Liquid biopsy acquired from bodily fluids near malignant cells provides valuable information without invasive tissue removal [25,26]. Especially for tumors that are difficult to approach, even a least invasive percutaneous biopsy may be uncomfortable and increase the risk of complications. Additionally, conventional biopsies may only cover a limited portion of the tumor, which may not accurately reflect it [27]. Blood is the most suitable bodily fluid for liquid biopsy, while urine, cerebrospinal fluid, and saliva may also be beneficial depending on the primary tumor [26]. Liquid biopsies from blood can be used for multiplexed cancer profiling analyses including circulating biomarkers (such as cross-linked type 1 collagen, bone sialoprotein, TRAcP5B, osteoprotegerin) and metabolites (such as pyridinoline and deoxypyridino). Three basic categories of actionable biological components may be acquired from a blood liquid biopsy, including CTCs, cell-free ctDNA, and EVs, or exosomes [28].

CTCs are tumor cells that enter the bloodstream. Ashworth initially recognized them almost 150 years ago, and they are now widely employed in therapeutic practice [29]. Isolating CTCs has been challenging due to their low number in the general circulation. However, new microfluidic platforms like the U.S. Food and Drug Administration (FDA)-approved Cell Search platform and next-generation sequencing allow for deep phenotyping of every isolated tumor cell [30]. Recent improvements enable liquid biopsies to reveal the genomic mosaicism, mutational landscape, epigenetics, and gene and protein expression of the original tumor.

The majority of ctDNA found in the general circulation is 180–200 bp in length, indicating that it originates from apoptotic and necrotic primary tumor cells [28]. Tumor DNA may reveal mutations and copy number variation, evaluating the need for specific-target treatment. For instance, the V600E mutation in BRAF is seen in several cancers such as metastatic colorectal cancer, melanoma, and papillary thyroid carcinoma [31,32,33,34]. Targeting the V600E form of BRAF with medicines such vemurafenib, dabrafenib, and trametinib may influence treatment choices [35].

EVs are lipid bilayer particles with a diameter of 30 to 1000 nm that may be divided into three types: apoptotic bodies, large EVs (microvesicles), and tiny EVs (exosomes). The biogenesis and biological function vary [36,37]. Cell-secreted EVs can include other biological components, such as DNA, RNA, and proteins, which can be used in determining a diagnosis and prognosis [36,38]. More EVs are produced in cancer cells than normal cells, and play an important role in bone metastases, osteosarcoma, and Ewing’s sarcoma [39,40]. In addition, normal and cancer cells show increased EV production due to hypoxia, increased intracellular calcium or pH, oxidative stress, ionizing radiation, and ultrasound [41]. Further, while RNA can degrade in circulation, encapsulating the RNA in EVs makes them more stable and amenable to transcriptional analyses [42]. Using various methods to gather biological data from patients’ blood may aid in accurate diagnosis, prognosis, and monitoring drug-resistant clones for informed treatment decisions [43]. Due to cost, scalability, repeatability, and separation procedures, bone liquid biopsy is not widely used in clinical trials, despite preclinical breakthroughs. However, it has shown substantial advances and preclinical and clinical uses in bone tumors including secondary and primary malignancies.

Circulating tumor DNA analysis may reveal the mutational landscape of metastatic disease and predict recurrence or response to therapy [44]. A retrospective study of primary breast cancer patients detected metastatic disease, including bone metastases, by measuring tumor-specific chromosomal rearrangements in ctDNA using droplet-based digital PCR technologies from plasma samples, nearly 1 year before clinical recurrence detection. The amount of ctDNA was directly proportional to disease progression. This suggests that ctDNA detection may be a useful technique for early metastasis detection. Liquid biopsy may also reveal minimum residual disease (MRD), indicating treatment and prognosis. Plasma tumor-associated ctDNA detection and analysis are effective indicators for identifying and monitoring MRD in breast cancer patients at high risk of recurrence [45,46]. Levels of ctDNA at baseline are linked to increased bone metastases and poor prognosis in non-small cell lung cancer (NSCLC) patients [47]. Late-stage NSCLC patients show increased ctDNA levels in patients with bone metastases [48], while higher ctDNA levels are detected in prostate cancer patients with visceral metastases than in those with bone metastases [49]. Detecting MRD using liquid biopsy is still in its infancy, and larger-scale longitudinal investigations are needed to investigate false positives and negative cases.

Circulating tumor cells have been shown to be beneficial for the diagnosis, prognosis, and monitoring of patients with metastatic breast cancer; in these patients, CTC detection and characterization has been linked to bone and hepatic metastases [50]. Multiple metastatic locations are associated with greater CTC counts, but bone-only metastatic breast cancer patients have lower CTC counts and improved prognoses [51,52]. Patients with one or two bone metastases show significantly fewer CTCs than those with more bone metastases [52].

CTCs exhibit a subset of metastasis-initiating cells expressing CD44, CD47, and c-MET. When transferred from a patient to immunocompromised animals, these cells cause metastases in the lungs, liver, and bones [53]. The discovery and quantification of CTCs may predict lung cancer prognosis, with high CTC counts as a predictor of bone metastases in advanced lung cancer patients [54,55] and in monitoring bone metastases in castration-resistant prostate cancer patients. Other studies indicate that CTC counts > 5 per 7.5 mL of blood predict bone metastases and worse overall survival [56,57,58]. Another study recommended a liquid biopsy method that detects and quantifies both CTCs and ctDNA simultaneously. Peripheral blood samples collected before and after treatment in a homogenous cohort of HER2-negative breast cancer patients [59] were examined prospectively in the COMET trial (NCT01745757). Compared to non-metastatic patients, greater CTCs and ctDNA mutations in tumor protein 53, phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha, and estrogen receptor 1 (ESR1) genes were observed in bone, liver, and brain metastasis patients [59]. Similarly, results of other investigations have indicated that the CTCs-ctDNA signature is effective for the diagnosis and prognosis of metastatic breast cancer [60,61,62,63].

Bhadresha et al. [64] identified 15 genes consistently upregulated in bone metastasis patients. Using serum-derived EVs, five genes (HSP90AA1, osteopontin, IL-3, VEGFA, and protein tyrosine kinase 2) were upregulated in breast and lung cancer patients with bone metastasis. The results indicated that EV-derived mRNA may be used to detect early bone metastases in cases of breast or lung cancer [64]. Plasma-derived EV miRNAs (hsa-miR-574-5p, 328-3p and 423-3p) from NSCLC patients were examined retrospectively as early biomarkers of bone metastases [65]. In that study, Yang et al. observed that EV-derived miR-181a-5p- is increased in prostate cancer patients with bone metastases. Bryant et al. obtained similar results for miR-141 and miR-375 [66]. Prostate microparticle-specific EVs have been reported to be more prevalent in metastatic prostate cancer than in no-metastatic disease and outperformed the FDA-approved CellSearch system in predicting CTCs [67]. ExoDX, a recently evaluated urine exosomal gene expression platform, outperformed the gold standard in predicting high-grade prostate cancer in patients with uncertain PSA scores (ROC AUC 0.7 vs. 0.62) and identifying benign prostate hyperplasia, reducing unnecessary biopsies [68].

Few studies have examined plasma-derived ctDNAs from osteosarcoma patients [69,70]. One study examined somatic mutations with tumor burden and prognosis using targeted next-generation sequencing (NGS) to detect tumor-specific somatic alterations in plasma samples at various stages of treatment, allowing for disease burden monitoring [69]. In the research of Shulman et al. [70], NGS hybrid capture assay ctDNA levels in peripheral blood samples of newly diagnosed localized osteosarcoma and Ewing sarcoma patients were linked to tumor burden, recurrence, and poor clinical outcomes; interestingly, ctDNA analysis revealed unexpected genetic characteristics of osteosarcoma, such as chromosomal arm 8q copy number increases [70]. Genetic mutations, including STAG2 and TP53 loss-of-function mutations, translocation events and fusion genes, have been identified in Ewing sarcoma patients, allowing for ctDNA monitoring of the bone malignancy [71]. In the study of Shulman et al. [70], ctDNA detection in plasma samples was linked to a poor clinical outcome in newly diagnosed Ewing sarcoma patients and revealed genomic information like EWSR1 fusion and STAG2 loss-of-function mutations [70]. Hayashi et al. observed that plasma EWSR1-FLI1 fusion gene levels are associated to tumor burden and therapeutic response in Ewing sarcoma patients, suggesting another potential use for liquid biopsy. In addition, EWS-FLI1 levels in the blood fall following chemotherapy or surgery and subsequently increase after tumor recurrence [72].

Osteosarcoma metastasis may be predicted by CTCs [73]. Li et al. [74] found additional CTCs in peripheral blood of metastatic osteosarcoma patients compared to those with localized disease in a prospective analysis. Additionally, CTC count has been shown to be negatively linked with the patient’s response following neoadjuvant chemotherapy [74]. Preclinical studies indicate that CTC count variations after treatment or surgery can indicate tumor sensitivity and metastasis [75,76]. An increased percentage of mesenchymal CTCs in peripheral blood of osteosarcoma patients after chemotherapy treatment has been linked to lower disease-free survival. This highlights the importance of monitoring changes in CTCs to assess treatment efficacy and detect disease recurrence or metastasis. In Ewing sarcoma patients, CTC characterization using tumor-specific markers (i.e., CD99 expression) and chromosomal translocations (e.g., EWSR1-FLI1 transcript fusion gene amplification) has been described [77,78]. In those patients, CTCs detected at diagnosis correlates with worse clinical outcomes and increased recurrence disease and metastasis [78,79].

A liquid biopsy has been used to study EVs as diagnostic or prognostic serum indicators in osteosarcoma. RNA analysis of circulating EVs in metastatic osteosarcoma samples has revealed various transcriptome changes, offering a novel therapeutically useful method for tracking metastatic osteosarcoma [80]. Osteosarcoma patients’ peripheral blood contains miRNAs that are known to partly circulate within EVs and have oncogenic or antitumor suppressive functions. Several biomarkers, including miR-148a [81], miR-574-3p, miR-214, miR-335-5p, miR-491, miR-221, miR-191, and miR-421, are becoming important diagnostic and prognostic indicators, while osteosarcoma patients have lower levels of miR-124, miR-101, and miR-195 in their blood compared to those of healthy persons [82,83,84,85,86]. These data may be used to develop a prognostic approach for osteosarcoma employing a mix of miRNAs.

Recently, Ewing sarcoma has been studied for circulating miRNAs. One example of a circulating miRNA linked to Ewing sarcoma development is miR-125b, which has been shown to be lower in patients’ blood after surgery compared to that in healthy controls [87]. Decreased expression of this gene has been linked to a poor chemotherapy response in the same study [87]. Research is now focusing on Ewing sarcoma-derived EVs cargo as a predictive biomarker source, notably protein content. Ewing sarcoma-derived tiny EVs may be biomarked by CD99, HINT1, and NGFR membrane proteins according to Samuel et al. They used these EV surface proteins to immuno-enrich Ewing sarcoma-associated tiny EVs and identify EWS-FLI1 and EWS-ERG fusion transcripts in plasma from localized and metastatic patients [88].

Recent research indicates that the tumor secretome, including DNA fragments from drug-resistant cells with mutations, is abundant in plasma, making blood-based liquid biopsy crucial [89]. Recent studies on plasma samples of small patient cohorts have identified resistance mutations during treatment. While the data is clinically informative about the therapy response, it has not yet been fully validated in clinical practice. Quantification and analysis of ctDNA are effective tools for analyzing various tumor types [90]. The potential of ctDNA for monitoring treatment efficacy can be demonstrated in the finding that breast cancer patients with metastases, treated with aromatase inhibitors, and carrying ESR1 mutations in ctDNA are likely to show resistance to endocrine therapy and experience shorter progression-free survival [91].

Liquid biopsy can also identify biomarkers linked to CDK inhibitor resistance and predict metastatic disease in advanced breast cancer patients with hormone receptor-positive/HER2-. Patients receiving CDK inhibitor and endocrine therapy show specific ctDNA mutations, such as retinoblastoma, ESR1, fibroblast growth factor receptor 1, or phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha alterations [92,93,94], potentially influencing disease outcomes and therapeutic decisions. Additionally, detecting and quantifying CTC acquired resistance may serve as a predictive marker for treatment outcomes [95]. In castration-resistant prostate cancer patients treated with docetaxel, CTC count in blood is a reliable indication of therapy sensitivity and survival [96].

Osteosarcoma treatment difficulty stems from genetic instability and the emergence of chemotherapy resistance after selection pressure; low levels of miR-375 in osteosarcoma patients have been associated with poor response to preoperative chemotherapy [97]. Recent research has linked tumor-associated miRNAs to osteosarcoma chemoresistance, including miR-491, which is decreased in serum from patients compared to that in healthy controls. This decrease is linked to increased metastasis, poor chemoresponse, and lower survival rates [98]. Serum miR-21 levels are considerably greater in osteosarcoma patients compared to those in controls and are associated with advanced Enneking stage and chemotherapy resistance [99]. Reduced miR-125b levels in Ewing sarcoma patients are linked to poor treatment response and chemoresistance development [87].

EVs play a crucial role in drug resistance transmission [100], making them valuable for monitoring its emergence during therapy. miRNAs may represent indicators for chemoresistance. High quantities of membrane transporter pump P-glycoprotein in EVs from doxorubicin-resistant osteosarcoma MG63 cells facilitate horizontal transmission of resistance to susceptible cancer cells [101]. Additionally, miR-25-3p overexpression in osteosarcoma patients’ blood has been linked to tumor development and medication resistance [102]. The levels of miR-222-3p in EVs derived from NSCLC patients’ blood may predict sensitivity to gemcitabine and identify individuals with advanced and resistant illness [103]. In addition, platinum-resistant NSCLC patients have increased EV-derived miR-425-3p in their blood compared to that of platinum-sensitive individuals [104]. Research indicates that miR-222 from doxorubicin-resistant breast cancer cells is transported by EVs to M2 macrophages, inducing polarization. In contrast, miR-222 overexpression suppresses the expression of the tensin homolog gene and phosphatase activity, leading to Akt phosphorylation and activation, which promotes the proliferation of cancer cells, as well as their migration and invasion through positive feedback. EVs from the plasma of chemoresistant breast cancer patients show increased levels of miRNA-222 [105], while paclitaxel-treated cells from the human osteotropic breast cancer cell line MDA-MB-231 release EVs enriched in Survivin [106]. Unexpectedly, EVs can directly inhibit anti-neoplastic drugs; EVs from HER2-positive breast cancer patients behave as decoy receptors for trastuzumab, affecting its activity [107].

EVs generated by cancer cells include HER2 on their surface that is bound by trastuzumab systemically, reducing the quantity of antibodies available for cell binding. Yang et al. identified increased GSTP1 mRNA levels in EVs from non-responding breast cancer patients treated with neoadjuvant chemotherapy compared to those of responders. GSTP1-containing EVs have been shown to transmit drug resistance horizontally, suggesting their potential as negative predictors of chemoresistance and clinical outcomes in breast cancer patients receiving anthracycline/taxane treatment [108].

Transient receptor potential channel 5 mRNA in EVs isolated from the blood of metastatic breast cancer patients may predict chemoresistance [109]. Kharaziha et al. [110] identified MDR-1, MDR-3, endophilin-A2, and poly(A) binding protein 4 as enriched proteins in EVs from both prostate cancer cells resistant to docetaxel and castration-resistant prostate cancer patient serum, suggesting potential as biomarkers for therapeutic response or resistance [110]. The field is promising in cancer research, although larger longitudinal studies are necessary to confirm the effect of biomarkers.

Despite the potential of liquid biopsies, difficulties must be solved before broad clinical use. Due to the sensitivity of the approaches, even little variations in sample collection or processing may significantly impact the final output. The use of serum instead of plasma may increase cell-free DNA from other sources, lowering the diagnostic power of NGS-based tests, particularly for uncommon variations [111]. Lifestyle variables may impact cell-free DNA release in the bloodstream, creating a complex set of confounding factors that are challenging to detect and define [112]. CTCs are uncommon and difficult to acquire, and although the CellSearch technique offers a uniform approach, it is limited in its viability. They are only suitable for DNA and FACS/Immunofluorescence investigations, not RNA-based or functional experiments such as patient-derived xenografts or in vitro drug sensitivity testing [113,114]. CTC analysis also has limitations similar to those of conventional biopsies, since it may not reflect the complete tumor, but rather a subset of cells that survived in the circulation. An experiment is underway to partly address this problem by selecting various blood collection locations. Current research suggests that arterial blood and blood from near the main tumor may provide more CTCs [115,116].

Despite efforts to develop guidelines for sample treatment, liquid biopsy requires training, specialized facilities, and expertise in the interpretation of results [117]. EVs have unique preanalytical obstacles, in addition to the broader issues mentioned above. They represent a new source of biological information; however, further research is needed to develop EVs as a liquid biopsy. EV isolation is a prime example, as several methods have been investigated for isolating EVs [118]. Unfortunately, there is no perfect strategy for EV separation, and findings may vary based on the investigator’s approach [118]. Additionally, lifestyle variables may also promote EV release, making tumor-specific exosome identification difficult [41].