Scoring of tumor-infiltrating lymphocytes: From visual estimation to machine learning

Deconstructing neutrophil to lymphocyte ratio (NLR) in early breast cancer: lack of prognostic utility and biological correlates across tumor subtypes

PURPOSE

The prognostic utility and biological correlates of neutrophil to lymphocyte ratio (NLR), a potential biomarker of the balance between immune response and the inflammatory status, are still uncertain in breast cancer (BC).

METHODS

We analysed a cohort of 959 women with early breast cancer, mostly treated with neoadjuvant or adjuvant chemotherapy. Clinical and pathological data, survival, NLR (continuous and categorical) and stromal tumor infiltrating lymphocytes (sTIL) were evaluated.

RESULTS

NLR was only weakly associated with Ki67, while no association was found for grade, histology, immunohistochemical subtype or stage. Lymphocyte infiltration of the tumor did not correlate with NLR (Rho: 0.05, p = 0.30). These results were similar in the whole group and across the different BC subtypes, with no differences in triple negative BC. Relapse free interval (RFI), breast cancer specific survival (BCSS) and overall survival (OS) changed according to pre-treatment NLR neither in the univariate nor in the multivariate Cox models (RFI: HR 0.948, p = 0.61; BCSS: HR 0.920, p = 0.57; OS: HR 0.96, p = 0.59).

CONCLUSION

These results question the utility of NLR as a prognostic biomarker in early breast cancer and suggest the lack of correlation of NLR with tumor microenvironment immune response.

Enhancing AI Research for Breast Cancer: A Comprehensive Review of Tumor-Infiltrating Lymphocyte Datasets

The field of immunology is fundamental to our understanding of the intricate dynamics of the tumor microenvironment. In particular, tumor-infiltrating lymphocyte (TIL) assessment emerges as essential aspect in breast cancer cases. To gain comprehensive insights, the quantification of TILs through computer-assisted pathology (CAP) tools has become a prominent approach, employing advanced artificial intelligence models based on deep learning techniques. The successful recognition of TILs requires the models to be trained, a process that demands access to annotated datasets. Unfortunately, this task is hampered not only by the scarcity of such datasets, but also by the time-consuming nature of the annotation phase required to create them. Our review endeavors to examine publicly accessible datasets pertaining to the TIL domain and thereby become a valuable resource for the TIL community. The overall aim of the present review is thus to make it easier to train and validate current and upcoming CAP tools for TIL assessment by inspecting and evaluating existing publicly available online datasets.

GLT8D2 is a prognostic biomarker and regulator of immune cell infiltration in gastric cancer

Because of the considerable tumor heterogeneity in gastric cancer (GC), only a limited group of patients experiences positive outcomes from immunotherapy. Herein, we aim to develop predictive models related to glycosylation genes to provide a more comprehensive understanding of immunotherapy for GC. RNA sequencing (RNA-seq) data and corresponding clinical outcomes were obtained from GEO and TCGA databases, and glycosylation-related genes were obtained from GlycoGene DataBase. We identified 48 differentially expressed glycosylation-related genes and established a prognostic model (seven prognosis genes including GLT8D2, GALNT6, ST3GAL6, GALNT15, GBGT1, FUT2, GXYLT2) based on these glycosylation-related genes using the results from Cox regression analysis. We found that these glycosylation-related genes revealed a robust correlation with the abundance of Tumor Infiltrating Lymphocytes (TILs), especially the GLT8D2 which is associated with many TILs. Finally, we employed immunohistochemistry and Multiplex Immunohistochemical to discover that GLT8D2 serves as a valuable prognostic biomarker in GC and is closely associated with macrophage-related markers. Collectively, we established a prognostic model based on glycosylation-related genes to provide a more comprehensive understanding of prediction for GC prognosis, and identified that GLT8D2 is closely correlated with adverse prognosis and may underscore its role in regulating immune cell infiltration in GC patients.

Proteomic characteristics of the treatment trajectory of patients with COVID-19

The ongoing COVID-19 pandemic caused by SARS-CoV-2 has prompted global concern due to its profound impact on public health and the economy. Effective treatment of COVID-19 patients in the acute phase or of those with long COVID is a major challenge. Using data-independent acquisition (DIA) technology, we performed proteomic profiling on plasma samples from 22 COVID-19 patients and six healthy controls at Dazhou Central Hospital. Random forest and least absolute shrinkage and selection operator algorithms were used for analysis at various COVID-19 treatment stages. We identified 79 proteins that were differentially expressed between COVID-19 patients and healthy controls, mainly involving pathways associated with cell processes and binding. Across different treatment stages of COVID-19, five proteins-PI16, GPLD1, IGFBP3, KRT19, and VCAM1-were identified as potential molecular markers for dynamic disease monitoring. Furthermore, the proteins BTD, APOM, IGKV2-28, VWF, C4BPA, and C7 were identified as candidate biomarkers for distinguishing between SARS-CoV-2 positivity and negativity. Analysis of protein change profiles between the follow-up and healthy control groups highlighted cardiovascular changes as a concern for patients recovering from COVID-19. Our study revealed the infection profiles of SARS-CoV-2 at the protein expression level comparing different phases of COVID-19. DIA mass spectrometry analysis of plasma samples from COVID-19 patients undergoing treatment identified key proteins involved in signaling pathways that might be used as markers of the recovery phase. These findings provide insight for the development of therapy options and suggest potential blood biomarkers for COVID-19.

Use of artificial intelligence for the prediction of lymph node metastases in early-stage colorectal cancer: systematic review

BACKGROUND

Risk evaluation of lymph node metastasis for early-stage (T1 and T2) colorectal cancers is critical for determining therapeutic strategies. Traditional methods of lymph node metastasis prediction have limited accuracy. This systematic review aimed to review the potential of artificial intelligence in predicting lymph node metastasis in early-stage colorectal cancers.

METHODS

A comprehensive search was performed of papers that evaluated the potential of artificial intelligence in predicting lymph node metastasis in early-stage colorectal cancers. Studies were appraised using the Joanna Briggs Institute tools. The primary outcome was summarizing artificial intelligence models and their accuracy. Secondary outcomes included influential variables and strategies to address challenges.

RESULTS

Of 3190 screened manuscripts, 11 were included, involving 8648 patients from 1996 to 2023. Due to diverse artificial intelligence models and varied metrics, no data synthesis was performed. Models included random forest algorithms, support vector machine, deep learning, artificial neural network, convolutional neural network and least absolute shrinkage and selection operator regression. Artificial intelligence models' area under the curve values ranged from 0.74 to 0.9993 (slide level) and 0.9476 to 0.9956 (single-node level), outperforming traditional clinical guidelines.

CONCLUSION

Artificial intelligence models show promise in predicting lymph node metastasis in early-stage colorectal cancers, potentially refining clinical decisions and improving outcomes.

PROSPERO REGISTRATION NUMBER

CRD42023409094.

Machine learning analysis of pathological images to predict 1-year progression-free survival of immunotherapy in patients with small-cell lung cancer

BACKGROUND

In small-cell lung cancer (SCLC), the tumor immune microenvironment (TIME) could be a promising biomarker for immunotherapy, but objectively evaluating TIME remains challenging. Hence, we aimed to develop a predictive biomarker of immunotherapy efficacy through a machine learning analysis of the TIME.

METHODS

We conducted a biomarker analysis in a prospective study of patients with extensive-stage SCLC who received chemoimmunotherapy as the first-line treatment. We trained a model to predict 1-year progression-free survival (PFS) using pathological images (H&E, programmed cell death-ligand 1 (PD-L1), and double immunohistochemical assay (cluster of differentiation 8 (CD8) and forkhead box P3 (FoxP3)) and patient information. The primary outcome was the mean area under the curve (AUC) of machine learning models in predicting the 1-year PFS.

RESULTS

We analyzed 100,544 patches of pathological images from 78 patients. The mean AUC values of patient information, pathological image, and combined models were 0.789 (range 0.571-0.982), 0.782 (range 0.750-0.911), and 0.868 (range 0.786-0.929), respectively. The PFS was longer in the high efficacy group than in the low efficacy group in all three models (patient information model, HR 0.468, 95% CI 0.287 to 0.762; pathological image model, HR 0.334, 95% CI 0.117 to 0.628; combined model, HR 0.353, 95% CI 0.195 to 0.637). The machine learning analysis of the TIME had better accuracy than the human count evaluations (AUC of human count, CD8-positive lymphocyte: 0.681, FoxP3-positive lymphocytes: 0.626, PD-L1 score: 0.567).

CONCLUSIONS

The spatial analysis of the TIME using machine learning predicted the immunotherapy efficacy in patients with SCLC, thus supporting its role as an immunotherapy biomarker.

Caveolin-1 promotes glioma proliferation and metastasis by enhancing EMT via mediating PAI-1 activation and its correlation with immune infiltrates

Glioma is typically characterized by a poor prognosis and is associated with a decline in the quality of life as the disease advances. However, the development of effective therapies for glioma has been inadequate. Caveolin-1 (CAV-1) is a membrane protein that plays a role in caveolae formation and interacts with numerous signaling proteins, compartmentalizing them in caveolae and frequently exerting direct control over their activity through binding to its scaffolding domain. Although CAV-1 is a vital regulator of tumour progression, its role in glioma remains unclear. Our findings indicated that the knockdown of CAV-1 significantly inhibits the proliferation and metastasis of glioma. Subsequent mechanistic investigations demonstrated that CAV-1 promotes proliferation and metastasis by activating the photoshatidylinositol 3-kinase/protein kinase B (PI3K/Akt) signaling pathway. Furthermore, we demonstrated that CAV-1 overexpression upregulates the expression of serpin peptidase inhibitor, class E, member 1 (SERPINE1, also known as PAI-1), which serves as a marker for the epithelial-mesenchymal transition (EMT) process. Further research showed that PAI-1 knockdown abolished the CAV-1 mediated activation of PI3K/Akt signaling pathway. In glioma tissues, CAV-1 expression exhibited a correlation with unfavorable prognosis and immune infiltration among glioma patients. In summary, our study provided evidence that CAV-1 activates the PI3K/Akt signaling pathway by upregulating PAI-1, thereby promoting the proliferation and metastasis of glioma through enhanced epithelial-mesenchymal transition (EMT) and angiogenesis, and CAV-1 is involved in the immune infiltration.

Deep Learning Approaches for Automatic Quality Assurance of Magnetic Resonance Images Using ACR Phantom

BACKGROUND

In recent years, there has been a growing trend towards utilizing Artificial Intelligence (AI) and machine learning techniques in medical imaging, including for the purpose of automating quality assurance. In this research, we aimed to develop and evaluate various deep learning-based approaches for automatic quality assurance of Magnetic Resonance (MR) images using the American College of Radiology (ACR) standards.

METHODS

The study involved the development, optimization, and testing of custom convolutional neural network (CNN) models. Additionally, popular pre-trained models such as VGG16, VGG19, ResNet50, InceptionV3, EfficientNetB0, and EfficientNetB5 were trained and tested. The use of pre-trained models, particularly those trained on the ImageNet dataset, for transfer learning was also explored. Two-class classification models were employed for assessing spatial resolution and geometric distortion, while an approach classifying the image into 10 classes representing the number of visible spokes was used for the low contrast.

RESULTS

Our results showed that deep learning-based methods can be effectively used for MR image quality assurance and can improve the performance of these models. The low contrast test was one of the most challenging tests within the ACR phantom.

CONCLUSIONS

Overall, for geometric distortion and spatial resolution, all of the deep learning models tested produced prediction accuracy of 80% or higher. The study also revealed that training the models from scratch performed slightly better compared to transfer learning. For the low contrast, our investigation emphasized the adaptability and potential of deep learning models. The custom CNN models excelled in predicting the number of visible spokes, achieving commendable accuracy, recall, precision, and F1 scores.

Multi-omics Analysis of Prognostic Significance and Immune Infiltration of FASTK Family Members in Kidney Renal Clear Cell Carcinoma

Objective

The Fas-activated serine/threonine kinase (FASTK) family of proteins has been recently found to be able to regulate mitochondrial gene expression post-transcriptionally. Nonetheless, there is a paucity of study about the role of the FASTK family in kidney renal clear cell carcinoma (KIRC). This study was conducted to explore the correlation of FASTK family genes with expression, prognosis, and immune infiltration in KIRC.

Methods

We collected the data from the UALCAN, GeneMANIA, STRING, CancerSEA, cBioPortal, Kaplan-Meier plotter, GEPIA, TISIDB and TIMER databases to evaluate the genetic alterations, differential expression, prognostic significance, and immune cell infiltration of FASTKs in patients with KIRC.

Results

In tumor tissues of KIRC, the mRNA expression level of FASTK and TBRG4 was elevated, whereas that of FASTKD1, FASTKD2, and FASTKD5 was lowered compared with normal tissues (P < .05). Patients with KIRC and high FASTK and Transforming growth factor β regulator 4 (TBRG4) expression had worse overall survival (OS) and disease specific survival (DFS), while those with lower expression of FASTKD2/3/5 had worse outcomes. FASTK was positively correlated with DNA damage. FASTKD1 was positively related to differentiation. FASTKD2 was inversely related to proliferation and FASTKD5 was inversely related to invasion and EMT in KIRC cells. FASTK expression in KIRC was inversely linked to the presence of several immune cells including Tgd, macrophages, Tcm, and Mast cells (P < .05).

Conclusions

Our research provided fresh insight and in-depth analysis to the selection of prognostic biological markers of FASTK family members in KIRC.

Artificial intelligence-powered spatial analysis of tumor-infiltrating lymphocytes for prediction of prognosis in resected colon cancer

Tumor-infiltrating lymphocytes (TIL) have been suggested as an important prognostic marker in colorectal cancer, but assessment usually requires additional tissue processing and interpretational efforts. The aim of this study is to assess the clinical significance of artificial intelligence (AI)-powered spatial TIL analysis using only a hematoxylin and eosin (H&E)-stained whole-slide image (WSI) for the prediction of prognosis in stage II-III colon cancer treated with surgery and adjuvant therapy. In this retrospective study, we used Lunit SCOPE IO, an AI-powered H&E WSI analyzer, to assess intratumoral TIL (iTIL) and tumor-related stromal TIL (sTIL) densities from WSIs of 289 patients. The patients with confirmed recurrences had significantly lower sTIL densities (mean sTIL density 630.2/mm2 in cases with confirmed recurrence vs. 1021.3/mm2 in no recurrence, p < 0.001). Additionally, significantly higher recurrence rates were observed in patients having sTIL or iTIL in the lower quartile groups. Risk groups defined as high-risk (both iTIL and sTIL in the lowest quartile groups), low-risk (sTIL higher than the median), or intermediate-risk (not high- or low-risk) were predictive of recurrence and were independently associated with clinical outcomes after adjusting for other clinical factors. AI-powered TIL analysis can provide prognostic information in stage II/III colon cancer in a practical manner.

The Value of Tumor Infiltrating Lymphocytes (TIL) for Predicting the Response to Neoadjuvant Chemotherapy (NAC) in Breast Cancer according to the Molecular Subtypes

INTRODUCTION

The antitumor host immune response is an important factor in breast cancer, but its role is not fully established. The role of tumor infiltrating lymphocytes (TIL) as an immunological biomarker in breast cancer has been significantly explored in recent years. The number of patients treated with neoadjuvant chemotherapy (NAC) has increased and the identification of a biomarker to predict the probability of pCR (pathological complete response) is a high priority.

MATERIALS AND METHODS

We evaluated 334 cases of BC treated with NAC followed by surgical resection from 2020-2022 at the Ist Clinic of Oncological Surgery, Oncological Institute "Prof Dr I Chiricuta" Cluj Napoca. Of the above, 122 cases were available for histological evaluation both in pre-NAC biopsy and post-NAC resection tissue. Evaluation of biopsy fragments and resection parts were performed using hematoxylin eosin (H&E). The TIL evaluation took place according to the recommendations of the International TIL Working Group (ITILWG).

RESULTS

There was a strong association between elevated levels of pre-NAC TIL. At the same time, there is a statistically significant correlation between stromal TIL and tumor grade, the number of lymph node metastases, the molecular subtype and the number of mitoses (p < 0.005). Intratumoral TIL showed a significant correlation with tumor size, distant metastasis, molecular subtype, number of mitosis, stage and lymph node metastasis (p < 0.005). We also demonstrated that high pre-NAC STIL represents a strong predictive marker for pCR.

CONCLUSION

This study reveals the role of TIL as a predictive biomarker in breast cancer not only for the well-established TNBC (triple negative breast cancer) and HER2+ (Her2 overexpressed) subtypes but also in Luminal A and B molecular subtypes. In this scenario, the evaluation of sTIL as a novel predictive and therapy-predicting factor should become a routinely performed analysis that could guide clinicians when choosing the most appropriate therapy.

Evaluation of tumour infiltrating lymphocytes in luminal breast cancer using artificial intelligence

BACKGROUND

Tumour infiltrating lymphocytes (TILs) are a prognostic parameter in triple-negative and human epidermal growth factor receptor 2 (HER2)-positive breast cancer (BC). However, their role in luminal (oestrogen receptor positive and HER2 negative (ER + /HER2-)) BC remains unclear. In this study, we used artificial intelligence (AI) to assess the prognostic significance of TILs in a large well-characterised cohort of luminal BC.

METHODS

Supervised deep learning model analysis of Haematoxylin and Eosin (H&E)-stained whole slide images (WSI) was applied to a cohort of 2231 luminal early-stage BC patients with long-term follow-up. Stromal TILs (sTILs) and intratumoural TILs (tTILs) were quantified and their spatial distribution within tumour tissue, as well as the proportion of stroma involved by sTILs were assessed. The association of TILs with clinicopathological parameters and patient outcome was determined.

RESULTS

A strong positive linear correlation was observed between sTILs and tTILs. High sTILs and tTILs counts, as well as their proximity to stromal and tumour cells (co-occurrence) were associated with poor clinical outcomes and unfavourable clinicopathological parameters including high tumour grade, lymph node metastasis, large tumour size, and young age. AI-based assessment of the proportion of stroma composed of sTILs (as assessed visually in routine practice) was not predictive of patient outcome. tTILs was an independent predictor of worse patient outcome in multivariate Cox Regression analysis.

CONCLUSION

AI-based detection of TILs counts, and their spatial distribution provides prognostic value in luminal early-stage BC patients. The utilisation of AI algorithms could provide a comprehensive assessment of TILs as a morphological variable in WSIs beyond eyeballing assessment.

Upregulated CANT1 is correlated with poor prognosis in hepatocellular carcinoma

BACKGROUND

CANT1, as calcium-activated protein nucleotidase 1, is a kind of phosphatase. It is overexpressed in some tumors and related to poor prognosis, but few studies explore its function and carcinogenic mechanism in hepatocellular carcinoma (HCC).

METHODS

The expression of CANT1 mRNA and protein was analyzed by the Cancer Genome Atlas (TCGA) database and immunohistochemistry(IHC) staining. The relationship between CANT1 expression and clinicopathology was evaluated by various public databases. The receiver operating characteristic (ROC) curve was used to assess the diagnostic accuracy of CANT1 by the area under curve (AUC). Univariate, multivariate Cox regression and Kaplan-Meier curves were applied to evaluate the predictive value of CANT1 on the prognosis of HCC. Methsurv was used to analyze gene changes and DNA methylation, and its impact on prognosis. The enrichment analysis of DEGs associated with CANT1 revealed the biological process of CANT1 based on Gene Set Enrichment Analysis (GSEA). The relationship between immune cell infiltration level and CANT1 expression in HCC was investigated using the single-sample GSEA (ssGSEA) method and the Tumor Immune Estimation Resource (TIMER) database. Finally, the association between CANT1 and immune checkpoints and drug sensitivity was also analyzed.

RESULTS

CANT1 was highly expressed in 22 cancers, including HCC, and CANT1 overexpression in HCC was confirmed by IHC. The expression of CANT1 was correlated with clinical features, such as histologic grade. Highly expressed CANT1 caused poor overall survival (OS) of HCC patients. Univariate and multivariate regression analysis suggested that CANT1 was an independent prognostic marker. Of the 31 DNA methylation at CpG sites, three CpG sites were associated with the prognosis of HCC. GSEA indicated that CANT1 was mainly involved in the cell cycle, DNA replication, and etc. Moreover, CANT1 expression was correlated with immune cell infiltration and independently associated with the prognosis of HCC patients. Finally, CANT1 expression was correlated with most immune checkpoints and drug sensitivity.

CONCLUSION

CANT1 may be a latent oncogene of HCC, and associated with immune cells and immune checkpoints, which may assist in HCC treatment.

The radiomic biomarker in non-small cell lung cancer: 18F-FDG PET/CT characterisation of programmed death-ligand 1 status

AIM

To present an integrated 2-[18F]-fluoro-2-deoxy-d-glucose (18F-FDG) positron-emission tomography (PET)/computed tomography (CT) radiomic characterisation of programmed death-ligand 1 (PD-L1) status in non-small-cell lung cancer (NSCLC).

MATERIALS AND METHODS

In this retrospective study, 18F-FDG PET/CT images and clinical data of 394 eligible patients were divided into training (n=275) and test sets (n=119). Next, the corresponding nodule of interest was segmented manually on the axial CT images by radiologists. After which, the spatial position matching method was used to match the image positions of CT and PET, and radiomic features of the CT and PET images were extracted. Radiomic models were built using five different machine-learning classifiers and the performance of the radiomic models were further evaluated. Finally, a radiomic signature was established to predict the PD-L1 status in patients with NSCLC using the features in the best performing radiomic model.

RESULTS

The radiomic model based on the PET intranodular region determined using the logistic regression classifier preformed best, yielding an area under the receiver operating characteristics curve (AUC) of 0.813 (95% CI: 0.812, 0.821) on the test set. The clinical features did not improve the test set AUC (0.806, 95% CI: 0.801, 0.810). The final radiomic signature for PD-L1 status was consisted of three PET radiomic features.

CONCLUSION

This study showed that an 18F-FDG PET/CT-based radiomic signature could be used as a non-invasive biomarker to discriminate PD-L1-positive from PD-L1-negative in patients with NSCLC.

Revolutionizing Cancer Research: The Impact of Artificial Intelligence in Digital Biobanking

BACKGROUND

Biobanks are vital research infrastructures aiming to collect, process, store, and distribute biological specimens along with associated data in an organized and governed manner. Exploiting diverse datasets produced by the biobanks and the downstream research from various sources and integrating bioinformatics and "omics" data has proven instrumental in advancing research such as cancer research. Biobanks offer different types of biological samples matched with rich datasets comprising clinicopathologic information. As digital pathology and artificial intelligence (AI) have entered the precision medicine arena, biobanks are progressively transitioning from mere biorepositories to integrated computational databanks. Consequently, the application of AI and machine learning on these biobank datasets holds huge potential to profoundly impact cancer research.

METHODS

In this paper, we explore how AI and machine learning can respond to the digital evolution of biobanks with flexibility, solutions, and effective services. We look at the different data that ranges from specimen-related data, including digital images, patient health records and downstream genetic/genomic data and resulting "Big Data" and the analytic approaches used for analysis.

RESULTS

These cutting-edge technologies can address the challenges faced by translational and clinical research, enhancing their capabilities in data management, analysis, and interpretation. By leveraging AI, biobanks can unlock valuable insights from their vast repositories, enabling the identification of novel biomarkers, prediction of treatment responses, and ultimately facilitating the development of personalized cancer therapies.

CONCLUSIONS

The integration of biobanking with AI has the potential not only to expand the current understanding of cancer biology but also to pave the way for more precise, patient-centric healthcare strategies.

TRAIP serves as a potential prognostic biomarker and correlates with immune infiltrates in lung adenocarcinoma

BACKGROUND

Lung adenocarcinoma (LUAD) is one of the major types of lung cancer with high morbidity and mortality. The TRAF-interacting protein (TRAIP) is a ring-type E3 ubiquitin ligase which has been recently identified to play pivotal roles in various cancers. However, the expression and function of TRAIP in LUAD remain elusive.

METHODS

In this study, we used bioinformatic tools as well as molecular experiments to explore the exact role of TRAIP and the underlying mechanism.

RESULTS

Data mining across the UALCAN, GEPIA and GTEx, GEO and HPA databases revealed that TRAIP was significantly overexpressed in LUAD tissues than that in adjacent normal tissues. Kaplan-Meier curve showed that high TRAIP expression was associated with poor overall survival (OS) and relapse-free survival (RFS). Univariate and multivariate cox regression analysis revealed that TRAIP was an independent risk factor in LUAD. And the TRAIP-based nomogram further supported the prognostic role of TRAIP in LUAD. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis demonstrated that TRAIP-associated genes were mainly involved in DNA replication, cell cycle and other processes. The immune infiltration analysis indicated that TRAIP expression was tightly correlated with the infiltration of diverse immune cell types, including B cell, CD8 + T cell, neutrophil and dendritic cell. Moreover, TRAIP expression was observed to be significantly associated with tumor infiltrating lymphocytes (TILs) and immune checkpoint molecules. In vitro experiments further confirmed knockdown of TRAIP inhibited cell migration and invasion, as well as decreasing chemokine production and inhibiting M2-like macrophage recruitment. Lastly, CMap analysis identified 10 small molecule compounds that may target TRAIP, providing potential therapies for LUAD.

CONCLUSIONS

Collectively, our study found that TRAIP is an oncogenic gene in LUAD, which may be a potential prognostic biomarker and promising therapeutic target for LUAD.

Grading of lung adenocarcinomas with simultaneous segmentation by artificial intelligence (GLASS-AI)

Preclinical genetically engineered mouse models (GEMMs) of lung adenocarcinoma are invaluable for investigating molecular drivers of tumor formation, progression, and therapeutic resistance. However, histological analysis of these GEMMs requires significant time and training to ensure accuracy and consistency. To achieve a more objective and standardized analysis, we used machine learning to create GLASS-AI, a histological image analysis tool that the broader cancer research community can utilize to grade, segment, and analyze tumors in preclinical models of lung adenocarcinoma. GLASS-AI demonstrates strong agreement with expert human raters while uncovering a significant degree of unreported intratumor heterogeneity. Integrating immunohistochemical staining with high-resolution grade analysis by GLASS-AI identified dysregulation of Mapk/Erk signaling in high-grade lung adenocarcinomas and locally advanced tumor regions. Our work demonstrates the benefit of employing GLASS-AI in preclinical lung adenocarcinoma models and the power of integrating machine learning and molecular biology techniques for studying the molecular pathways that underlie cancer progression.

Clinical Meaning of Stromal Tumor Infiltrating Lymphocytes (sTIL) in Early Luminal B Breast Cancer

Luminal breast cancer (BC) is associated with less immune activation, and the significance of stromal lymphocytic infiltration (sTIL) is more uncertain than in other BC subtypes. The aim of this study was to investigate the predictive and prognostic value of sTIL in early luminal BC. The study was performed with an observational design in a prospective cohort of 345 patients with predominantly high-risk luminal (hormone receptor positive, HER2 negative) BC and with luminal B features (n = 286), in which the presence of sTIL was analyzed with validated methods. Median sTIL infiltration was 5% (Q1-Q3 range (IQR), 0-10). We found that sTIL were associated with characteristics of higher biological and clinical aggressiveness (tumor and lymph node proliferation and stage, among others) and that the percentage of sTIL was predictive of pathologic complete response in patients treated with neoadjuvant chemotherapy (OR: 1.05, 95%CI 1.02-1.09, p < 0.001). The inclusion of sTIL (any level of lymphocytic infiltration: sTIL > 0%) in Cox regression multivariable prognostic models was associated with a shorter relapse-free interval (HR: 4.85, 95%CI 1.33-17.65, p = 0.016) and significantly improved its performance. The prognostic impact of sTIL was independent of other clinical and pathological variables and was mainly driven by its relevance in luminal B BC.

Image analysis-based tumor infiltrating lymphocytes measurement predicts breast cancer pathologic complete response in SWOG S0800 neoadjuvant chemotherapy trial

We assessed the predictive value of an image analysis-based tumor-infiltrating lymphocytes (TILs) score for pathologic complete response (pCR) and event-free survival in breast cancer (BC). About 113 pretreatment samples were analyzed from patients with stage IIB-IIIC HER-2-negative BC randomized to neoadjuvant chemotherapy ± bevacizumab. TILs quantification was performed on full sections using QuPath open-source software with a convolutional neural network cell classifier (CNN11). We used easTILs% as a digital metric of TILs score defined as [sum of lymphocytes area (mm2)/stromal area(mm2)] × 100. Pathologist-read stromal TILs score (sTILs%) was determined following published guidelines. Mean pretreatment easTILs% was significantly higher in cases with pCR compared to residual disease (median 36.1 vs.14.8%, p < 0.001). We observed a strong positive correlation (r = 0.606, p < 0.0001) between easTILs% and sTILs%. The area under the prediction curve (AUC) was higher for easTILs% than sTILs%, 0.709 and 0.627, respectively. Image analysis-based TILs quantification is predictive of pCR in BC and had better response discrimination than pathologist-read sTILs%.

Bridging clinic and wildlife care with AI-powered pan-species computational pathology

Cancers occur across species. Understanding what is consistent and varies across species can provide new insights into cancer initiation and evolution, with significant implications for animal welfare and wildlife conservation. We build a pan-species cancer digital pathology atlas (panspecies.ai) and conduct a pan-species study of computational comparative pathology using a supervised convolutional neural network algorithm trained on human samples. The artificial intelligence algorithm achieves high accuracy in measuring immune response through single-cell classification for two transmissible cancers (canine transmissible venereal tumour, 0.94; Tasmanian devil facial tumour disease, 0.88). In 18 other vertebrate species (mammalia = 11, reptilia = 4, aves = 2, and amphibia = 1), accuracy (range 0.57-0.94) is influenced by cell morphological similarity preserved across different taxonomic groups, tumour sites, and variations in the immune compartment. Furthermore, a spatial immune score based on artificial intelligence and spatial statistics is associated with prognosis in canine melanoma and prostate tumours. A metric, named morphospace overlap, is developed to guide veterinary pathologists towards rational deployment of this technology on new samples. This study provides the foundation and guidelines for transferring artificial intelligence technologies to veterinary pathology based on understanding of morphological conservation, which could vastly accelerate developments in veterinary medicine and comparative oncology.

A pathomic approach for tumor-infiltrating lymphocytes classification on breast cancer digital pathology images

Background and objectives

The detection of tumor-infiltrating lymphocytes (TILs) could aid in the development of objective measures of the infiltration grade and can support decision-making in breast cancer (BC). However, manual quantification of TILs in BC histopathological whole slide images (WSI) is currently based on a visual assessment, thus resulting not standardized, not reproducible, and time-consuming for pathologists. In this work, a novel pathomic approach, aimed to apply high-throughput image feature extraction techniques to analyze the microscopic patterns in WSI, is proposed. In fact, pathomic features provide additional information concerning the underlying biological processes compared to the WSI visual interpretation, thus providing more easily interpretable and explainable results than the most frequently investigated Deep Learning based methods in the literature.

Methods

A dataset containing 1037 regions of interest with tissue compartments and TILs annotated on 195 TNBC and HER2+ BC hematoxylin and eosin (H&E)-stained WSI was used. After segmenting nuclei within tumor-associated stroma using a watershed-based approach, 71 pathomic features were extracted from each nucleus and reduced using a Spearman's correlation filter followed by a nonparametric Wilcoxon rank-sum test and least absolute shrinkage and selection operator. The relevant features were used to classify each candidate nucleus as either TILs or non-TILs using 5 multivariable machine learning classification models trained using 5-fold cross-validation (1) without resampling, (2) with the synthetic minority over-sampling technique and (3) with downsampling. The prediction performance of the models was assessed using ROC curves.

Results

21 features were selected, with most of them related to the well-known TILs properties of having regular shape, clearer margins, high peak intensity, more homogeneous enhancement and different textural pattern than other cells. The best performance was obtained by Random-Forest with ROC AUC of 0.86, regardless of resampling technique.

Conclusions

The presented approach holds promise for the classification of TILs in BC H&E-stained WSI and could provide support to pathologists for a reliable, rapid and interpretable clinical assessment of TILs in BC.

Predictive Biomarkers for Response to Immunotherapy in Triple Negative Breast Cancer: Promises and Challenges

Triple negative breast cancer (TNBC) is a highly heterogeneous disease with a poor prognosis and a paucity of therapeutic options. In recent years, immunotherapy has emerged as a new treatment option for patients with TNBC. However, this therapeutic evolution is paralleled by a growing need for biomarkers which allow for a better selection of patients who are most likely to benefit from this immune checkpoint inhibitor (ICI)-based regimen. These biomarkers will not only facilitate a better optimization of treatment strategies, but they will also avoid unnecessary side effects in non-responders, and limit the increasing financial toxicity linked to the use of these agents. Huge efforts have been deployed to identify predictive biomarkers for the ICI, but until now, the fruits of this labor remained largely unsatisfactory. Among clinically validated biomarkers, only programmed death-ligand 1 protein (PD-L1) expression has been prospectively assessed in TNBC trials. In addition to this, microsatellite instability and a high tumor mutational burden are approved as tumor agnostic biomarkers, but only a small percentage of TNBC fits this category. Furthermore, TNBC should no longer be approached as a single biological entity, but rather as a complex disease with different molecular, clinicopathological, and tumor microenvironment subgroups. This review provides an overview of the validated and evolving predictive biomarkers for a response to ICI in TNBC.

Wiskott-Aldrich syndrome gene as a prognostic biomarker correlated with immune infiltrates in clear cell renal cell carcinoma

Introduction

The abnormal expression of the Wiskott-Aldrich syndrome protein (WASP) encoded by the Wiskott-Aldrich syndrome (WAS) gene has been implicated in tumor invasion and immune regulation. However, prognostic implications of WAS and its correlation tumor infiltrating in renal clear cell carcinoma (ccRCC) is not clear cut.

Methods

The correlation between WAS expression, clinicopathological variables and clinical outcomes were evaluated using The Cancer Genome Atlas (TCGA), Gene Expression Omnibus (GEO), Tumor Immune Estimation Resource (TIMER), UALCAN, Gene Expression Profiling Interaction Analysis (GEPIA), Kaplan-Meier (KM) plotter and other databases. Furthermore, we assessed the transcription expression of WAS in renal cancer tissues, various renal carcinoma cell lines and human renal tubular cells (HK2) using quantitative polymerase chain reaction (qPCR). A comprehensive analysis of multiple databases including TIMER, GEPIA, TISIDB, ESTIMATE algorithm, and CIBERSORT algorithm were performed to determine the correlation between WAS and tumor infiltrating immune cells in ccRCC.

Results

The results displayed an increase in WAS mRNA level in ccRCC compared to normal tissue. WAS protein level was found highly expressed in cancer tissues, particularly within renal tumor cells via the human protein atlas (HPA). Interestingly, we found that elevated WAS expression was significantly positively correlated with the infiltration of CD8+ T cells, B cells, Monocytes, Neutrophils, Macrophages, T cell regulation, NK cells, and Dendritic cells in ccRCC. Bioinformatics demonstrated a strong correlation between WAS expression and 42 immune checkpoints, including the T cell exhaustion gene PD-1, which is critical for exploring immunotherapy for ccRCC. We revealed that patients with high WAS expression were less sensitive to immunotherapy medications.

Conclusion

In conclusion, our study identified that WAS was a prognostic biomarker and correlated with immune infiltrates in ccRCC.

Computer-assisted tumor grading, validation of PD-L1 scoring, and quantification of CD8-positive immune cell density in urothelial carcinoma, a visual guide for pathologists using QuPath

Background

Advances in digital imaging in pathology and the new capacity to scan high-quality images have change the way to practice and research in surgical pathology. QuPath is an open-source pathology software that offers a reproducible way to analyze quantified variables. We aimed to present the functionality of biomarker scoring using QuPath and provide a guide for the validation of pathologic grading using a series of cases of urothelial carcinomas.

Methods

Tissue microarrays of urothelial carcinomas were constructed and scanned. The images stained with HE, CD8 and PD-L1 immunohistochemistry were imported into QuPath and dearrayed. Training images were used to build a grade classifier and applied to all cases. Quantification of CD8 and PD-L1 was undertaken for each core using cytoplasmic and membrane color segmentation and output measurement and compared with pathologists semi-quantitative assessments.

Results

There was a good correlation between tumor grade by the pathologist and by QuPath software (Kappa agreement 0.73). For low-grade carcinomas (by the report and pathologist), the concordance was not as high. Of the 32 low-grade tumors, 22 were correctly classified as low-grade, but 11 (34%) were diagnosed as high-grade, with the high-grade to the low-grade ratio in these misclassified cases ranging from 0.41 to 0.58. The median ratio for bona fide high-grade carcinomas was 0.59. Some of the reasons the authors list as potential mimickers for high-grade cases are fulguration artifact, nuclear hyperchromasia, folded tissues, and inconsistency in staining. The correlation analysis between the software and the pathologist showed that the CD8 marker showed a moderate (r = 0.595) and statistically significant (p < 0.001) correlation. The internal consistency of this parameter showed an index of 0.470. The correlation analysis between the software and the pathologist showed that the PDL1 marker showed a robust (r = 0.834) and significant (p < 0.001) correlation. The internal consistency of this parameter showed a CCI of 0.851.

Conclusions

We were able to demonstrate the utility of QuPath in identifying and scoring tumor cells and IHC quantification of two biomarkers. The protocol we present uses a free open-source platform to help researchers deal with imaging and data processing in the surgical pathology field.

Low Level FLT3LG is a Novel Poor Prognostic Biomarker for Cervical Cancer with Immune Infiltration

Introduction

The FMS-related tyrosine kinase 3 (FLT3) ligand (FLT3LG), a growth factor, binds to FLT3 on dendritic cell (DCs) to enhance their differentiation and expansion. It has shown great potential as an immunotherapy target for cancers. However, the expression and function of FLT3LG in cervical cancer remain largely unknown.

Materials and Methods

In this study, we obtained the expression of FLT3LG, the clinical prognosis in cervical cancer, via multiple databases, including The Cancer Genome Atlas (TCGA), the TISIDB database, and Tumor Immune Estimate Resource (TIMER). The results were further investigated using real-time quantitative PCR (qPCR) cytology specimens in 489 patients. Furthermore, Kaplan-Meier Cox regression and prognostic nomogram analyses were used to assess FLT3LG's clinical significance in cervical cancer patients. All calculations used the R package.

Results

As a result, FLT3LG expression decreased in cervical cancer compared with standard samples. And the low expression of FLT3LG was associated with a poor prognosis. Furthermore, Receiver Operating Characteristics (ROC) analysis indicated that FLT3LG might serve as a valuable diagnostic biomarker for cervical cancer. Additionally, it indicated that the FLT3LG had the highest odds ratio (OR=10.519; (7.371-27.071)) for detecting CIN 2+. In addition, our result also demonstrated that expression of FLT3LG was closely related to immune cells, immune inhibitors, immunostimulators, receptors, and chemokines in CESC.

Conclusion

Research on FLT3LG provided insight into its critical function. Hence, the low expression of FLT3LG may be a valuable biomarker in CESC patients linked with immune infiltration.

Integrated imaging and molecular analysis to decipher tumor microenvironment in the era of immunotherapy

Radiological imaging is an integral component of cancer care, including diagnosis, staging, and treatment response monitoring. It contains rich information about tumor phenotypes that are governed not only by cancer cellintrinsic biological processes but also by the tumor microenvironment, such as the composition and function of tumor-infiltrating immune cells. By analyzing the radiological scans using a quantitative radiomics approach, robust relations between specific imaging and molecular phenotypes can be established. Indeed, a number of studies have demonstrated the feasibility of radiogenomics for predicting intrinsic molecular subtypes and gene expression signatures in breast cancer based on MRI. In parallel, promising results have been shown for inferring the amount of tumor-infiltrating lymphocytes, a key factor for the efficacy of cancer immunotherapy, from standard-of-care radiological images. Compared with the biopsy-based approach, radiogenomics offers a unique avenue to profile the molecular makeup of the tumor and immune microenvironment as well as its evolution in a noninvasive and holistic manner through longitudinal imaging scans. Here, we provide a systematic review of the state of the art radiogenomics studies in the era of immunotherapy and discuss emerging paradigms and opportunities in AI and deep learning approaches. These technical advances are expected to transform the radiogenomics field, leading to the discovery of reliable imaging biomarkers. This will pave the way for their clinical translation to guide precision cancer therapy.

Quantifying Spatial Heterogeneity of Tumor-Infiltrating Lymphocytes to Predict Survival of Individual Cancer Patients

Tumor-infiltrating lymphocytes (TILs), identified on HE-stained histopathological images in the cancer area, are indicators of the adaptive immune response against cancers and play a major role in personalized cancer immunotherapy. Recent works indicate that the spatial organization of TILs may be prognostic of disease-specific survival and recurrence. However, there are a limited number of methods that were proposed and tested in analyses of the spatial structure of TILs. In this work, we evaluated 14 different spatial measures, including the one developed for other omics data, on 10,532 TIL maps from 23 cancer types in terms of reproducibility, uniqueness, and impact on patient survival. For each spatial measure, 16 different scenarios for the definition of prognostic factor were tested. We found no difference in survival prediction when TIL maps were stored as binary images or continuous TIL probability scores. When spatial measures were discretized into a low and high category, a higher correlation with survival was observed. Three measures with the highest cancer prognosis capability were spatial autocorrelation, GLCM M1, and closeness centrality. Most of the tested measures could be further tuned to increase prediction performance.

Deep Learning-Based Pathology Image Analysis Enhances Magee Feature Correlation With Oncotype DX Breast Recurrence Score

Background

Oncotype DX Recurrence Score (RS) has been widely used to predict chemotherapy benefits in patients with estrogen receptor-positive breast cancer. Studies showed that the features used in Magee equations correlate with RS. We aimed to examine whether deep learning (DL)-based histology image analyses can enhance such correlations.

Methods

We retrieved 382 cases with RS diagnosed between 2011 and 2015 from the Emory University and the Ohio State University. All patients received surgery. DL models were developed to detect nuclei of tumor cells and tumor-infiltrating lymphocytes (TILs) and segment tumor cell nuclei in hematoxylin and eosin (H&E) stained histopathology whole slide images (WSIs). Based on the DL-based analysis, we derived image features from WSIs, such as tumor cell number, TIL number variance, and nuclear grades. The entire patient cohorts were divided into one training set (125 cases) and two validation sets (82 and 175 cases) based on the data sources and WSI resolutions. The training set was used to train the linear regression models to predict RS. For prediction performance comparison, we used independent variables from Magee features alone or the combination of WSI-derived image and Magee features.

Results

The Pearson's correlation coefficients between the actual RS and predicted RS by DL-based analysis were 0.7058 (p-value = 1.32 × 10-13) and 0.5041 (p-value = 1.15 × 10-12) for the validation sets 1 and 2, respectively. The adjusted R 2 values using Magee features alone are 0.3442 and 0.2167 in the two validation sets, respectively. In contrast, the adjusted R 2 values were enhanced to 0.4431 and 0.2182 when WSI-derived imaging features were jointly used with Magee features.

Conclusion

Our results suggest that DL-based digital pathological features can enhance Magee feature correlation with RS.

Patient-level proteomic network prediction by explainable artificial intelligence

Understanding the pathological properties of dysregulated protein networks in individual patients’ tumors is the basis for precision therapy. Functional experiments are commonly used, but cover only parts of the oncogenic signaling networks, whereas methods that reconstruct networks from omics data usually only predict average network features across tumors. Here, we show that the explainable AI method layer-wise relevance propagation (LRP) can infer protein interaction networks for individual patients from proteomic profiling data. LRP reconstructs average and individual interaction networks with an AUC of 0.99 and 0.93, respectively, and outperforms state-of-the-art network prediction methods for individual tumors. Using data from The Cancer Proteome Atlas, we identify known and potentially novel oncogenic network features, among which some are cancer-type specific and show only minor variation among patients, while others are present across certain tumor types but differ among individual patients. Our approach may therefore support predictive diagnostics in precision oncology by inferring “patient-level” oncogenic mechanisms.

Building Efficient CNN Architectures for Histopathology Images Analysis: A Case-Study in Tumor-Infiltrating Lymphocytes Classification

Background

Deep learning methods have demonstrated remarkable performance in pathology image analysis, but they are computationally very demanding. The aim of our study is to reduce their computational cost to enable their use with large tissue image datasets.

Methods

We propose a method called Network Auto-Reduction (NAR) that simplifies a Convolutional Neural Network (CNN) by reducing the network to minimize the computational cost of doing a prediction. NAR performs a compound scaling in which the width, depth, and resolution dimensions of the network are reduced together to maintain a balance among them in the resulting simplified network. We compare our method with a state-of-the-art solution called ResRep. The evaluation is carried out with popular CNN architectures and a real-world application that identifies distributions of tumor-infiltrating lymphocytes in tissue images.

Results

The experimental results show that both ResRep and NAR are able to generate simplified, more efficient versions of ResNet50 V2. The simplified versions by ResRep and NAR require 1.32× and 3.26× fewer floating-point operations (FLOPs), respectively, than the original network without a loss in classification power as measured by the Area under the Curve (AUC) metric. When applied to a deeper and more computationally expensive network, Inception V4, NAR is able to generate a version that requires 4× lower than the original version with the same AUC performance.

Conclusions

NAR is able to achieve substantial reductions in the execution cost of two popular CNN architectures, while resulting in small or no loss in model accuracy. Such cost savings can significantly improve the use of deep learning methods in digital pathology. They can enable studies with larger tissue image datasets and facilitate the use of less expensive and more accessible graphics processing units (GPUs), thus reducing the computing costs of a study.

Potential Predictive and Prognostic Value of Biomarkers Related to Immune Checkpoint Inhibitor Therapy of Triple-Negative Breast Cancer

As an aggressive subtype of breast cancer, triple-negative breast cancer (TNBC) is associated with poor prognosis and lack of effective therapy, except chemotherapy. In recent years, immunotherapy based on immune checkpoint (IC) inhibition has emerged as a promising therapeutic strategy in TNBC. TNBC has more tumor-infiltrating lymphocytes (TILs) and higher rate of mutation and programmed cell death ligand-1 (PD-L1) expression than other subtypes of breast cancer have. However, previous studies have shown that monotherapy has little efficacy and only some TNBC patients can benefit from immunotherapy. Therefore, it is important to identify biomarkers that can predict the efficacy of IC inhibitors (ICIs) in TNBC. Recently, various biomarkers have been extensively explored, such as PD-L1, TILs and tumor mutational burden (TMB). Clinical trials have shown that PD-L1-positive patients with advanced TNBC benefit from ICIs plus chemotherapy. However, in patients with early TNBC receiving neoadjuvant therapy, PD-L1 cannot predict the efficacy of ICIs. These inconsistent conclusions suggest that PD-L1 is the best to date but an imperfect predictive biomarker for efficacy of ICIs. Other studies have shown that advanced TNBC patients with TMB ≥10 mutations/Mb can achieve clinical benefits from pembrolizumab. TILs also have potential predictive value in TNBC. Here, we select some biomarkers related to ICIs and discuss their potential predictive and prognostic value in TNBC. We hope these biomarkers could help to identify suitable patients and realize precision immunotherapy.

ITGAL as a Prognostic Biomarker Correlated With Immune Infiltrates in Gastric Cancer

Integrin alpha L (ITGAL) is a member of the integrin family in which the abnormal expression is linked with carcinogenesis and immune regulation. However, the relation between ITGAL and the prognosis of gastric cancer (GC) and tumor-infiltrating lymphocytes (TILs) are not well understood. The differential expressions of ITGAL in human tumors and the clinical prognosis in GC were systematically analyzed via multiple databases including Gene Expression Profiling Interaction Analysis (GEPIA), UALCAN, Tumor Immune Estimation Resource (TIMER), and Kaplan–Meier (KM) plotter. TIMER, GEPIA, and TISIDB databases were used to comprehensively investigate the correlation between ITGAL and tumor infiltration immune cells. Also, further results were investigated by immunohistochemistry, qRT-PCR, and Western blot. We found that ITGAL expression in GC samples was considerably increased than in peritumor samples. Sample type, subgroup, cancer stage, lymphatic node stage, and worse survival were strongly related to high ITGAL expression. Moreover, upregulated ITGAL expression was strongly connected with immunomodulators, chemokines, and infiltrating levels of CD8+, CD4+ T cell, B cell, monocyte, neutrophil, macrophage, T-cell regulatory, NK cell, and myeloid dendritic cell in stomach adenocarcinoma (STAD). Specifically, immunohistochemistry and bioinformatic analysis showed that ITGAL expression was shown to have strong relationships with various immunological marker sets including PD1 (T-cell exhaustion marker). In conclusion, ITGAL is a prognostic biomarker for GC patients. It might regulate tumor immune microenvironment leading to poor prognosis. Furthermore, studies are essential to explore therapeutic targeting ITGAL.

Prediction of lymph node metastasis in early colorectal cancer based on histologic images by artificial intelligence

Risk evaluation of lymph node metastasis (LNM) for endoscopically resected submucosal invasive (T1) colorectal cancers (CRC) is critical for determining therapeutic strategies, but interobserver variability for histologic evaluation remains a major problem. To address this issue, we developed a machine-learning model for predicting LNM of T1 CRC without histologic assessment. A total of 783 consecutive T1 CRC cases were randomly split into 548 training and 235 validation cases. First, we trained convolutional neural networks (CNN) to extract cancer tile images from whole-slide images, then re-labeled these cancer tiles with LNM status for re-training. Statistical parameters of the tile images based on the probability of primary endpoints were assembled to predict LNM in cases with a random forest algorithm, and defined its predictive value as random forest score. We evaluated the performance of case-based prediction models for both training and validation datasets with area under the receiver operating characteristic curves (AUC). The accuracy for classifying cancer tiles was 0.980. Among cancer tiles, the accuracy for classifying tiles that were LNM-positive or LNM-negative was 0.740. The AUCs of the prediction models in the training and validation sets were 0.971 and 0.760, respectively. CNN judged the LNM probability by considering histologic tumor grade.

Incorporation of TILs in daily breast cancer care: how much evidence can we bear?

One of the most important developments in the breast cancer field has been an improved understanding of prognostic and predictive biomarkers, of which TILs are increasingly gaining importance. The evaluation of TILs by light microscopy on a H&E-stained section is workable in a daily practice setting. Reproducibility of reporting TILs is good, but heterogeneity is a cause of variation. TILs provide clinicians with important prognostic information for patients with TNBC, as early-stage TNBC with high TILs have > 98% 5-year survival and TILs predict benefit to immunotherapy. Importantly, while TILs do not have level of evidence IA, TILs should be used as a prognostic factor with caution and with other accepted prognostic variables, such as tumour size and lymph node status, to inform clinicians and patients on their treatment options. A framework on how to use the TILs in daily practice is proposed, including a co-assessment with PD-L1 for its predictive role to immunotherapy.

Comprehensive analysis of PD-L1 expression, tumor-infiltrating lymphocytes, and tumor microenvironment in LUAD: differences between Asians and Caucasians

Backgrounds

The characteristics of programmed cell death protein-1 (PD-L1) expression, tumor-infiltrating lymphocytes (TILs), and tumor microenvironment (TME) in lung adenocarcinoma (LUAD) patients are closely related to immunotherapy, and there are differences between Asians and Caucasians.

Methods

Acquire the transcriptome data of the Cancer Genome Atlas and Chinese LUAD patients. R software was used to analyze the differential expression of genes, prognosis, and gene function. Use CIBERSORT for TIL-related analysis and ESTIMATE for TME-related analysis.

Results

The expression of PD-L1 in tumor tissues of Caucasian LUAD patients was lower than that in normal tissues, while there was no significant difference in Asians. There was no statistical difference between PD-L1 expression and prognosis. The composition of TILs between Caucasian and Asian LUAD patients was quite different. There was no correlation between TILs and prognosis in Caucasians. However, the higher content of resting mast cells indicated a better prognosis in Asians. The Caucasian patients with higher immune and estimate scores had a better prognosis (p = 0.021, p = 0.025). However, the Asian patients with a higher estimate score had a worse prognosis (p = 0.024). The high expression of COL5A2 (p = 0.046, p = 0.027) and NOX4 (p = 0.020, p = 0.019) were both associated with the poor prognosis in Caucasians and Asians.

Conclusion

There are many differences in the characteristics of PD-L1 expression, TILs, and TME between Caucasian and Asian LUAD patients. This provides a certain hint for the selection of specific immunotherapy strategies separately for Caucasian and Asian LUAD patients.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13148-021-01221-3.

A Pathologist-Annotated Dataset for Validating Artificial Intelligence: A Project Description and Pilot Study

Purpose:

Validating artificial intelligence algorithms for clinical use in medical images is a challenging endeavor due to a lack of standard reference data (ground truth). This topic typically occupies a small portion of the discussion in research papers since most of the efforts are focused on developing novel algorithms. In this work, we present a collaboration to create a validation dataset of pathologist annotations for algorithms that process whole slide images. We focus on data collection and evaluation of algorithm performance in the context of estimating the density of stromal tumor-infiltrating lymphocytes (sTILs) in breast cancer.

Methods:

We digitized 64 glass slides of hematoxylin- and eosin-stained invasive ductal carcinoma core biopsies prepared at a single clinical site. A collaborating pathologist selected 10 regions of interest (ROIs) per slide for evaluation. We created training materials and workflows to crowdsource pathologist image annotations on two modes: an optical microscope and two digital platforms. The microscope platform allows the same ROIs to be evaluated in both modes. The workflows collect the ROI type, a decision on whether the ROI is appropriate for estimating the density of sTILs, and if appropriate, the sTIL density value for that ROI.

Results:

In total, 19 pathologists made 1645 ROI evaluations during a data collection event and the following 2 weeks. The pilot study yielded an abundant number of cases with nominal sTIL infiltration. Furthermore, we found that the sTIL densities are correlated within a case, and there is notable pathologist variability. Consequently, we outline plans to improve our ROI and case sampling methods. We also outline statistical methods to account for ROI correlations within a case and pathologist variability when validating an algorithm.

Conclusion:

We have built workflows for efficient data collection and tested them in a pilot study. As we prepare for pivotal studies, we will investigate methods to use the dataset as an external validation tool for algorithms. We will also consider what it will take for the dataset to be fit for a regulatory purpose: study size, patient population, and pathologist training and qualifications. To this end, we will elicit feedback from the Food and Drug Administration via the Medical Device Development Tool program and from the broader digital pathology and AI community. Ultimately, we intend to share the dataset, statistical methods, and lessons learned.

Machine Learning Models Predict the Primary Sites of Head and Neck Squamous Cell Carcinoma Metastases Based on DNA Methylation

In head and neck squamous cell cancers (HNSCs) that present as metastases with an unknown primary (HNSC-CUPs), the identification of a primary tumor improves therapy options and increases patient survival. However, the currently available diagnostic methods are laborious and do not offer a sufficient detection rate. Predictive machine learning models based on DNA methylation profiles have recently emerged as a promising technique for tumor classification. We applied this technique to HNSC to develop a tool that can improve the diagnostic workup for HNSC-CUPs. On a reference cohort of 405 primary HNSC samples, we developed four classifiers based on different machine learning models (random forest (RF), neural network (NN), elastic net penalized logistic regression (LOGREG), support vector machine (SVM)) that predict the primary site of HNSC tumors from their DNA methylation profile. The classifiers achieved high classification accuracies (RF=83%, NN=88%, LOGREG=SVM=89%) on an independent cohort of 64 HNSC metastases. Further, the NN, LOGREG, and SVM models significantly outperformed p16 status as a marker for an origin in the oropharynx. In conclusion, the DNA methylation profiles of HNSC metastases are characteristic for their primary sites and the classifiers developed in this study, which are made available to the scientific community, can provide valuable information to guide the diagnostic workup of HNSC-CUP. This article is protected by copyright. All rights reserved.

The tale of TILs in breast cancer: A report from The International Immuno-Oncology Biomarker Working Group

The advent of immune-checkpoint inhibitors (ICI) in modern oncology has significantly improved survival in several cancer settings. A subgroup of women with breast cancer (BC) has immunogenic infiltration of lymphocytes with expression of programmed death-ligand 1 (PD-L1). These patients may potentially benefit from ICI targeting the programmed death 1 (PD-1)/PD-L1 signaling axis. The use of tumor-infiltrating lymphocytes (TILs) as predictive and prognostic biomarkers has been under intense examination. Emerging data suggest that TILs are associated with response to both cytotoxic treatments and immunotherapy, particularly for patients with triple-negative BC. In this review from The International Immuno-Oncology Biomarker Working Group, we discuss (a) the biological understanding of TILs, (b) their analytical and clinical validity and efforts toward the clinical utility in BC, and (c) the current status of PD-L1 and TIL testing across different continents, including experiences from low-to-middle-income countries, incorporating also the view of a patient advocate. This information will help set the stage for future approaches to optimize the understanding and clinical utilization of TIL analysis in patients with BC.

Interplay between copy number alterations and immune profiles in the early breast cancer Scandinavian Breast Group 2004-1 randomized phase II trial: results from a feasibility study

Emerging data indicate that genomic alterations can shape immune cell composition in early breast cancer. However, there is a need for complementary imaging and sequencing methods for the quantitative assessment of combined somatic copy number alteration (SCNA) and immune profiling in pathological samples. Here, we tested the feasibility of three approaches-CUTseq, for high-throughput low-input SCNA profiling, multiplexed fluorescent immunohistochemistry (mfIHC) and digital-image analysis (DIA) for quantitative immuno-profiling- in archival formalin-fixed paraffin-embedded (FFPE) tissue samples from patients enrolled in the randomized SBG-2004-1 phase II trial. CUTseq was able to reproducibly identify amplification and deletion events with a resolution of 100 kb using only 6 ng of DNA extracted from FFPE tissue and pooling together 77 samples into the same sequencing library. In the same samples, mfIHC revealed that CD4 + T-cells and CD68 + macrophages were the most abundant immune cells and they mostly expressed PD-L1 and PD-1. Combined analysis showed that the SCNA burden was inversely associated with lymphocytic infiltration. Our results set the basis for further applications of CUTseq, mfIHC and DIA to larger cohorts of early breast cancer patients.

Multiparameter persistent homology landscapes identify immune cell spatial patterns in tumors

Quantifying and comparing complex spatial biological datasets is crucial for medical applications and remains an active area of research. As datasets become more heterogeneous and complicated, so must the methods that are used to understand them. Multiparameter topology is built upon the assumption that the shape of data depends on multiple parameters, such as scale, outliers, or other parameters (e.g., cell density and oxygen levels in the case of tumors). A key difficulty encountered in multiparameter persistent homology (MPH) is interpreting and comparing data. The present work uses statistical MPH landscapes to overcome this difficulty and quantifies differences in synthetic data of immune cell infiltration as well as clinical tumor histology data of T cells, macrophages, and hypoxia.

Highly resolved spatial data of complex systems encode rich and nonlinear information. Quantification of heterogeneous and noisy data—often with outliers, artifacts, and mislabeled points—such as those from tissues, remains a challenge. The mathematical field that extracts information from the shape of data, topological data analysis (TDA), has expanded its capability for analyzing real-world datasets in recent years by extending theory, statistics, and computation. An extension to the standard theory to handle heterogeneous data is multiparameter persistent homology (MPH). Here we provide an application of MPH landscapes, a statistical tool with theoretical underpinnings. MPH landscapes, computed for (noisy) data from agent-based model simulations of immune cells infiltrating into a spheroid, are shown to surpass existing spatial statistics and one-parameter persistent homology. We then apply MPH landscapes to study immune cell location in digital histology images from head and neck cancer. We quantify intratumoral immune cells and find that infiltrating regulatory T cells have more prominent voids in their spatial patterns than macrophages. Finally, we consider how TDA can integrate and interrogate data of different types and scales, e.g., immune cell locations and regions with differing levels of oxygenation. This work highlights the power of MPH landscapes for quantifying, characterizing, and comparing features within the tumor microenvironment in synthetic and real datasets.

Deep learning in cancer diagnosis, prognosis and treatment selection

Deep learning is a subdiscipline of artificial intelligence that uses a machine learning technique called artificial neural networks to extract patterns and make predictions from large data sets. The increasing adoption of deep learning across healthcare domains together with the availability of highly characterised cancer datasets has accelerated research into the utility of deep learning in the analysis of the complex biology of cancer. While early results are promising, this is a rapidly evolving field with new knowledge emerging in both cancer biology and deep learning. In this review, we provide an overview of emerging deep learning techniques and how they are being applied to oncology. We focus on the deep learning applications for omics data types, including genomic, methylation and transcriptomic data, as well as histopathology-based genomic inference, and provide perspectives on how the different data types can be integrated to develop decision support tools. We provide specific examples of how deep learning may be applied in cancer diagnosis, prognosis and treatment management. We also assess the current limitations and challenges for the application of deep learning in precision oncology, including the lack of phenotypically rich data and the need for more explainable deep learning models. Finally, we conclude with a discussion of how current obstacles can be overcome to enable future clinical utilisation of deep learning.

Intra-Tumour Heterogeneity Is One of the Main Sources of Inter-Observer Variation in Scoring Stromal Tumour Infiltrating Lymphocytes in Triple Negative Breast Cancer

Stromal tumour infiltrating lymphocytes (sTILs) are a strong prognostic marker in triple negative breast cancer (TNBC). Consistency scoring sTILs is good and was excellent when an internet-based scoring aid developed by the TIL-WG was used to score cases in a reproducibility study. This study aimed to evaluate the reproducibility of sTILs assessment using this scoring aid in cases from routine practice and to explore the potential of the tool to overcome variability in scoring. Twenty-three breast pathologists scored sTILs in digitized slides of 49 TNBC biopsies using the scoring aid. Subsequently, fields of view (FOV) from each case were selected by one pathologist and scored by the group using the tool. Inter-observer agreement was good for absolute sTILs (ICC 0.634, 95% CI 0.539-0.735, p < 0.001) but was poor to fair using binary cutpoints. sTILs heterogeneity was the main contributor to disagreement. When pathologists scored the same FOV from each case, inter-observer agreement was excellent for absolute sTILs (ICC 0.798, 95% CI 0.727-0.864, p < 0.001) and good for the 20% (ICC 0.657, 95% CI 0.561-0.756, p < 0.001) and 40% (ICC 0.644, 95% CI 0.546-0.745, p < 0.001) cutpoints. However, there was a wide range of scores for many cases. Reproducibility scoring sTILs is good when the scoring aid is used. Heterogeneity is the main contributor to variance and will need to be overcome for analytic validity to be achieved.

Combining Machine Learning and Computational Chemistry for Predictive Insights Into Chemical Systems

Machine learning models are poised to make a transformative impact on chemical sciences by dramatically accelerating computational algorithms and amplifying insights available from computational chemistry methods. However, achieving this requires a confluence and coaction of expertise in computer science and physical sciences. This Review is written for new and experienced researchers working at the intersection of both fields. We first provide concise tutorials of computational chemistry and machine learning methods, showing how insights involving both can be achieved. We follow with a critical review of noteworthy applications that demonstrate how computational chemistry and machine learning can be used together to provide insightful (and useful) predictions in molecular and materials modeling, retrosyntheses, catalysis, and drug design.

Technological advances in cancer immunity: from immunogenomics to single-cell analysis and artificial intelligence

Immunotherapies play critical roles in cancer treatment. However, given that only a few patients respond to immune checkpoint blockades and other immunotherapeutic strategies, more novel technologies are needed to decipher the complicated interplay between tumor cells and the components of the tumor immune microenvironment (TIME). Tumor immunomics refers to the integrated study of the TIME using immunogenomics, immunoproteomics, immune-bioinformatics, and other multi-omics data reflecting the immune states of tumors, which has relied on the rapid development of next-generation sequencing. High-throughput genomic and transcriptomic data may be utilized for calculating the abundance of immune cells and predicting tumor antigens, referring to immunogenomics. However, as bulk sequencing represents the average characteristics of a heterogeneous cell population, it fails to distinguish distinct cell subtypes. Single-cell-based technologies enable better dissection of the TIME through precise immune cell subpopulation and spatial architecture investigations. In addition, radiomics and digital pathology-based deep learning models largely contribute to research on cancer immunity. These artificial intelligence technologies have performed well in predicting response to immunotherapy, with profound significance in cancer therapy. In this review, we briefly summarize conventional and state-of-the-art technologies in the field of immunogenomics, single-cell and artificial intelligence, and present prospects for future research.

Correlation of manual semi-quantitative and automated quantitative Ki-67 proliferative index with OncotypeDXTM recurrence score in invasive breast carcinoma

BACKGROUND

Ki-67 immunohistochemistry (IHC) staining is a widely used cancer proliferation assay; however, its limitations could be improved with automated scoring. The OncotypeDXTM Recurrence Score (ORS), which primarily evaluates cancer proliferation genes, is a prognostic indicator for breast cancer chemotherapy response; however, it is more expensive and slower than Ki-67.

OBJECTIVE

To compare manual Ki-67 (mKi-67) with automated Ki-67 (aKi-67) algorithm results based on manually selected Ki-67 "hot spots" in breast cancer, and correlate both with ORS.

METHODS

105 invasive breast carcinoma cases from 100 patients at our institution (2011-2013) with available ORS were evaluated. Concordance was assessed via Cohen's Kappa (κ).

RESULTS

57/105 cases showed agreement between mKi-67 and aKi-67 (κ 0.31, 95% CI 0.18-0.45), with 41 cases overestimated by aKi-67. Concordance was higher when estimated on the same image (κ 0.53, 95% CI 0.37-0.69). Concordance between mKi-67 score and ORS was fair (κ 0.27, 95% CI 0.11-0.42), and concordance between aKi-67 and ORS was poor (κ 0.10, 95% CI -0.03-0.23).

CONCLUSIONS

These results highlight the limits of Ki-67 algorithms that use manual "hot spot" selection. Due to suboptimal concordance, Ki-67 is likely most useful as a complement to, rather than a surrogate for ORS, regardless of scoring method.

Quantitative assessment of inflammatory infiltrates in kidney transplant biopsies using multiplex tyramide signal amplification and deep learning

Delayed graft function (DGF) is a strong risk factor for development of interstitial fibrosis and tubular atrophy (IFTA) in kidney transplants. Quantitative assessment of inflammatory infiltrates in kidney biopsies of DGF patients can reveal predictive markers for IFTA development. In this study, we combined multiplex tyramide signal amplification (mTSA) and convolutional neural networks (CNNs) to assess the inflammatory microenvironment in kidney biopsies of DGF patients (n = 22) taken at 6 weeks post-transplantation. Patients were stratified for IFTA development (<10% versus ≥10%) from 6 weeks to 6 months post-transplantation, based on histopathological assessment by three kidney pathologists. One mTSA panel was developed for visualization of capillaries, T- and B-lymphocytes and macrophages and a second mTSA panel for T-helper cell and macrophage subsets. The slides were multi spectrally imaged and custom-made python scripts enabled conversion to artificial brightfield whole-slide images (WSI). We used an existing CNN for the detection of lymphocytes with cytoplasmatic staining patterns in immunohistochemistry and developed two new CNNs for the detection of macrophages and nuclear-stained lymphocytes. F1-scores were 0.77 (nuclear-stained lymphocytes), 0.81 (cytoplasmatic-stained lymphocytes), and 0.82 (macrophages) on a test set of artificial brightfield WSI. The CNNs were used to detect inflammatory cells, after which we assessed the peritubular capillary extent, cell density, cell ratios, and cell distance in the two patient groups. In this cohort, distance of macrophages to other immune cells and peritubular capillary extent did not vary significantly at 6 weeks post-transplantation between patient groups. CD163+ cell density was higher in patients with ≥10% IFTA development 6 months post-transplantation (p < 0.05). CD3+CD8-/CD3+CD8+ ratios were higher in patients with <10% IFTA development (p < 0.05). We observed a high correlation between CD163+ and CD4+GATA3+ cell density (R = 0.74, p < 0.001). Our study demonstrates that CNNs can be used to leverage reliable, quantitative results from mTSA-stained, multi spectrally imaged slides of kidney transplant biopsies.

DIMEimmune: Robust estimation of infiltrating lymphocytes in CNS tumors from DNA methylation profiles

The interaction of CNS tumors with infiltrating lymphocytes plays an important role in their initiation and progression and might be related to therapeutic responses. Gene expression-based methods have been successfully used to characterize the tumor microenvironment. However, methylation data are now increasingly used for molecular diagnostics and there are currently only few methods to infer information about the microenvironment from this data type. Using an approach based on differential methylation and principal component analysis, we developed DIMEimmune (Differential Methylation Analysis for Immune Cell Estimation) to estimate CD4⁺ and CD8⁺ T cell abundance as well as tumor-infiltrating lymphocytes (TILs) scores from bulk methylation data. Well-established approaches based on gene expression data and immunohistochemistry-based lymphocyte counts were used as benchmarks. The comparison of DIMEimmune to the previously published MethylCIBERSORT and MeTIL algorithms showed an improved correlation with both gene expression-based and immunohistological results across different brain tumor types. Further, we applied our method to large datasets of glioma, medulloblastoma, atypical teratoid/rhabdoid tumors (ATRTs) and ependymoma. High-grade gliomas showed higher scores of tumor-infiltrating lymphocytes than lower-grade gliomas. There were overall only few tumor-infiltrating lymphocytes in medulloblastoma subgroups. ATRTs were highly infiltrated by lymphocytes, most prominently in the MYC subgroup. DIMEimmune-based estimates of TILs were a significant prognostic factor in the overall cohort of gliomas and medulloblastomas, but not within methylation-based diagnostic subgroups. To conclude, DIMEimmune allows for robust estimates of TIL abundance and might contribute to establishing them as a prognostic or predictive factor in future studies of CNS tumors.

The Correlation Between SPP1 and Immune Escape of EGFR Mutant Lung Adenocarcinoma Was Explored by Bioinformatics Analysis

Background

Immune checkpoint inhibitors have achieved breakthrough efficacy in treating lung adenocarcinoma (LUAD) with wild-type epidermal growth factor receptor (EGFR), leading to the revision of the treatment guidelines. However, most patients with EGFR mutation are resistant to immunotherapy. It is particularly important to study the differences in tumor microenvironment (TME) between patients with and without EGFR mutation. However, relevant research has not been reported. Our previous study showed that secreted phosphoprotein 1 (SPP1) promotes macrophage M2 polarization and PD-L1 expression in LUAD, which may influence response to immunotherapy. Here, we assessed the role of SPP1 in different populations and its effects on the TME.

Methods

We compared the expression of SPP1 in LUAD tumor and normal tissues, and in samples with wild-type and mutant EGFR. We also evaluated the influence of SPP1 on survival. The LUAD data sets were downloaded from TCGA and CPTAC databases. Clinicopathologic characteristics associated with overall survival in TCGA were assessed using Cox regression analysis. GSEA revealed that several fundamental signaling pathways were enriched in the high SPP1 expression group. We applied CIBERSORT and xCell to calculate the proportion and abundance of tumor-infiltrating immune cells (TICs) in LUAD, and compared the differences in patients with high or low SPP1 expression and wild-type or mutant EGFR. In addition, we explored the correlation between SPP1 and CD276 for different groups.

Results

SPP1 expression was higher in LUAD tumor tissues and in people with EGFR mutation. High SPP1 expression was associated with poor prognosis. Univariate and multivariate cox analysis revealed that up-regulated SPP1 expression was independent indicator of poor prognosis. GSEA showed that the SPP1 high expression group was mainly enriched in immunosuppressed pathways. In the SPP1 high expression group, the infiltration of CD8+ T cells was lower and M2-type macrophages was higher. These results were also observed in patients with EGFR mutation. Furthermore, we found that the SPP1 expression was positively correlated with CD276, especially in patients with EGFR mutation.

Conclusion

SPP1 levels might be a useful marker of immunosuppression in patients with EGFR mutation, and could offer insight for therapeutics.

An Open Source, Automated Tumor Infiltrating Lymphocyte Algorithm for Prognosis in Triple-Negative Breast Cancer

PURPOSE

Although tumor infiltrating lymphocytes (TIL) assessment has been acknowledged to have both prognostic and predictive importance in triple negative breast cancer (TNBC), it is subject to inter and intra-observer variability that has prevented widespread adoption. Here we constructed a machine-learning based breast cancer TIL scoring approach and validated its prognostic potential in multiple TNBC cohorts.

EXPERIMENTAL DESIGN

Using the QuPath open source software, we built a neural-network classifier for tumor cells, lymphocytes, fibroblasts and "other" cells on hematoxylin-eosin (H&E) stained sections. We analyzed the classifier-derived TIL measurements with five unique constructed TIL variables. A retrospective collection of 171 TNBC cases was used as the discovery set to identify the optimal association of machine-read TIL variables with patient outcome. For validation we evaluated a retrospective collection of 749 TNBC patients comprised of four independent validation subsets.

RESULTS

We found that all five machine TIL variables had significant prognostic association with outcomes (p{less than or equal to}0.01 for all comparisons) but showed cell specific variation in validation sets. Cox regression analysis demonstrated that all five TIL variables were independently associated with improved overall survival after adjusting for clinicopathological factors including stage, age and histological grade (p{less than or equal to}0.003 for all analyses).

CONCLUSIONS

Neural net driven cell classifier defined TIL variables were robust and independent prognostic factors in several independent validation cohorts of TNBC patients. These objective, open source TIL variables are freely available to download and can now be considered for testing in a prospective setting to assess clinical utility.

An Immune Model to Predict Prognosis of Breast Cancer Patients Receiving Neoadjuvant Chemotherapy Based on Support Vector Machine

Tumor microenvironment has been increasingly proved to be crucial during the development of breast cancer. The theory about the conversion of cold and hot tumor attracted the attention to the influences of traditional therapeutic strategies on immune system. Various genetic models have been constructed, although the relation between immune system and local microenvironment still remains unclear. In this study, we tested and collected the immune index of 262 breast cancer patients before and after neoadjuvant chemotherapy. Five indexes were selected and analyzed to form the prediction model, including the ratio values between after and before neoadjuvant chemotherapy of CD4⁺/CD8⁺ T cell ratio; lymphosum of T, B, and natural killer (NK) cells; CD3⁺CD8⁺ cytotoxic T cell percent; CD16⁺CD56⁺ NK cell absolute value; and CD3⁺CD4⁺ helper T cell percent. Interestingly, these characters are both the ratio value of immune status after neoadjuvant chemotherapy to the baseline. Then the prediction model was constructed by support vector machine (accuracy rate = 75.71%, area under curve = 0.793). Beyond the prognostic effect and prediction significance, the study instead emphasized the importance of immune status in traditional systemic therapies. The result provided new evidence that the dynamic change of immune status during neoadjuvant chemotherapy should be paid more attention.