The protocol was registered with the PROSPERO under the identification number CRD42023474749 (14). This research does not involve using humans or animals, so no institutional review board or ethics committee approval was deemed necessary for this review.

The search for relevant studies was conducted in multiple databases, including PubMed, Scopus, Web of Science, and Google Scholar. Keywords and search terms employed included “non-invasive biomarkers,” “cancer diagnosis,” “cancer treatment,” and variations thereof. Medical Subject Headings (MeSH) terms and Boolean operators were utilized to refine search results. The search encompassed articles published up to 2023. No language restrictions were applied, and non-English articles were considered. Grey literature sources, such as conference proceedings and preprint archives, were explored for potentially relevant studies. Manual searches included examining the reference lists of key articles and contacting experts in the field for additional studies.

Articles were included if they met the Population, Intervention, Comparators, Outcomes, and Study Design (PICOS) Criteria and provided relevant information on non-invasive biomarkers in cancer. Exclusion criteria included studies that did not pertain to cancer lacked full-text availability or were based on animal or in vitro models. No significant deviations from the PICOS framework were applied, ensuring a focused and comprehensive selection process. The PICOS framework employed to define inclusion and exclusion criteria are as follows:

The selection process involved the following steps: Initial screening of titles and abstracts to identify potentially relevant articles. This was followed by a full-text review of selected articles to assess eligibility based on inclusion and exclusion criteria. Discrepancies between reviewers during title and abstract screening were resolved through consensus, and any remaining disagreements during full-text review were addressed through discussion. A flow diagram ( Figure 1 ) illustrates the study selection process.

The authors conducted individual literature reviews and documented their discoveries. Data extraction was conducted independently by two reviewers (SZ and NKN), and disagreements were resolved through discussion and consensus. To facilitate information extraction, a standard table containing characteristics of the studies included in the systematic study analysis was created. To mitigate selection bias, the authors cross-referenced their extracted data after the revision stage and addressed any discrepancies, while duplicate entries were eliminated. If disagreements persisted, a senior researcher (OOO) re-examined the data extraction process. Extracted data items included study characteristics (author, publication year), study design, population characteristics, biomarker types, outcomes, and key findings.

A search across various databases, including PubMed, Scopus, Google Scholar, and Web of Science, yielded 2,500 records. After an initial screening of titles and abstracts, 1,500 records were excluded due to irrelevance to the topic. The remaining 1,000 records underwent full-text assessment. After this thorough review, 925 articles were excluded for reasons including insufficient relevance, not meeting the inclusion criteria, and lack of full-text availability. After a meticulous screening process based on predefined inclusion and exclusion criteria, a further 30 articles were excluded for lack of relevance or inadequate methodology. Ultimately, 45 studies were included in the systematic review.

Table 1 presents a list of selected biomarkers categorized by cancer type, stage, and classification, as identified in the reviewed studies. The included studies exhibited diverse characteristics:

Early cancer detection is the key to improved quality of life and survival and to reducing the financial burden of cancer treatments, which are greater at later stage detection (29). Liquid biopsy (LB) is a term used to describe the analysis of body fluid such as saliva, blood, urine, and cerebrospinal fluid, to identify specific biomarkers associated with cancer development and progression, for example, identification of Circulating Tumor Cells (CTC) as well as circulating tumor cells DNA in blood (29). In the context of cancer management, especially diagnosis and treatment, liquid biopsy which is the analysis of body fluids for circulating tumors cells, cell free nucleic acids, proteins or any other tumor fragments has opened new direction for early tumor detection and precision medicine (20, 21).

The first liquid biopsy application in cancer analysis was circulating tumor cell detection which has different biomarkers depending on the type of cancer. Since many cancers are from epithelial cells there is a universal biomarker used for CTC detection epithelial cell adhesion molecules (EpCAM proteins). Its expression differs from one cancer type to another. It’s mostly used for breast and prostate cancer diagnosis (29). Liquid biopsy offers a non-invasive means for multiple clinical applications in cancer management including early cancer detection, staging and monitoring of localized cancer, predicting relapse and metastatic progression, assessing therapy efficacy, distinguishing early responders from non-responders, and tracking tumor evolution and resistance mechanisms, all of which can significantly enhance patient care and treatment outcomes (30). Therefore, liquid biopsies have a numerous advantages over the traditional biopsy ( Table 2 ) as they help to obtain information from diagnosis to molecular profiling and response assessment without the need of tissue biopsy (31).

While liquid biopsies offer several advantages in cancer management compared to tissue biopsies, they are currently considered a secondary option in clinical applications. This is due to certain limitations ( Table 3 ) that have hindered their full approval as the ‘gold standard’ for cancer diagnosis in clinical settings.

Blood, cerebrospinal fluid (CSF), bone marrow (BM), saliva, sputum, cyst fluid, urine, and saliva are biological fluids that are relevant for liquid biopsy. These fluids can be analyzed to determine whether circulating tumor cells (CTCs), circulating tumor DNA (ctDNA), and other cancer-related biomarkers are present (30). Circulating tumor cells (CTCs) in peripheral blood are one of the emergent topics in cancer research since they can be used as a “liquid biopsy” technique. Noninvasive biomarkers have a clinical potential in the management of solid malignancies, such as prostate, ovarian, and breast cancer (7), for instance fibronectin (FN) which is present on the surface of extra vesicles released from human breast cancer cell lines, was considered as a potential biomarker candidate (32). This liquid biopsy technique to find FN on circulating EVs shows promise as a means of identifying specific markers for early breast cancer diagnosis. Thus, microRNAs enclosed in exosomes that are circulating in biofluids are an intriguing prospective biomarker for cancer because of their expression characteristics specific to cancer (22). In fact, Zhai LY et al. (22) reports an in situ detection of microRNA-1246 (miR-1246) in human plasma exosomes as breast cancer biomarker by a nucleic acid functionalized Au nanoflare probe (22). Recent investigations have shown that genetic alterations or epigenetic modifications in ctDNA could be used for cancer detection with a liquid biopsy like blood (33). Review of the existing literature reveals that methylation biomarkers play a critical role for the diagnosis and prognosis of certain malignancies. Additionally, certain publications describe a “PanCancer” a panel cancer detection technique for the simultaneous identification of multiple cancer types, demonstrating the promise of DNA methylation-based biomarkers for cancer detection and management (23). The use of liquid biopsies for cancer management has become more popular in clinical research and has many uses (24), including diagnosis, treatment and therapeutic monitoring. Given that DNA methylation in plasma can be identified early in the development of cancer pathogenesis, blood-based epigenetic biomarkers have a great promise for early cancer detection. Under specific conditions, those markers can break out of their dormant state and stimulate proliferation, which can ultimately result in a distant relapse and cancer-related death (34). As shown in Table 4 , different liquid biopsy approaches can be used to detect, characterize, and monitor minimal residual disease in breast cancer, prostate cancer, and melanoma (35).

Extracellular vesicles (EVs) are minute, lipid-bound particles secreted by cells under a variety of diseased and healthy conditions. They transport proteins and nucleic acid as part of their cargo (25). EVs from cancer cells aid in recruiting normal cells to promote tumor growth through proliferative signaling and apoptosis evasion. Research revealed that EVs containing RBM11 from glioblastoma cells induce oncogenic splicing in recipient cells, enhancing survival (36). In vivo studies on mice confirmed the malignancy-promoting potential of these EVs (37). Similarly, EVs from glioblastoma cells transfer CLIC1 protein to support tumor growth. Melanoma-derived EVs transfer PDGFR-β, activating PI3K/Akt pathway and inhibiting MAPK pathway in recipient cells, boosting proliferation. Bladder and gastric cancer cell EVs activate PI3K/AKT and MAP/ERK pathways, promoting proliferation and halting apoptosis in recipient cells. Extracellular vesicles may be used as circulating biomarkers for a variety of diseases, including various malignancies, there has been a notable upsurge in scientific interest in these molecules in recent years ( Figure 2 ).

Recent years have seen a significant increase in scientific interest in EVs because of their potential use as circulating biomarkers for a wide range of illnesses, including several cancers (39) (see Table 5 ). Studies have shown a significant correlation between the content of EVs, such as proteins, nucleic acids, and lipids, and various aspects of cancer biology, including tumor growth, angiogenesis, immune evasion, and drug resistance (45). The cargo carried by EVs can serve as potential biomarkers for cancer diagnosis, prognosis, and monitoring treatment response.

Exosomes: Serum exosomes from glioblastoma patients contain mutant EGFRvIII mRNA. Zhou and colleagues reported that the exosomal miR-15a-5p expression levels in endometrial cancer are 7–19 times higher than those in other cancer types (46). A poor prognosis was indicated by a substantial correlation between high exosomal miR-1247–3p expression and pulmonary metastases from liver cancer. Since exosomes are released by living cells and may reflect the pathophysiological condition of their parent cells, they are useful indicators for dynamic monitoring of disease progression (26). Increased miR-21 in circulating exosomes has been characterized as a potential biomarker in a number of malignancies, including liver, colorectal, gastric, breast, ovarian, and esophageal cancer. Higher levels of exosomal miR-21 found in urine have been correlated to bladder and prostate cancers (40).

Microvesicles: Microvesicles that developed on the cell surface of platelets were demonstrated to discharge lipid-rich molecules having procoagulant potential into their surroundings. The surface shedding, or “ectocytosis,” was later found to occur in a range of cell types, including tumor cells, the cells, neutrophils, and monocyte (47). Both benign as well as malignant colorectal tumor patients had significantly higher plasma concentrations of microvesicles. Microvesicles as biomarkers may improve the usefulness of colon cancer screening systems. Microvesicles are not only useful for detecting cancer, but also serve as biological markers that provide prognoses for many diseases. There are several neurological conditions such as Alzheimer’s disease, epilepsy that have been associated with increases in specific types of circulating microvesicles (48).

Ectosomes: Ectosomes are microscopic heterogeneous membrane vesicles that form when several cell types, most typically tumor cells, proliferate from the plasma membrane. They are characterized as a new form of intracellular interaction in which data is sent without physical touch between source and recipient cells. Beyond the plasma membrane, ectosomes are specialized, multipurpose carriers that expand the bounds of a cell. They establish communication networks that let cells share specific traits and information. Ectosomes as potential targets for biomarkers, diagnostic tools, and cancer therapy (41). The state of the living thing from which ectosomes emerge largely determines the specific composition of the substances they convey and their intended purpose. Because tumor-derived ectosomes are present in physiological fluids such as the blood and urine of cancer patients, they may prove to be useful prognostic and predictive biomarkers for breast and prostate cancers. Furthermore, a range of therapeutic modalities may target tumor-derived ectosomes (42).

Oncosomes: Extracellular vesicles called oncosomes are excessively large (1–10μm in diameter) and associated with severe disease. They are thought to have originated from malignancy. When membranes bleb shed, they are created. Fluorescence microscopy of large EVs revealed a morphology similar to oncosomes, indicating that these entities are oncosomes (43). One of the proteins localized in oncosomes, cytokeratin 18 (CK18), has been found to be highly prevalent (within the top fifth percentile) and was used in the creation of a test to detect oncosomes in tissues and circulatory of both human and mouse prostate cancer patients. These results imply that oncosomes are a distinct type of extracellular vesicles, or EVs, that can play a variety of roles in the growth of tumors and provide markers specific to malignancy. Potentially useful biomarkers for cancer diagnosis and prognosis are oncosomes. They are applied to prostate and breast cancer (49).

Prostasomes: Broadly expressed RNA, membrane, and cytosolic amino acids that are unique to the prostatectomy make up prostatesomes. Extracellular vesicles taken from people with prostate cancer have been shown to include altered levels of protein, mRNA, long non-coding RNA (lncRNA), and microRNA, both in regard to number and quality (27). RNA, which is membrane, and cytosolic amino acids that are unique to the prostatectomy and are broadly expressed make up prostatesomes. The extracellular vesicles taken from people with prostate cancer have been shown to include altered levels of protein, messenger RNA, long non-coding RNA (lncRNA), and microRNA, both in regard to number and quality (28).

Apoptotic bodies: Cells can perish by a variety of processes, including necrosis, autophagy, and catastrophic mitosis. Currently, however, it is believed that mediators of the death of apoptotic cells have tremendous potential as targets for cancer therapies. Morphological alterations include chromatin and its process of condensation, cell shrinkage, plasma membrane blebbing, and the creation of apoptotic bodies are indicative of apoptosis (50). Cytosolic cytochrome c activates the apoptosome complicated initiating caspase 9, and effector caspases. Apoptotic bodies are produced as a result of a sequence of irreversible events that include the fragmentation of cytokeratins (CKs) by caspase, poly (ADP-ribose) polymerase, with the reactivation of endonucleases to form nucleosomal DNA (nDNA) according to Ye et al, 2020 ( Figure 3 ). Furthermore, they promote the exterior of the plasma membrane to become exposed to phosphatidylserine, which allows phagocytes to identify dying cells. Indicators of apoptosis in breast cancer include circulating soluble FASL, granzyme B, and cytochrome C, which increase following treatment. Both intact PCa cells and apoptotic particles made of PCa cells are seen in urine (52). Patients receiving medication can eventually release these biomarker molecules into their circulation as shown in Figure 2 .

Autophagic EVS: Autophagy is one remarkably conserved method of cellular breakdown (53).Cancer cells’ autophagy stimulates the formation of tumors and the division of cancer cells, which inhibits the development of new cancers by killing cancer cells. The autophagic process is regulated by a number of proteins, including class III PtdIns3K complex, Bcl-2, Atg proteins. A degradative organelle known as the vacuole/lysosome receives portions of the cytosol and organelles that are sequestered into an autophagosome, a double-membrane vesicle, for eventual breakdown and recycling of the resultant macromolecules. Thus, autophagy both safeguards breast cancer cells and lessens their sensitivity to medicines. Consequently, autophagy may offer cytoprotection to breast cancer cells (19). Development of protocols to monitor autophagy could be useful in diagnosis and treatment monitoring.

The elements other than DNA sequence that influence gene expression and cellular phenotypes are referred to as epigenetics (54). The topic of how different phenotypes might be derived from the same genotype is addressed by epigenetics (18). The processes that add acetyl and methyl groups to histone tails, methylate DNA on cytosine residues, express non-coding RNA, and modify the structure of chromatin are the most well-understood epigenetic determinants of phenotype. To maintain the correct differentiation state, cells’ collective epigenetic status is strictly regulated (54–57). This precisely tuned genetic programming is upset in cancer, a process known as epimutation. This results in defective differentiation, unchecked cell division, and resistance to apoptosis (54). A heritable aberrant transcriptional suppression of gene activity that is unrelated to a DNA sequence is known as an epimutation. Although it can also occur in the germline, epimutation usually happens in somatic cells and shows up in the growth of tumors (18). Over the past forty years, epigenetic errors and their causes have become a prominent focus in cancer research when it was discovered that aberrant DNA methylation is associated with malignancy. It has been demonstrated that changes to the epigenome affect almost every stage of the development, growth, and management of tumors (54).

Histone alterations, non-coding DNA, and DNA methylation Since RNAs are found in all human cancer types and can manifest in the early stages of the disease, they make particularly appealing markers with a variety of diagnostic uses (58). Because of their notable stability over RNA and proteins, among other things, DNA methylation and microRNAs are more useful and feasible as biomarkers in therapeutic contexts (59). Specifically, great stability is provided by DNA methylation, microRNAs, and post-translational changes of histones in biofluids and low-quality materials like formalin-fixed paraffin embedded (FFPE) (60). Other advantages of epigenetic biomarkers over genetic or protein-based biomarkers are, that they are dynamic in nature, give more information about the function of the gene, thereby providing information about the specific genetic programs that alter during disease (60). By definition, an epigenetic biomarker is “any altered epigenetic mechanism or mark that is specifically stable and reproducible during sample processing and is generally used to evaluate health or disease status” (60).

The most extensively researched epigenetic modification in cancer is aberrant DNA methylation (61). In eukaryotic cells, aberrant hypermethylation of promoters can silence critical genes, including tumor suppressor genes, which in turn can cause illness. The reverse process can also have an impact on the development of cancer. Genes that are typically methylated, such as oncogenes, can have their expression elevated by hypomethylation (62). It is interesting to note that the first DNA methylation anomaly in human cancer to be discovered was hypomethylation (63). It has been found that 13% of sporadic colorectal cancer (CRC) show MLH1 hypermethylation, and a BRAF c.1799T>A, p.Val600Glu mutation has often also been identified in tumor DNA (64, 65). Though it is brought on by mutations in one of the DNA MMR genes, MSI and loss of MLS1 are both present in Lynch syndrome, the most frequent cause of hereditary colorectal cancer (66). It has been discovered that methylation of MGMT occurs in 40% of tumors in gliomas and CRC, but only in 25% of tumors in non-small cell lung carcinomas (NSCLCs), lymphomas, and head and neck carcinomas (67). When paired with IDH1 mutations, MGMT methylation status functions as a predictive biomarker. Patients with hypermethylated MGMT and the IDH1 p.R132H mutation had a better prognosis for their glioma (68). MGMT is a DNA repair gene that helps to eliminate harmful and mutagenic alkyl groups from O6-meG. Due to the fact that DNA alkylation causes mutations, MGMT shields cells from harm (67, 69).

The primary association of the RB1 gene with retinoblastoma is the loss of RB1 function. LOH or RB1 mutations are linked to the lack of expression of this gene in retinoblastoma and other malignancies, such as bladder carcinomas and malignant neuroendocrine lung carcinomas. However, methylation of RB1 might sometimes result in the suppression of its expression (70, 71). It has been said that RB1 methylation above the LOH and mutations are required for complete molecular diagnoses of retinoblastoma. It has been reported that 9% of spontaneous unilateral tumors are caused by RB1 hypermethylation, which is invariably acquired (72).

Circulating methylation SEPT9 DNA is one type of plasma epigenetic biomarker for colorectal cancer screening. SEPT9 is regarded as a tumor suppressor because it controls cell proliferation and inhibits unchecked cell division (73). Research has shown that SEPT9 methylation is linked to the pathophysiology of colorectal cancer (CRC), and that a decline in SEPT9 expression is connected with the advancement of neoplastic illness (74). SHOX2 hypermethylation has been noticed in the bronchial aspirates [43], pleural effusions [44], and blood plasma of patients with lung cancer (18, 75). DNA methylation analysis of SHOX2 combined with PTGER4 in blood plasma allows detection of lung cancer and differentiation of non-malignant diseases (75). In some cancer there are some epigenetic markers that has been discovered due to the fact that prostate cancers frequently contain methylation of the tumor suppressor genes GSTP1, RASSF1, and APC, these genes are regarded as cancer biomarkers ( Table 6 ) (18).

Generally, chromatin can be divided into two categories: euchromatin, which is more loose and contains actively transcribed genes, and heterochromatin, which is heavily compacted and contains dormant genes (18). The arrangement and functionality of chromatin are changed by covalent alteration of the histones that make up nucleosomes, which has an impact on the regulation and expression of genes (90). The six main roles of chromatin are transcription, repression, enhancer, insulator, promoter, and inactive chromatin. Histone modification is a significant factor in determining the function of chromatin (91). Phosphorylation, acetylation, methylation (mostly of lysine and arginine residues), ubiquitylation, glycosylation, SUMOylation, ADP (adenosine diphosphate)-ribosylation, and carbonylation are examples of modifications to histone structure (90, 92). The primary indicators of active chromatin are histone acetylation and methylation, which are frequently linked to a more relaxed chromatin conformation. Conversely, chromatin condensation is frequently linked to histone deacetylation and phosphorylation, which are indicators of inactive chromatin (90).

Globally methylated or non-modified histones were linked to a poor prognosis, whereas individuals with NSCLC who had global histone acetylation had a better prognosis in survival analysis (93). Human tumor cells are typically characterized by a complete loss of H4 histone Lys16 monoacetylation and Lys20 trimethylation, which is linked to DNA hypomethylation. A lower level of H4Lys20 methylation and H4Lys16 acetylation in breast cancer is associated with a worse prognosis (94, 95). The propensity for cancer can be increased by different isoforms of histone proteins found in the nucleosome as well as covalent alterations. For instance, genitourinary malignancies, which include bladder and prostate tumors, and undifferentiated cancers have been shown to overexpress the H2A histone isoform H2A.Z. Moreover, H2A.Z may contribute to endocrine resistance in individuals with breast cancer. The association between H2A.Z levels and short overall patient survival suggests that H2A.Z may be a valuable biomarker for tumor progression (96). In terms of post-translational histone changes, CRC patients’ blood had lower levels of H3K9me3 and H4K20me3 than that of cancer-free people (97). Global patterns of histone H3 and H4 modification are significant because they may serve as indicators of both disease-free survival and tumor recurrence ( Table 7 ).

The second most extensively researched epigenetic method of gene regulation, after methylation, is the interaction between microRNA (miRNA) and messenger RNA (mRNA). These are 18–25 nucleotide short, non-coding RNA molecules that are essential for controlling post-transcriptional gene expression. Translation is halted by miRNAs when they bind to the target mRNA molecule. It is believed that miRNA can regulate up to 60% of genes that encode proteins (18). Saliva, urine, serum, and plasma are among the bodily fluids into which tumor cells release miRNAs. Thus, the examination of circulating miRNAs in liquid biopsy samples offers potential biomarkers for non-invasive diagnostics in numerous human cancers, such as melanoma and rhabdomyosarcoma, as well as colorectal, lung, breast, prostate, gastric, pancreatic, esophageal, liver, thyroid, kidney, ovarian, endometrial, and cervical cancers (101). The control of numerous genes implicated in the genesis of cancer is greatly impacted by dysregulation of miRNA expression (18). The suppressor gene may be silenced as a result of overexpression of miRNA, which is implicated in the negative regulation of the suppressor gene. Conversely, overexpression of the oncogene occurs when the chromosomal loci encoding the miRNA that silences the oncogene are deleted. Therefore, oncogenes (oncomiRs) and miRNAs themselves can function as suppressors. It is also critical to keep in mind, that while a single miRNA molecule can control several genes, multiple miRNAs can target a single mRNA (18). Since miRNA activity and expression changes along the course of cancer development, miRNAs can function as biomarkers and be assessed in cancer patient blood and tumor tissue ( Table 8 ) (18).

The combination of two miRNA biogenesis genes (DICER1 and DROSHA) and four miRNAs (miR-30d, miR-21, miR-17, and miR-155) is one of the putative predictive indicators in non-small cell lung cancer (114). Specifically, miR-30d has been discovered to function as an oncomiR in cancer, and a substantial decrease in lifespan is associated with an increased copy number of miR-30d in cancer tissue (gains or amplifications compared to others) (114, 115). miR-31–3p is an additional intriguing prognostic and predictive biomarker in metastatic colorectal cancer (116). Blood alterations are a reflection of the dysregulation of many miRNAs found in malignancies. Let-7a-1, 7a-2, 7a-3, 7b, 7c, 7d, 7e, 7f-1, 7f-2, 7g, 7i, miR-98, and miR-202 are a group of well-known miRNAs that have changed expression in malignancies (117). Since some members of the let-7 family are downregulated in melanoma, pancreatic cancer, prostate cancer, and sarcoma, the let-7 family is thought to be a tumor suppressor (118). However, they can also be upregulated in lymphoma, mesothelioma, and breast cancer (118). It was discovered that miR-21 acts as an oncogene by suppressing the expression of many tumor suppressors (119). For instance, the miRNA targets PDCD4, which is linked to the prevention of neoplastic transformation as well as the promotion, invasion, and advancement of cancer (118–120). Among the other targets of miR21 are BCL2, PTEN, RECK, RHOB, and TPM1 (121). This miRNA has been linked to hematological malignancies, breast cancer, gastric cancer, ovarian cancer, pancreatic cancer, colorectal cancer, lung cancer, and liver cancer in terms of its diagnostic, predictive, and/or prognostic properties (18).

In recent years, extensive research has elucidated the multifaceted roles of ncRNAs in cancer. Non-coding RNAs (ncRNAs) have emerged as pivotal players in the intricate landscape of cancer biology, contributing significantly to both the understanding of malignancies and the development of innovative therapeutic strategies (122, 123). These molecules, including microRNAs (miRNAs), long non-coding RNAs (lncRNAs), and circular RNAs (circRNAs), exert regulatory functions in gene expression and cellular processes ( Figure 4 ). The dysregulation of ncRNAs has been linked to various malignancies, making them attractive candidates for diagnostic and therapeutic applications.

MiRNAs: MiRNAs, small non-coding RNA molecules, have garnered attention for their involvement in cancer pathogenesis (125–129). Notably, elevated levels of Phosphatase and Tensin Homolog (PTEN) induced by specific miRNAs can inhibit AKT signaling, activate apoptosis, and prevent malignancies such as renal cell carcinoma (130, 131). MiRNAs, including miR-29a, have demonstrated a major impact on oncogenicity in various neoplasms by regulating key genes involved in cancer progression (128). These molecules play crucial roles in physiological and pathological processes, including viral replication, cell proliferation, differentiation, apoptosis, fibrosis, angiogenesis, tumorigenicity, metastasis, and drug resistance. Research has identified specific miRNA panels with distinct expression patterns in the serum of cancer patients, offering potential as diagnostic indicators (132). For instance, a combination of miR-145, miR-155, and miR-382 demonstrated improved sensitivity and specificity, suggesting the potential of miRNA profiling for breast cancer screening (132). The translational applications of miRNAs extend to the management and survival improvement of oral cancer patients (129).

LncRNAs: Sometimes referred to as versatile biomarkers across cancers, long non-coding RNAs (lncRNAs) have emerged as versatile biomarkers with diagnostic potential across various cancers (133). A systematic review and meta-analysis highlighted lncRNA AFAP1-AS1 as a novel biomarker in different cancers (134). Specific lncRNAs, such as MALAT-1, HOTAIR, LINC00152, and others, have shown diagnostic potential in prostate, lung, colorectal, hepatocellular, gastric, renal, and colorectal cancers (135). LncRNA GIHCG has been linked to the etiology of numerous malignancies, offering promise as a biomarker (136). Additionally, circulating lncRNAs, like PVT1 and UCA1, exhibit significant multicancer diagnostic potential (137). These findings underscore the diversity of lncRNAs as biomarkers and their potential application in clinical setting.

CircRNAs: Circular RNAs (circRNAs) represent a novel class of non-coding RNAs with emerging roles as biomarkers and therapeutic targets in cancer (138). The discovery of their involvement in the onset and progression of malignancies has opened new frontiers in cancer research. CircRNAs such as circ0001955 and circ-LDLRAD3 have shown promise as diagnostic and prognostic markers in cervical and pancreatic cancers, respectively (139). Notably, exosomal circ_0044516 was found to be highly elevated in prostate cancer patients, influencing cancer cell proliferation and metastasis by modulating miR-29a-3p expression (140). These findings underscore the potential of circRNAs as valuable biomarkers and therapeutic targets in diverse cancer types.

Metabolomic biomarkers refer to specific small molecule metabolites produced by the organism in biological fluids that can be identified and analyzed to provide insights into physiological or pathological states. These biomarkers, detected through nontargeted metabolomic analysis, serve as indicators of metabolic changes associated with various conditions, including diseases or responses to treatments. Essentially, metabolomic biomarkers are measurable metabolic features that can be used for diagnostic, prognostic, or therapeutic purposes, providing valuable information about the biochemical status of an organism (141). These compounds typically weigh ≤ 1500 Da and span a diverse range, including peptides, oligonucleotides, sugars, nucleosides, organic acids, ketones, aldehydes, amines, amino acids, lipids, steroids, alkaloids, and occasionally drugs or xenobiotics (141). Metabolomic biomarkers have emerged as promising tools for non-invasive diagnostic and treatment monitoring in oncology. Metabolomics technologies have advanced our understanding of cancer metabolism, particularly in the context of the “Warburg effect,” which elucidates how cancer cells utilize glycolysis to support tumor proliferation and vascularization (142, 143).

The tumor metabolome provides insights into the interconnectedness of metabolome, proteome, and genome within cancer cells. Cancer cells exhibit altered metabolic processes, such as glycolysis, which converts glucose into pyruvate and subsequently ferments it into lactate ( Figure 5 ).

The flow of pyruvate through the TCA cycle is reduced in cancer cells. Additionally, pathways stemming from glycolysis, like the pentose phosphate pathway, generate essential building blocks to support the rapid growth of cancer cells. Specific genetic and enzyme-related behaviors play a role in this process. Enzymes highlighted in blue are crucial for the transition to a cancer metabolic phenotype, while those in orange indicate mutations found in cancer cells. Oncogenes, represented by green ovals, are up-regulated in cancer, whereas tumor suppressors, depicted by red ovals, are down-regulated.

In breast cancer, the integration of genomics, proteomics, and metabolomics has been proposed as a key approach for future biomarker discovery, highlighting the potential of metabolomics in this field (145). Furthermore, computational models applied to metabolomics data have hinted at the relevance of glutamine metabolism in breast cancer, emphasizing the potential of metabolomics in the development of new biomarkers for cancer (146). Metabolomics holds great promise for understanding the molecular determinants of cancer and advancing the development of new biomarkers for the diagnosis, prognosis, and treatment of neoplastic processes (147). Additionally, metabolomics offers a broad set of oncological applications, particularly in providing serum or imaging-based biomarkers for cancer (148). The utility of metabolomics in biomarker discovery for cancer has been demonstrated in various types of cancer, including colorectal cancer, where it has proven useful in early diagnostic biomarker discovery (149). The potential of metabolomics in cancer research extends to its application in precision medicine, as it can suggest new pathways and therapeutic targets for the targeted treatment of cancer (150).

Recent advances in metabolomics technologies have enabled a deeper investigation into cancer metabolism, providing a better understanding of how cancer cells utilize metabolic pathways for proliferation and vascularization (142), for instance, metabolic biomarkers for breast cancer, with a focus on glutamine metabolism (146). Additionally, metabolomics has been applied to urine and saliva for non-invasive cancer detection and biomarker discovery (151). Furthermore, metabolomics represents a potential strategy for the real-time selection and monitoring of patients treated with immunotherapy, indicating its relevance in treatment monitoring in oncology (152). The integration of metabolomic profiling with transcriptomics data has been proposed as a method for validating potential diagnostic biomarkers in cervical cancer, further emphasizing the potential of metabolomics in cancer research (153). Recent studies have shown the potential of metabolomics in identifying biomarkers for various cancers, such as endometrial cancer and primary glomerulonephritis sub-types (154) ( Table 9 ).

While the studies reviewed in this article provide compelling evidence for the utility of non-invasive biomarkers in cancer medicine, there is still much work to be done in this field. Despite the advances in cancer biomarker research, several limitations hinder their clinical application. One significant issue is the lack of standardized protocols for biomarker detection and quantification, leading to variability in results across different studies and clinical settings (166). Additionally, many biomarkers currently lack sufficient validation in large, diverse patient populations, which raises concerns about their generalizability and reliability (167). Future research should focus on validating these biomarkers in larger patient cohorts, exploring their potential for use in combination with existing diagnostic and treatment modalities, and developing new technologies to enhance their sensitivity and specificity. Another limitation is the complexity of cancer biology, which makes it challenging to identify biomarkers that are both highly specific and sensitive for early detection and prognosis. Current biomarkers often fail to capture the heterogeneity of cancer, leading to false positives and negatives (168). There is need for more extensive research into the underlying mechanisms of the origin of these biomarkers and their significance in cancer biology. This knowledge will be crucial for developing more effective therapies that target the specific molecular pathways involved in cancer development and progression. Early detection is very important for the improvement of life quality, survival, and to reduce the financial burden of cancer treatments, which are greater at later stage detection. Moreover, the integration of biomarkers into clinical practice is complicated by the lack of robust bioinformatic tools to analyze and interpret large-scale biomarker data (169). To address these challenges, future research should focus on the following areas:

Standardization and Validation: Develop standardized protocols for biomarker detection and validation through multi-center studies involving diverse patient populations. This will help ensure the reliability and reproducibility of biomarker tests across different settings (166).

Multi-omics Approaches: Integrate genomic, transcriptomic, proteomic, and metabolomic data to identify composite biomarker signatures that better reflect the complexity of cancer. This holistic approach can improve the specificity and sensitivity of biomarkers (167).

Advanced Bioinformatics: Invest in the development of advanced bioinformatic tools and machine learning algorithms to analyze complex biomarker data. These tools can help uncover novel biomarker patterns and improve the interpretation of existing data (169).

Ethical and Safe Integration of AI: Explore the ethical and safe integration of artificial intelligence (AI) in cancer research to enhance biomarker discovery and application. AI can help in the rapid analysis of large datasets and the identification of potential biomarkers with high clinical relevance (170).

Combining Biomarkers with Existing Therapies: Investigate the potential of combining biomarkers with existing therapeutic strategies to enhance treatment efficacy and overcome resistance. This includes exploring the role of biomarkers in predicting and monitoring treatment response (171).