Liquid biopsy – a narrative review with an update on current US governmental clinical trials targeting immunotherapy

Abstract

Aim

This study aims to present a comprehensive international analysis of the existing techniques used in liquid biopsies and their use in isolating tumor markers to detect, predict, and monitor the results of cancer treatment.

Materials and Methods

We conducted a narrative review using a scoping review model based on three databases, including PubMed/Medline, Scopus, and Cochrane. The search criteria included all articles on liquid biopsy of the last five years (June 30th, 2023–Oct 30, 2024) ((liquid Biopsy) AND ((“2023/06/30”[Date – Publication]: “2024/10/30”[Date – Publication]))). We also approached gray literature on this topic. We focused on review articles as an eligibility criterion for this narrative review, but we also carried out a United States registered clinical trials review targeting immunotherapy and liquid biopsy with the limitation “recruiting” and/or “not yet recruiting” (updated on March 31, 2025).

Results

We screened 2645 articles from PubMed/Medline, Scopus, and Cochrane and 45 articles from the gray literature. We retrieved the full text for 325 articles. Liquid biopsies involve the extraction of tumor-derived components such as circulating tumor cells, circulating tumor DNA, and tumor extracellular vesicles from the bodily fluids of cancer patients. We found 25 United States registered governmental clinical trials targeting immunotherapy and liquid biopsy, of which 20 trials are recruiting and five trials are not yet recruiting.

Discussion

Developments in medicine have led to a more comprehensive understanding of tumor features, including tumor load, tumor staging, heterogeneity, gene mutations, and clonal evolution. The utilization of liquid biopsies from cancer patients has provided novel opportunities for detection and ongoing monitoring, precision medicine-based therapy, and identification of markers for therapeutic resistance.

Article Highlights

• Liquid Biopsy Applications: Liquid biopsy, which analyzes biomarkers in blood or other body fluids, holds promise for early diagnosis, screening, prognosis, and monitoring treatment response in cancer.

• Liquid Biopsy Advantages over Tissue Biopsy: Liquid biopsy offers a less invasive alternative to tissue biopsy, allowing for serial sampling, longitudinal disease progression, and treatment response monitoring.

• Specific Applications: Liquid biopsy can be used to identify minimal residual disease, detect early relapse, and guide treatment decisions. It can analyze various tumor components and biomarkers, including circulating tumor cells (CTCs), circulating tumor DNA (ctDNA), and other cell-free products.

• Contemporary approaches for collecting tumor EVs from liquid biopsies: Multiple techniques have been published that efficiently separate EVs from diverse forms of cellular waste, exploiting distinct physical and biochemical features that aid in EV separation. Over 50% of EV isolation methods involve preparative ultracentrifugation. Differential, isopycnic, and moving zone ultracentrifugation reduce EV loss and contamination, improving EV purity. Nanomembrane ultrafiltration concentrators show a promising approach.

• Current US Clinical Trials of Immunotherapy and Liquid Biopsy: Twenty open recruiting clinical trials are using liquid biopsy for tumors requiring immunotherapy, and they are highly promising.

1. Introduction

Recently, molecular profiling of tumors derived from individual patients has demonstrated enhanced efficacy in selecting tailored cancer treatments, assessing patient responses, identifying drug resistance, and monitoring tumor recurrence [1,2]. The conventional way to profile cancers is to acquire resected tumor samples through invasive surgical procedures [3]. An inherent constraint of these invasive techniques is obtaining tumor samples that meet the required number and quality standards. Furthermore, the acquisition of biopsy samples using invasive techniques throughout therapy to track tumor response and recurrence also presents a significant obstacle in monitoring tumor evolution and tumor profiling [4]. The overall heterogeneity of excised tumor samples further restricts the application of invasive techniques [5].

Moreover, it is important to be able to monitor tumor evolution over time. In the scenario of metastasis, when tumors have disseminated and continuously undergo evolutionary changes affecting the molecular cancer drivers as a result of progression and therapy, it may be necessary to repeat several biopsies since it is challenging to get a comprehensive picture of a neoplasm [4]. In response to the difficulties encountered in conventional biopsies, contemporary oncology research has redirected its attention toward analyzing different biological fluids instead of complete tissues to identify tumor-derived components. This approach is known as liquid biopsy (LB). Blood continuously perfuses tumors in intimate contact, often acquiring components of the cancer, including cells and extracellular vesicles, as well as nucleic acids and proteins.

However, other bodily fluids such as mucosa, pleural effusions, urine, and cerebrospinal fluid (CSF) are also rich sources of tumor cell components and can be collected and examined [6]. Recently, laser beam surgery has offered improved diagnostic sensitivity and allows for repeated sampling during therapy in a more convenient and non-invasive manner [7,8].

Molecular profiling of tumors derived from individual patients has demonstrated enhanced efficacy in selecting tailored cancer treatments, assessing patient responses, identifying drug resistance, and monitoring tumor recurrence [1–3]. An inherent constraint of these invasive techniques is obtaining tumor samples that meet the required cellularity and quality standards. Furthermore, the acquisition of biopsy samples using invasive techniques chronologically and sequentially throughout therapy to track tumor response and recurrence also presents a significant obstacle in tumor profiling or transplantation programs [4,9]. The overall heterogeneity of excised tumor samples further complicates and restricts the application of invasive techniques [5].

In the scenario of metastasis, tumors have disseminated and continuously undergo changes in both location and time as a result of therapy. In this event, it may be necessary to repeat several biopsies since it is challenging for the pathologist to get a comprehensive picture of a tumor [4,10–22].

LB can be subjected to the evaluation of various tumor-derived components, including circulating tumor cells (CTCs) released by primary and metastatic tumors, circulating tumor DNA (ctDNA), tumor-derived extracellular vesicles (EVs) consisting of nucleic acids/proteins, tumor-educated platelets (TEPs), and circulating cell-free RNA (cfRNA) composed of small RNAs/miRNAs with advantages and disadvantages, which will be highlighted in the following sections. TEPs are platelets that have been altered by interactions with cancer cells, displaying changes in their RNA and protein profiles, and potentially playing a role in cancer progression and metastasis. Collectively, these components derived from tumors can offer vital longitudinal information and data to enhance the precision of pathologists in diagnosing both primary and metastatic malignancies. Some of the key actionable tumor-derived biomarkers include data such as DNA mutations, copy number alterations (CNAs) of essential genes, transcriptome/proteome profiling, epigenetic modifications, metabolite profiling, and more. CNAs, including gains or losses of DNA segments, are a hallmark of cancer and can alter gene expression, influencing tumor progression and therapy resistance. CNAs can be detected using various methods, including next-generation sequencing, and understanding these alterations is crucial for diagnosis, prognosis, and personalized treatment strategies. Bioinformatics techniques are used to analyze disease signatures and are increasingly popular on several research platforms [23–35].

This review presents a comprehensive examination of current and recent biomarker techniques, including the extraction of cancer stem cells, or CTCs, ctDNAs, and EVs from the blood of patients with different cancer types. It also highlights the applications of these extracted entities in identifying, predicting, and treating various types of cancers. We focused on review articles as an eligible criterion for this narrative review, but we also carried out a United States (US) registered clinical trials review targeting immunotherapy and liquid biopsy with the limitation “recruiting” and/or “not yet recruiting” (updated on March 31, 2025). Despite being exhaustive, this review will not review the details of specific techniques. Some details will be provided as Supplementary Material.

2. Materials and methods

A comprehensive narrative review was conducted using a scoping review-based model of three databases, including PubMed/Medline, Scopus, and Cochrane. The search criteria included all articles on liquid biopsy of the last five years (June 30th, 2023–Oct 30, 2024) ((liquid Biopsy) AND ((“2023/06/30”[Date – Publication]: “2024/10/30”[Date – Publication]))). We also approached gray literature on this topic. Grey literature encompasses information produced outside traditional commercial or academic publishing channels, including reports, conference proceedings, theses, and government documents. The more detailed breakdown of grey literature components includes the following items: (1) Reports, (2) Working Papers, (3) Government Documents, (4) Conference Proceedings and Abstracts, (5) Theses and Dissertations, (6) Technical Reports, (7) Patents, (8) Unpublished Clinical Trials, (9) Newsletters and other internal communications, (10) Blogs and Social Media Posts, (11) Market and Industry Research, and (12) Standards and Guidelines. This review remains a narrative review. We carried out a narrative review instead of a systematic or scoping review. The main objective of a systematic review is to formulate a well-defined research question. In a systematic review, qualitative and quantitative methods are used to analyze all the available evidence attempting to answer a specific question. In contrast, narrative reviews can address one or more questions simultaneously with a much broader scope. The efficacy of narrative reviews is irreplaceable in tracking the development of a scientific principle or a clinical concept or reviewing a condition using broad scopes. Apart from the author’s preferences, a narrative review structure must respect the journal style and conventions followed in the respective field (e.g., laboratory diagnostics). Moreover, to make the report compelling and attractive, and considering the expertise of the senior authors with randomized clinical trials, we carried out a US Government NCBI (National Center for Biotechnology Information) search for clinical trials (recruiting and not yet recruiting) dealing with immunotherapy and liquid biopsy.

3. Results

Two thousand six hundred forty-five articles have been collected through PubMed/Medline, Scopus, and Cochrane and screened. Gray literature identified 45 articles. We retrieved the full text for 325 articles. Liquid biopsies involve the extraction of tumor-derived entities such as CTCs, ctDNA, and tumor EVs from the bodily fluids of cancer patients. These obtained entities are then subjected to an examination of the genomic and proteomic information they carry. Significant advancements have recently been made in isolating and analyzing liquid biopsies. These developments have led to a more comprehensive understanding of tumor features, including tumor growth, tumor staging, heterogeneity, gene mutations, and clonal evolution. The utilization of liquid biopsies from cancer patients has provided novel opportunities in detection and ongoing monitoring, precision medicine-based therapy, and identification of markers for therapeutic resistance.

4. Liquid biopsy – advantages and disadvantages

Liquid biopsy is becoming standard practice quite diffusely. Liquid biopsy techniques, analyzing body fluids for tumor markers, offer minimally invasive, rapid, and potentially real-time tumor monitoring but can have limitations in sensitivity and specificity and may not identify all types of biomarkers. Table 1 illustrates the advantages and disadvantages of liquid biopsy. Substantially, the advantages of liquid biopsy include minimal invasiveness, rapid and real-time monitoring, accessibility and serial sampling, assessment of tumor heterogeneity, potential for early detection, and lower cost. On the other side, liquid biopsies harbor some disadvantages, including low sensitivity and specificity, limited biomarker detection, challenges in specificity, the necessity to have an initial histologic diagnosis, technical challenges, limited clinical utility in some settings, and, currently, the lack of standardization of procedures and data analysis methods for wider clinical application. Table 2 provides a summary of the clinical validity and reliability of liquid biopsy biomarkers across various cancer types.

Table 1.

Liquid biopsy advantages and disadvantages.

Paths	Advantages	Disadvantages	Information output
ctDNA	Simple isolation process TMB detection Methylation status	Low rate in blood NGS-based Time-consuming	Gene mutations Gene amplifications Gene translocations Gene deletions Gene methylation
ctRNA	Tumor tissue profiles consistency Alternative splice events	High instability Half-life: 15 sec No standardized extraction protocol	RNA splice variants LncRNAs miRNAs profiles
CTCs	Metastasis indication Neoplasm-specific Rapid informative data	Low rate High heterogeneity Cell capture procedure challenging	Theranostics (drug screening) Cytogenetics
Exosomes	Surface proteins expression Genomic contents protection from degradation Copious	Ad hoc isolation procedures (low standardized protocols) Detection limits (Ø: 40–160 nm)	Surface vs. Intravesicle proteins Genetic analysis RNA profiling
Proteome	Primary mediators in biologic activities Drug target availability Innovative detection protocols	Rigorous analysis Epigenetic controls RNA splicing & PTM Continuous shifts in	Clinic-based protein assays Proteomics

Notes: LncRNAs, long non-coding RNAs; NGS, next-generation sequencing; TMB, tumor mutational burden; PTM, Post-translational modifications. Extracellular vesicles (EVs) are released by all living cells and are formed from bilayer lipid membranes. EVs consist of a variety of subtypes, including exosomes, microvesicles (MVs), ectosomes, oncosomes, and apoptotic bodies; “EV” is understood to be a generic term that refers to secreted vesicles. The term “EV” is often used interchangeably with the term “exosome,” but exosomes have now become defined as a size-specific subset of EVs ranging from 30 to 100 nm in diameter. The tumor mutation load refers to the quantity of alterations (termed genetic mutations) present in the DNA of cancerous cells. Tumor mutational burden (TMB) is a biomarker that may assist physicians in assessing a patient’s response to medical treatment. Patients receiving immunotherapy may undergo tumor mutation burden assessment.

Table 2.

Summary of some scientific contributions displaying the clinical validity and reliability of liquid biopsy biomarkers across various cancer types.

Neoplasm	Biomarker (test)	TP/(TP + FN)	TN/(TN + FP)	INV.	# Studies	# Subj.	Notes	1st Au (YYYY)	REF
NSCLC	EGFR mt. (NGS)	81.0%	97.0%	SR	142	N.A.	High Specificity	van Delft (2020)	[36]
	EGFR mt. (ddPCR)	High	High	CS	1	3	Incomplete Data	Akhoundova (2020)	[37]
	MRD (ctDNA)	41–76%	86–95%	MA	16	1,251	P-TRT-RP	Zhong (2023)	[38]
GAS-CA	SpecGastro Test	76.6%	89.2%	Ob-S	1	282	DNA methyl.	Ma (2022)	[39]
	CTCs	42.0%	99.0%	MA	20	1,030	>2 CTCs/mL	Tang (2013)	[40]
	cfDNA (WGS)	94.9%	99.4%	Ob-S	1	131	Multi-omic profiling	Song (2025)	[41]
CRC	LB (multiple)	77.0%	89.0%	MA	1	18,739	CTCs, EVs, cfDNA	Zhu (2020)	[42]
PDAC	PAC-MANN-1	73.0%	98.0%	Ob-S	1	356	Cost-effective, fast	Montoya Mira (2025)	[43]
	ctDNA	High	High	NR	31	3,309	Pre-OP & Met	Jonnalagadda (2025)	[44]
	CTCs	84.62%	96.7%	Ob-S	1	52	High Specificity	Kim (2022)	[45]
	Exosomes	93.3%	91.0%	OB-S	1	105	Verita™ platform	Hinestrosa (2023)	[46]
OVA-CA	EVs (OC)	97.0%	73.5%	Ob-S	1	390	Strong design	Winn-Deen (2024)	[47]
Prost-Ca	SelectMDx	High	53.0%	NR	1	N.A.	mRNA-CA+PSA	Groen (2022)	[48]
	EPI	93.0%	26.1%	Ob-S	1	503	Urine, 3 RNAs	McKiernan (2018)	[49]
Multiple	Grail-LLC-Galleri	98.6%	99.6%	NR	15	N.A.	High Specificity	Ebbert (2025)	[50]
	PanSeer Test	95.0%	96.0%	NR	1	123,115	≤4 yrs pre-DGN	Beer (2020)	[51]
	CGGA-Galleri	67.3%	99.3%	NR	1	15,254	>50 cancers	Beer (2020)	[51]
	cfDNA+ (CancerSEEK)	33–98%	99.0%	NR	1	1,800	cfDNA + PrBMs	Beer (2020)	[51]
	cfDNA+ (CancerSEEK)	94.0%	92.0%	Ob-S	1	396	Stage ∼ Sensitivity	Nguyen (2025)	[52]

Notes: NSCLC, non-small cell lung cancer; GAS-CA, gastric carcinoma; CRC, colorectal carcinoma; PDAC, pancreatic ductal adenocarcinoma; OVA-CA, ovarian carcinoma; Prost-Ca, prostate carcinoma; Multiple, several cancer; SR, systematic review; CS, case series; MA, meta-analysis; Ob-S, observational study; NR, narrative review; PSA, prostate-specific antigen; DGN, diagnosis; PrBMs, protein biomarkers; Stage ∼ Sensitivity, sensitivity variably depending on staging; P-TRT-RP, post-treatment relapse-prediction; K19-MB-IF, K19-MB-IF, cytokeratin 19-coupled magnetic beads immunofluorescence. The sensitivity and specificity values may fluctuate depending on factors such as cancer stage, the biomarker examined, and the technological platform employed. Current research and clinical studies persist in improving these methodologies to augment their diagnostic precision and predictive efficacy in practical applications. This table summarizes liquid biopsy biomarkers’ clinical validity and reliability across different cancer types derived from existing research and reports. For a given test and disease/condition, its sensitivity is how well it can be positive among all those with it. Therefore, sensitivity = TP/(TP + FN), i.e., true positives/(all those with the disease). On the other hand, specificity is also critical for accuracy. The specificity for a given test and disease/condition is how well it can distinguish those with disease from those without. Therefore, specificity = TN/(TN + FP), i.e., true negatives/(all those without the disease). P-TRT-RP, post-treatment relapse predictor; EPI, ExoDx Prostate-IntelliScore; Zhong et al. determined that ctDNA MRD is a promising biomarker for predicting relapse in lung cancer patients’ post-definitive therapy, with excellent specificity but inadequate sensitivity, regardless of whether a landmark or monitoring method is employed. While surveillance ctDNA MRD analysis reduces specificity relative to the landmark technique, the reduction is negligible compared to the enhancement in sensitivity for predicting lung cancer relapse. Ma et al. showed that the SpecGastro test can detect three types of gastrointestinal cancers with high sensitivity and specificity. Sensitivities for CRC, Ga-Ca, and Eso-Ca were 87.8% (95% CI: 79.2–93.2%), 69.9% (95% CI: 61.3–77.3%), and 72.9% (95% CI: 57.9–84.3%), respectively. A high AUC of 0.903 (95% CI: 0.873–0.928) indicated that the SpecGastro test accurately distinguished gastrointestinal cancer patients from control subjects. Cheng et al. used cytokeratin-19-binding magnetic nanoparticles and immunofluorescence to detect CTCs in their patients’ blood (K19-MB-IF). Zhu et al. pooled studies showed high sensitivities and specificities. The sensitivities were 0.82 (95% CI 0.79–0.85), 0.76 (95% CI 0.72–0.80), and 0.76 (95% CI 0.75–0.77) for CTCs, exosomes, and cfDNA, respectively. The specificities were 0.97 (95% CI95% CI 0.95–0.99), 0.92 (95% CI 0.89–0.94), and 0.88 (95% CI 0.87–0.89) for CTCs, exosomes, and cfDNA, respectively. The Ebbert et al. study highlights that GRAIL LLC’s Galleri^® test utilizes a single biomarker class, cfDNA methylation, to detect a shared cancer signal across multiple neoplastic types while also predicting the source of the neoplastic signal. Galleri^® was initially developed and validated in the case-control Circulating Cell-free Genome Atlas (CCGA) study (NCT02889978), showing an overall specificity of 99.5%. In Winn-Deen’s study, the Mercy Halo Ovarian Cancer Test (OC Test) employs immunoaffinity capture of tumor-associated extracellular vesicles, subsequently utilizing proximity-ligation real-time quantitative PCR to identify combinations of up to three biomarkers, thereby enhancing specificity, while assessing multiple combinations to optimize sensitivity. The references of the studies included in this table are [36–52].

5. Proliferating malignant cells in liquid biopsies

The detection of tumor cells in the peripheral blood of patients dates back to the 1860s, and substantial advancements have been made since then in isolating cancer stem cells from a diverse range of blood cells [53]. Initially, CTCs are liberated from primary tumors in the tissue. They then migrate through the circulatory system and contribute to the formation of metastatic (or secondary) malignancies at remote locations in the body [54]. Quantitatively, the proportion of CTCs in the bloodstream is somewhat modest, with, on average, one CTC detected per million leukocytes [55]. Considerable research has demonstrated that colorectal cancer (CRC) exhibits a range of morphological forms influenced by the tumor’s stage and/or type [54].

Furthermore, circulating tumor cells (CTCs) can form clusters by binding to cells such as fibroblasts and platelets. These clusters have been observed to migrate to more remote locations in the body than individual CTCs. Cellular aggregates of this nature are, therefore, potentially shielded from oxidative damage and the adjacent immune system [56,57]. CTCs have become highly important in the identification of tumors, substituting invasive tissue biopsies not only because they are easy to sample but also because they provide real-time information on the state of the tumor. The quantity of CTCs exhibits a more dynamic pattern, providing a higher degree of accuracy compared to typical blood biomarkers, as they fluctuate in tandem with the treatment status of the tumor [58,59]. In a large cohort of breast cancer patients, it has been shown that lower numbers of CTCs are associated with improved overall survival, making them a more reliable indication of general prognosis and also specific patient response to therapy [60]. CTCs have potential value for the early detection of many forms of malignancies, such as lung cancer, albeit in a limited subset of individuals with chronic obstructive pulmonary disease [61]. The confirmation of their diagnostic worth was achieved by the identification of lung nodules and subsequent histologic examination of excised tissue samples. Recent studies have shown that lung biopsies in combination with CTCs may distinguish between the benign and malignant conditions of pulmonary lesions [62].

With regard to technological advancements in the separation of CTCs from liquid biopsies see also the Supplemental Information.

6. Utilization of liquid biopsy-derived cancer stem cells (CTCs) in clinical oncology

The analysis of mutational profiles in CTCs for genes associated with tumors offers vital insights into tumor characterization and the prediction of therapeutic response outcomes. Secondary point mutations in the epidermal growth factor (EGFR) gene, characterized by the substitution of methionine with a threonine residue (T790M), have been linked to tumor recurrence and provide resistance to otherwise efficacious treatment drugs like gefitinib and erlotinib in lung adenocarcinoma subjects [63]. Notably, these alterations can accurately forecast the effectiveness of newer irreversible EGFR tyrosine kinase inhibitors [64–66]. Characterizing the mutational profiles of colorectal CTCs for EGFR-related genes such as KRAS and PIK3CA suggests potential therapeutic effects in CRC [67,68]. These investigations demonstrate that differences in the expression and genetic changes of these genes among individuals may explain their different rates of response to treatment in CRC [69]. Therefore, the analysis of the molecular and genetic characteristics of CTCs obtained from patients helps to identify constantly changing alterations in the tumor gene composition in real-time (which cannot be detected by traditional tissue biopsy) and to develop more advanced and efficient treatment strategies.

CTCs obtained from lymph nodes might also offer significant insights into the epigenetic modifications of several tumor-related genes in malignancies [70]. Epigenetic modifications, such as DNA methylation, in the promoter regions of tumor/metastasis suppressor genes, including SOX17, BRMS1, and CST6, in EpCAM-positive CTCs obtained from patients with mammary malignancies, are associated with increased tumor metastasis and unfavorable prognosis [71–80]. Comparable changes in the methylation patterns of genes, such as VEGF and SFRP2, which are linked to the formation of blood vessels (angiogenetic properties), have been detected in CTCs obtained from patients with prostate cancer and colorectal cancer, respectively [81,82]. Moreover, studies have shown that CTCs isolated from liver cancer are a more effective diagnostic tool than tissue biopsies in identifying epigenetic changes in cancer-relevant genes [82]. Epigenetic modifications, including estrogen receptor 1 methylation, have been shown to indicate treatment resistance to chemotherapy regimens such as everolimus and exemestane in breast cancer patients, in addition to contributing to prognosis and diagnosis [83–85].

Likewise, the methylation patterns of non-coding RNAs (ncRNAs) linked to the shift from epithelial to mesenchymal lineages, such as miR-200, are increased in CTCs obtained from individuals with prostate cancer [86–88]. Alterations in the epigenetic patterns of different genes in CTCs serve as biomarkers that help guide prognosis, track tumor response, and indicate related changes in cellular processes that indicate tumor spread. It is essential to mention that epigenetic changes shown by cancer stem cells (or ctDNAs, as explained later in the study), while highly useful in predicting the cancer prognosis, may not always accurately represent the condition of primary tumors that undergo continuous evolution [72,73].

Tumor-derived CTCs can also be utilized in living organisms to create patient-derived tumor models that aid therapeutic interventions. Previous studies have demonstrated that breast cancer xenografts consisting of luminal breast cancer CTCs include microRNAs that can cause metastases in the bone, liver, and lungs of mice [89]. The study found an association between CTC surface markers, such as EPCAM positive, CD44 positive, CD47 positive, and MET positive, with increased metastatic sites and decreased survival rates. This data supports the development of improved diagnostic tools for treating metastatic breast cancer [74–76,90,91]. Comparable investigations on xenografts produced from LC CTCs have yielded a more comprehensive understanding of therapeutic drug trials, illness prognosis, and resistance mechanisms [92]. Research has also concentrated on producing uninterrupted cell lines from CTCs to address the issue of limited cells. For instance, the CTC-MCC-41 cell line derived from patients has demonstrated a consistent phenotype that shares characteristics with its original malignancies [93]. This enables the conduct of functional studies and both in vivo and in vitro drug therapy trials [93]. CTCs can also be useful for evaluating protein markers for treatment with Antibody-Drug Conjugates (ADC) an emerging important class of cancer drugs targeting many new cellular proteins including HER2, FGFRB2, Claudin18, NRG1, PDL1 and others [94]. Figure 1 shows the microfluidic technology for the molecular characterization of patient-derived circulating tumor cells at the single-cell level.

Figure 1.

Microfluidic technology for the molecular characterization of patient-derived circulating tumor cells at the single-cell level. Different procedures are displayed, including (a) a process connected with ClearCell^® FX. Analysis of single-cell genomics: A significant concordance rate of EGFR mutations (T790M and L858R) was observed between NSCLC circulating tumor cells and corresponding primary tumors. In the single-cell transcriptomic study, patient classification for breast cancer and/or NSCLC was conducted via comprehensive mRNA transcriptomic analysis and targeted gene expression profiling. In the single-cell metabolomics investigation, supervised principal component analysis (PCA) demonstrated distinct metabolic patterns between circulating tumor cells (CTCs) and lymphocytes in gastric and colorectal cancer individuals. In (b) single-cell proteome analysis a microfluidic single-cell western blotting (scWB) facilitated the swift examination of an eight-plex protein expression in ER+ breast cancer. In (c) a single cell secretomic analysis is performed. Here the integrated microfluidic on-chip device demonstrated significant heterogeneity in the expression profiles of two secreted proteins (IL-8 and VEGF) in CTCs from patients with lung cancer (Source: Lim SB, Di Lee W, Vasudevan J, Lim WT, Lim CT. Liquid biopsy: one cell at a time. NPJ Precis Oncol. 2019 Oct 2;3:23. doi: 10.1038/s41698-019-0095-0. PMID: 31602399; PMCID: PMC6775080).

7. Analysis of tumor DNA in liquid biopsies

Previous research conducted by Leon et al. initially showed that individuals diagnosed with pancreatic cancer had increased amounts of ctDNA in their blood samples, which appeared to decline after treatment [95]. Subsequent investigations unveiled that those tumors not only exhibited changes in levels of ctDNA but also in their ctDNA sequences. ctDNA samples obtained from the plasma of cancer patients revealed mutations in oncogenes, including KRAS [96]. Furthermore, multiple investigations have confirmed that ctDNA (or chromosomal fragments) can be transmitted horizontally by apoptotic bodies released by tumor cells, leading to host cell genetic alterations and facilitating cellular transformation and metastasis [97]. Significantly, ctDNA constitutes a mere 0.1–10% of the overall circulating cell-free DNA (cfDNA), with typical plasma concentrations ranging from 10–100 ng/ml [98]. Pathological conditions such as inflammation or physical activity are also recognized to increase levels of cfDNA, which may not always indicate the presence of underlying cancer [99]. Furthermore, the quantity of ctDNA in the plasma differs depending on the tumor amount, stage, and treatment response [100]. Furthermore, apart from measurement, the clinical use of ctDNA in precision medicine also enables the examination of ctDNA variations included in the plasma. Recent research has demonstrated that complementary DNA (cDNA) has a distinct length compared to the circulating cDNA pool. Reports suggest that ctDNA fractions in cancer patients range from 20 to 50 base pairs, somewhat shorter than cfDNA [101].

8. Utilization of liquid biopsy-derived cDNA in clinical oncology

Previous research has shown that cDNA offers a comprehensive perspective on the features and development of tumors originating from both primary and metastatic tumor sites (Table 3) [102,103]. Furthermore, alterations that were not discovered by traditional tissue sampling have been examined utilizing complementary DNAs from liver cancers [104]. Genetic sequencing of ctDNA has also helped identify tumor-specific changes in gene copy numbers in prostate cancer and exposes the dynamic character of cancer cell genomes, where gene amplifications are essential for cancer development [105]. Computational DNA profiling has also facilitated the monitoring of clonal variants in CRC patients, therefore aiding in the real-time assessment of tumor advancement and resistance to EGFR blocking therapy [106]. Clonal profiling of tumor cells using cDNA has been employed in prostate cancer to identify androgen receptor mutations that are resistant to chemotherapeutic treatments such as abiraterone or prednisolone. Chromosomal ctDNA profiles of clones in these investigations demonstrate both geographical and temporal tumor heterogeneity resulting from variations in resistance mechanisms at distinct tumor locations. Rare aggressive tumors, such as malignant angiomatoid fibrous histiocytoma [107], with uncertain malignant potential, are also critical to be monitored using cDNA (Figure 2).

Table 3.

Potential therapeutic uses of LB in different malignancies.

LB entity	Tissue	Diagnosis	References
EVs	Prostate	GPC1, KRAS	[95]
EVs	Prostate	miRNAs, CD44v6, CD104, Tspan8, EpCAM	[96]
EVs	Lunga	miR-23b-3p, miR-10b-5p, miR-21-5p	[77,97]
EVs	Lunga	miR-125b-5p	[98]
EVs	Lunga	miR-146a-5p	[99]
EVs	Skin	miR-211-5p	[100]
EVs	Skin	PD-1, CD28	[101]
CTCs	Lungb	EGFR	[102]
CTCs	CR	KRAS, PIK3CA	[56]
CTCs	Breast	SOX17, BRMS1, CST6	[64]
CTCs	Prostate, CR	VEGF and SFRP2 (promoter methylation)	[68,69]
CTCs	Breast	ER1 methylation	[103]
CTCs	Breast	EPCAM, CD44, CD47, MET expression	[104,105]
ctDNA	CRc	Genomic profiling	[93,106]
ctDNA	Prostated	AR	[93,106]
ctDNA	Bloode	DNA profiling	[107,108]
ctDNA	Ovary, CRe	DNA profiling	[109–111]
ctDNA	Breast, CRe	DNA profiling	[112,113]
ctDNA	CRf	KRAS, NRAS, BRAF	[114,115]
ctDNA	“Solid”g	PIK3CA, RB1, MED1, GAS6, EGFR	[116,117]

Notes: In non-small cell lung cancer (NSCLC) (^a) miR-23b-3p, miR-10b-5p, and miR-21-5p provide the biomarker diagnosis of a noninvasive phenotype. At the same time, miR-125b-5p predicts an improved T-cell activity, and miR-146a-5p predicts the chemosensitivity of the tumor. In lung adenocarcinoma (^b), EGFR gene mutations predict the sensitivity of the tumor to gefitinib and erlotinib. In CRC (^c), genomic profiling tracks clonal variation and therapeutic response. In prostate carcinoma (^d), AR gene mutations predict the tumor response to abiraterone or prednisolone, while DNA profiling of ctDNA (^e) in blood for B-cell lymphoma, ovarian carcinoma, breast cancer, and CRC determines the tumor subtypes, clinical outcome, residual disease, and relapse. The KRAS, NRAS, and BRAF gene mutations (^f) analysis predict the tumor response to panitumumab and cetuximab. In contrast, gene mutations in PIK3CA, RB1, MED1, GAS6, and EGFR (^g) predict the tumor response to paclitaxel, cisplatin, tamoxifen, lapatinib, and gefitinib.

Figure 2.

3D Neoplastic Angio-Chemotaxis. A malignant angiomatoid fibrous histiocytoma is seen invading a blood vessel and the 3D representation of the tumor cells approaching and invading (chemotaxis) a blood vessel is shown in the lower inset. A malignant angiomatoid fibrous histiocytoma refers to a rare, low-grade soft tissue neoplasm, also known as “Angiomatoid Fibrous Histiocytoma,” which typically presents as a slow-growing, painless mass, most found in the extremities of children and young adults. This tumor has a low potential for spreading, but chemotaxis has been demonstrated, and the surgical removal alone cannot rule out a subsequent malignant behavior. This tumor requires a biopsy to confirm the diagnosis based on the characteristic microscopic features, but close monitoring is highly recommended because surgical excision cannot predict alone the behavior of this neoplasm.

Furthermore, copy number DNA genotyping is recognized for its ability to identify tumor subtypes in individuals with B cell lymphoma, therefore aiding in the prediction of clinical outcomes and the provision of individualized therapy [48]. Comparatively, a greater quantity of cancer stem cells or several solid tumor biopsies would be necessary to get comparable results, hence emphasizing the effectiveness of ctDNA biopsies. Furthermore, in contrast to CTCs, ctDNAs have been shown to function as biomarkers that indicate the tumor size, as demonstrated by research conducted on patients with ovarian cancer and lung cancer [108,109]. It has been shown that the presence of ctDNA in numerous types of malignancies, including ovarian cancer [108], lung cancer [109], and CRC [110], is associated with rather poor clinical outcomes and tumor relapse, highlighting its prognostic importance in cancer progression and therapeutic response. Furthermore, the monitoring of ctDNA profiles in patients with breast cancer and CRC has not only provided a prognosis but also facilitated the identification of residual disease after treatment and the likelihood of recurrence. This, in turn, enables therapy adjustment and prevents excessive treatment. The examination of ctDNAs using Plasma-Seq demonstrates a diverse range of mutations or aberrations that serve as predictive indicators of resistance to treatments in different types of cancer [49,50]. KRAS-, NRAS-, and BRAF-related mutations in the plasma ctDNA of metastatic CRC patients cause first resistance five to six months after receiving anti-EGFR treatments such as panitumumab and cetuximab. In advanced malignancies, mutations in PIK3CA, retinoblastoma 1 (RB1), mediator complex subunit 1, growth arrest-specific 6, and EGFR result in resistance to medicines such as paclitaxel, cisplatin, tamoxifen, lapatinib, and gefitinib [51]. Curiously, although possessing distinct genetic origins, resistance mechanisms that emerge gradually appear to converge at critical signaling foci. Hence, ctDNAs have a multitude of uses in the fields of diagnosis, prognosis, therapy, and drug resistance control within precision oncology.

Epigenetic changes, including methylation patterns of tumor-derived ctDNA, have been identified as biomarkers for different types of malignancies and mutations [16,39–41,43–46,52,111–115]. Approaches for identifying cDNA methylation can be categorized into two types: targeted targeting of certain sections of the genome and a comprehensive genome-wide strategy, similar to screening ctDNA mutations [116]. In order to detect methylated DNA sequences, several PCR-based techniques, including methylation-specific PCR (MSP) [117], quantitative multiplexed MSP [118], and methylation on beads [119], necessitate prior knowledge of the targeted region. Studies have demonstrated that fluorescence-based real-time alterations to traditional molecularly imprinted polymers (MSPs) can effectively enable the quantitative identification of methylation patterns [120]. For detecting late-stage breast cancers and monitoring tumor growth, therapy response, and disease recurrence, assays such as cMethDNA, a modified version of QM-MSP, are highly effective [121]. This test offers improved methylation signals of new biomarkers in the sera of breast cancer patients, with more sensitivity and repeatability compared to traditional quantitative methylation-mass spectrometry (QM-MSPs) protocols. Methylation levels were demonstrated to be associated with treatment response and serve as a link between cDNA and its biological tissue of origin [40,41,111–115,121–138]. These investigations revealed that epigenetic traces were still present in the sera of patients many years after being diagnosed with the condition. Shotgun massively parallel bisulfite sequencing of the whole genome has demonstrated superior sensitivity and specificity in detecting cDNA methylations. Methylated CpG tandem amplification and sequencing have been used to identify many hypermethylated CpG islands in cDNA samples as low as 7.5 picograms in the plasma of patients with HCC.

9. Extracellular vesicles derived from liquid biopsies of tumors

EVs are small, membrane-bound, saucer-shaped vesicles measuring 30–100 nm that are extracellularly produced by cells and present in several body fluids, including plasma, urine, cerebrospinal fluid (CSF), saliva, and others, and have been useful for both tumor and non-tumor theranostic purposes [13,139–144]. Although formerly considered cellular waste products that help eliminate degraded endosomal or lysosomal components, EVs have since been shown to have important functions in several types of cell-to-cell communication [52]. EVs transported with various biomolecules, such as DNA, RNA, protein, etc., have been shown to have essential functions in intercellular communication [145–147]. Tumor-derived EVs have attracted significant interest due to their elucidated functions in facilitating tumor growth, epithelial-to-mesenchymal transition, metastasis, immunosuppression, angiogenesis, and other related processes [140,145,146,148,149]. Since tumors have been demonstrated to release large quantities of EVs, their concentration in the plasma of individuals with malignancies reaches exceptional levels [146]. Due to the presence of important cargo such as tumor-derived DNA, mRNA, ncRNAs, proteins, etc. [150], the study of tumor-derived EVs provides valuable information for tumor monitoring, prognosis, and therapy response [13,141,142,151].

Renal cell carcinoma (RCC) remains a formidable diagnostic challenge, especially in the context of small renal masses. The quest for noninvasive screening tools and biomarkers has steered research toward liquid biopsy, focusing on microRNAs (miRNAs), exosomes, and CTCs. MiRNAs, small non-coding RNAs, exhibit notable dysregulation in RCC, offering promising avenues for diagnosis and prognosis. Studies underscore their potential across various biofluids, including plasma, serum, and urine, for RCC detection and subtype characterization. Encouraging miRNA signatures show correlations with overall survival, indicative of their future relevance in RCC management. Exosomes, with their diverse molecular cargo, including miRNAs, emerge as enticing biomarkers, while CTCs, emanating from primary tumors into the bloodstream, provide valuable insights into cancer progression. Despite these advancements, clinical translation requires further validation and standardization, encompassing larger-scale studies and the generation of robust evidence. Currently lacking approved diagnostic assays for renal cancer, the potential future applications of liquid biopsy in follow-up care, treatment selection, and outcome prediction in RCC patients are profound.

Despite advancements, RCC diagnosis remains challenging, especially for small renal masses. The urgency for noninvasive screening tools and biomarkers led to the exploration of liquid biopsy, focusing on miRNAs, exosomes, and CTCs. MiRNAs exhibit dysregulation in RCC, offering diagnostic and prognostic potential. Promising miRNA signatures have been identified, correlating with overall survival, emphasizing their future role in RCC management. Similarly, exosomes have emerged as promising biomarkers due to their ability to carry diverse cargo, including miRNAs, while CTCs, shedding from primary tumors into the bloodstream, offer insights into cancer biology. While these liquid biopsy approaches hold promise, their clinical translation necessitates further validation and standardization. Challenges include the need for larger studies, robust evidence, and the identification of specific markers. Currently, there are no approved diagnostic assays for renal cancer, highlighting the potential future applications of liquid biopsy in follow-up care, therapeutic selection, and outcome prediction in RCC patients. Ongoing research is expected to refine these biomarkers and enhance their clinical utility in the comprehensive management of RCC.

10. Selected applications of EVs generated from liquid biopsies of tumors in clinical oncology

Protein or nucleic acid cargo analysis of EVs after enrichment and purification has shown effectiveness as diagnostic and prognostic indicators in various malignancies. Tumor-derived EVs that express biomarkers such as CD63 and caveolin-1 have been identified as potential indicators of melanoma [152–155]. EVs enriched with migration inhibitory factor are predictive markers of liver metastasis in patients with prostate cancer. Tumor-derived EVs containing markers like prostate-specific transglutaminase and stem cell antigen are indicative of tumor burden in patients with prostate cancer. New research has emphasized the significance of tumor-derived EVs in detecting early-stage prostate cancers. In these investigations, tumor-derived EVs were found to be rich in biomarkers such as glypican-1, a cell surface proteoglycan, and carried KRAS mutations [156]. In addition to markers attached to the cell membrane, exosomal contents such as DNA and RNA are also recognized as invaluable sources of information for diagnosing and assessing the response to treatment in patients with different forms of cancer. Previous studies have shown that increased levels of miRNA cargos from miR-1246, miR-4644, miR-3976, and miR-4306 serve as very sensitive and noninvasive indicators in individuals with prostate cancer [157,158]. These investigations demonstrated a comparable diagnostic capability for exosomal proteins such as CD44v6, Tspan8, EpCAM, and CD104. EVs that contain high levels of miRNAs, including miR-23b-3p, miR-10b-5p, and miR-21-5p, in the plasma of patients have been shown to serve as important noninvasive indicators for NSCLC [159]. Library-based exosomal miRNAs profiles also help assess therapy’s effectiveness against malignancies. Downregulation of immunosuppressor miRNA, such as miR-125b-5p, in post-treatment plasma EVs, has been associated with enhanced T-cell function and greater response to immunotherapy [160]. An analogous relationship between exosomal miRNAs and the effectiveness of chemotherapeutic medications has been established in the context of cisplatin used for NSCLC. Results indicated that higher levels of exosomal miR-146a-5p led to enhanced sensitivity of NSCLC to the chemotherapeutic treatment.

In contrast, the expression levels of miR-146a-5p were reported to decrease in both NSCLC cell lines and extracellular vesicles following cisplatin resistance. Furthermore, it has been proposed that increased miRNA profiles of miR-211-5p in melanoma and melanoma-derived vesicular secretome are suggestive of therapeutic resistance established against BRAF inhibitors, including vemurafenib, which is employed in the treatment of metastatic melanomas [161]. Furthermore, these results corroborated the notion that increased levels of miR-211-5p are associated with decreased susceptibility to BRAF inhibitors in living organisms. The exosomal payload, including DNA, also contributes to acquiring firsthand information about malignancies. Studies have shown that mutations in exosomal DNA at sites containing tumor-relevant genes such as KRAS and p53 can be a reliable indicator of prostate cancer [162]. Not only tumor-derived EVs but also serum EVs produced by immune cells have been shown to serve as biomarkers that can predict a clinical response to chemotherapeutic antibiotics. Elevated concentrations of surface markers PD-1 and CD28 on EVs produced by T-/dendritic cells in patients with melanoma have been associated with enhanced clinical response to chemotherapeutic medications such as ipilimumab [163]. Several recent investigations have reported that tiny EVs obtained from serum may play a significant role in the surveillance of brain tumors [140].

In addition to miRNAs, additional non-coding RNAs such as lncRNAs (long non-coding RNAs) and circRNAs (circular RNAs) found in tumor-derived EVs have also demonstrated effective functions in detecting and surveilling malignancies. Specifically, circular RNAs such as circ_0044516 have been found to increase blood exosomes generated from prostate cancer [164,165]. The suppression of circ_0044516 expression in prostate cancer cells was shown to be associated with a decrease in the proliferation and spread of cancer cells. Similar functions of non-coding RNAs, such as exosomal lncRNAs, have been reported in prostate cancer [166]. The upregulation of exosomal lncRNA H19 in serum has been seen in bladder cancer cases, suggesting its potential applicability as a significant diagnostic marker [167]. Likewise, increased concentrations of long non-coding RNAs PCA3 and BCAR4 have been documented in the blood samples of individuals diagnosed with CRC. Serum exosomal lncRNAs ENSG00000258332.1, LINC00635, and HEIH show potential diagnostic functions in liver cancer [168,169]. Furthermore, apart from their diagnostic role, exosomal lncRNAs also have utility in predicting the prognosis of tumors. For instance, lncRNA MALAT1 is related to epithelial ovarian cancer, while lncRNA MALAT1, PCAT-1 is associated with breast cancer [170,171]. Moreover, exosomal lncRNAs are utilized extensively in drug resistance surveillance in tumors, suggesting their possible clinical utility in cancer treatment. Tumor-derived RNA H19 is positively associated with gefitinib resistance in non-NSCLC patients [172]. Similarly, lncRNA PART1 is similarly associated with gefitinib resistance in esophageal squamous cell carcinoma [173,174]. LncRNA ARSR is specifically linked to sunitinib resistance in advanced renal cell carcinoma (RCC) [175]. Lastly, lncRNA UCA1 is expressed in cetuximab-resistant CRC cells [176]. Thus, exosomal lncRNA mediated lymph node biopsies have great potential for real-time monitoring of tumor diagnosis, progression, and recurrence in a diverse range of malignancies. In Table 2, the clinical validity and reliability of LB biomarkers across various cancer types are summarized.

This table provides a consolidated overview of the current clinical validity and reliability of LBbiomarkers across various cancer types, based on available studies and reports.

11. US governmental clinical trials for immunotherapy and liquid biopsy

We found 25 clinical trials dealing with immunotherapy and liquid biopsy, recruiting and not yet recruiting (Table 4).

Table 4.

US governmental clinical trials.

NCT #	Acronym	State	Conditions	-#	Type	Start date	End date
NCT05810402	LILIPSY	NYR	HCC	60	INT	2025-05-23	2025-05-27
NCT06849518	NA	RECR	L-NSCLC	30	INT	2025-02-11	2027-03-01
NCT04111029	HCCGenePanel	RECR	HCC	30	OBS	2020-01-16	2025-03-24
NCT05254132	SALMON	NYR	L-NSCLC	1000	OBS	2022-07-01	2025-06-30
NCT05704530	NA	RECR	Eso-Cancer	248	INT	2023-03-29	2026-12-31
NCT06710093	PerceIVe	NYR	Melanoma, Met+	100	OBS	2025-02-01	2029-01-01
NCT04790682	LIBERTYLUNG	RECR	Lung-Cancer, Met+	300	INT	2021-05-27	2028-06-01
NCT02511288	LIBIL	RECR	L-NSCLC	900	OBS	2025-07-15	2025-12-26
NCT05853887	iNUDGE	RECR	L-NSCLC, Met+	360	INT	2023-06-15	2025-12-25
NCT05099068	PLANET	RECR	Advanced Metastatic Solid Tumors, GBM, CLL	500	INT	2021-11-16	2025-09-15
NCT06869239	NA	RECR	LS-SCLC	65	OBS	2025-02-25	2028-01-03
NCT04966663	ctDNA Lung RCT	RECR	L-NSCLC, R0	66	INT	2022-03-28	2026-12-01
NCT05059444	ORACLE	RECR	Multiple Cancers	1050	OBS	2021-09-07	2025-02-28
NCT05427227	NA	RECR	Advanced or Late-Stage GI Cancer	500	OBS	2022-07-01	2025-07-01
NCT05889247	PRELUCA	RECR	L-NSCLC, Met+	350	INT	2023-07-28	2032-06-01
NCT04243720	IRIS	RECR	Cancer|Solid Tu|Met + Ca|IR	100	OBS	2020-08-26	2025-02-25
NCT06611072	ACTION	NYR	O-Cancers, IS	90	INT	2025-01-25	2025-12-27
NCT04138628	TOMBOLA	RECR	Bladder Ca|Bladder Ca, Met+	282	INT	2020-03-24	2029-11-01
NCT04965454	ExTRACT-HCC	RECR	HCC	80	INT	2022-03-28	2027-12-31
NCT06875609	NA	RECR	S-SqCC	60	OBS	2025-03-24	2029-04-30
NCT06717243	STRATUS	RECR	L-NSCLC, L-LCNE, TF	111	OBS	2025-02-20	2028-12-01
NCT06744075	CELERITY	NYR	BCL|Hodgkin Lymphoma	108	INT	2024-12-20	2027-12-20
NCT05088395	ALCINA4	RECR	Cancer	2050	INT	2022-05-19	2031-06-01
NCT04445532	NA	RECR	Liver Neoplasms & Precancerosis	450	OBS	2020-10-11	2025-06-01
NCT06490536	NA	RECR	Colon Cancer Stage II/III	700	INT	2024-10-22	2028-09-01

Notes: -#, number of subjects; NYR, not yet recruiting; RECR, recruiting; INT, interventional (red highlight); OBS, observational (green highlight); HCC, hepatocellular carcinoma; LB, liquid biopsy; LNEC, large cell neuroendocrine carcinoma; NSCLC, Non-Small Cell Lung Cancer; TF, therapy failure; R0 TNM staging; Met+, positive for metastasis; ES-SCLC, extensive-stage small cell lung cancer; GBM, glioblastoma multiforme; CLL, chronic lymphocytic leukemia; BCL, B-cell lymphoma; GI Cancer, gastrointestinal cancer; (See text for details).

NCT05810402: This prospective clinical trial aims to develop a predictive biomarker in patients with advanced hepatocellular carcinoma (HCC) (stages B and C) by a combinatorial LB technique. The primary inquiry it seeks to address is: * Can the multi-omic LB method identify a robust predictive biomarker for the efficacy of immunotherapy? Is there an association between tissue biopsy (PD-L1 expression levels) and LB (detection of CTCs expressing PD-L1) in patients with HCC? Blood samples from participants will be obtained at many time intervals.

NCT06849518: This is a prospective pilot study designed to evaluate the dynamic alterations of plasma cfRNA PD-L1 expression in lung cancer patients receiving immune checkpoint inhibitor (ICI) therapy. Outcomes will be associated with radiographic evaluations of immunotherapy treatment efficacy and plasma NGS ctDNA analysis.

NCT04111029: ctDNA containing tumor-specific sequence changes has been identified in the cell-free component of blood. Acquisition of HCC specimens is challenging, necessitating noninvasive techniques to evaluate disease progression and delineate underlying genetic characteristics. The utilization of LB through the evaluation of circulating cell-free DNA allows clinicians to provide targeted immunotherapy or signaling system inhibitors. It additionally provides a model to demonstrate responsiveness to locoregional or immunotherapy and to noninvasively forecast tumor recurrence.

NCT05254132: SALMON is a prospective, multi-center, multi-country study focused on biomarker validation. It integrates a comprehensive non-interventional biomarker discovery study involving diagnostic imaging and tissue biopsies of non-small cell lung cancer (NSCLC) known as rATLAS with a smaller minimally interventional biomarker study (aRECIST) that monitors patients with metastases through liquid biopsy and imaging follow-up over a two-year period. A total of 1120 patients will be screened to enroll 1000 participants in rATLAS, while a subset of 250 participants will be screened to recruit 150 participants for aRECIST. The study will conclude after a single visit for participants in rATLAS, whereas participants in aRECIST will undergo a follow-up period of two years. Participants will not receive any treatment specific to this study; however, they may receive standard of care therapy or investigational products as part of another clinical study after the baseline visit. The goals of enhancing AI-based tools for assessing EGFR status (rATLAS) and automated Response Evaluation Criteria in Solid Tumors 1.1 (RECIST 1.1) (aRECIST) will be accomplished through a trial design that integrates a biomarker discovery study (cross-sectional for rATLAS) with a reader study design (follow-up study in aRECIST). In the aRECIST cohort, medical treatments are not governed by the study protocol; instead, they are determined by clinicians in accordance with standard clinical practice.

NCT05704530: The objective of this study is to evaluate the efficacy of liquid biopsies, specifically the assessment of minimal residual disease (MRD) through the measurement of ctDNA at the time of diagnosis and throughout the multimodal and multidisciplinary curative-intent treatment of resectable esophageal cancer.

NCT06710093: The introduction of ICIs has significantly enhanced survival rates in patients with advanced melanoma over the past decade. Patients who demonstrate long-term responses after two years of immunotherapy treatment enter a surveillance phase involving regular radiological imaging every three to six months for a duration of five to ten years. This exposes patients with a low risk of recurrence to substantial ionizing radiation and increases the burden and costs on already strained radiology departments. This study evaluates the feasibility and patient experience of utilizing ctDNA through minimally invasive LB assays as a biomarker for identifying disease relapse or progression at the stage of radiological progression. The data obtained from this pilot study will inform the design of a subsequent validation study aimed at establishing the optimal LB for surveillance in patients with advanced melanoma.

NCT04790682: Patient diagnosed with histologically confirmed non-small cell lung cancer (NSCLC) at a metastatic stage, treatment-naive, and suitable for first-line therapy with an ICI. The combination with chemotherapy is possible. The identification of a mutation following NGS analysis is essential for ctDNA monitoring.

NCT02511288: This project aims to characterize the genetic profile of patients with advanced stage IIIB/IV NSCLC through the use of LB.

NCT05853887: This study extends the use of an electronic health record (EHR) “nudge” designed to encourage physicians to order molecular testing at the initial diagnosis for patients with certain types of advanced lung cancer. The primary objective is to obtain these test results before initiating treatment, enabling physicians to make informed treatment decisions. The second objective is to enhance understanding of the factors that influence the success of EHR-nudge implementation.

NCT05099068: This proposal outlines a prospective, multi-cohort study designed to analyze the molecular profiles and biological characteristics of advanced cancer patients throughout their disease progression, utilizing longitudinal and sequential analyses of tumor and liquid biopsies. This approach aims to develop a model for predicting tumor response and resistance under real-life conditions, enhancing the understanding of adaptive mechanisms. Additionally, it seeks to propose therapeutic options for enrolled patients based on the biological and molecular data generated during this study and reviewed in a Molecular Tumor Board in cases of disease progression. This study will encompass 12 cohorts based on tumor type and the standard treatment administered. The patient will be enrolled prior to the commencement of standard anti-cancer treatment.

NCT06869239: This study investigates the efficacy and safety of maintenance immunotherapy (PD-L1 inhibitor) following high-dose hyperfractionated simultaneous integrated boost radiotherapy combined with concurrent chemotherapy in patients diagnosed with limited-stage small cell lung cancer (LS-SCLC). This research constitutes a prospective observational study. LB technology will be utilized to identify biomarkers that predict the efficacy of PD-L1 inhibitors following chemoradiotherapy in LS-SCLC.

NCT04966663: This study examines the potential of ctDNA in blood as a predictor for the efficacy of adjuvant treatment post-surgery in reducing the recurrence risk of lung cancer.

NCT05059444: The objective of ORACLE is to illustrate the efficacy of a new ctDNA assay created by Guardant Health, an American biotechnology company, in identifying recurrence in patients who have undergone treatment for early-stage solid tumors. Linking ctDNA test results to clinical outcomes is essential to establish clinical validity for recurrence detection and to assess its value in a cost-constrained healthcare environment.

NCT05427227: Dynamic multiomics detection of plasma-derived exosomes to investigate the efficacy and mechanisms of anti-HER2 immunotherapy and anti-CLDN18.2 in gastrointestinal cancer.

NCT05889247: This research is a prospective randomized interventional study involving patients with advanced non-small cell lung cancer undergoing immunotherapy, aimed at optimizing treatment monitoring. This study investigates the clinical utility of liquid biopsy monitoring to decrease the incidence of ineffective treatments and unnecessary toxicity, while also examining the cost-effectiveness and cost-utility of implementing liquid biopsy monitoring in routine clinical practice.

NCT04243720: This prospective trial will cover immunotherapy-treated patients. This study examines whether immunotherapy-nonresponders and immunotherapy-responders have different genomic, transcriptomic, epigenetic, and immunophenotyping profiles. A fresh tumor biopsy will be done once. Serial blood samples (maximum 50 mL or 3 mL/kg in 8 weeks), archival tissue, and one stool sample will be taken.

NCT06611072: This multi-center explorative cross-sectional study aims to describe and phenotype the immunological status of ovarian cancer (OC) patients compared to controls without cancer, concentrating on hematopoietic organs and immune cells. OC monocytes and myeloid progenitor cells will be examined for transcriptional, epigenetic, and functional programming. Due to faulty trained immune responses, myeloid cells and their progenitors may play a major role in OC development. OC sufferers may have suppressive trained immunity. Researchers will compare (1) primary debulking surgery patients and (2) interval debulking surgery patients to (3) controls as blood and bone marrow donors to determine if circulating and intra-abdominal monocytes and bone marrow and spleen myeloid progenitor cells have different transcriptional, epigenetic, and functional signatures. Participants will have a venipuncture, bone marrow aspiration, tumor biopsy, and spleen biopsy. All patients’ peritoneal fluid will be analyzed.

NCT04138628: This trial is approved to study ICIs for metastatic bladder cancer as first- and second-line treatment. Here, the term “metastatic” refers to metastases detectable on standard CT scans, which require a particular tumor load. This initiative aims to find novel immunotherapy indications for metastatic bladder cancer patients. Patients with early metastatic disease will be identified using sensitive molecular methods for tumor DNA detection in the blood. A comprehensive biomarker study will also reveal therapy response predictions.

NCT04965454: This prospective clinical trial will assess PET/CT and genomic liquid biopsy biomarkers as predictors of ICI medication response in inoperable HCC patients. This diagnostic trial examines whether pretreatment fluorine-18 (18 F-) fluorocholine (FCH) PET/CT can predict “LOR” following 16 weeks of ICI therapy. LOR refers to an imaginary line connecting two detectors that registered a couple of annihilation photons at the same time.

NCT06875609: This study aims to evaluate the efficacy of a liquid biopsy assay in detecting residual disease post-surgery in patients with cutaneous squamous cell carcinoma and its capability to follow responses to immunotherapy treatment.

NCT06717243: This observational study investigates how genomic and epigenetic variables affect chemo-immunotherapy resistance in persons with extensive-stage small cell lung cancer (ES-SCLC) or metastatic large cell neuroendocrine carcinoma (LCNEC). ES-SCLC and LCNEC are aggressive lung cancers with poor prognoses and few therapy choices. Chemo-immunotherapy often works initially, but most patients acquire resistance within months, causing disease progression and poor prognosis. This study examines molecular and cellular alterations that cause resistance, which could lead to more tailored and successful treatment strategies. To investigate resistance mechanisms, the study analyzes genomic and methylation markers, CTCs, and ctDNA. Researchers gather and analyze biomarkers over time to identify patterns that distinguish long-term therapeutic beneficiaries from those who develop early resistance. These discoveries may enable novel diagnostics and treatments to predict and overcome chemo-immunotherapy resistance. This study asks: Do genomic or methylation patterns predict chemo-immunotherapy resistance in ES-SCLC and LCNEC? How do CTCs and ctDNA affect disease development, therapy response, and survival? What biological differences exist between long-term responders and early-stage resistant patients? An aim is to obtain baseline molecular profiles, sample blood and tumor tissue before therapy. Moreover, the investigators will track illness development and treatment response by visiting every 9 weeks for blood and imaging testing. They will provide re-biopsy tissue samples if the disease worsens to compare tumor biology over time. De-identified blood and tissue samples will be stored securely for genomic and epigenetic analysis. Tumor tissue samples will undergo genomic and methylation analysis, while blood samples will be tested for tumor cells and DNA. Researchers will employ modern molecular and bioinformatics methods to identify resistance patterns to better treatment strategies and produce more precise medicines. The study will assess patient data over three years, including treatment and illness progression. The study examines longitudinal samples to determine how tumor microenvironment dynamics affect treatment outcomes. This discovery is crucial since ES-SCLC and LCNEC treatment options are limited and there are no ways to predict chemo-immunotherapy response. Biomarkers of resistance could help oncologists adjust treatments to patients’ genetic profiles and increase survival. This work may help discover new biomarkers for resistance, improve early treatment failure detection, and develop novel therapeutics for resistant cancer cells. The STRATUS experiment could promote customized oncology by filling a major gap in resistance mechanism understanding.

NCT06744075: This study compares “LB” (monitoring of circulating tumor DNA, ctDNA) to PET-CT imaging at the end of first-line treatment (assessment of therapeutic response using the Deauville score according to the Lugano 2014 criteria) to determine if “LB” can predict patient outcomes at 1 and 2 years after. The study will show that ctDNA clearance can predict 1- and 2-year lymphoma progression or death in first-line B-cell or Hodgkin lymphoma patients. This may help establish minimal residual disease clearance as a key endpoint for evaluating therapeutic strategies in future clinical trials and guiding patient management (e.g., de-escalation or intensification, immunotherapy redirection, and theranostic approaches).

NCT05088395: It is a prospective multi-cohort exploratory study. Patient management and therapies remain unchanged during the ALCINA 4 study. Each cohort will have a sampling schedule. Blood sample time varies amongst cohorts depending on the clinical setting and biomarkers investigated. Patients can have four samples obtained for 12–18 months. A tumor sample will be taken once during the study if needed.

NCT04445532: As a control group, chronic hepatitis, cirrhosis, and healthy control participants must show tumor evolution and discover specific diagnostic markers. Hepatobiliary tumors have a poor prognosis and high individual heterogeneity, so finding important prognostic markers and screening out specific subgroups is crucial. Targeted therapy and immunotherapy expand cancer treatment, but only a portion of patients respond and benefit. These medications can produce treatment-related side effects and are pricey. Thus, accurate biomarker identification is necessary to screen individuals for improved response to these treatments and avoid wasting resources. Multi-omics research helps better understand hepatobiliary cancers and identify treatment targets. Thus, this investigation will comprise 450 hepatobiliary carcinoma patients over 18.

NCT06490536: Colon cancer is a primary cause of cancer mortality, and undetectable micro-metastases in stage II high-risk and stage III patients increase recurrence. Post-surgery LB detects leftover cancer DNA and tracks treatment response. Primary Objective is to show that tumor genetics and LB therapy improve outcomes and quality of life for high-risk stage II and III colon cancer patients relative to conventional therapy. Secondary Goals include 1) Recurrence time comparison. Assess quality of life and side effects. 2) Compare costs. 3) Verify LB precision. LB findings determine patient randomization into standard or tailored treatment groups. Treatments will include standard CAPOX (capecitabine and oxaliplatin) chemotherapy and FOLFOX (folinic acid, fluorouracil, and oxaliplatin). Immunotherapy includes nivolumab/ipilimumab. Moreover, trastuzumab/pertuzumab targeted treatment and FOLFOX with panitumumab will also be used. The population will include 700 operable stage III and high-risk stage II colon cancer patients. Exclusion Criteria: Previous malignancies within five years. Recent polyp removal or incomplete colonoscopy. Current experimental investigation or metastatic illness. Major cardiovascular disorders, intestinal obstruction, autoimmune diseases, neuropathy, HIV, active tuberculosis, or hepatitis B/C. Medical conditions preventing treatment. Endpoints: Primary: Assess two-year illness recurrence. Secondary: Evaluate disease recurrence and survival at 3 and 5 years. Assess therapy safety and tolerability. Verify LB precision. Assess life quality via questionnaires. The 5-year study will take place at 25–30 Italian, Spanish, and German institutions.

12. Limits of liquid biopsies

Contemporary advancements in noninvasive diagnostic and monitoring technologies, such as Raman spectroscopy, have garnered significant interest in cancer therapy [177]. Nevertheless, LBs are not conventionally regarded as a definitive method for verifying and diagnosing certain illnesses, such as cancer, and are mostly employed as a supplementary test to tissue biopsy, but we are going to eliminate a tissue biopsy for cancer probably by the end of this decade. An inherent constraint of LB is its limited sensitivity and precision in distinguishing different tumor types as compared to tissue biopsy. Furthermore, the extent to which LB offers a comprehensive representation of all genetic clones inside a single tumor or a particular sub-region remains uncertain. Several aspects of clonal evolution in tumors, including the emergence of drug resistance, alterations in the genome, gene expression, and epigenetics, have been effectively tackled by techniques such as single-cell analysis of cancer cells [178–180]. Furthermore, the abundance of CTCs, ctDNA, RNA, progenitor and mature endothelial cells, and TEPs is relatively low compared to other hematological components in the bloodstream. This scarcity poses a challenge in detecting latent benign tumors [181,182].

Therefore, it is necessary to develop a cost-effective pre-profiling approach to pre-select patients based on the low occurrence of target mutations (in the case of CTCs and cDNA) in a group of patients [183]. Moreover, several biomarkers detected using LB technology are considered “fragile” and pose challenges in their detection. Furthermore, the isolation of plasma necessitates the use of exact and sensitive techniques. However, there is a shortage of defined procedures or protocols for both isolation and interpretation of plasma levels [184]. A further limit noted with LBs is the presence of both false-positive and false-negative outcomes, which can impede the accurate assessment of the effectiveness of pharmacological therapy [185]. Furthermore, the liberation of biological substances (such as urine and blood) utilized for laboratory tests can be affected by microenvironmental variables [185].

The main objectives for detecting LB are CTCs, cancer-associated DNA, and EVs. Among them, EVs face difficulties transitioning from laboratory settings to clinical applications due to the absence of efficient enrichment technologies and accurate analytical procedures [41,186]. Higher morphological variability and counts of cancer-associated T-cells in various malignancies and patients present a significant obstacle to linear regression models. Typically, enhanced cancer stem cells are detected by analyzing biologic markers linked to the tumor, either at the protein or mRNA level. A challenge in identifying epithelial markers in patients with epithelial malignancies arises from their downregulation during the epithelial-to-mesenchymal transition, which might result in false-negative findings [187]. Furthermore, for the subsequent examination of individual or clusters of CTCs at the DNA, RNA, or protein level, micromanipulation or DEP (dielectrophoresis)-array technologies often necessitate a substantial initial concentration of CTCs [188]. While whole-genome amplification (WGA) can produce enough DNA for sequencing analysis, it can introduce bias by distorting the original template during amplification. Therefore, it is necessary to develop new WGA-free methods to enhance the reliability of these assays [189]. Establishing permanent cancer stem cell lines or creating xenograft models can provide a more profound understanding of the interactions between CTCs and the blood microenvironment. However, the limited presence of CTCs in the blood and the diversity of tumors make establishing cell lines and xenograft models difficult [190,191]. Furthermore, the construction of these models necessitates a substantial amount of time, has lower rates of effectiveness, and lacks cost-effectiveness, rendering them inappropriate for clinical research.

One further constraint is the limited selectivity of cDNAs caused by the existence of cfDNAs from healthy tissues, which might result in inaccurate positive or negative determinations [41,186]. Additionally, in order to prevent the accumulation of non-tumoral cfDNA, other pre-analytical procedures are necessary for the analysis. This is because the cfDNA released by normal cells serves as a diluent for the limited portion of ctDNA and can result in an imprecise sample of cfDNA [192]. Implementing optimized pre-analytical procedures, such as double plasma centrifugation, appropriate incubation duration, ambient temperatures, and specialized blood collection tubes for cfDNA, can minimize the interference from wild-type DNA and yield dependable testing outcomes [193,194]. Furthermore, identifying non-coding RNA signatures in the blood is more difficult than other widely used biomarkers because of their limited presence in bodily fluids, absence of appropriate reference analytes for non-coding RNA, and significant variation observed among patients. These constraints result in a lack of uniformity across biomarkers discovered in various scientific investigations [193,195]. To prevent excessive interpretation of the diagnostic data, laboratories conducting LB assays may also require an initial histological assessment of the tissue sample. Undoubtedly, LB is a very effective noninvasive diagnostic technique that can furnish detailed tumor-molecular profiling and immediate information on therapeutic cancer targets. However, it is necessary to create innovative methods and standardized approaches to address the constraints that impede the integration of LB into translational and clinical practice.

13. Conclusive remarks and future prospects

Existing compelling literature confirms that LB is a reliable and minimally invasive diagnostic method for early diagnosis and monitoring of treatment response, cancer screening in high-risk populations, evaluation of tumor heterogeneity, and identification of new cancer-driven mutations. Like tissue biopsy, LB offers molecular profiling of tumors, enabling early identification of tumor size ahead of traditional diagnostic procedures. In addition, LB has significant potential to track the heterogeneity inside tumors, identify the clonal origin of driving events and evolutionary processes in many early-stage malignancies, and function as a prediction indicator for hidden metastases. Moreover, the molecular profiling of tumors revealed by LB can monitor disease progression and prevent disease recurrences. Although LB has numerous advantages, its clinical use is hindered by several constraints, including insufficient specificity and sensitivity, the absence of standardized and isolated mechanisms, and increased economic expenses. Present technology affords just a cursory understanding of tumor activity and gene expression. A necessity exists to enhance the technology that enables the identification of cancer in several organs and the comprehensive study of tumors. Therefore, by tackling the difficulties related to the use of LB through advancements in research and technology, it is possible to achieve its very effective incorporation in clinical environments, bringing about a significant transformation in cancer research.

Nowadays we are all surrounded by and widely interacting in our daily lives with artificial intelligence (AI) in the form of consumer technology, whether we think about smartphones, wearable devices, search engines, social media channels, personalized advertising, facial recognition, autonomous vehicles, energy-efficient buildings, smart toys, and many more. The concept of AI is not new [41], but only in the past few years has it raised interest due to its potential as a virtual assistant able to serve various domains of human activity, especially the biomedical field. These AI experiences aim to improve human lives by increasing efficiency and by tailoring solutions for each individual. From this point of view, AI may be defined as a combination of theories, algorithms, and computing frameworks enabling a machine to independently reproduce intellectual processes associated with human cognition to decide on an action in response to its perceived environment to achieve a predetermined goal. Consequently, AI empowers humans to become faster in analyzing large amounts of data and smarter in decision-making by augmenting, however, without replacing, human intelligence and intuition [112]. At its core, AI is a branch of computer science that has gradually changed not only our daily lives but also the landscape of healthcare and biomedical research for medical image analysis, intraoperative imaging, and genomics.

Therefore, this first generation of AI systems has been improved by the integration of machine learning (ML) methods, which apply complex interactions to identify data patterns. ML is a subfield of AI that uses both mathematical and statistical approaches to improve the performance of computers. More specifically, ML resides in the development and implementation of dynamic algorithms that are able to take particular actions in response to particular inputs (environment) and analyze the data to determine the actions [119,120]. These algorithms are able to self-improve (learn) as more data is available, and the optimization process is called “training.” Depending on the tasks to be solved, basic ML algorithms can be divided into supervised, unsupervised [121], and reinforcement learning.

Deep learning (DL) is a subfield of ML that involves training an artificial neural network (ANN) with many layers. DL recapitulates the biological neural network of the human brain and uses a layered structure of algorithms to analyze data, identify patterns, draw conclusions, and make decisions. The basic architecture of deep neural networks (DNNs) consists of an input and an output layer together with a variable number of hidden layers in between. In such a network, the input is provided to the input layer, which transfers its computed value toward the hidden layers that are finally linked to an output layer. DL can be further classified into the deep neural network (DNN), recurrent neural network (RNN), and convolutional neural network (CNN). CNNs are particularly important to identify patterns from unprocessed images [130]. CNNs apply nonlinear transformations to structured data, for example, the raw pixels within an image, to automatically learn relevant features. In this approach, it is mandatory to process accurately the images before analysis in order to reduce the risk of the model learning from artifacts. CNNs basically use two main models: the first one uses images from a large collection, such as ImageNet, to train the initial layers, and the second one is based on an autoencoder where the model learns background features from a subset of representative images [131]. DL has been widely applied to empower medical procedures mainly involving analysis of images resulting from a wide range of procedures [132–135] and genomics [136].

It is foreseeable that AI will gain momentum in liqui biopsy in the next few years, probably re-classifying some neoplastic diseases and terminology [137–139].

Supplementary material.

Technological advancements in the separation of CTCs from liquid biopsies

Various technologies have been employed to identify live CTCs to gather tumor information [15,71,196–202]. An example of such an assay is the EPISPOT (EPithelialImmunoSPOT) test, which can identify circulating tumor cells as small as a single cell [203]. Extensive research has shown its effectiveness in a diverse range of patients with various types of tumors, including breast cancer, colorectal cancer, prostate cancer, and melanomas [204]. The assay entails the utilization of membrane-bound antibodies targeting the epithelial cell adhesion molecule (EpCAM, or CD326) found on tumor cells, followed by their cultivation and growth in both in vivo and stationary settings. This technique is prognostically significant in analyzing the protein secretome of live CTCs from breast cancer in living organisms.

The CellSearch^® system is a comparable positive selection cum enrichment technological approach for CTCs derived from lung cancer samples [58,90,205–208]. The method employs antibody-labeled magnetic beads to capture CTCs containing epithelial lineage markers such as EpCAM. The CellSearch^® technology has been instrumental in establishing a correlation between the number of CTCs and the prediction of prostate cancer patient survival [58]. There are drawbacks to this assay because not all cancer cells, including stem cells, might express EpCAM markers. Once formalin or alcohol-fixed, cancer stem cells are not viable for subsequent culture and functional tests in living organisms.

There are numerous advances in next generation sequencing (NGS) including, for example, The MiSeq^™ i100 Series sequencer, the most recent benchtop sequencer developed by Illumina [209]. It provides simplicity, quick analysis times, access to extensive panels analysis, and notable sustainability improvements. MiSeq^™ i100 Series has groundbreaking advancements has innovative uses in oncology and pathology, with numerous oncoReveal^™ panels, created at Pillar Bioscience [210–212]. Similar products have been commercialized by other companies, including Thermo, BGI, Signal Genomics, Element, Oxford Nanopore, Roche and others. Digital PCR and qPCR also provide rapid assessment for presence of circulating markers [111,213–218]. There are several commercial assays, including Guardant, Natera Signatera, Liquid Trace, Haystack, among others.

A further immunomagnetic-based enrichment technique for CTCs from lymph nodes is the AdnaTest [190,219–221]. Besides the EpCAM-labeled ferromagnetic beads employed in the CellSearch system, AdnaTest incorporates a polymerase chain reaction (PCR) process to identify mRNA transcripts specific to tumors [221]. For instance, when it comes to CTCs derived from prostate cancer, the test involves a PCR stage that utilizes primers targeting prostate-specific markers such as Prostate-Specific Antigen (PSA) and Prostate-Specific Membrane Antigen (PSMA). The assay was expanded to identify tumor-specific splicing variations of transcripts in CTCs that were enriched from liver biopsy samples. Specifically, the AdnaTest has been shown to identify the androgen receptor splice variant-7 (AR-V7) transcript that is overproduced in prostate cancer. This transcript does not contain the ligand domain ordinarily present as a transcription factor. Thus, the ligand-independent AR-V7 variation leads to a subsequent increase in the expression of genes regulated by AR [222]. Recent investigations have established a correlation between the identification of AR-V7+ variant CTCs and higher aggressiveness, worse prognosis, and resistance to numerous chemotherapy treatments (enzalutamide and abiraterone) in malignancies [223]. The clinical findings from these investigations, which indicated that non-AR-directed medicines would be more effective in treating AR-V7 + patients, were validated by further studies conducted by Onstenk et al., which showed the effective use of cabazitaxel in metastatic prostate cancer [224]. Moreover, the CellSearch System has successfully isolated CTCs targeting markers, including estrogen receptor, B-cell lymphoma 2 (BCL-2), Human Epidermal Growth Factor Receptor (EGFR) 2, and Ki67. These CTCs play a vital role in creating a new CTC-Endocrine Therapy Index, a parameter used to predict endocrine therapy response in breast cancer patients [225].

Microfluidic devices have been employed as an alternate method to antibody-labeled beads to positively select cancer-associated T lymphocytes in different malignancies [226,227]. The “CTC-Chip,” a device consisting of several small antibody-labeled micro-posts, has been employed to harvest CTCs that carry particular tumor antigens from the LB blood sample [228]. Recent iterations of “CTC-Chips” have shown an improved selection of microgroove patterns, which appear to enhance the duration of interaction between antibody-labeled micro-posts and CTCs, enhancing cellular entrapment. Microscopic CTCs separated from lung biopsy samples by the chip are then seen and examined [229,230].

Furthermore, ex-vivo bio-functional tests such as the Metastasis-Initiating-Cells (MIC) assay assess the invasive characteristics of CTCs derived from liver dissociated tissue biopsy into the surrounding tissue in living organisms, aiding in their subsequent characterization and L-MISC nanosensor is the ultimate technology [81,231,232]. Consequently, these studies help provide a comprehensive understanding of tumor stages and subtypes and develop innovative personalized therapeutic medications for the neoplastic disease [233]. Furthermore, other from ubiquitous nuclear and surface-specific markers that specifically target CTCs, it is also possible to use counterstain markers that specifically target cells other than CTCs, such as white blood cells (WBCs, leukocytes), platelets, red blood cells (RBCs, erythrocytes), etc., to selectively concentrate CTCs from blood samples. The chosen markers for counterstaining are CD45/CD66b for granulocytes, CD235a for red blood cells, CD41/CD61 for platelets, CD4/CD8 for lymphocytes, CD11b/CD14 for macrophages, and CD34 for hematopoietic progenitors/endothelial cells [234–237]. Technological solutions such as the EasySep Depletion Kit (StemCell Technologies) employ CD45-labeled magnetic beads to selectively remove leukocytes from the biopsy samples via negative selection [15,238]. Previous researchers have also devised comparable negative depletion technologies [239,240]. Alternative approaches, such as RosetteSep (StemCell Technologies), incorporate an extra density gradient centrifugation stage to enhance the enrichment of CTC [241]. Constraints of negative selection techniques include the possibility of other blood components, such as CD45 − ve endothelial cells, crossing and an increased chance of CTC loss in bulk leukocytic pulldowns [242].

While differentially expressed on cancer cells, tumor-associated markers are susceptible to being lost over time in CTCs due to molecular alterations and dedifferentiation, even in highly aggressive tumors [243]. In addition to differential expression of antigen markers, numerous other methods have been documented that have facilitated effective detection and isolation of CTCs, including variations in their physical characteristics compared to WBCs. Size exclusion-based separation techniques have been demonstrated to distinguish CTCs (mean diameter: 15.6 μm) from WBCs (diameter range: 7–15 μm), for which the former was comparatively larger [244]. In numerous instances, an inherent constraint was that CTCs are almost identical in size to WBCs, leading to a reduction of up to 50% in CTCs in methods that depend just on size exclusion [245]. Furthermore, it has been shown that smaller CTCs are associated with higher levels of metastasis in prostate cancer [246]. These constraints have been addressed by employing selective size amplification methods of CTCs that artificially enlarge their size using microbeads labeled with anti-EpCAM antibodies, enhancing cell retrieval and purity [247]. Divergences in deformability between CTCs and healthy blood cells have been utilized to separate CTCs by enabling their transit via microfluidic channels. These investigations have shown that the disparities in deformability between WBCs and cancerous melanoma cells are far more significant (CTCs being more deformable than WBCs) than between CTCs, facilitating effective identification and separation of CTCs [248]. Recent advancements have introduced newer techniques, such as Celsee^® systems, which utilize size disparities and deformability to separate and evaluate CTCs. The systems include microfluidic devices with fluidic channels and capture chambers designed to ensnare relatively larger tumor cells, allowing normal cells like white blood cells to flow through [249]. Moreover, being able to capture unlabeled cells, the technology can also be employed for subsequent examination of CTCs using diverse techniques like immunocytochemistry or in situ hybridization methods, like “FISH.” Indeed, these microfluidic-based devices (Celsee^® systems) have demonstrated superior sensitivity in detecting CTCs, as indicated by their higher CTC counts compared to cell surface marker-based technologies (CellSearch^® system).

Nevertheless, these systems still tend to lose very tiny CTCs. Antibody-independent techniques to separate and recover CTCs from blood include the ApoStream^™ device [250]. This device evaluates the variations in surface charge and polarizability between CTCs and non-tumoral blood cells. Subsequent investigations have once again shown that these dielectrophoresis-based systems lead to improved identification and retrieval of CTCs in prostate cancer compared to the surface marker-based systems (CellSearch^® system) [250].

Furthermore, post-enrichment techniques such as the DEPArray^™ System have been proven to effectively separate and retrieve individual cancer stem cells (such as Melan A positive melanoma cells) from low-flow samples of whole blood [251]. Furthermore, the Ion Torrent PGM^™ system, which consists of the Ion AmpliSeq^™ Cancer Hotspot Panel, enables NGS analysis directly on CTCs [21,252]. This system offers improved mutational analysis and eliminates the need for error-prone techniques such as whole genome amplification (WGA), enhancing screening accuracy [251].

Contemporary methodologies for the analysis of ctDNA from liquid biopsies

The comparison of data from several sources remains a significant challenge in clinical ctDNA analysis due to the absence of fully standardized procedures for sample handling, ctDNA isolation, and analysis and the lack of a comprehensive analytical consensus. Variations in the amount and quality of ctDNA should be attributed to biological changes associated with the tumor rather than artifacts caused by differences in sample handling. Therefore, studies have concentrated on examining the impact of several “preanalytical” variables such as clotting, DNA leakage from leukocytes, freeze-thawing, DNAse activity (blood), PCR reagent compatibility, time-lapse between blood drawing and analysis, and temperature on ctDNA analysis. Restrictions related to separating and analyzing tiny quantities of cDNA have been significantly lowered by continuously advancing technical applications [116].

Two primary methodologies have been explored for ctDNA analysis: targeted methods that concentrate on specific gene rearrangements or gene mutations in particular genomic regions that serve as loci of variation in a particular tumor type, or untargeted methods that provide a more comprehensive analysis and monitoring of the tumor genome, including information on nucleotide changes, copy number aberrations, chromosomal changes, etc., without relying on any previous data on molecular changes. Amplification-based techniques, such as droplet digital PCR and BEAMing, have demonstrated exceptional sensitivity ranging from 1% to 0.001% in identifying somatic point mutations [193,253–262]. During droplet digital PCR, the sample DNA (target and background DNA) is divided into several separate divisions or droplets. The target sequence is amplified using end-point PCR in each droplet, and the relative fractions of positive and negative droplets are tallied using fluorescent probes to produce a relative quantification of the target samples. The detection of ctDNA by digital PCR has been demonstrated in over 75% of patients with advanced CRC, breast cancer, and prostate cancer, and to a considerable degree in patients with confined tumors [263].

BEAMing, short for beads, emulsions, amplification, and magnetics, is a variant of emulsion PCR in which certain templates are amplified in separate compartments (or emulsion droplets) within a single tube [193,258–262]. These templates are affixed to primer-bound beads and then recovered using a magnetic field or centrifugal force. PCR-based tests that specifically identify genomic rearrangements linked to the tumor genome have demonstrated encouraging outcomes in terms of sensitivity and specificity when employing cDNA. The assays personalized analysis of rearranged ends (PARE), which employ primers surrounding the breakpoint area, have demonstrated effective detection of mutant ctDNA (rearranged sequences) at concentrations as low as 0.001% in plasma samples of patients with CRC and breast cancer [264,265]. PARE analysis of ctDNA helps monitor the disease’s magnitude and identify biomarkers specific to solid tumors in patients [265]. Recent developments in NGS have introduced techniques that provide a more extensive screening of genomic regions and improved resolutions in identifying mutations in cDNA samples. The tagged-amplicon deep sequencing (TAm-Seq) assay, created by Forshew et al., can identify ctDNA mutations in plasma with extremely low allelic frequencies (around 2%) and with a high level of sensitivity (about 97%) [261,266]. Sequence-specific primers are used to amplify several areas of the targeted location in the genome to enable the representation of different alleles in the template material. This process helps to narrow down the pool of amplified products. The various products are once again amplified for enrichment, labeled with adaptors, and then subjected to sequencing [266]. Previous research has detected genetic alterations in the tumor suppressor p53 and EGFR areas in the plasma of individuals diagnosed with ovarian cancer, which are not often identifiable by invasive solid biopsies.

Furthermore, TAm-Seq has also facilitated the long-term monitoring of tumor mutations in the plasma of patients with breast cancer over periods of several months [261]. Newman et al. have developed deep sequencing techniques, such as CAPP-Seq [14,261,267,268], which have enabled the identification of ctDNA mutant fractions as low as 0.02% with a high specificity of approximately 95% in patients diagnosed with non-small-cell lung cancer (NSCLC) [269]. Quantifying ctDNA using CAPP-Seq analysis demonstrated superior correlation with tumor burden, detection of residual disease, and early tumor response compared to conventional radiography techniques [14,261,267–269]. Targeting specific sections of DNA is achieved by using tagged complementary oligonucleotide probes that may be retrieved.

Unlike the targeted approaches mentioned before that concentrate on primer-specified genomic areas, untargeted methods are comparatively more thorough in studying the tumor genome. Shotgun massively parallel sequencing of ctDNA from plasma has demonstrated capabilities to provide comprehensive genome profiling for copy number changes (CNA) and mutations in patients diagnosed with hepatocellular carcinoma (HCC), breast cancer, and ovarian cancer [270]. Comprehensive analysis of plasma ctDNA using high-throughput Illumina MiSeq [271] has demonstrated the identification of several CNAs (androgen receptor amplification) and chromosomal rearrangements (TMPRSS2-ERG fusion; 8p loss/8q gain) in patients with both castration-resistant and castration-sensitive prostate cancer [59]. Utilizing massively parallel sequencing of plasma ctDNA, whole-genome analysis has also facilitated the identification of comparable changes in individuals with CRC and breast cancer [243].

Contemporary approaches for extracting tumor EVs from liquid biopsies

Multiple methodologies have been documented that efficiently separate EVs from different types of cellular waste, utilizing distinct physical and biochemical characteristics that aid in EV isolation. Preparative ultracentrifugation is a widely used technique for isolating EVs, making up over 50% of all the EV isolation methods already in use [272,273]. The approach capitalizes on variations in particles’ density, shape, and size. Various investigations have demonstrated variations such as differential, isopycnic, and moving zone ultracentrifugation that effectively reduce EV loss and contamination, thereby enhancing EV purity [274,275]. An alternative method for isolating EVs based on size is ultrafiltration, which utilizes membrane filters with predetermined molecular weight cutoffs [276]. While ultrafiltration-based separation techniques eliminate the need for specialized equipment such as centrifuges, the vesicles are susceptible to rupture or distortion caused by the applied force, which impacts subsequent analysis [277]. Nanomembrane ultrafiltration concentrators have demonstrated significant efficacy in isolating urine EVs from samples as small as 500 µl and aiding in diagnosing renal problems [278]. The ExoMir^™ commercial kit from Bio Scientific utilizes syringe filters to collect EVs from cell-free fluids such as sera, CSF, and cell culture media [279]. Next, RNA extraction reagents are employed to break down and liberate the contents of exosomes, which are subsequently subjected to qRT-PCR examination. Alternative methods, such as sequential filtering employing electron microscopy and mass spectrometry for subsequent analysis, enable the extraction of EVs with enhanced purity and integrity [280]. Type exclusion chromatography is a size-based technique used to isolate EVs. In this procedure, EVs are separated from the other components in the sample by being excluded from the pores in the stationary phase and eluted before the other fractions. Such size-exclusion techniques have been used to isolate EVs released by mesenchymal stem cells in response to cardiac infarction [281]. Comparable size exclusion fractionation techniques have been employed to isolate tumor-derived EVs that can inhibit T-cells from the ascites of patients with ovarian cancer. These EVs are then verified for biomarkers using Western blots. Furthermore, research has demonstrated that the use of size exclusion chromatography, together with ultracentrifugation, enhances the production of urine EVs and helps identify prognostic biomarkers that can predict the outcomes of renal diseases [282]. Innovative size-based methods for separating EVs include flow field-flow fractionation, which is used to isolate EVs from brain stem cells, and hydrostatic filtration dialysis, which successfully separates and enriches urine microvesicles to create larger sample volumes. Furthermore, beyond size-based isolation techniques, immunoaffinity-based tests like ELISA have been documented to isolate and analyze EVs from various bodily fluids including plasma, urine, CSF, and others. These approaches utilize the existence of membrane-bound surface biomarkers that are either expressly present or upregulated on extracellular vesicles [272]. Furthermore, RNA obtained from EVs separated by immunoaffinity-based tests showed higher yield than traditional size-based techniques such as ultracentrifugation. These methodologies can extract exosomal RNA from as little as 100 μl of the sample [272]. Modifications of these approaches, such as the magneto-immunocapture technique utilizing antibody-coated magnetic particles, have been shown to achieve yields that are 10–15 times superior to UC [272].

CD63, a tetraspanin abundantly present on the membranes of EVs, can be used to selectively extract EVs from complex sample mixtures utilizing these methods [152–154,283–286]. Furthermore, it has been hypothesized that CD34 positive blast-derived EVs obtained by comparable magneto-immunocapture methods can serve as biomarkers and demonstrate utility in the surveillance of disease advancement and therapy efficacy in acute myeloid leukemia. Remarkably, EVs produced by these techniques were demonstrated to possess biological activity and the ability to induce immunological suppression. Several scientific investigations have also emphasized several versions of magneto-immunocapture-based methods that provide improved efficiency and sensitivity in capturing tumor extracellular vesicles [287–290]. Successful precipitation techniques utilizing polymers, such as polyethylene glycol, have been employed to isolate EVs from biological fluids [291]. Commercial isolation kits that rely on the precipitation of EVs include ExoQuick PLUS (System Biosciences). These kits can precipitate EVs from serum, plasma, and other biological fluids in a comparatively shorter time and with fewer residual contaminants. The separated EVs are subsequently examined for their protein composition using Western blots and for RNA for quantitative real-time polymerase chain reaction (qRT-PCR) [292,293]. Novel EV isolation kits, such as ThermoFisher, CUSABIO, iZON, qEVSingle, and 101Bio, have been created to accommodate various samples, including urine, saliva, CSF, ascites, and amniotic fluid. Included in the pre- and post-isolation procedures for EVs purification are measures to eliminate non-EVs impurities such as lipoproteins and polymeric substances, respectively [13,141–143,151,186,272,294–304]. Furthermore, apart from polymers, lectins that have a strong affinity for oligosaccharide residues on the membranes of EVs have also been shown to effectively separate urine EVs [305]. The miRNA profiles of EVs produced by lectin-based agglutination techniques have been proposed as important biomarkers for diagnosing prostate cancer. In addition, microfluidics-based enrichment techniques, such as those devised by Wang et al., employ ciliated micropillars that specifically capture EVs while excluding other outside vesicles, proteins, and trash [306]. Similarly, commercially accessible microfluidic hardware like ExoChip, created by Kanwar et al., effectively separate EVs from the blood of patients with prostate cancers for the purpose of miRNA study [307]. These electronics employ a CD63 antibody-coated immunochip to capture EVs, which are then stained with membrane-specific fluorescent dyes. The analysis of exosomal cargo is conducted using either Western blot for proteins or RT-PCR for RNA. Analogous immunochip-based tests utilizing antibody-labeled magnetic beads utilize plasma samples as little as 30 microliters, with an assay duration of less than 100 minutes, for the purpose of enriching and analyzing EVs [308]. Notwithstanding the swift progress achieved in the extraction and enrichment of EVs as previously explained, the technologies remain restricted in some therapeutic applications. Researchers in this field face significant hurdles related to sample pretreatment, isolation efficiency, standardization, heterogeneity of EVs, and the yield of exosomal cargo. However, the importance of EVs in therapeutic applications such as tumor diagnosis, monitoring, and treatment should not be underestimated due to its substantial value as a minimally invasive approach in these fields, as supported by several sources reported below.

Viable alternatives in liquid biopsies excluding blood

Furthermore, apart from circulatory fluids such as plasma or serum, several empirical investigations have demonstrated the substantial utility of additional bodily fluids like saliva and urine in LB. In terms of accessibility, noninvasiveness, and cost-effectiveness in sampling, saliva provides practical advantages that surpass those of plasma or serum [309]. The field of salivary molecular diagnostics has seen significant advancements in the last 10 years, showing immense promise in cancer detection, monitoring, and the implementation of point-of-care treatment [310]. Novel electrochemical sensor-based technologies, such as the electric field-induced release and measurement (EFIRM) developed by the Wong lab, have demonstrated the ability to identify EGFR mutations (specifically in the tyrosine kinase domain) in saliva samples from patients diagnosed with NSCLC [311]. Previous studies have employed comparable EFIRM based methodologies to identify salivary biomarkers such as Foxp1 and Gng2 to detect pancreatic cancer [312]. Also demonstrated to be applicable in liquid biopsies utilizing saliva are indicators that are not based on the genome. Analysis of salivary metabolites using spectroscopic methods has shown that elevated porphyrin levels are a reliable indicator of oral squamous cell carcinoma [313]. Moreover, recent research have established a correlation between alterations in the oral microbiome and the development of oral squamous cell carcinoma [314].

The entirely noninvasive characteristic of urine sample, compared to tissue or even blood, renders it a highly valuable option in liver cancers, especially when frequent sampling is necessary to track tumor advancement and treatment results [315]. Furthermore, there has been a growing focus on using urine in liver cancer research because of the abundance of clinically significant cell-free molecular components such as proteins, circulating DNA/RNA, and EVs that can be used to track tumor proliferation. Research has shown that urine liquid biomarkers are highly valuable in identifying malignancies from both urological and non-urological sources. The sensitivity of urine liquid biomarkers is generally similar to that of blood liquid biomarkers. Considering that most biomolecules released by urological malignancies are likely to be directly eliminated into the urinary tract, urine LBs provide a convenient and uninterrupted means of monitoring these tumors [316,317]. Well-established urine biomarkers for prostate and bladder cancer include Nuclear Matrix Protein 22 (NMP22), expression of the TMPRSS2:ERG fusion gene, and Prostate Cancer gene 3 (PCA3). Urinary lncRNAs such as FR0348383, UCA1, and MALAT1 have been identified as far more effective biomarkers in prostate cancer compared to prostate-specific antigens obtained from serum [318–320]. Urine-derived cell-free nucleic acids such as IQGAP3 and UBE2C have been recently identified as diagnostic indicators for bladder cancer [321]. PiRNA (piR-823), a non-coding RNA implicated in transposon silencing, has been found to be modified in serum and urine samples and has been proposed to have potential diagnostic value in patients with renal cell carcinoma [322]. Additionally, circular RNAs such as PRMT5, known to stimulate the transition from epithelial to mesenchymal cells by absorbing miRNAs, have been proposed to significantly impact the development of bladder urothelial carcinoma. Consequently, they are positively associated with the advanced stage of the disease and decreased survival in patients with urothelial carcinoma [323]. Further investigation of urine as a source of LB for urothelial cancer is now being conducted in ongoing clinical trials such as NCT04432909 (ClinicalTrials.gov). Non-urological malignancies such as lung, gastric system, colorectal, breast, etc., have also been demonstrated to be detected and closely followed using urine lymphobiosis biomarkers. Urinary loop benders, similar to plasma loop benders, have shown considerable efficacy in identifying epigenetic modifications such as DNA methylation at specific gene sites linked to NSCLC [324]. Furthermore, the sensitivity of tissue and urine LBs were calculated to be similar (around 75%) in identifying EGFR mutations in NSCLC. These and other studies have also shown that urinary lysis products are somewhat reliable predictors of the chemotherapeutic response of malignancies to certain medications such as rociletinib and osimertinib [325,326]. The identification of DNA markers linked with HCC (TP53 249 T; RASSF1A and GSTP1 methylations) utilizing urine lymphocytes has also helped in the surveillance of tumor recurrence [327].

Liquid biopsy is emerging as a pivotal tool in precision oncology, offering a noninvasive and comprehensive approach to cancer diagnostics and management. By harnessing biofluids such as blood, urine, saliva, cerebrospinal fluid, and pleural effusions, this technique profiles key biomarkers including circulating tumor DNA, circulating tumor cells, microRNAs, and extracellular vesicles. This review discusses the extended scope of liquid biopsy, highlighting its indispensable role in enhancing patient outcomes through early detection, continuous monitoring, and tailored therapy. While the advantages are notable, we also address the challenges, emphasizing the necessity for precision, cost-effectiveness, and standardized methodologies in its broader application. The future trajectory of liquid biopsy is set to expand its reach in personalized medicine, fueled by technological advancements and collaborative research.

Identification of emerging analytes for liquid biopsies

The phenomenon of tumor-mediated platelet education has been documented as tumors’ modification of platelet behavior [328]. Moreover, new biomarkers linked to these tumor-expansion polymorphisms (TEPs) have emerged as promising analytes that aid in the noninvasive diagnosis of malignancies [328,329]. RNA-Seq studies of RNA produced from platelets were shown to accurately identify both early- and late-stage NSCLC with an accuracy rate of above 80% [330]. Moreover, since the growth of tumors is linked to certain systemic shifts that result in metabolic modifications, the levels of circulating metabolites in the plasma of patients might be used as biomarkers for cancer. An association has been established between increased concentrations of metabolites, such as branched-chain amino acids, and the initial stages of human pancreatic adenocarcinoma [331]. Previous research has linked this phenomenon to decreased use of these circulating metabolites in individuals with prostate cancer [331]. However, malignancies like NSCLC actively using these metabolites are linked to reduced quantities of branched-chain amino acids. The modulation of metabolite levels due to their differential use by tumors may have a significant impact on the early identification of distinct types of malignancies particular to certain tissues [332].

Moreover, like methylation patterns (mentioned previously), the nucleosome location of cfDNA has recently been shown to vary among different cells and provide useful information about target genes. Novel genome-wide maps of nucleosome occupancy on cfDNA have been generated by deep sequencing cfDNA samples obtained from blood plasma [333]. The occupancy of nucleosomes and the DNA footprints of complementary fragment DNAs (cfDNAs) can be linked to the expression of several target genes, including cancer drivers, and provide important information about the tissue from which they originate [333,334]. Recent research has demonstrated a direct relationship between a reduction in nuclear cfDNA levels and a shift toward longer fragments, resulting in enhanced responsiveness to chemotherapy in patients with colorectal cancer [335]. In contrast, higher and shorter nuclear cfDNA content was associated with tumor recurrence [335]. Comparative functional DNA (cfDNA) analysis presents potential advancements in noninvasive LB techniques [263]. Recirculating cell-free miRNAs linked to EVs, apoptotic bodies, lipoproteins, or antigen-presenting molecules (miRISC components) are well-established to be highly stable in various bodily fluids such as plasma or serum [336]. Contemporary research has progressively concentrated on utilizing miRNA signatures from bodily fluids as diagnostic modality for cancer identification. The “miR-Test” method, created by Montani et al., examines blood miRNA signatures and has shown considerable effectiveness in detecting early liver cancers, particularly in high-risk patients, compared to conventional and costly techniques like computed tomography [337]. Relevant clinical studies have indicated the importance of plasma miRNA signatures in enhancing the diagnostic and prognostic assessment effectiveness of patients with lung cancer [338,339]. Hence, the assessment of miRNA signatures in bodily fluids is a developing field of liquid biopsies that could have significant utility in the detection of many types of malignancies. The increasing repertoire of substances that can be utilized in liquid biopsies offers new biomarkers that have significant potential in the fields of diagnosis, prognosis, and the development of treatment interventions for a diverse range of malignancies [340–351].

Funding Statement

This research has been funded by the generosity of the Children’s Hospital of Eastern Ontario, Ottawa, Ontario, and the Stollery Children’s Hospital Foundation and supporters of the Lois Hole Hospital for Women through the Women and Children’s Health Research Institute (WCHRI, Grant ID #: 2096). The author is a Grant Receiver from the Natural Science Foundation of Hubei Province for Hubei University of Technology (100-Talent Grant for Recruitment Program of Foreign Experts Total Funding: Digital PCR and NGS-based diagnosis for infection and oncology, 2017-2022), Österreichische Krebshilfe Tyrol (Krebsgesellschaft Tirol, Austrian Tyrolean Cancer Research Institute, 2007 and 2009 – “DMBTI and cholangiocellular carcinomas” and “Hsp70 and HSPBP1 in carcinomas of the pancreas”), Austrian Research Fund (Fonds zur Förderung der wissenschaftlichen Forschung, FWF, Grant ID L313-B13), Canadian Foundation for Women’s Health (“Early Fetal Heart-RES0000928”), Cancer Research Society (von Willebrand factor gene expression in cancer cells), Canadian Institutes of Health Research (Omega-3 Fatty Acids for Treatment of Intestinal Failure Associated Liver Disease: A Translational Research Study, 2011-2014, CIHR 232514), and the Saudi Cultural Bureau, Ottawa, Canada.

Authors’ contributions

FS and CMS conceptualized the study, collected the data, and interpreted the data. CMS drafted the initial manuscript and revised the manuscript. FS revised the clinical trials data and methodology. CMS was responsible for the intramural funding and revised the manuscript. SZ, AM, and BC revised the manuscript. All authors meet the ICMJE requirements for authorship, approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

Disclosure statement

AM is president of Protean BioDiagnostics Inc, Orlando, Florida, USA. All the other authors have no conflicts of interest to declare.