1. Study Design Overview Study Type: Single-center, prospective, observational, diagnostic study. Primary Objective: To validate whether Super-Resolution Ultrasound Imaging (SRUS) can accurately predict pathological complete response (pCR) to neoadjuvant chemotherapy (NAC) in breast cancer at an early stage (after the first cycle of chemotherapy).
Sample Size: A total of 150 patients are planned to be enrolled. Study Period: March 2026 - March 2028. 2. Core Methodology: Cohort Allocation This study employs a classic "Model Development-Validation" cohort design to construct a predictive model and verify its generalizability.
Allocation Method: Systematic random allocation. Allocation Ratio: 7:3. Randomization: Patients will be assigned based on the sequence of enrollment using a computer-generated random sequence.
Blinding Principle: Allocation information will be concealed (blinded) from patients and the clinical treatment team. Only the research coordinators and statisticians will have access to the grouping data to prevent information leakage.
Validation Set (Independent Validation Cohort): Comprising 30% of the sample (45 cases). The data will remain "sealed" until model construction is finalized. It will be used for unbiased, objective performance evaluation of the final model (e.g., calculating AUC, sensitivity). 3. Study Procedures and Visits
The study workflow strictly adheres to the chemotherapy timeline, with core data collection points focused on the early phase of treatment:
V0 (Screening): Confirmation of eligibility criteria. V1 (Baseline, Pre-chemotherapy): Initial SRUS examination to acquire baseline tumor data.
V2 (Early Visit, 48-72 hours after the 1st cycle): The critical data point for the predictive model; the first follow-up examination.
V3 (Mid-term Visit, Pre-4th cycle): The second follow-up examination. V4 (Surgery, 3-4 weeks after the last cycle): Radical surgery is performed. V5 (Endpoint Assessment, 2-4 weeks post-surgery): Acquisition of pathological results to confirm pCR status (the gold standard). 4. Key Technology and Statistics Key Technology: The Mindray Resona A20 ultrasound system will be used in conjunction with Sulfur Hexafluoride (SF6) microbubble contrast agents to extract quantitative parameters such as tumor microvascular density and blood volume.
Statistical Analysis:
In the Training Set: LASSO regression will be used for feature selection to construct a logistic regression predictive model.
In the Validation Set: The formula derived from the training set will be directly applied to calculate the Area Under the ROC Curve (AUC), calibration curves, and other metrics to evaluate model performance. 5. Eligibility Criteria Inclusion Criteria: Females aged 18-75, histologically confirmed invasive breast cancer, scheduled for standard neoadjuvant chemotherapy, with lesions clearly visible on baseline ultrasound.
Exclusion Criteria: History of prior breast cancer treatment (surgery, radiotherapy, chemotherapy), presence of other active malignancies, pregnancy or lactation, severe organ dysfunction, or poor image quality.
Age and Gender: Female patients aged 18 to 75 years.
Diagnosis: Histologically confirmed invasive breast cancer via core needle biopsy.
Treatment Plan: Scheduled to receive standard neoadjuvant chemotherapy at our institution.
Imaging: The lesion is clearly visible on baseline ultrasound. Consent: Willing to participate in the study and sign the written informed consent form.
Prior Treatment: History of any prior treatment for ipsilateral breast cancer (e.g., surgery, radiotherapy, chemotherapy, or targeted therapy).
Other Malignancies: Presence of other active malignancies.
Pregnancy/Lactation: Pregnant or breastfeeding women.
Health Status: Severe cardiac, hepatic, or renal insufficiency, or psychiatric disorders that preclude cooperation with the examination.
This study will enroll female patients aged 18 to 75 years with histologically confirmed invasive breast cancer who are scheduled to receive neoadjuvant chemotherapy (NAC) prior to surgery. Participants must have at least one measurable tumor lesion suitable for super-resolution ultrasound imaging. Patients with evidence of distant metastasis (Stage IV) or those who have received prior systemic therapy for the current cancer diagnosis will be excluded. The study aims to recruit approximately 150 participants from the Anhui Provincial Cancer Hospital. This population represents a typical cohort of locally advanced breast cancer patients undergoing standard NAC protocols.
Cohort
Prospective
Not listed
Discriminative Ability of the Super-Resolution Ultrasound Prediction Model for Pathological Complete Response (pCR)
The discriminative ability of the final prediction model (derived from super-resolution ultrasound parameters) in identifying pCR, assessed in the independent validation cohort. Performance is quantified by the Area Under the Receiver Operating Characteristic Curve (AUC). An AUC value closer to 1 indicates stronger discriminative ability.
Diagnostic performance of the model.
In the validation set, the optimal cutoff value is determined using the Youden index. Performance metrics including sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) are calculated.
Predictive value of early parameter changes.
The ability of early changes in super-resolution ultrasound parameters (measured after the first chemotherapy cycle) to predict final pCR status. Evaluated using univariate ROC analysis to calculate the AUC.
Diagnostic Accuracy, Sensitivity, and Specificity of SRUS Model for pCR
The secondary outcomes include the diagnostic accuracy, sensitivity, and specificity of the SRUS prediction model. These metrics will be calculated based on the confusion matrix derived from the independent validation cohort.
Clinical utility of the model.
The net clinical benefit of using the prediction model to guide clinical decision-making, assessed across a range of threshold probabilities. Evaluated using Decision Curve Analysis (DCA).
Click on the lead sponsor to view all of their trials.
This trial is not recruiting at the moment. You can still explore other options:
