Ultra-fast confocal fluorescence microscopy for neck lymph node imaging in head and neck cancer

Objectives: We evaluated ultra-fast confocal fluorescence microscopy (UFCM) as a new modality for pathology practice in head and neck cancer (HNC). This was assessed through an ex vivo study to estimate the accuracy, specificity, and sensitivity of interpretation of UFCM images by pathologists for the detection of metastatic lymph nodes in HNC patients undergoing sentinel lymph node biopsy or selective or complete neck dissection. Materials and Methods: 44 patients with 32 cN0 and 12 cN+ HNC were included. The macroscopically non-invaded fresh bisected lymph nodes were stained with acridine orange and imaged with Histolog® Scanner (UFCM). Two pathologists interpreted independently the UFCM images postoperatively and gave a consensus diagnosis in case of disagreement. The gold standard was the diagnosis based on hematoxylin-eosin-saffron (HES) sections. Results: 201 lymph nodes were imaged by UFCM. Thirty nodes (15 %) were invaded on final histology: 3 with micrometastases and 27 with macrometastases. The concordance rate between the pathologists on the UFCM images was 192/201 = 95.5 % and the Cohen kappa coefficient was 0.80. The accuracy of UFCM was 95.5 % (95 %CI: 91.7 %-97.9 %) with a high specificity at 98.8 % (95 %CI: 95.8 %-99.9 %) but an insufficient sensitivity at 76.7 % (95 %CI: 57.7 %-90.1 %). The three micrometastases and four of the 27 macrometastases were missed on UFCM images. Conclusion: The UFCM is providing promising detection values with a very good specificity and moderate sensitivity carrying room for improvement.