This PDF file contains the front matter associated with SPIE Proceedings Volume 10862, including the Title Page, Copyright information, Table of Contents, and Author and Conference Committee lists.

Full-Field Optical Coherence Tomography (FFOCT) is an effective technique for tissue anatomy imaging, allowing cancer detection through the observation of disorders in the tissue microarchitecture. Moreover, with the temporal analysis of the FFOCT signal variations, a functional dimension is added, allowing the distinction of cell types through their intracellular activity. This complementary imaging mode, called Dynamic Cell Imaging (DCI), has shown the ability to identify normal cells, cancer cells and immune cells in different types of tissue such as breast, liver or lung. On samples ranging from cell cultures to entire tissue resections, DCI signals are recorded and analyzed in order to characterize the involved endogenous biomarker at the subcellular scale. Longitudinal studies of these samples over a few hours are performed under different environment perturbations intended to modify the cell metabolism. The potential of bimodal FFOCT is evaluated through a clinical study organized at the department of breast surgery of Peking University People’s Hospital in Beijing. More than 200 breast samples and lymph nodes are included. A part of the images will be directly compared to histology to identify reading criteria and build a reference atlas. The other part will be read on a blind study to measure the ability of cancer detection with respect to histology. To take full advantage of the FFOCT and the DCI information richness for faster cancer assessment during surgeries, a computer-aided diagnostic system based on Machine Learning, and more specifically Deep Learning, is investigated.

Intraoperative visualization of molecular processes delineated by fluorescence contrast agents offers the potential for increased surgical precision and better patient outcomes. To exploit fully the clinical potential for targeted fluorescence guidance, there is a growing need to develop high-resolution, quantitative imaging systems suitable for surgical use. Diffuse optical fluorescence tomography (DOFT) systems in pre-clinical and diagnostic imaging applications have demonstrated improvements in fluorescence quantification with the addition of a priori data from structural imaging modalities (e.g., MR, CT). Here, we investigate the use of a cone-beam CT (CBCT) surgical guidance system to generate spatial priors for intraoperative DOFT. Imaging and localization data is incorporated directly into a finite element method DOFT implementation (NIRFAST) at multiple stages. First, CBCT data from an intraoperative flat-panel C-arm is used to generate tetrahedral meshes. Second, optical tracking of laser and camera devices enables an adaptable non-contact DOFT approach to accommodate various anatomical sites and acquisition geometries. Finally, anatomical segmentations from CBCT are included in the optical reconstruction process using Laplacian-type regularization (“soft spatial priors”). Calibration results showed that light rays between the tissue surface and navigated optical devices were mapped with sub-millimeter accuracy. Liquid phantom experiments determined the improvements in quantification of fluorescence yield, with errors of 85% and <20% for no priors and spatial priors, respectively. CBCT-DOFT fusion in a VX2-tumor rabbit model delineated contrast enhancement using a dual CT/optical liposomal nanoparticle. These developments motivate future translation and evaluation in an ongoing CBCT-guided head and neck surgery patient study.

Monitoring tissue oxygenation is important in clinical applications such as breast surgery, Surgical reconstruction with free flaps is complex: a flap is taken from a healthy area of the body to be transposed into the damaged area. This operation requires microsurgical reconstruction of the vascular network. Tissue alterations may occur if the blood supply of the flap is not normal. This must be rapidly detected in order to avoid irreversible and extremely serious damages. We developed a Time-Resolved approach to get information in depth and separate the contribution of the upper layer from the lower layer of interest (flap). The TR system is detailed in [1-2]. We designed a specific optical stethoscope probe built from materials compliant with the clinical constraints. Distances between sources and detectors (SD) are 6, 8 and 13mm. We focused on the 750-800 and 850nm wavelengths. Abdominal flap were collected and buried on the left lateral abdominal muscle of the pig (~1cm deep of skin, fat and highly absorbing muscle) [3]. Either arterial or venous occlusions were performed (20 minutes of rest / 20 minutes of occlusion / 10-20 minutes release). US control of the upper layer and flap thicknesses has been systematically done. The two layers (surface layer and flap) were systematically controlled and monitored by two invasive LICOX (Integra laboratory) that measure the tissue pressure in Oxygen. We measured and analyzed 16 pigs (32 flaps) with arterial and venous occlusions; clinical control and LICOX were systematically performed. For each acquired data, we compared the results obtained in NIRS (integral of the TR signal) and in Resolved Time. Our results show that while the detection of occlusion depends on its depth in NIRS, we were able to detect occlusions whatever the sample (from the surface down to 1.23 mm) using the Resolved Time signals. In addition, the Oxy and Deoxy concentrations information provided by TR made it possible to identify the type of occlusion, which LICOX does not always allow. References: [1] Planat-Chrétien A., et al. Diffuse Optical Imaging V, H. Dehghani and P. Taroni, eds., Vol. 9538 of SPIE Proceedings (Optical Society of America, 2015), paper 953806 (2015). [2] Planat-Chrétien A., et al., Biomedical Optics Congress 2018, Clinical and Translational, OSA Technical Digest (Optical Society of America, 2018), paper JW3A.27 (2018). [3] Lartizien R, et al.,J Stomatol Oral Maxillofac Surg. 2017 Financial support ANR-15-CE19_0010 (Agence Nationale de Recherche, France).

Experimental setup geometry in Monte Carlo (MC) simulations is often simplified to shorten computation times. We investigate the effect of these simplifications on the accuracy of the spatial frequency domain (SFD) reflectance. We also introduce a new detection scheme in the MC method that eliminates the often overlooked errors arising from the Hankel transform of the spatially discretized reflectance profiles to SFD reflectance. Finally, we propose and evaluate an artificial neural network-based framework capable of estimating high-definition maps of optical properties in real-time.

Pattern generalization was proposed recently as an avenue to increase the acquisition speed of single-pixel imaging setups. This approach consists of designing some positive patterns that reproduce the target patterns with negative values through linear combinations. This avoids the typical burden of acquiring the positive and negative parts of each of the target patterns, which doubles the acquisition time. In this study, we consider the generalization of the Daubechies wavelet patterns and compare images reconstructed using our approach and using the regular splitting approach. Overall, the reduction in the number of illumination patterns should facilitate the implementation of compressive hyperspectral lifetime imaging for fluorescence-guided surgery.

Fluorescence imaging using near-infrared (NIR) fluorescent contrast agents is increasingly being investigated as intraoperative tool to visualize, in real-time, tissues of interest such as tumors, lymph nodes or nerve bundles. Generally, spectral imaging systems are used that measure the intensity of fluorescent signals. However, to aid in a more specific detection of these fluorescent signals, fluorescence lifetime can be added to the image. The lifetime is independent of the intensity and in addition, multiple tracers emitting around the same wavelengths can still be distinguished based on their difference in lifetime. Imaging lifetimes, however, requires a much more advanced imaging system. None of the currently approved fluorescence guidance systems support fluorescence lifetime and today’s available lifetime imaging technology (TCSPC, ICCD) does not allow imaging the sub-nanosecond lifetimes of NIR dyes with the efficiency needed to reach video frame rates. In this paper, we present a 32×32-pixel proof-of-concept camera based on our novel CAPS-pixel based gated image sensor. This camera is specifically targeted at imaging fluorescence lifetimes at NIR wavelengths with high efficiency for the use in fluorescence-guided surgery and is a first step towards a full camera with video resolution at video frame rates. We describe the camera system and how it is used to image fluorescence lifetimes. Next, we show the lifetime imaging capability by imaging the lifetimes of two different nanosecond visible dyes (fluorescein and acridine orange) in cuvette and two more challenging (sub-)nanosecond NIR dyes (ICG and IRDye800CW). Lastly, we validate the camera by imaging NIR fluorescence phantoms in a mouse.

Pituitary adenomas are benign tumours of the pituitary gland, a pea-sized organ situated behind the nose, attached to the base of the brain. During transsphenoidal endoscopic surgery to remove pituitary adenomas, it is important to distinguish normal pituitary tissue from adenoma, both to maximize the completeness of resection, and to minimize damage to healthy tissue, thus preserving endocrine function. Standard of care intraoperative white light imaging often displays low contrast between healthy pituitary tissue and adenoma. Multispectral imaging (MSI), which allows simultaneous collection of morphological (spatial) and biochemical (spectral) information, can help to more effectively delineate tissue types. This motivated the design and construction of a compact, clinically translatable endoscope capable of capturing multispectral images during transsphenoidal surgery. The multispectral endoscope employs a spectrally resolved detector array (SRDA) with 9 spectral filters (8 narrow bands; average FWHM 30nm, center wavelengths 553, 587, 629, 665, 714, 749, 791, 829nm; 1 broadband; 500–850nm). The SRDA was coupled to a standard clinically approved 4mm rigid endoscope, through which broadband (400–750nm) and narrowband (400–480nm) illumination were supplied sequentially for reflectance and fluorescence imaging respectively. Subjects due to undergo transsphenoidal surgery to resect pituitary adenoma were enrolled for experimental imaging in a pilot clinical study (MAPS). Images were captured before, during and after resection of the adenoma. Here, we present the results from these first-in-human tests, including evaluation of the image quality and classification potential of the multispectral image cubes.

Short-wave infrared imaging in tissue in the 1000-2000 nm range is characterized by reduced photon scatter and comparable or higher absorption compared to the NIR-I regime. These characteristics have implications for the performance of fluorescence molecular tomography (FMT) techniques, potentially improving the resolution of subsurface structure, possibly at the expense of depth sensitivity. To examine these questions, we have developed a SWIR small animal fluorescence tomography system. This instrument acquires multi-angle SWIR projection images of a stationary platform through a rotating gantry technique. These images are then processed for tomographic reconstruction of the SWIR fluorescence activity. Herein, we describe the development of this system and show multi-angle images from a mouse carcass containing a SWIR-specific fluorophore inclusion.

Fluorescence guidance systems are used for different indications, and the relevant range of concentrations and the relevant dye used varies considerably for human clinical trials. Most systems are designed for high concentration ICG imaging, where 0.1 mg/kg is injected IV and perfusion is imaged, yet other applications such as delayed uptake (or second temporal window) ICG imaging can have tissue concentrations an order of magnitude or more below this. Additionally, IRDye800 is used in both antibody imaging and small protein imaging in clinical trials, but injected doses vary from therapeutic to microdose levels. In this study, the sensitivity to both perfusion dose and microdose ranges were tested for ICG and IRDye800CW, using the Spy Elite system (Stryker) and the Solaris (Perkin Elmer). The sensitivity to ICG was significantly different and the sensitivity to IRDye800 was also different but with opposite choice of the optimal system. The concentrations tested ranged from 0.1 mg/kg down to 0.1 ug/kg. The signal to background limited the sensitivity in both cases, and the ambient light effects were significantly different in the two systems. The in vivo testing in lymphatic and vascular transport was assessed to determine limits to detection for vessel size.

Near Infrared Fluorescence (NIRF) molecular imaging is widely used in research and increasingly employed in the clinic for intraoperative imaging to assist in tumor resection and avoiding critical structures. However, emerging data indicate that it may also find use in early stage diagnosis for diseases such as breast cancer and rheumatoid arthritis. The potential of NIRF probes to provide both molecular and spatial information using non-ionizing radiation in a fast and inexpensive format for clinical diagnosis has significant advantages over blood tests, anatomical and physiological imaging modalities (e.g. MRI, ultrasound), and nuclear medicine. Recently, we have published papers showing oral delivery and subcutaneous injection of NIRF probes are feasible in animal models of breast cancer and rheumatoid arthritis. These routes of administration are less expensive and potentially safer than intravenous delivery, further lowering the barrier to implementation. Despite the potential of self-administered diagnostic probes, there are several challenges before this approach could be broadly applied in the clinic. The principal challenge is depth of imaging, limiting applications to disease sites near an accessible body surface. Additional challenges include the selection of biomarkers to improve diagnosis over current practice and identifying targeting ligands suitable for less-invasive delivery. In this work, we discuss the challenges associated with delivering NIRF molecular imaging agents by oral and subcutaneous injection for the early or subclinical diagnosis of rheumatoid arthritis. We outline a strategy for multiple probes that could provide complementary information and describe how optical advances could further improve the resolution for more accurate diagnosis.

Accidental nerve transection or injury is a significant morbidity associated with many surgical interventions, resulting in persistent postsurgical numbness, chronic pain, and/or paralysis. Nervesparing can be a difficult task due to patient-to-patient variability and the difficulty of nerve visualization in the operating room. Fluorescence image-guided surgery to aid in the precise visualization of vital nerve structures in real time during surgery could greatly improve patient outcomes. To date, all nerve-specific contrast agents emit in the visible range. Developing a nearinfrared (NIR) nerve-specific fluorophore is poised to be a challenging task, as a NIR fluorophore must have enough “double-bonds” to reach the NIR imaging window, contradicting the requirement that a nerve-specific agent must have a relatively low molecular weight to cross the blood-nervebarrier (BNB). Herein we report our efforts to investigate the molecular characteristics for the nervespecific oxazine fluorophores, as well as their structurally analogous rhodamine fluorophores. Specifically, optical properties, physicochemical properties and their in vivo nerve specificity were evaluated herein.

We reported the uptake value of a near-infrared imaging agent in an orthotopic glioma mouse models. The imaging agent was IRDy680®. The mice were injected with a trace amount of this imaging agent and imaged on a magnetic resonance imaging device that is coupled with a fluorescence imaging tomography system. By applying a reconstruction method, at each time point, we got a single value for the uptake of this imaging agent in the tumor region. The mean of the uptake values in the mice is reported here.

Identification of tumor margins in the operating room in real time is a critical challenge for surgical procedures that serve as cancer cure. Breast conserving surgery (BCS) is particularly affected by this problem, with current reexcision rates above 25%. Due to a lack of clinically available methodologies for detection of involved or close tumor margins, much effort is focused on developing intraoperative margin assessment modalities that can aid in addressing this unmet clinical need. BCS provides a unique opportunity to design contrast-based technologies that are able to assess tumor margins independent from the patient, providing a rapid pathway from bench to bedside at a much lower cost. Since resected tissue is removed from the patient’s blood supply, non-specific contrast agent uptake becomes a challenge due to the lack of clearance. Therefore, a dual probe, ratiometric fluorescence imaging approach was taken in an effort to reduce non-specific signal, and provide a modality that could demonstrate rapid, robust margin assessment on resected patient samples. Termed, dual-stain difference specimen imaging (DDSI), DDSI includes the use of spectrally unique, and fluorescently labeled target-specific, as well as non-specific biomarkers. In the present study, we have applied epidermal growth factor receptor (EGFR) targeted DDSI to tumor xenografts with variable EGFR expression levels using a previously optimized staining protocol, allowing for a quantitative assessment of the predictive power of the technique under biologically relevant conditions. Due to the presence of necrosis in the model tumors, ring analysis was employed to characterize diagnostic accuracy as measured by receiver operator characteristic (ROC) curve analysis. Our findings demonstrate the robust nature of the DDSI technique even in the presence of variable biomarker expression and spatial patterns. These results support the continued development of this technology as a robust diagnostic tool for tumor margin assessment in resected specimens during BCS.

In the pursuit of reducing re-excision rates in breast conserving surgery, a dual probe specimen staining technique has emerged as a promising approach to identify positive margins during surgery. This approach generally involves staining the tissue with a fluorescent dye targeted to a biomarker of interest, such as a cell surface receptor, and an untargeted counterpart, imaging both dyes and using the two images together to compensate for instrumentation inhomogeneities and non-specific uptake. A growing body of literature suggests that this approach can effectively discriminate tumor and normal tissue in gross fresh specimens in reasonable timeframes. However, the robustness of the staining protocol is still under investigation as all parameters have not been fully evaluated. In this paper, we examine the effect of staining temperature on diagnostic performance. Tumor (overexpressing EGFR) and normal fresh specimens were stained at room temperature or 37 °C and diagnostic performance compared using area under the curve (AUC) from receiver operator characteristic (ROC) analysis. The results suggest that the use of Licor IRDye800CW-labeled anti-EGFR antibody and Licor IRdye680RD-labeled control antibody as the probe pair is not significantly affected by staining temperature, in contrast to our experience with quantum-dot labeled antibodies. The robustness of the technique using these stains is reassuring and simplifies the staining protocol.

The discovery of new tumor targeting agents is desirable to expand imaging and drug delivery platforms. Cobalamins, vitamin B12 derivatives, selectively accumulate in tumor versus benign tissue due to overexpression of transcobalamin receptors in a variety of cancer types. Multiple forms of this vitamin are taken into cells via transport through transcobalamin receptors on the cell surface. Alkylcobalamins are light-activatable, and we have discovered that the wavelength of this light activation is tunable via appendage of a fluorophore. We have been able to harness this cobalamin platform to release drugs with a variety of wavelengths of light, including those within the optical window of tissue. This cobalamin drug delivery platform provides selective spatiotemporal activation of drug only where needed, thereby diminishing side effects of traditional chemotherapy. A Bodipy650-cobalamin was synthesized and utilized to study the tumor targeting ability of cobalamin derivatives in athymic nude mice with subcutaneous MCF-7 and MIA PaCa-2 tumors, which have been demonstrated to overexpress transcobalamin receptors. The fluorescently labeled cobalamin was injected intravenously into the mice and allowed to incubate for a series of time points. Fluorescence imaging revealed that this cobalamin conjugate selectively accumulated in both tumor types. We utilized this cobalamin platform for tumor selective, light-activated delivery of the pancreatic cancer drugs erlotinib and SN38. We determined light-induced apoptosis in MIA PaCa-2 cells in vitro and explored the reduction of MIA PaCa-2 tumors in vivo utilizing these cobalamin drug conjugates. This cobalamin platform provides potential for development of new theranostic tools for drug delivery.

In proof-of-concept studies, the anti-CEA M5A-IR800 conjugate demonstrated rapid and effective near infrared (NIR) imaging of human colon cancer and pancreatic cancer primary and metastatic lesions in mouse models. A limitation observed from these studies is the antibody-dye conjugate’s rapid clearance from the blood due to the increased hydrophobicity of the IR800 dye. This is a bottleneck for clinical applications, requiring high doses to be administered and a short surgical time window for intraoperative imaging. As a result, we developed a new prototype anti-CEA-swPEG-IR800 conjugate, that incorporates a PEGylated sidearm linker to shield or mask the IR800 dye’s hydrophobicity, a novel approach to extend the blood circulation half-life and in doing so increase tumor sensitivity as well as lower normal hepatic uptake. Results of the anti-CEA-swPEG-IR800 in an orthotopic human pancreatic cancer mouse model demonstrated exceptional optical imaging at lower doses, a much longer in vivo half-life enabling increased tumor fluorescence and higher tumor to background ratios. We propose that our novel anti-CEA-swPEG-IR800 is capable of enhanced optical imaging than currently available agents and will become the next generation optical imaging agent for safe and effective intraoperative image-guided surgery in CEA expressing GI cancers.

Improved methods to determine tumor localization and disease extension are critical factors in the management of pancreatic ductal adenocarcinoma (PDAC). Thus, contrast-enhanced fluorescence-guided detection of these lesions could improve the treatment of PDAC. A current limitation to fluorescence enhancement of PDAC is non-specific signal due to the clearance of the imaging dye, e.g. indocyanine green, like the liver, spleen or other local, intraperitoneal organs where metastatic spread may be present. We report surgical imaging agents that provide strong enhancement of PDAC compared to healthy tissue, but also have significantly reduced nonspecific background signal in clearance organs of the peritoneal cavity.

Fluorescent image-guided surgery has the potential to revolutionize surgery by providing direct tissue visualization using compact imaging systems and targeted contrast agents. Nerve targeting contrast agents are currently under development, where clinical use would decrease morbidity and benefit post-surgical outcomes for a variety of surgical procedures. Herein, we have applied a previously optimized direct administration methodology to resected human prostate specimens to validate the human nerve cross-reactivity of the nerve-specific fluorophore Oxazine 4. Nerves were clearly identified on the posterior lateral surfaces of stained prostate specimens, consistent with prostate neuroanatomy. Additionally, nerve signal-to-background ratios were consistent with the relevant postperfusion murine nerve models, suggesting that contrast levels will match results in murine models when Oxazine nerve-specific fluorophore is applied to human nerves in vivo. The positive human nerve staining demonstrated herein on resected prostate specimens using the optimized direct administration methodology provides a human nerve cross reactivity screening tool for future translational studies of near-infrared nerve specific contrast agents.

The presence of lymph node metastases played as a critical prognostic factor in breast cancer treatment and guiding the future adjuvant treatment. The possibility of missed micrometastases by conventional pathology was estimated around 20-60% cases has created a demand for the development of more accurate approaches. Here, a paired-agent imaging approach is presented that employs a control imaging agent to allow rapid, quantitative mapping of microscopic populations of tumor cells in lymph nodes to guide pathology sectioning. To test the feasibility of this approach to identify micrometastases, healthy rat and human lymph nodes were stained with targeted and control imaging agent solution to evaluate the potential for the agents to diffuse into and out of intact nodes. Erbitux, an EGFR specific antibody was labeled with IRDye-700DX(LICOR) as targeted agent and IRDye-800CW was labeled to rat IgG as control agent. Lymph nodes were stained for 60 min, followed by 30 min rinsing, and the uptake and washout of fluorescence were recorded. Subsequently, lymph nodes were frozen-sectioned and imaged under an 80- um resolution fluorescence imaging system (Pearl, LICOR) to confirm equivalence of spatial distribution of both agents in the entire node. Both imaging agents correlated well with each other(r=0.877) and the binding potential of targeted agent was found to be 0.08 ± 0.22 along the lymph node in the absence of binding. The results demonstrate this approach’s potential to enhance the sensitivity of lymph node pathology by detecting fewer than 1000 cell in a whole human lymph node.

Magnetic resonance imaging (MRI) of gadolinium (Gd)-based contrast agents plays a central role in managing the treatment of intracranial tumors. These images are involved in diagnosis, surgical planning, surgical navigation, and postoperative assessment of extent of resection. Replicating the information from Gd-MRI in the visual surgical field using fluorescent agents that behave similar to gadolinium in vivo would represent a major advance for surgical intervention of these tumors, and could provide robust compensation information to update pre-operative MRI images during surgery. In this paper, we examine the uptake of a Gd-based contrast agent in orthotopic tumor models and compare this behavior to two fluorescein-based contrast agents; specifically, clinical-grade sodium fluorescein (NaFl) and a 900 Da pegylated form of fluorescein. We show that the pegylated form of fluorescein is a more promising Gd-analog candidate.

Near-infrared fluorescence imaging is a promising intraoperative technique for real-time visualization of vital structures and tumor tissue during surgery. This manuscript describes the applications and limitations of NIR fluorescence imaging, and provides a general overview of two novel fluorescent agents and the process of clinical translation. The process of clinical translation of novel fluorescent agents is an essential part in the evolution of NIR fluorescence guided surgery. Treatments and surgeries are constantly advancing, which can occasionally cause challenges or difficulties for the surgeons. Poor visualization of tumors during surgery is one of the major challenges surgeons often face in oncologic patients, mainly due to the improved neo-adjuvant treatment patients receive. In these cases, NIR fluorescence imaging with the use of tumor-targeted fluorescent agents can play an essential role and help provide better results or outcomes. However, before this technique can be implemented in standard of care, optimal tumor-targeted fluorescent agents need to be developed and novel fluorescent agents need to undergo a successful process of clinical translation.

Introduction: Each year, nearly 100,000 patients proceed to the operating room for pulmonary resection, though identification of pulmonary nodules can frequently be challenging. We hypothesize that targeted intraoperative molecular imagining can improve identification of pulmonary nodules at the time of surgery. To test the safety of our novel targeted optical contrast agent, we have performed a Phase I trial. Methods: OTL38 is a near-infrared imaging agent that targets FRα, a receptor upregulated by 10,000-fold in 85-90% of patients with pulmonary adenocarcinoma. Twenty patients with a biopsy-proven lung adenocarcinoma were enrolled in a Phase I clinical trial. Prior to surgery, patients were systemically administered OTL38 (0.0025mg/kg) by intravenous infusion. During surgery, tumors were imaged in situ and ex vivo. Tumor fluorescence was quantified using tumor-to-background ratio (TBR). Results: In human patients receiving OTL38 prior to resection, we observed only minor Grade 1 toxcities: itching and a rash. We identified 18 nodules (90%) in 20 patients, and the mean tumor size was 2.5 cm (range 0.5-10.5cm). 22% of the fluorescent nodules measured less than 1cm. Mean TBR of fluorescent tumors was 3.2 (range 1.7-4.6). Tumor size did not correlate with TBR (p>0.05). In 2 patients, intraoperative imaging identified synchronous subcentimeter (5mm, 9mm) nodules which were not detected by pre-operative CAT or PET scanning. Conclusion: Our Phase I clinical trial showed targeted molecular imaging with OTL38 is safe and has only minor Grade I toxicities. In addition, this study showed that the optical contrast agent is capable of detecting subcentimeter pulmonary nodules in humans. Our group is currently conducting a multicenter, Phase II study to better understand the implications of intraoperative molecular imaging using OTL38.

Introduction: Glioblastoma (GBM) is the most common and devastating primary brain tumor. The recurrence rate remains high with a median survival of 15 months. GBM’s infiltrative nature results in ill-defined margins that makes maximal tumor resection with minimal morbidity a challenge. Epidermal growth factor receptor (EGFR) is the most frequently amplified gene in GBM (35-45% of tumors) and is associated with overexpression in about 40-98% of cases, a characteristic of more aggressive phenotypes. We hypothesize that fluorescence labeled anti-EGFR monoclonal antibodies (mAb), panitumumab-IRDye800 (pan800) and cetuximab-IRDye800 (cet800), could be leveraged to enhance tumor contrast during surgical resection and improve patient outcome. Methods: 50mg fluorescently labeled corresponding study drugs, pan800 and cet800 respectively, were administered 1-2 days in glioblastoma patients with contrast enhancing (CE) tumors prior to surgery following 100 mg loading dose of unlabeled cetuximab or panitumumab. Near-infrared fluorescence imaging of tumor and histologically negative peri-tumoral tissue was performed intraoperatively and ex vivo. Fluorescence was measured as mean fluorescence intensity (MFI), and tumor-to-background ratios (TBRs) were calculated by comparing MFIs of tumor and histologically uninvolved tissue. Results: Despite heterogeneous drug uptake across all resected brain tissues, mean fluorescence intensity (MFI) correlated strongly (R^2=0.97) with tumor volume among histologically confirmed tumor tissues. The smallest detectable tumor size in a closed-field setting was 4.2 x 2.7 mm^2 (8.2 mg) for pan800 and 8.5 x 6.6 mm^2 (70mg) for cet800. Tumor tissues from pan800 infusion had significantly higher mean TBR (8.1 ± 4.6) than cet800 infused ones in intraoperative imaging (3.3 ± 2.7; P = 0.004). NIR fluorescence from both test drugs provided high contrast to identify as few as a cluster of (5 ± 1) tumor cells in macroscopic imaging of whole sections of paraffin embedded tissues. Sensitivity and specificity of MFI for viable tumor detection was calculated and fluorescence was found to be highly sensitive (64.4% for pan800, 73.0% for cet800) and specific (98.0% for pan800, 66.3% for cet800) for viable tumor tissue while normal peri-tumoral tissue showed minimal fluorescence. No related grade-2 adverse events were observed 30 days beyond the infusion of either study drugs. Conclusion: EGFR antibody based imaging for contrast-enhanced glioblastomas proved safe in human patients and specific intratumoral delivery of NIR fluorescence provided high optical contrast and resolution for intraoperative image-guided resection. Fully humanized panitumumab-IRDye800 demonstrated superior detection sensitivity and tumor specificity over the chimeric cetuximab-IRDye800.

Many tumors for which fluorescence guided surgery (FGS) has been developed are surface tumors, where direct visualization by the surgeon is straightforward. On the other hand, cancers such as soft-tissue sarcomas, are present at a subsurface level. Resection of these sub-surface tumors is performed using ‘wide local excision’ where a single, complete mass is removed with an intact zone of normal tissue (~ 1 cm ‘margin’). We used a phantom model for sarcoma with near infrared fluorophore IRDye800 CW that defined different tissue properties. We compare the detection sensitivity of two commercially available near infrared (NIR) surgical imaging systems, Solaris (Perkin Elmer) and SPY PHI (Novadaq) using the phantom models of sarcoma. We also determine targeted fluorescence signal on both systems for blinded surgical phantom dissection by a surgeon. The fluorescence intensities are higher for Solaris than for SPY-PHI. On average, the fluorescence increased with an increase in intralipid concentration and decreased with an increase in blood concentration. The depth of imaging was higher for Solaris than for SPY PHI. Using the target values, the surgeon successfully dissected all phantoms using Solaris. Using fat phantoms for SPY PHI, the surgeon cut through four out of the total. Further improvement in FGS will improve cancer recurrence and morbidity.

Introduction: Distinguishing glioblastoma multiforme (GBM) tumor cells from normal brain remains a significant clinical challenge that limits the efficacy of treatment planning and resection of GBM. Developing agents that specifically target GBM for both non-invasive and intraoperative imaging is an attractive strategy to guide maximal safe resection and other therapy strategies to improve the prognosis for patients with GBM. Matrix metalloproteinase (MMP)-14 is a membrane-bound collagenase that is overexpressed in GBM with negligible expression in normal brain, presenting MMP-14 as an attractive biomarker for imaging GBM. These studies explored the utility of a novel peptide probe containing an MMP-14 binding sequence, an MMP-14-activatable near infrared fluorescence (NIRF) reporter, and a chelate for labeling with positron emission tomography (PET) radionuclides for dual-modality imaging of GBM in preclinical models. Methods and Results: Immunofluorescence, western blot, and gel zymography studies showed varying in vitro expression and activity of MMP-14 in D54, U87, and U251 GBM cell lines. At 24 h after i.v. injection of the peptide in athymic nude mice bearing s.c. xenografts of the GBM cell lines, NIRF signals in the tumors correlated with the cells’ in vitro expression of MMP-14. At 4 h after i.v. injection of the 64Cu-labeled peptide in mice bearing orthotopic patient derived xenograft (PDX) GBM tumors, PET/CT imaging showed accumulation in the intracranial tumors that was significantly reduced (p<0.05) by co-injecting an excess of the non-labeled peptide as a blocking agent. There was a linear correlation (p<0.05) between in vivo PET and ex vivo NIRF signals in the PDX tumor regions. Microscopic immunohistochemistry showed co-localization of MMP-14 expression and NIRF signal in PDX tumors. Conclusions: The novel peptide probes successfully targeted MMP-14 for imaging GBM models in mice, warranting continued development of the probes for image-guided resection of GBM in future preclinical studies.

Molecular image-guided surgery has the potential for translating the tools of molecular pathology to real-time guidance in surgery. As a whole, there are incredibly positive indicators of growth, including the first US FDA clearance of an enzyme-biosynthetic-activated probe for surgery guidance, and a growing number of companies producing agents and imaging systems. The strengths and opportunities must be continued but are hampered by important weaknesses and threats within the field. The ultimate potential may require multiple probes, as are used in molecular pathology, and a combination with ultra-high resolution imaging and image recognition systems which capture the data at a finer granularity than is possible by the surgeon. size-fits-all’ concept, similar to metabolic aberrations as exploited in FDG-PET (i.e. Warburg effect) or tumor acidity. Finally, methods to approach the problem of production cost minimization and regulatory approvals in a manner consistent with the potential revenue of the field will be important. In this area, some solid steps have been demonstrated in the use of fluorescent labeling commercial antibodies and separately in microdosing studies with small molecules.

Near-infrared (NIR) fluorescence combined with targeting epidermal growth factor receptor (EGFR) overexpression for surgical guidance in many cancers is gaining momentum. ABY-029 is an anti-EGFR Affibody molecule conjugated to IRDye 800CW that is FDA approved as an exploratory Investigational New Drug (eIND 122681). ABY-029 has a short plasma half-life (~15-20 minutes), which allows for administration of the imaging agent and excision surgery to occur on the same day unlike fluorescent antibodies. This fast tissue clearance may provide the means necessary to achieve clinically relevant tumor-to-normal tissue contrast levels using microdosing administration schemes. Pre-clinical studies have indicated that tumor-to-normal tissue contrast peaks between 4-8 hours depending on EGFR expression. Additionally, the No Observable Adverse Effect Level (NOAEL) was determined in pre-clinical toxicity studies to be 1,000X the microdose, or 30 micromole, whereas mild adverse events are common in antibody imaging studies. A number of promising first-in-human clinical Phase 0 trial microdose evaluation of ABY-029 have been initiated for recurrent glioma, soft-tissue sarcoma, and head and neck cancers (NCT0290925, NCT03154411, and NCT03282461, respectively). Here, we provide an update on our experience using ABY-029 for surgical resection at microdose levels and describe tissue contrast and correlation of ABY-029 fluorescence to EGFR tissue expression. Current progress indicates that moderate TBRs (~3-5) are observable at the microdose administration level but increased administration doses (3X and 6X microdoses) are being investigated. In addition, ex vivo tissue fluorescence is highly correlated to EGFR staining in both intensity and spatial localization.

The most common complication of autologous breast reconstructions using a deep inferior epigastric artery (DIEP) flap after mastectomy is fat necrosis due to ischemia. DIEP flap perfusion assessment can be optimized by using nearinfrared fluorescence (NIRF) imaging with indocyanine green (ICG). Several studies demonstrated diminished incidence of fat necrosis, partial flap loss and other complications. Studies with large patient numbers and standardized outcomes are lacking. To determine the correlation of incidence of fat necrosis and NIRF imaging with ICG we performed: 1) a review to evaluate relevant literature, and 2) a pilot study of NIRF imaging. With the aim to decide whether a randomized controlled trial (RCT) should be started. Seven studies were included in the review, showing NIRF imaging was feasible and incidence of fat necrosis was diminished. However, most studies were retrospective and with small patient numbers. In the pilot study, 18 flaps, assessed with NIRF imaging were prospectively included and compared to 18 retrospectively included flaps solely assessed based on clinical findings. There were no significant differences in patient and surgery characteristics. Incidence of complications and fat necrosis decreased from 30% to 6% (p-value: 0.07) in the NIRF imaging group. No strong conclusions can be drawn from the pilot study given the low patient number, neither from the reviewed studies. But based on the results a multicenter RCT would be recommended to determine the actual value of NIRF imaging for perfusion assessment of DIEP flaps. An RCT could also aid in wider implementation of this accessible technique in a standardized matter.

Chemo- and/or radiotherapy remain treatment standards for cancer. Clinical challenges including poor tumour specificity and cause dose-dependent side effects. The proposed new treatment modality combining radiation and radiation-triggered nanoconstructs can selectively target and efficiently destroy deep-seated tumours. To achieve this, we combined existing clinical techniques used in cancer treatment–X-ray radiation and other treatment modalities including chemotherapy and photodynamic therapy by using various nanoparticles which we successfully developed. They included scintillator nanoparticles (CeF3), Au nanoparticles, Poly Lactic-co-Glycolic Acid (PLGA) polymer and liposome nanoparticles. A photosensitive molecule, verteporfin was either conjugated (in the case of CeF3) or loaded inside the nanoparticles (in the case of PLGA polymer and liposomes). Their unique combination enables effective ablation of cancer cells in vitro and control of tumour growth in vivo at a low X-ray dose, which was conclusively demonstrated by our recent work.

Firstly, gold nanosphere and 5-aminolevulinic acid (ALA) conjugant was used to study the photodynamic therapy efficiency and its mechanism. We found that the conjugant can improve the cell killing efficiency that mainly attribute to the carriers function of the gold nanoparticles. In order to shift the absorption peak to red light that offer deeper penetration of tissues, gold nanorod and Hematoporphyrin monomethyl Ether (HMME) was conjugated to employ photodynamic therapy of KB cells. During the research, the 808 nm laser light and xenon lamp were used to irradiate the sample, which offered better therapy efficiency than gold nanosphere. Since TiO2 can be used to effectively generate reactive oxygen species (ROS) for photodynamic application with the absorption in the ultraviolet range without oxygen, TiO2 nanoparticles (NPs) are sensitized by linking with the photosensitizer, HMME, to form HMME-TiO2 nanocomposites (NCs) for demonstrating the photodynamic effects under the illumination of white light. The HMME-TiO2 NCs of different composition ratios are prepared for maximizing the generation of ROS and optimizing the inactivation effect of KB cells. The material characteristics and the ROS generation capability of the HMME-TiO2 NCs with the optimized combination ratio show their merits in a photodynamic process under white light irradiation. The application of such NCs to KB cell experiments results in a higher inactivation efficiency when compared to pure HMME of the same concentration. In general, different nanoparticles can improve PDT efficiency with different increasement, mechanism, advantage and disadvantage, we should choose different nonaparticle according to different applications.

Cherenkov emission, which is generated during radiation therapy, can be utilized for imaging that is synergistic with radiation therapy. Cherenkov light can be utilized to excite phosphors, which then can be imaged utilizing Cherenkov excited luminescence scanned imaging (CELSI). Europium chelate microspheres, which exhibit bright luminescence with long luminescent lifetime, were appended with multiple copies of cetuximab. This will allow for selective imaging of EGFR overexpressing tumors during the course of radiation therapy via CELSI. We have characterized the functionality of the cetuximab loaded microspheres in vitro via ELISA, as well as via fluorescence microscopy in EGFR overexpressing A431 cells. These microspheres were intravenously injected into athymic nude mice bearing A431 flank tumors and allowed to incubate for a series of time points. They were then imaged first via standard fluorescence imaging to determine the ideal time point for visualizing tumors via CELSI. After demonstrating selective accumulation in tumors, imaging was then undertaken in vivo via CELSI. These antibody conjugated europium microspheres provide promise to image tumors selectively with CELSI. Future studies involve conjugating other antibodies to the europium microspheres to utilize in CELSI.

X-ray luminescence computed tomography (XLCT) and X-ray fluorescence computed tomography (XFCT) are two emerging technologies in X-ray imaging. In these modalities, images are formed through detection of secondary emissions (light in XLCT, or secondary X-rays in XFCT) following X-ray excitations. XLCT and XFCT enable us to leverage the widely used X-ray imaging for simultaneous in vivo molecular and functional imaging. Depending on the geometry of the excitation X-ray beam (pencil-, fan-, and cone-beam or coded apertures), optimal tradeoff between imaging efficiency and spatial resolution can be achieved. The novel imaging principles of XLCT/XFCT make it possible to achieve a spatial resolution comparable to that of anatomical X-ray imaging. Here, we summarize our studies in this area in the past decade and discuss their prospects.

Cherenkov-excited luminescence scanned imaging (CELSI) is achieved with External Beam Radiotherapy, to map out molecular luminescence intensity or lifetime in tissue. In order to realize a deeper imaging depth with a reasonable spatial resolution, we optimized the original scanning gesture to do in a similar way to computed tomography (CT) and the image reconstruction was instead used a customized Maximum-likelihood expectation maximization (ML-EM) for CELSI. In tomographic CELSI (TCELSI), tomographic images are generated by irradiating the subject using a sequence of programmed X-ray beams at a fixed projection angle, while sensitive measurement is to take a sum for all image pixels from an intensified charge-coupled device. By restricting the X-ray excitation to a single, narrow beam of radiation, the origin of the optical photons can be inferred regardless of where these photons were detected, and how many times they scattered in tissue. Measurement geometry was designed for clinical expectation: CT scanning was achieved by a clinical linear accelerator (LINAC), where X-ray beam sequence and multiple projections were realized with multi leaf collimator (MLC) and gantry movement, respectively. Furthermore, in most modern External Beam Radiotherapy, MLC movement is synchronized with gantry angle to release a uniform radiation, and some of treatment plans, e.g., Intensity Modulated Radiation Therapy (IMRT), have a potential to match the scanning way mentioned. By including Cherenkov imaging results, medium surface profile can be additionally acquired, which can be used as boundary reference to do depth correction and co-register with molecular images. Resolution phantom studies showed that a 0.3 mm diameter capillary tube containing 0.01 nM luminescent nanospheres could be recognized at a depth of 21 mm into tissue-like media. Small animal imaging with a 1 mm diameter cylindrical target demonstrated that fast 3D data acquisition was achieved by a multi-pinhole collimator to image local luminescence 20mm deep.

During radiotherapy, X-ray beams induce Cherenkov light emission in tissue as part of the dose delivery. This light can be used for dosimetry, in order to track and image the dose as it happens. The Cherenkov light levels are in the range of 10⁻⁶ to 10⁻⁹ W∕cm², which makes it challenging to detect in a clinical environment. However, because the radiation is pulsed in 4 microsecond bursts, time-gated acquisition of the signal allows for robust detection, even in the presence of ambient room lighting. Thus, imaging sensors for this application must be highly sensitive and must be able to time gate faster than a microsecond. In this study, the use of a solid-state detector composed of 64×32 single photon avalanche diodes (SPADs) was examined. The advantages of this technology were intra-chip amplification, high dynamic range, superior X-ray noise rejection and fast temporal gating of the acquisition. The results show that the SPAD camera was sensitive enough to detect Cherenkov radiation despite the 3% fill factor. 2D oversampling (×25) was also used to increase final image resolution to 320×160. In this work we demonstrate the SPAD camera performance in imaging Cherenkov emission from a tissue optical phantom and one patient undergoing radiotherapy. The results show that the SPAD camera was sensitive enough to detect Cherenkov radiation emitted from patient’s surface with signal-to-noise ratio of 14 after 6s acquisition. The SPAD camera sensors could be a viable alternative for Cherenkov imaging, as compared to current imaging methods that are mostly focused around image intensifier-based cameras and so have a range of non-linearities and instabilities which could be solved by an all solid-state camera sensor.

Background: Cherenkov luminescence (CL) has been used in the field of biomedical imaging since 2009 and has attracted more and more attentions in recent years. However, the weak signal intensity and the spectrum distribution whose energy mainly located at 400-500 nm limited the widespread application of CLI. In this study, a novel nanoparticle which contained europium atoms doped into gadolinium oxide (GdO: Eu) was used to be excited by radiopharmaceuticals to convert the weak blue light into stronger long-wavelength fluorescence for improving imaging performance. Methods: First, both EO nanoparticles and GdO: Eu nanoparticles of the same mass in two Eppendorf (EP) tubes were excited simultaneously by 18F-FDG with the activity of 466 µCi and and the emitted fluorescence intensity of two EP tubes ßwere compared. Next, a series of in vitro experiments were desinged and conducted to investigate the cause of production of the long wavelength fluorescence of GdO: Eu under radiopharmaceutical excitation. Lastly, the tumor mouse models (n=6) were constructed, injected with the novel nanoparticles through the tail vein, and received the in vivo fluorescence imaging. Results: Compared with EO, the novel nanoparticle had a better performance in emitted fluorescence intensity. The fluorescence intensity increased with the decreasing distance between the nanoparticle and the radiopharmaceutical, or the increasing activity of 18F-FDG. Conclusion: The results showed that gadolinium oxide nanoparticle doped with europium atoms can convert the short-wavelength Cerenkov light of 18F-FDG into a long-wavelength light with better performance for in vivo imaging compared with CLI. This imaging strategy showed great potential for tumor imaging and detection.

In this study, we use Monte Carlo modelling to investigate the effect of tissue optical properties on Cherenkov emission detected from tissue surface. MC simulations are performed for wavelength between 400-1000nm and the values of absorption coefficient at each wavelength are determined based on the molar extinction coefficients of oxy- and deoxy-hemoglobin, with varying total haemoglobin concentration and tissue oxygen saturation of 70%. Tissue reduced scattering coefficient is approximated using μ_s (λ) = Aλ^-0.838. A range of clinically relevant tissue optical properties was investigated, with absorption coefficient between 0.1 and 1 cm^-1 and reduced scattering coefficient between 5 and 40 cm^-1 at 665nm. The angular distribution, depth of origins and the effect of tissue optical properties on Cherenkov emission on tissue surface are evaluated.

Solid tumors often exhibit abnormal morphology which can be characterized by increased permeability and low perfusion. The resulting tumor hypoxia has been correlated with poor prognosis, which may be due to ineffective therapy or survival of more aggressive phenotypes. External beam radiation therapy (EBRT) is often used to treat such tumors, where radiation dose is delivered on a daily fractionated basis over the course of weeks. A non-contact optical method for measuring in vivo oxygen levels during EBRT treatments has been developed to provide early indications of hypoxic tumor environments. This method uses a time-gated intensified imaging device to measure both Cherenkov emissions, which are generated in tissue by high energy electrons traveling faster than the phase-velocity of the medium, and Cherenkov-excited luminescence generated by the oxygen-sensitive phosphorescent compound, PtG4. Murine models have shown the ability to discriminate phosphorescence lifetime changes before and after animal sacrifice. Pixel-maps of the estimated pO2 can be generated from this data to show high spatial variability within a region of interest. By further camera optimization, this method can be expanded to show pO2 distributions for other physiological conditions in near real-time. Our imaging method has the unique ability to be integrated within existing clinical applications while providing a wide-field mapping of oxygen saturation, which is currently unavailable with existing point probes.

Fluorescence-guided brain tumour resection, notably using 5-aminolevulinic acid (ALA)-induced protoporphyrin IX (PpIX) for high-grade gliomas, has been demonstrated to provide better tissue differentiation, thereby improving patient outcomes when compared to white-light guidance. Novel fluorescence imaging devices aiming to increase detection specificity and sensitivity and targeting applications beyond high-grade gliomas are typically assessed by measurements using tissue-mimicking optical phantoms. The field currently lacks adequate phantoms with well-characterised tuneable optical properties. In this study, we developed soft tissue-mimicking fluorescence phantoms (TMFP) highly suitable for this purpose. We investigated: 1) the ability to independently tune optical and fluorescent properties; 2) the stability of the fluorescence signal over time; and 3) the potential of the proposed phantoms for imaging device validation. The TMFP is based on gel-wax which is an optically transparent mineral-oil based soft material. We embedded TiO2 as scattering material, carbon black oil-paint as background absorber, and CdTe Quantum Dots (QDs) as fluorophore because of its similar fluorescence spectrum to PpIX. Scattering and absorption properties were measured by a spectrophotometer, while the fluorescence was assessed by a wide-field fluorescence imaging system (WFFI) and a spectrometer. We demonstrated that: 1) the addition of QDs didn’t alter the phantom’s scattering which was only defined by the concentration of TiO2, whereas its absorption was defined by both QDs and colour oil paint; 2) the measured fluorescence intensity was linearlyproportional to the concentration of QDs; 3) the fluorescence intensity was stable over time (up to eight months); and 4) the fluorescence signal measured by the WFFI were strongly correlated to spectrometer measurements.

Breast cancer patients that experience complete removal of the primary tumor, or negative surgical margins (NSMs), benefit from decreased rates of local recurrence and increased survival. However, intraoperative margin detection is limited to visualization, palpation, and experience to identify malignant vs. healthy tissue. As a result, roughly 1/3 of patients treated with breast conserving surgery (BCS) have residual cancer cells left at the resection border, or positive surgical margins (PSMs). Fluorescence image-guided surgery (FIGS) is a promising alternative for intraoperative margin detection, providing surgeons with real-time feedback on tumor location, increasing the likelihood of achieving NSMs. Our past work has demonstrated that the use of self-assembled hyaluronic acid (HA) nanoparticles improves the delivery of indocyanine green (ICG) to breast tumors, enhancing intraoperative tumor signal and contrast. This study built upon these findings by assessing the surgical efficacy of ICG-loaded HA nanoparticles (NanoICG) for the image-guided resection of orthotopic iRFP+/luciferase+ 4T1 breast tumors in BALB/c mice. Tumors were resected with FIGS in mice treated with ICG or NanoICG and compared to bright light surgery (BLS) or sham controls. Tumor growth and recurrence were monitored with bioluminescence imaging. NanoICG improved complete resection and prolonged tumorfree survival. Additionally, NanoICG provided greater intraoperative contrast in malignant tissue than ICG or BLS. Furthermore, NanoICG demonstrated a greater ability to identify small, occult lesions than ICG. Overall, the use of NanoICG for the fluorescence image-guided resection of breast tumors could potentially decrease PSM rates and improve complete tumor removal.

Luminescence molecular tomography with Cherenkov excitation offers the ability to non-invasively image and quantify temporal changes in fluorescence throughout the body, and then further realize tumor localization. This can be done in radiotherapy to determine the response to treatment in fractionated therapy. To obtain high signal-to-background or signal-to-noise ratio measurement, it is critical to know the best post time point of in-vivo agent-based molecular imaging, which could account on a high signal ratio of target to skin (TSR). For this purpose, ex-vivo murine experiments were performed to quantify the biokinetics and biodistribution of the major organs, plasma, tumor, and skin.