FAP PET/CT for Staging Patients With Breast Cancer

Nodal and distant metastases will be identified on a per-patient and per-lesion basis using [⁶⁸Ga]Ga-FAP-2286 PET/CT in 65 patients undergoing baseline staging. PET/CT images will be read blinded. The reference standard is histopathology where available; if histopathology is not available, a composite of clinical and imaging follow-up will be used. Sensitivity and specificity with 95% confidence intervals will be calculated per patient and per lesion.

In approximately 50 patients undergoing post-neoadjuvant therapy PET/CT prior to surgery, axillary lymph node status will be classified as (a) uptake on PET/CT with histologically confirmed residual disease, or (b) no uptake on PET/CT with pathological complete response (pCR). PET/CT images will be read blinded, with surgical pathology as the gold standard. Concordance (proportion correctly classified) and appropriate concordance statistics will be calculated.

For baseline staging (approximately 65 patients), TNM stage and MDT treatment decisions will be compared before and after availability of [⁶⁸Ga]Ga-FAP-2286 PET/CT results. Proportions of patients with changes will be reported with 95% confidence intervals (Wilson-score method). Data will be extracted from eCRF documentation.

All baseline scans will be independently read by two blinded physicians. Lesions will be scored using a standardized Likert-type scale for presence, risk of malignancy, and clinical relevance on a per-patient and per-lesion basis. Quantitative uptake will be measured using SUVmax, SUVmean, and lesion-to-background ratio (LBR). Interobserver agreement for categorical measures will be quantified with (linearly weighted) Cohen's kappa; for continuous measures, Bland-Altman analysis and intraclass correlation coefficients (ICC) will be used.

Adverse events will be recorded for 48 hours post-injection, assessed per CTCAE v6.0. Data will include incidence, severity, and causality of AEs, SAEs, and SUSARs. Descriptive statistics will summarize findings by type, severity, and relation to the tracer.

ROI-based quantification will be performed for primary tumors, axillary lymph nodes, and suspected distant lesions. Uptake metrics (SUVmax, SUVmean, LBR) will be summarized per lesion, stratified by histological and molecular subtype, and anatomical site using descriptive statistics.

Incidental findings will be documented per patient at baseline , and for patients receiving neoadjuvant chemotherapy, on pre-surgery scans. This will include numbers of incidental findings, type, and clinical consequences. Proportions of incidental findings leading to additional diagnostic procedures (e.g., biopsy, MRI, CT) or clinical intervention will be reported descriptively.