Endoscope-assisted visualization of 5-aminolevulinic acid fluorescence in surgery for brain metastases.

Objective: Fluorescence-guided resection of cerebral metastases has been proposed as an approach to visualize residual tumor tissue and maximize the extent of resection. Critics have argued that tumor cells at the resection margins might be overlooked under microscopic visualization because of technical limitations. Therefore, an endoscope, which is capable of inducing fluorescence, has been applied with the aim of improving exposure of fluorescent tumor tissue. In this retrospective analysis, authors assessed the utility of endoscope assistance in 5-aminolevulinic acid (5-ALA) fluorescence-guided resection of brain metastases.
Methods: Between June 2013 and December 2016, a standard 20-mg/kg dose of 5-ALA was administered 4 hours prior to surgery in 26 patients with suspected single brain metastases. After standard neuronavigated microsurgical tumor resection, a microscope capable of inducing fluorescence was used to examine tumor margins. The authors classified the remaining fluorescence into 3 grades (0 = none, 1 = weak, and 2 = strong). Endoscopic assistance was employed if no or only weak fluorescence was visualized at the resection margins under the microscope. Endoscopically identified fluorescent tissue at the margins was resected and evaluated separately via histological examination to prove or disprove tumor infiltration.
Results: Under the microscope, weakly fluorescent tissue was seen at the margins of the resection cavity in 15/26 (57.7%) patients. In contrast, endoscopic inspection revealed strongly fluorescent tissue in 22/26 (84.6%) metastases. In 11/26 (42.3%) metastases no fluorescence at the tumor margins was detected by the microscope; however, strong fluorescence was visualized under the endoscope in 7 (63.6%) of these 11 metastases. In the 15 metastases with microscopically weak fluorescence, strong fluorescence was seen when using the endoscope. Neither microscopic nor endoscopic fluorescence was found in 4/26 (15.4%) cases. In the 26 patients, 96 histological specimens were obtained from the margins of the resection cavity. Findings from these specimens were in conjunction with the histopathological findings, allowing identification of metastatic infiltration with a sensitivity of 95.5% and a specificity of 75% using endoscope assistance.
Conclusions: Fluorescence-guided endoscope assistance may overcome the technical limitations of the conventional microscopic exposure of 5-ALA-fluorescent metastases and thereby increase visualization of fluorescent tumor tissue at the margins of the resection cavity with high sensitivity and acceptable specificity.