Intraoperative diagnosis of sentinel lymph node metastases in breast cancer treatment with one-step nucleic acid amplification assay (OSNA)

Accurate Evaluation of Feature Contributions for Sentinel Lymph Node Status Classification in Breast Cancer

The current guidelines recommend the sentinel lymph node biopsy to evaluate the lymph node involvement for breast cancer patients with clinically negative lymph nodes on clinical or radiological examination. Machine learning (ML) models have significantly improved the prediction of lymph nodes status based on clinical features, thus avoiding expensive, time-consuming and invasive procedures. However, the classification of sentinel lymph node status represents a typical example of an unbalanced classification problem. In this work, we developed a ML framework to explore the effects of unbalanced populations on the performance and stability of feature ranking for sentinel lymph node status classification in breast cancer. Our results indicate state-of-the-art AUC (Area under the Receiver Operating Characteristic curve) values on a hold-out set (67%) while providing particularly stable features related to tumor size, histological subtype and estrogen receptor expression, which should therefore be considered as potential biomarkers.

One-Step Nucleic Acid Amplification System in Comparison to the Intraoperative Frozen Section and Definitive Histological Examination Among Breast Cancer Patients: A Retrospective Survival Study

Simple Summary

Implementing intraoperative assessment of sentinel lymph nodes by one-step nucleic acid amplification in early breast cancer can reduce the surgical burden to the patient and the costs to the health system. However, only limited data are available in terms of long-term disease-free survival and overall survival. Therefore, this study aims to compare disease-free survival and overall survival between one-step nucleic acid amplification, frozen section, and definitive histology. These results could impact the healthcare community, adding further proof to the body of evidence supporting the broader adoption of this innovative technology that enables a safe reduction in patient surgical burden and healthcare costs.

Background

The one-step nucleic acid amplification (OSNA) system is a novel molecular technique, which consents to quick intraoperative detection of sentinel lymph node metastases by the amplification of cytokeratin 19 mRNA. Our study aims to evaluate the OSNA method in comparison with frozen section (FS) and definitive histological examination of the sentinel lymph node biopsy among early breast cancer patients considering disease-free survival (DFS) and overall survival (OS).

Methods

In this study, we included all women who underwent sentinel lymph node biopsy (SLNB) for breast cancers classified as TNM stage I and II in our center between January 2005 and January 2017, and the follow-up was collected up to January 2019. We divided patients among three groups based on SLNB evaluation: definitive histological examination, intra-operative FS, or OSNA.

Results

We included 2412 SLNBs: 727 by definitive histological examination, 697 by FS, and 988 by OSNA. Isolated tumor cells were found in 2.32% of cases, micrometastasis in 9.12%, and macrometastases in 13.64%. Surgical procedure duration was significantly shorter in OSNA than in FS (42.1 minutes ±5.1 vs. 70.1 minutes ±10.5, p <0.05). No significant differences have been observed among the three groups regarding OS, DSF, cumulative local, or distant metastases. In particular 5-year DFS was 96.38% in definitive histology (95% C.I. 95.02-97.75%), 96.37% in FS (95% C.I. 94.98-97.78%), and 96.51% in OSNA group (95% C.I. 95.32-97.72%).

Conclusions

No difference in OS and DFS was found comparing OSNA, FS, and definitive histology. Furthermore, reduced operative time was found in the OSNA group.

A ultrasound-based radiomic approach to predict the nodal status in clinically negative breast cancer patients

In breast cancer patients, an accurate detection of the axillary lymph node metastasis status is essential for reducing distant metastasis occurrence probabilities. In case of patients resulted negative at both clinical and instrumental examination, the nodal status is commonly evaluated performing the sentinel lymph-node biopsy, that is a time-consuming and expensive intraoperative procedure for the sentinel lymph-node (SLN) status assessment. The aim of this study was to predict the nodal status of 142 clinically negative breast cancer patients by means of both clinical and radiomic features extracted from primary breast tumor ultrasound images acquired at diagnosis. First, different regions of interest (ROIs) were segmented and a radiomic analysis was performed on each ROI. Then, clinical and radiomic features were evaluated separately developing two different machine learning models based on an SVM classifier. Finally, their predictive power was estimated jointly implementing a soft voting technique. The experimental results showed that the model obtained by combining clinical and radiomic features provided the best performances, achieving an AUC value of 88.6%, an accuracy of 82.1%, a sensitivity of 100% and a specificity of 78.2%. The proposed model represents a promising non-invasive procedure for the SLN status prediction in clinically negative patients.

Sentinel Lymph Node Metastasis on Clinically Negative Patients: Preliminary Results of a Machine Learning Model Based on Histopathological Features

The reported incidence of node metastasis at sentinel lymph node biopsy is generally low, so that the majority of women underwent unnecessary invasive axilla surgery. Although the sentinel lymph node biopsy is time consuming and expensive, it is still the intra-operative exam with the highest performance, but sometimes surgery is achieved without a clear diagnosis and also with possible serious complications. In this work, we developed a machine learning model to predict the sentinel lymph nodes positivity in clinically negative patients. Breast cancer clinical and immunohistochemical features of 907 patients characterized by a clinically negative lymph node status were collected. We trained different machine learning algorithms on the retrospective collected data and selected an optimal subset of features through a sequential forward procedure. We found comparable performances for different classification algorithms: on a hold-out training set, the logistics regression classifier with seven features, i.e., tumor diameter, age, histologic type, grading, multiplicity, in situ component and Her2-neu status reached an AUC value of 71.5% and showed a better trade-off between sensitivity and specificity (69.4 and 66.9%, respectively) compared to other two classifiers. On the hold-out test set, the performance dropped by five percentage points in terms of accuracy. Overall, the histological characteristics alone did not allow us to develop a support tool suitable for actual clinical application, but it showed the maximum informative power contained in the same for the resolution of the clinical problem. The proposed study represents a starting point for future development of predictive models to obtain the probability for lymph node metastases by using histopathological features combined with other features of a different nature.

A Proposal of Quantum-Inspired Machine Learning for Medical Purposes: An Application Case

Learning tasks are implemented via mappings of the sampled data set, including both the classical and the quantum framework. Biomedical data characterizing complex diseases such as cancer typically require an algorithmic support for clinical decisions, especially for early stage tumors that typify breast cancer patients, which are still controllable in a therapeutic and surgical way. Our case study consists of the prediction during the pre-operative stage of lymph node metastasis in breast cancer patients resulting in a negative diagnosis after clinical and radiological exams. The classifier adopted to establish a baseline is characterized by the result invariance for the order permutation of the input features, and it exploits stratifications in the training procedure. The quantum one mimics support vector machine mapping in a high-dimensional feature space, yielded by encoding into qubits, while being characterized by complexity. Feature selection is exploited to study the performances associated with a low number of features, thus implemented in a feasible time. Wide variations in sensitivity and specificity are observed in the selected optimal classifiers during cross-validations for both classification system types, with an easier detection of negative or positive cases depending on the choice between the two training schemes. Clinical practice is still far from being reached, even if the flexible structure of quantum-inspired classifier circuits guarantees further developments to rule interactions among features: this preliminary study is solely intended to provide an overview of the particular tree tensor network scheme in a simplified version adopting just product states, as well as to introduce typical machine learning procedures consisting of feature selection and classifier performance evaluation.

Predicting of Sentinel Lymph Node Status in Breast Cancer Patients with Clinically Negative Nodes: A Validation Study

Simple Summary

Sentinel lymph node biopsy procedure is time consuming and expensive, but it is still the intra-operative exam capable of the best performance. However, sometimes, surgery is achieved without a clear diagnosis, so clinical decision support systems developed with artificial intelligence techniques are essential to assist current diagnostic procedures. In this work, we evaluated the usefulness of a CancerMath tool in the sentinel lymph nodes positivity prediction for clinically negative patients. We tested it on 993 patients referred to our institute characterized by sentinel lymph node status, tumor size, age, histologic type, grading, expression of estrogen receptor, progesterone receptor, HER2, and Ki-67. By training the CancerMath (CM) model on our dataset, we reached a sensitivity value of 72%, whereas the online one was 46%, despite a specificity reduction. It was found the addiction of the prognostic factors Her2 and Ki67 could help improve performances on the classification of particular types of patients.

Abstract

In the absence of lymph node abnormalities detectable on clinical examination or imaging, the guidelines provide for the dissection of the first axillary draining lymph nodes during surgery. It is not always possible to arrive at surgery without diagnostic doubts, and machine learning algorithms can support clinical decisions. The web calculator CancerMath (CM) allows you to estimate the probability of having positive lymph nodes valued on the basis of tumor size, age, histologic type, grading, expression of estrogen receptor, and progesterone receptor. We collected 993 patients referred to our institute with clinically negative results characterized by sentinel lymph node status, prognostic factors defined by CM, and also human epidermal growth factor receptor 2 (HER2) and Ki-67. Area Under the Curve (AUC) values obtained by the online CM application were comparable with those obtained after training its algorithm on our database. Nevertheless, by training the CM model on our dataset and using the same feature, we reached a sensitivity median value of 72%, whereas the online one was equal to 46%, despite a specificity reduction. We found that the addition of the prognostic factors Her2 and Ki67 could help improve performances on the classification of particular types of patients with the aim of reducing as much as possible the false positives that lead to axillary dissection. As showed by our experimental results, it is not particularly suitable for use as a support instrument for the prediction of metastatic lymph nodes on clinically negative patients.

Comparing scraping cytology with touch imprint cytology and frozen section analysis in the intraoperative diagnosis of sentinel lymph node metastasis in breast cancer

BACKGROUND

Breast cancer is a common cancer in the female population. Sentinel lymph node (SLN) in breast cancer patients can be evaluated by different methods including intraoperative frozen section analysis (FSA), touch imprint cytology (TIC), and scraping cytology (SC). In this study, in addition to investigating TIC and FSA, we focused on SC to evaluate the diagnostic value of this almost new method.

MATERIALS AND METHODS

In this retrospective study, the quadrantectomy and sentinel lymph node resection of 150 specimens with mammography and core needle biopsy-confirmed breast cancer were examined. Of the 150 participants, 77 and 73 had negative and positive results for the permanent pathology of SLN metastasis, respectively. Intra-operative FSA, TIC, and SC for SLN were performed for the patients and all were confirmed by permanent pathology.

RESULTS

All the specimens were taken from females aged between 25 and 82 years. The sensitivity and specificity of TIC, FS, and SC were 73% and 50%, 92.6% and 50%, and 92.1% and 50%, respectively. Among the three techniques, TIC had the lowest positive and negative predictive values. However, FS had the highest positive predictive value whereas SC had the highest negative predictive value.

CONCLUSION

In addition to FS and TIC as rapid, cost-effective, and reliable diagnostic methods in SLN metastasis, SC is an acceptable and highly sensitive method. A combination of these methods may provide a more favorable diagnostic value for SLN assessment in breast cancer patients.

A novel approach for preventing recurrence of malign pleural effusion: early phase pleurodesis

INTRODUCTION

The effective control of malignant pleural effusion (MPE) is of paramount importance in the treatment of patients with disseminated cancer. In this study, we compared two different approaches (early pleurodesis versus late pleurodesis) to MPE.

MATERIAL AND METHODS

Patients (188 cases) whose primary tumor type was known and who were confirmed to have MPE, were included in the study and were separated into two groups. One group consisted of patients who were asymptomatic and who underwent early phase pleurodesis (group I, n = 79). The other group (group II, n = 109) was composed of patients who were symptomatic and whose pleurodesis was performed later. In all cases, pleural effusion was evaluated by means of direct radiography. Computed tomography was performed with the goal of confirming the parenchymal or mediastinal lesions accompanying the pleural fluid.

RESULTS

The rate of complete success in group I cases was observed to be higher, while the rate of recurrence was lower (p = 0.001 and p = 0.002, respectively) than group II. In multiple logistic regression analysis, co-morbidities and the group that patient belong were found to be significant in terms of pleurodesis success (p = 0.02 and p = 0.03). There was a significant difference in survival time between group I and group II, with group I exhibiting longer average survival time (log rank test p < 0.001).

CONCLUSIONS

We observed that the success rate was lower and the rate of recurrence higher in the late pleurodesis group, whose members already had greater volumes of pleural effusion.

One-Step Nucleic Acid Amplification (OSNA): A fast molecular test based on CK19 mRNA concentration for assessment of lymph-nodes metastases in early stage endometrial cancer

INTRODUCTION

The aim of the current study is to evaluate the detection rate of micro- and macro-metastases of the One-Step Nucleic Acid Amplification (OSNA) compared to frozen section examination and subsequent ultra-staging examination in early stage endometrial cancer (EC).

MATERIAL AND METHODS

From March 2016 to June 2016, data of 40 consecutive FIGO stage I EC patients were prospectively collected in an electronic database. The sentinel lymph node mapping was performed in all patients. All mapped nodes were removed and processed. Sentinel lymph nodes were sectioned and alternate sections were respectively examined by OSNA and by frozen section analysis. After frozen section, the residual tissue from each block was processed with step-level sections (each step at 200 micron) including H&E and IHC slides.

RESULTS

Sentinel lymph nodes mapping was successful in 29 patients (72.5%). In the remaining 11 patients (27.5%), a systematic pelvic lymphadenectomy was performed. OSNA assay sensitivity and specificity were 87.5% and 100% respectively. Positive and negative predictive values were 100% and 99% respectively, with a diagnostic accuracy of 99%. As far as frozen section examination and subsequent ultra-staging analysis was concerned, we reported sensitivity and specificity of 50% and 94.4% respectively; positive and negative predictive values were 14.3% and 99%, respectively, with an accuracy of 93.6%. In one patient, despite negative OSNA and frozen section analysis of the sentinel node, a macro-metastasis in 1 non-sentinel node was found.

CONCLUSIONS

The combination of OSNA procedure with the sentinel lymph node mapping could represent an efficient intra-operative tool for the selection of early-stage EC patients to be submitted to systematic lymphadenectomy.

Comparison of CK-IHC assay on serial frozen sections, the OSNA assay, and in combination for intraoperative evaluation of SLN metastases in breast cancer

BACKGROUND

Intraoperative evaluations of sentinel lymph node (SLN) metastases are performed for providing appropriate and immediate axillary treatments in breast cancer patients who do not meet Z0011 criteria; however, standard intraoperative procedure has not yet been established.

METHODS

We consecutively performed intraoperative evaluation for SLN metastases using both a cytokeratin immunohistochemistry (CK-IHC) assay on serial frozen sections and a one-step nucleic acid amplification (OSNA) assay of the remaining whole node in patients with invasive breast cancer. In this article, we compared the intraoperative diagnostic ability of CK-IHC assay, the OSNA assay, and in combination.

RESULTS

Between August 2009 and May 2017, 1,103 SLNs from 499 consecutive clinically node-negative invasive breast cancers were intraoperatively evaluated for SLN metastases using an OSNA and CK-IHC assay. The detection rates of SLN metastases by the OSNA and CK-IHC assays and in combination were 11.8, 12.1, and 14.5%, respectively. The concordance rate between the intraoperative SLN findings of the OSNA and CK-IHC assays was 94.9% (95% confidence interval 93.6-96.2%). The false negative rate for the OSNA assay was 3.1% (30/973), including 3 (0.3%) macrometastases and 27 (2.8%) micrometastases, and for the CK-IHC assay was 2.7% (26/969), including 1 (0.1%) OSNA ++ and 25 (2.6%) OSNA +.

CONCLUSIONS

The CK-IHC assay and the OSNA assay have compatible diagnostic abilities in intraoperative evaluations for SLN metastases. The low incidence of false negative results with limited disease burden suggests that both assays can be reliable techniques for intraoperative diagnoses of SLN metastases in breast cancer patients.

Imprint cytology versus frozen section analysis for intraoperative assessment of sentinel lymph node in breast cancer

Introduction

Sentinel lymph node (SLN) biopsy is the gold standard for surgical staging of the axilla in breast cancer (BC). Frozen section (FS) remains the most popular means of intraoperative SLN diagnosis. Imprint cytology (IC) has also been suggested as a less expensive and equally accurate alternative to FS. The aim of our study was to perform a direct comparison between IC and FS on the same SLNs of BC cases operated in a single center by the same surgical team.

Materials and methods

Into this prospective study we enrolled 60 consecutive patients with histologically proven T1–T3 BC and clinically negative axilla. Sentinel nodes were detected using a standard protocol. The SLN(s) was always assessed by IC as well as FS analysis and immunohistochemistry. Nevertheless, all intraoperative decisions were based on FS analysis.

Results

During the study period 60 patients with invasive BC were registered, with 80 SLNs harvested. Mean number of SLN(s) identified for each patient was 1.33. The sensitivity and specificity were 90% and 100%, respectively, for IC, and 80% and 100% for FS. Relevant positive/negative predictive values were 100%/98% for IC and 100%/96.15%, respectively, for FS. Overall accuracy was 98% for IC and 97% for FS. Therefore, statistically significant difference between the two methods in the detection of positive nodes was not elucidated (p=1.000).

Conclusions

IC appeared to be marginally more sensitive than FS in detecting SLN metastatic activity. Overall accuracy was 98.75%. With regard to the primary lesion characteristics, we conclude that initial lesion size and lymphovascular invasion play a pivotal role in metastatic involvement of the SLN with the dimensions of metastasis bearing no correlation with tumor size. Therefore, IC appears to be a sensitive and accurate method for the intraoperative assessment of SLN in BC patients, but further studies are required to confirm this interesting data.