Description

This is a prospective, single-center, single-arm study of 200 unique participants undergoing breast cancer surgery with axillary lymph node dissection independently indicated by the clinical team. Earlier observations from an internal, unregistered protocol are not included in this study population. The first 50 consecutive enrolled participants form a prespecified descriptive device-feasibility subset; they remain part of the full 200-patient cohort and are not a separate arm.

Before AR/VR imaging, the operating surgeon records the indication and extent of the planned axillary dissection and declares that the standard procedure and inspection are complete. A brief ICG near-infrared fluorescence second look is then performed with the wearable system within the already exposed operative field. Each additional fluorescent focus receives a unique identifier, anatomic location, image or video reference when available, and a record of whether it was visible or palpable under standard lighting. The surgeon reassesses the focus under standard lighting and removes it only when safe and clinically appropriate. Each removed focus is submitted separately for pathology. Negative scans, foci left in place, interrupted scans, device failures, and added operating time are recorded.

The primary outcome remains the patient-level rate of pathology-confirmed clinically significant events (CSE) already defined in this record: at least one additional fluorescent focus not planned for removal before the second look, removed after standard-light reassessment, and confirmed to be a lymph node containing isolated tumor cells, micrometastasis, or macrometastasis. These pathology categories are reported separately. For the clinical analysis, pathology determines the nodal category using the standard dissection specimens alone and again after adding separately labeled second-look specimens. The patient-level rate of change in pathological nodal category is reported separately from the CSE rate. Any documented change in the postoperative treatment recommendation is recorded separately from a pathology finding. Procedure- and device-related adverse events are followed for 30 days.

The first 50 participants support a descriptive report of device feasibility and workflow, including scan completion, localization of additional foci, pathology of removed foci, time, and failures. The full 200-patient cohort supports estimation of CSE yield and nodal-category change with confidence intervals. There is no conventional near-infrared camera comparator. This single-arm study does not establish an improvement in recurrence, survival, or other patient outcomes.