Intraoperative nerve tissue fluorescence imaging

Abstract

Iatrogenic nerve injury represents one of the most feared surgical complications and remains a major morbidity across many surgical specialties. Currently, no clinically approved technique can directly enhance intraoperative nerve visualization, where intraoperative nerve identification continues to challenge even experienced surgeons. Fluorescence-guided surgery (FGS) has been successfully integrated into clinical medicine to improve safety and efficacy in the surgical arena. A number of tissue- and disease-specific contrast agents are in the clinical translation pipeline for future FGS integration. Within this context, a diverse repertoire of fluorescent tracers have been developed to improve surgeons’ intraoperative vision. This review aims to convey the recent developments for nerve-specific FGS and its potential for clinical translation.

INTRODUCTION

Current statistics estimate that worldwide surgical volume annually exceeds 300 million procedures, illustrating the prominent role of surgical intervention in clinical medicine.^{1, 2} The ultimate goal of surgery is to repair or remove damaged and diseased tissues, while minimizing comorbidities by preserving vital structures such as nerves and blood vessels as an important secondary aim. However, iatrogenic nerve damage, resulting in loss of sensation or function as well as pain, is common and continues to plague otherwise essential interventions. Up to 8% of all patients experience iatrogenic nerve injuries, accounting for an annual 25 million peripheral nerve traumas globally.^3–6

Alleviating the consequences of nerve damage remains challenging in the vast majority of cases, where functional improvements following nerve injury are slow, even with medical intervention.⁷ Permanent disability, marked by diminished or lost motor/sensory functions, and development of chronic neuropathy, often ensues, limiting patient quality of life, employment prospects, and capacity to participate in daily activities. Currently, a surgeon’s means for intraoperative nerve detection are limited, but include electromyographic (EMG) monitoring and ultrasound, which both lack contrast, specificity, and sensitivity. Although EMG has been utilized to assist in the identification of nerve tissues during surgery, the implementation of real-time, nerve-specific mapping could offer substantial benefits in preventing iatrogenic nerve injuries.

Fluorescence-guided surgery (FGS) has emerged as a rapidly-expanding technology that provides intraoperative guidance for real-time visualization of the tissue of interest with high spatial resolution using targeted probes. Currently, most clinical intraoperative FGS vision systems are designed to detect the two FDA-approved near-infrared (NIR, 650–900 nm) contrast agents (i.e., 700-nm methylene blue and 800-nm indocyanine green, Figure 1a, b). Importantly, NIR imaging offers advantages such as minimal autofluorescence, photon scattering and absorption compared to the visible spectrum, enabling up to a 1 cm photon penetration depth in tissue (Figure 1c).^8–17 For the last two decades, tremendous efforts have been put forth in the development of nerve-specific probes (Figure 1d). This review focuses on the needs and progress in fluorescent probe development for intraoperative nerve tissue identification and visualization, as well as the latest efforts towards clinical translation of nerve-specific FGS technology.

Figure 1. Photophysical property consideration for the development of nerve-specific fluorescent contrast agents.

a Absorption spectra of hemoglobin and water as a function of wavelength motivating in vivo fluorescence imaging in the NIR window. Adapted from Ref. [8]. Copyright 2001, Springer Nature. b Chemical structures and photophysical properties of the two FDA-approved NIR fluorophores, methylene blue and indocyanine green. c Wavelength-dependent autofluorescence illustration and comparison of vital organs and bodily fluids showcase the benefits of NIR fluorescence for in vivo imaging. Adapted from Ref. [9]. Copyright 2003, Elsevier. d Summary of contrast agents developed for nerve-specific fluorescence imaging.

Iatrogenic nerve injury

A systematic review of reported, medically-induced nerve injuries reveal incidence rates are stratified according to anatomic location of the nerve, the surgical procedure, or even the specific method of dissection or approach. Iatrogenic nerve damage occurs through various insults, including direct transection/crushing, mechanical tension through tissue retraction, thermal damage from electrosurgical devices, or ligation/entrapment with suture/screws/tacks.⁶ Notably, certain surgical procedures are more prone to postoperative neuropathic pain, reportedly occurring in 20% to 60% of patients after thoracotomy, mastectomy and limb amputation procedures.^18–20 Furthermore, the delicate cavernous nerves responsible for continence and potency in men are so small, that they prove difficult to recognize even under surgical camera magnification during prostatectomy. Difficulty in sparing prostatic nerves leads to nerve damage in up to 70% of patients postoperatively.^21–23 Avoidance of another important peripheral structure, the ilioinguinal nerve, during routine inguinal hernia repair involves navigation around the nerve-rich anatomic landmark colloquially referred to as the “Triangle of Pain”, where surgical complications can result in comorbidities similar to those following radical prostatectomy, including chronic post-herniorraphy inguinal pain with a 10–12% incidence rate.²⁴ These statistics illustrate the continued need for improved intraoperative nerve identification and visualization.

Protein-based fluorescent contrast agents

The use of immunofluorescence to detect and visualize antigens in cells and tissue samples is a well-established microscopic technique. To improve margin assessment for cancer resection, antibody-fluorophore conjugates have been translated to clinical medicine for the identification and visualization of cancer-specific biomarkers (e.g., epidermal growth factor receptor [EGFR], carbohydrate antigen 19–9 [CA19–9], Carcinoembryonic antigen [CEA], etc.). These antibody-based probes have been studied extensively preclinically as well as in clinical trials and are well-established as a leading class of FGS probes.^25–27 However, the utility of antibody-fluorophores conjugates for nerve-specific imaging in vivo is limited due to the tight blood-nerve-barrier (BNB) junction, which generally prevents crossing of macromolecules (Figure 2a).²⁸ Notably, there are limited successful pre-clinical examples of nerve imaging using protein-based probes. The Mohs’ laboratory has demonstrated intraneural or topical application of recombinant human Nerve Growth Factor (rhNGF), a protein-based probe termed Nervelight^™, which is labeled with an 800 nm fluorophore that showed fluorescence in rat sciatic, cavernous, facial, and axillary nerve tissues (Figure 2b). The targeting mechanism of the Nervelight^™ probe was hypothesized to be due to the probe’s affinity for Tropomyosin receptor kinase A (TrkA), an overexpressed and high-affinity receptor for nerve growth factor (NGF) at the distal ends of the nerve.²⁹ In a similar fashion, the Sheikh laboratory used an anti-ganglioside monoclonal antibody with specificity for GT1b (GT1b-2b) as an axonal labeling target. Their fluorescently labeled version (GT1b-2b-550, ex/em=562/576 nm) used a visible fluorophore dylight550, to delineate intact motor, sensory, and autonomic nerve tissue fluorescence in mice six days post intraperitoneal injection (Figure 2c).³⁰

Figure 2. Protein- and peptide-based fluorescent contrast agents for nerve tissue imaging.

a Anatomical illustration of a typical peripheral nerve. Reprinted from Ref. [28]. Copyright 2016, image courtesy of Myoscience. b White light and fluorescence images of sciatic and cavernous nerves in rats following topical administration of a protein-based fluorescent contrast agent, Nervelight^™. Adapted from Ref. [29]. Copyright 2020, SPIE. c Brightfield and corresponding fluorescence images of a mouse sciatic nerve taken at day six post intraperitoneal injection of a fluorescently labeled anti-ganglioside antibody, GT1b-2b-550. Adapted from Ref. [30]. Copyright 2015, Springer Nature. d Whole-body survey of nerve tissues in mice 4 h after systemic injection with a peptide-based fluorescent contrast agent FAM-NP41. Screening data showed probe accumulation in various nerve tissues in the PNS with no blood-brain-barrier (BBB) cross. Adapted from Ref. [32]. Copyright 2011, Springer Nature. e Nerve-specific peptide FAM-HNP401 with affinity for human nerve tissue showed equivalent nerve-specific contrast in rats when compared to the previous FAM-NP41 nerve-specific peptide. Adapted with permission from Ref. [34]. Copyright 2018, Ivyspring International. f Brightfield and corresponding fluorescence images of a mouse sciatic nerve 30 min after systemic administration of a BODIPY-FL labeled Na_v1.7-selective peptide, Hsp1a-FL. Adapted from Ref. [35]. Copyright 2019, American Chemical Society. g In vivo nerve specificity demonstration of a myelin protein zero (P0) derived peptide sequence, Cy5-P0_101–125, following an intraneural injection at the site of a mouse sciatic nerve. Adapted from Ref. [38]. Copyright 2021, Springer.

Peptide-based fluorescent contrast agents

Analogous to antibody-based probes, peptides can also be used as targeting vectors for the protein of interest with intrinsic advantages, including lower immunogenicity, higher loading density, and improved tissue penetration.³¹ The Nguyen laboratory developed a nerve-specific peptide (NP41), discovered using phage display and in vitro selection against excised murine peripheral nerve tissue to yield the top-performing peptide sequence.^{32, 33} NP41 was then conjugated to a visibly emissive fluorophore, fluorescein and a NIR fluorophore, Cy5, resulting in two fluorescently labeled nerve-specific peptides, FAM-NP41 (ex/em=490/525 nm) and Cy5-NP41 (ex/em=649/666 nm), respectively. Both versions were validated in murine models, showing specificity for a number of peripheral nerves, including brachial plexus, sciatic, phrenic, and dorsal cutaneous nerves (Figure 2d). Fluorescence microscopy further demonstrated that the primary localization of Cy5-NP41was in the epineurium rather than myelin or axonal membranes as the common target for nerve tissue labeling. Importantly, NP41 generated nerve-specific contrast within 2–3 h following systemic injection and was completely cleared from the body within 24 h with no pharmacological or toxic effects observed at the imaging dose. Given these exquisite properties, the team subsequently developed a next-generation nerve peptide, HNP401 with specificity for the human peripheral nerves with future clinical translation in mind. Similar to FAM-NP41, FAM-HNP401 (ex/em=490/525 nm) provided equivalent nerve-specific contrast but much-improved signal intensities in rodents (Figure 2e).³⁴ Notably, this nerve-specific probe is currently in clinical trial for intraoperative use during head and neck surgery (ALM-488, NCT05377554).

In addition to the epineurium, the axon is another plausible targeting compartment in peripheral nerve bundles. The Reiner laboratory developed a fluorescently labeled Na_v1.7-selective peptide, Hsp1a, obtained from the venom of the Peruvian tarantula (Figure 2f). This peptide is a highly potent inhibitor for the voltage-gated sodium channel Na_v1.7 that is expressed in both human and mouse tissues.³⁵ This Hsp1a peptide was conjugated to a visible fluorophore BODIPY-FL, resulting in a fluorescently labeled version (Hsp1a-FL, ex/em=503/512 nm), that demonstrated nerve-specificity in mouse sciatic nerves following systemic administration. Histological and confocal microscopy analysis of the resected tissues further confirmed the Na_v1.7-specific binding affinity of this novel probe in the axons. The team subsequently switched fluorophore labeling from BODIPY-FL to Cy7.5 dye to enable NIR nerve visualization.³⁶ The utility of this peptide for nerve targeting has been expanded, encompassing other imaging modalities that are already present in the clinic beyond fluorescence for the nerve tissue imaging.³⁷

More recently, the van Leeuwen laboratory explored the utility of truncated homotypic myelin protein zero (P0) peptide sequences as the targeting vector for the peripheral nervous system.³⁸ They first designed seven P0 peptides derived from the crystal structure of the extracellular portion of P0. Screening these fluorescently labeled peptide candidates enabled the identification of Cy5-P0_101–125 as a potent nerve-specific agent for myelin under in vitro, ex vivo, and in vivo conditions (Figure 2g). Like the other two aforementioned peptide-based probes, Cy5-P0_101–125 can also be injected systemically, producing contrast within hours post injection.

Small-molecule fluorophore-based contrast agents

Histological staining using small molecule dyes for pathology and medical examination has been established for over a century. Inspired by the histological stains for myelin (e.g., Luxol Fast Blue), structural derivatization of these dyes has produced a number of fluorescent small-molecule fluorophores with an inherent affinity for the peripheral or central nervous system tissues, including stilbene,^{39, 40} coumarin,⁴¹ styryl pyridinium,⁴² distyrylbenzene,^43–46 carbocyanine,⁴⁷ and oxazine fluorophores.^48–52 The Frangioni laboratory generated two libraries of fluorescent dyes based on the styryl pyridinium and distyrylbenzene scaffolds, where they discovered first-generation nerve-specific small-molecule `fluorophore BMB (ex/em=390/501 nm), a UV-Vis fluorophore with affinity for both the peripheral and central nervous systems (Figure 3a).^{42, 43, 45} BMB nerve-to-muscle signal-to-background ratio (SBR) was high, but overall nerve specificity was suboptimal due to excessive uptake in the adipose tissues. Following quantitative structure-activity relationships (QSAR) screening and analysis of the styryl pyridinium and distyrylbenzene chemical libraries, the Choi laboratory judiciously selected a number of oxazine and rhodamine fluorophores with chemical similarities to the lead nerve-specific BMB probe. They discovered that a commercially available far-red dye, oxazine 4 (ex/em=616/635 nm), exhibited excellent nerve SBRs against both muscle and adipose tissues. However, oxazine 1, a NIR structural analog of oxazine 4, displayed a drastically different biodistribution pattern resulting in minimal nerve-specific contrast following systemic administration despite the high structural similarities (Figure 3b).⁴⁸

Figure 3. Small-molecule fluorophores for nerve tissue imaging.

a A distyrylbenzene-based small molecule fluorophore exhibited affinity for the brachial plexus and sciatic nerves in rats. Adapted from Ref. [43]. Copyright 2011, Decker Publishing. b A far-red fluorophore, oxazine 4 showed strong nerve specificity, while the NIR structural analog oxazine 1 showed pan-tissue fluorescence after systemic administration. Adapted from Ref. [48]. Copyright 2014, Ivyspring International. c Directed synthetic tuning of oxazines 1 and 4 enabled the development of NIR 700 nm nerve-specific oxazine fluorophores. d Photographs of conventional white-light illumination and fluorescence visualization of buried swine iliac and lumbar nerves using intravenously injected NIR nerve-specific LGW01–08 (left) and LGW05–75 (right). e Mock FGS in swine showcased fluorescent imaging could readily highlight a contiguous sub–1-mm nerve (blue arrow and dotted yellow lines) running between the surgical scissor blades that was in danger of transection by white light illumination alone. Figure panels in c-d were adapted from Ref. [50]. Copyright 2020, American Association for the Advancement of Science.

The Gibbs laboratory crafted a systemically modified oxazine fluorophore library, consisting of 66 hybrid structures derived from oxazines 1 and 4, to understand the SAR and afford a NIR nerve-specific fluorophore for intraoperative use as the end goal. Following a tiered screening assay including topical and systemic administration strategies in rodent models, four first-in-class NIR nerve-binding fluorophores (i.e., LGW01–08, LGW05–75, LGW04–31, and LGW03–76) were identified (Figure 3c). In a series of fluorescence-guided swine surgeries using clinical surgical systems, these lead candidates readily delineated buried lumbar and iliac nerves (Figure 3d). This represents a direct benefit of shifting the spectral properties of the oxazine probe into the NIR wavelengths, which enhances imaging depth. As the surgeon dissected through surrounding tissues, the visibility of the buried nerve tissue improved, resulting in highly sensitive nerve imaging. Notably, a small nerve branch at high risk for transection using conventional white light visualization was easily avoided using nerve-specific fluorescence guidance (Figure 3e), further highlighting the clinical promise of the FGS for nerve-sparing procedures to reduce iatrogenic injuries and subsequent comorbidities.

Administration routes

Several administration routes for nerve contrast agents exist and are applicable to the FGS procedures, including but not limited to intravascular, topical, intraperitoneal, intraneural, and intrathecal administration.⁵³ Systemic administration completed either in the pre-operative arena or at the beginning of the procedure can provide contrast throughout the nerve-sparing process without interruption to existing surgical workflow, making it highly amenable to the established surgical workflows. Unlike protein- and peptide-based contrast agents, nerve-specific small molecules tend to be lipophilic with low aqueous solubility, necessitating the development of formulation strategies to enable clinical translation potential. The Alani and Gibbs laboratories have developed micellar dispersion, liposomes, and inclusion complex formulations for the systemic administration of the distyrylbenzene and oxazine nerve dyes, replacing the previously established laboratory grade co-solvent formulation that hampered the clinical translation of probes with poor water solubility (Figure 4a).^{54, 55}

Figure 4. Preclinical testing and clinical translation of the nerve-specific contrast agents.

a A number of clinically viable systemic administration formulations have been developed to replace the laboratory grade co-solvent formulation while maintaining the sensitivity and specificity of the NIR nerve-specific probes. Adapted from Ref. [51]. Copyright 2021, Wiley-VCH GmbH. b Schematic and representative images of hydrogel formulation viscosity assessment in swine for the topical administration of contrast agents. Adapted from Ref. [52]. Copyright 2022, Elsevier. c Representative color and fluorescence images of the swine iliac plexus stained via topical administration of co-solvent, 22% Pluronics and 6.5% Na Alginate formulated oxazine 4. Adapted from Ref. [52]. Copyright 2022, Elsevier. d Schematic layout for perfusion of the amputated human limb model for preclinical human biomarker/tissue cross reactivity confirmation. Representative fluorescence image of human adipose, muscle, and peroneal nerve samples following a perfusion procedure of a contrast agent in the amputated human limb model. Adapted from Ref. [57]. Copyright 2023, SPIE.

Not all procedures would necessarily benefit from systematically administered nerve-specific contrast during surgery, especially when large nerve bundles are involved, which can be readily identified and well-differentiated with normal anatomy, placing patients at minimal risk for nerve injury. However, pre-existing tumors or prior trauma can significantly change the anatomical path of nerves, complicating the process of nerve identification and differentiation. In such scenarios, both the surgeon and patients could benefit from fast-acting, nerve-specific contrast in real-time, where a local administration strategy presents unique intraoperative decision-making opportunities in iatrogenic nerve injury prevention. To enable intraoperative administration at the surgical site, the Alani and Gibbs laboratories developed a gel-based formulation strategy that permits nerve-trajectory mapping within minutes of low-dose fluorophore application. Furthermore, the formulated probes can be easily applied as a liquid followed by rapid gelling for subsequent tissue hold at the site of application regardless of the tissue surface angle or composition (Figure 4b). This formulation is especially advantageous when the nerves are presented on irregular surfaces or steep angles, where liquid formulation would result in reduced tissue contact and substantial pooling, along with increased non-specific background tissue uptake, diminishing the overall nerve contrast (Figure 4c).⁵² Importantly, the substantially reduced imaging dose required for topical administration of nerve-specific probes for nerve-sparing FGS applications could facilitate clinical translation via the exploratory investigational new drug (eIND) guidance, significantly reducing the timeline and overall cost of first-in-human studies.⁵⁶ While intraneural injections could also enable local fluorophore uptake, a priori identification of the nerve tissue is required for successful administration, negating some of the contrast agent utility and presenting the potential for nerve damage due to direct injection.

Many aforementioned fluorescent contrast agents have demonstrated clinical utility for nerve-sparing surgery in animal models. However, translation to the clinic requires rigorous safety testing and costly trials. Target validation and efficacy testing in preclinical models that more closely resemble human physiology can further de-risk the technology and boost confidence for successful clinical translation. The Henderson laboratory recently developed a preclinical perfusion model for the evaluation of the molecularly targeted probes using amputated human limbs that can be perfused with contrast agent using the native vasculature, allowing for the collection of human testing data with zero risk to human patients (Figure 4d).⁵⁷ This ex vivo human tissue perfusion model presents additional opportunities for determining human biomarker/tissue cross reactivity, assessment of any tissue toxicity and evaluating pharmacokinetics profiles.^{58, 59}

FUTURE OUTLOOK

Driven by the discovery of molecularly targeted fluorescent contrast agents, the nascent field of FGS presents a variety of exciting opportunities to advance clinical medicine. Notably, the armory of nerve-specific contrast agents has expanded considerably over the last few years. While FDA-approved FGS vision systems exist, clinical nerve FGS still lags behind cancer-specific contrast agent development and translation due to the slower development of NIR nerve-specific fluorescent contrast agents. Probes with demonstrated safety, sensitivity, and compatibility with the clinical FGS systems would have tremendous adoption potential. Improving photon penetration with NIR nerve-specific fluorescence can provide an additional safeguard for buried nerve tissue, where the fluorescence signal would be visible from the commencement of the surgical procedures and would progressively brighten as the surgical plane nears the nerve tissues. While both antibody- and peptide-based targeting ligands have flexibility in the fluorophore conjugation selection for imaging system compatibility, a number of studies have shown fluorophores can dramatically alter the intrinsic properties of the targeting ligand following bioconjugation, diminishing targeted contrast and impeding their clinical potential.⁶⁰ Likewise, small-molecule nerve-specific fluorophores present core limitations with inherent optical properties (e.g., excitation and emission wavelength, brightness, etc.) and physicochemical properties (e.g., solubility, stability, etc.). Thus, much work remains before the nerve-specific FGS technology will have a major impact in medicine. In light of personalized precision medicine, nerve-specific FGS will benefit from probes with different photophysical and pharmacokinetic profiles more amenable to the specific surgical needs, where the size, type and location of nerve tissue as well as imaging time could be vastly different. For instance, visibly emissive contrast agents enable surface-weighted intraoperative nerve detection, while nerve tissues buried beneath blood or surrounding tissue will benefit from their NIR counterparts. Moreover, imaging in the central nervous system presents additional opportunities for creative design and molecular engineering of BBB permeable fluorescent contrast agents, propelling a broader impact of FGS for patient care.