Sacituzumab govitecan vs. chemotherapy for metastatic breast cancer: a meta-analysis on safety outcomes.

Aim: There is limited data available regarding the comparison of Sacituzumab govitecan (SG) vs. chemotherapy in metastatic breast cancer patients. Materials & methods: We performed a systematic review and meta-analysis aimed to assess the safety profile of SG vs. chemotherapy for metastatic breast cancer (mBC) clinical trials. Results: The pooled odds ratio for outcomes such as grade 3-4 and all grade neutropenia, leukopenia, anemia and other non-hematological adverse events showed a higher risk for patients receiving SG. No statistically significant differences were reported in terms of grade 3-4 fatigue, all grade nausea, febrile neutropenia and treatment discontinuation due to adverse events. Conclusion: Our data, coupled with a statistically and clinically meaningful survival benefit, support the use of SG for mBC.

[Box: see text].

Eribulin Mesylate as Third or Subsequent Line Chemotherapy for Elderly Patients with Locally Recurrent or Metastatic Breast Cancer: A Multicentric Observational Study of GIOGer (Italian Group of Geriatric Oncology)-ERIBE.

BACKGROUND: Metastatic breast cancer (MBC) is highly prevalent in middle-aged or elderly patients. Eribulin is a nontaxane microtubule inhibitor, approved for the treatment of pretreated MBC. This multicentric study (sponsored by GIOGer, Italian Group for Geriatric Oncology) was designed to assess the efficacy and tolerability of eribulin, according to parameters usually used in geriatric oncology. SUBJECTS, MATERIALS, AND METHODS: An observational study was conducted on 50 consecutive elderly patients with MBC. The primary endpoint was to evaluate the change in items score of comprehensive geriatric assessment (CGA) and health-related quality of life (HRQL). Italian versions of the CGA and HRQL questionnaires were administered at baseline, before the third and fifth cycles, and then every three cycles until treatment discontinuation. Secondary endpoints were efficacy and safety. RESULTS: Overall, both EQ-5D scores and EQ-5D-3 L visual analogic scale did not significantly change from baseline; the percentage of subjects without problems doing usual activities tended to decrease during treatment (p for linear trend .018), and the percentage of patients with minor problems performing usual activities tended to increase (p for linear trend.012). Among CGA items, Instrumental Activities of Daily Living tended to decrease during treatment and Geriatric Depression Scale tended to increase. After 12 months follow-up, 24 patients (out of 47) showed clinical benefits; median progression-free survival was 4.49 months (2.10-10.33) and median OS was 7.31 months (3.70-14.03). The treatment was associated with mild toxicity. CONCLUSION: Eribulin treatment preserved quality of life and geriatric parameters included in the CGA, except for instrumental functioning and geriatric depression, in elderly patients with MBC. IMPLICATIONS FOR PRACTICE: A collaboration between oncologist and geriatric specialists is essential in the management of patients with metastatic breast cancer, who are frequently elderly or frail. The assessment of geriatric parameters in the decision-making process can contribute to direct toward the most appropriate therapeutic plan and preserve the quality of life of patients. Eribulin does not seem to affect quality of life or worsen the overall geriatric status; therefore, it can be considered a suitable option for elderly patients with metastatic breast cancer.

Efficacy and safety of lenvatinib in an elderly patient with metastatic papillary thyroid carcinoma and cardiological comorbidity: a case report.

Lenvatinib significantly prolonged progression-free survival versus placebo in patients with radio-iodine refractory differentiated thyroid carcinoma. However, the primary adverse effects of any grade that occurred in >40% of patients in the lenvatinib group of the Phase III SELECT trial was hypertension (67.8%). Therefore, this drug should be used with caution in patients with cardiological morbidity. Here, we describe the case of a 73-year-old man with hypertension, obesity and chronic atrial fibrillation, who received lenvatinib for 6 months in the absence of cardiological symptoms.

Dose intensity and efficacy of the combination of everolimus and exemestane (EVE/EXE) in a real-world population of hormone receptor-positive (ER+/PgR+), HER2-negative advanced breast cancer (ABC) patients: a multicenter Italian experience.

AIM: This retrospective analysis focused on the effect of treatment with EVE/EXE in a real-world population outside of clinical trials. We examined the efficacy of this combination in terms of PFS and RR related to dose intensity (5 mg daily versus 10 mg daily) and tolerability. METHODS: 163 HER2-negative ER+/PgR+ ABC patients, treated with EVE/EXE from May 2011 to March 2016, were included in the analysis. The primary endpoints were the correlation between the daily dose and RR and PFS, as well as an evaluation of the tolerability of the combination. Secondary endpoints were RR, PFS, and OS according to the line of treatment. Patients were classified into three different groups, each with a different dose intensity of everolimus (A, B, C). RESULTS: RR was 29.8% (A), 27.8% (B) (p = 0.953), and not evaluable (C). PFS was 9 months (95% CI 7-11) (A), 10 months (95% CI 9-11) (B), and 5 months (95% CI 2-8) (C), p = 0.956. OS was 38 months (95% CI 24-38) (A), median not reached (B), and 13 months (95% CI 10-25) (C), p = 0.002. Adverse events were stomatitis 57.7% (11.0% grade 3-4), asthenia 46.0% (6.1% grade 3-4), hypercholesterolemia 46.0% (0.6% grade 3-4), and hyperglycemia 35.6% (5.5% grade 3-4). The main reason for discontinuation/interruption was grade 2-3 stomatitis. CONCLUSIONS: No correlation was found between dose intensity (5 vs. 10 mg labeled dose) and efficacy in terms of RR and PFS. The tolerability of the higher dose was poor in our experience, although this had no impact on efficacy.

Efficacy and safety of eribulin in taxane-refractory patients in the 'real world'.

AIM: Recent clinical, randomized and observational studies showed that eribulin, an analogous of Halichondrin B, was beneficial and well-tolerated in heavily pretreated metastatic breast cancer patients. Here, we aim to evaluate the effectiveness and safety of eribulin in taxane-refractory metastatic breast cancer patients. PATIENTS & METHODS: In this subanalysis of the ESEMPIO study database, we selected 91 subjects with well-defined taxane refractoriness and complete data available. RESULTS: 41 patients (45.2%) showed clinical benefit; one complete response (2.2%) and 16 partial responses (17.6%) were observed. Median progression-free survival and median overall survival were 3.1 and 11.6 months, respectively. The most experienced adverse event was asthenia/fatigue (58%), followed by neutropenia (30%). The treatment-related toxicity led to eribulin-dose reduction in 19 patients and suspension in nine. CONCLUSION: This study shows that eribulin is effective and well tolerated also in taxane-refractory patients in clinical practice.

Explainable 3D CNN based on baseline breast DCE-MRI to give an early prediction of pathological complete response to neoadjuvant chemotherapy.

BACKGROUND: So far, baseline Dynamic Contrast-Enhanced Magnetic Resonance Imaging (DCE-MRI) has played a key role for the application of sophisticated artificial intelligence-based models using Convolutional Neural Networks (CNNs) to extract quantitative imaging information as earlier indicators of pathological Complete Response (pCR) achievement in breast cancer patients treated with neoadjuvant chemotherapy (NAC). However, these models did not exploit the DCE-MRI exams in their full geometry as 3D volume but analysed only few individual slices independently, thus neglecting the depth information. METHOD: This study aimed to develop an explainable 3D CNN, which fulfilled the task of pCR prediction before the beginning of NAC, by leveraging the 3D information of post-contrast baseline breast DCE-MRI exams. Specifically, for each patient, the network took in input a 3D sequence containing the tumor region, which was previously automatically identified along the DCE-MRI exam. A visual explanation of the decision-making process of the network was also provided. RESULTS: To the best of our knowledge, our proposal is competitive than other models in the field, which made use of imaging data alone, reaching a median AUC value of 81.8%, 95%CI [75.3%; 88.3%], a median accuracy value of 78.7%, 95%CI [74.8%; 82.5%], a median sensitivity value of 69.8%, 95%CI [59.6%; 79.9%] and a median specificity value of 83.3%, 95%CI [82.6%; 84.0%], respectively. The median and CIs were computed according to a 10-fold cross-validation scheme for 5 rounds. CONCLUSION: Finally, this proposal holds high potential to support clinicians on non-invasively early pursuing or changing patient-centric NAC pathways.

Machine learning survival models trained on clinical data to identify high risk patients with hormone responsive HER2 negative breast cancer.

For endocrine-positive Her2 negative breast cancer patients at an early stage, the benefit of adding chemotherapy to adjuvant endocrine therapy is not still confirmed. Several genomic tests are available on the market but are very expensive. Therefore, there is the urgent need to explore novel reliable and less expensive prognostic tools in this setting. In this paper, we shown a machine learning survival model to estimate Invasive Disease-Free Events trained on clinical and histological data commonly collected in clinical practice. We collected clinical and cytohistological outcomes of 145 patients referred to Istituto Tumori "Giovanni Paolo II". Three machine learning survival models are compared with the Cox proportional hazards regression according to time-dependent performance metrics evaluated in cross-validation. The c-index at 10 years obtained by random survival forest, gradient boosting, and component-wise gradient boosting is stabled with or without feature selection at approximately 0.68 in average respect to 0.57 obtained to Cox model. Moreover, machine learning survival models have accurately discriminated low- and high-risk patients, and so a large group which can be spared additional chemotherapy to hormone therapy. The preliminary results obtained by including only clinical determinants are encouraging. The integrated use of data already collected in clinical practice for routine diagnostic investigations, if properly analyzed, can reduce time and costs of the genomic tests.

Prognostic power assessment of clinical parameters to predict neoadjuvant response therapy in HER2-positive breast cancer patients: A machine learning approach.

BACKGROUND: About 15%-20% of breast cancer (BC) cases is classified as Human Epidermal growth factor Receptor type 2 (HER2) positive. The Neoadjuvant chemotherapy (NAC) was initially introduced for locally advanced and inflammatory BC patients to allow a less extensive surgical resection, whereas now it represents the current standard for early-stage and operable BC. However, only 20%-40% of patients achieve pathologic complete response (pCR). According to the results of practice-changing clinical trials, the addition of trastuzumab to NAC brings improvements to pCR, and recently, the use of pertuzumab plus trastuzumab has registered further statistically significant and clinically meaningful improvements in terms of pCR. The goal of our work is to propose a machine learning model to predict the pCR to NAC in HER2-positive patients based on a subset of clinical features. METHOD: First, we evaluated the significant association of clinical features with pCR on the retrospectively collected data referred to 67 patients afferent to Istituto Tumori "Giovanni Paolo II." Then, we performed a feature selection procedure to identify a subset of features to be used for training a machine learning-based classification algorithm. As a result, pCR to NAC was associated with ER status, Pgr status, and HER2 score. RESULTS: The machine learning model trained on a subgroup of essential features reached an AUC of 73.27% (72.44%-73.66%) and an accuracy of 71.67% (71.64%-73.13%). According to our results, the clinical features alone are not enough to define a support system useful for clinical pathway. CONCLUSION: Our results seem worthy of further investigation in large validation studies and this work could be the basis of future study that will also involve radiomics analysis of biomedical images.

Analyzing breast cancer invasive disease event classification through explainable artificial intelligence.

Introduction: Recently, accurate machine learning and deep learning approaches have been dedicated to the investigation of breast cancer invasive disease events (IDEs), such as recurrence, contralateral and second cancers. However, such approaches are poorly interpretable. Methods: Thus, we designed an Explainable Artificial Intelligence (XAI) framework to investigate IDEs within a cohort of 486 breast cancer patients enrolled at IRCCS Istituto Tumori "Giovanni Paolo II" in Bari, Italy. Using Shapley values, we determined the IDE driving features according to two periods, often adopted in clinical practice, of 5 and 10 years from the first tumor diagnosis. Results: Age, tumor diameter, surgery type, and multiplicity are predominant within the 5-year frame, while therapy-related features, including hormone, chemotherapy schemes and lymphovascular invasion, dominate the 10-year IDE prediction. Estrogen Receptor (ER), proliferation marker Ki67 and metastatic lymph nodes affect both frames. Discussion: Thus, our framework aims at shortening the distance between AI and clinical practice.

Downstream Signaling of Inflammasome Pathway Affects Patients' Outcome in the Context of Distinct Molecular Breast Cancer Subtypes.

Inflammasomes are protein complexes involved in the regulation of different biological conditions. Over the past few years, the role of NLRP3 in different tumor types has gained interest. In breast cancer (BC), NLRP3 has been associated with multiple processes including epithelia mesenchymal transition, invasion and metastization. Little is known about molecular modifications of NLRP3 up-regulation. In this study, in a cohort of BCs, the expression levels of NLRP3 and PYCARD were analyzed in combination with CyclinD1 and MYC ones and their gene alterations. We described a correlation between the NLRP3/PYCARD axis and CyclinD1 (p < 0.0001). NLRP3, PYCARD and CyclinD1's positive expression was observed in estrogen receptor (ER) and progesterone receptor (PgR) positive cases (p < 0.0001). Furthermore, a reduction of NLRP3 and PYCARD expression has been observed in triple negative breast cancers (TNBCs) with respect to the Luminal phenotypes (p = 0.017 and p = 0.0015, respectively). The association NLRP3+/CCND1+ or PYCARD+/CCND1+ was related to more aggressive clinicopathological characteristics and a worse clinical outcome, both for progression free survival (PFS) and overall survival (OS) with respect to NLRP3+/CCND1− or PYCARD+/CCND1− patients, both in the whole cohort and also in the subset of Luminal tumors. In conclusion, our study shows that the NLRP3 inflammasome complex is down-regulated in TNBC compared to the Luminal subgroup. Moreover, the expression levels of NLRP3 and PYCARD together with the alterations of CCND1 results in Luminal subtype BC'ss poor prognosis.

An Invasive Disease Event-Free Survival Analysis to Investigate Ki67 Role with Respect to Breast Cancer Patients' Age: A Retrospective Cohort Study.

Characterization of breast cancer into intrinsic molecular profiles has allowed women to live longer, undergoing personalized treatments. With the aim of investigating the relation between different values of ki67 and the predisposition to develop a breast cancer-related IDE at different ages, we enrolled 900 patients with a first diagnosis of invasive breast cancer, and we partitioned the dataset into two sub-samples with respect to an age value equal to 50 years. For each sample, we performed a Kaplan−Meier analysis to compare the IDE-free survival curves obtained with reference to different ki67 values. The analysis on patients under 50 years old resulted in a p-value < 0.001, highlighting how the behaviors of patients characterized by a ki67 ranging from 10% to 20% and greater than 20% were statistically significantly similar. Conversely, patients over 50 years old characterized by a ki67 ranging from 10% to 20% showed an IDE-free survival probability significantly greater than patients with a ki67 greater than 20%, with a p-value of 0.01. Our work shows that the adoption of two different ki67 values, namely, 10% and 20%, might be discriminant in designing personalized treatments for patients under 50 years old and over 50 years old, respectively.

Adenosine pathway inhibitors: novel investigational agents for the treatment of metastatic breast cancer.

INTRODUCTION: The adenosine pathway has been suggested to play a key role in several carcinogenetic processes, with the metabolism of adenosine-5'-triphosphate (ATP) and its derivatives reported to be dysregulated in breast cancer. Preclinical evidence has supported the role of adenosine in the pathogenesis of this malignancy as well as the development of selective adenosine pathway inhibitors. AREAS COVERED: The paper overviews the evidence regarding the use of adenosine pathway inhibitors in breast cancer; a literature search was conducted in January 2022 of Pubmed/Medline, Cochrane library, and Scopus databases. EXPERT OPINION: The adenosine pathway regulates inflammation, apoptosis, metastasis, and cell proliferation in breast cancer cells, and adenosine pathway inhibitors have yielded encouraging results in early-phase clinical trials. Well-designed, multicenter studies focused on monotherapies and combination therapies (which include immune checkpoint inhibitors) are warranted in this setting.

Robustness Evaluation of a Deep Learning Model on Sagittal and Axial Breast DCE-MRIs to Predict Pathological Complete Response to Neoadjuvant Chemotherapy.

To date, some artificial intelligence (AI) methods have exploited Dynamic Contrast-Enhanced Magnetic Resonance Imaging (DCE-MRI) to identify finer tumor properties as potential earlier indicators of pathological Complete Response (pCR) in breast cancer patients undergoing neoadjuvant chemotherapy (NAC). However, they work either for sagittal or axial MRI protocols. More flexible AI tools, to be used easily in clinical practice across various institutions in accordance with its own imaging acquisition protocol, are required. Here, we addressed this topic by developing an AI method based on deep learning in giving an early prediction of pCR at various DCE-MRI protocols (axial and sagittal). Sagittal DCE-MRIs refer to 151 patients (42 pCR; 109 non-pCR) from the public I-SPY1 TRIAL database (DB); axial DCE-MRIs are related to 74 patients (22 pCR; 52 non-pCR) from a private DB provided by Istituto Tumori "Giovanni Paolo II" in Bari (Italy). By merging the features extracted from baseline MRIs with some pre-treatment clinical variables, accuracies of 84.4% and 77.3% and AUC values of 80.3% and 78.0% were achieved on the independent tests related to the public DB and the private DB, respectively. Overall, the presented method has shown to be robust regardless of the specific MRI protocol.

A machine learning ensemble approach for 5- and 10-year breast cancer invasive disease event classification.

Designing targeted treatments for breast cancer patients after primary tumor removal is necessary to prevent the occurrence of invasive disease events (IDEs), such as recurrence, metastasis, contralateral and second tumors, over time. However, due to the molecular heterogeneity of this disease, predicting the outcome and efficacy of the adjuvant therapy is challenging. A novel ensemble machine learning classification approach was developed to address the task of producing prognostic predictions of the occurrence of breast cancer IDEs at both 5- and 10-years. The method is based on the concept of voting among multiple models to give a final prediction for each individual patient. Promising results were achieved on a cohort of 529 patients, whose data, related to primary breast cancer, were provided by Istituto Tumori "Giovanni Paolo II" in Bari, Italy. Our proposal greatly improves the performances returned by the baseline original model, i.e., without voting, finally reaching a median AUC value of 77.1% and 76.3% for the IDE prediction at 5-and 10-years, respectively. Finally, the proposed approach allows to promote more intelligible decisions and then a greater acceptability in clinical practice since it returns an explanation of the IDE prediction for each individual patient through the voting procedure.

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.

Decision support systems for the prediction of lymph node involvement in early breast cancer.

The prediction of lymph node involvement represents an important task which could reduce unnecessary surgery and improve the definition of oncological therapies. An artificial intelligence model able to predict it in pre-operative phase requires the interface among multiple data structures. The trade-off among time consuming, expensive and invasive methodologies is emerging in experimental setups exploited for the analysis of sentinel lymph nodes, where machine learning algorithms represent a key ingredient in recorded data elaboration. The accuracy required for clinical applications is obtainable matching different kind of data. Statistical associations of prognostic factors with symptoms and predictive models implemented also through on-line softwares represent useful diagnostic support tools which translate into patients quality of life improvement and costs reduction.

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

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.

Advancement study of CancerMath model as prognostic tools for predicting Sentinel lymph node metastasis in clinically negative T1 breast cancer patients.

PURPOSE: Sentinel lymph node biopsy (SLNB) is an invasive surgical procedure and although it has fewer complications and is less severe than axillary lymph node dissection, it is not a risk-free procedure. Large prospective trials have documented SLNB that it is considered non-therapeutic in early stage breast cancer. METHODS: Web-calculator CancerMath (CM) allows you to estimate the probability of having positive lymph nodes valued on the basis of tumour size, age, histologic type, grading, expression of estrogen receptor, progesterone receptor. We collected 595 patients referred to our Institute resulting clinically negative T1 breast cancer characterized by sentinel lymph node status, prognostic factors defined by CM and also HER2 and Ki-67. We have compared classification performances obtained by online CM application with those obtained after training its algorithm on our database. RESULTS: By training CM model on our dataset and using the same feature, adding HER2 or ki67 we reached a sensitivity median value of 71.4%, 73%, 70.4%, respectively, whereas the online one was equal to 61%, without losing specificity. The introduction of the prognostic factors Her2 and Ki67 could help improving performances on the classification of particularly type of patients. CONCLUSIONS: Although the training of the model on the sample of T1 patients has brought a significant improvement in performance, the general performance does not yet allow a clinical application of the algorithm. However, the experimental results encourage future developments aimed at introducing features of a different nature in the CM model.

Early prediction of neoadjuvant chemotherapy response by exploiting a transfer learning approach on breast DCE-MRIs.

The dynamic contrast-enhanced MR imaging plays a crucial role in evaluating the effectiveness of neoadjuvant chemotherapy (NAC) even since its early stage through the prediction of the final pathological complete response (pCR). In this study, we proposed a transfer learning approach to predict if a patient achieved pCR (pCR) or did not (non-pCR) by exploiting, separately or in combination, pre-treatment and early-treatment exams from I-SPY1 TRIAL public database. First, low-level features, i.e., related to local structure of the image, were automatically extracted by a pre-trained convolutional neural network (CNN) overcoming manual feature extraction. Next, an optimal set of most stable features was detected and then used to design an SVM classifier. A first subset of patients, called fine-tuning dataset (30 pCR; 78 non-pCR), was used to perform the optimal choice of features. A second subset not involved in the feature selection process was employed as an independent test (7 pCR; 19 non-pCR) to validate the model. By combining the optimal features extracted from both pre-treatment and early-treatment exams with some clinical features, i.e., ER, PgR, HER2 and molecular subtype, an accuracy of 91.4% and 92.3%, and an AUC value of 0.93 and 0.90, were returned on the fine-tuning dataset and the independent test, respectively. Overall, the low-level CNN features have an important role in the early evaluation of the NAC efficacy by predicting pCR. The proposed model represents a first effort towards the development of a clinical support tool for an early prediction of pCR to NAC.

Early Prediction of Breast Cancer Recurrence for Patients Treated with Neoadjuvant Chemotherapy: A Transfer Learning Approach on DCE-MRIs.

Cancer treatment planning benefits from an accurate early prediction of the treatment efficacy. The goal of this study is to give an early prediction of three-year Breast Cancer Recurrence (BCR) for patients who underwent neoadjuvant chemotherapy. We addressed the task from a new perspective based on transfer learning applied to pre-treatment and early-treatment DCE-MRI scans. Firstly, low-level features were automatically extracted from MR images using a pre-trained Convolutional Neural Network (CNN) architecture without human intervention. Subsequently, the prediction model was built with an optimal subset of CNN features and evaluated on two sets of patients from I-SPY1 TRIAL and BREAST-MRI-NACT-Pilot public databases: a fine-tuning dataset (70 not recurrent and 26 recurrent cases), which was primarily used to find the optimal subset of CNN features, and an independent test (45 not recurrent and 17 recurrent cases), whose patients had not been involved in the feature selection process. The best results were achieved when the optimal CNN features were augmented by four clinical variables (age, ER, PgR, HER2+), reaching an accuracy of 91.7% and 85.2%, a sensitivity of 80.8% and 84.6%, a specificity of 95.7% and 85.4%, and an AUC value of 0.93 and 0.83 on the fine-tuning dataset and the independent test, respectively. Finally, the CNN features extracted from pre-treatment and early-treatment exams were revealed to be strong predictors of BCR.