Introduction
Hepatocellular carcinoma (HCC) is a significant public health concern worldwide. With over 780,000 cases diagnosed and more than 740,000 deaths each year, HCC is ranked as the sixth most common neoplasm and the third leading cause of cancer-related death.1 Globally, the incidence and mortality rates of HCC are nearly equal, indicating poor treatment efficacy.2 The majority of HCC cases occur in Eastern Asia and sub-Saharan Africa, where there is a high prevalence of hepatitis B virus infection3 and many individuals are infected with the virus at birth.4,5 Additionally, due to the increasing incidence of hepatitis C virus infection, the incidence of HCC is also rising in Western countries and Japan.6 Other risk factors include heavy alcohol consumption, non-alcoholic fatty liver disease, obesity, and diabetes.7–10 Certain genetic conditions, such as hemochromatosis, also increase the risk of HCC.11
Hepatectomy is one of the most effective treatments for HCC, and achieving radical resection of tumor lesions is critical for a successful outcome.1,12 Otherwise, the presence of residual cancer cells after surgery can lead to cancer recurrence and poor prognosis. Therefore, surgeons rely on imaging technologies to locate the tumor lesion.13 In the past several decades, imaging technologies such as computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography have provided surgeons with valuable preoperative planning information for surgical procedures. However, real-time visualization of the tumor during surgery has been limited due to equipment constraints. Unlike these imaging technologies, intraoperative ultrasound (IOUS) can provide real-time feedback to identify the number and location of tumor lesions, guiding the surgical approach during the procedure.14 However, there are several limitations to IOUS for HCC. Firstly, IOUS is highly dependent on the surgeon's skill level. Inexperienced surgeons may struggle to interpret the images and identify the tumor margins, which can result in incomplete tumor removal or damage to healthy liver tissue. Secondly, IOUS is a time-consuming procedure that can prolong surgery, increasing the risk of complications such as bleeding and infection. Thirdly, IOUS does not provide continuous real-time visualization because surgeons must perform intermittent ultrasonic detection during hepatectomy. In other words, the surgeon must transect the hepatic parenchyma after ultrasonic detection rather than under continuous ultrasonic monitoring.15 For better surgical outcomes, the development of intraoperative imaging technology is a critical issue in liver resection for HCC.
In recent years, fluorescence-guided surgery (FGS) has gained attention as a novel technique that utilizes fluorescent dyes to highlight cancerous tissue during surgery, enabling more precise and thorough resection.16–19 One of the main advantages of FGS in HCC surgery is its ability to differentiate between tumor and normal tissue with high sensitivity and specificity.20 By administering a fluorescent dye that selectively accumulates in cancer cells, surgeons can visualize tumor borders and identify small satellite lesions that may otherwise go undetected. More importantly, the fusion image is displayed on a high-definition screen by integrating the fluorescent signal from the fluorescent dyes with standard unaided vision using white light imaging, thanks to advanced imaging system software. This allows surgeons to receive real-time guidance throughout the entire surgical procedure.21
In this review, we provide a summary of recent advances in fluorescent probes for the detection and treatment of HCC (Scheme 1). First, we introduce the commonly used fluorescence navigation methods in clinical practice. Next, we highlight the progress of novel fluorescent probes, which offer improved efficacy. Finally, we discuss the challenges and opportunities associated with using novel fluorescent probes in HCC surgery, with the goal of achieving more precise and thorough removal of cancerous tissue and improving overall survival rates.
Indocyanine green (ICG)-based FGS for HCC
ICG is currently the most widely used near-infrared (NIR) fluorescent dye certified by the US FDA, and it has been discovered and applied in clinical practice for nearly 60 years.22 In recent years, due to the extensive development and application of new fluorescence imaging systems and fluorescence navigation equipment in clinical practice, ICG-based fluorescence imaging technology has rapidly advanced and gradually become more popular in China.23 After intravenous injection, ICG binds to polymeric proteins such as albumin and lipoprotein in the plasma, where it neither undergoes metabolic breakdown nor alters the molecular structure of the bound protein, providing good intravascular stability. As ICG circulates through the blood vessels around the liver, it is actively transported and specifically taken up by liver cells. It is then secreted into the biliary system and eventually excreted into the intestine and eliminated from the body.24 Normal liver tissue can completely clear ICG within 12–24 h, while ICG remains in tumor tissue for several weeks due to the impaired bile excretion function of liver cells in HCC tissue.25 In practical applications, the complex biological environment can interfere with the fluorescence signal of probes. For example, experimental studies have shown that most of the light is absorbed and neutralized by hemoglobin (wavelengths less than 700 nm) and water (wavelengths greater than 900 nm) as fluorescent light penetrates biological tissues.26 ICG molecules are excited by light with a wavelength of 750–810 nm and emit fluorescence with a peak wavelength of 840 nm, which falls within the "window" boundary (700–900 nm) of the deep red and NIR light spectrum, making it highly tissue-penetrating.27 Under this optimal emission wavelength, the fluorescent signal from ICG can highlight tumor tissue, thus guiding the intraoperative resection of HCC.
In clinical practice, ICG-based FGS has been commonly utilized to identify and demarcate tumor boundaries at the cellular functional level due to the difference in excretion between HCC lesions and normal liver tissue.28 With ICG injected intravenously several days before the operation, the tumor boundary can be delineated using ICG fluorescence detection during surgery. This allows the surgeon to maintain adequate surgical margins from the tumor lesion under the guidance of ICG-based NIR fluorescence imaging.23 Besides, the suspected tumor nodule identified by preoperative imaging techniques can be intraoperatively diagnosed based on fluorescence imaging, potentially modifying the clinical stage of HCC. As illustrated in Figure 1A, the preoperative imaging using enhanced CT scanning demonstrates an HCC lesion measuring 7 cm × 5 cm located in the caudate lobe of the liver. During surgery, this lesion, along with another lesion, was identified and localized through the strong fluorescence signal from ICG accumulated in the tumor (Fig. 1B). According to the fluorescence guidance, the tumor boundaries were distinguished from the left pedicle of the liver, the inferior vena cava, and the papillary process of the caudate lobe, and then the tumor lesion was dissected along these boundaries (Fig. 1C). Additionally, due to the low uptake ability of ICG in poorly differentiated HCC tissues, the fluorescence signal provided by the lesion itself is usually weak. However, because of the compression of the tumor on normal liver tissue, the excretion of ICG is delayed, resulting in a characteristic ring-shaped fluorescence around the cancer tissue. Partial cells in moderately differentiated HCC tissues lose their uptake function, typically showing a partial fluorescence signal. Highly differentiated HCC tissues retain a specific uptake ability for ICG, but their bile excretion function is abnormal, leading to prolonged fluorescence display and a homogeneous fluorescence signal.29 Based on these anatomical abnormalities and dysfunctions, HCC lesions can be distinguished from noncancerous nodules during surgery, as illustrated by the strong fluorescence emitted by the HCC lesion, while the necrotic lesion displays little to no fluorescence signal (Fig. 2).
In addition, for patients with unresectable HCC at the current stage, the primary treatment options include local ablation, chemotherapy, radiotherapy, and interventional embolization, with transcatheter arterial embolization (TACE) being the most widely employed in clinical medicine.30 TACE involves dissolving chemotherapy drugs, for instance, doxorubicin or carboplatin in an embolic agent (commonly lipiodol) and injecting the mixture into the tumor’s blood supply via the hepatic artery. This leads to significant tumor cytotoxicity, ischemia, and hypoxia, while simultaneously inhibiting the proliferation of tumor cells and killing them.31 Compared to systemic chemotherapy, TACE significantly increases the drug concentration in HCC tissue and reduces the probability of systemic adverse events. Additionally, when ICG is added to the embolic agent, clinicians can observe changes in the tumor post-TACE treatment and facilitate intraoperative tumor navigation, leading to improved therapeutic effect. However, for unresectable HCC tumors larger than 10 cm in diameter, the therapeutic effects are often unsatisfactory. Common reasons include insufficient drug load, rapid drug clearance, and biotoxicity.32
In recent years, a research team developed a new super-stable homogeneous intermixed formulation technology to improve the stability of drugs in lipiodol (Fig. 3).33 At a certain temperature and pressure, carbon dioxide is continuously injected into the reactor to make the fluid reach the supercritical stage above the critical temperature. Under these conditions, the team successfully mixes clinically available drugs with lipiodol, achieving uniform and stable dispersion of the drugs within the lipiodol.33 With the help of this technology, ICG can be more effectively dispersed in lipiodol (lipiodol-ICG formulation). Assisted by interventional procedures, the formulation is delivered to the local tumor region. Compared to free ICG of the same concentration, the fluorescence stability of the formulation is significantly improved. As a highly decentralized drug system, super-stable homogeneous intermixed formulation technology can overcome the aggregation effects that lead to fluorescence quenching.34,35 This capability has been confirmed in both in vitro and in vivo experiments.33 Another important point is the enhanced targeting for primary tumor tissues. The lipiodol-ICG formulation benefits from drug delivery in situ and the tumor cells' specific uptake of lipiodol, allowing for targeted delivery to tumor tissues. This will help the surgeon to remove the tumor tissue as completely as possible, helping to prevent tumor recurrence and metastasis.
Although ICG-based FGS has many advantages for HCC, there are still several areas for improvement in clinical practice, as follows: (i) ICG has limited detection ability for deep lesions. As the intensity of fluorescence decreases while passing through tissue, the proportion of light reaching the camera is greatly reduced. Thus, only surfaces or fluorescent structures less than 10 mm deep can be visualized.36 (ii) ICG has poor tumor targeting ability and is completely passive. Its distribution in the liver is affected by blood flow and other physiological factors, such as the degree of cirrhosis, making it difficult to determine the optimal timing for preoperative administration.37 If the degree of cirrhosis is severe and the administration time is too short, a large number of cirrhotic nodules will show fluorescence, as seen in Figure 4. (iii) Benign liver tumors also emit fluorescence due to ICG retention in tumor tissues because they lack bile ducts. Therefore, ICG fluorescence imaging cannot be used as a diagnostic method to distinguish between benign and malignant liver tumors. As shown in Figure 5, hepatic adenomas and focal nodular hyperplasia, verified by postoperative pathological examination, can also be illuminated following preoperative intravenous administration of ICG to assess liver function.
In conclusion, when using ICG for surgical navigation of HCC, surgeons need to consider the above factors and combine them with other imaging modalities and clinical findings to ensure the accuracy and safety of surgical navigation. Therefore, the development of innovative fluorescent probes for FGS of HCC with higher specificity, greater signal-to-noise ratio, and more convenient use has garnered significant attention in recent years to address the aforementioned deficiencies. In the following sections, we will summarize recent advances in this area.
Conventional fluorescent molecules for FGS of HCC
Under normal conditions, human cells and their microenvironment maintain a stable state at the molecular level. However, after HCC occurs, some biomolecules, including enzymes, reactive oxygen species (ROS), reactive sulfur species (RSS), pH, and other biomarkers, may fluctuate to a certain degree. These changes serve as specific targets that enable accurate monitoring of HCC.38,39 Based on this, scientists have synthesized and verified a series of molecular fluorescent probes with better fidelity, high signal-to-noise ratio, and biological imaging specificity.
Enzyme-responsive fluorescence imaging
Enzymes play a crucial role in maintaining the balance and stability of biological systems. Abnormal enzyme activity index is directly associated with the occurrence of various tumors, and can be used as a key index for clinical diagnosis and evaluation of specific diseases.40 Over the past decade, significant progress has been made in fluorescent probes for detecting HCC overexpression enzymes. Here are several notable overexpressed enzymes and their corresponding molecular fluorescent probes.
Cyclooxygenase-2 (COX-2) is an enzyme typically absent in normal tissue but becomes activated and expressed when triggered by factors such as tumors or infections. Research has revealed that COX-2 plays a significant role in oncogenesis, proliferation, and migration in HCC.41 Therefore, COX-2 may be the most suitable candidate target for fluorescence imaging of HCC. In 2017, Wang et al.42 reported a COX-2-specific probe (COX-2 FP) and tested whether it could distinguish HCC both in vitro and in vivo (Fig. 6A). They determined the enhanced expression of COX-2 in human HCC cell lines (Fig. 6A-b). Fluorescence emission of COX-2 FP was then tested in HCC cell lines and a heterograft mouse model, showing that both could be clearly distinguished from normal liver cells or tissues based on the fluorescence signal (Fig. 6A-c, d). These results suggest that COX-2 FP could be a potential method to distinguish HCC, particularly in intraoperative imaging.
Histone deacetylases (HDACs) are a family of epigenetic enzymes that regulate gene expression, cell cycle progression, and apoptosis.43 Consequently, they have emerged as an attractive target for novel cancer diagnostic and therapeutic strategies. In 2019, Tang et al. designed and synthesized an HDAC-targeted NIR probe, IRDye800CW-SAHA, for liver cancer imaging and FGS.44 This probe has strong affinity for HDAC and good biocompatibility. In vivo and ex vivo fluorescence imaging, IRDye800CW-SAHA showed high specificity and sensitivity, and tumor fluorescence signals could still be seen after subcutaneous accumulation for up to 48 h. In image-guided surgical studies, liver tumors could be precisely resected under fluorescence guidance. These findings demonstrate that IRDye800CW-SAHA can detect HDAC expression levels in tumors, enabling targeted imaging of HCC and FGS, with potential for clinical translation.
β-glucuronidase (GLU) is a lysosomal glycosidase whose main function is to participate in the degradation of glucuronidoglycans containing glucuronic acid and is associated with various pathophysiological conditions in the body.45 Additionally, the active expression of GLU is closely related to the invasion, apoptosis, metastasis and proliferation of tumor cells. Compared with normal hepatocytes, GLU activity was significantly enhanced when hepatocytes became cancerous.46 In 2022, Wang et al.47 reported a fluorescent probe, DCDNO2, for the real-time, precise detection of GLU (Fig. 6B). DCDNO2 exhibited excellent fluorescence characteristics, such as high specificity, rapid response, and good biocompatibility. As shown in Figure 6B-b, when DCDNO2 was added to HepG2 cells, a strong fluorescence signal was observed. After baicalin pretreatment, the red fluorescence in HepG2 cells decreased significantly. In addition, DCDNO2 can accurately assist and guide the surgical removal of tumors after local spraying of DCDNO2 in mouse models of liver cancer (Fig. 6B-c). Therefore, DCDNO2 plays an important role in assisting HCC surgical resection.
ROS-responsive fluorescence imaging
According to previous literature, liver cancer cells can produce more ROS than normal liver cells.48 Hypochlorous acid (HClO) is one of the most essential ROS and a critical target in current probe synthesis research. It is generated with the assistance of myeloperoxidase, catalyzing the oxidation reaction of hydrogen peroxide and chloride ion in the biological system.49 The high reactivity of endogenous HClO makes it a double-edged sword. On the one hand, it plays a key role in defending against invading pathogens; On the other hand, its abnormal accumulation can cause or aggravate various inflammation-related diseases.50,51 Recent studies have shown that inflammatory injuries or diseases are major driving forces for cancer, and inflammatory signals may lead to carcinogenesis.52,53 Therefore, tracking the level of HClO can be used to target liver cancer cells. Li et al.54 designed and synthesized a near-infrared fluorescent probe (MBH-MT) (Fig. 7). The probe targets liver cells by transferring the melatonin fragment onto the methylene blue fluorophore (Fig. 7A). Four normal cells (HUVEC, LO2, RAW264.7, and HK-2) and one tumor cell (HepG2) were selected to evaluate whether probe MBH-MT can differentiate between normal and tumor cells. Tumor cells showed much higher fluorescence intensity than normal cells (Fig. 7B). In vitro, the fluorescent signal of HepG2 cell grafted tumors was significantly stronger than that of normal liver organs (Fig. 7C). Moreover, the probe successfully assisted liver tumor resection surgery in nude BALB/c mice with transplanted tumors (Fig. 7D, E). Based on the above results, the probe is expected to be an effective tool for HCC diagnosis and surgical treatment.
RSS-responsive fluorescence imaging
RSS mainly includes cysteine (Cys), homocysteine (Hcy), and glutathione (GSH), which have high reactivity and play irreplaceable roles in physiological and pathological processes.55 They are also commonly used as scavengers of oxidants and are crucial for maintaining the redox state of biological systems.56 Excessive RSS can trigger a series of related diseases such as inflammation, cancer, and Alzheimer's disease.57
In 2014, Ren et al.58 developed a new ratiometric fluorescence probe for detecting GSH in a cancerous biomatrix (Fig. 8). The probe demonstrated high selectivity and sensitivity for detecting intracellular glutathione in four cancer cell lines (HepG2, Hep3B, Huh7, and HeLa) without causing cell toxicity (Fig. 8B). They constructed a quantitative model of intracellular GSH content using the probe ratio (B/Y). The probe was validated in the liver of normal mice, tumor xenografted mice, and tumor tissues of xenografted mice, with results as expected (Fig. 8C, D). The probe has the potential to be used as a highly sensitive and accurate biosensor for real-time intraoperative monitoring of HCC.
For targeting Cys, He et al.59 designed a NIR fluorescent probe, Cy-Q. After Cy-Q was incubated with different analytes at 37°C for 2 h, a significant fluorescence increase was observed with the addition of Cys, Hcy, and GSH. At the same time, no obvious changes appeared with the other 16 analytes. The fluorescence intensity of Cys was twice as strong as the other two thiols, allowing effective differentiation of Cys from Hcy and GSH. Its level can be quantitatively monitored with specificity. The Cys fluorescence imaging of Cy-Q probe on normal mice, H22 tumor bearing mice and lipopolysaccharide (LPS) -induced mice was studied. The results showed that CY-Q probe preferently responded to liver tumor sites, which verified that the probe could easily distinguish liver tumor from normal tissue and inflammation. This work provides a promising NIR fluorescent probe with desired sensing properties for HCC imaging in biological systems.
Other fluorescent probes for HCC imaging
In addition to fluorescent probes activated by enzymes, ROS, and RSS, some probes are activated by other species, such as pH and polarity, which are also essential to living organisms. The pH value of the tumor microenvironment (6.2–6.9)60 is slightly lower than the pH of the physiological microenvironment (7.2–7.4). This phenomenon is caused by the uncontrolled proliferation of cancer cells in the hypoxic environment.61 Thus, precisely distinguishing subtle pH changes offers an ideal and practical way to differentiate cancer cells from normal cells. In 2019, Zhang et al.62 developed a new dual-hepatocyte-targeted fluorescent probe (HPL-1). The molecule is "on" in the tumor microenvironment (pH 6.5) but "off" in the physiological microenvironment (pH 7.4). HPL-1 can distinguish HCC cells from other tissue cells and is the first example of small molecule fluorescent probes targeting pH changes in HCC cells. This heralds further possibilities for molecular fluorescent probe development in this area. Polarity is also a significant microenvironmental parameter, reflecting multiple complex biological processes, such as protein composition and signal transduction. It has been reported that polarity disorders are closely related to liver diseases.63,64 In 2021, Lin et al.65 developed a fluorescent probe ERNT for monitoring endoplasmic reticulum polarity. The probe can detect polarity changes during liver injury, including fatty liver, liver fibrosis, liver cirrhosis, and liver cancer, providing valuable insights into liver tumors and environmental polarity.
Additionally, ligand-modified fluorophores are another strategy to enhance the anchoring effect of fluorophores onto tumor cells. Integrin α6 is an adhesion molecule that exists on the cell surface and is involved in the process of cell attachment and tumor invasion and metastasis.66 Evidence of overexpression of integrin α6 in approximately 94% of clinical early-stage HCC cases67 suggests that it is a potential target for molecular fluorescence imaging in HCC assays. Feng et al.68 previously identified a tumor-targeting peptide, CRWYDENAC (RWY), through phage display technology, confirming its high specificity and affinity for integrin α6. The researchers then labeled the RWY peptide with cyanine 7 (Cy7) to create a new molecular fluorescent probe, Cy7-RWY, targeting HCC.69 The results showed that Cy7-RWY had high affinity to HCC-LM3 cells in vitro. By near infrared imaging, intravenous Cy7-RWY was only observed at the tumor site with a strong fluorescent signal in both subcutaneous HCC-LM3 tumor models and HCC orthotopic transplantation tumor models. These findings indicate the specific targeting ability of Cy7-RWY for HCC tumors in vivo. Based on these validations, the Cy7-RWY probe may play a significant role in accurately delineating liver tumor lesions during future surgical resection of HCC.
Aggregation-induced emission luminogens (AIEgens)-based FGS for HCC
Although fluorescence molecular imaging has been widely used in simulated experiments for clinical image-guided surgery of liver tumors, its application in aqueous systems is greatly limited due to the aggregation-caused quenching (ACQ) effect. In high-concentration solutions or solid states, the formation of aggregates leads to close intermolecular contact, which alters the energy levels of electrons inside the molecules and disrupts the luminous pathway of the fluorescent compound. As a result, the amount of emitted light is reduced or even completely quenched. Additionally, the intensity of traditional fluorescent probes is affected by various factors, such as chemical reagents, ambient temperature, pH, and UV excitation light stability, which greatly reduce the accuracy and stability of detection and imaging. This limitation has driven the development of new fluorescent probes to achieve superior imaging capabilities.70–75 Aggregation-induced emission (AIE) (including aggregation-enhanced emission (AEE)) has gained tremendous interest in research since its discovery in 2001.76 Unlike conventional fluorophores that suffer from the ACQ effect, AIE fluorophores are not emissive in solution but become highly emissive in the aggregated or solid state due to the restriction of intramolecular motions.77,78 The simple synthesis of AIE molecules and its remarkable fluorescence properties make it very popular in the fields of chemical sensors, biosensors, cell imaging, smart materials, and therapeutics.79–91 After 20 years of design and development, AIE materials now exhibit strong photobleaching resistance, with luminescence increasing as the concentration rises.92,93 Notably, light regulation of different emission bands has been achieved through flexible chemical modifications, and several near-infrared probes have been developed to produce high-resolution images in cell imaging and related bioimaging techniques.94 These advantages make AIE probes highly suitable for the future diagnosis and guidance of liver cancer tumors.
AIEgens for HCC imaging
In 2020, Wang et al.95 reported an aqueous-soluble AIE probe (HOTN) for detecting HClO (Fig. 9). Based on the excellent selectivity of HOTN for ClO− in vitro (Fig. 9B), HOTN has been successfully applied to imaging liver tumors in situ. The fluorescence intensity in liver of mice with liver cancer was obviously higher than that of healthy mice, and the fluorescence in cancer spot was the brightest (Fig. 9C–E), suggesting that HOTN could serve as a powerful tool for detecting HClO in HCC.
In the same year, We report that a novel AIE fluorescent probe hexaphenyl-1, 3-butadiene derivative (ZZ-HPB-NC) has a highly selective and immediate response to HCC.96 In the frozen section of HCC, the fluorescence of ZZ-HPB-NC can be quickly lit by simple spraying, but there is no obvious fluorescence change in the liver cirrhosis, benign nodule tissue and normal liver tissue. Compared with the most authoritative hematoxylin and eosin staining, the detection speed and accuracy of ZZ-HPB-NC are higher. The results showed that ZZ-HPB-NC staining provided experimental data basis for the intraoperative diagnosis of HCC, and had potential clinical application value. Based on the successful detection of liver tumor lesions using ZZ-HPB-NC, its isomer, all-trans configuration HPB-NC (EE-HPB-NC), was designed and synthesized.97 Remarkably, EE-HPB-NC can identify normal liver tissue, and as an auxiliary tool, it can significantly improve the accuracy of ZZ-HPB-NC in identifying liver tumors (Fig. 10A).97 In the HepG2/LO2 co-culture system, ZZ-HPB-NC and EE-HPB-NC can accurately label HepG2 and LO2 cells, respectively (Fig. 10B). Similarly, at the fresh tissue level, tumor lesions gradually fluoresce in ZZ-HPB-NC staining, reaching the maximum fluorescence intensity within 2 minutes, while normal liver tissues in the same dish remain unchanged. Moreover, EE-HPB-NC can also label normal tissues within 2 minutes (Fig. 10C). To effectively use HPB-NC for surgical navigation of solid hepatic tumors in situ, mouse orthotopic liver tumor models were established. Significant green fluorescence appeared in the tumor lesions with ZZ-HPB-NC staining, including tumors that had metastasized subcutaneously. EE-HPB-NC accurately labeled normal liver tissue to help guide complete tumor resection (Fig. 10D). These two probes complement each other to construct a configuration-induced cross-identification fluorescence imaging strategy, demonstrating their potential for clinical applications in real-time surgical fluorescence navigation.
Nanoprobes decorated with AIEgens as fluorophores for FGS of HCC
Although AIE probes have superior luminescence characteristics compared to traditional ACQ fluorescent dyes and can detect HCC lesion sites as described above, there is still a long way to go from laboratory research to clinical application. The main reason is that the complex in vivo environment presents a bottleneck in selecting drug delivery routes. For example, due to the complex distribution of tumors in vivo, imaging clearance is often difficult and may even be almost synchronized with the liver tumor, which affects the surgeon's judgment of tumor boundaries and can lead to surgical failure. Additionally, during the administration process, AIE probes may become unstable and degrade prematurely, which weakens the luminescence or causes non-specific deposition in other tissues and organs, resulting in inaccurate imaging. Therefore, there is an urgent need to develop AIE probes with stronger targeting, more stability, and more convenient drug delivery for clinical applications. AIE nanoprobes, with their relatively stable core structure, stronger fluorescence, adjustable size, and modifiable outer layer, may meet this challenge.
α-fucosidase (AFU) is widely present in various tissues and cells, as a member of the glycosidase family, AFU can hydrolyze α-L-focusidase residues on lysosomes and their terminal glycolipids or glycoproteins.It is highly active within a pH range of 4−6.5.98 Previous studies have shown that detecting overexpression of serum AFU levels diagnosed 85% of HCC patients six months earlier than ultrasonography.99 Therefore, real-time detection of AFU could be a promising strategy for early diagnosis of HCC. In 2022, Situ et al.100 constructed an AIE molecular fluorescent probe (α-Fuc-DCM), as shown in Figure 11. The probe was assembled from DCM-O− and α-L-fucose residues. The authors confirmed that AFU effectively catalyzes the cleavage of α-L-fucose residues, thereby triggering fluorescence (Fig. 11A). Results from live cell imaging indicated that much brighter NIR fluorescence images were captured in HepG2 cells compared to HepG2 cells with an inhibitor and LO2 cells (Fig. 11B). In a mouse liver cancer model imaging experiment, α-Fuc-DCM allowed imaging of approximately 1 mm tumors. In comparison, after the subcutaneous injection of α-Fuc-DCM nanoparticles, there was minimal fluorescence(Fig. 11C, D), which also verified the specificity of the probe. Furthermore, immediately after killing the mice, fluorescence images of the tumor and other organs in vitro were recorded (Fig. 11E), confirming that the NIR fluorescent signal was emitted from the DCM-O– product. In conclusion, the authors established an effective method for monitoring AFU activity in vitro and in vivo, which could contribute to the early diagnosis and treatment of HCC.
γ-glutamyl transpeptidase (GGT) is a biological marker for the diagnosis of HCC and is the preferred reagent for early detection and intraoperative navigation.101 In 2019, Liu et al.102 designed and synthesized a novel NIR fluorescent nanoprobe (ABTT-Glu) with a coupled AIE effect and excited-state intramolecular proton transfer (ESIPT) effect for specific detection and imaging of GGT. This was the first demonstration of an "AIE + ESIPT" NIR nanoprobe for detecting GGT with high selectivity and sensitivity. The breakage of the γ-glutamyl bond catalyzed by GGT can alter the dispersed state of the probe and trigger both the AIE and ESIPT effects. ABTT-Glu has good specificity and sensitivity to GGT, and the detection time is short and the detection limit is low . The authors successfully imaged GGT activity in HepG2 cells and human liver tumor tissues with high specificity and long-term sustainability, meeting clinical needs and with great potential for precisely guiding human liver tumor surgery.
Multispectral optoacoustic tomography (MSOT) is a functional photoacoustic imaging method that works by irradiating a sample through multiple wavelengths and detecting ultrasonic signals from a light absorber. MSOT can produce three-dimensional (3D) images by capturing a stack of cross-sectional (chromatographic) images as a maximum intensity projection (MIP) image. In 2020, Zeng et al.103 developed an activatable nanoprobe, BH-NO2@BSA, which responds explicitly to an overexpressed nitroreductase enzyme found in tumor cells and generates strong MSOT and NIR-I/NIR-II signals (Fig. 12A). The probe has been successfully used for in vivo fluorescence and MSOT imaging of subcutaneous liver tumors and liver tumors in situ (Fig. 12A-b), to guide preoperative precise localization of in situ liver tumors and accurate intraoperative identification of tumor margins (Fig. 12A-c), ensuring complete resection of liver tumors.
In the absence of oxygen, N-oxides can be converted into corresponding amines after binding with heme iron of various cytochrome P450 enzymes (CYP450), and irreversible two-electron reduction occurs. Interestingly, TPE-tetra(4-(diethylamino) phenyl)ethene (TPE-4NE) has shown AIE behavior and strong proton capture capability, which was inspired by Zhang et al. that TPE-4NE may be a promising candidate for deep tumor penetration.104 In 2021, they synthesized an AIE nanoprobe (TPE-4NE-O),105 which specifically turned on fluorescence in the presence of cytochrome P450 reductase (Fig. 12B-a). The probe can selectively illuminate hepatoma cells and show enhanced deep tumor penetration both in vitro and in vivo by charge conversion. After modification with FA-DSPE-PEG, FA-DSPE/TPE-4NE-O accumulated better within the tumor (Fig. 12B-b), enabling resection of micro tumors in Hepa1-6 tumor-bearing mice (Fig. 12B-c). More importantly, the probe exhibited good biocompatibility and negligible organ damage, without the risk of hemolysis. This simple yet promising probe provides new strategies for the visualization of minute liver tumors, showing great potential in HCC clinical surgery.
Future perspective
Fluorescent probes are important tools for the visualization of physiopathological processes in vivo. In the past few years, significant progress has been made in developing fluorescent probes for guiding intraoperative resection of HCC. In this review, based on previous work, we systematically summarized the ICG widely used in clinical practice, molecular fluorescent probes targeted to identify biomarkers, including enzymes, ROS, RSS, etc., and the latest fluorescent probes based on the AIE effect. These probes showed excellent optical performance for the intraoperative real-time detection of simulated HCC.
Nevertheless, several key points remain to be addressed in the design of better fluorescent probes for the clinical transformation of HCC: (i) Optimize properties of fluorophores to enhance the tissue penetration depth of the detected signal. As is well-known, fluorophores in the NIR region have deeper tissue penetration than those in the visible region. Thus, the development of novel NIR fluorophores, especially in the NIR II region (1,000–1,700 nm), will provide a potential means to overcome tissue barriers. Besides, multimodal imaging, combining fluorescence signals with CT, MRI, photoacoustic imaging, etc., has also attracted attention as a means to achieve intraoperative visualization with higher resolution and deeper tissue penetration.106 (ii) Improve the specific probe recognition to HCC to enhance the signal-to-noise ratio. Ligand-modified probes can increase the binding of probes to cancer cells and decrease distribution in normal tissue. As a result, tumor tissue emits strong fluorescence, while normal tissues show almost no fluorescence. On the other hand, “turn-on” fluorescence probes can light up tumor lesion when activated in the tumor microenvironment, but remain dark in normal tissues because they cannot be activated there. (iii) Develop new types of materials and technologies. For example, AIE technology can prevent probe quenching after aggregation, thereby improving fluorescence intensity. Additionally, AIE technology offers a novel strategy for highlighting tumor tissues through specific binding between AIEgens and specific biomacromolecules in tumor cells, triggering restriction of intramolecular motions to inducing emission. (iv) Explore new approaches for probe administration. The current mode of administration is usually intravenous injection, but some probes still distribute in normal tissues, resulting in undesired fluorescent signals. Therefore, a more effective approach would be to improve the accumulation of fluorescent probes in tumor lesions, enhancing their distribution and retention. For example, nanocarriers can accurately deliver probes into tumor cells through the EPR effect and active targeting, where they accumulate and emit fluorescence. Additionally, a homogenous lipiodol-ICG formulation, developed using superstable homogeneous intermixed formulation technology for transcatheter arterial embolization, has been developed as an alternative administration method. This allows for local deposit in HCC tissue, enalbling the imaging of the full tumor region and boundaries in real-time and guiding laparoscopic hepatectomy.107 (v) Design novel multifunctional probes for integrated diagnosis and treatment. For instance, fluorescent probe with photosensitivity can accurately destroy tumor lesions through photodynamic therapy while detecting them, or kill remaining tumor cells at the surgical margin during resection, improving the surgical treatment outcome.