How can artificial intelligence models assist PD-L1 expression scoring in breast cancer: results of multi-institutional ring studies

Artificial intelligence enhances whole-slide interpretation of PD-L1 CPS in triple-negative breast cancer: A multi-institutional ring study

BACKGROUND AND AIMS

Evaluation of the programmed cell death ligand-1 (PD-L1) combined positive score (CPS) is vital to predict the efficacy of the immunotherapy in triple-negative breast cancer (TNBC), but pathologists show substantial variability in the consistency and accuracy of the interpretation. It is of great importance to establish an objective and effective method which is highly repeatable.

METHODS

We proposed a model in a deep learning-based framework, which at the patch level incorporated cell analysis and tissue region analysis, followed by the whole-slide level fusion of patch results. Three rounds of ring studies (RSs) were conducted. Twenty-one pathologists of different levels from four institutions evaluated the PD-L1 CPS in TNBC specimens as continuous scores by visual assessment and our artificial intelligence (AI)-assisted method.

RESULTS

In the visual assessment, the interpretation results of PD-L1 (Dako 22C3) CPS by different levels of pathologists have significant differences and showed weak consistency. Using AI-assisted interpretation, there were no significant differences between all pathologists (P = 0.43), and the intraclass correlation coefficient (ICC) value was increased from 0.618 [95% confidence interval (CI) = 0.524-0.719] to 0.931 (95% CI = 0.902-0.955). The accuracy of interpretation result is further improved to 0.919 (95% CI = 0.886-0.947). Acceptance of AI results by junior pathologists was the highest among all levels, and 80% of the AI results were accepted overall.

CONCLUSION

With the help of the AI-assisted diagnostic method, different levels of pathologists achieved excellent consistency and repeatability in the interpretation of PD-L1 (Dako 22C3) CPS. Our AI-assisted diagnostic approach was proved to strengthen the consistency and repeatability in clinical practice.

An algorithm-based technique for counting mitochondria in cells using immunohistochemical staining of formalin-fixed and paraffin-embedded sections

PURPOSE

Visualizing mitochondria in cancer cells from human pathological specimens may improve our understanding of cancer biology. However, using immunohistochemistry to evaluate mitochondria remains difficult because almost all cells contain mitochondria and the number of mitochondria per cell may have important effects on mitochondrial function. Herein, we established an objective system (Mito-score) for evaluating mitochondria using machine-based processing of hue, saturation, and value color spaces.

METHODS

The Mito-score was defined as the number of COX4 (mitochondrial inner membrane) immunohistochemistry-positive pixels divided by the number of nuclei per cell. The system was validated using four lung cancer cell lines, normal tissues, and lung cancer tissues (199 cases).

RESULTS

The Mito-score correlated with MitoTracker, a fluorescent dye used to selectively label and visualize mitochondria within cells under a microscope (R2 = 0.68) and with the number of mitochondria counted using electron microscopy (R2 = 0.79). Histologically, the Mito-score of small cell carcinoma (57.25) was significantly lower than that of adenocarcinoma (147.5, p < 0.0001), squamous cell carcinoma (120.6, p = 0.0004), and large cell neuroendocrine carcinoma (111.8, p = 0.002).

CONCLUSION

The Mito-score method enables the analysis of the mitochondrial status of human formalin-fixed paraffin-embedded specimens and may provide insights into the metabolic status of cancer.

Concordance of assessments of four PD-L1 immunohistochemical assays in esophageal squamous cell carcinoma (ESCC)

OBJECTIVE

Given real-world limitations in programmed death-ligand 1 (PD-L1) testing, concordance studies between PD-L1 assays are needed. We undertook comparisons of PD-L1 assays (DAKO22C3, Ventana SP263, Ventana SP142, E1L3N) among observers in esophageal squamous cell carcinoma (ESCC) to provide information on the analytical and clinical comparability of four PD-L1 IHC assays.

METHODS

Paraffin embedded samples of 50 cases of esophageal squamous cell carcinoma were obtained, satined with all four PD-L1 assays. PD-L1 was evaluated by 68 pathologists from 19 different hospitals. PD-L1 expression was assessed for combined positive score (CPS).

RESULTS

The expression sensitivity of SP263 was the highest in ESCC, followed by 22C3, E1L3N and SP142. Taking CPS 10 as the critical value, inter-observer concordance for CPS scores among 68 physicians was assessed for the 22C3, SP263, SP142, and E1L3N assays, yielding values of 0.777, 0.790, 0.758, and 0.782, respectively. In the comparison between assays, the overall CPS scores concordance rates between 22C3 and SP263, SP142, and E1L3N were 0.896, 0.833, and 0.853, respectively. 22C3 and SP263 have high concordance, with OPA of 0.896, while E1L3N and SP142 have the highest concordance, with OPA of 0.908.

CONCLUSION

In ESCC, the concordance of PD-L1 evaluation among observers is good, and the immune cell score is still an important factor affecting the concordance of interpretation among observers. Cases near the specific threshold are still the difficult problem of interpretation. SP263 had the highest CPS score of the four assays. SP263 cannot identify all 22C3 positive cases, but had good concordance with 22C3.E1L3N and SP142 showed high concordance.

Weakly Supervised Deep Learning Predicts Immunotherapy Response in Solid Tumors Based on PD-L1 Expression

Programmed death-ligand 1 (PD-L1) IHC is the most commonly used biomarker for immunotherapy response. However, quantification of PD-L1 status in pathology slides is challenging. Neither manual quantification nor a computer-based mimicking of manual readouts is perfectly reproducible, and the predictive performance of both approaches regarding immunotherapy response is limited. In this study, we developed a deep learning (DL) method to predict PD-L1 status directly from raw IHC image data, without explicit intermediary steps such as cell detection or pigment quantification. We trained the weakly supervised model on PD-L1-stained slides from the non-small cell lung cancer (NSCLC)-Memorial Sloan Kettering (MSK) cohort (N = 233) and validated it on the pan-cancer-Vall d'Hebron Institute of Oncology (VHIO) cohort (N = 108). We also investigated the performance of the model to predict response to immune checkpoint inhibitors (ICI) in terms of progression-free survival. In the pan-cancer-VHIO cohort, the performance was compared with tumor proportion score (TPS) and combined positive score (CPS). The DL model showed good performance in predicting PD-L1 expression (TPS ≥ 1%) in both NSCLC-MSK and pan-cancer-VHIO cohort (AUC 0.88 ± 0.06 and 0.80 ± 0.03, respectively). The predicted PD-L1 status showed an improved association with response to ICIs [HR: 1.5 (95% confidence interval: 1-2.3), P = 0.049] compared with TPS [HR: 1.4 (0.96-2.2), P = 0.082] and CPS [HR: 1.2 (0.79-1.9), P = 0.386]. Notably, our explainability analysis showed that the model does not just look at the amount of brown pigment in the IHC slides, but also considers morphologic factors such as lymphocyte conglomerates. Overall, end-to-end weakly supervised DL shows potential for improving patient stratification for cancer immunotherapy by analyzing PD-L1 IHC, holistically integrating morphology and PD-L1 staining intensity.

SIGNIFICANCE

The weakly supervised DL model to predict PD-L1 status from raw IHC data, integrating tumor staining intensity and morphology, enables enhanced patient stratification in cancer immunotherapy compared with traditional pathologist assessment.

Refining PD-1/PD-L1 assessment for biomarker-guided immunotherapy: A review

Anti-programmed cell death ligand 1 (anti-PD-L1)  immunotherapy is an increasingly crucial in cancer treatment. To date, the Federal Drug Administration (FDA) has approved four PD-L1 immunohistochemistry (IHC) staining protocols, commercially available in the form of "kits", facilitating testing for PD-L1 expression. These kits comprise four PD-L1 antibodies on two separate IHC platforms, each utilizing distinct, non-interchangeable scoring systems. Several factors, including tumor heterogeneity and the size of the tissue specimens assessed, can lead to PD-L1 status misclassification, potentially hindering the initiation of therapy. Therefore, the development of more accurate predictive biomarkers to distinguish between responders and non-responders prior to anti-PD-1/PD-L1 therapy warrants further research. Achieving this goal necessitates refining sampling criteria, enhancing current methods of PD-L1 detection, and deepening our understanding of the impact of additional biomarkers. In this article, we review potential solutions to improve the predictive accuracy of PD-L1 assessment in order to more precisely anticipate patients' responses to anti-PD-1/PD-L1 therapy, monitor disease progression and predict clinical outcomes.

Artificial Intelligence in Breast Cancer Diagnosis and Personalized Medicine

Breast cancer is a significant cause of cancer-related mortality in women worldwide. Early and precise diagnosis is crucial, and clinical outcomes can be markedly enhanced. The rise of artificial intelligence (AI) has ushered in a new era, notably in image analysis, paving the way for major advancements in breast cancer diagnosis and individualized treatment regimens. In the diagnostic workflow for patients with breast cancer, the role of AI encompasses screening, diagnosis, staging, biomarker evaluation, prognostication, and therapeutic response prediction. Although its potential is immense, its complete integration into clinical practice is challenging. Particularly, these challenges include the imperatives for extensive clinical validation, model generalizability, navigating the "black-box" conundrum, and pragmatic considerations of embedding AI into everyday clinical environments. In this review, we comprehensively explored the diverse applications of AI in breast cancer care, underlining its transformative promise and existing impediments. In radiology, we specifically address AI in mammography, tomosynthesis, risk prediction models, and supplementary imaging methods, including magnetic resonance imaging and ultrasound. In pathology, our focus is on AI applications for pathologic diagnosis, evaluation of biomarkers, and predictions related to genetic alterations, treatment response, and prognosis in the context of breast cancer diagnosis and treatment. Our discussion underscores the transformative potential of AI in breast cancer management and emphasizes the importance of focused research to realize the full spectrum of benefits of AI in patient care.

Computational pathology to improve biomarker testing in breast cancer: how close are we?

The recent advancements in breast cancer precision medicine have highlighted the urgency for the precise and reproducible characterization of clinically actionable biomarkers. Despite numerous standardization efforts, biomarker testing by conventional methodologies is challenged by several issues such as high inter-observer variabilities, the spatial heterogeneity of biomarkers expression, and technological heterogeneity. In this respect, artificial intelligence-based digital pathology approaches are being increasingly recognized as promising methods for biomarker testing and subsequently improved clinical management. Here, we provide an overview on the most recent advances for artificial intelligence-assisted biomarkers testing in breast cancer, with a particular focus on tumor-infiltrating lymphocytes, programmed death-ligand 1, phosphatidylinositol-3 kinase catalytic alpha, and estrogen receptor 1. Challenges and solutions for this integrative analysis in pathology laboratories are also provided.

Immune Biomarkers in Triple-Negative Breast Cancer: Improving the Predictivity of Current Testing Methods

Triple-negative breast cancer (TNBC) poses a significant challenge in terms of prognosis and disease recurrence. The limited treatment options and the development of resistance to chemotherapy make it particularly difficult to manage these patients. However, recent research has been shifting its focus towards biomarker-based approaches for TNBC, with a particular emphasis on the tumor immune landscape. Immune biomarkers in TNBC are now a subject of great interest due to the presence of tumor-infiltrating lymphocytes (TILs) in these tumors. This characteristic often coincides with the presence of PD-L1 expression on both neoplastic cells and immune cells within the tumor microenvironment. Furthermore, a subset of TNBC harbor mismatch repair deficient (dMMR) TNBC, which is frequently accompanied by microsatellite instability (MSI). All of these immune biomarkers hold actionable potential for guiding patient selection in immunotherapy. To fully capitalize on these opportunities, the identification of additional or complementary biomarkers and the implementation of highly customized testing strategies are of paramount importance in TNBC. In this regard, this article aims to provide an overview of the current state of the art in immune-related biomarkers for TNBC. Specifically, it focuses on the various testing methodologies available and sheds light on the immediate future perspectives for patient selection. By delving into the advancements made in understanding the immune landscape of TNBC, this study aims to contribute to the growing body of knowledge in the field. The ultimate goal is to pave the way for the development of more personalized testing strategies, ultimately improving outcomes for TNBC patients.

Artificial Intelligence in the Pathology of Gastric Cancer

Recent advances in artificial intelligence (AI) have provided novel tools for rapid and precise pathologic diagnosis. The introduction of digital pathology has enabled the acquisition of scanned slide images that are essential for the application of AI. The application of AI for improved pathologic diagnosis includes the error-free detection of potentially negligible lesions, such as a minute focus of metastatic tumor cells in lymph nodes, the accurate diagnosis of potentially controversial histologic findings, such as very well-differentiated carcinomas mimicking normal epithelial tissues, and the pathological subtyping of the cancers. Additionally, the utilization of AI algorithms enables the precise decision of the score of immunohistochemical markers for targeted therapies, such as human epidermal growth factor receptor 2 and programmed death-ligand 1. Studies have revealed that AI assistance can reduce the discordance of interpretation between pathologists and more accurately predict clinical outcomes. Several approaches have been employed to develop novel biomarkers from histologic images using AI. Moreover, AI-assisted analysis of the cancer microenvironment showed that the distribution of tumor-infiltrating lymphocytes was related to the response to the immune checkpoint inhibitor therapy, emphasizing its value as a biomarker. As numerous studies have demonstrated the significance of AI-assisted interpretation and biomarker development, the AI-based approach will advance diagnostic pathology.

The slow-paced digital evolution of pathology: lights and shadows from a multifaceted board

Objective

The digital revolution in pathology represents an invaluable resource fto optimise costs, reduce the risk of error and improve patient care, even though it is still adopted in a minority of laboratories. Barriers include concerns about initial costs, lack of confidence in using whole slide images for primary diagnosis, and lack of guidance on transition. To address these challenges and develop a programme to facilitate the introduction of digital pathology (DP) in Italian pathology departments, a panel discussion was set up to identify the key points to be considered.

Methods

On 21 July 2022, an initial conference call was held on Zoom to identify the main issues to be discussed during the face-to-face meeting. The final summit was divided into four different sessions: (I) the definition of DP, (II) practical applications of DP, (III) the use of AI in DP, (IV) DP and education.

Results

Essential requirements for the implementation of DP are a fully tracked and automated workflow, selection of the appropriate scanner based on the specific needs of each department, and a strong commitment combined with coordinated teamwork (pathologists, technicians, biologists, IT service and industries). This could reduce human error, leading to the application of AI tools for diagnosis, prognosis and prediction. Open challenges are the lack of specific regulations for virtual slide storage and the optimal storage solution for large volumes of slides.

Conclusion

Teamwork is key to DP transition, including close collaboration with industry. This will ease the transition and help bridge the gap that currently exists between many labs and full digitisation. The ultimate goal is to improve patient care.

Predicting Response of Triple-Negative Breast Cancer to Neoadjuvant Chemotherapy Using a Deep Convolutional Neural Network-Based Artificial Intelligence Tool

PURPOSE

Achieving a pathological complete response (pCR) to neoadjuvant chemotherapy (NAC) is associated with improved patient outcomes in triple-negative breast cancer (TNBC). Currently, there are no validated predictive biomarkers for the response to NAC in TNBC. We developed and validated a deep convolutional neural network-based artificial intelligence (AI) model to predict the response of TNBC to NAC.

MATERIALS AND METHODS

Whole-slide images (WSIs) of hematoxylin and eosin-stained core biopsies from 165 (pCR in 60 and non-pCR in 105) and 78 (pCR in 31 and non-pCR in 47) patients with TNBC were used to train and validate the model. The model extracts morphometric features from WSIs in an unsupervised manner, thereby generating clusters of morphologically similar patterns. Downstream ranking of clusters provided regions of interest and morphometric scores; a low score close to zero and a high score close to one represented a high or low probability of response to NAC.

RESULTS

The predictive ability of AI score for the entire cohort of 78 patients with TNBC ascertained by receiver operating characteristic analysis demonstrated an area under the curve (AUC) of 0.75. The AUC for stages I, II, and III disease were 0.88, 0.73, and 0.74, respectively. Using a cutoff value of 0.35, the positive predictive value of the AI score for pCR was 73.7%, and the negative predictive value was 76.2% for non-pCR patients.

CONCLUSION

To our knowledge, this study is the first to demonstrate the use of an AI tool on digitized hematoxylin and eosin-stained tissue images to predict the response to NAC in patients with TNBC with high accuracy. If validated in subsequent studies, these results may serve as an ancillary aid for individualized therapeutic decisions in patients with TNBC.

PD-L1 testing by immunohistochemistry in immuno-oncology

Immunotherapy, based on immune checkpoint inhibitors targeting the Programmed cell death ligand 1 (PD-L1) and/or Programmed Death Receptor 1 (PD-1), has substantially improved the outcomes of patients with various cancers. However, only ~30% of patients benefit from immune checkpoint inhibitors. Tumor PD-L1 expression, assessed by immunohistochemistry, is the most widely validated and used predictive biomarker to guide the selection of patients for immune checkpoint inhibitors. PD-L1 assessment may be challenging due to the necessity for different companion diagnostic assays for required specific immune checkpoint inhibitors and a relatively high level of inter-assay variability in terms of performance and cutoff levels. In this review, we discuss the role of PD-L1 immunohistochemistry as a predictive test in immunotherapy (immuno-oncology), highlight the complexity of the PD-L1 testing landscape, discuss various preanalytical, analytical and clinical issues that are associated with PD-L1 assays, and provide some insights into optimization of PD-L1 as a predictive biomarker in immuno-oncology.

Deep learning-based image analysis predicts PD-L1 status from H&E-stained histopathology images in breast cancer

Programmed death ligand-1 (PD-L1) has been recently adopted for breast cancer as a predictive biomarker for immunotherapies. The cost, time, and variability of PD-L1 quantification by immunohistochemistry (IHC) are a challenge. In contrast, hematoxylin and eosin (H&E) is a robust staining used routinely for cancer diagnosis. Here, we show that PD-L1 expression can be predicted from H&E-stained images by employing state-of-the-art deep learning techniques. With the help of two expert pathologists and a designed annotation software, we construct a dataset to assess the feasibility of PD-L1 prediction from H&E in breast cancer. In a cohort of 3,376 patients, our system predicts the PD-L1 status in a high area under the curve (AUC) of 0.91 - 0.93. Our system is validated on two external datasets, including an independent clinical trial cohort, showing consistent prediction performance. Furthermore, the proposed system predicts which cases are prone to pathologists miss-interpretation, showing it can serve as a decision support and quality assurance system in clinical practice.

Artificial Intelligence in Digital Pathology to Advance Cancer Immunotherapy

Immune-checkpoint inhibitors (ICIs) have revolutionized the treatment of many malignancies. For instance, in lung cancer, however, only 20~30% of patients can achieve durable clinical benefits from ICI monotherapy. Histopathologic and molecular features such as histological type, PD-L1 expression, and tumor mutation burden (TMB), play a paramount role in selecting appropriate regimens for cancer treatment in the era of immunotherapy. Unfortunately, none of the existing features are exclusive predictive biomarkers. Thus, there is an imperative need to pinpoint more effective biomarkers to identify patients who may achieve the most benefit from ICIs. The adoption of digital pathology in clinical flow, as being powered by artificial intelligence (AI) especially deep learning, has catalyzed the automated analysis of tissue slides. With the breakthrough of multiplex bioimaging technology, researchers can comprehensively characterize the tumor microenvironment, including the different immune cells' distribution, function, and interaction. Here, we briefly summarize recent AI studies in digital pathology and share our perspective on emerging paradigms and directions to advance the development of immunotherapy biomarkers.

Understanding resistance to immune checkpoint inhibitors in advanced breast cancer

INTRODUCTION

The addition of immune checkpoint inhibitors (ICIs) to frontline chemotherapy has improved survival for patients with advanced triple-negative breast cancer (TNBC) expressing programmed death-ligand 1 (PD-L1). Nonetheless, most patients develop resistance, with outcomes remaining poor for this population. Moreover, unsatisfactory activity has been observed with ICIs in PD-L1-negative TNBC and in other breast cancer (BC) subtypes, warranting a deeper understanding of resistance to ICIs in BC.

AREAS COVERED

We discuss the immune landscape of distinct BC subtypes, review the clinical activity of immunotherapy in BC, and highlight strategies under development to overcome resistance to ICIs.

EXPERT OPINION

Activity and resistance to ICIs in BC are strongly related to the intrinsic immunophenotype of the tumor tissue. Several promising biomarkers reflecting the immunological state of BC are emerging, with only PD-L1 expression currently adopted into clinical practice. However, limitations make of PD-L1 a sub-optimal biomarker for patient selection, which require efforts to integrate this marker with other immunological features. Concomitantly, a wide variety of drug combinations designed to overcome immune-resistance are being evaluated, with some encouraging signals observed in early-phase trials. Combination strategies tailored to patient and tumor immunophenotype may allow to overcome resistance and fully exploit the potential of ICIs.