, addressing technical challenges, and conducting large-scale clinical validation studies to facilitate the translation of computational pathology into routine clinical practice for BC patients.

Radiosensitivity is associated with antitumor immunity in estrogen receptor-negative breast cancer

PURPOSE

This study evaluated radiosensitivity and the tumor microenvironment (TME) to identify characteristics of breast cancer patients who would benefit most from radiation therapy.

METHODS

We analyzed 1903 records from the Molecular Taxonomy of Breast Cancer International Consortium cohort using the radiosensitivity index and gene expression deconvolution algorithms, CIBERSORT and xCell, that estimates the TME composition of tumor samples. In this study, patients were stratified according to TME and radiosensitivity. We performed integrative analyses of clinical and immuno-genomic data to characterize molecular features associated with radiosensitivity.

RESULTS

Radiosensitivity was significantly associated with activation of antitumor immunity. In contrast, radioresistance was associated with a reactive stromal microenvironment. The immuno-genomic analysis revealed that estrogen receptor (ER) pathway activity was correlated with suppression of antitumor immunity. In ER-negative disease, the best prognosis was shown in the immune-high and radiosensitive group patients, and the lowest was in the immune-low and radioresistant group patients. In ER-positive disease, immune signature and radiosensitivity had no prognostic significance.

CONCLUSION

Taken together, these results suggest that tumor radiosensitivity is associated with activation of antitumor immunity and a better prognosis, particularly in patients with ER-negative breast cancer.

A spatial attention guided deep learning system for prediction of pathological complete response using breast cancer histopathology images

MOTIVATION

Predicting pathological complete response (pCR) to neoadjuvant chemotherapy (NAC) in triple-negative breast cancer (TNBC) patients accurately is direly needed for clinical decision making. pCR is also regarded as a strong predictor of overall survival. In this work, we propose a deep learning system to predict pCR to NAC based on serial pathology images stained with hematoxylin and eosin and two immunohistochemical biomarkers (Ki67 and PHH3). To support human prior domain knowledge-based guidance and enhance interpretability of the deep learning system, we introduce a human knowledge-derived spatial attention mechanism to inform deep learning models of informative tissue areas of interest. For each patient, three serial breast tumor tissue sections from biopsy blocks were sectioned, stained in three different stains and integrated. The resulting comprehensive attention information from the image triplets is used to guide our prediction system for prognostic tissue regions.

RESULTS

The experimental dataset consists of 26 419 pathology image patches of 1000×1000 pixels from 73 TNBC patients treated with NAC. Image patches from randomly selected 43 patients are used as a training dataset and images patches from the rest 30 are used as a testing dataset. By the maximum voting from patch-level results, our proposed model achieves a 93% patient-level accuracy, outperforming baselines and other state-of-the-art systems, suggesting its high potential for clinical decision making.

AVAILABILITY AND IMPLEMENTATION

The codes, the documentation and example data are available on an open source at: https://github.com/jkonglab/PCR_Prediction_Serial_WSIs_biomarkers.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

The Impact of Tumor Infiltrating Lymphocytes Densities and Ki67 Index on Residual Breast Cancer Burden following Neoadjuvant Chemotherapy

To avoid unnecessary neoadjuvant chemotherapy in case anticipating a poor therapy response, it is essential to find the pathological parameters that would predict pathological complete response or at least a decrease in tumor burden following neoadjuvant chemotherapy. The purpose of this study is to investigate the hypothesis that tumor infiltrating lymphocytes can predict the efficacy of neoadjuvant chemotherapy and to find the Ki67 cutoff value that best predicts the benefit of chemotherapy. 153 cases of breast cancer were chosen, based on their molecular subtype: triple negative subtype (77) and luminal, HER2-ve subtype (76). Histopathological assessment of pretherapy core biopsies was conducted to assess variable pathological parameters including TILs rates with the aid of immunohistochemical staining for CD20 and CD3. Moreover, core biopsies were stained for Ki67, and the findings were compared to the residual cancer burden following neoadjuvant chemotherapy. On analyzing and contrasting the two groups, a significant association between molecular subtype and pathological complete response was confirmed, while tumor-infiltrating lymphocytes in either group had no effect on therapy response. We used receiver operating characteristic curve analysis to determine that a cutoff of 36% for Ki67 is the most accurate value to predict complete therapy response.

The effect of neoadjuvant chemotherapy on tumor-infiltrating lymphocytes in patients with breast cancer

Aim: This study investigated the effect of neoadjuvant chemotherapy (NAC) on stromal tumor-infiltrating lymphocytes (sTILs) and their treatment response. Materials & methods: One hundred fifteen patients with pre-NAC core biopsies and post-NAC surgical resection specimens were reviewed. Results: There was no significant change between pre- and post-treatment sTILs. Both pre- and post-NAC sTILs were significantly lower in patients with luminal A subtype. An increase in sTILs was observed in 21 (25.9%) patients after NAC, a decrease in 29 (35.8%) and no change in 31 (38.3%; p = 0.07). Pretreatment sTIL density was independent predictor of pathological complete response in multivariate analyses (odds ratio: 1.025, 95% CI: 1.003-1.047; p = 0.023). Conclusion: High sTIL density in core biopsies was independently related to pathological complete response. In addition, ER appears to be the most crucial factor determining the rate of sTIL.

Immune Checkpoint Blockade in Hormone Receptor-Positive Breast Cancer: Resistance Mechanisms and Future Perspectives

Anti-programmed cell death protein 1 immunotherapy has been incorporated in the treatment algorithm of triple-negative breast cancer (TNBC). However, clinical trial results for patients with hormone receptor (HR)-positive disease appear less compelling. HR-positive tumors exhibit lower levels of programmed death-ligand 1 expression in comparison with their triple-negative counterparts. Moreover, signaling through estrogen receptor alters the immune microenvironment, rendering such tumors immunologically "cold." To explain differential responses to immune checkpoint blockade, this review interrogates differences between HR-positive and TNBC. Starting from distinct genomic features, we further present disparities concerning the tumor microenvironment and finally, we summarize early-phase clinical trial results on promising novel immunotherapy combinations.

Dissecting Tumor-Immune Microenvironment in Breast Cancer at a Spatial and Multiplex Resolution

The tumor immune microenvironment (TIME) is an important player in breast cancer pathophysiology. Surrogates for antitumor immune response have been explored as predictive biomarkers to immunotherapy, though with several limitations. Immunohistochemistry for programmed death ligand 1 suffers from analytical problems, immune signatures are devoid of spatial information and histopathological evaluation of tumor infiltrating lymphocytes exhibits interobserver variability. Towards improved understanding of the complex interactions in TIME, several emerging multiplex in situ methods are being developed and gaining much attention for protein detection. They enable the simultaneous evaluation of multiple targets in situ, detection of cell densities/subpopulations as well as estimations of functional states of immune infiltrate. Furthermore, they can characterize spatial organization of TIME—by cell-to-cell interaction analyses and the evaluation of distribution within different regions of interest and tissue compartments—while digital imaging and image analysis software allow for reproducibility of the various assays. In this review, we aim to provide an overview of the different multiplex in situ methods used in cancer research with special focus on breast cancer TIME at the neoadjuvant, adjuvant and metastatic setting. Spatial heterogeneity of TIME and importance of longitudinal evaluation of TIME changes under the pressure of therapy and metastatic progression are also addressed.

Negative Relationship between Post-Treatment Stromal Tumor-Infiltrating Lymphocyte (TIL) and Survival in Triple-Negative Breast Cancer Patients Treated with Dose-Dense Dose-Intense NeoAdjuvant Chemotherapy

Background: Patients with triple-negative breast cancers (TNBC) have a poor prognosis unless a pathological complete response (pCR) is achieved after neoadjuvant chemotherapy (NAC). Few studies have analyzed changes in TIL levels following dose-dense dose-intense (dd-di) NAC. Patients and methods: From 2009 to 2018, 117 patients with TNBC received dd-di NAC at our institution. We aimed to identify factors associated with pre- and post-NAC TIL levels, and oncological outcomes relapse-free survival (RFS), and overall survival (OS). Results: Median pre-NAC and post-NAC TIL levels were 15% and 3%, respectively. Change in TIL levels with treatment was significantly correlated with metabolic response (SUV) and pCR. High post-NAC TIL levels were associated with a weak metabolic response after two cycles of NAC, with the presence of residual disease and nodal involvement at NAC completion. In multivariate analyses, high post-NAC TIL levels independently predicted poor RFS and poor OS (HR = 1.4 per 10% increment, 95%CI (1.1; 1.9) p = 0.014 and HR = 1.8 per 10% increment 95%CI (1.3−2.3), p < 0.0001, respectively). Conclusion: Our results suggest that TNBC patients with TIL enrichment after NAC are at higher risk of relapse. These patients are potential candidates for adjuvant treatment, such as immunotherapy, in clinical trials.

Multi-omic machine learning predictor of breast cancer therapy response

Breast cancers are complex ecosystems of malignant cells and the tumour microenvironment1. The composition of these tumour ecosystems and interactions within them contribute to responses to cytotoxic therapy2. Efforts to build response predictors have not incorporated this knowledge. We collected clinical, digital pathology, genomic and transcriptomic profiles of pre-treatment biopsies of breast tumours from 168 patients treated with chemotherapy with or without HER2 (encoded by ERBB2)-targeted therapy before surgery. Pathology end points (complete response or residual disease) at surgery3 were then correlated with multi-omic features in these diagnostic biopsies. Here we show that response to treatment is modulated by the pre-treated tumour ecosystem, and its multi-omics landscape can be integrated in predictive models using machine learning. The degree of residual disease following therapy is monotonically associated with pre-therapy features, including tumour mutational and copy number landscapes, tumour proliferation, immune infiltration and T cell dysfunction and exclusion. Combining these features into a multi-omic machine learning model predicted a pathological complete response in an external validation cohort (75 patients) with an area under the curve of 0.87. In conclusion, response to therapy is determined by the baseline characteristics of the totality of the tumour ecosystem captured through data integration and machine learning. This approach could be used to develop predictors for other cancers.

TMB and TCR Are Correlated Indicators Predictive of the Efficacy of Neoadjuvant Chemotherapy in Breast Cancer

Immune characteristics were reported correlated to benefit neoadjuvant chemotherapy (NAC) in breast cancer, yet integration of comprehensive genomic alterations and T-cell receptors (TCR) to predict efficacy of NAC needs further investigation. This study simultaneously analyzed TMB (Tumor Mutation Burden), TCRs, and TILs (tumor infiltrating lymphocyte) in breast cancers receiving NAC was conducted in a prospective cohort (n = 22). The next-generation sequencing technology-based analysis of genomic alterations and TCR repertoire in paired breast cancer samples before and after NAC was conducted in a prospective cohort (n = 22). Fluorescent multiplex immunohistochemistry was used to stain CD4, CD8, PD1, TIM3, and cytokeratins simultaneously in those paired samples. TMB in pretreatment tumor tissues and TCR diversity index are higher in non-pCR patients than in pCR patients (10.6 vs. 2.3; p = 0.043) (2.066 vs. 0.467; p = 0.010). TMB and TCR diversity index had linear correlation (y = 5.587x − 0.881; r = 0.522, p = 0.012). Moreover, infiltrating T cells are significantly at higher presence in pCR versus non-pCR patients. Dynamically, the TMB reduced significantly after therapy in non-pCR patients (p = 0.010) but without TCR index change. The CDR3 peptide AWRSAGNYNEQF is the most highly expressed in pre-NAC samples of pCR patients and in post-NAC samples of non-pCR patients. In addition to pCR, high clonality of TCR and high level of CD8+ expression are associated with disease-free survival (DFS). TCR index and TMB have significant interaction and may guide neo-adjuvant treatment in operable breast cancers. Response to NAC in tumors with high TCR clonality may be attributable to high infiltration and expansion of tumor-specific CD8 positive effector cells.

Deep learning-based predictive biomarker of pathological complete response to neoadjuvant chemotherapy from histological images in breast cancer

Background

Pathological complete response (pCR) is considered a surrogate endpoint for favorable survival in breast cancer patients treated with neoadjuvant chemotherapy (NAC). Predictive biomarkers of treatment response are crucial for guiding treatment decisions. With the hypothesis that histological information on tumor biopsy images could predict NAC response in breast cancer, we proposed a novel deep learning (DL)-based biomarker that predicts pCR from images of hematoxylin and eosin (H&E)-stained tissue and evaluated its predictive performance.

Methods

In total, 540 breast cancer patients receiving standard NAC were enrolled. Based on H&E-stained images, DL methods were employed to automatically identify tumor epithelium and predict pCR by scoring the identified tumor epithelium to produce a histopathological biomarker, the pCR-score. The predictive performance of the pCR-score was assessed and compared with that of conventional biomarkers including stromal tumor-infiltrating lymphocytes (sTILs) and subtype.

Results

The pCR-score derived from H&E staining achieved an area under the curve (AUC) of 0.847 in predicting pCR directly, and achieved accuracy, F1 score, and AUC of 0.853, 0.503, and 0.822 processed by the logistic regression method, respectively, higher than either sTILs or subtype; a prediction model of pCR constructed by integrating sTILs, subtype and pCR-score yielded a mean AUC of 0.890, outperforming the baseline sTIL-subtype model by 0.051 (0.839, P  =  0.001).

Conclusion

The DL-based pCR-score from histological images is predictive of pCR better than sTILs and subtype, and holds the great potentials for a more accurate stratification of patients for NAC.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12967-021-03020-z.

Quantitative Image Analysis for Tissue Biomarker Use: A White Paper From the Digital Pathology Association

Tissue biomarkers have been of increasing utility for scientific research, diagnosing disease, and treatment response prediction. There has been a steady shift away from qualitative assessment toward providing more quantitative scores for these biomarkers. The application of quantitative image analysis has thus become an indispensable tool for in-depth tissue biomarker interrogation in these contexts. This white paper reviews current technologies being employed for quantitative image analysis, their application and pitfalls, regulatory framework demands, and guidelines established for promoting their safe adoption in clinical practice.

The Immunology of Hormone Receptor Positive Breast Cancer

Immune checkpoint blockade (ICB) has revolutionized the treatment of cancer patients. The main focus of ICB has been on reinvigorating the adaptive immune response, namely, activating cytotoxic T cells. ICB has demonstrated only modest benefit against advanced breast cancer, as breast tumors typically establish an immune suppressive tumor microenvironment (TME). Triple-negative breast cancer (TNBC) is associated with infiltration of tumor infiltrating lymphocytes (TILs) and patients with TNBC have shown clinical responses to ICB. In contrast, hormone receptor positive (HR+) breast cancer is characterized by low TIL infiltration and minimal response to ICB. Here we review how HR+ breast tumors establish a TME devoid of TILs, have low HLA class I expression, and recruit immune cells, other than T cells, which impact response to therapy. In addition, we review emerging technologies that have been employed to characterize components of the TME to reveal that tumor associated macrophages (TAMs) are abundant in HR+ cancer, are highly immune-suppressive, associated with tumor progression, chemotherapy and ICB-resistance, metastasis and poor survival. We reveal novel therapeutic targets and possible combinations with ICB to enhance anti-tumor immune responses, which may have great potential in HR+ breast cancer.

A review and comparison of breast tumor cell nuclei segmentation performances using deep convolutional neural networks

Breast cancer is currently the second most common cause of cancer-related death in women. Presently, the clinical benchmark in cancer diagnosis is tissue biopsy examination. However, the manual process of histopathological analysis is laborious, time-consuming, and limited by the quality of the specimen and the experience of the pathologist. This study's objective was to determine if deep convolutional neural networks can be trained, with transfer learning, on a set of histopathological images independent of breast tissue to segment tumor nuclei of the breast. Various deep convolutional neural networks were evaluated for the study, including U-Net, Mask R-CNN, and a novel network (GB U-Net). The networks were trained on a set of Hematoxylin and Eosin (H&E)-stained images of eight diverse types of tissues. GB U-Net demonstrated superior performance in segmenting sites of invasive diseases (AJI = 0.53, mAP = 0.39 & AJI = 0.54, mAP = 0.38), validated on two hold-out datasets exclusively containing breast tissue images of approximately 7,582 annotated cells. The results of the networks, trained on images independent of breast tissue, demonstrated that tumor nuclei of the breast could be accurately segmented.

Introduction to Artificial Intelligence and Machine Learning for Pathology

CONTEXT.—

Recent developments in machine learning have stimulated intense interest in software that may augment or replace human experts. Machine learning may impact pathology practice by offering new capabilities in analysis, interpretation, and outcomes prediction using images and other data. The principles of operation and management of machine learning systems are unfamiliar to pathologists, who anticipate a need for additional education to be effective as expert users and managers of the new tools.

OBJECTIVE.—

To provide a background on machine learning for practicing pathologists, including an overview of algorithms, model development, and performance evaluation; to examine the current status of machine learning in pathology and consider possible roles and requirements for pathologists in local deployment and management of machine learning systems; and to highlight existing challenges and gaps in deployment methodology and regulation.

DATA SOURCES.—

Sources include the biomedical and engineering literature, white papers from professional organizations, government reports, electronic resources, and authors' experience in machine learning. References were chosen when possible for accessibility to practicing pathologists without specialized training in mathematics, statistics, or software development.

CONCLUSIONS.—

Machine learning offers an array of techniques that in recent published results show substantial promise. Data suggest that human experts working with machine learning tools outperform humans or machines separately, but the optimal form for this combination in pathology has not been established. Significant questions related to the generalizability of machine learning systems, local site verification, and performance monitoring remain to be resolved before a consensus on best practices and a regulatory environment can be established.

Impact of BRCA Mutation Status on Tumor Infiltrating Lymphocytes (TILs), Response to Treatment, and Prognosis in Breast Cancer Patients Treated with Neoadjuvant Chemotherapy

INTRODUCTION

Five to 10% of breast cancers (BCs) occur in a genetic predisposition context (mainly BRCA pathogenic variant). Nevertheless, little is known about immune tumor infiltration, response to neoadjuvant chemotherapy (NAC), pathologic complete response (pCR) and adverse events according to BRCA status.

MATERIAL AND METHODS

Out of 1199 invasive BC patients treated with NAC between 2002 and 2012, we identified 267 patients tested for a germline BRCA pathogenic variant. We evaluated pre-NAC and post-NAC immune infiltration (TILs). Response to chemotherapy was assessed by pCR rates. Association of clinical and pathological factors with TILs, pCR and survival was assessed by univariate and multivariate analyses.

RESULTS

Among 1199 BC patients: 46 were BRCA-deficient and 221 BRCA-proficient or wild type (WT). At NAC completion, pCR was observed in 84/266 (31%) patients and pCR rates were significantly higher in BRCA-deficient BC (p = 0.001), and this association remained statistically significant only in the luminal BC subtype (p = 0.006). The interaction test between BC subtype and BRCA status was nearly significant (P_interaction = 0.056). Pre and post-NAC TILs were not significantly different between BRCA-deficient and BRCA-proficient carriers; however, in the luminal BC group, post-NAC TILs were significantly higher in BRCA-deficient BC. Survival analysis were not different between BRCA-carriers and non-carriers.

CONCLUSIONS

BRCA mutation status is associated with higher pCR rates and post-NAC TILs in patients with luminal BC. BRCA-carriers with luminal BCs may represent a subset of patients deriving higher benefit from NAC. Second line therapies, including immunotherapy after NAC, could be of interest in non-responders to NAC.

Quantitative multiplex immunofluorescence analysis identifies infiltrating PD1+ CD8+ and CD8+ T cells as predictive of response to neoadjuvant chemotherapy in breast cancer

Background

This study aimed to explore the potentially predictive role and dynamic changes of immune checkpoints on T cell subsets in patients with breast cancer receiving neoadjuvant chemotherapies.

Methods

Fluorescent multiplex immunohistochemistry (mIHC) was used to stain CD4, CD8, PD1, TIM3, and cytokeratins simultaneously in paired breast cancer samples before and after neoadjuvant therapies (NAT) in a prospective cohort (n = 50). Singleplex IHC was conducted to stain for CD3 in 100 cases with inclusion of extra retrospective 50 cases. Cell levels were correlated with clinicopathological parameters and pathological complete response (pCR).

Results

In pretreatment tumors, the percentages of infiltrating CD8⁺, PD1⁺, PD1⁺CD8⁺, and the ratio of PD1⁺CD8⁺/CD8⁺ cells, were higher in pCR than non‐pCR patients in either the stromal or intratumoral area, but PD1⁺CD4⁺, TIM3⁺CD4⁺, TIM3⁺CD8⁺ cells and CD4⁺/CD8⁺ ratio was not. Multivariate analyses showed that the percentage of intratumoral CD8⁺ cells (OR, 1.712; 95% CI: 1.052–2.786; P = 0.030) and stromal PD1⁺CD8⁺/CD8⁺ ratio (OR, 1.109; 95% CI: 1.009–1.218; P = 0.032) were significantly associated with pCR. Dynamically, reduction in the percentages of PD1⁺, CD8⁺ and PD1⁺CD8⁺ cells after therapy strongly correlated with pCR. Notably, incremental percentages of PD1⁺CD8⁺ cells, rather than TIM3⁺CD8⁺, were shown in tumors from non‐pCR patients after NAT. CD3 staining confirmed the percentage of T cells were associated with pCR.

Conclusions

PD1⁺CD8⁺ rather than TIM3⁺CD8⁺ cells are main predictive components within tumor‐infiltrating T cells in NAT breast cancer patients. Dynamically incremental levels of PD1⁺CD8⁺ cells occurred in non‐pCR cases after NAT, suggesting the combination of chemotherapy with PD1 inhibition might benefit these patients.

Key points

Significant findings of the study

PD1⁺CD8⁺, rather than TIM3⁺CD8⁺, T cells are the main component to predict the response of neoadjuvant therapies in breast cancer.

What this study adds

Incremental levels of PD1⁺CD8⁺ T cells in non‐pCR post‐NAT tumors suggest PD1 inhibition might benefit in the neoadjuvant setting.

Priming the tumor immune microenvironment with chemo(radio)therapy: A systematic review across tumor types

BACKGROUND

Chemotherapy (CT), radiotherapy (RT), and chemoradiotherapy (CRT) are able to alter the composition of the tumor immune microenvironment (TIME). Understanding the effect of these modalities on the TIME could aid in the development of improved treatment strategies. Our aim was to systematically review studies investigating the influence of CT, RT or CRT on different TIME markers.

METHODS

The EMBASE (Ovid) and PubMed databases were searched until January 2019 for prospective or retrospective studies investigating the dynamics of the local TIME in cancer patients (pts) treated with CT, RT or CRT, with or without targeted agents. Studies could either compare baseline and follow-up specimens - before and after treatment - or a treated versus an untreated cohort. Studies were included if they used immunohistochemistry and/or flow cytometry to assess the TIME.

RESULTS

In total we included 110 studies (n = 8850 pts), of which n = 89 (n = 6295 pts) compared pre-treatment to post-treatment specimens and n = 25 (n = 2555 pts) a treated versus an untreated cohort (4 studies conducted both comparisons). For several tumor types (among others; breast, cervical, esophageal, ovarian, rectal, lung mesothelioma and pancreatic cancer) remodeling of the TIME was observed, leading to a potentially more immunologically active microenvironment, including one or more of the following: an increase in CD3 or CD8 lymphocytes, a decrease in FOXP3 Tregs and increased PD-L1 expression. Both CT and CRT were able to immunologically alter the TIME.

CONCLUSION

The TIME of several tumor types is significantly altered after conventional therapy creating opportunities for concurrent or sequential immunotherapy.

Combining measures of immune infiltration shows additive effect on survival prediction in high-grade serous ovarian carcinoma

BACKGROUND

In colorectal and breast cancer, the density and localisation of immune infiltrates provides strong prognostic information. We asked whether similar automated quantitation and combined analysis of immune infiltrates could refine prognostic information in high-grade serous ovarian carcinoma (HGSOC) and tested associations between patterns of immune response and genomic driver alterations.

METHODS

Epithelium and stroma were semi-automatically segmented and the infiltration of CD45RO+, CD8+ and CD68+ cells was automatically quantified from images of 332 HGSOC patient tissue microarray cores.

RESULTS

Epithelial CD8 [p = 0.027, hazard ratio (HR) = 0.83], stromal CD68 (p = 3 × 10-4, HR = 0.44) and stromal CD45RO (p = 7 × 10-4, HR = 0.76) were positively associated with survival and remained so when averaged across the tumour and stromal compartments. Using principal component analysis, we identified optimised multiparameter survival models combining information from all immune markers (p = 0.016, HR = 0.88). There was no significant association between PTEN expression, type of TP53 mutation or presence of BRCA1/BRCA2 mutations and immune infiltrate densities or principal components.

CONCLUSIONS

Combining measures of immune infiltration provided improved survival modelling and evidence for the multiple effects of different immune factors on survival. The presence of stromal CD68+ and CD45RO+ populations was associated with survival, underscoring the benefits evaluating stromal immune populations may bring for prognostic immunoscores in HGSOC.

Predictive factors and prognostic value of pathologic complete response of ipsilateral supraclavicular lymph nodes in breast cancer after neoadjuvant chemotherapy

Background

Breast cancer with ipsilateral supraclavicular lymph node metastasis is one of the indicators of poor prognosis. Patients who attain pathologic complete response in breast and axillary sites have improved survival and are highest in aggressive HR-HER2- and HER2-positive tumor subtypes. However, there is no study on the related factors and prognostic value of supraclavicular pathologic complete response in breast cancer after neoadjuvant chemotherapy. The aim of our work was to investigate the factors and prognostic significance of pathologic complete response of ipsilateral supraclavicular lymph node metastasis in breast cancer after neoadjuvant chemotherapy.

Methods

A total of 214 patients with breast cancer who had primary ISLN metastasis, receiving NAC and subsequent ISLN dissection, were retrospectively and consecutively reviewed. Univariate and multivariate analyses were performed using χ² test and the logistic regression model, and the prognosis was analyzed by Kaplan-Meier curve.

Results

All patients included were women who were 26-74 years old. The rate of supraclavicular pathologic complete response (pCR) was 53.7%. Multivariate analysis showed that the expression of Ki67, breast pCR, and axillary pCR were independent predictors of supraclavicular pCR (P<0.05). After a median follow-up of 16.2 months, the risk of recurrence and metastasis in patients with supraclavicular pCR was half reduced compared to that of the non-pCR group (HR 0.51, 95% CI, 0.32-0.80, P<0.01), mainly manifested in HR-HER2- and HER2-positive disease.

Conclusions

The expression level of Ki67, breast pCR, and axillary pCR were independent predictors of supraclavicular pCR. Supraclavicular pCR was an independent predictor of disease-free survival (DFS). Surgical removal of supraclavicular lymph nodes can accurately evaluate the rate of supraclavicular pCR, which is of great significance for patient prognosis.

Interaction between Molecular Subtypes and Stromal Immune Infiltration before and after Treatment in Breast Cancer Patients Treated with Neoadjuvant Chemotherapy

PURPOSE

High levels of tumor-infiltrating lymphocytes (TIL) before neoadjuvant chemotherapy (NAC) are associated with higher pathologic complete response (pCR) rates and better survival in triple-negative breast cancer (TNBC) and HER2-positive breast cancer. We investigated the value of TIL levels by evaluating lymphocyte infiltration before and after NAC.

EXPERIMENTAL DESIGN

We assessed stromal TIL levels in 716 pre- and posttreatment matched paired specimens, according to the guidelines of the International TIL Working Group.

RESULTS

Pre-NAC TIL levels were higher in tumors for which pCR was achieved than in cases with residual disease (33.9% vs. 20.3%, P = 0.001). This was observed in luminal tumors and TNBCs, but not in HER2-positive breast cancers (P _Interaction = 0.001). The association between pre-NAC TIL levels and pCR was nonlinear in TNBCs (P = 0.005). Mean TIL levels decreased after chemotherapy completion (pre-NAC TILs: 24.1% vs. post-NAC TILs: 13.0%, P < 0.001). This decrease was strongly associated with high pCR rates, and the variation of TIL levels was strongly inversely correlated with pre-NAC TIL levels (r = -0.80, P < 0.001). Pre-NAC TILs and disease-free survival (DFS) were associated in a nonlinear manner (P < 0.001). High post-NAC TIL levels were associated with aggressive tumor characteristics and with impaired DFS in HER2-positive breast cancers (HR, 1.04; confidence interval, 1.02-1.06; P = 0.001), but not in luminal tumors or TNBCs (P _Interaction = 0.04).

CONCLUSIONS

The associations of pre- and post-NAC TIL levels with response to treatment and DFS differ between breast cancer subtypes. The characterization of immune subpopulations may improve our understanding of the complex interactions between pre- or post-NAC setting, breast cancer subtype, response to treatment, and prognosis.

Update Breast Cancer 2019 Part 3 - Current Developments in Early Breast Cancer: Review and Critical Assessment by an International Expert Panel

The treatment of breast cancer patients in a curative situation is special in many ways. The local therapy with surgery and radiation therapy is a central aspect of the treatment. The complete elimination of tumour cells at the site of the primary disease must be ensured while simultaneously striving to keep the long-term effects as minor as possible. There is still focus on the continued reduction of the invasiveness of local therapy. With regard to systemic therapy, chemotherapies with taxanes, anthracyclines and, in some cases, platinum-based chemotherapies have become established in the past couple of decades. The context for use is being continually further defined. Likewise, there are questions in the case of antihormonal therapy which also still need to be further defined following the introduction of aromatase inhibitors, such as the length of therapy or ovarian suppression in premenopausal patients. Finally, personalisation of the treatment of early breast cancer patients is also being increasingly used. Prognostic tests could potentially support therapeutic decisions. It must also be considered how the possible use of new therapies, such as checkpoint inhibitors and CDK4/6 inhibitors could look in practice once study results in this regard are available. This overview addresses the backgrounds on the current votes taken by the international St. Gallen panel of experts in Vienna in 2019 for current questions in the treatment of breast cancer patients in a curative situation.