Sentinel Lymph Node in Breast Cancer: Review Article from a Pathologist's Point of View

Frozen section analysis in community settings: Diagnostic challenges and key considerations

The frozen section is a useful tool for pathologists and surgeons as it provides rapid microscopic evaluation of tissue in the intraoperative setting. The diagnosis rendered on the frozen section assists the surgeon in making intraoperative decisions. In community hospitals, which usually have a general surgical pathology coverage model, frozen section requests occur in various clinical settings. This review covers the everyday situations that require frozen section analysis in a community-based hospital setting. It outlines the frozen section procedure, discusses histologic evaluation, and highlights the diagnostic challenges that may arise during this process.

Out-of-distribution generalization for segmentation of lymph node metastasis in breast cancer

Pathology provides the definitive diagnosis, and Artificial Intelligence (AI) tools are poised to improve accuracy, inter-rater agreement, and turn-around time (TAT) of pathologists, leading to improved quality of care. A high value clinical application is the grading of Lymph Node Metastasis (LNM) which is used for breast cancer staging and guides treatment decisions. A challenge of implementing AI tools widely for LNM classification is domain shift, where Out-of-Distribution (OOD) data has a different distribution than the In-Distribution (ID) data used to train the model, resulting in a drop in performance in OOD data. This work proposes a novel clustering and sampling method to automatically curate training datasets in an unsupervised manner with the aim of improving model generalization abilities. To evaluate the generalization performance of the proposed models, we applied a novel use of the Two One-sided Tests (TOST) method. This method examines whether the performance on ID and OOD data is equivalent, serving as a proxy for generalization. We provide the first evidence for computing equivalence margins that are data-dependent, which reduces subjectivity. The proposed framework shows the ensembled models constructed from models that generalized across both tumor and normal patches enhanced performance, achieving an F1 score of 0.81 for LNM classification on unseen ID and OOD samples. Interactive viewing of slide-level segmentations can be accessed on PathcoreFlow™ through https://web.pathcore.com/folder/18555?s=QTJVHJuhrfe5 . Segmentation models are available at https://github.com/IAMLAB-Ryerson/OOD-Generalization-LNM .

Prognosis of isolated tumor cells and use of molecular classification in early stage endometrioid endometrial cancer

OBJECTIVE

We assessed the prognosis and molecular subtypes of early stage endometrioid endometrial cancer with isolated tumor cells within sentinel lymph nodes (SLNs) compared with node negative disease.

METHODS

Patients diagnosed with stage IA, IB, or II endometrioid endometrial cancer and primary surgical management were identified from January 1, 2007 to December 31, 2019. All SLNs underwent ultrastaging according to the institutional protocol. Patients with cytokeratin positive cells, micrometastases, and macrometastases were excluded. Clinical, pathology, and molecular subtype data were reviewed.

RESULTS

Overall, 1214 patients with early stage endometrioid endometrial cancer met the inclusion criteria, of whom 1089 (90%) had node negative disease and 125 (10%) had isolated tumor cells. Compared with node negative disease, the presence of isolated tumor cells had a greater association with deep myometrial invasion, lymphovascular space invasion, receipt of adjuvant therapy, and adjuvant chemotherapy with or without radiation (p<0.01). There was no significant difference in survival rates between patients with isolated tumor cells and node negative disease (3 year progression free survival rate 94% vs 91%, respectively, p=0.21; 3 year overall survival rate 98% vs 96%, respectively, p=0.45). Progression free survival did not significantly differ among patients with isolated tumor cells who received no adjuvant therapy or chemotherapy with or without radiation (p=0.31). There was no difference in the distribution of molecular subtypes between patients with isolated tumor cells (n=28) and node negative disease (n=194; p=0.26). Three year overall survival rates differed significantly when stratifying the entire cohort by molecular subtype (p=0.04).

CONCLUSIONS

Patients with isolated tumor cells demonstrated less favorable uterine pathologic features and received more adjuvant treatment with similar survival compared with patients with nodenegative disease. Among the available data, molecular classification did not have a significant association with the presence of isolated tumor cells, although copy number-high status was a poor prognostic indicator in early stage endometrioid endometrial cancer.

Clonal dominance defines metastatic dissemination in pancreatic cancer

Tumors represent ecosystems where subclones compete during tumor growth. While extensively investigated, a comprehensive picture of the interplay of clonal lineages during dissemination is still lacking. Using patient-derived pancreatic cancer cells, we created orthotopically implanted clonal replica tumors to trace clonal dynamics of unperturbed tumor expansion and dissemination. This model revealed the multifaceted nature of tumor growth, with rapid changes in clonal fitness leading to continuous reshuffling of tumor architecture and alternating clonal dominance as a distinct feature of cancer growth. Regarding dissemination, a large fraction of tumor lineages could be found at secondary sites each having distinctive organ growth patterns as well as numerous undescribed behaviors such as abortive colonization. Paired analysis of primary and secondary sites revealed fitness as major contributor to dissemination. From the analysis of pro- and nonmetastatic isogenic subclones, we identified a transcriptomic signature able to identify metastatic cells in human tumors and predict patients' survival.

Association between postoperative pathological results and non-sentinel nodal metastasis in breast cancer patients with sentinel lymph node-positive breast cancer

OBJECTIVE

For patients with 1-2 positive sentinel lymph nodes (SLN) identified by biopsy, the necessity of axillary lymph node dissection (ALND) remains a matter of debate. The primary aim of this study was to investigate the association between postoperative pathological factors and non-sentinel lymph node (NSLN) metastases in Chinese patients diagnosed with sentinel node-positive breast cancer.

METHODS

This research involved a total of 280 individuals with SLN-positive breast cancer. The relationship between postoperative pathological variables and non-sentinel lymph node metastases was scrutinized using univariate, multivariate, and stratified analysis.

RESULTS

Among the 280 patients with a complete count of SLN positives, 126 (45.0%) exhibited NSLN metastasis. Within this group, 45 cases (35.71%) had 1 SLN positive, while 81 cases (64.29%) demonstrated more than 1 SLN positive. Multivariate logistic regression analysis revealed that HER2 expression status (OR 2.25, 95% CI 1.10-4.60, P = 0.0269), LVI (OR 6.08, 95% CI 3.31-11.14, P < 0.0001), and the number of positive SLNs (OR 4.17, 95% CI 2.35-7.42, P < 0.0001) were positively correlated with NSLNM.

CONCLUSION

In our investigation, the risk variables for NSLN metastasis included LVI, HER2 expression, and the quantity of positive sentinel lymph nodes. However, further validation is imperative, including this institution, distinct institutions, and diverse patient populations.

The Role of Nodes and Nodal Assessment in Diagnosis, Treatment and Prediction in ER+, Node-Positive Breast Cancer

The majority of breast cancers are oestrogen receptor-positive (ER+). In ER+ cancers, oestrogen acts as a disease driver, so these tumours are likely to be susceptible to endocrine therapy (ET). ET works by blocking the hormone's synthesis or effect. A significant number of patients diagnosed with breast cancer will have the spread of tumour cells into regional lymph nodes either at the time of diagnosis, or as a recurrence some years later. Patients with node-positive disease have a poorer prognosis and can respond less well to ET. The nodal metastases may be genomically similar or, as is becoming more evident, may differ from the primary tumour. However, nodal metastatic disease is often not assessed, and treatment decisions are almost always based on biomarkers evaluated in the primary tumour. This review will summarise the evidence in the field on ER+, node-positive breast cancer, including diagnosis, treatment, prognosis and predictive tools.

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.

Predictive factors for dissection-free sentinel node micrometastases in early oral squamous cell carcinoma

This sentinel node (SN) biopsy trial aimed to assess its effectiveness in identifying predictive factors of micrometastases and to determine whether elective neck dissection is necessary in oral squamous cell carcinoma. This retrospective study included 55 patients from three previous trials, with positive SNs. The relationship between the sizes of the metastatic focus and metastasis in non-sentinel node (NSN) was investigated. Four of the 55 largest metastatic focus were isolated tumor cells, and the remaining 51 were ranged from 0.2 to 15 mm, with a median of 2.6 mm. The difference of prevalence between 46 negative- and 9 positive-NSN was statistically significant with regard to age, long diameter of primary site and number of cases with regional recurrence. In comparing the size of largest metastatic focus dividing the number of positive SN, with metastaic focus range of < 3.0 mm in one-positive SN group, there were 18 (33%) negative-NSN and no positive-NSN. Regarding prognosis, 3-year overall survival rate of this group (n = 18) and other (n = 37) were 94% and 73% (p = 0.04), and 3-year recurrence free survival rate of this group and other were 94% and 51% (p = 0.03), respectively. Absolutely a further prospective clinical trial would be needed, micrometastases may be defined as solitary SN metastasis with < 3.0 mm of metastatic focus, and approximately 33% of neck dissections could be avoided using these criteria.

Generalization of Deep Learning in Digital Pathology: Experience in Breast Cancer Metastasis Detection

Poor generalizability is a major barrier to clinical implementation of artificial intelligence in digital pathology. The aim of this study was to test the generalizability of a pretrained deep learning model to a new diagnostic setting and to a small change in surgical indication. A deep learning model for breast cancer metastases detection in sentinel lymph nodes, trained on CAMELYON multicenter data, was used as a base model, and achieved an AUC of 0.969 (95% CI 0.926-0.998) and FROC of 0.838 (95% CI 0.757-0.913) on CAMELYON16 test data. On local sentinel node data, the base model performance dropped to AUC 0.929 (95% CI 0.800-0.998) and FROC 0.744 (95% CI 0.566-0.912). On data with a change in surgical indication (axillary dissections) the base model performance indicated an even larger drop with a FROC of 0.503 (95%CI 0.201-0.911). The model was retrained with addition of local data, resulting in about a 4% increase for both AUC and FROC for sentinel nodes, and an increase of 11% in AUC and 49% in FROC for axillary nodes. Pathologist qualitative evaluation of the retrained model´s output showed no missed positive slides. False positives, false negatives and one previously undetected micro-metastasis were observed. The study highlights the generalization challenge even when using a multicenter trained model, and that a small change in indication can considerably impact the model´s performance.

Independent assessment of a deep learning system for lymph node metastasis detection on the Augmented Reality Microscope

Several machine learning algorithms have demonstrated high predictive capability in the identification of cancer within digitized pathology slides. The Augmented Reality Microscope (ARM) has allowed these algorithms to be seamlessly integrated within the pathology workflow by overlaying their inferences onto its microscopic field of view in real time. We present an independent assessment of the LYmph Node Assistant (LYNA) models, state-of-the-art algorithms for the identification of breast cancer metastases in lymph node biopsies, optimized for usage on the ARM. We assessed the models on 40 whole slide images at the commonly used objective magnifications of 10×, 20×, and 40×. We analyzed their performance across clinically relevant subclasses of tissue, including breast cancer, lymphocytes, histiocytes, blood, and fat. Each model obtained overall AUC values of approximately 0.98, accuracy values of approximately 0.94, and sensitivity values above 0.88 at classifying small regions of a field of view as benign or cancerous. Across tissue subclasses, the models performed most accurately on fat and blood, and least accurately on histiocytes, germinal centers, and sinus. The models also struggled with the identification of isolated tumor cells, especially at lower magnifications. After testing, we reviewed the discrepancies between model predictions and ground truth to understand the causes of error. We introduce a distinction between proper and improper ground truth for analysis in cases of uncertain annotations. Taken together, these methods comprise a novel approach for exploratory model analysis over complex anatomic pathology data in which precise ground truth is difficult to establish.

Spatiality Sensitive Learning for Cancer Metastasis Detection in Whole-Slide Images

Metastasis detection in lymph nodes via microscopic examination of histopathological images is one of the most crucial diagnostic procedures for breast cancer staging. The manual analysis is extremely labor-intensive and time-consuming because of complexities and diversities of histopathology images. Deep learning has been utilized in automatic cancer metastasis detection in recent years. Due to the huge size of whole-slide images, most existing approaches split each image into smaller patches and simply treat these patches independently, which ignores the spatial correlations among them. To solve this problem, this paper proposes an effective spatially sensitive learning framework for cancer metastasis detection in whole-slide images. Moreover, a novel spatial loss function is designed to ensure the consistency of prediction over neighboring patches. Specifically, through incorporating long short-term memory and spatial loss constraint on top of a convolutional neural network feature extractor, the proposed method can effectively learn both the appearance of each patch and spatial relationships between adjacent image patches. With the standard back-propagation algorithm, the whole framework can be trained in an end-to-end way. Finally, the regions with high tumor probability in the resulting probability map are the metastasis locations. Extensive experiments on the benchmark Camelyon 2016 Grand Challenge dataset show the effectiveness of the proposed approach with respect to state-of-the-art competitors. The obtained precision, recall, and balanced accuracy are 0.9565, 0.9167, and 0.9458, respectively. It is also demonstrated that the proposed approach can provide more accurate detection results and is helpful for early diagnosis of cancer metastasis.

Improving Cancer Metastasis Detection via Effective Contrastive Learning

The metastasis detection in lymph nodes via microscopic examination of H&E stained histopathological images is one of the most crucial diagnostic procedures for breast cancer staging. The manual analysis is extremely labor-intensive and time-consuming because of complexities and diversities of histopathological images. Deep learning has been utilized in automatic cancer metastasis detection in recent years. The success of supervised deep learning is credited to a large labeled dataset, which is hard to obtain in medical image analysis. Contrastive learning, a branch of self-supervised learning, can help in this aspect through introducing an advanced strategy to learn discriminative feature representations from unlabeled images. In this paper, we propose to improve breast cancer metastasis detection through self-supervised contrastive learning, which is used as an accessional task in the detection pipeline, allowing a feature extractor to learn more valuable representations, even if there are fewer annotation images. Furthermore, we extend the proposed approach to exploit unlabeled images in a semi-supervised manner, as self-supervision does not need labeled data at all. Extensive experiments on the benchmark Camelyon2016 Grand Challenge dataset demonstrate that self-supervision can improve cancer metastasis detection performance leading to state-of-the-art results.

Comparative Analysis of Postoperative Complications of Sentinel Node Identification Using the SentiMag® Method and the Use of a Radiotracer in Patients with Breast Cancer

(1) Background: The purpose of the study was a retrospective, comparative assessment of complications of the surgical sentinel node biopsy (SNB) procedure in breast cancer using the radiotracer method and the SentiMag® method on groups of patients after 3.5 years of use. (2) Methods: The material was a group of 345 patients with primary surgical breast cancer who underwent the SNB procedure with the use of a radiotracer in combination with wide local excision (WLE), simple amputation (SA) with SNB and an independent SNB procedure in the period from May 2018 to January 2021 in the Department of Oncological Surgery. Of the patients who were monitored in the Hospital Outpatient Clinic, 300 were enrolled. The analyzed group was compared in terms of the occurrence of the same complications with the group of 303 patients also operated on in our center in the period from January 2014 to September 2017, in which SN identification was performed using the SentiMag® method. (3) Results: The most common complications found were sensation disorders in the arm, which occurred in 16 (14.1%) patients using the radiotracer method, SentiMag®-11 (9.9%). By comparing the complication rate between the methods with the radiotracer (n = 300) and SentiMag® (n = 303), no significant differences were found. (4) Conclusions: Sentinel node (SN) identification using the radiotracer method and the SentiMag® method are comparable diagnostic methods in breast cancer, with a low risk of complications.

Fast Segmentation of Metastatic Foci in H&E Whole-Slide Images for Breast Cancer Diagnosis

Breast cancer is the leading cause of death for women globally. In clinical practice, pathologists visually scan over enormous amounts of gigapixel microscopic tissue slide images, which is a tedious and challenging task. In breast cancer diagnosis, micro-metastases and especially isolated tumor cells are extremely difficult to detect and are easily neglected because tiny metastatic foci might be missed in visual examinations by medical doctors. However, the literature poorly explores the detection of isolated tumor cells, which could be recognized as a viable marker to determine the prognosis for T1NoMo breast cancer patients. To address these issues, we present a deep learning-based framework for efficient and robust lymph node metastasis segmentation in routinely used histopathological hematoxylin−eosin-stained (H−E) whole-slide images (WSI) in minutes, and a quantitative evaluation is conducted using 188 WSIs, containing 94 pairs of H−E-stained WSIs and immunohistochemical CK(AE1/AE3)-stained WSIs, which are used to produce a reliable and objective reference standard. The quantitative results demonstrate that the proposed method achieves 89.6% precision, 83.8% recall, 84.4% F1-score, and 74.9% mIoU, and that it performs significantly better than eight deep learning approaches, including two recently published models (v3_DCNN and Xception-65), and three variants of Deeplabv3+ with three different backbones, namely, U-Net, SegNet, and FCN, in precision, recall, F1-score, and mIoU (p<0.001). Importantly, the proposed system is shown to be capable of identifying tiny metastatic foci in challenging cases, for which there are high probabilities of misdiagnosis in visual inspection, while the baseline approaches tend to fail in detecting tiny metastatic foci. For computational time comparison, the proposed method takes 2.4 min for processing a WSI utilizing four NVIDIA Geforce GTX 1080Ti GPU cards and 9.6 min using a single NVIDIA Geforce GTX 1080Ti GPU card, and is notably faster than the baseline methods (4-times faster than U-Net and SegNet, 5-times faster than FCN, 2-times faster than the 3 different variants of Deeplabv3+, 1.4-times faster than v3_DCNN, and 41-times faster than Xception-65).

Ultrasonographic Algorithm for the Assessment of Sentinel Lymph Nodes That Drain the Mammary Carcinomas in Female Dogs

The status of sentinel lymph nodes (SLNs) is decisive in staging, prognosis, and therapeutic approach. Using an ultrasonographic examination algorithm composed of B-mode, Doppler technique, contrast-enhanced ultrasound (CEUS) and elastography, this study aimed to determine the diagnostic performance of the four techniques compared to histopathological examination. 96 SLNs belonging to 71 female dogs with mammary gland carcinomas were examined. After examinations, mastectomy and lymphadenectomy were performed. Histopathological examination confirmed the presence of metastases in 54 SLNs. The elasticity score had the highest accuracy-89.71%, identifying metastases in SLNs with 88.9.9% sensitivity (SE) and 90.5% specificity (SP), ROC analysis providing excellent results. The S/L (short axis/long axis) ratio showed 83.3% SE and 78.6% SP as a predictor of the presence of metastases in SLN having a good accuracy of 81.2%. On Doppler examination, the resistivity index(RI) showed good accuracy of 80% in characterizing lymph nodes with metastases versus unaffected ones; the same results being obtained by CEUS examination. By assigning to each ultrasonographic parameter a score (0 or 1) and summing up the scores of the four techniques, we obtained the best diagnostic performance in identifying lymph node metastases with 92.2% accuracy. In conclusion, the use of the presented algorithm provides the best identification of metastases in SLNs, helping in mammary carcinoma staging and appropriate therapeutic management.

Attempting to define sentinel node micrometastasis in oral squamous cell carcinoma

OBJECTIVE

The aim of this supplemental study of a sentinel node (SN) biopsy (SNB) trial for oral squamous cell carcinoma (OSCC) was to assess the effectiveness in identifying micrometastasis and determining whether elective neck dissection (END) is necessary.

MATERIALS AND METHODS

Twenty-three patients with pathologically positive SNs were included. The sizes of the metastatic lesions in positive SNs (SMSNs) were classified and the rates of occult metastasis of non-SNs were compared.

RESULTS

The patients were divided according to the SMSN:<0.2 mm (group A, n=3);0.2 mm to <2.0 mm (group B, n=7);and ≥2.0 mm (group C, n=13). The rates of occult metastasis in groups A, B, and C were 0% (0/3), 14% (1/7) and 23% (3/13), respectively.

CONCLUSION

Rare cancer cell distribution to nodes other than SNs was observed in the patients with SN metastatic lesions of at least smaller than 0.2 mm in size, suggesting the possibility of defining SN micrometastasis in N0 OSCC.

The spectrum of pathological diagnoses in non-sentinel axillary lymph node biopsy: A single institution's experience

Although axillary lymphadenopathy is a common clinical encounter, systemic evaluation of non-sentinel lymph node biopsy is sparse. We reviewed our institution's 15-year experience to delineate the spectrum of diagnoses in non-sentinel axillary lymph nodes. 1165 non-sentinel axillary lymph node biopsies were retrieved and the diagnosis and relevant clinical information was reviewed. This spectrum of diagnoses was further stratified by gender, age, and oncologic history. The spectrum of diagnoses included: breast carcinoma (27.6%), lymphoma (29.2%), melanoma (3.5%), other carcinoma (2.9%), sarcoma (0.4%), and benign changes (36.3%). The most common diagnoses in men were lymphoma (61.8%) and benign changes (23.6%); while in women they were benign change (41.2%), breast carcinoma (37.8%) and lymphoma (16.7%). Besides benign changes, lymphoma and breast carcinoma were most common in women younger and older than 30 years, respectively. In patients with a history of malignancy, the most common diagnoses were metastasis from the known tumor and benign change; while in patients with a negative oncologic history and female patients without a history of breast cancer, the diagnosis was generally either lymphoma or benign change. Anaplastic large cell lymphoma was rare but may be mistaken as metastatic carcinoma thus a high index of suspicion is warranted. Thus through retrospective review of a large cohort of non-sentinel axillary lymph node biopsies, we described the spectrum of pathological entities based on the gender, age, and clinical history, which could provide valuable information for further work-up of axillary lymph node biopsy.

Prognostic Value of EndoPredict in Women with Hormone Receptor-Positive, HER2-Negative Invasive Lobular Breast Cancer

PURPOSE

Invasive lobular carcinoma (ILC) accounts for approximately 5%-15% of all invasive breast cancer cases. Most of the correlations between multigene assays and patient outcome were derived from studies based on patients with invasive ductal carcinoma (IDC) or without distinction between the subtypes. Here, we investigate the prognostic value of EndoPredict (EPclin) in a large cohort of ILCs pooled from three phase III randomized trials (ABCSG-6, ABCSG-8, TransATAC).

EXPERIMENTAL DESIGN

The primary objective of this analysis was to determine the prognostic value of EPclin for distant recurrence (DR) in years 0-10 in postmenopausal women with ILC. The primary outcome was DR.

RESULTS

470 women (17.9%) presented with ILC, 1,944 (73.9%) with IDC, and 216 (8.2%) with other histologic types. EPclin was highly prognostic in women with ILC [HR = 3.32 (2.54-4.34)] and provided more prognostic value than the Clinical Treatment Score [CTS; HR = 2.17 (1.73-2.72)]. 63.4% of women were categorized into the low EPclin risk group and they had a 10-year DR of 4.8% (2.7-8.4) compared with 36.6% of women in the high-risk group with a 10-year DR risk of 26.6% (20.0-35.0). EPclin also provided highly prognostic information in women with node-negative disease [HR = 2.56 (1.63-4.02)] and node-positive disease [HR = 3.70 (2.49-5.50)].

CONCLUSIONS

EPclin provided highly significant prognostic value and significant risk stratification for women with ILC. Ten-year DR risk in the EPclin low-risk groups were similar between ILC and IDC. Our results show that EPclin is informative in women with ILC and suggest that it is equally valid in both histologic subtypes.

Independent risk factors for axillary lymph node metastasis in breast cancer patients with one or two positive sentinel lymph nodes

Background

The benefit of axillary lymph node dissection (ALND) in breast cancer patients with one or two positive sentinel lymph nodes (SLNs) remains inconclusive. The purpose of this study was to identify risk factors independently associated with axillary lymph node (ALN) metastasis.

Methods

We retrospectively analyzed data from 389 Chinese breast cancer patients with one or two positive SLNs who underwent ALND. Univariate and multivariate logistic regression analyses were performed to identify ALN metastasis-associated risk factors.

Results

Among the 389 patients, 174 (44.7%) had ALN metastasis, while 215 (55.3%) showed no evidence of ALN metastasis. Univariate analysis revealed significant differences in age (< 60 or ≥ 60 years), human epidermal growth factor receptor-2 (Her-2) status, and the ratio of positive to total SLNs between the ALN metastasis and non-metastasis groups (P < 0.05). The multivariate analysis indicated that age, the ratio of positive to total SLNs, and occupations were significantly different between the two groups. Lastly, younger age (< 60 years), a higher ratio of positive to total SLNs, and manual labor jobs were independently associated with ALN metastasis (P < 0.05).

Conclusions

The risk of ALN metastasis in breast cancer patients with one or two positive SLNs can be further increased by younger age, manual labor jobs, and a high ratio of positive to total SLNs. Our findings may also aid in identifying which patients with one or two positive SLNs may not require ALND.

Isolated tumor cells in the regional lymph nodes in patients with squamous cell carcinoma of the esophagus are rarely observed but often represent part of a true metastasis

The most common malignancy of the esophagus is squamous cell carcinoma (SCC) and regional lymph node metastases are an important prognostic factor. Isolated tumor cells (ITCs) are defined as single tumor cells or small clusters of tumor cells not exceeding 0.2 mm. The prognostic role of ITCs is not clear. This study aimed to determine the prevalence of ITCs in regional lymph nodes in patients with esophageal SCC and to investigate how frequently ITCs represent part of a true metastasis. Surgical specimens from 100 patients with SCC of the esophagus were included. All original H&E stained slides containing lymph nodes were reviewed by two gastrointestinal pathologists. In lymph nodes containing ITCs, additional levels were cut and stained with a H&E- and a cytokeratin stain. Areas of tumor cells that measured >0.2 mm on the deeper sections were classified as metastases. A total of 2460 lymph nodes were examined. ITCs were detected in 10 lymph nodes (0.4%) from nine patients (9%). Deeper sections revealed metastases in five out of the 10 lymph nodes (50%). ITCs in regional lymph nodes of patients with SCC of the esophagus is a rare finding compared with patients with adenocarcinoma of the esophagogastric junction. However, deeper sections often revealed metastases. Therefore, in patients with SCC of the esophagus, we recommend additional sectioning and immunohistochemical examination of lymph nodes when ITCs are detected on the first slide.

Impact of Deep Learning Assistance on the Histopathologic Review of Lymph Nodes for Metastatic Breast Cancer

Supplemental Digital Content is available in the text.

Advances in the quality of whole-slide images have set the stage for the clinical use of digital images in anatomic pathology. Along with advances in computer image analysis, this raises the possibility for computer-assisted diagnostics in pathology to improve histopathologic interpretation and clinical care. To evaluate the potential impact of digital assistance on interpretation of digitized slides, we conducted a multireader multicase study utilizing our deep learning algorithm for the detection of breast cancer metastasis in lymph nodes. Six pathologists reviewed 70 digitized slides from lymph node sections in 2 reader modes, unassisted and assisted, with a wash-out period between sessions. In the assisted mode, the deep learning algorithm was used to identify and outline regions with high likelihood of containing tumor. Algorithm-assisted pathologists demonstrated higher accuracy than either the algorithm or the pathologist alone. In particular, algorithm assistance significantly increased the sensitivity of detection for micrometastases (91% vs. 83%, P=0.02). In addition, average review time per image was significantly shorter with assistance than without assistance for both micrometastases (61 vs. 116 s, P=0.002) and negative images (111 vs. 137 s, P=0.018). Lastly, pathologists were asked to provide a numeric score regarding the difficulty of each image classification. On the basis of this score, pathologists considered the image review of micrometastases to be significantly easier when interpreted with assistance (P=0.0005). Utilizing a proof of concept assistant tool, this study demonstrates the potential of a deep learning algorithm to improve pathologist accuracy and efficiency in a digital pathology workflow.

Performance of Molecular Lymphosonography for Detection and Quantification of Metastatic Involvement in Sentinel Lymph Nodes

OBJECTIVES

To assess the performance of molecular lymphosonography with dual-targeted microbubbles in detecting and quantifying the metastatic involvement in sentinel lymph nodes (SLNs) using a swine melanoma model.

METHODS

Targeted microbubbles were labeled with P-selectin and αV β3 -integrin antibodies. Control microbubbles were labeled with immunoglobulin G antibodies. First lymphosonography with Sonazoid (GE Healthcare, Oslo, Norway) was used to identify SLNs. Then dual-targeted and control microbubbles were injected intravenously to detect and quantify metastatic disease in the SLNs. Distant non-SLNs were imaged as benign controls. All evaluated lymph nodes (LNs) were surgically removed, and metastatic involvement was characterized by a histopathologic analysis. Two radiologists blinded to histopathologic results assessed the baseline B-mode images of LNs, and the results were compared to the histologic reference standard. The mean intensities of targeted and control microbubbles within the examined LNs were measured and compared to the LN histologic results.

RESULTS

Thirty-five SLNs and 34 non-SLNs from 13 Sinclair swine were included in this study. Twenty-one SLNs (62%) were malignant, whereas 100% of non-SLNs were benign. The sensitivity of B-mode imaging for metastatic LN diagnosis for both readers was relatively high (90% and 71%), but the specificity was very poor (50% and 58%). The sensitivity and specificity of molecular lymphosonography for metastatic LN detection were 91% and 67%, respectively. The mean intensities from dual-targeted microbubbles correlated well with the degree of metastatic LN involvement (r = 0.6; P < 0.001).

CONCLUSIONS

Molecular lymphosonography can increase the specificity of metastatic LN detection and provide a measure to quantify the degree of metastatic involvement.

Application of Circulation Tumor Cells Detection in Diagnosis and Treatment of Breast Tumors

In recent years, the clinical application of circulating tumor cell (CTCs) detection has become one of the research hotspots in the field of precision medicine. CTCs detection is noninvasive, easy to obtain, can be repeatedly collected, and highly repeatable with other advantages. It not only can be a real-time comprehensive monitoring of cancer treatment but also can have a large number of applications, including early diagnosis of tumor, timely evaluation of efficacy, condition monitoring, resistance factor analysis, prognosis judgment, individualized treatment of tumors, drug guidance, and so on. At present, many large-scale clinical studies at home and abroad run through all stages of breast cancer diagnosis and treatment. For different treatment stages of breast cancer, the application value of CTCs detection is different. Compared with traditional detection methods, CTCs have advantages in dynamic monitoring of disease changes and efficacy evaluation in real-time. In the era of breast cancer classificational and individualized treatment, CTCs detection can provide patients with the most timely and optimized treatment plan.

Artificial Intelligence-Based Breast Cancer Nodal Metastasis Detection: Insights Into the Black Box for Pathologists

CONTEXT.—

Nodal metastasis of a primary tumor influences therapy decisions for a variety of cancers. Histologic identification of tumor cells in lymph nodes can be laborious and error-prone, especially for small tumor foci.

OBJECTIVE.—

To evaluate the application and clinical implementation of a state-of-the-art deep learning-based artificial intelligence algorithm (LYmph Node Assistant or LYNA) for detection of metastatic breast cancer in sentinel lymph node biopsies.

DESIGN.—

Whole slide images were obtained from hematoxylin-eosin-stained lymph nodes from 399 patients (publicly available Camelyon16 challenge dataset). LYNA was developed by using 270 slides and evaluated on the remaining 129 slides. We compared the findings to those obtained from an independent laboratory (108 slides from 20 patients/86 blocks) using a different scanner to measure reproducibility.

RESULTS.—

LYNA achieved a slide-level area under the receiver operating characteristic (AUC) of 99% and a tumor-level sensitivity of 91% at 1 false positive per patient on the Camelyon16 evaluation dataset. We also identified 2 "normal" slides that contained micrometastases. When applied to our second dataset, LYNA achieved an AUC of 99.6%. LYNA was not affected by common histology artifacts such as overfixation, poor staining, and air bubbles.

CONCLUSIONS.—

Artificial intelligence algorithms can exhaustively evaluate every tissue patch on a slide, achieving higher tumor-level sensitivity than, and comparable slide-level performance to, pathologists. These techniques may improve the pathologist's productivity and reduce the number of false negatives associated with morphologic detection of tumor cells. We provide a framework to aid practicing pathologists in assessing such algorithms for adoption into their workflow (akin to how a pathologist assesses immunohistochemistry results).