CRISPR-based platforms for detecting tumor-associated genetic materials in clinical samples

ABSTRACT

Tumor-associated genetic markers are useful for early cancer screening, diagnosis, and treatment monitoring. However, traditional detection methods are complex in operation procedures, time-consuming, and the equipment costs are expensive. CRISPR/Cas systems are becoming emerging detection tools for tumor detection due to their programmability, rapid reaction, high targeting specificity, and the ability to amplify the signals. CRISPR/Cas has made breakthroughs in the detection of tumor-associated genetic materials including gene mutations, DNA methylation, miRNA, lncRNA, and circRNA detection. Herein, we critically discuss these advancements and describe the key concepts of each CRISPR/Cas system for detecting tumor-associated genetic materials. The significance of these tumor-associated genetic materials in cancer diagnosis and prognosis is highlighted.

1. Introduction

The incidence of malignant tumors is on the rise worldwide [1]. Delay in cancer detection is a major factor contributing to high mortality rates. Early detection of cancer through tumor biomarkers can significantly improve patient outcomes and reduce treatment burden. Tumor biomarkers are biological substances secreted by tumors or the body’s response to tumors during tumor development and progression. They can be found in blood, bodily fluids, or tissues, and their characteristics can be altered by cancerous cells. Therefore, it can indicate the presence, stage, or progression of cancer. It is also useful for tumor prognosis prediction, recurrence detection, and monitoring of treatment effects [2].

Traditionally, tumor diagnosis methods mainly rely on imaging examinations [3] and the detection of tumor markers using PCR or ELISA [4]. These methods often require expensive and bulky detection equipment, as well as complex operations. Their detection sensitivity and specificity are limited, especially when detecting early-stage cancer. Liquid biopsies, which analyze tumor biomarkers in bodily fluids like blood, urine, saliva, and pleural effusion, are attractive for early cancer detection [5]. Compared with traditional methods, liquid biopsy has the advantages of being noninvasive, highly reproducible, and having high detection sensitivity. The main components of tumor-associated genetic materials in liquid biopsy include circulating tumor DNA (ctDNA) and other tumor-derived materials such as extracellular vesicles and RNA molecules. The ctDNA can be used for gene mutation detection and methylation analysis, while tumor-associated RNAs, including miRNA, lncRNA, and circRNA are used for quantification of their expression levels. These tumor-associated genetic materials have been used for early cancer screening, diagnosis, prognosis, and treatment monitoring.

The clustered regularly interspaced short palindromic repeats (CRISPR)-associated system can be a promising diagnostic tool for detecting these tumor-associated genetic materials in liquid biopsy. CRISPR offers the advantages of programmability, rapid detection, high targeting specificity, and the ability to amplify the signals [6–8]. CRISPR/Cas systems, including CRISPR/Cas9, CRISPR/Cas12, CRISPR/Cas13, and CRISPR/Cas14, have been used to detect tumor-associated genetic materials. CRISPR/Cas9 was initially developed for targeted gene editing [9–11]. It is then harnessed for nucleic acid detection [12,13]. It can specifically recognize PAM and sequence-complementary targets and cleave targets under the guidance of sgRNA. Later, the field of molecular diagnostics achieved significant milestones, driven by the emergence of CRISPR/Cas12 [14–16] and CRISPR/Cas13 [17–19] technologies. Cas12 recognizes the target dsDNA and ssDNA, whereas Cas13 possess the ability to target the ssRNA. They execute cis- and trans-cleavage activities when are activated by target nucleic acids. The collateral cleavage of nontarget fluorescent probe reporters by active Cas12 or Cas13 can amplify the signals for detecting low abundance nucleic acids. Due to their signal amplification capability, Cas12 and Cas13 have been harnessed for pathogen detection in CRISPR/Cas12-based DETECTR [15] and CRISPR/Cas13-based SHERLOCK [20] detection platforms. Thereafter, CRISPR has been integrated with various isothermal amplification technologies [21] or amplification-free signal transduction technologies [22] to develop next-generation CRISPR-based diagnostics (CRISPR-Dx). CRISPR-Dx has gained popularity for cancer detection [23,24], pathogen detection [25,26], and food safety monitoring [27,28].

Therefore, the advancements and the key strategies of each CRISPR-Dx detection system for the detection of diverse tumor-associated genetic materials are comprehensively reviewed. The technical shortcomings and hurdles associated with these CRISPR-Dx systems are critically discussed. We also provide countermeasures and future perspectives of CRISPR-Dx for clinical applications and point-of-care diagnostics of tumor-associated genetic materials. Different types of tumor-associated genetic materials relevant to cancer diagnostics are also described.

2. CRISPR-Dx for detection of tumor-associated DNAs

2.1. Detection of germline mutation and somatic mutation in genomic DNA

The occurrence and development of tumors is often a multi-step process where cells acquire multiple mutations in different genes, such as oncogenes and tumor suppressor genes [29]. These mutations accumulate over time, with each mutation providing a slight growth advantage to the cell. It drives the transformation of normal cells into malignant phenotypes by regulating key pathways such as cell proliferation, apoptosis, metabolism, and microenvironment remodeling [30]. Most of these mutations are related to oncogenes and tumor suppressor genes. A mutation in an oncogene can promote cell growth and uncontrolled cell proliferation, whereas a mutation in a tumor suppressor gene can disrupt normal cell function and induce uncontrolled cell division. Different types of cancer can be associated with specific gene mutations. Therefore, gene mutation detection enables the diagnosis of specific types of cancer. The traditional methods to detect gene mutation include polymerase-chain reaction (PCR) and DNA sequencing [31]. However, these detection approaches have the drawbacks of low detection accuracy, low sensitivity, long detection time, and high cost. CRISPR-Cas system can circumvent the limitations associated with the existing methods for detecting gene mutation. CRISPR-Cas can target and differentiate between DNA sequences with even subtle variations like single-nucleotide polymorphisms (SNPs). An SNP is an inherited genetic variation that can increase a person’s cancer susceptibility and affect predisposition, while somatic mutations are acquired and can sometimes drive cancer development. CRISPR-Cas offers the convenience of requiring only a constant, low temperature for nucleic acid detection. It also offers the potential for rapid, on-site diagnostics, with results available in minutes rather than hours. However, preamplification of the template is often required for detecting trace nucleic acids with sufficient sensitivity. This additional preamplification step can increase the risk of carryover contamination, prolong the detection time, and lengthy multi-step operation procedures.

The abundance of mutant DNA in clinical samples is extremely low (< 0.1%), and amplification is often interfered with by an excess amount of wild-type DNA. The excess wild-type DNA will compete with mutant DNA for amplification, leading to low detection sensitivity and specificity. Various allele-specific PCR or mutant enrichment PCR have been developed to overcome this issue. For example, Wang et al. [32] proposed a single-base mutation detection method that combines allele-specific PCR with the CRISPR/Cas12a system (Figure 1(A)). They designed primers with hairpin structures to effectively inhibit the amplification of wild-type sequences during the amplification process. At the same time, the PAM sequence was introduced through PCR, allowing CRISPR/Cas12a to recognize and cut the target dsDNA, thereby releasing fluorescent signals through its trans-cleavage activity. It can overcome the limitations of traditional Cas12a’s dependence on the PAM sequence and broaden the scope of mutation analysis. Their approach can stably detect a mutation ratio of 0.1%. It has been verified using a clinical sample with a detection accuracy rate of 100%. In another approach, MscI restriction endonuclease was used to digest and degrade the wild-type DNA, while leaving the mutant DNA undigested [33] (Figure 1(B)). Because the wild-type DNA has an enzyme cleavage site while the mutant type does not, the endonuclease will specifically cut the wild-type DNA. Then, recombinase-aided amplification (RAA) was used to isothermally amplify the low-abundance mutant gene. When integrated with a microfluidic chip, rapid and ultra-sensitive detection of the desired mutation was achieved.

Figure 1.

Strategies for gene mutation detection. (A) a single-base mutation detection method combining allele-specific PCR and CRISPR/Cas12a system. MT, mutant; WT, wild-type. (B) method for detecting mutations using CRISPR-Cas12a combined with the restriction enzyme MsCI. MRE, msci restriction enzyme; RAA, recombinase-aided amplification. (C) a mutation site detection with engineered Cas12a. ERA, editing, repression, and activation.

In addition to single-base mutations, CRISPR-Dx has also been employed to detect base insertions and deletions. The frequency of insertions and deletions in cancer genomes is only about 10% of that of base substitutions, but they are more likely to cause frameshifts and produce abnormal proteins that can promote cancer development [34]. For example, Liu’s research team has developed a one-stop ultra-sensitive ERA-CRISPR system called CoHIT (Cas12a-based One-for-all High-speed Isothermal Test) to detect insertion/deletion mutations (indels) [35] (Figure 1(C)). They engineered AsCas12a to increase mismatch tolerance and recognize an NNCV/TTTV PAM sequence. As a consequence, only a single crRNA was needed to detect different indels at a common site. The system has a detection limit of 0.01% for detecting 4-bp insertions in acute myeloid leukemia (AML). The detection was completed within 30 minutes without wild-type cross-reaction. Using this method, they have successfully detected 30 mutations in 108 AML patient samples, with 100% accordance with NGS (Next-Generation Sequencing) results. This CRISPR-Dx approach has also been integrated with lateral flow test strips and microfluidic chips to realize point-of-care diagnostics of cancer. This integrated system allows rapid detection, simplifies the analysis, improves the detection accuracy, a miniature device, and reduces the diagnosis cost.

2.2. Detection of cancer somatic mutation in ctDNA

Circulating tumor DNA (ctDNA) is DNA fragments released from tumor cells into the bloodstream. It is one of the most important tumor biomarkers in liquid biopsy. It carries DNA fragments with specific gene mutations and methylation signatures, allowing us to gain insights into the tumor’s genetic makeup and tailor treatment strategies [36]. Therefore, ctDNA is useful for cancer detection, monitoring treatment response, and identifying drug resistance mechanisms. Analysis of ctDNA liquid biopsy enables noninvasive detection of cancer, early detection of cancer for timely interventions, and treatment effectiveness monitoring by tracking ctDNA levels over time. It can also capture the diversity of genetic mutations within a tumor, which can be underrepresented by single-site tissue biopsies. While promising, analysis of ctDNA liquid biopsies faces several obstacles such as low ctDNA concentrations and the need for standardized workflows and analytical methods. Recently, CRISPR-Dx is being increasingly used in the analysis of ctDNA liquid biopsies [37].

In 2020, the first CRISPR-Cas platform for ctDNA detection was reported [38]. This platform covalently fixed inactivated Cas9 (dCas9) and synthetic guide RNA (sgRNA) on a graphene oxide screen-printed electrode (GPHOXE). It detected PIK3CA exon 9 mutations by electrochemical impedance spectroscopy (EIS). It detected 2–20 nM of 120-bp ctDNA within 40 seconds, with a detection limit of 0.65 nM and a quantification limit of 1.92 nM. The recovery rate in real blood samples exceeded 96%, providing a rapid and highly selective method for liquid biopsy of ctDNA.

However, ctDNA usually accounts for only a small part of the total circulating free DNA (cfDNA) in the blood (0.01–10%), and the mutated DNA derived from tumors is often masked by wild-type DNA fragments, challenging the analysis of ctDNA liquid biopsies [39]. To address this issue, an ultra-sensitive ctDNA mutation detection technology called HiCASE has been developed [40]. It combines PCR amplification, restriction enzyme digestion, and CRISPR-Cas13a system for target gene mutation detection in the plasma of patients with cancer (Figure 2(A)). HiCASE has a detection sensitivity of 0.01% (40 ng cfDNA input) for mutations such as EGFR L858R, 19dels, and T790M. It has demonstrated a clinical sensitivity of 88.1% and a specificity of 100% in 140 clinical samples, which is superior to Super-ARMS and digital PCR detection platforms. In addition, this technology can accurately distinguish the cis/trans positions of T790M/C797S mutations, providing crucial information for clinical targeted therapy.

Figure 2.

Strategies for ctDNA detection. (A) HiCASE detects mutated genes in ctDNA. MT, mutant; WT, wild-type; RE, restriction enzyme. (B) MME-CRISPR schematic diagram. MME-CRISPR, manganese metal-organic frameworks (MOFs)-enhanced CRISPR.

For improving the detection sensitivity of the CRISPR-Cas system, the majority of studies focused on Cas protein modification, crRNA structure optimization, and signal output mode innovation, but ignored the stability of Cas enzymes in harsh environments such as high temperature and in the presence of interference factors (e.g., inhibitors and organic solvents). To address this issue, Cas12a-crRNA was encapsulated in manganese metal organic frameworks (Mn-MOFs) to form an enzyme-MOF complex for electrochemical analysis [41] (Figure 2(B)). The MOF can effectively protect the Cas12a enzyme from the pressure of the external environment. The framework can quickly release Mn2+ to enhance the activity of Cas12a and achieve ultrasensitive detection of gene mutations in ctDNA. This method has a LOD of 0.28 fM without resorting to the preamplification step. The recovery rate in clinical samples exceeds 96%, providing a new and efficient liquid biopsy method for cancer diagnosis.

2.3. DNA methylation analysis

DNA methylation is the transfer of methyl groups to cytosine by methyltransferase (MTase) to form 5-methylcytosine (5-mC). DNA methylation functions to regulate gene expression in the human genome. Abnormal DNA methylation can lead to tumorigenesis [42]. The conventional method for DNA methylation analysis involves treating DNA with sodium bisulfite, which converts unmethylated cytosines to uracils while leaving methylated cytosines unchanged. The bisulfite-treated DNA is then subjected to PCR and DNA sequencing. However, this method is cumbersome, has lengthy operation procedures, and has a long detection time. The harsh experimental conditions during bisulfite treatment can cause DNA degradation, leading to low reproducibility and inaccurate methylation analysis. Methyl-specific antibodies or methyl-binding domain proteins have also been used to enrich methylated DNA [43]. Since methylation often occurs in the cytosine phosphate guanine (CpG) islands of the promoter regions, this method tends to enrich DNA fragments with high CpG density. Methylation-sensitive restriction endonuclease (MSRE) has also been used to distinguish methylated and unmethylated DNA at the target site [44]. MSRE cleaves DNA at specific sequences, but its activity is inhibited by methylation of cytosine bases within that sequence. Therefore, it selectively cleaves unmethylated DNA, leaving methylated DNA as a template for qPCR. It offers the advantages such as bisulfite-free methylation analysis, high detection sensitivity when combined with PCR, short detection time because bisulfite conversion is not needed, and relatively cheaper compared to other DNA methylation analysis techniques.

CRISPR-Dx is increasingly being used for DNA methylation analysis. For example, CRISPR-Cas12b has been used to detect methylated DNA [45]. In this case, the DNA samples were first treated with bisulfite followed by PCR amplification. Subsequently, the PCR products activated the trans-cleavage activity of the CRISPR-Cas12b to generate the fluorescent signals. However, it has disadvantages such as high reaction temperature, the need for real-time PCR, and cumbersome processing. Later, dual methylation-sensitive restriction endonucleases were coupled with a recombinase polymerase amplification (RPA)-assisted CRISPR/Cas13a system (DESCS) for specific detection of methylation sites [46]. The dual methylation-sensitive restriction endonuclease system can specifically digest unmethylated DNA fragments, while methylated DNA is unaffected. The intact methylated DNA fragments can be isothermally amplified by RPA and transcribed into a large amount of ssRNA under the action of T7 RNA polymerase. The ssRNA then activated the trans-cleavage activity of Cas13/crRNA to cut the surrounding reporters and generate strong fluorescent signals (Figure 3(A)). This DESCS system can detect methylated DNA at a LOD of 86.4 aM and effectively distinguish between unmethylated and methylated DNA. When integrated with a lateral flow sensor, it allows visual detection of methylated DNA and achieves a LOD of 200 aM. Thereafter, the MSRE-CRISPR platform has been integrated with different isothermal amplification techniques for DNA methylation analysis [47–49].

Figure 3.

Strategies for DNA methylation analysis. (A) A platform for detecting methylation sites using CRISPR-Cas and MSRE. RT-RPA, reverse transcription-recombinase polymerase amplification. (B) Specific enrichment of methylated DNA using chemically modified probes and detection using RCA-Cas12a one-pot approach. MB, magnetic bead.

The CRISPR-Cas platform has also been integrated with various enrichment methods to enrich methylated DNA prior to analysis. For example, a chemical modification coupled to an isothermal CRISPR detection system (CAICas12a assay) has been proposed to achieve ultra-sensitive quantitative detection of 5hmC [49]. It uses the multifunctional photosensitive probe PC-DBCO-PEG3-biotin to specifically label and enrich 5-hydroxymethylcytosine (5hmC) DNA (Figure 3(B)). Rolling circle amplification (RCA) was then used to amplify this enriched methylated DNA. The RCA products then activated the trans-cleavage activity of CRISPR-Cas12a for fluorescent signal generation. This method achieves a detection limit of 11 fM and a linear range of 6.38 pM to 50 fM. Only 10 ng of input DNA is required to detect global 5hmC in biological samples.

3. CRISPR-Dx for detection of tumor-associated RNAs

3.1. miRNA detection

MicroRNA (miRNA) is an endogenous non-coding RNA molecule with a length of about 20–25 nucleotides. They do not participate in protein coding, but regulate gene expression at the post-transcriptional level by binding to the 3’ untranslated region (3‘UTR) of the target mRNA [50]. Dysregulation of miRNA expression is closely related to the occurrence and development of many diseases such as cancers [16,51]. In cancer, miRNA plays a unique dual role. Some miRNAs act as “tumor suppressors” which can inhibit tumor progression by targeting the oncogenes, while others act as “oncogenes” which can promote tumorigenesis by inhibiting the tumor suppressor genes [52]. These diversity and specificity features make miRNA a highly promising biomarker for cancer diagnosis and treatment. Circulating miRNAs are stably present in body fluids such as blood and urine and present in great excess in extracellular vesicles. Their expression patterns can be indicative of cancer. MicroRNAs offer several advantages as biomarkers for cancer diagnosis, including higher detection sensitivity than other tumor markers, can classify cancer subtypes due to cancer-specific miRNA expression profiles, early detection of cancer before clinical symptoms appear, and the ability to monitor treatment response by tracking the dynamic changes in miRNA expression.

However, miRNAs are short and have highly homologous sequences among families, which makes their detection very challenging [53]. Furthermore, the difficulty of maintaining sample integrity during extraction and analysis complicates the detection procedures of miRNA. The existing approaches to detecting miRNA have some drawbacks. For example, microarrays are susceptible to cross-hybridization, therefore, they cannot detect all miRNA isoforms. Next-generation sequencing can be biased due to PCR amplification during library preparation, leading to inaccurate miRNA quantification. Quantitative PCR (qPCR) can be affected by sequence variations and isomiR heterogeneity. CRISPR-based methods offer several advantages for miRNA quantification, including high sensitivity, specificity, and the ability to discriminate miRNA isoforms and target multiple miRNAs simultaneously. It can distinguish between miRNAs with even single-base differences. Although CRISPR can detect miRNAs at the single-molecule level, preamplification of target miRNA is often required to detect low-abundance miRNA at a high signal-to-noise ratio. Examples of amplification approaches that have been integrated into the CRISPR-Dx system are including rolling circle amplification (RCA) [16,54,55], isothermal recombinase polymerase (IRP) [56–58], hybridization chain reaction (HCR) [59–61], strand displacement amplification (SDA) [62,63], loop-mediated isothermal amplification (LAMP) [58] and catalytic hairpin assembly (CHA) [64–66].

However, these additional amplification steps require the introduction of multiple enzymes and primers, complicating the detection procedures of CRISPR-Dx and prone to carryover contamination. In contrast, amplification-free CRISPR-Dx can streamline the workflow, shorten the detection time, and reduce the detection cost, making it suitable for point-of-care diagnostics in various settings. To realize amplification-free detection and quantification of miRNA, various signal enhancement strategies have been developed using a positive feedback loop within the CRISPR-Cas system to enhance the signal generated by target DNA. These signal enhancement strategies include cascade reactions, auto-catalysis, and dual signal amplifications [6]. Some amplification-free CRISPR-Dx systems also incorporate logic functions to achieve multiplexed detection [67,68]. For example, Cas13a and Cas14a were used together to achieve cascade amplification of the signal for miRNA detection [69] (Figure 4(A)). Cas13a was first used to recognize the target miRNA. The activated trans-cleavage activity was then used to cleave the hairpin structure ST-HP containing unpaired rU, converting it from a stable hairpin to a double-stranded structure with a 5’ toehold. This ST-HP can then react with Cas14a/sgRNA through a strand displacement reaction, hydrolyzing the fluorescent reporter probe to generate fluorescent signals. This method enables one-pot detection of miRNA, avoiding unnecessary contamination caused by multi-step procedures. The detection limit of miR-17 is 1.33fM, and it can distinguish homologous family members with single-base differences. It can also successfully detect breast cancer patients by detecting miRNA in serum.

Figure 4.

Strategies for miRNA and circRNA detection. (A) One-pot detection of miRNA through cascade amplification reaction. ST-HP, locked-trigger for Cas14a/sgRNA. (B) Detection of miRNA by Cas12-mediated trans-cleavage activity. AuNP, gold nanoparticle; SA-biotin, streptavidin-biotin. (C) Electrochemical probe platform based on pyramid structure of DNA nanocages (tetrahedra), combined with CRISPR/Cas13a for detecting circRNA in urine. (D) PddCas13a platform for detecting ciricRNA.

Recently, amplification-free CRISPR-Dx beyond cascade systems have also been developed to detect target nucleic acids by integrating CRISPR/Cas system with electrochemical microfluidic [70–74], droplet digital [75–79], or nanopore [80] technologies. For example, CRISPR-empowered electrochemical biosensor enables amplification-free and ultrasensitive detection of miRNA through dual-signal amplifications [72]. When the target miRNA is present, the trans-cleavage activity of CRISPR/Cas13a is activated. The active Cas13a then triggers the primer exchange reaction (PER) on the electrode surface, leading to second round of signal amplification and transducing electrochemical signal. CRISPR/Cas13a has also been integrated with electrochemical microfluidic biosensor for amplification-free and multiplex detection of miRNA by utilizing multiple chip designs and amperometric readout [71]. In another examples, CRISPR-electrochemical biosensor was integrated with DNA framework to realize amplification-free and all-on-chip detection of miRNA [73,74]. Meanwhile, the use of droplet digital platform in CRISPR-Dx enables amplifications-free or even single molecule detection of miRNA by partitioning the sample into thousands of tiny, independent reaction chambers (droplets). This microdroplet-enabled confinement effect can enrich and ultralocalize the target molecules and improve the signal-to-noise ratio [77–79].

As the length of miRNA is short, direct detection of miRNA with Cas13a/crRNA is almost impossible due to the restriction of the PAM sequence requirement. CRISPR-Cas13a uses ssRNA probe for detecting miRNA, but the ssRNA probe is susceptible to degradation by RNase nuclease and chemical breakdown. Therefore, CRISPR-Cas12a is more popular than CRISPR-Cas13a for the detection of miRNA. For example, a miRNA detection platform has been developed using CRISPR-Cas12a to overcome the shortcomings associated with the preamplification step and PAM sequence requirement [81] (Figure 4(B)). This CRISPR-Dx system has also been integrated with reverse fluorescence enhanced lateral flow test paper to conveniently and sensitively achieve a detection limit of 5.53fM.

3.2. circRNA detection

Circular RNA (circRNA) is characterized by a closed-loop structure RNA without 5‘and 3’ ends [82,83]. This unique structure makes circRNAs highly stable and resistant to exonuclease degradation. CircRNA is generated through reverse splicing, making it essentially different from linear RNA in configuration. While a few circRNAs are now known to have protein-coding potential, they are primarily classified as non-coding RNAs. CircRNAs play diverse roles in cellular processes, including acting as miRNA sponges and protein scaffolds, regulating gene transcription and splicing, involved in protein–protein interactions, and even encoding proteins. Dysregulation of circRNA expression is linked to various cancers, with some circRNAs acting as oncogenes and others as tumor suppressors [84–86]. Their stability and tissue-specific expression make them promising biomarkers for cancer diagnosis and therapy [87]. The traditional approaches to detecting circRNA are RNA sequencing and RT-PCR. However, they have some drawbacks, such as low detection sensitivity and specificity. RNA sequencing has low detection efficiency, and requires a substantial amount of RNA input and appropriate sequencing depth for accurate quantification of circRNA. RT-PCR can generate false-positive results due to the interference of linear RNA with secondary structure at the 3′ end, and reverse transcription of circRNA can generate repeated sequences, thereby overestimating the abundance of circRNA. CRISPR technology offers powerful tools for detecting and quantifying circRNAs due to its high targeting specificity and fast turnaround time.

For example, CRISPR/Cas13a was integrated with a pyramid-structured DNA nanocage (tetrahedron)-based electrochemical probe platform to achieve direct detection of circRNA in urine [88] (Figure 4(C)). A detection platform based on polydisperse droplet digital CRISPR/Cas13a (PddCas13a) has also been developed to realize portable quantitative analysis of circRNA and miRNA [89] (Figure 4(D)). In this case, the RNA sample was quickly dispersed into droplets, followed by Cas13a-mediated trans-cleavage of reporter molecules to generate the fluorescent signals. By modifying crRNA with phosphorothioate to enhance its stability, PddCas13a exhibited excellent quantitative accuracy in serum/exosome samples compared with traditional digital PCR. In addition, it was integrated with a set of automated droplet generation portable equipment to realize rapid on-site detection and broad clinical applications.

3.3. lncRNA detection

Long noncoding RNA (lncRNA) has a length of more than 200 nucleotides, but it does not code for proteins [90]. Although it does not translate proteins, lncRNA plays diverse roles in cellular processes, including gene expression regulation, chromatin remodeling, and epigenetic modifications. Studies have shown that these noncoding molecules are of irreplaceable importance in life activities [91]. In addition, abnormal expression of lncRNA is closely related to a variety of diseases, including malignant tumors, cardiovascular diseases, and nervous system disorders [92]. LncRNA is emerging as a promising biomarker for cancer diagnosis, prognosis, and potential therapeutic targets. Dysregulation of lncRNAs in various cancers, coupled with their stability in body fluids, makes lncRNAs attractive for noninvasive cancer detection. While promising, detecting lncRNAs presents several challenges due to their low abundance compared to protein-coding mRNAs, variable expression across different cell types and tissues, diverse functions with multiple isoforms, and complex interactions with other cellular components such as DNA and protein. Traditional approaches to detect and quantify lncRNA include microarrays and next-generation RNA sequencing (RNA-seq). RNA-seq faces challenges in accurately quantifying low abundance lncRNAs due to inherent noise and the need for high sequencing depth. Microarrays have poor sensitivity and accuracy for detecting lncRNA isoforms and low abundance lncRNA transcripts, and rely on known sequences for probe design. CRISPR-Dx is increasingly being utilized in lncRNA detection.

For example, CRISPR-Cas12a has been combined with RT-RAA for detecting lncRNA RMRP in urine exosomes of bladder cancer patients [93]. A novel mini-crRNA-mediated CRISPR/Cas12a system (MCM-CRISPR/Cas12a) has also been proposed to detect lncRNA at high sensitivity [94]. This system used truncated crRNA, chain displacement amplification (CHA), and hybridization chain reaction (HCR) to detect trace amounts of miRNA and lncRNA (Figure 5(A)). It has a good detection specificity and can be directly used to detect lncRNA in serum and cell samples. In another example, isothermal amplification by circular extension (IACE) was integrated with a three-way junction structure (3WJ) and a nicking enzyme to construct a one-step fluorescent biosensor for detecting lncRNA [95] (Figure 5(B)). This sensor achieved a detection limit of 37.5 fM in clinical samples and completed the detection within 2 hours. It has high detection specificity and high agreement with RT-qPCR results. This one-step approach can also reduce the risk of contamination.

Figure 5.

Strategies for lncRNA detection. (A) Mini crRNA-mediated CRISPR/Cas12a system (MCM-CRISPR/Cas12a). CHIA, chain displacement amplification; HCR, hybridization chain reaction. (B) Cas12a detection of LncRNA by IACE.

4. Future perspective

Despite encouraging results from laboratory-based research for the use of CRISPR-Dx in detecting tumor-associated genetic materials (Table 1), multiple challenges need to be addressed before it can be translated into clinical settings and commercialization. First, CRISPR reagents, including Cas9 protein and guide RNAs, have varying stability depending on the storage conditions. They are easily inactivated in harsh clinical environments with high temperatures, organic solvents or inhibitors, which restricts their applicability in complex biological sample processing. Second, the detection bottleneck of low-abundance targets, such as ctDNA, which only accounts for 0.01%–10% of circulating free DNA (cfDNA), thereby being easily masked by wild-type sequences. Amplification or enrichment approaches are often needed to detect this rare event. Cascade CRISPR systems, such as Cas13a-Cas14a, can amplify signals through two-step cutting to improve detection sensitivity without target amplification. Third, the PAM requirement in CRISPR-Dx systems restricts the targetable sites within a nucleic acid template. This issue can be addressed by engineering Cas variants to broaden the PAM recognition range and improve mismatch tolerance. Fourth, single-base targeting specificity is difficult to achieve by CRISPR-Dx in some cases. For example, tedious optimization and crRNA screening are required to distinguish highly homology miRNA and point mutations in a certain sequence context. Fifth, multiple steps, enzymes, or primers are required in DNA methylation analysis and miRNA detection, leading to compatibility issues due to enzyme/primer competition and different buffer requirements. This also leads to operational complexity and incompatibility with the existing clinical workflows. The cost and scalability of CRISPR-Dx can be a barrier to widespread adoption, especially in resource-limited settings. Sixth, CRISPR-Dx faces several challenges in multiplexing detection, such as maintaining specificity without cross-reactivity, optimizing sensitivity in the presence of multiple primers, multi-enzyme compatibility in one-pot reactions, and streamlining workflows for high-throughput applications [96]. However, accurate diagnosis of cancer often requires the detection of multiple cancer-related biomarkers simultaneously. This multiplexing detection issue could be addressed using orthogonal CRISPR system [97,98] or spatial barcoding [99]. For example, multiple orthogonal Cas proteins (e.g Cas12a, Cas13a, and Csm6) have been used together for multiplex detection in a single reaction tube [97]. The spatial barcoding strategy based on a digital microfluidics platform can also realize multiplex detection with low-volume samples [99]. Other recently developed amplification-free CRISPR-Dx systems are including machine learning-assisted, dual-channel CRISPR/Cas12a biosensor [100] and SERS-signaled, CRISPR/Cas-powered bioassay [101].

Table 1.

CRISPR-Dx for detection of tumor-associated genetic materials.

Nucleic acid	Detection target	Detection strategy	LOD	Sample type	Clinical accuracy	Reference
DNA	Mutation	CRISPR/Cas12a-PCR	0.1%	Tissue	100%	[32]
DNA	Mutation	CRISPR/Cas12a-RAA	0.001%	Blood	100%	[33]
DNA	Mutation	CRISPR/Cas12a	0.01%	Blood	100%	[35]
DNA	Mutation	CRISPR/dCas9 powered impedimetric biosensor	0.65 nM	Blood	> 96%	[38]
DNA	Mutation	CRISPR/Cas13a-PCR	0.01%	Plasma	100%	[40]
DNA	Mutation	CRISPR/Cas12a-based electrochemical biosensor	0.28 fM	Plasma	100%	[41]
DNA	Mutation	SPEAR: CRISPR (NEAR-CRISPR/Cas12b)	0.1%	Blood	100%	[116]
DNA	Mutation	DNA hydrogel-based CRISPR/Cas12a	0.1%, 5 aM	Tissue	100%	[117]
DNA	Mutation	CRISPR/Cas12a-triple-mismatch differentiating (TMD)	0.0001%	Blood	100%	[118]
DNA	Mutation	CRISPR/FnCas9	0.034%	Blood	94.7%	[119]
DNA	Mutation	TtAgo-coupled-multiplex-digtal-RPA-CRISPR/Cas12a	0.1%	Serum	–	[120]
DNA	Methylation	CRISPR/Cas12b-LAMP	10−8 nM	Cell	–	[45]
DNA	Methylation	CRISPR/Cas13a-RPA	0.01%	Cell	–	[46]
DNA	Methylation	CRISPR/Cas12a-RPA	0.05%	Cell	–	[47]
DNA	Methylation	CRISPR/Cas12a-DC-SDA	1.28 × 10−13 M, 0.1%	Serum	100%	[48]
DNA	Methylation	CRISPR/Cas12a-SDA	8.19 fM	Tissue	100%	[49]
DNA	Methylation	MSRE-assisted glycerol-enhanced RPA-CRISPR/Cas12a	100 aM, 0.1%	Cell	–	[121]
DNA	Methylation	CRISPR/Cas12a-MSRE-based cascade exponential amplification	1 aM, 0.1%	Cell	–	[122]
DNA	Methylation	CRISPR/Cas12a-mediated DNAzyme/split-aptamer cascade	1.74 nM	Tissue Blood	–	[123]
DNA	Methylation	meHOLMES-CRISPR/Cas12a	–	Cell	–	[124]
RNA	miRNA	CRISPR/Cas13a-CRISPR/Cas14a	1.33 fM	Serum	100%	[69]
RNA	miRNA	CRISPR/Cas13a-powered electrochemical microfluidic biosensor	10 pM	Serum	100%	[70]
RNA	miRNA	CRISPR/Cas13a-powered electrochemical microfluidic multiplexed biosensor	2–18 pM	Synthetic miRNA	–	[71]
RNA	miRNA	CRISPR/Cas13a-empowered electrochemical biosensor	30.2 fM	Plasma	100%	[72]
RNA	miRNA	Tetrahedral DNA framework based CRISPR-electrochemical biosensor	10 pM	Serum	100%	[74]
RNA	miRNA	CRISPR/Cas12a system based on fragment complementary activation	5.53 fM	Serum	100%	[81]
RNA	miRNA	‘splice-at-will’ CRISPR RNA (crRNA) engineering mechanism	6.0 pM	Cell	–	[125]
RNA	miRNA	Low-background CRISPR/Cas12a sensor	0.96 pM	Cell	–	[126]
RNA	miRNA	“pseudo hybrid DNA-RNA” (PHD) mediated CRISPR/Cas12a	7.7 pM	Vaginal secretion	100%	[127]
RNA	miRNA	5’-end engineering of CRISPR/Cas12a activators	–	Tissue	100%	[128]
RNA	miRNA	CRISPR/Cas12a-aptamer/hemin	0.95 pM	Tissue	–	[129]
RNA	circRNA	Tetrahedron supported CRISPR/Cas13a-electrochemical	0.089 fM	Urine	–	[88]
RNA	circRNA	Polydisperse droplet digital CRISPR/Cas13a	10 aM	Serum	–	[89]
RNA	circRNA	CRISPR-Cas13a/Cas12a	0.013 pM	Cell	–	[130]
RNA	circRNA	Modular RCA-CRISPR/Cas12a	0.125 copies/µL	Blood	95.3%	[131]
RNA	circRNA	Dual Cas12a and multiplex crRNA CRISPR	0.5 fM	Blood	100%	[132]
RNA	circRNA	One-pot Cas13a/13b-RPA	100 copies	Plasma	–	[133]
RNA	circRNA	CRISPR/Cas13a-photoelectrochemical	0.5 fM	Blood	–	[134]
RNA	circRNA	RISPR/Cas12a-coupled nested PCR	10−6 fM	Plasma	AUC = 0.8977	[135]
RNA	lncRNA	RT-RAA-CRISPR/Cas12a	–	Urine	–	[93]
RNA	lncRNA	CRISPR/Cas12a-HCR or CRISPR/Cas12a-CHA	81.6 fM	Serum	100%	[94]
RNA	lncRNA	CRISPR/Cas12a-IACE	37.5 fM	Tissue	100%	[95]
RNA	lncRNA	RT-RPA-CRISPR/Cas12a	0.1 copies/µL	Plasma	–	[136]
RNA	lncRNA	CRISPR/Cas13a-enhanced PEC biosensor	25.5 aM	Blood	93.3%	[137]

Abbreviations: CHA, catalytic hairpin assembly; ctDNA, circulating tumor DNA; DC-SDA, double cascaded strand displacement amplification; HCR, hybridization chain reaction; IACE, isothermal amplification by cyclic extension; LAMP, loop-mediated isothermal amplification; LOD, limit of detection; NEAR, nicking enzyme amplification reaction; PCR, polymerase chain reaction; PEC, photocurrent-polarity-switching photoelectrochemical; RPA, recombinase polymerase amplification; RT-RAA, reverse transcription recombinase-aided amplification; SPEAR, specific point mutation evaluation via CRISPR-Cas assisted recognition.

Lastly, the lack of standardized protocols has also hindered widespread commercialization and clinical applications of the CRISPR-Dx. The operation can be simplified and portable reading can be achieved by combining CRISPR-Dx with microfluidic chips or lateral flow technology. The development of a synchronous detection platform compatible with multiple types of nucleic acids and integrated with liquid biopsy technology can improve the efficiency and flexibility of early tumor screening. Nevertheless, with continued refinement of CRISPR-Dx technology, it holds immense potential for revolutionizing cancer diagnosis and treatment. Apart from nucleic acid-based tumor markers, CRISPR/Cas-based biosensors have been recently developed for detecting protein and enzyme tumor markers. These protein and enzyme tumor markers are substances made by normal cells in response to cancer or by the cancer cells themselves. They can also be detected in bodily fluids to assist cancer diagnosis, treatment monitoring, prognosis assessing, and recurrence tracking. Examples of these protein and enzyme tumor markers are oncofetal antigens, hormones, antigens, DNA repair enzymes, and lineage-specific proteins. Some of the key examples of protein and enzyme tumor markers that have been detected using CRISPR/Cas technology are including PSA (prostate-specific antigen) [102–104], AFP (alpha-fetoprotein) [105–107], CEA (carcinoembryonic antigen) [104,108–110], SCCA (squamous cell carcinoma antigen) [110], VEGF (human vascular endothelial growth factor) [111], APE1 (apurinic/apyrimidinic endonuclease 1) [112], UDG (uracil-DNA glycosylase) [113], PNKP (polynucleotide kinase/phosphatase) [114], and cancer stem cell markers [115]. In general, aptamer, exonuclease III, DNAzyme, DNA nanomachine, nanoparticles, or antibodies were firstly used to recognize the target proteins or enzymes on cancer cell or exosome secreted by tumor cell, followed by activation of CRISPR/Cas system for generating detection signals.

5. Conclusion

CRISPR/Cas hold greats promise for detecting diverse types of tumor-associated genetic materials, including mutated DNA fragment, methylated DNA fragment, ctDNA, miRNA, lncRNA, and circRNA. Despite a number of challenges need to be tackle before it can be translated into clinical settings and commercialization, the continued refinement of these CRISPR/Cas tools will pave the road for developing next-generation cancer diagnosis and prognosis technologies.

Funding Statement

This work was supported by Shantou University Research Initiation Fund Project [NTF20030], Guangdong Provincial Natural Science Foundation General Project [2023A1515011906], Xiamen Municipal Bureau of Science and Technology [QN2023021001L], Natural Science Foundation of Chongqing [CSTB2024NSCQ-MSX0490], Joint project of Chongqing Health Commission and Science and Technology Bureau [2025MSXM075], and Scientific and Technological Research Program of Chongqing Municipal Education Commission [KJZD-M202300102].

Article highlights

• Summarize the key strategies of CRISPR/Cas for detecting tumor-associated genetic materials in clinical samples

• Highlight recent advances and advantages of CRISPR/Dx for cancer diagnosis and prognosis

• Discuss current challenges and technical shortcomings of CRISPR/Dx for clinical implementation and commercialization

• Provide countermeasures and future perspectives to translate CRISPR/Dx into real-world applications

Author contributions

Xiaoqing Chen, Qiaoyuan Ye, Qingle Liang: Writing-review & editing, Writing-original draft, Conceptualization.

Jiahui Li, Yumei Huang, Qiyuan Xia, Jianrong Xiao, Chenyu Liao: Writing-review & editing, Validation.

Cia-Hin Lau, Haibao Zhu: Writing-review & editing, Validation, Supervision, Resources, Project administration, Funding acquisition, Conceptualization.

Disclosure statement

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Writing assistance

No writing assistance was utilized in the production of this manuscript

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Data availability statement

No data was used for the research described in the article.