Early diagnosis of lung cancer using magnetic nanoparticles-integrated systems

2.1 Classification, synthesis, and functionalization of MNPs

MNPs are categorized into the following types:

1. Metals and their oxides (Fe, Co, Ni; Fe₃O₄, Fe₂O₃, NiO).

2. Alloys (FePt, FePd).

3. Ferrites (CoFe₂O₄, MgFe₂O₄, MnFe₂O₄, ZnFe₂O₄, NiFe₂O₄, CuFe₂O₄).

SPIONs belong to a ferrimagnetic group of MNPs with broad significance in the areas of biomedicine and bioengineering. While several SPION-based diagnostic methods are under clinical and preclinical investigation, others have been already commercialized for diagnostic purposes (Table 1). Fe₃O₄, γ-Fe₂O_3, and α-Fe₂O₃ are a few examples of SPIONs [20]. The presence of ferric ions (Fe²⁺ and Fe³⁺) has resulted in the dominance of SPIONs for utilization in biomedical applications. Different types of MNPs with their synthesis methods and unique features are elaborated, as shown in Table S1. The synthesis of MNPs has been achieved by diverse processes as listed in Tabe S2. A number of these processes have led to extraordinary properties like monodispersity, shape control, and stabilization. Co-precipitation or hydrothermal synthesis, as well as the sol-gel technique, are the most widely employed methods. Thermal breakdown, microemulsion, sonolysis, and biosynthesis are examples of some novel and modern techniques. The type of MNPs has a greater influence on the optimal synthesis pathway. The most used MNPs for diagnostic purposes are often spherical in shape. However, researchers have investigated different morphologies that imparted additional features to the nanoparticles such as hollow rod shape for drug delivery [80] and nanocubes for guided chemotherapy and photothermal therapy (PTT) [81].

2.2 Functionalization of MNPs

Functionalization of MNPs is required to improve their applicability for further applications. Functionalization is the process that involves the addition of new features, properties, and functions to a material by changing its surface chemistry [8,82,83,84,85]. This is a fundamental method used throughout materials science, chemistry, pharmacy, textile industry, bioengineering, nanotechnology, and biomedicine. Since the cores of MNPs are made up of toxic materials, hence it is required to reduce their toxicity for targeting and imaging purposes. To achieve this, the bare cores of MNPs are covered with biocompatible capping agents that act as shields/shells to lower their toxicity. For example, nanoshells that have been used as dielectric cores usually made up of silicon (Si), are coated with a thin layer of the shell (gold or Au) surrounding them. It has a size range between 10 and 300 nm [86]. The working principle of these nanoshells involves the conversion of electrical energy mediated by plasma into photon energy that can be optically tuned through UV-IR emission/absorption arrays. Although the large-size nanoshells have limited usage, they are desirable because of their imaging process and are being devoid of heavy metal toxicity [87]. Both active and passive ways of targeting cells by Au nanoparticles were demonstrated in the research conducted by Nunes et al. [87,88]. The working principle of passive targeting has been mediated by the assembly of Au nanoparticles to enhance imaging by permeability and retention effect (EPR) in tumors [88]. On the other hand, for active targeting, Au nanoparticles get coupled with tumor-targeted specific drugs. Earlier findings have significantly influenced the early diagnosis of cancer because these allow the detection of small tumors of millimeters (mm) size in the patient’s body [87]. By using a one-step solvothermal method, Qian et al. [89] synthesized Fe₃O₄ nanospheres/reduced graphene oxide (rGO) nanocomposites. In contrast to Fe₃O₄ microspheres/rGO and Fe₃O₄ nano-polyhedrons/rGO, the resultant nanocomposites had the most peroxidase-like activity toward the usual peroxidase substrate (3,3′,5,5′-tetramethylbenzidine) oxidation. This suggests that using their peroxidase-like catalytic activity, selective synthesis of Fe₃O₄/graphene composites with desirable properties is crucial. Based on their high activity, the Fe₃O₄ nanospheres/rGO nanocomposites were used to construct an economical colorimetric platform for the sensitive and selective detection of acetylcholine. Gao et al. [90] described the synthesis of MNPs coated with chitosan (CSMNPs) for use in a capture detection immunosorbent assay in which antigen may be collected, sorted, and enriched prior to the assay method. The carcinoembryonic antigen (CEA) was identified by CSMNPs with a LOD of up to 1 ng/mL. Chitosan was employed as both a ligand and a surface-modifying agent in a one-step solvothermal method to make CSMNPs. The presence of surface amine groups in CSMNPs provided high dispersibility in aqueous solution as well as suitable locations for covalently attaching antibodies to the MNPs. They also have catalytic characteristics for catalyzing color reactions in immunoassays, as well as magnetic properties for capturing, separating, and enriching antigens prior to the assay procedure. Table S3 in Supplementary shows different types of capping agents that serve as shell material to coat MNPs. The motif of coating MNPs with capping agents is to improve its applicability by preventing aggregation of MNPs and to provide protection to cells from toxic cores. This can be attained by the proper choice of shell materials that are hydrophilic in nature (e.g., silica) and allows diffusion of MNPs in a liquid environment. The three most important methods of functionalization include [8,82,83,84,85] the following:

1. Ligand addition and functionalization via bioconjugation.

2. Ligand-exchange strategy to MNPs.

3. Coating of hydrophilic materials like silica.

2.2.1 Ligand addition and functionalization via bioconjugation

The stabilization of MNPs can further be increased by coating organic and inorganic materials. The presence of hydroxyl groups (–OH) of nanoparticles produced by coprecipitation or other synthetic techniques might readily explain functionalization via bioconjugation at the MNP surface. Depending on the pH of the medium, it can be positively or negatively charged. If the –OH ligands are free, nonpeptizable particles will be produced at pH 7.5; however, ligands stay linked to MNPs in the pH range of 6–10. As a result, the free hydroxyl groups on the surface of MNPs help in the binding with biomolecules. Some of the clinically approved iron oxide MNP (IONP)-conjugated compounds with their applications are listed in Table 1, and the summarized form of MNPs as nanocarriers for cancer diagnosis is shown in Figure 2.

Figure 2

Classification of MNPs employed as carriers for drug delivery in cancer therapy [8].

IONPs/SPIONs are usually stabilized by the addition of stabilizers or surfactants like oleylamine and/or oleic acid that are generally hydrophobic. Now, these nanoparticles are often made hydrophilic for biomedical applications by functionalizing their surface via surfactant/ligand exchange or stabilizers [8,82,83,84,85]. The addition of surfactant is achieved through the adsorption of the amphiphilic molecules on the nanoparticle surface that contains both hydrophobic and hydrophilic segments [82]. A double-layer structural arrangement is formed by the hydrophobic component with the initial hydrocarbon chain, while there is an exposure of hydrophilic groups outside of the nanoparticles that render them to be dispersed in water easily.

3.1 Bronchoscopy

Bronchoscopy is an invasive technique that involves introducing a camera-connected illuminated tube (Bronchoscope) into the patient’s lungs through the nose or mouth [71]. As the bronchoscope exposes the interior tissues, bronchi, and bronchioles, it is possible to identify many abnormalities in the lungs, such as tumors, signs of infection, and excess mucus, using bronchoscopy. Rigid and flexible are the two different forms of bronchoscopy [71]. Rigid bronchoscopy uses a bigger, more rigid piece of equipment to access the proximal airways. It is usually done to remove foreign objects and airway stents in order to manage severe hemoptysis. Also, it has application in tumor debulking and airway dilatation [91]. Flexible bronchoscopy, on the other hand, accesses the lower airways using a smaller, more flexible piece of equipment. Bronchoscopy is commonly suggested for the early detection of LC because it provides a thorough review of distinct characteristics of the lungs from the inside [92]. Fiber-optic bronchoscopy is extensively utilized in practice because it is safe (0.12% complication rate) and simple to conduct [93].

3.2 Endobronchial ultrasound (EBUS)

EBUS is a minimally invasive technique that is used to identify a variety of lung diseases [71]. The interior tissues and the airwall of the lungs are seen using EBUS in conjunction with routine bronchoscopy [95]. EBUS is widely applied in normal practice for early identification because of its low risk and high diagnostic value. Linear EBUS and radial EBUS are the two EBUS systems available [93]. The ultrasonic transducer is utilized in the distal end of linear EBUS to visualize the curvilinear pattern of the airways using a fixed array. A mechanical radial microprobe is utilized in radial EBUS to visualize the characteristics of the lungs’ periphery. EBUS, in combination with frequent bronchoscopy, has been shown to be a highly successful technique for detecting the degree and depth of invasion in the LC.

3.3 Confocal endoscopy

Confocal endoscopy, also known as confocal laser endoscopy (CLE), is a cutting-edge imaging method that allows for real-time imaging of cellular and subcellular features in the mucosa and live cells in the lung tissue [71]. CLE might be used to investigate the architecture of alveolar elastic fibers, microvessels in lung tissues, and bronchus mucous membranes. For the reliable diagnosis of LC, Comino et al. [96] investigated probe-based CLE (pCLE) combined with computer-aided diagnostics techniques. The obtained data were used to study the deep-learning feature spaces to provide improved CLE pictures (83.4% accuracy) of lung tissue neoplastic cells.

3.4 Biopsy

Lung biopsy is a procedure that involves removing a sample of tissue from the lungs for evaluation. A specific biopsy needle or surgery is used to remove a tiny portion of the lung. A tissue biopsy is a sort of biopsy that is utilized to detect several lung diseases, including LC [71]. To identify lung disorders, many types of biopsies are available [71]. A needle biopsy, for example, involves inserting a needle into the lung and extracting a sample. A bronchoscope (a long, thin tube with a small camera) is used to perform a transbronchial biopsy. The thoracoscope, which transmits the chest picture to the computer monitor, is used in video-assisted thoracoscopic surgery. The thoracoscope is inserted into the chest cavity through a tiny incision in this minimally invasive procedure. Open biopsy, on the other hand, necessitates a bigger incision in the patient’s skin in order to access and remove a tiny section of the lung for further investigation. For years, tissue samples have been used to detect/diagnose LC; however, their invasive nature restricts its utility.

3.5 CT

In radiology, CT is used to obtain detailed pictures of the body in a noninvasive manner [12]. CT scan uses a spinning X-ray tube and detectors to measure the X-ray attenuations of various tissues inside the body. X-ray readings are often obtained many times from various angles and then processed on a computer to reconstruct the algorithms for creating tomographic pictures of the body. Contrast agents (iodine- and gadolinium-based contrast agents are the most common ones) for CT are administered intravenously. On the other hand, patients with renal impairment have been documented to be extremely susceptible to widely used contrast agents because they cause toxicity in cells and tissues. As a result, MNPs were considered as a possible replacement for iodine-based contrast agents [97,98].

In CT imaging, gold-coated iron oxide glycol nanoparticles can be employed as a contrast agent [12]. It was discovered that nanoscale MNPs are extremely biocompatible, biodegradable, and have X-ray attenuation properties, as well as pose a minimal toxicity risk when exposed to low levels over time. They are a good candidate for CT and MRI imaging because of these features [99,100]. Naha et al. [99] synthesized a nanocomposite of bismuth-IONPs with a dextran covering and tested the nanoconjugate’s cytotoxicity, accumulation, and half-life. HepG2 (human liver cancer cell line) showed less cytotoxicity, and CT imaging was used to examine the contrast in blood arteries and the heart following overnight incubation with nanoparticles. The nanoconjugate was discovered to have high X-ray attenuation, a long circulation half-life in the body, and biocompatibility, which makes it a promising candidate for use as a CT and MRI contrast agent.

Reguera et al. [101] synthesized Janus MNPs that were employed as a contrast agent in a range of techniques including CT, MRI, TEM, PAI (photo acoustic imaging), surface-enhanced Raman spectroscopy (SERS), and optical imaging. The results showed that these can collect the most cellular information, which make them an excellent imaging tool in the biomedical platform [101]. On the other hand, the negative consequences of ionizing radiation might occasionally limit their usage in patients.

3.6 PET

PET is an imaging technique wherein radioactive materials known as radiotracers are employed to envision and determine the changes in metabolic reactions [70]. Radioisotopes used in PET imaging possess a short half-life with ¹¹C of 20 min, ¹³N of 10 min, and ¹⁵O of 2 min. ⁶⁴Cu was widely studied due to its half-life up to 12.7 h. Researchers found that cross-linked dextran nanoparticles had a noteworthy affinity to tumor-associated macrophages [102,103]. ⁶⁴Cu-labeled dextran nanoparticles by macrocyclic chelators can be applied in PET imaging for clinical oncologic diagnosis. However, the poor stability of radiometal chelator complexes in vivo [104] greatly influenced the physicochemical properties of nanoparticles in PET imaging. Hence, the chelator-free ⁶⁴Cu nanoclusters were developed through a simple one-pot chemical reduction method by employing bovine serum albumin as a framework for PET LC detection to improve the stability and accumulation [105]. Although Cu-based radionuclide has been studied extensively and made some progress, the blemishes of its instability and higher accumulation in the liver are still the focus of lung diagnosis studies.

3.7 MRI

In vivo imaging of the cells is required to track and monitor disease treatment, immune cell distribution into the body, diffusion time, proliferation, and migration rate. PET, CT, and MRI are some of the imaging methods utilized for this purpose [12,70]. MRI, in particular, has been demonstrated to have a high spatial resolution [106], making it an excellent imaging method for in vivo imaging studies. It is a high-resolution imaging method that uses a strong magnetic field and radio waves to create pictures of numerous organs in human and animal bodies. In experimental contexts, it has been frequently utilized in clinical radiology with nonionizing radiation. In addition, the modality is tomographic, allowing for high-resolution soft tissue penetration. The magnetic field, radio waves, or electric fields are used in MRI to show the body’s precise interior structure. T1 and T2 contrast agents are the two types of MRI contrast agents [12]. T1 agents affect water protons’ longitudinal relaxation time, whereas T2 agents affect their transverse relaxation time. It shortens the T1 and has a mild influence on T2, resulting in a brighter picture when positive T1-weighted. Negative T2/T2*-weighted contrast, on the other hand, reduces T2 relaxation time and results in dark pictures [107]. Because of their superparamagnetic characteristics, MNPs lower the relaxation period of surrounding protons, making them good candidates for MRI contrast agents.

MRI cannot yet be directly applied in LC detection due to the motion artifacts, numerous susceptibility gradients, and low proton density [108]. By combining with the optimized proton MRI sequences based on ultrashort echo time (UTE), ultrashort echo-time magnetic resonance imaging (UTE-MRI) can be applied in lung tissue imaging [109]. Gadolinium is clinically used as the MRI contrast material, showing the noninvasive detection of NSCLC by UTE-MRI that can be achieved via the orotracheal administration of nebulized gadolinium nanoparticles with the enhanced signal. Moreover, gadolinium can be selectively deposited in tumor tissues while removed by healthy tissues [110]. Accurate and detailed detection information can be obtained through the simultaneous usage of two MRI contrast agents [70]. Longitudinal relaxation contrast medium (T1 contrast medium), such as Gd-DTPA and Mn-DPDP, and transverse relaxation contrast medium (T2 contrast medium), such as superparamagnetic iron oxide, among which T1 contrast medium can effectively decrease the T1 relaxation time by interactions with the neighboring T2 contrast medium [70]. The strong magnetic coupling between the T1 and T2 contrast medium could disturb the relaxation effect of the paramagnetic T1 contrast medium, leading to an undesirable weakening and quenching of the magnetic resonance signal. According to this, a smart MRI contrast agent with Fe₃O₄ nanoparticles in core (T2 contrast medium) and the silica shell containing water-soluble Mn-porphyrin (T1 contrast medium) and anticancer drug doxorubicin in the shell was constructed (Figure 3(a)) [111]. After the modification by poly(acrylic) acid (PAA) and c(RGDyK) peptides (cRGD), the dual-mode MRI contrast medium was equipped with functions of tumor-specific target and pH response (Figure 3(b)) [112]. When the contrast medium was internalized by cancer cells, the tumor acidic microenvironment facilitated the release of porphyrin and recovered the quenched signal caused by the combination of Fe₃O₄ and Mn-porphyrin (Figure 3(a)) [111]. Compared to single-contrast medium imaging, multimodal imaging with several contrast mediums can provide complementary imaging information for cancer diagnosis. Co-delivery of various contrast mediums without imaging signal interference is a great challenge. USRPs-Cy5.5 was constructed by covalently conjugating cyanine 5.5 on nebulized gadolinium nanoparticles for fluorescence tomography and UTE-MRI to detect LC noninvasively. Melanin nanoparticles in photoacoustic imaging (PAI) can be used as nanocarriers to codeliver¹²⁴I (PET contrast agent) and Mn²⁺ (MRI contrast agent) by an electrophilic substitution reaction and a chelation reaction, respectively, which is an ideal vector for tri-mode imaging (Figure 3(c)) to improve the efficiency of LC diagnosis at an early-stage effectively.

Figure 3

(a) Schematic presentation of synthesis, release, and imaging process of Fe₃O₄@SiO₂@PAA-cRGD. Reprinted from ref. [111] by Creative Commons License. (b) Drug release study using UV-Vis spectra. Reprinted from ref. [112] by Creative Commons License. (c) Schematic presentation of the synthesis of contrast agent and imaging mode (MRI/PET/PAI). Reprint from ref. [70] by Creative Commons License.

3.8 Magnetic particle imaging (MPI)

MPI is a noninvasive tomographic method for detecting tracer particles with superparamagnetic characteristics. Diagnostics, imaging, and material characteristics are all areas where MPI might be useful. MPI may be used to create a signal by using nonionizing radiation to find and quantify the number of nanoparticles at any depth within the body. When superparamagnetic nanoparticles are magnetized, MPI signals are created, which are 107 times more sensitive than MRI signals. When SPIONs were used for lactoferrin conjugation, the magnetic particle spectrometer (MPS) provides the potential for 3D in vivo imaging with excellent spatial resolution [113]. A custom-built MPS was used to record the MPS signal [113,114]. Overexpressed cancer cells release particular proteases like trypsin and matrix metalloprotease-2 that can be identified by MPS. SPIONs were aggregated in the presence of biotin-labeled peptides, which were then preferentially recognized and broken by particular proteases contained in the peptides, causing the SPIONs to disperse. MPS signals detected the aggregation or dispersion state of SPIONs based on magnetic relaxation properties. As a result, MPS may be utilized in biomedical applications including fast detection of proteases in biological materials like tissue extracts, urine, blood, and cell culture medium for LC diagnosis [115].

4.1 Biomarkers detection

In the area of LC detection based on MNPs, one vital step is biomarkers recognition. Cancer biomarkers are the molecular structures (aptamers, peptides, DNAs, proteins) responsible for disease development and its evolution. Therefore, the selection of these molecules is of crucial importance. Autoantibodies (AAbs), nasal signature, circulating tumor DNA (ctDNA), microRNA (miRNA), proteins, methylated DNA, and complement fragments are among the most common biomarkers [69,71]. There are several specific biomarkers depending on the type of LC, with the main biomarkers being CEA related cell adhesion molecule-1 (CEACAM), cytokeratin 19 fragment antigen 21-1 (CYFRA21-1), human epidermal growth receptor-2 (HER2), vasculo-endothelial growth factor receptors (VEGFR), epidermal growth factor receptor (EGFR), class 1 phosphoinositide 3-kinase (PI3K), neuron-specific enolase (NSE), and V-raf murine sarcoma viral oncogene homolog B (BRAF) [10]. Protein and peptide-based biomarkers identified in blood and bronchoscopic samples are the most commonly utilized and trustworthy biomarkers [116,117]. Different biological specimens, including blood, urine, sputum, saliva, and exhaled air, are used to extract and quantify biomarkers [10]. Biomarkers based on genetic materials, such as DNA, miRNA, and others, are primarily detected in blood, bronchoscopic samples, sputum, and nasal epithelium, whereas volatile organic compounds (VOCs) are mostly found in exhaled breath. Table 3 lists the various widely used biomarkers.

Various sampling methodologies for biomarkers detection are [10]

1. Liquid biopsy.

2. Sputum cytology (SC).

3. Breath analysis.

4.2 Microfluidic chip

The microfluidic chip is a well-organized, competent, efficient, rapid, and accurate diagnostic tool for LC patients. The manipulation of fluids on a microscopic scale is possible by using a microfluidic chip to control relevant parameters of cell culture [9]. This provides a better simulation of the tumor microenvironment in vivo. More specifically, it is possible to operate the cells precisely using the microscopic structure of a microfluidic chip and it is easy to conduct high-yield output analysis by the use of multiplexing microstructures [9]. The microfluidic control is beneficial to impersonate the internal fluidic environment, and its specific material characteristics can better mimic the microenvironment of tumor tissues. In precision medicine, the application of microfluidic chips can become a powerful assisting equipment to be realized in diagnostic forms [177], especially in the tumor organoid culture field single-cell sequencing, cancer biomarkers detection, and nanoparticles synthesis. In view of these functions, the combination of the microfluidic chip with downstream analysis can better identify cancer progression by studying its cellular, molecular, and biophysical properties [178].

Microfluidic devices have a remarkable prospect for development in the field of cancer metastasis because of their customizable properties, and they can meet the demands of various research works worldwide. These chips have rapid and high throughput, require a small volume of sample, have lower LOD, are flexible, are of small footprint, and are suitable for longitudinal collection. In the organoid culture field, microfluidic devices are used to culture two or more organoids to mimic cancer progression and can be applied to different types of cells. The separation of different organoids is conducted by the use of some definite biomaterials, like polydimethylsiloxane (PDMS), while the channels and controllable fluids are used to connect them with each other. The design and construction of a multiorgan microfluidic chip have been reported by Xu et al. [179] to mimic LC metastasis to different body organs such as the brain, bone, and liver. For this purpose, the organoids were divided into different chambers with an upper chamber for lung organoids and a lower for the other three organoids. Each chamber was seeded with various cell types to culture different organoids and the side channels were used to link each organoid. For simulation of blood circulation, the culture medium flowed through microvascular channels, and simultaneously a circulating vacuum was connected to mimic the physiological breathing as shown in Figure 5(a) [179]. This system helps effectively in exploring the complex process of underlying LC metastasis mechanisms. As depicted in Figure 5(b), another research reveals a microfluidic bone chip (BC) to explore metastasis of breast cancer to bone marrow.

Figure 5

Cancer models based on microfluidic chips; (a) a multiorgan microfluidic chip that mimics LC metastasis in the brain, liver, and bone; (b) cancer model to study breast cancer metastasized to bone marrow. Reprinted from ref. [9] by Creative Commons License.

Based on the simultaneous growth dialysis principle, there are two areas in the space of BC one for osteoblastic tissue growth and the other for culture medium flow [180]. This cancer model presented a natural bone microenvironment that allows the formation of an extraordinary thick layer devoid of artificial scaffolds by biomineralization of osteoblastic tissues or in vivo growth step. Moreover, researchers also developed osteoblastic tissue in the BC for co-culturing the metastatic human breast cancer cells and perceived various noteworthy characteristics of bone metastasis in breast cancer. Thus, the microfluidic BC has the potential to become powerful equipment in the in vitro study of bone cancer metastasis.

4.3 Biosensors

The functionalized MNP surfaces offer linkage groups to permit binding events with the complementary biomolecules. As discussed in Section 2.2.1, different strategies for bioconjugation include [8]:

1. Physical interactions (hydrophilic–hydrophobic, affinity interactions, and electrostatic interactions).

2. Chemical interactions (covalent bonds).

These interactions make immobilization on MNPs or transducers easy and simple. Specific interactions between protein and their recognition elements have shown their utilization in the development and designing of biosensors. There are different sensing/biosensing methods to determine the level of cancer biomarkers in blood, plasma, or diseased tissues. Some of them are listed below:

1. Electrophoresis.

2. Optical methods (fluorescence, electrochemiluminescence, colorimetric assay, SPR, SERS, etc.) (Table 4) [181,182,183].

3. Immunological methods (ELISA, polymerase chain reaction (PCR), etc.) [165].

4. Microcantilevers.

5. Electrochemical assay, electro-conductive, piezoelectric, and amperometric biosensors [184,185,186].

There are different labeling approaches to outline biosensing events depending on the method (electroactive molecules, fluorescent labels, nano/microparticles, enzymes, etc.). The following are the two primary techniques for integrating MNPs into the biosensor design [8]:

1. Direct labeling: In this approach, MNPs are immobilized at the transducing element via an affinity recognition reaction such as MNPs functionalized with single-stranded DNA as a capture probe is used to modify the sensor surface, and the hybridization reaction occurs when complementary oligonucleotides come into contact with the sensor, resulting in a physical/chemical response.

2. Indirect labeling: The ELISA technique is used in this procedure such as primary antibodies that are complementary to the targeted protein are immobilized on the sensing surface, followed by an affinity reaction with a biomarker-containing solution. Later on, secondary biotin-labeled antibodies are added to the system, allowing for an affinity interaction between streptavidin-labeled MNPs and the sensor surface.

Microelectronics’ contribution to sensor technology is helpful in the creation of devices that can detect LC biomarkers. Several biosensors, including optical and electrical conductance sensors, have been proven for the diagnosis of cancer-associated proteins, in recent years [10]. Simultaneously, the importance of investigating volatile metabolites that might be directly examined from the breath to identify LC is increasing [10]. Capuano et al. [187] conducted a survey of current state-of-the-art volatile chemical sensors and biosensors. CYFRA21-1 and NSE are two of the most reliable LC indicators [70]. These indicators are well-known for distinguishing between the two most common types of LC: NSCLC and SCLC. In serum having standardized immunosorbent tests, the identification of these compounds can be detected. However, such methods need the tagging of the target molecule in order to identify it. The fast label-free technique of detection can identify immunoselective interactions. Cheng et al. [188] developed a field-effect transistor biosensor in which the human antigen CFYFRA21-1 and NSE were covalently linked on the sensitive surface. Because of the lack of a labeling technique and the great sensitivity of the device, these markers can only be detected at concentrations of around 1 ng/mL. In a study, the development of an ultrasensitive aptamer for the diagnosis of A549 LC in human blood serum with high sensitivity (8 cells/mL) was reported by Wu et al. [189]. A new aptasensor for early detection and monitoring of LC was developed using cyanine dye and an S6 quadruplex aptamer with increased fluorescence. Furthermore, a recent study by Zhao et al. [190] reported aptamer cocktails for the CTC detection in LC by the use of microfluidic assay and the detection process employed aptamers. Aptamers are defined as highly specific, selective, sensitive, and precise techniques for LC monitoring with a scope for expansion in the field of future sensing. With the rapid application of biosensors in diagnostics and monitoring systems, IONPs/SPION- based biosensors are under investigation for their prospective in LC diagnostics. On the same lines, Kumar et al. [191] reported an electrochemical paper-based sensor using IONPs, and Yu et al. [192,193] designed an aptamer conjugated with thermally cross-linked SPIONs showing promising leads in terms of LC detection.

4.4 Nanotheranostics

Theranostics is a new field of biomedicine and bioengineering that specializes in disease diagnosis and therapy, particularly cancer [10]. Nanoparticles based on theranostics are becoming a more advanced platform for overcoming the limits of a traditional cancer diagnosis. For the real-time study of illnesses like cancer, theranostics combines several diagnostic imaging techniques with theranostic monomolecular agents. Multifunctional carbon nanodots for gene delivery in the treatment and diagnosis of LC were developed in recent research [194]. The photoluminescence characteristics of these nanodots are used to deliver siRNA in the treatment of NSCLC. These nanodots are injected into the body and deposit at the tumor site, emitting blue light fluorescence between 300 and 400 nm, and the siRNA contained in them induces tumor regression by apoptosis. This new theranostics nanoagent has shown promise in the detection and therapy of NSCLC [194]. Another intriguing study focused on the theranostic implication of boron neutron capture treatment and its effectiveness in LC [195]. The research involves the creation of boron/gadolinium-LDL adducts with detectable radioactivity and efficient cytotoxicity against lung tumors utilizing MRI. This enhancing approach may also be used to assess tumors with a diameter of less than 0.2 mm [195]. Researchers also synthesized a hyaluronic acid-based nanogel for PTT for LC diagnostics with continuous doxorubicin release. pH-responsive hyaluronic acid, photoresponsive graphene, and doxorubicin were used for the preparation of the nanogel. Graphene accumulates in tumor cells when this conjugate is injected into the body, and irradiation causes doxorubicin to be released sequentially from the conjugation. This is a very sensitive method of detecting and treating LC.

Another technique for early detection of LC involves the functionalization of MNPs collectively with antibodies and chemotherapeutic drugs [196]. Antibodies are highly specific for a particular antigen; thus, they will help in the targeted delivery of the drug. Photodynamic treatment, PTT, and magnetically induced hyperthermia have all been used with MNPs as a nanotheranostics technique for the diagnosis of LC [8]. These methods are used in the branch of medicine called oncology. The combined effect, on the other hand, typically confirms the most important diagnostic outcomes. This has been ascribed to MNPs’ modular architecture, which allows them to perform many functions. Figure 6(a) shows an example of how MRI was utilized to diagnose cancer early. Chemotherapeutic therapy combined with MRI resulted in better results [8]. The protocol used in the synthesis process regulates and changes the final characteristics of MNPs. The application of an external magnetic field inhibited MNPs’ clustering, and this magnetization was lost as soon as the field was withdrawn, preventing MNPs’ aggregation [8]. As discussed earlier, some MNPs have a magnetocaloric effect, in which the temperature of the MNPs changes in the presence of an external magnetic field. The magnetocaloric effect, along with MNPs’ high surface-to-volume ratio, allowed for efficient heat exchange with the environment. This enabled the development of the cancer treatment method known as “hyperthermia.” Figure 6(b) shows how MNPs’ unique features are completely subdued in tissue labeling and targeted drug delivery locations. The drugs were delivered to the precise target cell by placing the magnetic field outside of the anatomical structures and using an in vivo transportation approach [8]. MNPs absorb heat generated by electromagnetic waves in alternating cycles, making this magnetic material suitable for tumor detection in general. MNPs’ small size has a significant influence. The role of MNPs in LC diagnosis is linked to their use as a distinguishing agent in MRI and as a heating intermediary in hyperthermia (Figure 6(c)) [8]. Figure 6(d and e) further depicts the unique functions and properties of MNPs as a platform for antibody immobilization in the design of biosensors. Affinity ligands such as aptamers, hEGF, lectin, and folic acid have been associated with the surface of MNPs in order to guide them into tumors. As demonstrated in Figure 6(f), this promoted the accumulation of MNPs in a specific cell or tissue site [8,86,87,88]. The use of these strategies in conjunction with guided medication administration and magnetic hyperthermia has resulted in a synergetic effect in the accurate identification of LC.

Figure 6

Schematic presentation of various applications of MNPs: (a) MRI, (b) drug delivery, (c) magnetic hyperthermia, (d) as biosensors, (e) bioseparation, and (f) tissue engineering in biomedicine. Reprinted from ref. [8] by Creative Commons License.

Wang et al. [196] demonstrated successful coupling of synthesized MNPs with pan cytokeratin antibody, abbreviated as pan-ck Ab (Figure 7), as well as two additional types of prepared QDs conjugated with Lunx and SPC-A-1 antibodies. The simultaneous usage of QDs with double-labeled antibodies was applied for detecting the CTCs of NSCLC patients assembled by MNPs combined with pan-ck (MNP-pan-ck). MNPs and QDs were used to successfully construct a unique and creative approach for evaluating micrometastases of LC in peripheral blood. With its separation and visibility, the coupling of MNPs with QDs allows for the simple identification of specific cancer cells. Excess fluorescence multilabeling in combination with functionalized MNPs might be used to produce images with multicolor fluorescent molecules and magnetic variations. The conjugation of specific CTCs or CTCs [197] occurs when the anti-epithelial cell adhesion molecule antibody (EpCAM Abs) is immobilized. It depicts a schematic of micrometastasis detection in LC using the combined influence of MNPs and QDs [196]. It was found that QDs with double-labeled antibodies successfully identified LC cells A549 and SPC-A-1. As a result, a total of 32 cases of NSCLC were discovered. Twenty-six patients had improved CTCs, and 21 patients were effectively identified by QDs. As a result, a novel approach was developed in which CTCs were collected using MNP-pan-ck, and CTCs from NSCLC patients were identified using QDs with double-labeled antibodies.

Figure 7

Schematic presentation of recognition and separation of tumor cells using MNP-coupled pan cytokeratin antibody and QDs with double-labeled antibody [196] (PBMC: peripheral blood mononuclear cells). Reprinted ref. [196] under Creative Common License.

4.5 AI

Many aspects of the healthcare industry have been profoundly affected by AI and ML [14]. Technology advancements have paved the path for cost- and time-effective analysis of large datasets. AI is proving to be beneficial in clinical oncology and research [198]. NGS systems were introduced to meet these needs and they have changed the future of precision oncology [199]. NGS has numerous clinical uses, including risk prediction, early disease detection, diagnosis by sequencing and medical imaging, accurate prognosis, biomarker identification, and therapeutic target identification for new drug development. NGS creates enormous datasets, which necessitate the use of specialist bioinformatics tools to evaluate the data that is therapeutically useful. Cancer diagnosis and prognosis prediction are improved using NGS and high-resolution medical imaging, thanks to these AI applications. Regardless of technological advancements, AI’s pattern recognition and sophisticated algorithm skills may be used to get important clinical information, reducing diagnostic and treatment mistakes [200]. In cancer, ML is a useful technology with many applications in precision medicine. Complex neural networks can predict disease and treatment results by generating diagnostic pictures and genetic analysis data [200,201]. Deep learning algorithms are employed in the healthcare sector for the curation of vast amounts of data and the creation of improved data-driven diagnostics. Biomarkers are found in medical imaging and can be used to screen patients for cancer. Image analysis entails identifying the image of interest as well as significant sections of the image. Data from datasets and the results of patient screening can be utilized to detect malignant tumors automatically. Using AI algorithms, we can thus customize our treatment options in the event of a cancer diagnosis [14,198]. Deep learning is the most often used AI technique in radiomics, a field of computers that extract diagnostic images to discover malignant tumors that are not visible to the naked eye. The combined efforts of radiomics and deep learning will improve diagnostic picture analysis accuracy [201]. When AI and ML are used in healthcare, they have the potential to transform illness management and deliver effective medical care. AI research has progressed to the point where it can make decisions in a human-like fashion (Figure 8). AI can improve the sensitivity, repeatability, and accuracy of tumor detection not just through automated segmentation but also through the rapid increase in computing speed and by enhancing AI algorithms. It is conceivable that a separate segmentation analysis of suspicious images will become obsolete. Consequently, AI systems can evaluate image data from the whole body. In this sense, a full-body approach will allow a more detailed study of the organs’ properties that can be modified by pathological processes but are invisible to human sight. This means that these algorithms could be used in various aspects of early LC diagnosis, such as information from serological studies and tissue biomarkers.

Figure 8

Application of AI in cancer imaging to create better data-driven diagnoses using deep learning algorithms.

In the search for surrogate markers of LC, researchers have looked at biomarkers derived from body fluids and tissues such as whole blood, plasma, bronchial lavage, urine, sputum, and biopsy specimens. Studies have shown that CTCs, AAbs, miRNAs, EB biomarkers, and blood proteomic profiling are promising molecular candidates for the early detection of LC [69,70,71]. Finally, high-throughput technologies such as epigenomics, transcriptomics, and metabolomics are also being explored as potential indicators of early LC [202]. The combination of numerous “omics” with data from medical imaging can give useful information for understanding human disease, including LC (Figure 9).

Figure 9

AI algorithms have the ability to enhance the early detection of LC by combining numerous data sources [13].

ML and deep learning algorithms have been effectively employed in many research contexts to combine diverse omics, imaging, and clinical data [203,204]. Using deep learning models to combine imaging and omics data might aid in early LC diagnosis. Using “radiogenomics,” researchers have discovered that high-dimensional features gained from CT scans are connected to the existence of particular mutations in tumor tissues [205]. For example, specific CT scan imaging characteristics, such as anaplastic lymphoma kinase (ALK), EGFR, c-ros oncogene 1 (ROS1), KRAS, and rearranged during transfection proto-oncogene (rearranged during transfection), have been linked to the presence of tumor driving mutations in LC [205,206,207,208]. As a result, the extraction, integration, and interpretation of the large volume of data generated by AI-based technologies will be supplemented in a cancer screening program to help its early diagnosis.

4.6 Wearable devices

Wearable materials and devices are one of the specialized areas of therapy and diagnostics that are progressing. In terms of point-of-care diagnostics, smartphone-linked applications and sensors, the advent of mobile health have the potential to alter the current healthcare landscape. With more than 95% of the population having access to mobile networks [207], telehealth and mobile health are gaining popularity among academics, and the WHO has formally defined them. A unique self-renewing sensor for detecting VOCs through breath and skin was created in a recent study. This new gold nanoparticle-based sensor has the ability to self-heal injured tissue caused by abrasion [209]. Despite the enormous hurdles that this sector of wearable sensors in oncology faces in terms of development, a large range of smartphone-based applications is accessible [210]. In addition, Cancer Research UK, in partnership with the British Thoracic Society, released “The Pulmonary Nodule Risk App,” a smartphone-based tool developed for clinical oncologists to be used for LC evaluation and monitoring [211]. In the realm of LC diagnosis, this area of wearable equipment is underutilized.

The integration of the IoT with cloud computing, big data analytics, and their potentialities in healthcare has resulted in the development of smart health monitoring systems, which have the potential to advance almost every domain of science and technology, and their fusion with nanotechnology is considered the next evolutionary step of the 21st century [15]. For early detection of LC, several real-time-based nano-enabled sensors employ biomarkers found in breath, blood, saliva, and perspiration. Cloud connectivity is used in these smart healthcare monitoring systems because it provides processing, storage, and data analysis, which aids in informing medical authorities at the right moment. Google has announced the Cloud Healthcare Application Programming Interface, with the goal of making the process of collecting, storing, and accessing healthcare-related data easier for health organizations [15]. Google recently inked a big cloud computing contract with Flex, a publicly traded electronics company that makes medical device components all around the world. Various sensors have been embedded in watches, tattoos, wristbands, cellphones, belts, and soft lenses to gather, monitor, and transmit bodily motions, SPO2 levels, glucose levels, pulse rate, and blood pressure to authorized individuals such as physicians, health businesses, and consultants. Furthermore, by integrating differential pulse voltammetry and cyclic voltammetry, a smartphone-based integrated voltammetry system was created to detect uric acid, ascorbic acid, and dopamine in the body, as shown in Figure 10 [212]. rGO and gold nanoparticles were used to modify the electrodes. The detector turned the analog impulses into digital signals and sent them over Bluetooth to the smartphone. The findings demonstrated that the smartphone-based system could efficiently detect several biomolecules at the same time and that it has promise in point-of-care testing.

Figure 10

The operational process of the smartphone-associated voltammetry system. Modification of screen-printed electrodes to employ it as the sensor for detection of various samples [15,212]. (Redrawn with permission from Ji et al.[212].).