Intraoperative Delineation of Tumor-resection Margins

Head and Neck tumors


Head and Neck Squamous Cell Carcinoma (HNSCC) is the 6th most common cancer worldwide. Despite advances in surgical and nonsurgical treatment, overall survival for these tumors has not improved significantly. ~600,000 people will be diagnosed this year with HNSCC of the oral cavity, oropharynx, larynx, hypopharynx, and nasopharynx. Only 40-50% of these patients will survive for 5 years. Challenges in improving these statistics are multiple and include tumor recurrence at the local and regional level, distant metastasis, and second tumors.

The fluorescence lifetime techniques developed in our laboratory target noninvasive and possibly earlier and more accurate diagnosis and detection of the extent of the neoplastic area in the pre- and intra-operative setting. We have integrated fiber-based FLIm with a DaVinci Surgical System for exploration of in vivo tumor margins. Transoral robotic surgery is implemented in head and neck cancer operations to allow surgeons better access to the small confines of the human oral cavity, particularly in cases where cancer is suspected deep in the throat.

Combining FLIm measurements with transoral robotic surgery could guide more restricted surgical procedures and less aggressive nonsurgical treatment, thus improving survival and reducing morbidity.



Breast tumors


Breast cancer is the second most common cancer worldwide and by far the most frequent cancer among women. Breast conserving therapy tends to be the preferred surgical procedure after early breast cancer diagnosis. During breast conserving surgery the surgeon attempts to excise the entire tumor volume including a surrounding layer (margins) of normal tissue, to minimize the risk of local recurrence.

To date, there is no clinically established method to intraoperatively delineate breast tumor margins leading to re-excisions due to positive margins identified during histologic examination. New technologies able to address this requirement are needed.

Our lab is developing fluorescence lifetime techniques that have potential to analyze the breast tissue molecular makeup and to resolve distinct types of brain tumors from the surrounding normal regions during surgical interventions.

We are using state-of-the-art machine learning techniques to turn the fluorescence lifetime signature into a probabilistic output providing an intraoperative real-time feedback for the surgeon. Likewise, we develop registration procedures to match histology, the gold standard to grade tissue on a benign-malignancy scale, with the video image to provide a reliable classifier training.

The combination of fluorescence lifetime techniques and machine learning could result in the reduction of the re-excision rates, minimizing the cost, pain and mental distress that repeated surgeries can cause. Thus, this technique could play a significant role in the management and treatment of breast cancer.

Comments are closed.