Gold Nanoparticle Effects on the Hounseld Unit of Computed Tomography in Hepatocellular Carcinoma: Systematic Review

Background: The present study has attempted to gather all the original and relevant data on the application of gold nanoparticles aimed at the improvement of computed tomography image quality and Hounseld unit in hepatocellular carcinoma. We performed a systematic review on the studies indexed in PubMed from January 2000 to January 2020. Afterwards, the study design and quality were evaluated. Results: An increase in the nanoparticles concentration and incubation time was associated with improved image quality and the Hounseld Unit of computed tomography. Conclusion: This study highlights the considerable diagnostic role of gold nanoparticle as novel contrast agents in the early detection of hepatocellular carcinoma.


Introduction
Liver cancer (Hepatocellular Carcinoma (HCC) is the fth most common type of malignancy, and the third leading cause of cancer-related death. The annual incidence and mortality of liver cancer from 1992 to 2016, as reported by the American Cancer Society (ACS), was about 8.8 and 6.5 per 100,000 individuals per year, respectively. In 2020, 42,810 new cases of liver cancer were identi ed in the US, and 30,160 were reported to have died from this malignancy, indicating an incidence and mortality of 2.4% and 5.2% among cancers, respectively (https://cancer.gov). It has been estimated that liver cancer will surpass breast, prostate, and colorectal malignancies, in terms of prevalence, and become the third leading cause of cancer-related death by 2030 1 . The overall 5-year survival rate for patients with liver cancer is only 19.6% after the disease is diagnosed. Regrettably, the majority of cases are detected in late stages, resulting in signi cant di culties in the treatment process [2][3][4][5] . Interestingly, the establishment of new biomarkers can be considered for the prediction of cancer patient treatment and prognosis [6][7][8] .
In this regard, administration of contrast-promoting agents could be considered a helpful strategy for making a valid diagnosis with high contrast 11,[23][24][25] . The routine clinical CT contrast agents, as well as small iodinated compounds, are imbued with a number of challenges including short imaging time due to rapid renal clearance, renal toxicity, and inappropriate vascular permeability that have limited their applications in speci c tumor imaging 26-28 . Polymer-coated bismuth sul de (Bi 2 S 3 ) has been demonstrated to be a good candidate for incorporation as CT contrast agent with a notable e cacy and safety higher than that of iodinated compounds 29 . However, modi cation of size, shape and surface area of this compound is di cult 28 .
Since the prerequisite of dense matter has been regarded as a drawback for CT, it has recently been proposed to adopt materials with a high atomic number.
The present study evaluates the functional outcomes reported in the publications of the last two decades regarding the application of AuNPs as contrast agents in CT of animal models. Considering the reports included in this systematic review, we sought to provide a comprehensive discussion on the effect of properties of AuNPs such as size, shape, dosage, coating elements, effects of different injection modes, time and Houns eld unit (HU) on liver tumor imaging.
To our knowledge, no systematic review has been conducted over the past twenty years, thus, this study would be the rst systematic review of the various effects of AuNPs on the quality of CT imaging of liver tumors ( Figure 1).

Materials And Methods
The present study follows the PRISMA statement for transparent and comprehensive reporting of methodology and results 78 .

Search strategy
All original investigations exploring the effects of AuNPs on the imaging quality of CT of liver tumors published between January 2000 and January 2020 were considered. Our keywords included "gold" AND "liver cancer" OR "hepatocarcinoma" AND "nanoparticles" OR "Nano" OR "Nano structure" OR "Nano rod" OR "Nano cage" OR "Nano shell" OR "Nano prism" OR "Nano star" OR "Nano sphere" AND "tomography" OR "computed tomography" OR "CT".

Inclusion and exclusion criteria
All English reports concerned with the application of AuNPs as contrast agents in CT of liver tumors in both in vitro and in vivo conditions from January 2000 to January 2020 were included. In the next step, except original studies, other types of reports such as reviews, commentaries, simulation, book chapters and perspectives, studies in congresses, conferences, symposiums and incomplete articles, and papers written in languages other than English were excluded.
This process was controlled and revised independently by four authors.

Data extraction
From each study, the following information were extracted: rst author, year of publication, in vivo/in vitro model, administration route, AuNPs geometry/size/zeta potential/coating-conjugating agent/dose of usage/hydrodynamic size/targeting ligand/time and HU.

Quality of study
To assess the quality of observational studies included in this systematic review, a checklist as presented in Table 1, was adopted, containing 12 items with scores "Yes=1" and "No=0". On a scale of 0 to 12 in terms of cumulative quality score, only studies with a minimum score of 8 were included in this systematic review.

Statistical analysis
Accordingly, we opted to discuss the relationship between NPs dose (mg/ml) and HU rate (r), and the correlation between NP administration time followed by CT imaging and HU rate (r).

Search strategy results
As explained in Figure 2, articles were eventually included in this study 79 . According to exclusion and inclusion criteria, 13 papers were remained for analysis (Table 2 and Table 3). According to the PRISMA diagram, 741 records were found. After removing 5 duplicates, 736 records remained for further assessment.
After removing 15 records including reviews, guidelines and editorials, 721 original records remained, of which 49 were irrelevant, and 657 were duplicates.
The nal 17 reports, after the application of the exclusion criteria, were reduced to 13 reports with potential for further evaluation.

NP concentration, incubation time and HU rate
In articles reviewed the dose/concentration of AuNPs varied from below 1mg/ml 83 and 1mg/ml 3, 5, 80, 85 to 82 mg/ml 28 . The post-injection time followed by CT imaging varied from 0.08h 3, 10 to 48h 10 , resulting in alterations in the HU from 7 2 to 758.8 84 without targeting, and 24 10 to 96 3 with targeting.
Our systematic review concluded that by increasing the NPs incubation time without targeting in animal subjects, HU rate was elevated.

Quality assessment
The scores attributed to the quality of the 13 articles included in this study fall within the acceptable range of 9 to 12 (Table 1). Discussion with tunable surface plasmon resonance (SPR). The small size renders them suitable for being accumulated within the tumor microenvironment by means of passive strategies in the treatment of cancer. AuNPs are compatible with an appreciable array of materials, have a higher X-ray absorption coe cient, and response well to different conjugates, which is particularly useful in active targeting 3,28,32,80 . The present review suggests that AuNP-based contrast agents can be useful in x-ray-based computed tomography that may lead to an improvement in the HU rate.
Despite all the merits, important questions regarding the e cacy of AuNPs at improving the CT imaging quality and HU rate are yet to be answered.
The 2007 in vitro analysis by Kim et. al. on antibiofouling PEGylated AuNPs revealed an X-ray attenuation capacity that was 5.7 and 2 times higher than that of the routine iodine-based CT contrast agent, Ultravist, and Bi 2 S 3 NPs-based agents, respectively. The fabricated NPs were used to discern hepatoma from normal liver tissue with a considerable contrast (~two-fold) 5min after injection, without imposing any alterations in the relative contrast of CT images up to 24h later 28 . In another study (2011), Kim et. al. demonstrated signi cant CT attenuation due to the hybrid constituent of amphiphilic polymer-coated Au-Fe3O4 NPs. At the same Au concentration, the HU values of the hybrid NPs were higher (723 at 100mM Au) than those of pure AuNPs (528 at 100mM Au), which was attributed to the presence of IONPs with an x-ray absorption coe cient of 0.37cm2 g −1 at 100keV. Moreover, the designed NPs generated a signi cant contrast between the hepatoma cells and normal parenchyma in CT/MRI imaging. An hour following the injection of NPs, a good contrast enhancement (~1.6fold) was achieved and the relative difference between the contrasts of the healthy and tumor tissues remained unchanged for the next 24h 85 . In 2014, Liu et. al. treated hepatocarcinoma cells with LA-Au DENPs, and observed a sharper enhancement in CT contrast than that elicited by nontargeted NPs at the same concentrations. An hour following the injection, a higher X-ray attenuation capacity was observed compared to that of the routine clinical iodine-based CT contrast agents at same concentrations 3  re-shieldable targeting system based on the pH-responsive self-assembly/disassembly of AuNPs. In vitro analyses showed an appreciable X-ray attenuation rate for AuNPs@Re-GA, with the CT values of the samples linearly proportionate to the concentration of AuNPs@Re-GA. The ndings from preliminary CT imaging in Kunming mice with xenografted tumors disclosed that despite the CT images of tumors in both groups becoming gradually brighter, the AuNPs@Re-GA group performed better with a visual difference in the brightness compared with that of the control group. An ΔHUmax value of 41 was reported as the maximum contrast enhancement for the group treated with AuNPs@Re-GA, which was shown to be signi cantly higher (roughly 1.7-fold higher) than that of the control group (ΔHUmax = 23). Put together, the enhanced CT images of the tumors further indicated that the intelligent re-shieldable targeting system might be more .5HU for non-targeted), which then started to decrease slightly due to background metabolic processes. Surprisingly, the tumor HU values of the mice recorded with the RGD-Au-PENPs were remarkably greater than those recorded with the non-targeted Au PENPs at the same time points. The HU values of normal liver were also considerably greater than the orthotopic hepatic carcinoma at the same time juncture post-injection, since these particles were mainly cleared by the RES in the liver 9 . In another study, published in the same year, Zhou et. al. noticed that 3h post-injection, normal hepatic parenchyma (102.0±5.0HU) yielded more than 2.5 times a higher CT contrast enhancement than HCC (40.6±3.2HU). However, HCC showed a reduced tissue density as a result of the partial necrosis inside the tumor, hence, the lower CT value compared to that of the normal liver before injection 11 .
Qin et. al. developed a novel strategy based on "seed-mediated growth" that comprised both small and large mesopores in the shell and core, respectively, with xable Localized Surface Plasmon Resonance (LSPR) absorption and good colloidal stability in aqueous solutions, with an appreciably insigni cant cytotoxic impact. Furthermore, the PEGylated multiple gold nanoparticle-encapsulated dual-mesoporous silica nanospheres PEGylated MGNRs@DMSSs, displayed a better performance as contrast agents, resulting in an image enhancement (HU=267.4) that was signi cantly higher than that attained with PEGylated gold nanorods (GNRs) (HU=194.8) at the same Au concentration 83 . In 2019, Zhang et. al. designed "Janus nanoparticles" (JNPs) with a unique surface. They investigated a distinctive nanoplatform known as LA-GNR/zeolitic imidazolate framework-8 (LA-AuNR/ZIF-8), that comprised LA-modi ed metallic onedimensional nanorod/metal organic framework (1DNR/MOF) JNPs. With a one-dimentional (1D) nanostructure based on AuNRs conjugated with PAA on one side, and modi ed LA on the other, LA-AuNR/ZIF-8 was shown to be an effective contrast agents, and thus, a suitable platform for targeting tumor cells, owing to its remarkable effect on the image quality of CT (HU=758.8) 84 .
It is noteworthy that the enhancement of CT image quality is positively correlated with the content of contrast agent at the imaging site 81 . The retention of NPs in tumors could be enhanced by passive targeting (via the EPR effect) and active targeting, while both mechanisms work in concert to decrease the nonspeci c tissue biodistribution 81, 86 . However, tumor uptake of particles through active targeting has been reported to be more e cient than passive targeting. Therefore, a better quality of CT imaging can be attained with active targeting 87-89 . Traditional CT imaging agents such as barium (37.4keV) or iodine (Z=53, 33.2keV) have high X-ray absorption coe cients, nonetheless, they are known to be nephrotoxic 82 . In this regard, one may consider the conjugation, encapsulation or stabilization strategies in development of nanoplatforms facilitated with targeting ligands, imaging elements, functional moieties and antibiofouling agents 9,28 . Generally biocompatible materials, AuNPs provide greater contrast than iodinated contrast agents, owing to their higher atomic number (Z=79) and k-edge value (80.7keV). Furthermore, the X-ray attenuation of AuNPs was not signi cantly reduced in water. These ndings suggest a strong potential for AuNPs to serve as contrast agents in CT imaging 90 . According to an investigation by Boote et al, trivial concentrations of AuNPs, as low as a few hundred micrograms per every gram of background material (agarose), were able to result in the attenuation of X-Rays su cient enough to change the Hu value 91 .
In summary, we found that lower concentrations of AuNPs as contrast agents, compared with their traditional counterparts, result in better CT image quality and signal intensi cation. The elemental properties of gold in AuNPs are the major driving forces that positively affect CT contrast.

Conclusions And Future Perspective
Despite the developments in diagnostic methods, early detection of cancer is still a formidable challenge. Most of the patients are diagnosed with the disease at advanced stages with metastasis, where the current therapies are rendered ineffective. Hence, development of effective contrast agents is still a challenging task for precise early diagnosis. In this respect, increasing the accumulation of imaging probes at the tumor site and decreasing the unwanted uptake by normal organs or the immune system are key factors for manipulating the CT image quality in a desirable way. Based on novel approaches for development of nanocomplexes, the effective AuNPs with considerable X-ray absorption, long circulation time and biocompatibility have been developed for passive and active targeting aims in CT imaging of human HCC.
It can be concluded that application of gold in the form of gold nanoparticles can positively affect CT contrast compared with the more conventional contrast agents that are more heavily dependent on concentration and incubation time. Thus, application of AuNPs as contrast agents in CT could signi cantly enhance the image quality and provide better diagnostic results. Our systematic review showed that AuNPs could be applied for accurate and early diagnosis of malignant liver tumors and real-time monitoring of therapeutic response in near future. Different CT modalities, e.g., dual-energy CT, might as well bene t from the positive effects of AuNP-based contrast agents in the future.

Declarations
Ethics approval and consent to participate Application of AuNPs in theranostics.

Figure 2
Tree diagram of study selection based on PRISMA criteria