New Breakthrough in Predicting Axillary Lymph Node Metastasis in Early-Onset Breast Cancer
Breast cancer is the most common malignancy affecting women worldwide, with a notable rise in cases diagnosed before the age of 50. These early-onset cases often display more aggressive behaviors and less favorable outcomes compared to later-onset breast cancer, emphasizing the urgent need for improved diagnostic and prognostic tools tailored to younger women.
The Challenge of Predicting Axillary Lymph Node Metastasis in Young Women
Despite extensive research into breast cancer, predicting axillary lymph node metastasis (ALNM) in early-onset breast cancer remains unresolved. Most studies focus on general breast cancer populations, overlooking the unique biological characteristics and clinical outcomes of young patients. This subgroup’s higher rates of ALNM, potentially due to more aggressive tumor phenotypes, require dedicated studies to understand and predict metastasis effectively.
Developing a Clinical-Radiomics Nomogram
A recent study aimed to fill this gap by creating a clinical-radiomics nomogram that combines DCE-MRI-derived radiomic features with traditional clinical indicators to predict ALNM in early-onset breast cancer patients. This research integrates advanced imaging techniques with robust clinical parameters to establish a predictive model that enhances clinical decision-making and personalizes treatment strategies for young patients.
Study Methodology
The study involved female patients diagnosed with early-onset breast cancer (before age 50) at Shanxi Bethune Hospital between March 2020 and February 2024. It used a retrospective design with anonymized data, avoiding the need for ethics approval. Participants included newly diagnosed patients with complete clinical records and high-quality DCE-MRI scans.
The cohort was divided into a training set (198 patients) and a validation set (99 patients) to develop and validate the predictive models. Data included demographic information, detailed clinical profiles, and radiological imaging.
DCE-MRI Image Acquisition
DCE-MRI scans were conducted using a 3.0 Tesla MRI system equipped with a dedicated breast coil. The protocol involved axial T1-weighted and T2-weighted sequences, along with dynamic sequences post-injection of a gadolinium-based contrast agent. Image quality was rigorously controlled by two experienced radiologists.
Image Segmentation and Feature Extraction
Nodule segmentation was performed using ITK-SNAP software, and radiomics feature extraction was done using 3D-Slicer software. This process produced 912 radiomics features, including first-order statistics, gray level dependence matrix (GLDM), gray level co-occurrence matrix (GLCM), gray level run length matrix (GLRLM), and gray level size zone matrix (GLSZM).
Feature Selection and Radiomics Score Construction
The study employed a systematic approach, filtering features with low interclass correlation coefficients (ICC ≤ 0.75). The minimum redundancy maximum relevance (mRMR) technique and Least Absolute Shrinkage and Selection Operator (LASSO) method were used to select the most predictive radiomic features. A radiomics score was developed using four key features.
Development of the Nomogram
A clinical model incorporating demographic and tumor characteristics was developed using multivariable logistic regression. This was integrated with radiomics features to create a comprehensive clinical-radiomics nomogram. The nomogram’s discriminative capacity was assessed using the area under the receiver operating characteristic (ROC) curve (AUC).
Finding and Validation
The final nomogram was validated in an independent dataset, confirming its predictive performance and utility. The model demonstrated high predictive accuracy, as evidenced by substantial AUC values and high testing accuracy in the validation set.
Patient Characteristics
Table 1 compares patient characteristics between the training set and the validation set, showing no significant differences in key parameters like menopause status, tumor location, histologic grade, multifocality, MRI-reported axillary lymph node status, and receptor status (ER, PR, HER2).
| Characteristics | Training Set (n (198) | Validation (99) | p-value |
|---|---|---|---|
| Menopausal Status | … | … | … |
| Tumor Location | … | … | … |
| Histologic Grade | … | … | … |
| Multifocality | … | … | … |
| Molecular Subtypes | … | … | … |
Radiomics Score Building
After filtering and refining features using mRMR and LASSO regression analyses, four essential radiomics features were identified. The LASSO regression analysis, as shown in Figure 2, identified seven pivotal features initially.
Model Building and Validation
The clinical model included multifocality, molecular subtypes, tumor size, and MRI-reported axillary lymph node status as significant predictors. In the clinical-radiomics model, the inclusion of a radiomics score significantly enhanced predictive power.
| Predictor | Clinical Model (OR, p-value) | Clinical-Radiomics Model (OR, p-value) |
|---|---|---|
| Multifocality | … | … |
| Molecular Subtypes | … | … |
| Tumor Size | … | … |
| MRI-reported ALN Status | 3.17, 0.003 | 6.12, 0.024 |
| Radiomics Score | – | 2.63, 0.009 |
Model Performance
The clinical-radiomics model showed superior performance compared to the clinical model in both training and validation datasets. The ROC curves in Figure 3 demonstrate that the clinical-radiomics model significantly outperforms the clinical model, with AUC values of 0.892 and 0.877 in the training and validation sets, respectively.
The Clinical-Radiomics Nomogram
Figure 4 illustrates the clinical-radiomics nomogram developed for predicting ALNM in young-onset breast cancer patients. The nomogram integrates clinical and radiomic data, providing a predictive tool. Calibration curves for both the training and validation sets indicate that the nomogram is well-calibrated, matching observed outcomes closely.
Decision Curve Analysis
Figure 5 showcases a decision curve analysis, confirming that the clinical-radiomics nomogram delivers higher net benefits across various threshold probabilities compared to standard models.
Discussion
Predicting ALNM in young-onset breast cancer is crucial for guiding therapeutic strategies and improving outcomes in this aggressive subgroup. Enhancing the accuracy of ALNM prediction is vital for optimizing treatment plans and improving survival rates.
The study’s clinical-radiomics nomogram enhances ALNM prediction by integrating DCE-MRI-derived radiomic features with traditional clinical indicators. The effectiveness of this model underscores the benefits of incorporating radiomic analysis, which extracts a vast array of quantitative data from imaging to provide a more nuanced understanding of tumor behavior.
Each predictor included in the nomogram—such as multifocality, molecular subtypes, tumor size, and MRI-reported ALN status—has distinct clinical significance. The role of radiomics in enhancing ALNM prediction has been increasingly documented, with radiomic features revealing subtle patterns in tissue architecture not detectable by the naked eye.
Compared to existing methods like ultrasound-based prediction models, DCE-MRI offers superior soft tissue contrast and spatial resolution, making it advantageous for detecting subtle changes in axillary lymph nodes. Incorporating DCE-MRI-derived radiomic features in predictive models provides a more standardized and comprehensive approach.
Study Limitations
Despite these advancements, the study has limitations. The lack of direct pathological correlation limits the analysis, and the retrospective design and single-center constraint reduce the model’s broader applicability. Future research should focus on multicentric and prospective validations and the integration of genomic data to enhance predictive power.
Conclusion
The clinical-radiomics nomogram represents a significant step forward in personalizing the management of young-onset breast cancer. It provides a scientifically validated, non-invasive tool that enhances the prediction of ALNM, enabling more precise and tailored treatment strategies. This approach not only improves clinical decision-making but also holds the potential to significantly impact patient outcomes by aligning treatment with individual risk profiles.
By leveraging advanced imaging and radiomic techniques, this research opens new possibilities for precision medicine in oncology. The nomogram’s ability to predict lymph node metastasis more accurately can lead to better prognosis, more informed treatment decisions, and ultimately better outcomes for young women with breast cancer.
