A new and simple predictive formula for non-sentinel lymph node metastasis in breast cancer patients with positive sentinel lymph nodes, and validation of 3 different nomograms in Turkish breast cancer patients

Deep learning radiomics based prediction of axillary lymph node metastasis in breast cancer

This study aimed to develop and validate a deep learning radiomics nomogram (DLRN) for the preoperative evaluation of axillary lymph node (ALN) metastasis status in patients with a newly diagnosed unifocal breast cancer. A total of 883 eligible patients with breast cancer who underwent preoperative breast and axillary ultrasound were retrospectively enrolled between April 1, 2016, and June 30, 2022. The training cohort comprised 621 patients from Hospital I; the external validation cohorts comprised 112, 87, and 63 patients from Hospitals II, III, and IV, respectively. A DLR signature was created based on the deep learning and handcrafted features, and the DLRN was then developed based on the signature and four independent clinical parameters. The DLRN exhibited good performance, yielding areas under the receiver operating characteristic curve (AUC) of 0.914, 0.929, and 0.952 in the three external validation cohorts, respectively. Decision curve and calibration curve analyses demonstrated the favorable clinical value and calibration of the nomogram. In addition, the DLRN outperformed five experienced radiologists in all cohorts. This has the potential to guide appropriate management of the axilla in patients with breast cancer, including avoiding overtreatment.

Preoperative ultrasound-based radiomics score can improve the accuracy of the Memorial Sloan Kettering Cancer Center nomogram for predicting sentinel lymph node metastasis in breast cancer

PURPOSE

To develop a combined nomogram by incorporating the Memorial Sloan Kettering Cancer Center (MSKCC) nomogram and ultrasound (US)-based radiomics score (Radscore) for predicting sentinel lymph node (SLN) metastasis in invasive breast cancer.

MATERIALS AND METHODS

This retrospective study was approved by the ethics committee of our institution, and written informed consent was waived. A total of 452 patients with invasive breast cancer who received SLN Biopsy in a single center were included between January 2016 and December 2019. The patients were divided into a training set (n = 318) and a validation set (n = 134). A total of 1216 features were extracted from the regions of interest (ROIs) of the tumors on conventional ultrasound. The maximum relevance minimum redundancy (mRMR) and the least absolute shrinkage and selection operator (LASSO) algorithm were used to build the Radscore. Afterward, the diagnostic performance was assessed and validated. Comparison of receiver operating characteristic (ROC) curves and decision curve analysis (DCA) were performed to evaluate the incremental value of the combined model.

RESULTS

Obtained from 18 features, the Radscore indicated a favorable discriminatory capability in the training set with an area under the curve (AUC) of 0.834, whereas a value of 0.770 was observed in the validation set. The AUC of the combined model was 0.901 (95 % confidence interval (95 % CI): 0.865-0.938) in the training set and 0.833 (95 % CI: 0.788-0.878) in the validation set. Both of them were superior to MSKCC or imaging Radscore alone (P < 0.05). DCA demonstrated that the combined model was superior to the others in terms of clinical practicability.

CONCLUSION

Preoperative US-based Radscore can improve the accuracy of clinical MSKCC nomogram for SLN metastasis prediction in breast cancer.

Management of the axilla in breast cancer: outcome analysis in a series of ductal versus lobular invasive cancers

INTRODUCTION

Axillary lymph node dissection (ALND) has been considered essential for the staging of breast cancer (BC). As the impact of tumor biology on clinical outcomes is recognized, a surgical de-escalation approach is being implemented. We performed a retrospective study focused on surgical management of the axilla in invasive lobular carcinoma (ILC) versus invasive ductal carcinoma (IDC).

MATERIALS AND METHODS

1151 newly diagnosed BCs, IDCs (79.6%) or ILCs (20.4%), were selected among patients treated at our Breast Cancer Unit from 2012 to 2018. Tumor characteristics and clinical information were collected and predictors of further metastasis after positive sentinel lymph node biopsy (SLNB) analyzed in relation to disease-free survival (DFS) and overall survival (OS).

RESULTS

27.5% of patients with ILC had ≥ 3 metastatic lymph nodes at ALND after positive SLNB versus 11.48% of IDCs (p = 0.04). Risk predictors of further metastasis at ALND were the presence of > 2 positive lymph nodes at SLNB (OR = 4.72, 95% CI 1.15-19.5 p = 0.03), T3-T4 tumors (OR = 4.93, 95% CI 1.10-22.2, p = 0.03) and Non-Luminal BC (OR = 2.74, 95% CI 1.16-6.50, p = 0.02). The lobular histotype was not associated with the risk of further metastasis at ALND (OR = 1.62, 95% CI 0.77-3.41, p = 0.20).

CONCLUSIONS

ILC histology is not associated with higher risk of further metastasis at ALND in our analysis. However, surgical management decisions should be taken considering tumor histotype, biology and expected sensitivity to adjuvant therapies.

Can Nomograms Predict Preoperative Axillary Lymph Node Metastasis in Patients With Breast Cancer to Guide Second Look Ultrasonography?

OBJECTIVES

The low sensitivity of ultrasonography (US) for diagnosing axillary lymph node metastasis in patients with breast cancer has led to the development of multiple tools in an attempt to increase preoperative sensitivity, including second-look US. We compared axillary lymph node metastasis predictor scores with postsurgical findings, using the Memorial Sloan Kettering Cancer Center (MSKCC; New York, NY) and Evidencio (www.evidencio.com) nomograms: 2 freely available online predictor tools.

METHODS

We retrospectively evaluated 450 patients with breast cancer and analyzed data from 194 patients. Sonograms were evaluated to measure lymph node cortical thickness, transverse diameter, and hilum status. Patients were divided into 3 groups: namely 0, 1, and 2 based on the number of postoperative positive nodes (0, 1 and ≥2, respectively). One-way analysis of variance was used to analyze the differences in mean scores across the 3 nodal groups for both nomograms. P < .05 was considered statistically significant.

RESULTS

There were significant differences in mean scores across the 3 nodal groups when using MSKCC (P < .001) as well as Evidencio (P < .001). However, there was an overlap of scores across the 3 groups; thus, mutually exclusive values were not obtained. A strong positive correlation was found between MSKCC and Evidencio (P < 0.001). Tumor size and the presence of lymphovascular invasion were significantly associated with axillary nodal disease (P < .001; P = .003, respectively).

CONCLUSIONS

The use of nomograms to predict axillary nodal involvement in patients with breast cancer can guide discussions, but in their present state, these scores cannot guide clinical decisions or direct second-look US of axilla.

Preoperative prediction of sentinel lymph node metastasis in breast cancer based on radiomics of T2-weighted fat-suppression and diffusion-weighted MRI

OBJECTIVES

To predict sentinel lymph node (SLN) metastasis in breast cancer patients using radiomics based on T₂-weighted fat suppression (T₂-FS) and diffusion-weighted MRI (DWI).

METHODS

We enrolled 146 patients with histologically proven breast cancer. All underwent pretreatment T₂-FS and DWI MRI scan. In all, 10,962 texture and four non-texture features were extracted for each patient. The 0.623 + bootstrap method and the area under the curve (AUC) were used to select the features. We constructed ten logistic regression models (orders of 1-10) based on different combination of image features using stepwise forward method.

RESULTS

For T₂-FS, model 10 with ten features yielded the highest AUC of 0.847 in the training set and 0.770 in the validation set. For DWI, model 8 with eight features reached the highest AUC of 0.847 in the training set and 0.787 in the validation set. For joint T₂-FS and DWI, model 10 with ten features yielded an AUC of 0.863 in the training set and 0.805 in the validation set.

CONCLUSIONS

Full utilisation of breast cancer-specific textural features extracted from anatomical and functional MRI images improves the performance of radiomics in predicting SLN metastasis, providing a non-invasive approach in clinical practice.

KEY POINTS

• SLN biopsy to access breast cancer metastasis has multiple complications. • Radiomics uses features extracted from medical images to characterise intratumour heterogeneity. • We combined T ₂ -FS and DWI textural features to predict SLN metastasis non-invasively.

Factors Influencing Non-sentinel Node Metastasis in Patients with Macrometastatic Sentinel Lymph Node Involvement and Validation of Three Commonly Used Nomograms

OBJECTIVE

Omitting axillary lymph node dissection (ALND) in a subgroup of patients with sentinel lymph node (SLN) metastasis is becoming a widely accepted practice. Avoiding the well-known complications of ALND is the sole aim without compromising the curative intention of surgery.

MATERIALS AND METHODS

The data were probed for breast cancer patients that were operated on between February 2014 and June 2016. SLN biopsies were performed in 507 patients and out of 157 patients who underwent ALND for a metastatic SLN, 151 were found eligible for the analyses as having macrometastatic (>2mm) SLN. MD Anderson, Memorial Sloan Kettering Cancer Center and Helsinki nomograms were also tested in our patient population.

RESULTS

Pathologic tumor size greater than 2 cm, the ratio of metastatic SLN to dissected SLN, metastatic tumor greater than 1 cm and tumors that extended outside the SLN's capsule were found to be associated with non-sentinel node metastasis in both univariate and multivariate tests. MD Anderson nomogram performed well with an area under the curve (AUC) value of 0.72.

CONCLUSION

Our results suggest that ALND should be considered in patients with macrometastatic SLN greater than 10 mm in size, have extracapsular extension, have metastatic SLNs at a rate of more than 50% and whose primary tumor is greater than 2 cm.

Prediction of nonsentinel lymph node metastasis in breast cancer patients with one or two positive sentinel lymph nodes

OBJECTIVE

The aim of the present study was to investigate the association between non sentinel lymph node metastasis (NSLNM) and clinicopathological factors, particularly in the case of sentinel lymph node (SLN) metastasis in one or two, in clinically node negative patients with breast cancer.

METHODS

Between 10/2010 and 10/2014, 350 sentinel lymph node biopsy (SLNB) were performed in patients with histologically proven primary breast cancer in our clinic. The data collection includes the following characteristics: age, pathological tumor size, histological type, histological grade, lymphovascular invasion (LVI), number of positive SLN, size of the SLN metastasis (macrometastasis, micrometastasis, isolated tumor cells), multifocality (MF), extracapsuler invasion (ECI) of the SLN, the estrogen receptor (ER) status, the progesterone receptor (PR) status and the Her 2 receptor status, Ki 67 reseptor status. Data were collected retrospectively and then analyzed.

RESULTS

A successful SLN biopsy were performed in 345 (98.5%) cases. SLN metastases were detected in 110 (31.8%) cases. These patients then underwent axillary dissection; among these patients, 101 (91.8%) had only one to two positive SLNs. Of the 101 patients with positive SLN biopsies, 32 (31.6%) had metastases in the NSLNs. Univariate and multivariate analysis showed that lymphovascular invasion, extracapsular invasion (ECI), Her-2 receptor positive, and Ki-67 > 14% were related to NSLNM (p<.0.05).

CONCLUSION

The predicting factors of NSLNM were LVI, ECI, Ki-67 level, Her-2 reseptor positive and but should be further validated in our institutions, different institutions and different patient groups prospectively.