Neurosurg Focus 36 (2):E9, 2014 ©AANS, 2014

Laser scanning confocal in the neurosurgical operating room: a review and discussion of future applications

Michael A. Mooney, M.D.,2 Aqib H. Zehri, B.S.,1,3 Joseph F. Georges, B.S.,1 and Peter Nakaji, M.D.2 1Neurosurgery Research Laboratory, 2Division of Neurological Surgery, Barrow Neurological Institute, St. Joseph’s Hospital and Medical Center, Phoenix; and 3College of Medicine, The University of Arizona, Phoenix, Arizona

Laser scanning confocal endomicroscopy (LSCE) is an emerging technology for examining brain neoplasms in vivo. While great advances have been made in macroscopic in recent years, the ability to perform in vivo expands the potential of fluorescent tumor labeling, can improve intraoperative tissue diagnosis, and provides real-time guidance for tumor resection intraoperatively. In this review, the authors highlight the technical aspects of confocal endomicroscopy and fluorophores relevant to the neurosurgeon, provide a compre- hensive summary of LSCE in animal and human neurosurgical studies to date, and discuss the future directions and potential for LSCE in neurosurgery. (http://thejns.org/doi/abs/10.3171/2013.11.FOCUS13484)

Key Words • brain neoplasm • resection • glioma • pathology • surgery • intraoperative • in vivo • confocal microscopy • diagnosis • neuronavigation

ntraoperative tissue analysis is essential in guiding the cellular level can be gathered in the operating room. diagnosis and treatment decisions in the neurosurgi- In this review, we provide a background on intraoperative cal setting. Frozen tissue is the most widely confocal devices and fluorophore technology, describe Iused approach for intraoperative tissue analysis, but its the application of this technology in both animal mod- limitations (for example, sampling bias, waiting time, els and human studies, and discuss the future directions and freezing artifacts) are frequently encountered and and potential applications of confocal endomicroscopy in widely recognized.40,42 Advances in optical have neurosurgery. expanded the tools available to both the neurosurgeon and the neuropathologist, and their implementation in the neurosurgical operating room has the potential to facili- Technical Considerations: Confocal Endoscopes tate, and also refine, intraoperative diagnosis, treatment and Fluorophores decisions, and surgical resections. Confocal microscopy is one of these valuable tools, Laser scanning confocal microscopy (LSCM) is an as it can provide high-resolution imaging of tissue mor- optical fluorescence imaging modality used for imaging phology, cytoarchitecture, and intracellular elements at thick in vivo and ex vivo tissues. In LSCM, a specific various depths within a tissue sample. The application of wavelength laser is raster scanned across a tissue, and flu- confocal imaging into a handheld endomicroscope with orophores within the tissue are excited. Photons emitted in vivo utility has spurred the exploration of this technol- from the excited fluorophores pass through an objective ogy in the management of brain tumors intraoperatively. and then a confocal aperture (pinhole) that spatially re- When combined with intravenous and topical fluoro- stricts photons emitted from above and below the point of phores, real-time information about brain neoplasms at focus. This allows generation and visualization of optical sections, which are images of thin tissue planes gener- Abbreviations used in this paper: GBM = glioblastoma; ICG = ated without physical sectioning of the examined tissue. indocyanine green; IGFBP7 = insulin-like growth factor-binding Tissue sections from various depths within the tissue can protein–7; LSCE = laser scanning confocal endomicroscopy; LSCM be visualized by translating the imaging objective along = laser scanning confocal microscopy; 5-ALA = 5-aminolevulinic a vertical axis.5 Recently, this technology has been minia- acid. turized into clinically available endomicroscopy systems

Neurosurg Focus / Volume 36 / February 2014 1

Unauthenticated | Downloaded 09/30/21 08:56 AM UTC M. A. Mooney et al.

(LSCE), which have proven utility in gastrointestinal croendoscopes and reported visualization of rodent hip- tract in vivo histopathology.11 pocampal and thalamic capillaries following fluorescein Clinical confocal endomicroscopes generate contrast injection, as well as visualization of YFP-expressing py- when coupled to appropriate fluorescent agents. Fluores- ramidal neurons in the rodent hippocampus.20 Since that cent agents with neurosurgical use include fluorescein, time, confocal imaging approaches have been implement- indocyanine green (ICG), and 5-aminolevulinic acid (5- ed to study a variety of molecular and cellular processes ALA). Of these, the first two are FDA approved in the US in the rodent brain in vivo.27,30 for nonneurosurgical applications, and the third is under The use of LSCE for specific neurosurgical purposes, research evaluation. In Europe and many other countries, however, has been much more limited, making this field formal approval has been obtained for 5-ALA. Specific ripe for further exploration. To date, 4 reports have been fluorophore properties and clinical uses are outlined in published utilizing commercially available LSCEs in ani- Table 1. Fluorescein and ICG generally contrast tissue ar- mal brain ex vivo or in vivo. chitecture similarly to H & E staining. ICG can be used The first published report came out of our institution to visualize deeper structures within tissue due to its in- in 2010.35 In this proof-of-principle study, LSCE (Opti- frared excitation-emission, which travels further through scan FIVE 1) was used with intravenous fluorescein and tissue without scattering.3 5-ALA, on the other hand, pro- topical acriflavine dye to visualize glioblastoma xeno- vides tumor-specific contrast that is more similar to im- grafts in mouse brain in vivo. In addition to demonstrat- munohistochemistry. It is preferentially concentrated in ing the ability of LSCE to visualize normal cerebral cap- neoplastic cells and converted to the endogenous fluoro- illary beds and distinguish normal gray and white matter phore protoporphyrin IX intracellularly, which provides in the brain in vivo, this study was the first demonstration fluorescent contrast of neoplastic cell cytoplasm. 5-ALA of LSCE’s ability to distinguish between neoplastic tissue is best excited with ultraviolet (UV) light and emits far- and normal brain, highlighting the potential implications red photons, which can travel further through tissue with- of this technology for distinguishing tumor boundaries in out scattering. This makes appropriate coupling to excita- the brain in vivo. tion/emission filters particularly important for minimiz- This study was corroborated in 2011–2012 with 2 ing detection of autofluorescence generated by UV light studies demonstrating the application of ICG to LSCE and collection of far-red photons specific to 5-ALA.21 tumor imaging in vivo.10,25 Indocyanine green had been Clinically available confocal endomicroscopy systems previously demonstrated to enhance intraoperative imag- contain with precise excitation wavelengths and ing of gliomas in vivo,12–14 but the application of LSCE dichroic filters for detecting appropriate emission wave- and near-infrared imaging in these 2 2011–2012 studies lengths. The experimental use of these systems for neuro- allowed for improved visualization of tumor transition surgical applications is gaining in popularity as they allow zones and the interface between neoplasm and normal real-time visualization of tissue cytoarchitecture in the op- brain, demonstrating the potential of this technology for erating room. Currently, Optiscan and Cellvizio produce guiding intraoperative resection and decision making. In commercially available laser scanning confocal endomi- the study by Martirosyan et al.,25 ICG was clearly shown croscopes. These systems both use laser scanning systems to contrast the cytoplasm of neoplastic cells, allowing for coupled to optical fibers. The Optiscan device is available visualization of pleomorphism and mitoses in these ar- in a 488-nm and infrared excitation platform. Cellvizio eas, as well as extracellular tumor matrix, allowing for produces single-laser and dual-laser systems containing visualization of hypercellularity and necrosis in regions 488-nm and 560-nm excitation. Both of these manufactur- of tumor. In the study by Foersch et al.,10 extravasation ers use specific dichroic mirrors for filtering excitation and of ICG in regions of tumor, as well as small intratumoral emission photons. Technical specifications of these 2 sys- hemorrhages, was demonstrated with LSCE, and further tems and available probes are listed in Table 2. cellular and nuclear morphology was discerned through the use of acriflavine hydrochloride. Notably, one limita- Animal Models tion of intravenous ICG fluorescence is its time depen- dence, and this phenomenon was cited in both studies. Laser scanning confocal microscopy has been exten- Although ICG is visualized within blood vessels, capil- sively applied in the field of basic neuroscience research, laries, and extracellular tumor matrix in the early phase and numerous groups have used endoscopic LSCM tech- following administration, over time, leakage into normal niques to image fluorescence in the animal brain in vivo. tissues does occur, complicating image interpretation. In 2004, Jung et al. incorporated confocal probes into mi- Foersch et al. demonstrated one strategy to circumvent TABLE 1: Fluorophore properties

Tested in Tested Fluorophore Excitation (nm) Emission (nm) Route of Administration Localization Animals Clinically fluorescein 494 521 intravenous extracellular yes yes ICG 778 820 intravenous extracellular yes no* 5-ALA 410 635 & 704 oral intracellular yes yes

* Indocyanine green has been used clinically for visualization of vasculature but not with LSCE.

2 Neurosurg Focus / Volume 36 / February 2014

Unauthenticated | Downloaded 09/30/21 08:56 AM UTC Laser scanning confocal endomicroscopy

TABLE 2: Confocal endomicroscope and probe properties

Wavelength Lat Resolution Axial Resolution Working Distal Tip System Probes Detected (nm) (μm) (μm) FOV (μm) Distance (μm) Diameter (mm) Optiscan/Pentax confocal endomicro- 488 & 790 0.7 7 475 × 475 250 5.0 ISC-1000 scope Cellvizio S 488 & 660 5 20 600 × 500 0 1.5 HD 488 & 660 2–3 20 240 × 240 30–80 1.8 Z 488 & 660 30 30 600 × 500 70 0.94–1.8 MinoO 488 & 660 1 3 240 × 240 30 2.6–4.2 AlveoFlex 488 & 660 5 100 600 × 500 0 1.4 this phenomenon, which was through direct expression of ditional features seen in GBM, such as apoptotic figures eYFP in tumor cell lines. Tumor cells were readily identi- in perinecrotic palisading tumor cells, giant cells, and fi- fied with this strategy through LSCE, and even single tu- brillary tumor matrix/blood vessels in select specimens. mor cells migrating from the main tumor into normal tis- Notably, however, not all WHO criteria were identified sue could be visualized. This demonstrates the potential in all 9 patients included in this study. While cell den- for improved visualization of neoplastic cells with LSCE sity and cell pleomorphism criteria were met in all 9 pa- through advances in fluorescence-labeling techniques tients, microvascular proliferation was only visualized (discussed below). in 4 of 9 patients, and mitosis was only visualized in 2 In a divergence from animal models of glioma, Peyre of 9 patients. These findings may not be a limitation of et al. published a recent study examining the ability of LSCE, however, since this study was not designed to di- LSCE and ICG to identify histological properties of me- rectly compare LSCE diagnosis to H & E diagnosis, and ningiomas ex vivo in an animal model.31 This study used more tissue samples were collected for H & E staining a LSCE (Optiscan) to examine tumor specimens ex vivo than for LSCE in these patients. Furthermore, excellent for evidence of brain invasion in 2 animal models of ag- agreement between LSCE and H & E–stained images gressive meningioma (a transgenic mouse model and a was demonstrated for numerous tissue samples collected xenograft model). Laser scanning confocal microscopy from the same tumor resection bloc in various patients reliably differentiated between histological subtypes of in this study. The limitations of the study included the meningioma in the transgenic model and was able to use of topical acriflavine hydrochloride, as well as the ex detect invasion of Virchow-Robin spaces and brain pa- vivo application of LSCE technology: Although topical renchyma (with the addition of ICG), demonstrating the acriflavine readily distinguishes superficial cell borders potential application of LSCE in meningioma surgery for and their nuclei for LSCE analysis,15 mutagenesis with improving the detection invasive disease. this agent has been demonstrated in cell culture,8 and it is not currently used in vivo in the CNS. Furthermore, while the ex vivo approach can be justified for testing the Human Studies feasibility of LSCE tissue analysis, the use of LSCE for Alongside studies of LSCE in animal models, this guiding tissue biopsy and maximizing extent of resec- technology has been explored in the study of human brain tion requires an in vivo application of this technology. tumors. To date, 5 reports have been published utilizing Notably, one other study examined ex vivo application of LSCE in either ex vivo or in vivo human studies. LSCE in human brain tumors, with similar results.10 Ad- In 2010, Schlosser et al. reported the first use of ditionally, at least 3 studies examining LSCM for ex vivo LSCE, which they termed neurolasermicroscopy, in hu- analysis of brain tumors have been reported with promis- man brain tissue.37 This study examined 9 patients with ing results.38,46,48 glioblastoma (GBM) and compared LSCE with conven- Sanai et al. expanded upon the application of LSCE tional histopathology in ex vivo brain tissue samples. in 2011 when they used this technology with intravenous LSCE was used intraoperatively immediately following fluorescein to examine a variety of brain tumor subtypes excision of the brain tumor specimen, and, thereafter, in vivo at our institution.33 In their prospective feasibility the specimen was sent for conventional histopathologi- study, any patient with an intracranial mass undergoing cal examination (that is, H & E, periodic acidic Schiff, craniotomy for biopsy and/or resection was considered silver-impregnation, and/or immunohistochemical stain- eligible, and 33 patients with a variety of lesions (for ex- ing). Acriflavine hydrochloride was applied topically to ample, meningiomas, oligodendrogliomas, astrocytomas, ex vivo tissue samples prior to LSCE examination. and metastases, among others) were enrolled. Following With this technique, Schlosser et al. were able to surgical exposure of the tumor and intravenous fluores- identify all WHO microscopic criteria for GBM diag- cein administration, LSCE images were acquired over nosis (that is, cell number/density criteria, cell pleomor- the course of approximately 10 minutes. Following image phism, mitotic figures, microvascular proliferation, and acquisition, tissue biopsies were obtained at the imaging pseudopalisading necrosis) using LSCE ex vivo. Laser site and stained with H & E (as well as other stains, when scanning confocal endomicroscopy also identified ad- applicable) for histological analysis and comparison.

Neurosurg Focus / Volume 36 / February 2014 3

Unauthenticated | Downloaded 09/30/21 08:56 AM UTC M. A. Mooney et al.

Utilizing this approach, specific confocal features (Fig. 1). Furthermore, in 6 of the 10 patients, microscopic of high-grade glioma (specifically, neovascularization, fluorescence was visualized at the margins of the resec- dense cellularity, and irregular cellular phenotypes) were tion cavity after a gross-total resection was thought to be readily identified in vivo and confirmed histologically. achieved—this finding suggests that there may be a sig- Notably, it was reported that these features were readily nificant role for intraoperative LSCE and 5-ALA imag- apparent to both the surgeon and the neuropathologist in- ing in achieving full extent of resection in patients with traoperatively, allowing for integration of this informa- low-grade glioma in the future. The Barrow 5-ALA In- tion into the operative plan. Additionally, specific cellular traoperative Confocal (BALANCE) study is a Phase IIIa, features of other tumor subtypes, including low-grade multicenter, randomized, placebo-controlled trial that is glioma, neurocytoma, hemangioblastoma, and menin- currently underway to evaluate this approach (Clinical- gioma, were identified on LSCE images, demonstrating Trials.gov ID no. NCT01502280). the potential for LSCE to intraoperatively diagnose and guide treatment decisions for these lesions. Following this feasibility analysis, a prospective Future Directions study of LSCE and intravenous fluorescein was per- Despite the successful application of LSCE in vivo to formed at our institution to determine the accuracy and date, much work remains to fully incorporate this tech- reliability of real-time intraoperative diagnosis of a vari- nology into the neurosurgical operating room. We be- ety of brain tumors in vivo.7 Fifty patients were included lieve that future advances will come in 3 main realms: in the study in which image acquisition and tumor biopsy fluorescent and nonfluorescent technology for imaging specimens were obtained in a similar fashion to the afore- neoplasms in vivo; confocal endoscope technology for mentioned study. This was followed by a blinded assess- improving imaging quality, depth, and ease of use; and ment of 2 neuropathologists’ ability to properly diagnose improved incorporation of LSCE into the neurosurgery tissue specimen based on intraoperative LSCE images. and neuropathology work flow. Tumors in this study included 24 meningiomas, 12 high- Fluorescein and ICG are primarily visualized grade gliomas, 8 low-grade gliomas, 4 schwannomas, 1 through the enhanced permeability and retention effect, hemangioblastoma, and 1 ependymoma. Features identi- which means that fluorophores are preferentially taken up fied on confocal imaging were extensively described for by tumor tissue due to increased breakdown and leakage these tumors and their subtypes utilizing LSCE and H & in the blood-brain barrier. 5-ALA, on the other hand, pro- E comparison slides. Interestingly, in the blinded portion vides detection of neoplastic cells through intracellular of this study, 26 (92.9%) of 28 LSCE images were cor- fluorescence, but its ability to demarcate tumor margins rectly diagnosed, including 9 of 9 glioma images and 2 of is somewhat subjective, partly due to difficulty interpret- 2 infiltrating glioma edge images. This provides evidence ing levels of fluorescence near tumor boundaries.19,24,47 for the role of in vivo LSCE for intraoperative diagnosis All 3 of these currently available fluorescent agents also and also demonstrates its potential for in vivo identifica- have limited circulation time and readily diffuse into and tion of infiltrating tumor edge and maximizing extent of out of interstitial space.41 Intraoperative confocal micros- resection in infiltrative tumors. copy combined with molecular imaging probes specific While initial results from studies utilizing LSCE with for tumor biomarkers could play a significant role in fu- intravenous fluorescein are encouraging, technical limita- ture neurosurgical applications of LSCE. tions in visualization have been recognized. Specifically, Molecular probes that have been investigated for use the ability to visualize nuclear morphology, cytoplasm in brain neoplasms can be divided into 3 functional cat- characteristics, and nuclear-to-cytoplasm ratios is limited egories: peptides, antibodies, and nanoparticles. The ef- when using fluorescein as a . 33 Administration ficacy and applicability of these probes is based on their of 5-ALA offers one potential solution to this limitation, as selectiveness for tumor tissue, resistance to photo bleach- it provides intracellular fluorescent signal preferentially in ing and autofluorescence (by using near-infrared probes), tumor cells. The capacity of 5-ALA for macroscopic fluo- and capacity to be safely administered to patients.32 rescence in high-grade gliomas and its utility in resection Peptide probes have been used to label multiple cel- of these tumors has been demonstrated and extensively dis- lular targets because of their high affinity and specificity cussed elsewhere.39 Its utility in low-grade glioma surgery, for intracellular proteins. One promising example of this on the other hand, was previously limited due to the lack is a peptide for the cell-adhesion glycoprotein, of macroscopic fluorescence in these cells.9,47 Interestingly, integrin avb3. Integrin avb3 is expressed in tumor cells despite the lack of macroscopic fluorescence in low-grade and plays a role in tumor growth, angiogenesis, and me- gliomas, microscopic fluorescence has been observed,9,19 tastasis.16 Its expression is low in normal cells but high in and in a recent study of LSCE paired with 5-ALA for low- tumors cells and tumor vessels, and fluorescently labeled grade glioma, visualization of fluorescently labeled tumor near-infrared peptide probes have been used to specifi- cells was achieved.34 cally target integrin avb3 in glioblastomas and medul- This study by Sanai et al. examined 10 patients with loblastomas in vivo in animal models.4,16,17,26,49 These newly diagnosed low-grade glioma and evaluated the studies demonstrate the ability to specifically tag tumor ability of LSCE to detect intraoperative 5-ALA fluores- cells with fluorescent peptide probes and help differen- cence in vivo in these patients. While macroscopic tumor tiate between neoplasm and normal brain tissue. How- fluorescence was not present in any patient, microscopic ever, heterogeneous expression of integrins in nontumor fluorescence with LSCE was present in 10 of 10 patients regions makes the specificity of the probes an ongoing

4 Neurosurg Focus / Volume 36 / February 2014

Unauthenticated | Downloaded 09/30/21 08:56 AM UTC Laser scanning confocal endomicroscopy

Fig. 1. Microscopic 5-ALA fluorescence detected in a WHO Grade II glioma using intraoperative LSCE. A: Intraoperative MR images with neuronavigation highlighting the edge of the resection cavity. B: Intraoperative placement of confocal probe at the edge of the resection cavity. C: LSCE detection of 5-ALA fluorescence in this region. Adapted from Sanai N et al: Intra- operative confocal microscopy in the visualization of 5-aminolevulinic acid fluorescence in low-grade gliomas. Clinical article. J Neurosurg 115(4):740–748, 2011. issue. Other promising peptide probes include those tar- of demarcating the brain-tumor interface, further experi- geted toward cysteine cathepsin (GB119)6 and epidermal mentation in clinical models is needed to establish the safe- growth factor receptor (EGRF),1 but use of these probes ty profile and long-term effects before translation into the for guiding neurosurgical resection or tissue sampling in- clinical setting can proceed. Notably, confocal microscopy traoperatively must still be explored. techniques that do not require the use of exogenous fluoro- Antibodies have also been developed to target brain phores have been described and show exciting potential.48 tumor molecular targets. A 2010 study used single-do- In addition to advances in tumor-specific fluoro- main antibodies (sdAb) targeted to anti–insulin-like phores, the future success of LSCE depends upon ad- growth factor-binding protein–7 (IGFBP7), which accu- vances in confocal endoscope technology. Future LSCE mulates in the basement membranes of glioblastoma en- technology must facilitate deeper tissue imaging, greater dothelium.18 Due to their small size and nanomolar affin- detection of weak fluorescent signals, and faster acqui- ity, IGFBP7 antibodies displayed rapid accumulation to sition than what is currently available. Current LSCE human and ani­mal GBM vessels after systemic injection. systems have 1–2 mm of imaging depth in the brain, but Anti-IGFBP7 sd­Abs have advantages compared with advances in fluorophores and available detection spectra other predominant­­ly peptide-based strategies, including in endoscopes will improve these imaging properties and high specificity to the tar­get, and the appropriate pharma- allow for more useful incorporation of LSCE into the cokinetic characteristics for imaging applications, such as neurosurgical operating room. Systems that have tunable PET scan.18 filters for specific fluorophores with intelligent algorithms Antibodies conjugated to fluorophores have also for multispectral imaging are being developed to enhance been used to target vascular endothelial growth factor fluorescence detection. receptor–1 (VEGFR-1) in a transgenic mouse model of Lastly, the implementation of LSCE requires changes medulloblastoma.44 Results of using this conjugated an- in neurosurgical and neuropathological training, as well as tibody demonstrated preferential binding to tumor tissue the neurosurgical workflow. Intraoperative fluorescence- and microscopic delineation of tumor margins using a guided surgery provides real-time information that can confocal microscopic in vivo. Further studies incorpo- overcome the issues of standard frozen-section pathol- rating this strategy and using a miniaturized microscope ogy; however, interpretation of LSCE-generated images intraoperatively have exciting potential. requires a shift from interpreting routine H & E–stained Nanoparticles are organic or inorganic particles of tissue. This transition requires neurosurgeons and/or neu- nanometer size that have also been demonstrated to provide ropathologists to be trained in interpreting LSCE images, intraoperative imaging of brain neoplasms. Nanoparticle which would facilitate both intraoperative diagnosis and strategies have been used to demarcate tumor margins and tumor resection by eliminating frozen-section processing have been incorporated into multimodal imaging strate- time and decreasing sampling error. Initial studies have gies by conjugating near-infrared fluorochromes (quantum already documented the ability of neuropathologists to be dots)2,29,36,45 and traditional imaging molecules (fluorescein trained to interpret confocal images.7,38 In the absence of and ICG)24 with MRI-detectable iron oxide or polymer- trained neuropathologists at a given institution, LSCE lends based cores.22,28,32,41,43,50,51 Magneto-optical nanoparticles itself to tele-pathology through the real-time transmission permit visualization of tomographic landmarks on MRI for of digital images.7 Although the use and incorporation of surgical planning while subsequently allowing for intraop- LSCE technology requires significant change in practice erative fluorescent visualization of tumor boundaries.23,28 up front, the potential for this technology to improve the Although these imaging probes provide a precise method neurosurgical approach to brain neoplasms is extensive.

Neurosurg Focus / Volume 36 / February 2014 5

Unauthenticated | Downloaded 09/30/21 08:56 AM UTC M. A. Mooney et al.

Conclusions 11. Gheonea DI, Cârţână T, Ciurea T, Popescu C, Bădărău A, Săftoiu A: Confocal laser endomicroscopy and immunoen- Laser scanning confocal endomicroscopy has great doscopy for real-time assessment of vascularization in gas- potential to improve the way we approach and treat brain trointestinal malignancies. World J Gastroenterol 17:21–27, neoplasms in the neurosurgical operating room. Despite 2011 the wide array of neurosurgical applications of LSCE, the 12. Haglund MM, Berger MS, Hochman DW: Enhanced optical in vivo applications of this technology remain limited in imaging of human gliomas and tumor margins. Neurosur- gery 38:308–317, 1996 both animals and humans, and there is a need for future 13. Haglund MM, Hochman DW, Spence AM, Berger MS: En- high-impact studies in this area. Through improved mo- hanced optical imaging of rat gliomas and tumor margins. lecular labeling and confocal endoscope technology, as Neurosurgery 35:930–941, 1994 well as through efficient and effective translation of this 14. Hansen DA, Spence AM, Carski T, Berger MS: Indocyanine technology into the operating room, LSCE will provide a green (ICG) staining and demarcation of tumor margins in a powerful tool for achieving better and safer patient out- rat glioma model. Surg Neurol 40:451–456, 1993 comes in the neurosurgical operating room in the future. 15. Hoffman A, Goetz M, Vieth M, Galle PR, Neurath MF, Kiesslich R: Confocal laser endomicroscopy: technical status and current indications. Endoscopy 38:1275–1283, 2006 Disclosure 16. Hsu AR, Hou LC, Veeravagu A, Greve JM, Vogel H, Tse V, Funding for this project was provided by Barrow Neurological et al: In vivo near-infrared fluorescence imaging of integrin Foundation to Dr. Nakaji. alphavbeta3 in an orthotopic glioblastoma model. Mol Imag- Author contributions to the study and manuscript preparation ing Biol 8:315–323, 2006 include the following. Conception and design: Mooney. Acquisi- 17. Huang R, Vider J, Kovar JL, Olive DM, Mellinghoff IK, May- tion of data: Mooney, Zehri, Georges. Drafting the article: Mooney, er-Kuckuk P, et al: Integrin αvb3-targeted IRDye 800CW Zehri, Georges. Critically revising the article: Nakaji. Reviewed near-infrared imaging of glioblastoma. Clin Cancer Res 18: submitted version of manuscript: Nakaji. Approved the final version 5731–5740, 2012 of the manuscript on behalf of all authors: Nakaji. Study supervision: 18. Iqbal U, Albaghdadi H, Luo Y, Arbabi M, Desvaux C, Veres Nakaji. T, et al: Molecular imaging of glioblastoma multiforme us- ing anti-insulin-like growth factor-binding protein-7 single- domain antibodies. Br J Cancer 103:1606–1616, 2010 References 19. Ishihara R, Katayama Y, Watanabe T, Yoshino A, Fukushima T, 1. Agnes RS, Broome AM, Wang J, Verma A, Lavik K, Basilion Sakatani K: Quantitative spectroscopic analysis of 5-aminolev- JP: An optical probe for noninvasive molecular imaging of or- ulinic acid-induced protoporphyrin IX fluorescence intensity in thotopic brain tumors overexpressing epidermal growth factor diffusely infiltrating astrocytomas. Neurol Med Chir (Tokyo) receptor. Mol Cancer Ther 11:2202–2211, 2012 47:53–57, 2007 2. Arndt-Jovin DJ, Kantelhardt SR, Caarls W, de Vries AH, 20. Jung JC, Mehta AD, Aksay E, Stepnoski R, Schnitzer MJ: In Giese A, Jovin Ast TM: Tumor-targeted quantum dots can vivo mammalian brain imaging using one- and two-photon help surgeons find tumor boundaries. IEEE Trans Nanobio- fluorescence microendoscopy.J Neurophysiol 92:3121–3133, science 8:65–71, 2009 2004 3. Behbahaninia M, Martirosyan NL, Georges J, Udovich JA, 21. Juzenas P, Juzeniene A, Kaalhus O, Iani V, Moan J: Noninva- Kalani MY, Feuerstein BG, et al: Intraoperative fluorescent sive fluorescence excitation spectroscopy during application imaging of intracranial tumors: a review. Clin Neurol Neu- of 5-aminolevulinic acid in vivo. Photochem Photobiol Sci rosurg 115:517–528, 2013 1:745–748, 2002 4. Chen X, Conti PS, Moats RA: In vivo near-infrared fluores- 22. Kircher MF, de la Zerda A, Jokerst JV, Zavaleta CL, Kempen cence imaging of integrin alphavbeta3 in brain tumor xeno- PJ, Mittra E, et al: A brain tumor molecular imaging strat- grafts. Cancer Res 64:8009–8014, 2004 egy using a new triple-modality MRI-photoacoustic-Raman 5. Claxton NS, Fellers TJ, Davidson MW: Laser Scanning Con- nanoparticle. Nat Med 18:829–834, 2012 focal Microscopy. Olympus, 2013 (http://www.olympuscon 23. Kircher MF, Mahmood U, King RS, Weissleder R, Josephson focal.com/theory/LSCMIntro.pdf) [Accessed December 17, L: A multimodal nanoparticle for preoperative magnetic reso- 2013] nance imaging and intraoperative optical brain tumor delinea- 6. Cutter JL, Cohen NT, Wang J, Sloan AE, Cohen AR, Panneer- tion. Cancer Res 63:8122–8125, 2003 selvam A, et al: Topical application of activity-based probes 24. Li Y, Rey-Dios R, Roberts DW, Valdés PA, Cohen-Gadol AA: for visualization of brain tumor tissue. PLoS ONE 7:e33060, Intraoperative fluorescence-guided resection of high-grade gli- 2012 omas: a comparison of the present techniques and evolution of 7. Eschbacher J, Martirosyan NL, Nakaji P, Sanai N, Preul MC, future strategies. World Neurosurg [epub ahead of print], 2013 Smith KA, et al: In vivo intraoperative confocal microscopy 25. Martirosyan NL, Cavalcanti DD, Eschbacher JM, Delaney for real-time histopathological imaging of brain tumors. Clin- PM, Scheck AC, Abdelwahab MG, et al: Use of in vivo near- ical article. J Neurosurg 116:854–860, 2012 infrared laser confocal endomicroscopy with indocyanine 8. Ferenc T, Janik-Spiechowicz E, Bratkowska W, Denys A: Muta- green to detect the boundary of infiltrative tumor. Laboratory genic activity of 3,6-diamino-10-methyl-9,10-dihydroacridine investigation. J Neurosurg 115:1131–1138, 2011 in Salmonella typhimurium cells. Int J Occup Med Environ 26. Moore SJ, Hayden Gephart MG, Bergen JM, Su YS, Rayburn Health 12:67–72, 1999 H, Scott MP, et al: Engineered knottin peptide enables nonin- 9. Floeth FW, Sabel M, Ewelt C, Stummer W, Felsberg J, Reif- vasive optical imaging of intracranial medulloblastoma. Proc enberger G, et al: Comparison of (18)F-FET PET and 5-ALA Natl Acad Sci U S A 110:14598–14603, 2013 fluorescence in cerebral gliomas. Eur J Nucl Med Mol Im- 27. Navarrete M, Perea G, Fernandez de Sevilla D, Gómez-Gon- aging 38:731–741, 2011 zalo M, Núñez A, Martín ED, et al: Astrocytes mediate in 10. Foersch S, Heimann A, Ayyad A, Spoden GA, Florin L, vivo cholinergic-induced synaptic plasticity. PLoS Biol 10: Mpoukouvalas K, et al: Confocal laser endomicroscopy for e1001259, 2012 diagnosis and histomorphologic imaging of brain tumors in 28. Olson ES, Jiang T, Aguilera TA, Nguyen QT, Ellies LG, vivo. PLoS ONE 7:e41760, 2012 Scadeng M, et al: Activatable cell penetrating peptides linked

6 Neurosurg Focus / Volume 36 / February 2014

Unauthenticated | Downloaded 09/30/21 08:56 AM UTC Laser scanning confocal endomicroscopy

to nanoparticles as dual probes for in vivo fluorescence and 41. Tréhin R, Figueiredo JL, Pittet MJ, Weissleder R, Josephson MR imaging of proteases. Proc Natl Acad Sci U S A 107: L, Mahmood U: Fluorescent nanoparticle uptake for brain tu- 4311–4316, 2010 mor visualization. Neoplasia 8:302–311, 2006 29. Orringer DA, Koo YE, Chen T, Kim G, Hah HJ, Xu H, et al: 42. Uematsu Y, Owai Y, Okita R, Tanaka Y, Itakura T: The useful- In vitro characterization of a targeted, dye-loaded nanodevice ness and problem of intraoperative rapid diagnosis in surgical for intraoperative tumor delineation. Neurosurgery 64:965– neuropathology. Brain Tumor Pathol 24:47–52, 2007 972, 2009 43. Veiseh O, Sun C, Fang C, Bhattarai N, Gunn J, Kievit F, et al: 30. Pérez-Alvarez A, Araque A, Martín ED: Confocal microsco- Specific targeting of brain tumors with an optical/magnetic py for astrocyte in vivo imaging: recycle and reuse in micros- resonance imaging nanoprobe across the blood-brain barrier. copy. Front Cell Neurosci 7:51, 2013 Cancer Res 69:6200–6207, 2009 31. Peyre M, Clermont-Taranchon E, Stemmer-Rachamimov A, 44. Wang D, Chen Y, Leigh SY, Haeberle H, Contag CH, Liu JT: Kalamarides M: Miniaturized handheld confocal microscopy Microscopic delineation of medulloblastoma margins in a identifies focal brain invasion in a mouse model of aggressive transgenic mouse model using a topically applied VEGFR-1 meningioma. Brain Pathol 23:371–377, 2013 Probe. Transl Oncol 5:408–414, 2012 32. Pogue BW, Gibbs-Strauss S, Valdés PA, Samkoe K, Roberts 45. Wang J, Yong WH, Sun Y, Vernier PT, Koeffler HP, Gunder- DW, Paulsen KD: Review of neurosurgical fluorescence im- sen MA, et al: Receptor-targeted quantum dots: fluorescent aging methodologies. IEEE J Sel Top Quantum Electron probes for brain tumor diagnosis. J Biomed Opt 12:044021, 16:493–505, 2010 2007 33. Sanai N, Eschbacher J, Hattendorf G, Coons SW, Preul MC, 46. Whitson WJ, Valdes PA, Harris BT, Paulsen KD, Roberts DW: Smith KA, et al: Intraoperative confocal microscopy for brain Confocal microscopy for the histological fluorescence pattern tumors: a feasibility analysis in humans. Neurosurgery 68 (2 of a recurrent atypical meningioma: case report. Neurosur- Suppl Operative):282–290, 2011 gery 68:E1768–E1773, 2011 34. Sanai N, Snyder LA, Honea NJ, Coons SW, Eschbacher JM, 47. Widhalm G, Wolfsberger S, Minchev G, Woehrer A, Krssak M, Smith KA, et al: Intraoperative confocal microscopy in the vi- Czech T, et al: 5-Aminolevulinic acid is a promising marker sualization of 5-aminolevulinic acid fluorescence in low-grade for detection of anaplastic foci in diffusely infiltrating gliomas gliomas. Clinical article. J Neurosurg 115:740–748, 2011 with nonsignificant contrast enhancement. Cancer 116:1545– 35. Sankar T, Delaney PM, Ryan RW, Eschbacher J, Abdelwahab 1552, 2010 M, Nakaji P, et al: Miniaturized handheld confocal microsco- 48. Wirth D, Snuderl M, Sheth S, Kwon CS, Frosch MP, Curry W, py for neurosurgery: results in an experimental glioblastoma et al: Identifying brain neoplasms using dye-enhanced mul- model. Neurosurgery 66:410–418, 2010 timodal confocal imaging. J Biomed Opt 17:026012, 2012 36. Sarin H, Kanevsky AS, Wu H, Brimacombe KR, Fung SH, 49. Xiao W, Yao N, Peng L, Liu R, Lam KS: Near-infrared opti- Sousa AA, et al: Effective transvascular delivery of nanopar- cal imaging in glioblastoma xenograft with ligand-targeting ticles across the blood-brain tumor barrier into malignant alpha 3 integrin. Eur J Nucl Med Mol Imaging 36:94–103, glioma cells. J Transl Med 6:80, 2008 2009 37. Schlosser HG, Suess O, Vajkoczy P, van Landeghem FK, Zeitz 50. Yan H, Wang J, Yi P, Lei H, Zhan C, Xie C, et al: Imaging M, Bojarski C: Confocal neurolasermicroscopy in human brain tumor by dendrimer-based optical/paramagnetic nano- brain – perspectives for neurosurgery on a cellular level (in- probe across the blood-brain barrier. Chem Commun (Camb) cluding additional comments to this article). Cent Eur Neu- 47:8130–8132, 2011 rosurg 71:13–19, 2010 51. Yan H, Wang L, Wang J, Weng X, Lei H, Wang X, et al: 38. Snuderl M, Wirth D, Sheth SA, Bourne SK, Kwon CS, An- Two-order targeted brain tumor imaging by using an optical/ cukiewicz M, et al: Dye-enhanced multimodal confocal im- paramagnetic nanoprobe across the blood brain barrier. ACS aging as a novel approach to intraoperative diagnosis of brain Nano 6:410–420, 2012 tumors. Brain Pathol 23:73–81, 2013 39. Stummer W, Pichlmeier U, Meinel T, Wiestler OD, Zanella F, Reulen HJ: Fluorescence-guided surgery with 5-aminolevu- Manuscript submitted October 15, 2013. linic acid for resection of malignant glioma: a randomised Accepted November 26, 2013. controlled multicentre phase III trial. Lancet Oncol 7:392– Please include this information when citing this paper: DOI: 401, 2006 10.3171/2013.11.FOCUS13484. 40. Tilgner J, Herr M, Ostertag C, Volk B: Validation of intraop- Address correspondence to: Peter Nakaji, M.D., Neuroscience erative diagnoses using smear preparations from stereotactic Publications, Barrow Neurological Institute, St. Joseph’s Hospital brain biopsies: intraoperative versus final diagnosis—influ- and Medical Center, 350 W. Thomas Rd., Phoenix, AZ 85013. ence of clinical factors. Neurosurgery 56:257–265, 2005 email: [email protected].

Neurosurg Focus / Volume 36 / February 2014 7

Unauthenticated | Downloaded 09/30/21 08:56 AM UTC