Emerging Multifunctional Carbon‐Nanomaterial‐Based Biosensors for Cancer Diagnosis

Despite significant advancements in medical technology, cancer remains the world's second‐leading cause of death, largely attributed to late‐stage diagnoses. While traditional cancer detection methodologies offer foundational insights, they often lack the specificity, affordability, and sensitivity for early‐stage identification. In this context, the development of biosensors offers a distinct possibility for the precise and rapid identification of cancer biomarkers. Carbon nanomaterials, including graphene, carbon nitride, carbon quantum dots, and other carbon‐based nanostructures, are highly promising for cancer detection. Their simplicity, high sensitivity, and cost‐effectiveness contribute to their potential in this field. This review aims to elucidate the potential of emerging carbon‐nanomaterial‐based biosensors for early cancer diagnosis. The relevance of the various biosensor mechanisms and their performance to the physicochemical properties of carbon nanomaterials is discussed in depth, focusing on demonstrating broad methodologies for creating performance biosensors. Diverse carbon‐nanomaterial‐based detection techniques, such as electrochemical, fluorescence, surface plasmon resonance, electrochemiluminescence, and quartz crystal microbalance, are emphasized for early cancer detection. At last, a summary of existing challenges and future outlook in this promising field is elaborated.


Introduction
According to the World Health Organization, cancer refers to a wide range of diseases that may originate in any organ or tissue of the body.3][4] The nature of cancer is the rapid proliferation of abnormal cells with the ability to spread to other organs, [5][6][7] and its development generally follows several steps.First, normal genes in cells are mutated and become cancer-related genes or oncogenes; this process is also known as proto-oncogenesis.The formation of oncogenes through mutation or replication of normal genes is known as the activation of oncogenes.Subsequently, these activated oncogenes increase cell division, cell growth deregulation, and cancer formation, while tumor suppressor genes are downregulated.[13][14] In the year 2020, there were 19.3 million new cancer cases diagnosed and over 10 million recorded cancer-related deaths worldwide.Among these instances, female breast cancer constituted the most frequently diagnosed type at approximately 2.3 million (11.7%), followed by lung cancer (11.4%), colorectal (10.0%), prostate (7.3%), and stomach (5.6%) cancers.[17] Despite recent medical and technological advancements, cancer morbidity and mortality rates remain high, mainly due to late-stage diagnoses and poor prognosis.[26] Conventional cancer diagnosis methods typically employ noninvasive nuclear-based imaging techniques, including magnetic resonance imaging (MRI), computed tomography (CT), ultrasonography, positron emission tomography (PET), and single-photon emission computed tomography.Moreover, invasive biopsy and histopathological examination are employed to identify cancer types and determine their stages. [27]Although these methods have provided the foundation for cancer diagnosis and treatment, they are nonspecific, costly, and heavily dependent on the growth rate of cancer tumors. [27]In addition, a wide variety of techniques within the field of molecular biology, such as enzyme-linked immunosorbent assay, [28] radioimmunoassay, [29] immunohistochemistry, [30,31] flow cytometry, [32] and DNA/RNA-based hybridization/sequencing approaches, [33][34][35][36] have also been widely used to detect molecular signatures/biomarkers in cancer cells.Although current molecular diagnostic techniques have been clinically validated, they frequently exhibit low sensitivity, extended detection times, potential health risks, and the requirement for sophisticated equipment and skilled operators.[39][40][41] Therefore, developing novel detection approaches with efficient nanomaterials that are simple, sensitive, and rapid for early-stage cancer diagnosis is of enormous clinical significance.
In the field of biosensing, the choice of materials becomes crucial not only for their intrinsic properties but also for their adaptability to interface with various biomolecules. [42]To this end, carbon nanomaterials have a distinct edge: their carbonaceous nature lends itself to a plethora of functional groups that can be introduced either during synthesis or through postsynthetic modifications.Such inherent versatility allows for conjugating a diverse range of biomolecules directly onto the carbon-based scaffold.[47][48][49][50][51] Furthermore, the diverse allotropes of carbon, from graphene to carbon nanotubes, offer varying configurations and densities of these functional sites, [52,53] ensuring that a suitable carbon nanomaterial with the optimal density and orientation of functional groups can be chosen based on the specific requirements of the biosensing applications.
To date, biosensor development is becoming a prevalent approach for effectual point-of-care testing with high-level accuracy, reproducibility, reliability, stability, affordability, and disposability. [54,55]In particular, biosensor research is deemed as a required field to detect the toxins and cancer cells in the blood, and numerous researchers have developed biosensors to sense the different types of biomarkers responsible for cancer and other diseases. [56][71][72][73] This ease and compatibility of functionalization, combined with their intrinsic properties, make carbon nanomaterials uniquely suited for biosensing applications. [74]e high tunability of the surface chemistry for carbon nanomaterials enables them to be specific with target biomolecules, further enhancing their sensitivity and selectivity as diagnostic tools. [75,76]Carbon-based nanostructures have been extensively employed in biosensors and therapy fields, [77][78][79] which are essential for food, agriculture, and environmental monitoring applications. [80,81][84][85] Thereby, carbon-based biosensors play an essential role in healthcare fields nowadays, especially in cancer research, where they are exploited to detect the concentration of the multiple targets/biomarkers responsible for particular cancer diseases.[91] Besides, other carbon nanomaterials, such as porous carbon, [92] graphic carbon nitride (g-C 3 N 4 ), [93,94] fullerene, [95] and carbon nanotubes, [85] also present a massive potential for biosensing applications.
Even though several reviews have explored the potential of carbon-nanomaterial-based biosensors in medical diagnostics, [85,89,[96][97][98][99] this field is highly active, and a large amount of research papers have been published recently.The number of publications on the three most predominant carbon nanostructures, i.e., graphene, carbon dots, and carbon nanotubes, depicts a booming trend in the last years, as shown in  performance, advantages, and limitations of carbon nanomaterials across these mechanisms.A unique highlight of our work is the emphasis on the structure-property-performance correlation of carbon nanomaterials for suitable sensor architectures and high selectivity/sensitivity, guiding researchers toward the future design of high-performing biosensors.To begin with, we elaborate on various prevailing biomarkers and cancer diseases, followed by introducing different sensing mechanisms and recent experimental realizations based on carbon-based nanomaterials.The advantages of various carbon nanomaterials for different technologies have been emphasized. [100,101]At last, we conclude with an in-depth discussion and a summary of the challenges associated with developing a practical technology for carbon-nanomaterial-based biosensors.

Types of Cancer Biomarkers
Biomarkers are biomolecules that have been widely used as identifiers for particular physiological processes or disease conditions. [102]They can be found in biological fluids, including blood, serum, saliva, or tissues, making them essential tools for screening or diagnosing cancer. [103]In the field of oncology, biomarkers have numerous potential applications, such as evaluating risk, conducting screenings, distinguishing between different diagnoses, determining prognosis, predicting treatment responses, and tracking disease progression. [102,103]otably, with the advancement of analytical technology, a variety of new biomarkers have been developed and serve as diagnostic, predictive, and prognostic markers in cancer detection. [103]Based on the molecular constituents of cancer biomarkers, this review will specifically focus on genomic biomarkers and protein biomarkers (Figure 3).

Genomic Biomarkers
A genetic biomarker is an indicator for analyzing the changes in deoxyribonucleic acid (DNA) or ribonucleic acid (RNA) in normal biological and pathogenic processes. [104]DNA biomarkers  have been extensively employed to analyze DNA profiles and single nucleotide polymorphisms (SNPs) with abilities to identify the exact variations in genomic DNA. [105,106]The main advantage of DNA biomarkers is the ability to detect somatic mutation in patients efficiently.Somatic mutation or DNA sequence variations frequently occur in oncogenes, tumor suppressor genes, and mismatch repair genes. [107]Therefore, DNA biomarkers, such as BRCA1 and BRCA2, can quickly determine these changes in DNA sequences related to disease states. [108]On the other hand, RNA biomarkers are well known as transcriptomic biomarkers since they are able to detect the expression of RNA profiles covering a set of RNA molecules generated in the cells, such as mRNA, rRNA, tRNA, and other noncoding RNA (i.e., microRNA). [109,110]They represent a novel category of cancer-specific biomarkers that are extensively used in fingerprint identification and clinical trials.The main advantages of RNA biomarkers lie in their solid ability to analyze the variation in posttranscriptional regulation of gene expression.Examples of RNA biomarkers include microRNAs such as miR-21, miR-122, miR-16, and miR-155. [111]

Protein Biomarkers
Protein biomarkers are of paramount significance due to the ubiquitous presence of proteins and their critical involvement in the metabolic and essential processes of both normal and tumor cells. [112]Advanced mass-spectrophotometry-based proteomics has unveiled numerous potential protein biomarkers in the literature.Despite this, a small portion has obtained approval from the United States Food and Drug Administration (FDA). [113]Compared to DNA and RNA biomarkers, protein biomarkers are often recognized with high sensitivity and specificity for low-concentration detection. [114]As a result, these biomarkers perform a vital role in disease recognition and dramatically slow the progression of the disease. [115]n the early diagnosis of cancer, varieties of protein biomarkers have been exploited experimentally to detect disease states. [116]For instance, alpha-feto protein (AFP), [117] carcinoembryogenic antigen (CEA), [118] carbohydrate antigen 15 (CA 15-3), [119] prostate-specific antigen (PSA), [120] and New York esophageal squamous cell carcinoma-1 (NY-ESO1) [121] are typical examples of protein biomarkers in cancer diseases. [122]Among these, AFP is a glycoprotein with a size of 70 kDa, primarily produced in the fetal liver and yolk sac toward the end of the first trimester of pregnancy.AFP is found in the fetal serum with a concentration of 3000-5000 μg mL À1 .A high level of AFP is found in newborns but deficient in healthy adults. [123]After the age of 2 years, the expression of albumin-like AFP is turned down with a drop in serum concentration of 5 Â 10 À4 -1.5 Â 10 À2 μg mL À1 . [124]FP is the most frequently utilized biomarker to detect hepatocellular carcinoma, the leading primary liver malignancy. [125]FP is used to detect multiple cancers, including pancreas, lung, colon, and ovaries or testicle cancer, during the diagnosis and treatment.It can also help determine the disease stage and observe the treatment response. [126]The high level of AFP in serum is frequently linked to the presence of tumor. [127,128]EA is a cell surface glycoprotein frequently employed in clinical trials.It is a 201 kDa glycoprotein secreted through the intestinal lumen, expressed at the apical portion of colonic epithelial cells, and released through the colonic lumen.CEA comprises a single polypeptide chain containing 641 amino acids, with lysine located at the N-terminal position. [129,130]It is a glycoprotein usually produced by gastrointestinal tissue during fetal development. [131]CEA can be expressed on the cell surface and circulate in the bloodstream. [132]The average range of CEA in the blood is <2.5 ng mL À1 .This protein produced by abnormal cells leads to cancer in a particular group. [131]A high level of CEA could indicate certain types of cancers, like rectum, prostate, colorectal, ovary, lung, and liver cancer.Grunnet et al. illustrated that CEA was a tumor antigen first identified in tumor tissue extracts and found in fetal gastrointestinal tract epithelial cells. [133]

Most Common Types of Cancers and Their Biomarkers
Cancer is generally divided into five stages (0-IV). [134]Stage zero (0) denotes cancer cells confined to the origin site and not extended to neighboring regions. [135]This stage is usually treatable and is examined precancerous by health professionals.In stage I, cancer is limited to a small area without involvement of lymph nodes or adjacent tissues.Stage II is categorized as early-stage cancer, indicating local growth but no spread.Stage III is characterized by significant growth and potential spread to lymph nodes or surrounding tissues.Notably, stage IV is referred as metastatic or advanced cancer, and it has already spread to different organs or regions of the human body. [134]ifferent stages of cancer are also accompanied by the release or expression of specific biomarkers, cells, and metabolites in various body fluids, which can be used for early and rapid cancer diagnosis. [136]As an effective detecting approach, biomarkers have significant potential in clinical use for estimating the risk of the disease and screening the primary stage of cancer or early diagnosis. [137]As mentioned in Section 2.1 and 2.2, numerous genomic and protein biomarkers have been attempted for cancer diagnosis.In this section, we discuss in detail five common types of cancers, i.e., breast, ovarian, prostate, lung, and pancreas cancers and corresponding biomarkers that have been used to detect or confirm the presence of these diseases.

Breast Cancer
Breast cancer results from the abnormal proliferation of breast tissue cells.Typically, it originates from the epithelial lining cells of the ducts, lobules within the glandular tissue of the breast, or the surrounding connective tissues, and it has the potential to spread rapidly to the adjacent breast tissues. [138]Invasive ductal carcinoma and invasive lobular carcinoma are the prevailing types of breast cancer.The former involves cancer cells spreading beyond the ducts into surrounding breast tissue, while the latter sees cancer cells spreading from lobules to other breast tissues.It is worth noting that in 2020, breast cancer affected over 2.3 million women, resulting in approximately 685 000 deaths worldwide. [139]Thereby, effective diagnoses are warranted so patients can access treatment earlier and have improved prognoses.
BRCA1 and BRCA2 serve as major prognostic biomarkers in breast cancer detection. [140]In many cases, the overexpression or mutation of BRCA1 and BRCA2 genes hints at the formation of breast cancer.These genetic biomarkers are mainly applied to predict high-risk hereditary breast cancer. [141]More importantly, genetically unstable BRCA genes in which physical deletion of wild-type alleles can result in breast cancer. [142]BRCA1 is a tumor suppressor gene or anti-ancogene that contains a protein of 1863 amino acids, which control transcriptional activation, DNA repair, apoptosis, and chromosomal remodeling.Similarly, the BRCA2 produces a protein of 3418 amino acids that helps to repair damaged DNA and modulate transcriptional activation and cell growth. [143]The inactivation or physical deletion of BRCA2 leads to the tumor by duplication of the mutant allele, gene conversion, and mitotic recombination. [142]iomarkers linked to breast cancer at the protein level encompass CA, CA 15-3, CA 27-29, CEA, estrogen receptor (ER), progesterone receptor (PR), and HER2. [30]Among them, CA15-3 and HER2 are standard biomarkers for breast cancer patients.CA15-3 is the most often utilized biomarker for patients with symptoms. [144]It is a carbohydrate-containing protein biomarker called Mucin, a large transmembrane glycoprotein, highly O-linked glycosylated protein with extracellular domains consisting of highly conserved 20 repeated units of amino acids. [145]A 27-29 is a breast cancer-associated antigen produced by the MUC 1 gene.It is worth mentioning that up to 80% of women's cancer is due to overexpression of CA 27-29 protein. [146]ER2 is usually expressed in organs' epithelial cells, and it is also known as Neu, featuring an extracellular binding domain, a single transmembrane domain (E), and an intracellular tyrosine kinase region. [147]The receptor of HER2 takes part in the proliferation of cells and cell-to-cell communication through signal transduction that affects the transcription of genes by the phosphorylation process. [148]Notably, HER2 can be found in breast tumor cells at a low level, in which HER2 receptor protein increases the cell growth and malignant growth of breast cancer cells overexpressed in %20-30% of breast cancers, enabling it to breast cancer detection in the earliest stages. [149]Therefore, HER2 protein is considered a significant predictive biomarker for monitoring the presence and aggressiveness of breast cancer cells.

Ovarian Cancer
Ovarian cancer typically originates in one of the three types of ovarian tissue, namely epithelial, stromal, or germ cell tissue, with epithelial tumors accounting for 90% of all ovarian cancer cases. [150]Ovarian cancer protein biomarkers comprise CA-125, osteopontin, CEA, vascular endothelial growth factor (VEGF), and human epididymis protein 4 (HE4).Moreover, genetic markers involve germline variations in both BRCA1 and BRCA2 genes, which may substantially increase ovarian cancer's lifespan threat. [151]RCA 1 and BRCA 2 biomarkers are also tumor suppressor genes in ovarian cancer as a mutation of these genes may increase the risk of ovarian cancer progression.[152] CA-125 is overexpressed and presents abnormally in ovarian cancer patients.[152] To date, the CA-125 biomarker blood test has become a routine tool for ovarian monitoring and high-risk group detection, in which a high level of CA-125 indicates a high possibility of ovarian cancer.[153] Nowadays, researchers have already achieved a low limit of detection (LOD) of CA-125 smaller than 26 U mL À1 .[154] Nevertheless, the exact detection rate is still unclear because the majority of the patients have a low value at early detection.In contrast, HER4 is a standard biomarker providing high sensitivity and specificity for diagnosing and progressing ovarian cancer with a detection limit of 55 pmol L À1 .[155] Additionally, HER4 protein was known to induce tumor progression in both breast and ovarian cells, [156,157] and its serum biomarker was used to diagnose early-stage ovarian cancer.60] VEGF stimulates endothelial cell proliferation and acts as a mediator of angiogenesis in this cancer, with a relatively low LOD of 0.5 pg mL À1 .[161,162]

Pancreatic Cancer
Pancreatic cancers primarily occur in two main types: exocrine pancreatic cancer and neuroendocrine pancreatic cancer. [163,164]denocarcinoma, commonly known as ductal carcinoma, is an exocrine cancer that predominantly arises in the lining of the pancreatic ducts.On the other hand, pancreatic neuroendocrine tumors (NETs) originate from cells in the pancreatic endocrine gland. [165]In this regard, the cancer antigen 19-9 (CA19-9) test becomes the standard diagnostic blood test for pancreatic cancers.The high expression level of CA19-9 in the blood is an indicator for patients with pancreatic cancer symptoms. [166]CA19-9 and pulse amplitude modulation 4-level (PAM4) are specific protein-based biomarkers for pancreatic cancer.Some other serum protein biomarkers are linked to the pancreatic disease diagnosis, including MUC1, MUC4, CEA, CAM1, MMP-9 (OPG), osteopontin (OPN), CA 242, CA 50, CEA, CA 72-4, and CA 494. [167]mong these biomarkers, CA19-9 has been approved for pancreatic cancer detection by the FDA. [168]According to Thapa et al., the lowest detection limit of CA19-9 is around 37 U mL À1 , providing sensitivity of 70-90% and selectivity of 68-91% in the serum sample. [169]Similarly, CA 242 is an indicator for pancreatic cancer diagnosis and gives a sensitivity of 74% and selectivity of 9 L%, which is slightly lower than CA19-9. [170]P53, deleted in pancreatic carcinoma 4 (DPC4), p16, and BRCA2 are a few common signature genetic biomarkers for pancreatic cancer, in which they detect the mutation of particular genes such as chromosomal instability and telomere abnormalities in most patients. [171]

Prostate Cancer
In Australia, prostate cancer stands as the third most common cause of death attributed to cancer and remains the second most prevalent disease among men. [19,20,22,172]It is classified as either acinar adenocarcinoma or ductal adenocarcinoma.Acinar adenocarcinoma originates from gland cells that line the prostate gland, whereas ductal adenocarcinoma tends to spread at a faster rate than acinar adenocarcinoma, which arises in cells lining the ducts of the prostate gland. [173]PSA and prostatic acid phosphatase (PAP) are protein-based biomarkers specifically associated with prostate cancer.PSA is usually present in body fluids, and a high PSA level can hint at prostate cancer. [174]Therefore, PSA is utilized as the primary tumor biomarker for the early diagnosis of prostate cancer in clinical settings.Typically, a blood PSA level higher than 4.0 ng mL À1 is considered high risk. [175]On the other hand, genetic markers of prostate cancer include prostate cancer antigen 3 (PCA3), p53, and glutathione-S-transferase P1 (GSTP1). [176]bnormal cancer cells circulating in the bloodstream are used as a detection modality, which allows for a shorter detection period for tumor cells in the blood. [177]E-cadherin is a serum-based biomarker for prostate cancer, and it acts as a mediator of cell-to-cell adhesion or cell-cell signalling. [178]Leman and Getzenberg reported that numerous genes were involved in cell adhesion and controlled prostate cancer cell growth, in which E-cadherin showed relatively low activity in prostate cancer than in normal tissue. [178]Typically, the mutation of tumor suppressor genes such as p53 and phosphatase with TENsin homology (PTEN), which regulate cell growth, can contribute to the advancement of prostate cancer.

Lung Cancer
Lung cancer is one of the most prevalent forms of tumors and continues to be the primary contributor to cancer-related deaths, responsible for 18.4% of all total fatalities. [21,179]Lung cancer can be classified into two main types: small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC).NSCLC is the more common type, accounting for over 80% of cases.It is further divided into three subtypes: large cell carcinoma, adenocarcinoma, and squamous cell carcinoma.Conversely, SCLC comprises about 20% of cases and can be categorized into subtypes, including small cell carcinoma, mixed small cell/large cell carcinoma, and combined small cell carcinoma. [180]The leading cause of lung cancer is smoking tobacco and drinking alcohol, although environmental pollutants and genetic factors also play a role in the incidence of lung cancer. [181]everal noninvasive approaches have been devised for diagnosing lung cancer, including ultrasound, MRI, PET imaging, low-dose helical CT scans, and bone studies.Invasive approaches such as sputum cytology, bronchoscopy, and needle biopsy are also employed for diagnosis. [180]These conventional methods, however, cannot detect lung cancer at an early stage because their detecting principles rely on the tumor's phenotypic properties.Biomarkers have emerged as promising tools for early-stage detection of lung cancer, and a diverse range of lung cancer biomarkers have been documented in scientific literature.Genomic-based biomarkers encompass various types, such as cyclooxygenase-2 (COX-2), interleukin-8 (IL-8) mRNA, death-associated protein kinase (DAPK), polypyridyl complexes 3 (PRCS3), p53 mutant, and epidermal growth factor receptor (EGFR), with particular emphasis on c-ErbB-1 and c-ErbB-2. [182]Neuron-specific enolase (NSE), progastrin-releasing peptide (ProGRP), cytokeratins 21-1 (CYFRA 21-1), epithelial cell adhesion molecule (EPCAM), CEA, and VEGF are essential protein biomarker for lung cancer diagnosis. [183,184]Okamura et al. demonstrated that the sensitivity and specificity of the CEA biomarker could be up to 69% and 68%, respectively, while these performance values of CYFRA 21-1 were 43% and 89%, respectively. [185]The minimum detectable levels for NSE, ProGRP, CYFRA 21-1, and VEGF were 0.05, 63, 0.32 ng mL À1 , and 31.25 pg mL À1 , respectively.Additionally, CA 125 and CA 19-9 are also utilized in lung cancer diagnosis. [182]ng cancer treatment is known for its unique challenges due to factors such as early detection, stage at diagnosis, and treatment options.Recently, the synergistic combination of biosensors and biological markers has played an important role in lung cancer diagnosis and treatment. [186]Zamay et al. demonstrated that CEA, CYFRA 21-1, and ProGRP could serve as clinical biomarkers for lung cancer diagnosis, while they could not be functional alone for lung cancer detection due to their low concentration level in serum. [187]To this end, the combination of two biomarkers (e.g., CEA and CYFRA 21-1) could significantly enhance the sensitivity and specificity for disease detection. [188]

Multifunctional Carbon-Based Nanomaterials for Biosensing
Biosensors are a type of high sensitive device capable of detecting various biological analytes. [189]According to the International Union of Pure and Applied Chemistry (IUPAC), a biosensor is a self-contained integrated system with a bioreceptortransducer system, capable of delivering quantitative analytical information through bioreceptors like enzymes, DNA probes, and antibodies. [190,191]The bioreceptor is a biomolecule that recognizes the target analyte, and the transducer can convert the recognized information into quantitative signals, as demonstrated in Figure 4. [192] However, the conventional methods of detecting biomarkers, like immunological assays and biopsies, suffer from low sensitivity, long detection time, and are hazardous to health with the necessity of sophisticated instruments and trained operators.Therefore, the development of highly effective biosensors for cancer diagnosis is a major challenge.
Recently, nanomaterials, especially carbon-based nanostructures, have been highly effective for sensing cancer biomarkers because of their increased specific surface area, surface binding sites, and electrochemical properties. [193]The surface active sites of these nanostructures can be used for selectively binding/ sensing different biomolecules, including antigens, peptides, and other cancer-specific biomarkers.In this section, we will discuss various carbon-based biosensors, including ECL, electrochemical, fluorescence, QCM, and SPS biosensors.

Electrochemiluminescence Biosensors
ECL, also known as electrogenerated chemiluminescence, is a highly efficient analytical technique employed in various fields. [194]ECL is based on the generation of electro-active species near the electrode surface, followed by high-energy electrontransfer reactions that lead to the formation of excited states, resulting in light emission. [195]As the light is produced from the biomolecules (luminophores), the concentration of biomolecules can be detected from the intensity of emission light. [196,197]uminophores can be the atom or functional group in a chemical compound responsible for having luminescent or light-emitting properties and are typically loaded on biomolecules like antigens and antibodies. [198]Semiconductor QDs, Ru(bpy) 3 2þ , luminol, and their analogs are frequently employed as ECL luminophores.Compared to other sensing methods, ECL is a sensitive and potential method that can efficiently analyze cancer biomarkers in clinical samples. [199,200]The high sensitivity, low background, and ease of control are the signature characteristics of this type of biosensor. [201]Over the past decade, ECL techniques utilizing carbon nanomaterials have been extensively investigated owing to their numerous advantages.These nanomaterials demonstrate remarkable conductivity of electricity, consistent electrochemical properties, and a comparatively high specific surface area, contributing to their high ECL activity.Additionally, the availability of surface functional categories and defects in their lowdimensional structures enables versatile functionalization, making them a novel class of luminescent electrode materials suitable for quantitative analysis of biomolecules such as DNAs, enzymes, and cancer proteins. [202,203]

Graphene-Based ECL Biosensors
As the most famous carbon nanomaterial, graphene has been most investigated for ECL-based cancer biosensors.The contribution of graphene boosts the performance of the ECL sensing system originating from its extraordinary physical properties. [204,205]In many cases, graphene may increase the surface area of sensing platforms, improve chemical processes (such as the binding of biomolecules), speed up the passage of electrons across electrode interfaces, and amplify electron signals during sensing. [206]The efficient binding of antigen-antibody takes place on the graphene surface layer and exhibits highaccuracy detection of particular analytes. [207]These properties enable graphene for high-accuracy sensing in ECL immunosensors and as carriers to load ECL labels on biomolecules.For example, Wu and co-workers constructed SnO 2 /reduced graphene oxide (rGO)/gold nanoparticles (AuNPs) composites for determining the presence of insulin, using the SiO 2 @Polydopamine as the ECL signal quencher. [208]In this system, SnO 2 /rGO behaved as the support material and helped to load a high density of AuNPs owing to its high stability, good reversible capacity, and high specific surface area, while AuNPs were responsible for combining with antibodies by forming Au-N covalent bond.
Graphene can also act as an amplifier that improves the rate of electron transfer at the electrode.In this case, the graphenemodified electrode in the ECL biosensing system promotes a higher sensitivity of cancer biomarkers than other detection techniques. [209]For instance, Zhang et al. described an ECL biosensor utilizing graphene functionalized with a gold-silver nanocomposite, in which graphene substantially increases the surface area, enabling greater binding of primary antibodies and accelerating electron transfer rates. [210]Furthermore, the secondary antibodies labeled with CdTe QDs-coated carbon microspheres amplified ECL intensity and stabilized luminescence properties, enabling accurate and highly sensitive CA 125 detection with satisfied linear relationships ranging from 0.008 to 50 U mL À1 .Similarly, Wang et al. introduced an ECL sensor with CdSe QDs/GO-chitosan for cytochrome C detection. [211]In this system, the high ECL sensitivity stems from the synergistic effects of three components: GO guarantees high stability and conductibility, and its porous structure not only enables a great rough particular surface area to load significant quantities of CdSe QDs but also promotes the diffusion of K 2 S 2 O 8 into the interface of electrode and accelerate reaction; chitosan (CS) with amino and hydroxy groups enhances the stability of the composites because of its excellent water permeability and good biocompatibility; CdSe QDs reacts with K 2 S 2 O 8 and generates ECL emission.The fabricated ECL sensor exhibited a high sensitivity for cytochrome C spanning from 4.0 to 324 μM, coupled with LOD 1.5 μM.
The ECL-based biosensor, employing a graphene/Au-CdS:Eu QDs architecture, effectively detected AFP biomarkers as well. [212]ndeed, the ECL immunosensor adopts a dual-quenching approach, with horseradish peroxidase molecules on Au/Ab2 initiating the redox interaction involving substrate hydroquinone and H 2 O 2 , resulting in ECL intensity quenching as the concentration of co-reactant H 2 O 2 decreases.Additionally, the ECL resonance energy transfer (ECL-RET) phenomenon takes place between the Au and CdS:Eu QDs, resulting in additional ECL intensity quenching owing to the spectral overlap between the two components.Specifically, the emitted light from rGO/ Au/CdS:Eu QDs (donor) is reabsorbed by Au nanorods (acceptor). [213]Consequently, this ternary system exhibited the linear range of AFP detection between 0.05 pg mL À1 and 1 ng mL À1 at LOD of 0.05 pg mL À1 .It also demonstrated a 96% to 110% restoration of AFP in serum.Similarly, Nasrollahpour et al. conducted a label-free ECL immunosensor assay on a rGO and CS composite to detect the presence of the HER-2 protein in breast cancer. [214]Initially, rGO was electrodeposited on the surface of a glassy carbon electrode (GCE) using cyclic voltammetry (CV).Subsequently, a biopolymer solution of CS/[Ru(bpy) 3 ] 2þ was coated on the modified electrode, with CS covalently interacting with rGO.Monoclonal antibodies (Abs) specific to HER-2 were attached to the amine groups of CS/[Ru(bpy) 3 ] 2þ /rGO/GCE through amide bonds using EDC/NHS chemistry (Figure 5a).The efficiency of the rGO electrode was evaluated, and ECL signals were recorded with/without rGO electrodeposition (Figure 5b).The presence of rGO significantly enhanced ECL intensities, while the absence of rGO led to very low ECL intensities, indicating substantial quenching of emitted materials.The various steps in the electrode separation process of the ECL sensor are depicted in Figure 5c.The ECL intensity decreased when the Abs and the target HER-2 protein (1 fM) were introduced onto the electrode, which can be attributed to steric hindrance that reduced the ECL intensity due to the tripropylamine (TPrA) obstruction from reaching the electrode surface.The ECL calibration curve, obtained using GCE/rGO/CS/[Ru(bpy) 3 ] 2þ , included seven different concentrations of HER-2 protein (ranging from 1 fM to 0.000001 nM) (Figure 5d).The sensitivity of the detection is determined by the slope of the calibration curve (y = À45.233Xþ 36.633)(Figure 5e).This result demonstrated that the developed technique exhibited high sensitivity in detecting HER-2 protein.The prepared immunosensor was validated by analyzing several real samples collected from breast cancer patients at different stages.To assess the reliability of the system, the obtained results were compared with pathological examinations, as shown in Figure 5f. [214]1.2.g-C 3 N 4 -Based ECL Biosensors Apart from graphene, g-C 3 N 4 with highly flexible electronic structures was also intensively investigated for ECL biosensor applications.[215][216][217][218] In 2014, a pioneered work by Li et al. first employed g-C 3 N 4 as a luminophore to construct a label-free g-C 3 N 4 /graphene-based ECL biosensor for squamous cell carcinoma antigen (SCCA) detection.[219] In this study, g-C 3 N 4 was closely anchored on the surface of graphene through electrostatic interactions, where graphene was utilized to increase the sensitivity of the ECL sensor further.The introduction of SCCA and its corresponding antibodies resulted in a linear reduction of ECL intensity, encompassing the span from 0.025 to 10 ng mL À1 , and featuring an LOD of 8.53 pg mL À1 .
Shortly after this work, Wu et al. developed an innovative ECL biosensor of the "in-electrode" type to detect SCCA, employing AuNPs/g-C 3 N 4 and nanoFe 3 O 4 @GO. [220]In this system, the nanoFe 3 O 4 @GO and the SCCA primary antibodies (Ab 1 ) played the role of capture probe, and AuNPs/g-C 3 N 4 labeled the Ab 2 sary antibodies behaved as the signal tag (Figure 6a).In this case, AuNPs enhanced the ECL intensity of g-C 3 N 4 thanks to the increased conductivity and accelerated electron transfer rate between the electrode and g-C 3 N 4 .After Fe 3 O 4 @GO catches SCCA at the Ab 1 site, the Ab 2 -AuNPs/g-C 3 N 4 in the immunocomplex can be conjugated sequentially with the antigen and thereby fixed on the surface of the electrode.The immunological reaction between SCCA and its antibody demonstrated high  [214] Copyright 2021, The Authors, published by Springer Nature.
sensitivity and selectivity, yielding a low LOD of 0.4 pg mL À1 , as shown in Figure 6b.Additionally, the same research group designed an ECL immunosensor based on g-C 3 N 4 for the detection of CA125, a marker linked to ovarian cancer.Through the application of amino-coated Fe 3 O 4 nanoparticles and anti-CA125 onto single-use screen-printed carbon electrodes, they enabled efficient electron transfer between g-C 3 N 4 and the electrode, resulting in a notable augmentation of ECL intensity. [221]ruitful achievements have been reported in g-C 3 N 4 -based nanocomposites ECL systems, including promising LOD values. [222]For example, the CA125 could be detected by Anti-CA125-nanoFe 3 O 4 @g-C 3 N 4 composite at concentrations between 0.001 and 5 U mL À1 , with an LOD of 0.4 mU mL À1 . [223]dris et al. discovered that g-C 3 N 4 nanosheets/gold electrodes could form an ultrasensitive ECL immunosensor for the detection of DNA with LOD of 0.001 fg mL À1 . [224]Similarly, Liu et al. designed an ECL sensor utilizing graphitic QDs (g-CNQDs)/AuNPs, demonstrating an impressive LOD of 0.01 fM for DNA. [225]Undoubtedly, g-C 3 N 4 is a potential ECL luminophore and can open an avenue for future biosensor applications.

Carbon-Dot-Based ECL Biosensors
Carbon dots (CDs) are another effective ECL luminophores widely used in detection technologies because of the carboxylic acid groups on their surface.The inherent fluorescence of CDs can be significantly suppressed due to significant electron transfer transpiring at the interface between biomolecules and CDs.This characteristic renders CDs highly promising for biomolecule detection, such as nucleic acids, amino acids, and vitamins, as well as various biomarkers like cancer cells, cancer antigens, and microRNA. [226]Remarkably, they can play the roles of both luminophores and quenchers.In most cases, CDs serve as novel luminescent reagents owing to their ECL properties.For example, Chai's team fabricated a novel ECL biosensor using amino-modified CDs and TiO 2 NPs functionalized with AuNPs (CDs-Au-PEI-TiO 2 ) for miRNA-21 detection. [227]Original CDs-Au-PEI@TiO 2 composites presented a strong initial ECL signal due to the effective immobilization of CDs by PEI@TiO 2 as support, while the target miRNA-21 presence repeatedly caused hybridization of H1, which significantly reduced the ECL intensity with the miRNA-21 concentrations between 0.1 fM and 10 pM, with an LOD of 0.03 fM.In a separate investigation, Sun and his team integrated graphene QDs with molybdenum disulfide (MoS2-GQDs) to design an ECL biosensor with an "on/ off" switch, enabling the detection of specific DNA sequences. [228]Interestingly, the combination of 2D MoS2 nanosheets and long-chain hairpin DNA significantly increased the loading capacity of GQDs.During the ECL process, charge transfer occurred between GQDs and MoS 2 , resulting in the "on" phase.Simultaneously, exonuclease III and target DNA were more easily bound by DNA walker cyclic amplification, leading to increased ECL intensity.On the other hand, DNA2-Fc-DNA1 served as a quencher for the ECL signal, resulting in an "off" phase.The newly developed ECL biosensor had an outstanding  [220] Copyright 2016, Elsevier.c) Schematic representation of S-g-CNQDs dual-band ECL biosensor mechanism.d) Relationship between the ECL intensity and different concentrations (0, 50 fM, 0.1, 1 pM, 0.01, 0.1, 0.5-1 nM at 555 and 620 nm) of K-RAS gene.e) The corresponding linear relationship of the ECL intensity and K-RAS gene concentrations.c-f ) Reproduced with permission. [229]Copyright 2019, American Chemical Society.linear detection range of 1 nM to 100 aM and a remarkable LOD of 25.1 aM.In 2019, Zhang et al. constructed a wavelengthdependent dual-band biosensor by using sulfur-doped graphitic phase g-C 3 N 4 QDs (S-g-CNQDs) to identify the K-RAS gene, a genomic biomarker linked to pancreatic cancer, using surface plasmon coupling ECL (SPC-ECL) detection. [229]As depicted in Figure 6c, the ECL signal at 620 nm is similar to common g-CNQDs or other semiconductor materials, which is related to its conduction band and valence band.However, another ECL signal at 555 nm should be ascribed to the sulfur doping vacancies, which generate an adding energy trap state positioned above the valence band of QDs, effectively accepting holes.Since an improved spectrum overlapping of 555 nm peak with the surface plasmon absorption peak of AuNPs, the ECL intensity at this wavelength shows a much higher enhancement (Figure 6d), demonstrating a wider linear range between 50 fM and 1 nM, with an LOD of 16 fM at 555 nm, compared to these of 0.1 pM to 1 nM and 30 fM at 620 nm (Figure 6e).
Notably, CDs can also act as quenching materials in ECL systems.For instance, in 2020, Yang's team first employed CDs to dampen the ECL signal of a Ru(bpy) 3 2þ -tripropylamine system. [230]In this case, CDs connected with complementary DNA were attached to the electrode surface and acted as a quenching probe.Without the target miRNA 142-3p, CDs-DNA stayed at the electrode and caused weak ECL intensity; with miRNA 142-3p, on the other hand, DNA could bind with targets and form DNA/RNA duplexes.More importantly, DNA would be hydrolyzed afterward, and the target RNA could be released again and bind with other DNAs.This repeated hydrolyzed DNA process can effectively remove CDs-DNA composites from the electrode surface, thus resulting in a much-enhanced ECL intensity and a sensitive platform for detecting microRNA 142-3p with a linear range between 10 fM and 1.0 nM, with an LOD of 10 fM.Table 1 summarizes various reports on ECL biosensors based on carbon materials for cancer detection.

Electrochemical Biosensors
The electrochemical biosensor broadly works on the principle of the change in biological signal converting to the electrical signal.This technique is particularly efficient for the protein biomarkers immunoassays due to their size, ease of use, high sensitivity, and cost-effectiveness. [11]The electrochemical biosensor system consists of three major components: 1) A recognition element or bioreceptor that interacts with the analyte; 2) An electrode that works as a signal transducer, which displays a change when the analyte interacts with the recognition element; 3) An electronic system which assists in the data management. [231]

Graphene-Based Electrochemical Biosensors
Graphene remains one of the most sought-after materials because of its strong mechanical strength, flexibility, theoretically high surface area, good electrical, and thermal conductivity. [232]eing a 2D carbon allotrope, the 2D surface of graphene forms a base for the adsorption of particular biomolecules.Owing to these merits, graphene has been intensively used for the detection of cancers. [233]The sensitivity and selectivity can be finely enhanced using techniques such as oxygen plasma treatment, graphene hybrids formation, heteroatom doping, covalent bonding with dienophiles, and non-covalent bonding with pyrene derivatives. [234]In most cases, graphene behaves as a conductive template to immobilize the bio-recognition units in the biosensor system.AuNPs have dominated the biosensor field thanks to their adequate chemical stability, biocompatibility, and high activities, making them excellent scaffolds with graphene for biosensor construction. [235]For example, Chen et al. introduced a sandwich-type electrochemical biosensor using carboxyl graphene nanosheets to detect both CEA and AFP biomarkers concurrently immobilized on a CS-Au nanoparticles (CHIT-AuNPs) modified electrode.The sensor exhibited LOD of 0.1 and 0.05 ng mL À1 for CEA and AFP, respectively. [236]The combination of graphene and AuNPs also increases the electron transport for effectual antibody loading and offers a large surface area for antigen binding. [237]By self-assembly on the GCE, grapheneconjugated AuNPs were developed for AFP biomarker detection. [238]Also, rGO and AuNP composites were developed on the surface of the electrode for impedimetric detection as DNA biosensors, with both complementary and noncomplementary sequences. [239]In a study by Bai et al., a label-free electrochemical biosensor for DNA/miRNA detection was developed utilizing AuNPs-toluidine blue-GO nanocomposites, and the biosensor demonstrated an impressive BOD of 2.95 pM. [240]Apart from DNA detection, label-free electrochemical biosensors have also been exploited for miRNA detection.By employing AgNPs in polyaniline and N-doped graphene, a LOD of miRNA of 0.2 fM has been achieved. [241]n another work, Wei et al. established an electrochemical detection technique for VEGFR2 protein identification. [242]This study modified the working electrode with CS-functionalized rGO to significantly increase the electrical conductivity for detecting VEGFR2 protein by electrochemical biosensing platform (Figure 7a).The effect of CS in RGO was analyzed and confirmed by transmission electron microscope (TEM), as demonstrated in Figure 7b.The rGO without CS encountered significant agglomeration problems, while the chitosan-rGO sample with a CS membrane enclosing a few layers of graphene nanosheets exhibited no agglomeration.This structural analysis demonstrated that CS plays a crucial role in enabling the aqueous dispersion of rGO through electrostatic interactions.As depicted in Figure 7c, the peak current of pulse voltammetry (DPV) has demonstrated an increase with the ascending VEGFR2 protein concentration from 0.4 to 86 pM.The reduction current peak showed a logarithmic proportionality to the VEGFR2 protein concentration (Figure 7d).The VEGFR2 protein exhibited a wide concentration, from 0.4 to 86 pM, with a high correlation coefficient of 0.999 and an LOD of 0.28 pM for VEGFR2 protein at a signal-to-noise ratio of 3. The interaction between the drug and VEGFR2 involves stabilized hydrophobic interactions and polar residues, playing a significant role in drug stabilization through H-bonds and electrostatic interactions, as illustrated in Figure 7e.Additionally, the interaction between sorafenib and the tyrosine kinase region of VEGFR2 becomes more specific and selective due to the extra hydrogen bonds involved.
Different biomarkers have been successfully detected by graphene-derived systems.For example, a reusable GO-modified  [242] Copyright 2014, Springer Nature.
electrode was fabricated to detect VEGF in the blood sample, which exhibited a low detection of 31.25 pg mL À1 . [243]In this case, GO acts as an efficient carrier for the Avastin antibody, which in turn alters the amperometric signal to detect the analyte.Likewise, Sharafeldin et al. fabricated the electrochemical biosensor by forming GO-Fe 3 O 4 nanocomposites to detect PSA and PSMA cancer biomarkers.The GO-Fe 3 O 4 hybrid generated electrical signals, enabling the detection of PSA and PSMA biomarkers at LOD of 15 and 4.8 fg mL À1 , respectively. [244]

g-C 3 N 4 -Based Electrochemical Biosensors
Thanks to the photoactive nature, high surface area, tunable porosity, and facile synthesis nature, highly sensitive electrochemical biosensors have been realized based on g-C 3 N 4 .Recently, Low et al. fabricated a photoelectrochemical biosensor wherein AuNP and g-C 3 N 4 were cast on the indium tin oxide (ITO) electrode surface for recognizing microRNA-319a. [245]he sensing unit consists of anti-DNA:RNA antibodies as a target analyte biorecognition unit and g-C3N4 as a photoactive nanomaterial.The developed biosensing system demonstrated outstanding identification sensibility, achieving an LOD of 2.26 fM. [246]A similar electrochemical nanobiosensor system based on g-C 3 N 4 / AuNPs has also been utilized to detect the PSA biomarker with a low detection limit. [247,248]In another study, aptamer molecules were cast on the GCE surface with g-C 3 N 4 /AuNPs, providing high sensitivity and an LOD of 1.6 pg mL À1 for the VEGF biomarkers. [249,250]In another report, Luo et al. demonstrated AuNPs/hexagonal carbon nitride tubes (HCNT)-based photoelectrochemical biosensor for HER2 breast cancer biomarker detection.251a] The excellent electrochemical properties of g-C 3 N 4 enable it in other applications beyond cancer sensing.Organic dyes find widespread application in food preservation, bakery products, and beverages.Tartrazine, an azo dye commonly utilized in commonly consumed foods, has been implicated in causing mutagenic and carcinogenic ailments.Consequently, the need for monitoring these chemicals has become crucial in sensor technology.Karimi et al. [251b] exhibited the straightforward pyrolysis of melamine to fabricate g-C 3 N 4 sheets, serving as a sensitive material for detecting tartrazine.A strong catalytic response in the reduction of tartrazine was achieved through the direct adsorption strategy on the g-C 3 N 4 nanosheets, examined using CV and DPV techniques.The notable finding from the study indicated that a thin film of graphite/g-C 3 N 4 nanosheets could enhance the electrocatalytic effect, while a thick film might impede the response.The sensor's sensitivity, stability, and real-time applicability were validated using saffron fake powder. [252]Phenolic compounds, including phenol, catechol, quercetin (QR), vanillin, and resveratrol, are largely produced as by-products in various chemical industries such as pharmaceuticals, antioxidants, cosmetics, agricultural pesticides, dyes, leather, petrochemicals, and textiles.Regrettably, these compounds pose significant toxicity to various biological species, human health, and the environment, leading to complications resulting from their degradation.Therefore, the development of electrode materials for monitoring the levels of these phenolic compounds is considered crucial.Selvarajan et al. presented a composite material of g-C 3 N 4 /NiO for the electrochemical detection of QR.The g-C 3 N 4 was synthesized by heating melamine to 550 °C for 4 h, followed by ultrasonication and mixing with nickel chloride to achieve even distribution and form NiO on the g-C 3 N 4 nanosheets.The peak current of the electrocatalytic oxidation of QR increased from 0.010 to 250 mM, with LOD of 0.002 mM achieved through DPV analysis.Upon analysis with real samples, the recovery results demonstrated the high feasibility, selectivity, and sensitivity of the sensor electrode material for electrochemical QR sensing.The electrochemical detection of QR was performed using a material consisting of silver nanoparticles deposited on porous ultrathin g-C 3 N 4 nanosheets (AgNPs@g-CN). [253]

Carbon-Dot-Based Electrochemical Biosensors
CDs are utilized as efficient electrochemical biosensors because of their excellent electronic properties, low intrinsic toxicity, high solubility, and good biocompatibility. [254]In a typical work, Gao and co-workers fabricated a CDs-based electrochemical biosensor using polyamidoamine dendrimers capped-CDs (PAMAM-CDs) and Au nanocrystal nanocomposites. [255]The amine groups in graphitic PAMAM-CDs interact with each other to create PAMAM-CDs/Au nanocrystal composites, enabling AFP detection with an impressively low LOD of 0.025 pg mL À1 .Furthermore, the impedimetric approach was employed using bimetallic ZrHf metal-organic framework composites with CDs (CDs-ZrHf-MOF) to detect HER2 and MCF-7 breast cancer cells.The LOD were 19 fg mL À1 and 23 cells mL À1 for HER2 and MCF-7 cells, respectively. [256,257]Table 2 summarizes typical electrochemical biosensors performance based on carbon-based nanomaterials and their detection limits for various biological markers of cancer.

Fluorescent Biosensors
In recent years, the fluorescent biosensor is becoming increasingly popular as a sensing platform for detecting cancer biomarkers in real samples. [258]The fluorescent biosensor is an optically based sensor that measures the changes in analyte concentration through changes in fluorescence.They are mainly separated into two categories: turn-on and turn-off, wherein the analyte causes an increase or decrease in the fluorescence of the probe, respectively.Compared to conventional optical sensors based on the change in the optical absorbance with the addition of analytes, fluorescent sensors exhibit minimal interference arising from the analyte itself.Fluorescent biosensors have gained significant importance over the years due to the advent of multiple fluorescent agents, such as carbon QDs, metallic nanoparticles, fluorophores, and polymeric nanoparticles.Fluorescent biosensors are highly user-friendly due to their quick turnaround time for the detection of proteins, enzymes, glucose, food toxins, metal ions, and medicine. [259]Fluorescent sensing converts the change in the biological recognition system to a change in electrical signals.The fluorescent signal transparency depends upon the usage of fluorescent dyes, fluorescent protein, and fluorescent nanoparticles. [260]luorescence resonance energy transfer (FRET) involves energy transmission between the donor and acceptor fluorescent molecules.In the excited state, the donor transfers its energy to the acceptor through nonradiative dipole-dipole couplings. [261]he energy transfer occurs as the donor chromophore from a higher excitation state transfers the energy to the acceptor via dipole-dipole interactions. [262]FRET forms the basis for many fluorescent sensors, in which the change in fluorescence signal is recorded to demonstrate its selectivity with respect to the acceptor molecule.The fluorescent biosensor has been mainly used in the medical field to detect disease-specific biomarkers, offering an easy and potential diagnosing technique for identifying various cancer biomarkers. [263]66]
reducing the background signal and increasing sensitivity. [270]ompared to other molecular quenchers, graphene showed a high quenching efficiency of d À4 , where d denotes the separation between the fluorophore and the graphene. [271]The finely tunable surface of graphene to incorporate extended functionalities empowers it to obtain excellent antigen-antibody specificity.Furthermore, although graphene is inherently not fluorescent, its 0D counterpart, GQDs, are a kind of fluorescent material, and they have the merits of being simple to synthesize, stable, quick to internalize into tissues, and biocompatible. [254]By coupling GO with a molecular aptamer beacon probe and employing enzyme-assisted signal amplification, the detection of VEGF165 and ATP at concentrations of 1 pM and 4 nM, respectively, was successfully achieved.GO serves as a super-quencher, efficiently reducing the high background signal. [272]Shi et al. designed a GQD-based FRET sensor conjugated with graphene oxide; through careful regulation of their interaction, the synthesized system exhibited an excellent LOD of 75 pM toward DNA. [273]3.2.g-C 3 N 4 -Based Fluorescence Biosensors g-C 3 N 4 , as an excellent fluorescent nanoprobe, is able to directly interact with the analyte of interest to quench the fluorescence, so the analyte concentration is subsequently determined.[274] Notably, g-C 3 N 4 nanosheets have a large surface area, enabling them to bind with DNA targets and multicolor dye-labeled DNA probes to develop the DNA biosensor.[215] In a typical case, the fluorescence quenching ability of g-C 3 N 4 assisted in the detection of 15-mer DNA fragments and 18-mer DNA fragments, with an LOD of 75 and 62 pM, respectively.[275] g-C 3 N 4 has not been explored much in the fluorescent sensor route due to the relatively lower fluorescence.However, the nitrogen content, along with rich carbon, can act as a very efficient platform for the donor and acceptor electrons to interact with the analyte in various fields.For example, many g-CN nano architectures emit blue fluorescence that may potentially disrupt biological selffluorescence, leading to a degradation of the actual signal.Acknowledging this, Liu et al. introduced a ratiometric fluorescence aptasensor utilizing phenyl-doped g-CN nanosheets, which display robust green fluorescence as an internal reference fluorophore.This system operates through a PET quenching effect on 6-carboxy-X-rhodamine fluorescent dyelabeled anti-adenosine aptamer, facilitating the detection of adenosine.[276]

CDs-Based Fluorescence Biosensors
CDs are extremely small carbon-based nanomaterials (usually in a size range of <10 nm), which exhibit excellent fluorescence due to the quantum confinement effect.CDs are signal tags that are frequently used in cancer biomarker detection.CDs have been proven to efficiently label with capturing antibody/primary antibody against MCF-7 surface protein, which is responsible for breast cancer. [30]Incorporating secondary antibody/detecting antibody labeled with magnetic beads permits their magnetic separation to achieve fluorescence emission spectra.In one typical study, CDs-MnO 2 composites were used to detect miRNA155 and MCF-7 breast cancer cells, demonstrating notably low LOD of 0.1 aM and 600 cells mL À1 , respectively.This was achieved through the incorporation of CDs onto the DNA probe, with MnO 2 serving as the quenching agent. [277]Similarly, Ghadareh et al. constructed a fluorescence biosensor based on CDs conjugated with PAMAM-Dendrimers/AuNPs. [278] The CDs were attached with the CA125 antibody to determine the CA125 biomarker, demonstrating a relatively low LOD of 0.5 fg mL À1 . [279]n another case, Wu fabricated a fluorescence biosensing system using CDs with silica nanoparticles (CD-SNPs) and fluorescein isothiocyanate (FITC).The antibody was labeled with FITC and used to detect the AFP biomarker at an LOD of 0.317 μg dL À1 . [280]Ds with Au nanocomposites were also used to detect the CA19-9 pancreatic cancer tumor marker, with notable high sensitivity and selectivity.The anti-CA19-9 biomarker was labeled with horseradish peroxidase enzyme in this study and immobilized on the CQDs/Au nanocomposite surface. [281]he operational mechanism of fluorescent biosensors based on monoclonal NSE antibody (anti-NSE)/amine-N-GQDs-AuNPs for detecting small cell lung cancer biomarkers is depicted in Figure 8. [282] In this study, amine-functionalized nitrogen-doped GQDs were synthesized via a hydrothermal method initiated by citric acid and diethylenetriamine, resulting in an average size of 3 nm.These GQDs were then covalently conjugated with anti-NSE.The detection of NSE biomarkers relied on the nano surface energy transfer mechanism between AuNPs and the antibody-conjugated GQDs.In the beginning, the blue emission of the anti-NSE with the GQDs system experienced quenching (OFF state) due to the presence of highly efficient fluorescent acceptors, AuNPs.However, upon the incorporation of the NSE antigen and hybrid solution, the distance between AuNPs and the conjugated antibody increased, leading to fluorescence restoration (ON state).The resultant anti-NSE-AuNPs combination has a very low LOD of 0.09 pg mL À1 and a wide linear range between 0.1 pg mL À1 and 1000 ng mL À1 .In addition, this fluorescent biosensing system was an excellent platform for analyzing the biomarkers within human serum samples, yielding a standard recovery value of 94.69%.Table 3 enlists the specification and detection limits of fluorescence biosensors developed by carbon-based materials for various cancer detection.

Surface Plasmon Resonance Biosensors
SPR-based biosensors are considered an attractive biomolecule detection technique due to their numerous important properties. [283]First, SPR is a label-free sensing method that eliminates the need for labeling to generate the signal. [284]Additionally, it is capable of detecting the trace amount of desired analytes from composites and analyzing them at sub-nanogram levels in real-time. [285]Moreover, these sensors are favored for proteome profiles and biomolecule interactions based on affinity toward each other, encompassing phenomena such as the binding of antigens and antibodies, the kinetics of ligand-receptor interactions, enzyme-substrate reactions, and the cartography of epitopes. [286]Last but not least, SPR can sense biomolecules with incredibly high selectivity and specificity. [287]Based on these advantages, the SPR biosensor is undoubtedly considered a powerful biomedical and clinical analysis tool. [286]PR-based sensing method allows the arrangement of various transducer formations, such as classical Kretschmann and Otto configurations. [288]The classical Kretschmann configuration is the most widely used, which includes SPR imaging, [289] nanoparticle-based LSPR, [290] long-range SPR, [291] fiber-optic sensing, [292] and phase sensing. [293]The typical structure of an SPR sensor contains a thin metals layer (e.g., Au and Ag) coated on a prism, which separates the prism from the sensing medium, in which the immobilization of analyte on the gold surface material, as well as the binding of the analyte and receptor, takes place on the surface. [294]ancer biomarker sensing via SPR biosensors has been increased due to their high selectivity and sensitivity for some particular analytes. [295]Several materials have been assessed for their suitability as the sensing element of the biosensor.
Nanomaterials have emerged as a promising option for developing highly sensitive SPR biosensing devices with excellent refractive index sensitivity.Among the various nanomaterials studied, 2D graphene stands out as one of the most attractive nanomaterials due to the unique aforementioned features, such as high surface area, good strength, mechanical flexibility, high optical transparency, and carrier mobility. [296]Graphene is an absorbing or binding material with a high surface area, making it an essential component in biosensing systems.Hence, incorporating graphene into a biosensor system effectively enhances the biosensor performance. [297]he basic mechanism of the SPR-based biosensor relies on changes in the material properties or refractive index due to antigen-antibody interaction, which generates different signals at different concentrations. [298]SPR is an optical phenomenon  [282] Copyright 2020, American Chemical Society.that occurs when a metal surface is struck by an incident light photon.Notably, the AuNPs surface has exhibited superior sensing accuracy, primarily influenced by three critical factors, an augmentation in the absolute mass during each binding event, an elevation in the bulk refractive index of the analyte, and plasmon fluctuations creating an electric field between the metal surface and the sample, [299,300] facilitating an electromagnetic interaction connecting the localized surface plasmon (LSP) found in metallic nanoparticles with the SPR exhibited by the sensing film. [289,301]A graphene layer was applied to the gold surface in order to increase the detection capacity of SPR-based biosensors.This was accomplished because of the extensive surface area, superior biomolecule adsorption of the graphene, and significant changes in refractive index at the graphene-gold interface. [302]The resultant graphene-based biosensors presented 1 þ 0.4 L (in prism waveguide, L represents the quantity of graphene layers) and 1 þ 0.45 L (in the planar waveguide) folds with higher sensitivity compared to traditional SPR biosensors.
Conventionally, SPR-based immunosensors have been utilized to detect breast cancer biomarkers like CA15-3 for over a decade. [303]The SPR surface could be further altered by incorporating a nanocomposite of gold and zinc oxide (Au/ZnO) or a traditional gold/chromium (Au/Cr) film to enhance chip performance. [304]As for graphene, Stebunov et al. pioneeringly introduced an airbrushing technique to deposit 2D GO on as the linking film to tightly adsorb biomolecules, as displayed in Figure 9a. [305]It turns out that the developed airbrushing method realized a tunable and homogeneous GO layer with large surface areas, and its large number of binding sites for the biomolecules ultimately increases the sensitivity.As demonstrated by SPR curves in Figure 9b, there are obvious shifts for the bare gold chip, graphene chips, and GO-covered chip, and their corresponding dependencies of refractive index changes (SRI) (Figure 9c) unravel 32% and 20% enhanced sensitivity for graphene and GO-based chips compared to the bare gold chip.Similarly, Chiu et al. demonstrated that adding carboxyl groups could effectively improve and control GO SPR-based immunoaffinity biosensors through the plasmonic coupling mechanism, achieving a low LOD of 0.01 pg mL À1 . [306,307]In another case, Li et al. proposed a GO-AuNPs composite-based SPR biosensor, in which the sensor chip was equipped with DNA-GO-AuNPs to identify miRNA and adenosine, achieving a heightened sensitivity with an LOD as low as 0.1 fM. [308]Additionally, cytokeratin 19 was detected in a GO-based SPR sensor for lung cancer diagnosis.Through the functional group on the GO surface, the sensor system fixes large amounts of antibodies, enabling an LOD of 1 fg mL À1 . [240,307]he SPR platform can also be enhanced by incorporating lowdimensional nanomaterials, such as 2D g-C 3 N 4 nanosheets and  [305] Copyright 2015, American Chemical Society.d) Schematic illustration of QCM-based immunosensor.e) QCM curves (frequency vs time) of (i) anti-IL-6-Ab1/AuNPs/S-GQD/QCM, (ii) IL-6/anti-IL-6-Ab1/AuNPs/S-GQD/QCM, and (iii) anti-IL-6-Ab2/ASP-h-ZnS-CdS NC conjugated to IL-6/anti-IL-6-Ab1/AuNPs/S-GQD/QCM.f ) Concentration effect on QCM frequency.The inset shows the corresponding calibration curve.d-f ) Reproduced with permission. [321]Copyright 2021, Elsevier.
0D MoS 2 QDs combined with CS-stabilized AuNPs (CS-AuNPs) placed on the surface. [309]The MoS 2 QDs/g-C 3 N 4 CS-AuNPsbased sensor exhibits a strong binding affinity with the PSA biomarker, and it has the highest selectivity and reliability with the lowest LOD of 0.77 ng mL À1 . [310]5.Quartz Crystal Microbalance Biosensors QCM biosensors, also known as mass-based or piezoelectric biosensors, are label-free mass-sensitive devices.[311] Accurate quantification of biomarker concentrations is possible because the mass of absorbed analytes on the piezoelectric crystal correlates directly with the resonance frequency of the crystals.[312,313] Typically, QCM biosensors contain a quartz crystal coated with electrode layers on both the top and bottom sides, and it is attached to an oscillating circuit that applies an external electrical field to the crystals.[314][315][316] The presence of any mass deposition on the electrode leads to alterations in the crystal's frequency response, which can be observed through the oscillating circuit system.7] It is well acknowledged that QCM-based biosensors are one of the most sensitive devices with antigens detection limitations in the picogram range, enabling them as powerful tools for the analysis of tumor marker detection and differentiation between cancer cells and normal cells.[318] Carbon-based nanomaterials are widely used as supporting agents to increase the performance of AuNP-based QCM biosensors and prevent AuNPs aggregation.These carbon materials are hybridized with AuNPs and coated on the crystal surface.For example, the combination of graphene oxide with AuNPs was developed for CEA detection.[319,320] GO-AuNPs were deposited on the QCM electrode by in situ synthesis method, and anti-CEA was coated on the surface of the electrode to react with CEA.In this system, GO provides quick transduction of signals from biochemical reactions to electrode outputs, while AuNPs help to immobilize anti-CEA by forming stable Au-S covalent interactions with the sulfur atom in mercaptoacetic acid. GO-AuNPs crease the sensitivity of the reaction with a low LOD of 0.09 ng mL À1 , and the value could further decrease to 0.06 ng mL À1 with simply oxygen plasma treatment due to a higher density of AuNPs.
Recently, Atar and Yola demonstrated that nanocomposites consisting of hollow ZnS-CdS nanocages (h-ZnS-CdS NC) and sulfur-doped GQD functionalized with gold nanoparticles (AuNPs/S-GQD) provide excellent QCM biosensors for determining the presence of interleukin-6 (IL-6), as depicted in Figure 9d. [321]Here, S-GQDs not only prevent AuNPs aggregation but also act as a supporting layer for AuNPs.Due to the amino-gold affinity, AuNPs/S-GQD could strongly conjugate with anti-IL-6 antibodies, which subsequently reacted with h-ZnS-CdS NC in the presence of target IL-6.In Figure 9e, while sensorgram b of the QCM curves shows a rise in QCM frequencies as an outcome of the immobilization of antigen IL-6 with anti-IL-6-Ab1 on the surface of QCM gold, sensogram a of the QCM curves shows the starting point for anti-IL-6-Ab1/ AuNPs/S-GQD/QCM, while Sensogram b exhibits an increase in QCM frequencies.This sensitivity is ascribed to the essential and distinct interactions involving IL-6/anti-IL-6-Ab1/AuNPs/ S-GQD and anti-IL-6-Ab2/ASP-h-ZnS-CdS NC, resulting in heightened frequency fluctuations.The concentration effect of IL-6 on QCM frequencies was also investigated, as shown in Figure 9f.The developed QCM immunosensor exhibited an impressively low LOD of 3.33 fg mL À1 , along with a linear range spanning from 0.01 to 2.0 pg mL À1 .Moreover, it demonstrated excellent storage stability for over eight weeks.In another report, Jandas et al. demonstrated GO/AuNP-coated QCM sensors utilizing the CEA antibody for precise and real-time quantification of CEA.The designed biosensor recognizes and captures CEA biomolecules with LOD of 0.06 and 0.09 nm mL À1 of CEA.Notably, the invented device's sensitivity may be significantly enhanced after treatment with oxygen plasma as it increased the surface active sites responsible for interacting with the antibody of CEA. [319]Despite the numerous merits of carbon-based QCM biosensors, they have been rarely used for biomarker sensors until now, presumably because of the complication within fabricating a consistent carbon nanomaterial thin layer with excellent friction to transducers, as well as the high cost of electrodes.Table 4 compiles the carbon-material-based SPR and QCM biosensors that have been developed for the purpose of detecting distinct cancer biomarkers with low LOD, including ssDNA, miRNA, CA15-3, CEA, interleukin-6, etc.However, more efforts are still required before this type of biosensor becomes a practical analytical tool.

Conclusions and Outlook
Early diagnosis of cancer is the ultimate goal of biosensor technology.A simple and sensitive diagnostic method that needs to detect multiple cancer biomarkers at low concentrations in body Au NPs, gold nanoparticle; GO, graphene oxide; S-GQDs, sulfur graphene quantum dots; CDs, carbon dots; NC, nitrogen carbon; ZnS, zinc sulfide; SPR, surface plasmon resonance; QCM, quartz crystal microbalance.
fluids is pursued by numerous scientists, while carbonnanomaterial-based biosensors have mostly fulfilled these requirements, with additional advantages including portable, biocompatible, enabling efficient detection, diagnosis, management, and treatment for cancer diseases.Until now, various biosensor technologies based on carbon nanomaterials have been devised to realize early cancer diagnosis, and significant advancements in this research field have been achieved.The efficient analysis of cellular alterations and rapid cancer detection at the onset can ultimately change the strategies of following prognosis and treatment, greatly improving the patient's survival rate.In this review, we first comprehensively reviewed common tumor biomarkers, followed by introducing their corresponding cancers.Later, the development of multiple detection techniques for identifying cancer biomarkers is elaborated in detail, demonstrating excellent detection limits in a broad range of fg mL À1 to ng mL À1 , fast signal transduction, noninvasive detection, realtime monitoring, biocompatibility, and economic friendly, which position carbon-based nanomaterials as the next-generation materials for crafting advanced biosensors capable of highly sensitive cancer biomarker detection. [74]espite the considerable efforts and advancements made in utilizing carbon nanomaterials in biosensor applications, their usage is still in the early stages.To enhance performance and enable future commercialization, further efforts are needed to address the following challenges: 1) Developing a reproducible and economical synthesis approach for oxygen-free and largearea carbon nanomaterials with phase purity is crucial, which is also a prerequisite for future large-scale commercialization.A lot more effort has to be made to explore the single structural phase of the carbon material family rather than mixed-phase formation in case of reversibly reducing the sensitivity and selectivity of biosensors.2) Currently, various carbon-nanomaterialbased sensing systems have been tested successfully only under laboratory conditions for analyzing the quantification of cancer biomolecules/cells and for clinical applications like imaging and therapeutics.However, cancer biomarker detection in real biological specimens, such as blood plasma, blood serum fluid, saliva, or urine, is much more challenging because these samples are way more complex, with various proteins, ions, and other chemicals able to increase false-positive responses in assays.Consequently, experiments under realistic conditions with practical device fabrication approaches and operation procedures must be developed for the future reliable applications of these carbon-based biosensors.3) Carbon-based nanomaterials need to be further developed in the biosensing area to realize multiple biomarker analyses at the same time.The diagnosis of specific cancers frequently requires multiplexed analysis.The strategies for concurrently detecting multiple cancer biomarkers or dual-mode response detection have been demonstrated by other nanomaterials, [322,323] but their underlying mechanisms and realization in carbon-based biosensors are still not well established so far.4) Even though outstanding achievements have been made, there is still a far distance from commercializing carbon-based biosensors.For practical clinical use, more investigation should be focused on the long-term stability and antidisturbance properties of carbon-based biosensors.
To overcome these challenges, purification and templating methods could be one solution to increase the sensitivity of the sample in the biosensing system, in which the materials' purities, their specific surface area, and conductivity can be significantly enhanced.Another straightforward approach is to explore new compositions and morphologies of carbon nanomaterials.Carbon-based nanomaterials could be flexibly hybridized with other functional materials to overcome their intrinsic drawbacks by modifying the structure/morphology and enriching the functionality by virtue of their synergistic effects, thereby enhancing sensing performance.Apart from the carbon nanomaterials introduced in this review, more investigation should focus on designing biosensors based on fullerene materials, [324][325][326][327][328] which are observed as promising candidates to identify cancer biomarkers according to recent progress. [329,330]It is believed that fullerene-based sensors are able to use molecularly imprinted polymers and dendrimers as recognition agents rather than biomolecules, which are the starting line in sensing technology and are essential for developing new analyzers. [331]Other emerging nanomaterials, like 2D borophene, [332] silanol, [44] metal nitrides, [333] are extremely promising to hybridize with carbon materials for cancer detection.Moving forward, we believe that different families of nanomaterials can make these sensing devices highly potential and allow for rapid, simple, sensitive, and cost-effective cancer biosensing.
Ajayan Vinu is a Professor at the University of Newcastle and Director of the Global Innovative Center for Advanced Nanomaterials.He introduced a new field of research on nanoporous nitrides and developed novel methods for making nanoporous materials, culminating in multiple reports of world's first mesoporous carbon nitrides, boron nitrides, boron carbon nitrides, biomolecules, and fullerenes for various applications.

Figure 2 .
Our review should be the first work to cover five different mechanisms of biosensors with the latest advancements in utilizing carbon-based nanomaterials for cancer diagnosis applications and critically analyze the

Figure 1 .
Figure 1.Different types of carbon nanomaterial biosensors based on their transducer operation.

Figure 2 .
Figure 2. The number of journal publications on three carbon nanostructures in the last years.The data of publication numbers are taken from the Web of Science.

Figure 4 .
Figure 4. Schematics of the working process of carbon-nanomaterial-based biosensors for early cancer diagnosis.

Figure 5 .
Figure 5. a) Graphical representation of rGO-based immunosensor for HER-2 protein.b) Effect of rGO on the intensity of ECL signals.c) The three consecutive ECL graphs of Bare, Bare þ rGO-CS-Ru, Bare þ rGO-CS-RuþAb, and Bare þ rGO-CS-Ru þ AbþHER-2.d) The ECL curves for GCE/rGO/CS-[Ru(bpy) 3 ] 2þ /Ab-HER-2 for different concentrations of HER-2 protein (0.000001 to 1 nM).e) Calibration curve of ECL intensity of the final modified electrode at the corresponding concentrations of HER-2 protein.f ) ECL plots of investigation of the HER-2 in untreated serum samples from breast cancer patients.a-f ) Reproduced under the terms of the CC-BY Creative Commons Attribution 4.0 International license (https://creativecommons.org/ licenses/by/4.0).[214]Copyright 2021, The Authors, published by Springer Nature.

Figure 7 .
Figure 7. a) The electrochemical biosensing platform designed for protein detection is shown schematically.b) A comparison of the rGO with (left) and without (right) chitosan using TEM images.c) DPV curves for detecting a range of VEGFR2 concentrations, from: a) 0 to b) 0.3, c) 0.4, d) 0.9, e) 4.3, f ) 8.6, g) 43, h) 65, i) 86, to j) 103 pM.d) The associated curve of calibration for (c).e) The sorafenib and vandetanib chemical formulas.Below is a representation of the respective binding models for VEGFR2-sorafenib and VEGFR2-vandetanib.a-e) Reproduced with permission.[242]Copyright 2014, Springer Nature.

Figure 8 .
Figure 8. Schematic diagram depicting the operational mechanism of the fluorescent biosensor for detecting small cell lung cancer biomarkers by neuron-specific enolase antibodies (anti-NSE)/amine nitrogen-doped graphene quantum dots with AuNP.Reproduced with permission.[282]Copyright 2020, American Chemical Society.

Table 1 .
Summary of ECL biosensors used to detect cancer biomarkers.

Table 2 .
Electrochemical biosensors produced for cancer biomarkers detection. a)

Table 4 .
Summary of SPR and QCM biosensors for cancer biomarker detection. a)