Optical Signatures Derived From Deep UV to NIR Excitation Discriminates Healthy Samples From Low and High Grades Glioma.

Among all the tumors of the central nervous system (CNS), glioma are the most deadly and the most malignant. Surgical resection is the standard therapeutic method to treat this type of brain cancer. But the diffusive character of these tumors create many problems for surgeons during the operation. In fact, these tumors migrate outside the tumor solid zone and invade the surrounding healthy tissues. These infiltrative tissues have the same visual appearance as healthy tissues, making it very difficult for surgeons to distinguish the healthy ones from the diffused ones. The surgeon, therefore, cannot properly remove the tumor margins increasing the recurrence risk of the tumor. To resolve this problem, our team has developed a multimodal two-photon fibered endomicroscope, compatible with the surgeon trocar, to better delimitate tumor boundaries by relying on the endogenous fluorescence of brain tissues. In this context, and in order to characterize the optical signature of glioma tumors, this study offers multimodal and multi-scaled optical measurements from healthy tissues to high grade glioma. We can interrogate tissue from deep ultra-violet to near infrared excitation by working with spectroscopy, fluorescence lifetime imaging, two-photon fluorescene imaging and Second Harmonic Generation (SHG) imaging. Optically derived ratios such as the Tryptophan/Collagen ratio, the optical redox ratio and the long lifetime intensity fraction, discriminated diseased tissue from its normal counterparts when fitted by Gaussian ellipsoids and choosing a threshold for each. Additionally two-photon fluorescence and SHG images were shown to display similar histological features as Hematoxylin-Eosin stained images.

Multimodal imaging to explore endogenous fluorescence of fresh and fixed human healthy and tumor brain tissues.

To complement a project toward label-free optical biopsy and enhanced resection which the overall goal is to develop a multimodal nonlinear endomicroscope, this multimodal approach aims to enhance the accuracy in classifying brain tissue into solid tumor, infiltration and normal tissue intraoperatively. Multiple optical measurements based on one- and two-photon spectral and lifetime autofluorescence, including second harmonic generation imaging, were acquired. As a prerequisite, studying the effect of the time of measurement postexcision on tissue's spectral/lifetime fluorescence properties was warranted, so spectral and lifetime fluorescences of fresh brain tissues were measured using a point-based linear endoscope. Additionally, a comparative study on tissue's optical properties obtained by multimodal nonlinear optical imaging microscope from fresh and fixed tissue was necessary to test whether clinical validation of the nonlinear endomicroscope is feasible by extracting optical signatures from fixed tissue rather than from freshly excised samples. The former is generally chosen for convenience. Results of this study suggest that an hour is necessary postexcision to have consistent fluorescence intensities\lifetimes. The fresh (a,b,c) vs fixed (d,e,f) tissue study indicates that while all optical signals differ after fixation. The characteristic features extracted from one- and two-photon excitation still discriminate normal brain (a,d) cortical tissue, glioblastoma (GBM) (b,e) and metastases (c,f).

Real-time Brain Tumor imaging with endogenous fluorophores: a diagnosis proof-of-concept study on fresh human samples.

The primary line of therapy for high-grade brain tumor is surgical resection, however, identifying tumor margins in vivo remains a major challenge. Despite the progress in computer-assisted imaging techniques, biopsy analysis remains the standard diagnostic tool when it comes to delineating tumor margins. Our group aims to answer this challenge by exploiting optical imaging of endogenous fluorescence in order to provide a reliable and reproducible diagnosis close to neuropathology. In this study, we first establish the ability of two-photon microscopy (TPM) to discriminate normal brain tissue from glioblastomas and brain metastasis using the endogenous fluorescence response of fresh human brain sample. Two-photon fluorescence images were compared to gold standard neuropathology. "Blind" diagnosis realized by a neuropathologist on a group of TPM images show a good sensitivity, 100%, and specificity, 50% to discriminate non tumoral brain tissue versus glioblastoma or brain metastasis. Quantitative analysis on spectral and fluorescence lifetime measurements resulted in building a scoring system to discriminate brain tissue samples.

The Impact of Compressed Femtosecond Laser Pulse Durations on Neuronal Tissue Used for Two-Photon Excitation Through an Endoscope.

Accurate intraoperative tumour margin assessment is a major challenge in neurooncology, where sparse tumours beyond the bulk tumour are left undetected under conventional resection. Non-linear optical imaging can diagnose tissue at the sub-micron level and provide functional label-free histopathology in vivo. For this reason, a non-linear endomicroscope is being developed to characterize brain tissue intraoperatively based on multiple endogenous optical contrasts such as spectrally- and temporally-resolved fluorescence. To produce highly sensitive optical signatures that are specific to a given tissue type, short femtosecond pulsed lasers are required for efficient two-photon excitation. Yet, the potential of causing bio-damage has not been studied on neuronal tissue. Therefore, as a prerequisite to clinically testing the non-linear endomicroscope in vivo, the effect of short laser pulse durations (40-340 fs) on ex vivo brain tissue was investigated by monitoring the intensity, the spectral, and the lifetime properties of endogenous fluorophores under 800 and 890 nm two-photon excitation using a bi-modal non-linear endoscope. These properties were also validated by imaging samples on a benchtop multiphoton microscope. Our results show that under a constant mean laser power, excitation pulses as short as 40 fs do not negatively alter the biochemical/ biophysical properties of tissue even for prolonged irradiation.

Two-photon optical imaging, spectral and fluorescence lifetime analysis to discriminate urothelial carcinoma grades.

In the framework of urologic oncology, mini-invasive procedures have increased in the last few decades particularly for urothelial carcinoma. One of the essential elements in the management of this disease is still the diagnosis, which strongly influences the choice of treatment. The histopathologic evaluation of the tumor grade is a keystone of diagnosis, and tumor characterization is not possible with just a macroscopic evaluation. Even today intraoperative evaluation remains difficult despite the emergence of new technologies which use exogenous fluorophore. This study assessed an optical multimodal technique based on endogenous fluorescence, combining qualitative and quantitative analysis, for the diagnostic of urothelial carcinoma. It was found that the combination of two-photon fluorescence, second harmonic generation microscopy, spectral analysis and fluorescence lifetime imaging were all able to discriminate tumor from healthy tissue, and to determine the grade of tumors. Spectral analysis of fluorescence intensity and the redox ratio used as quantitative evaluations showed statistical differences between low-grade and high-grade tumors. These results showed that multimodal optical analysis is a promising technology for the development of an optical fiber setup designed for an intraoperative diagnosis of urothelial carcinoma in the area of endo-urology.

Multimodal optical analysis discriminates freshly extracted human sample of gliomas, metastases and meningiomas from their appropriate controls.

Delineating tumor margins as accurately as possible is of primordial importance in surgical oncology: extent of resection is associated with survival but respect of healthy surrounding tissue is necessary for preserved quality of life. The real-time analysis of the endogeneous fluorescence signal of brain tissues is a promising tool for defining margins of brain tumors. The present study aims to demonstrate the feasibility of multimodal optical analysis to discriminate fresh samples of gliomas, metastases and meningiomas from their appropriate controls. Tumor samples were studied on an optical fibered endoscope using spectral and fluorescence lifetime analysis and then on a multimodal set-up for acquiring spectral, one and two-photon fluorescence images, second harmonic generation signals and two-photon fluorescence lifetime datasets. The obtained data allowed us to differentiate healthy samples from tumor samples. These results confirmed the possible clinical relevance of this real-time multimodal optical analysis. This technique can be easily applied to neurosurgical procedures for a better delineation of surgical margins.

Multimodal optical analysis of meningioma and comparison with histopathology.

Meningioma is the most frequent primary central nervous system tumor. The risk of recurrence and the prognosis are correlated with the extent of the resection that ideally encompasses the infiltrated dura mater and, if required, the infiltrated bone. No device can deliver real-time intraoperative histopathological information on the tumor environment to help the neurosurgeon to achieve a gross total removal. This study assessed the abilities of nonlinear microscopy to provide relevant and real-time data to help resection of meningiomas. Nine human meningioma samples (four World Health Organization Grade I, five Grade II) were analyzed using different optical modalities: spectral analysis and imaging, lifetime measurements, fluorescence lifetime imaging microscopy, fluorescence emitted under one- and two-photon excitation and the second-harmonic generation signal imaging using a multimodal setup. Nonlinear microscopy produced images close to histopathology as a gold standard. The second-harmonic generation signal delineated the collagen background and two-photon fluorescence underlined cell cytoplasm. The matching between fluorescence images and Hematoxylin and Eosin staining was possible in all cases. Grade I meningioma emitted less autofluorescence than Grade II meningioma and Grade II meningioma exhibited a distinct lifetime value. Autofluorescence was correlated with the proliferation rates and seemed to explain the observed differences between Grade I and II meningiomas. This preliminary multimodal study focused on human meningioma samples confirms the potential of tissue autofluorescence analysis and nonlinear microscopy in helping intraoperatively neurosurgeons to reach the actual boundaries of the tumor infiltration. Correspondence between H&E staining (top pictures) and the two-photon fluorescence imaging (bottom pictures).

ADAPTIVE OPTICS IMAGING OF FOVEAL SPARING IN GEOGRAPHIC ATROPHY SECONDARY TO AGE-RELATED MACULAR DEGENERATION.

BACKGROUND: To describe adaptive optics (AO) imaging of foveal sparing in geographic atrophy (GA) secondary to age-related macular degeneration. METHODS: Flood-illumination AO infrared (IR) fundus images were obtained in four consecutive patients with GA using an AO retinal camera (rtx1; Imagine Eyes). Adaptive optics IR images were overlaid with confocal scanning laser ophthalmoscope near-IR autofluorescence images to allow direct correlation of en face AO features with areas of foveal sparing. Adaptive optics appearance of GA and foveal sparing, preservation of functional photoreceptors, and cone densities in areas of foveal sparing were investigated. RESULTS: In 5 eyes of 4 patients (all female; mean age 74.2 ± 11.9 years), a total of 5 images, sized 4° × 4°, of foveal sparing visualized on confocal scanning laser ophthalmoscope near-IR autofluorescence were investigated by AO imaging. En face AO images revealed GA as regions of inhomogeneous hyperreflectivity with irregularly dispersed hyporeflective clumps. By direct comparison with adjacent regions of GA, foveal sparing appeared as well-demarcated areas of reduced reflectivity with less hyporeflective clumps (mean 14.2 vs. 3.2; P = 0.03). Of note, in these areas, en face AO IR images revealed cone photoreceptors as hyperreflective dots over the background reflectivity (mean cone density 3,271 ± 1,109 cones per square millimeter). Microperimetry demonstrated residual function in areas of foveal sparing detected by confocal scanning laser ophthalmoscope near-IR autofluorescence. CONCLUSION: Adaptive optics allows the appreciation of differences in reflectivity between regions of GA and foveal sparing. Preservation of functional cone photoreceptors was demonstrated on en face AO IR images in areas of foveal sparing detected by confocal scanning laser ophthalmoscope near-IR autofluorescence.

Appearance of medium-large drusen and reticular pseudodrusen on adaptive optics in age-related macular degeneration.

PURPOSE: To investigate the appearance of medium-large drusen and reticular pseudodrusen on adaptive optics (AO). METHODS: In 14 consecutive patients, AO infrared (IR) images were overlaid with confocal scanning-laser-ophthalmoscope IR reflectance images and IR-referenced spectral-domain optical coherence tomography. RESULTS: In eight eyes of six patients, a total of 19 images of medium-large drusen were investigated by AO imaging. En face AO revealed medium-large drusen as highly hyper-reflective round/oval lesions, always centred and/or surrounded by a continuous/discontinuous hyporeflectivity. Cone photoreceptors were detected overlying drusen, appearing either as continuous 'bright' hyper-reflective dots over a 'dark' hyporeflective background, or as continuous 'dark' hyporeflective dots over a 'bright' hyper-reflective background. In eight eyes from eight patients, a total of 14 images of pseudodrusen were investigated by AO imaging. En face AO revealed reticular pseudodrusen as isoreflective lesions, always surrounded by a continuous/discontinuous hyporeflectivity. Cone photoreceptors were detected overlying pseudodrusen as 'bright' hyper-reflective dots over either a hyporeflective or isoreflective background. No 'dark' hyporeflective dots were detected in eyes with reticular pseudodrusen only. Cone photoreceptors were counted on the border of the drusen and pseudodrusen, respectively, and in a visibly healthy zone in its absolute vicinity. A similar decrease in cone appearance was observed for drusen and pseudodrusen (15.7% vs 16.2%). CONCLUSIONS: AO allows differences in reflectivity between medium-large drusen and reticular pseudodrusen to be appreciated. The cone mosaics may be detected as continuous 'bright' hyper-reflective dots overlying/on the border of drusen and pseudodrusen deposits, and possibly as continuous 'dark' hyporeflective dots overlying drusen only.