Rapid multimodal imaging and ML identify taurine as biomarker for breast cancer margins

Overview

This study integrates femtosecond label-free imaging microscopy and imaging mass spectrometry with machine learning to rapidly assess breast cancer surgical margins. A biomarker panel including taurine was identified, with taurine abundance correlating with positive margins and poorer overall survival.

Background

Breast-conserving surgery (BCS) is a standard treatment for early-stage breast cancer, where accurate surgical margin assessment is critical to reduce local recurrence and avoid re-excision. Current intraoperative margin assessment methods like frozen section analysis are time-consuming and require skilled pathologists. There is a need for rapid, sensitive, and reliable techniques and biomarkers to guide margin evaluation intraoperatively. Multiphoton microscopy and mass spectrometry imaging offer label-free, high-resolution visualization of tissue morphology and molecular profiles, respectively, but specific biomarkers for margin detection remain lacking.

Data Highlights

Key Findings

• Femtosecond label-free imaging (FLI) microscopy accurately delineates breast cancer and tumor-adjacent noncancerous breast tissue morphology without staining.
• Integration of FLI microscopy with imaging mass spectrometry (IMS) enables rapid visualization of cellular structures and metabolites in unlabeled tissues.
• A biomarker panel comprising taurine, threonate, and glutamate effectively classifies breast cancer from adjacent noncancerous tissue.
• Taurine is identified as a key biomarker with higher abundance in positive surgical margins and is associated with poorer overall survival in breast cancer patients.
• Functional studies confirm the pro-tumorigenic effect of taurine in breast cancer cell lines.
• The combined multimodal imaging and machine learning approach offers potential for real-time intraoperative margin assessment to improve surgical outcomes.

Clinical Implications

The identification of taurine as a biomarker for positive breast cancer surgical margins provides a promising target for rapid intraoperative margin assessment. The multimodal imaging platform integrating FLI microscopy and IMS can facilitate real-time visualization of tissue morphology and metabolite distribution without the need for staining, potentially reducing re-excision rates and improving patient outcomes. This approach may streamline surgical decision-making and optimize breast-conserving surgery.

Conclusion

This study demonstrates that combining femtosecond label-free imaging, mass spectrometry imaging, and machine learning can rapidly and accurately assess breast cancer surgical margins. Taurine emerges as a novel biomarker with clinical relevance for margin evaluation and prognosis.

References

1. Study Authors/Year -- Rapid multimodal imaging integrated with machine learning reveals taurine as a potential biomarker for assessing breast cancer surgical margins