57% decline in Rhode Island invasive breast cancer mortality between 1987 and 2017: mammography predominates in preventing mortality

Evaluation of the peritumoral features using radiomics and deep learning technology in non-spiculated and noncalcified masses of the breast on mammography

OBJECTIVE

To assess the significance of peritumoral features based on deep learning in classifying non-spiculated and noncalcified masses (NSNCM) on mammography.

METHODS

We retrospectively screened the digital mammography data of 2254 patients who underwent surgery for breast lesions in Harbin Medical University Cancer Hospital from January to December 2018. Deep learning and radiomics models were constructed. The classification efficacy in ROI and patient levels of AUC, accuracy, sensitivity, and specificity were compared. Stratified analysis was conducted to analyze the influence of primary factors on the AUC of the deep learning model. The image filter and CAM were used to visualize the radiomics and depth features.

RESULTS

For 1298 included patients, 771 (59.4%) were benign, and 527 (40.6%) were malignant. The best model was the deep learning combined model (2 mm), in which the AUC was 0.884 (P < 0.05); especially the AUC of breast composition B reached 0.941. All the deep learning models were superior to the radiomics models (P < 0.05), and the class activation map (CAM) showed a high expression of signals around the tumor of the deep learning model. The deep learning model achieved higher AUC for large size, age >60 years, and breast composition type B (P < 0.05).

CONCLUSION

Combining the tumoral and peritumoral features resulted in better identification of malignant NSNCM on mammography, and the performance of the deep learning model exceeded the radiomics model. Age, tumor size, and the breast composition type are essential for diagnosis.

Prognostic Role of Tumor Budding in Breast Cancer Patients Receiving Neo-Adjuvant Therapy

Background: Isolated tumor cells or small clusters of tumor cells observed in the vicinity of the main tumor mass in pathology sections, termed tumor budding, are common in cancers and have been associated with prognosis in some settings. This study examined the clinical associations and treatment efficacy implications of tumor budding in breast cancer patients receiving neo-adjuvant therapy. Methods: Breast cancer patients that received neo-adjuvant therapy before definitive surgical treatment in a single cancer center over a 7-year period were included, and their records were reviewed. Data extracted included patient demographics, tumor characteristics and pathologic response to treatment at surgery. The initial breast cancer biopsy before any therapy was reviewed by two pathologists, and a hot spot area was evaluated for tumor budding (defined as 1 to 5 cancer cells observed detached from the main tumor mass). Results: Seventy-five patients who received neo-adjuvant therapy (73 received chemotherapy and 2 received hormonal therapy) were included. Tumor budding was observed in two-thirds of the patients. There were no significant differences in patient (age and menopause status) and tumor (stage, histology and molecular sub-type equivalent) characteristics between the group that had tumor budding and the group that did not have tumor budding in the pre-treatment biopsy. Likewise, no statistically significant differences were observed in the frequency of complete or partial responses between the two groups. Conclusion: In this cohort of breast cancer patients receiving neo-adjuvant therapy, tumor budding was frequent, but it was not associated with tumor characteristics or pathologic responses to treatment. The value of tumor budding as a prognostic factor in the neo-adjuvant setting within the general breast cancer population could not be confirmed, but such a value in specific sub-groups deserves further investigation, given the pathophysiologic rationale and data from other settings.