Integration of Nanotechnology with ctDNA: New Advances for Monitoring Neoadjuvant Chemotherapy

Abstract

The evaluation of tumor response in the neoadjuvant chemotherapy (NACT) is still difficult as traditional imaging techniques can only provide morphological information at discrete intervals, and pathological response is usually assessed only after surgery. Although circulating tumor DNA (ctDNA) has the potential to provide a minimally invasive approach to monitor molecular response during treatment, its clinical use is hampered by low presence, high fragmentation, rapid clearance and the presence of high levels of non-tumor cell-free DNA (cfDNA). In this review, we aim to discuss the challenges of ctDNA detection in analytical and biological settings and how nanotechnology may be able to help address these challenges, particularly during the NACT window. The analytical performance and the potential integration of nanotechnology platforms, including gold nanoparticle plasmonic sensing, magnetic nanoparticle (MNP) extraction and microfluidic integration, fluorescent quantum dots for multiplexing, carbon-based electrochemical sensors and nano‑based CRISPR biosensing, enable a rapid detection of ctDNA. Benefits of these nanotechnological platforms are not just that technologies exist, but that early ctDNA clearance predicts response, persistence signals resistance, and nanotechnology makes routine, sensitive, and rapid monitoring feasible, provided certain translational hurdles are cleared. While there is great promise for the use of nanotechnology to enable fast, sensitive detection of ctDNA during NACT, significant issues remain in standardization of assays, reproducible manufacturing, prospective validation, and cost-effectiveness before clinical use is a reality.

Introduction

The growing burden of the cancer and clinical need to act earlier in the course of disease have led to the neoadjuvant chemotherapy (NACT).^1,2 The updated worldwide estimates show that approximately 20.6 million new cancer cases are expected for 2024 and approximately 9.8 million people are expected to die from cancer during the same period, while the demographic projections show an increase in the annual number of cancer patients by mid of this century.³ These figures indicate that the treatment strategies used to eradicate the micrometastatic disease at an early stage and to achieve local control are required and for the solid tumors the adaptive therapy prioritizing these strategies should be implemented. The NACT is clinically useful as it can downstage the tumours to allow non-morbid surgical downstaging, and act as an in vivo stress test of tumour sensitivity to allow the response adapted postoperative escalation.⁴ The neoadjuvant regimens also allow the pathologic response endpoints to be applied to a variety of tumors to enhance the assessment of the outcome early in treatment and the development of therapy.⁵ For example, in the early stage of triple-negative breast cancer, the neoadjuvant chemo-immunotherapy supported by randomized evidence shows the improved pathologic response and survival when pembrolizumab is added to chemotherapy in the KEYNOTE-522 regimen.⁶ For HER2 positive cancer, same applies regarding neoadjuvant HER2 oriented therapies, and the post neoadjuvant decisions illustrate how the response assessment directly influences the next treatment.⁷ But the possibility of relapse remains, putting pressure on the field to improve the measurement of the response.⁸ Another example is colorectal cancer in which total neoadjuvant therapy (TNT) has been used to enhance the systemic therapy (chemoradiation) and hence disease control.⁹ The resectable non small cell lung cancer (NSCLC) has also undergone such neoadjuvant metamorphosis. The addition of nivolumab to the NACT regimen showed benefits in terms of pathologic complete response and event-free survival, along with presurgical clearance of the circulating tumor DNA (ctDNA) with overall favorable survival.¹⁰ The expansion of the neoadjuvant chemo-immunotherapy is a good example of the need for frequent and real-time monitoring of response to identify early nonresponders and for guidance on post-operative intensification, where molecular disease persists, as in the case of NSCLC.¹¹

In the conventional approach to response assessment of the NACT, imaging and pathology are used and typically offer an incomplete and delayed picture of the effects of the therapy. The imaging techniques mostly assess anatomic regression and indirect physiologic traits that may be delayed from true biologic responses.¹² Therefore, the performance will differ between the various tumor subtypes, therapy regimens and timepoints.¹³ The imaging techniques might not be accurate enough to support the significant changes in therapy that occur when the treatment is introduced early, stepped up or avoided altogether based on image analysis, without the additional biomarkers that reflect closer on the biology of the tumors.¹⁴ Although the pathologic assessment occurs after surgery, it is unable to do real-time adaptation during the neoadjuvant therapy.⁵ Therefore, imaging and/or pathology alone will not accurately reflect minimal residual disease and the volume of tumor below the level of imaging and/or pathological detection and drives recurrence.¹⁵ To fill this gap, the ctDNA has emerged as a candidate biomarker which is capable of tracking the tumor kinetics in near real time.¹⁴

The ctDNA is the part of the circulating cell free DNA (cfDNA) that comes from the tumor, as the tumor cells apoptose, necrose and otherwise dump their DNA into the bloodstream.¹⁶ It can provide a molecular lens into the evolution of the tumor and its response to treatment before changes can even be seen on a scan.¹⁴ The serial measurement of ctDNA in breast cancer has revealed clinically relevant association with the pathological response and the risk of recurrence during NACT.¹⁷ Moreover, there is growing evidence that ctDNA can also be used to detect the high risk biology in patients with good pathologic outcomes.¹⁸ This makes ctDNA one of the most intensively investigated biomarkers in the CRC. A study has demonstrated that the dynamics of ctDNA can be tied to therapeutic response; however, its accuracy alone is variable and faces challenges of performance and specificity/sensitivity.⁷ Yet, the promise of ctDNA is not undermined by these limitations, as they make it clear that greater analytical sensitivity and seamless incorporation into clinical endpoints are essential for routine use of neoadjuvant ctDNA.¹⁹

The emergence of neoadjuvant chemo-immunotherapy in resectable NSCLC has sparked renewed interest in the use of ctDNA as a dynamic response biomarker and as a perioperative minimal residual disease (MRD) tool. One takeaway from the long-term follow-up in CheckMate 816 is that there was an association between improved survival outcomes and the presence of presurgical ctDNA clearance and pathologic response, suggesting that perhaps ctDNA could be a tool to distinguish true biologic responders from patients with ambiguous tissue findings or radiographic stability.²⁰ Independent studies have likewise identified a relationship between ctDNA levels during neoadjuvant immunotherapy and pathologic response and recurrence-free survival, confirming that in general, ctDNA levels can be a marker of therapeutic sensitivity.²¹ What this data illustrates together is a consistent story: ctDNA is clinically interesting because it can provide serial, minimally invasive monitoring signals that can be a proxy for tumor dynamics on clinically relevant timescales for neoadjuvant decision-making.¹⁴ However same biology that makes ctDNA useful also poses many technical hurdles that prevent routine use especially for MRD and early detection of on treatment response.²²

The underlying analytical challenge is simple but critical. In many clinically important situations, the amount of ctDNA may be very small, with variant allele frequencies (VAF) as low as or below 0.01%, and may be detectable or not depending on the timing of sampling, shedding of the tumor and the design of the assay.²³ Concurrently, both cfDNA and ctDNA are highly fragmented, with a characteristic modal size of 166 base pairs in plasma and tumor-derived fragments often being shorter, impacting extraction efficiency and detection.²³ Today’s ctDNA field has been enabled by conventional molecular platforms, droplet digital PCR (ddPCR) and next-generation sequencing (NGS), which have respective limitations that are most evident in the neoadjuvant MRD and early response settings.²⁴ Therefore, highly sensitive, small sample size, fast, and scalable ctDNA workflows need to be developed and repeated with a high frequency and reliability during NACT.²⁵

Nanotechnology is not a single better detector, but a toolkit that can enhance ctDNA analysis throughout the entire analysis process, from capture, enrichment, amplification, to readout, often via integrated devices with reduced sample and reagent requirements.²⁶ In the front end, nanomaterial surfaces and magnetic nanoparticle (MNP) systems can improve the extraction and size selection, and maximize the recovery of ctDNA and background DNA removal, which is particularly important when the amount of ctDNA is limited.²⁷ The nano enabled amplification of signal at the stage of detection can push the readout into single molecule or ultra low concentration regime. The plasmonic nanostructures are enabling surface-enhanced Raman spectroscopy (SERS) substrates that are highly sensitive to nucleic acid detection, and the last few studies highlight the development of engineered substrates and assay design and their transition to complex biofluid applications.²⁸ Often based on graphene or other nanostructured interfaces, electrochemical and field-effect transistor (FET) biosensors provide label-free or minimally labeled detection with promising prospects for miniaturization and point-of-care translation, yet robust and standardized clinical translation is an ongoing challenge.²⁹

However, there is still a need to explicitly state the translation bar. In addition to sensitivity, neoadjuvant monitoring also needs to be reproducible over time, reproducible in terms of preanalytics, harmonized, validated for clinical thresholds, and demonstrated to be able to improve outcomes.³⁰ While current evidence indicates a role for ctDNA in prognosis across cancers and a correlation between on-treatment ctDNA dynamics and pathological response, clinical actionability is still a developing field with a number of ongoing trials and key discussions highlighting the need for prospective validation in a decision-making setting.³¹ Thus, the potential of nanotechnology is to render ctDNA monitoring technically achievable with the sensitivity and speed and cost point needed to facilitate clinical studies that would demonstrate (or not) its true usefulness for neoadjuvant application.²⁶

In this review, the technological evidence for the above ctDNA uses is summarized: baseline ctDNA for risk stratification prior to NACT, early on-treatment ctDNA kinetics as a chemosensitivity or chemoresistance marker, presurgical ctDNA clearance as a correlate of deep response, postoperative ctDNA as MRD for recurrence prediction, and the improved feasibility and clinical reliability of nano-enabled detection platforms. The aim is to concentrate on the NACT window, where early molecular response data could impact treatment assessment prior to surgery, which makes it distinct from previous reviews that broadly summarize the applications of ctDNA. This review questions the ability of nanoparticle platforms to consistently identify low abundance and fragmented ctDNA, if pathological response or recurrence could be associated with changes in ctDNA, and if the results can be used as a basis for clinical decision-making.

The Biological Basis of the ctDNA in Monitoring the Neoadjuvant Chemotherapy

ctDNA is derived from the cfDNA in blood and other biofluids that is derived from the tumor and contains molecular information from malignant cells that can be sampled repeatedly during treatment.³² Interpretation of ctDNA in the context of NACT needs to be done in an explicit context of biology as the measured signal is a dynamic balance between the shedding of tumor DNA into the system and the clearance of the DNA from the system, both of which can change rapidly when turnover of tumor cells or the tumor microenvironment is perturbed by cytotoxic therapy.³³ It is this biology of ctDNA that makes it appealing for monitoring response in neoadjuvant window and can provide an early molecular readout of whether treatment is effectively suppressing viable tumor clones.³³

The Biological Aspects of ctDNA

In healthy people, cfDNA is thought to be derived for the most part from normal turnover of cells, particularly nucleosome-associated DNA fragments released as a by-product of the normal apoptosis of hematopoietic cells.³⁴ The relative amount of ctDNA in total cfDNA varies from very low levels in localized cancer to much higher levels in advanced cancer and/or in high-burden states, influenced by tumor load, tumor vascularity, tumor cell death, metastatic tumor distribution and therapy effects.¹⁴ During the course of NACT, the amount of ctDNA can rapidly change as a function of both the amount of tumor cells and the major mechanism of tumor cell kill, making abundance of ctDNA a dynamic measure of tumor biology, rather than a static biomarker.³⁵

Apoptosis is a major contributor to cfDNA and its release. The characteristic nucleosome-protected fragments that result from programmed cell death underlie the presence of a modal peak corresponding to one nucleosome length of cfDNA in plasma, and subsequent peaks corresponding to di- and tri-nucleosomes.^16,34 In the context of effective chemotherapy, the levels of apoptosis in sensitive subclones could be increased in the initial phase, leading to an overall decrease in the levels of ctDNA as the viable tumor cells diminish.³⁶

Necrosis is another source of release of tumour DNA, particularly in the hypoxic areas of a tumour and in fast growing tumours with poor perfusion.³⁴ Necrotic release generally has been thought of as yielding less orderly fragmentation, for instance, longer molecules of DNA or more variable sizes of fragments, but even extensive extracellular nuclease activity can degrade these before they are detected.³⁷ From a clinical perspective, necrosis can also be regulated by ischemic damage, immune mediated injury and microenvironmental stress. Thus the shedding of ctDNA may differ in response to a given therapy and vary between individuals with nominally similar neoadjuvant therapy.³⁵

In addition to the passive release by dying cells, active secretion is suggested, such as the export of DNA from viable cells or the packaging in extracellular vesicles, such as exosomes or microparticles.³⁸ The biological significance of active release is that ctDNA sometimes can reflect persistent tumour activity, in the absence of extensive cell death, and therefore may provide additional sensitivity to monitor the presence of resistant clones after therapy selection.³⁹

The heterogeneity of tumours is closely related to biological value of ctDNA. Spatial heterogeneity (differences across regions and metastatic sites) and temporal heterogeneity (clonal evolution over time) are two traits of solid tumors that can contribute to therapeutic response and resistance during NACT.⁴⁰ ctDNA may overcome some of the sampling bias of a single-site tissue biopsy and capture the changing subclonal architecture during treatment.⁴¹ But another source of uncertainty is the heterogeneity of ctDNA itself. Different lesions shed at different rates and highly vascular, necrotic or rapidly cycling lesions may be over represented in blood while indolent or low shedding lesions may be under represented.⁴²

Kinetics of ctDNA During Chemotherapy

cfDNA is sometimes referred to as a real-time biomarker as a result of the rapid turnover of ctDNA, which is estimated to be minutes to a few hours.³² This brief persistence of cfDNA/ctDNA is mechanistically related to clearance through the action of circulating nucleases (DNase activity), uptake and processing of nucleosomes, and filtration/clearance by organs such as liver and kidneys; disruption of renal or hepatic function may affect the concentration of cfDNA/ctDNA and make longitudinal interpretation difficult.³³ These clearance kinetics have potential implications in neoadjuvant setting as ctDNA levels may fluctuate over days or even hours after exposure to chemotherapy. This allows for the early assessment of response but also makes sampling time and physiologic variation critical factors.³⁵

The ctDNA kinetics are influenced by competing mechanisms during chemotherapy: as treatment leads to the death of tumor cells, the amount of ctDNA may temporarily rise, but when treatment is effective, the viable tumor mass from which ctDNA is produced is reduced.³⁶ There are mechanistic modeling studies that suggest that an early peak or transient increase in ctDNA, due to an acute wave of apoptosis, can occur at the onset of therapy and then decrease below baseline if therapy is effective.³⁶ However, early kinetics are clinically choppy, and wider observations highlight that interpretation of molecular response should consider the intrinsic nature of the variability of ctDNA assays, assay limitations, and patient-to-patient differences in shedding.⁴²

The early clearance and persistence are most clinically relevant kinetic states in neoadjuvant monitoring. On-treatment clearance of ctDNA at an early timepoint is called early clearance and persistence of ctDNA despite therapy is called persistence.⁴³ In several studies, and for various tumor types, the decrease in ctDNA has repeatedly been correlated with better response and outcomes and, conversely, the persistence of ctDNA was often correlated with poor pathologic response, residual disease, and early relapse, suggesting the biological concept that resistant clones survive and continue to shed DNA.⁴³

The Application of ctDNA as a Biomarker of Therapeutic Response

One of the major benefits of ctDNA in NACT is to predict early on whether the patient will respond or not. Within the setting of early breast cancer, the serial evaluation of ctDNA in the setting of neoadjuvant breast cancer treatment has shown that early changes in ctDNA can be correlated with final pathological response and long-term outcomes in high-risk early breast cancer.⁸ Studies provide an overview of neoadjuvant and perioperative settings, highlighting that clearance of ctDNA may precede radiographic response and that it may be useful during the black box period of neoadjuvant treatment.⁴³ In the neoadjuvant immunotherapy and chemo-immunotherapy contexts in solid tumors, ctDNA clearance has also been tested as a surrogate of pCR, further supporting the general idea that early molecular response is a good indicator of deep tumor regression.⁴⁴

ctDNA also serves as a biomarker for persisting disease following neoadjuvant treatment, especially if sampled after NACT and prior to surgery. Retrospective analyses of data from tumor-informed assays have demonstrated a strong association between ctDNA positivity after neoadjuvant therapy (pre-surgery) and recurrence in locally advanced rectal cancer (in that data set, 100% of patients were positive for recurrence if they were also positive for ctDNA after neoadjuvant therapy), with significantly worse disease-free survival hazard ratios seen for post-NACT and post-surgery ctDNA positivity.⁴⁵ In rectal cancer, there has also been some interesting research on ctDNA in the longitudinal setting, including after neoadjuvant chemoradiation (NAC), with interesting findings that the dynamics of ctDNA may correlate with pathological response; however, it may be imperfectly specific in some settings, which is a big caveat to keep in mind when considering decisions for non-operative management.⁴⁶ Figure 1 shows a schematic representation of traditional neoadjuvant approach where treatment is administered without real time insight into tumour response versus the ctDNA-guided neoadjuvant approach, where the decreasing levels of ctDNA represent clearance and stable or rising levels indicate persistence in early therapy adaptation.

Figure 1.

A schematic representation of (A) traditional neoadjuvant approach versus (B) the ctDNA-guided neoadjuvant approach in early therapy adaptation.

Most consistent clinical association of ctDNA in the realm of solid tumors is the prognostic value of MRD. Postoperatively, ctDNA positivity is repeatedly associated with much higher recurrence rates than postoperatively, ctDNA negativity.⁴⁷ In colorectal cancer (CRC) per se, clinical studies have paved the way for the use of ctDNA as a strong prognostic tool and a novel tool for adjuvant decision making and surveillance intensity.⁴⁸ Post neoadjuvant ctDNA has been shown to correlate with recurrence risk in breast cancer, highlighting how the detection of ctDNA after neoadjuvant therapy and surgery can help to stratify risk, even in patients who would be considered favourable based on conventional response metrics.⁷

A biologically based interpretation of these data is that ctDNA reflects two clinically relevant signals, namely the completeness of tumor killing (the extent of suppression of viable shedding clones by treatment), and the presence of clinically relevant, residual, viable disease that could reseed relapse.⁴⁴ However, ctDNA is not a direct census of all tumor cells at the same time. Shedding heterogeneity and physiologic variability are important to consider, as ctDNA negativity does not always represent biologic cure, especially in low shedding tumors or when below the limits of detection of the assay; interpretation is best when using ctDNA kinetic alongside tumor biology, preanalytical rigor and complementary clinical end points.⁴²

Detection of ctDNA with Various Nanotechnology Platforms

While the analytical promise of ctDNA is obvious (serial minimally invasive access to tumor derived genomic information), the practical measurement is limited by low abundance, short fragment size and high background from non-tumor cfDNA.⁴⁹ These constraints are most important in clinical situations where they make a difference in the monitoring of neoadjuvant chemotherapy (early on-treatment kinetics, deep responses, and minimal residual disease) and where preanalytical losses may be the predominant source of signal.²⁶ To overcome these challenges throughout the ctDNA workflow, nanotechnology platforms are being developed. Different platforms of nanotechnology have been of interest to the current research in ctDNA biosensing, such as plasmonic gold nanoparticle systems, MNP enrichment systems, quantum dot and fluorescent nanobiosensors, carbon based nano materials, and nano-enabled CRISPR biosensors.²⁶ Figure 2 illustrates four key nanotechnological modules to address the low abundance and complexity of ctDNA. The biofluid samples serve as non invasive source for liquid biopsy and high surface area nanotechnology platforms, along with high affinity recognition elements, are utilized to overcome low target concentration of ctDNA. Signal transduction strategies convert subtle molecular binding events into quantifiable signals to significantly enhance detection of ctDNA while convergence of nanostructures with CRISPR/Cas systems enables high specificity in diagnosis.²⁶ This section focuses on the principles of operation, recent developments and translation-related limitations of nanotechnology platform in the detection of ctDNA.

Figure 2.

An illustrative overview of the nanotechnology enabled detection of ctDNA.²⁶

Gold Nanoparticle-Based Biosensors

One of the most popular classes of nanomaterials for nucleic acid biosensing is gold nanoparticles (AuNPs), which have optical properties that allow its readouts to be strong and tunable without the need for bulky instrumentation.⁵⁰ The important phenomenon is localized surface plasmon resonance (LSPR). Collective oscillation of conduction electrons under illumination that creates a characteristic extinction band that changes as a function of particle size or shape, local refractive index and aggregation state. This physical basis enables plasmonic sensing, surface plasmon resonance (SPR) and colorimetric AuNP assays relevant to ctDNA.⁵⁰

Plasmonic sensing strategies rely on engineered AuNP geometries and surfaces that provide the maximum near-field enhancement and sensitivity to binding events. For ctDNA applications, localized plasmonic AuNP arrays can be on-chip or immobilized to enhance local fields and thus hybridization or sandwich capture detection at low target concentrations.⁵¹ Label-free detection can be achieved by measuring refractive-index changes near the surface, which are characteristic of the binding of biomolecules to the SPR biosensors, which are typically thin gold films coated on a prism or waveguide.⁵² For ctDNA, SPR’s strengths are real-time kinetic measurements and straightforward surface chemistry, while weaknesses include challenges in achieving ultra-low detection limits in complex plasma without amplification.⁵² To fill this gap, AuNPs serve as mass and refractive-index amplifiers, thereby enhancing the SPR signal when they are used as labeled reporters in sandwich assays or as plasmonic enhancers that are deposited on the sensing surface.⁵³ The most translation-tempting plasmonic format is colorimetric AuNP assays that can give a visual readout, thus allowing for low-cost and decentralized testing.⁵⁰ In the typical designs, target DNA hybridization causes AuNP aggregation (causing a change in color from red to blue), or it hinders salt-induced AuNP aggregation (maintaining the dispersed red state).⁵⁰ A recent example involved a fast, non-crosslinking AuNP aggregation assay that leverages mutation-specific hybridization to generate a strong optical signal for detection of a frequently found EGFR exon 19 deletion in non-small cell lung cancer.⁵³

Magnetic Nanoparticles for ctDNA Enrichment and Isolation

Though the focus of the AuNP platforms is often readout, MNPs are frequently used upstream to address an equally important aspect of recovery and enrichment of ctDNA from plasma with minimal loss and contamination.⁵⁴ The silica like surface, polymers or affinity ligands can be attached to superparamagnetic particles to allow for control of DNA adsorption and release based on salt or alcohol conditions or on sequence-specific capture chemistry; the use of an outside magnet can facilitate rapid washing and elution.⁵⁴

One of the key improvements made to ctDNA is size-selective magnetic extraction. Short DNA fragments predominate in the ctDNA compared to background cfDNA, and decreasing the amount of longer genomic DNA contamination leads to better sequencing and sensor specificity.⁵⁴ A size-selective cfDNA extraction platform used a nano-magnetic approach that integrated size selection and extraction into one step, achieving >90% recovery of cfDNA with low levels of genomic DNA contamination and in patient plasma, demonstrating proportionally more ctDNA (as measured by targeted panel sequencing) consistent with pragmatic enrichment of tumor-derived fragments. In neoadjuvant monitoring, in which ctDNA may be detectable at low levels occasionally, pre-analytical recovery may be critical for whether a trend is detectable or not.⁵⁵

To enhance the possibility of reproducibility and throughput, integration of microfluidic is gradually being adopted. As of 2026, an MNP integrated microfluidic chip has been developed to reliably isolate plasma cfDNA, with advantages including manual handling minimization and minimized manual labor compared to batch extraction.⁵⁶ Additionally, MNPs can be used for direct coupling of enrichment to downstream biosensing. A model study combined magnetic enrichment of unprocessed plasma with electrochemical signal enhancement and molecular amplification to enable multi-ctDNA single nucleotide variant (SNV) detection; magnetics were used here as a tool to mitigate matrices and facilitate faster workflows.⁵⁶ Likewise, integrated microfluidic systems have been reported that both extract and automatically detect and quantify specific mutations (eg, BRCA1 biomarker) from plasma, proving the feasibility of closed-loop processing from raw plasma to mutation readout.⁵⁷

Quantum Dots and Fluorescent Nanoprobes

Nanocrystals of semiconductors (quantum dots, QDs) can be made to emit a narrow-band, bright fluorescence with a wavelength tunable by size and composition, and, critically for multiplexing, several different emission colors with a single excitation. Another benefit of QDs is its ability to be used in electrochemiluminescence (ECL) formats with low optical background, which is beneficial in complex biological matrices.⁵⁸ A ctDNA-relevant improvement is the use of QDs in resonance energy transfer schemes (ECL-RET or FRET-like designs) in which the distances between the QDs in the system are modulated by the hybridization-induced structural changes, which are translated into measurable luminescence changes. The biosensor was designed based on CdS QDs and hairpin DNA probes for the detection of the clinically important EGFR T790M ctDNA, which shows how the photophysics of QDs can be combined with mutation-selective hybridization.⁵⁹ Such approaches are appealing for neoadjuvant applications as they could in principle be designed to have low-input and quick readout; the limitation here is clinical-grade robustness.

In addition to classical QDs, fluorescent nanoprobes are increasingly represented by perovskite nanocrystals and carbon dots, as well as 2D nanomaterial-probe hybrids whose fluorescence intensity can be quenched or restored by the binding of a target. In 2024, the label-free fluorescent biosensor was developed based on the modulation of the interactions between the perovskite nanostructures and a 2D quencher via probe hybridization, and was used for the detection of EGFR exon 19 deletion ctDNA.⁶⁰ These formats highlight an underlying design approach in ctDNA sensing, namely, to provide a fluorescence gain or more efficient quenching through nanomaterials, and to provide a molecular-level gain through nucleic-acid engineering (eg, molecular-level hairpins, walkers, HCR/CHA circuits).⁶¹

Carbon-Based Nanomaterials

The main method used for ctDNA detection using carbon nanomaterials is electrical transduction, particularly with graphene and carbon nanotubes (CNTs). They are ideal for electrochemical sensors and field-effect transistor (FET) biosensors that involve hybridization events close to their surface that result in changes in charge distribution, current, impedance, or redox signaling due to their high conductivity and large surface-to-volume ratio.⁶² One such demonstration, with a particular focus on ctDNA, is the application of CNT transistors with DNA nanostructures to enhance the orientation of probes and hybridization efficiencies. One of the methods for achieving this is to employ DNA architecture to control the presentation of nanoscale probes and minimize steric/electrostatic barriers, as was done for a biosensor based on a tetrahedral DNA nanostructure that utilizes an all-carbon nanotube transistor for high performance detection of ctDNA.⁶³ This architecture is in line with the general strategies of FET biosensing, which involve the design of probe layers (such as peptide nucleic acids or DNA nanostructures) to reduce Debye screening and increase mismatch discrimination in physiological ionic-strength.⁶⁴

Other applications for mutation detection are with solid tumor mutation detection using graphene based electrochemical platforms. An electrochemical DNA biosensor, for instance, was fabricated based on the reduced graphene oxide (rGO)/metal-organic framework (MOF) composite, which was used to detect the clinically important KRAS G12D oncogenic driver mutation with ultra-high sensitivity.⁶⁵ Many of these studies are first tested in buffered oligonucleotide systems; however, the design directions, antifouling coatings, nanostructured electrodes, and nuclease-assisted amplification are more and more geared towards biofluids that need little or no processing.⁶² The advantages, disadvantages, detection mechanisms, and analytical performance of each type of nanoparticle platform are summarized in Table 1.

Table 1.

The Detection Mechanisms, Advantages, Disadvantages, and Analytical Performance of AuNPs, MNPs, QDs, and Carbon-Based Nanomaterials for the Detection of ctDNA

Platform	Detection Mechanism	Advantages	Disadvantages	Analytical Performance	References
AuNPs	Probe-functionalized AuNPs recognize complementary sequences of ctDNA. When hybridization occurs a color change or ultraviolet-visible absorbance change is observed due to aggregation or dispersion of nanoparticles.	Strong plasmonic properties, high optical stability, compatibility with colorimetric, electrochemical, potential for rapid and instrument-light testing, suitable for catalytic hairpin assembly and DNA-walker amplification, and capacity for single-nucleotide discrimination.	Salt concentration, pH, temperature and incubation time could alter the aggregation of nanoparticles; non-specific aggregation can lead to false-positive signals; size and morphology of nanoparticles can have a significant impact on optical responses	A gravitationally condensed gold nanoparticles-based catalytic walker DNA assay was reported to detect a deletion in the EGFR gene with a limit of detection of 38.5 aM and a mutant to wild-type discrimination of approximately 0.1%.	[66–69]
MNPs	Probes or coating with DNA-binding sequences were functionalised on Fe3O4-Au core-shell nanoparticle magnetic beads, where the external magnetic field is applied to separate ctDNA from the plasma. The enriched DNA are then subjected to PCR, electrochemical detection, giant magnetoresistive sensing or optical analysis and sequencing.	Fast magnetic separation; high efficiency for low abundance ctDNA; low background from proteins and non-target cfDNA; large surface area for probe immobilisation; automatic and microfluidic; possibility of integrating enrichment, amplification and detection in a single platform	Loss of target during repeated manipulations; additional surface-coating and washing steps; possible bias towards particular fragment lengths; dependence on magnetic-field strength and mixing efficiency; nanoparticle aggregation and oxidation; and non-specific adsorption of normal cfDNA and plasma proteins.	EGFR-mutated ctDNA was detected at a mutant allele fraction of 0.01% by a giant magnetoresistive nanosensor, which was then tested in plasma of 36 non-small-cell lung cancer patients. A Fe3O4-Au magnetic nanoelectrode platform that integrated surface-confined amplification with electrochemical detection was able to detect mutant ctDNA at the attomolar level.	[67,70–72]
QDs	Target hybridisation changes the fluorescence via Förster resonance energy transfer, electrochemiluminescence resonance energy transfer, quenching or recovery of fluorescence or aggregation of differently coloured quantum dots.	High fluorescence intensity and quantum yield; strong resistance to photobleaching; broad excitation with narrow and size tunable emission spectra; simultaneous detection of differently coloured probes; compatible with fluorescence imaging and electrochemiluminescence.	Fluorescence blinking and signal heterogeneity; potential spectral overlap; quenching by plasma constituents; need for optical filters, cameras or electrochemiluminescence devices; complicated surface passivation and bioconjugation; potential toxicity of cadmium containing quantum-dots	ECL-RET sensor based on CdS quantum dot and AuNPs achieved the detection of EGFR T790M mutation in single and double-stranded DNA fragments, even mutations occurring as close as seven nucleotides from the end of a fragment.	[59,73,74]
Carbon-based nanomaterials	High surface area interfaces for probe immobilisation. ctDNA hybridisation can lead to modulation of electrical conductance, impedance, quenching of fluorescence, scattering of Raman or change of electrochemical current. Carbon nanomaterials can also be used to support metal nanoparticles or catalytic amplification labels.	High electrical conductivity; rapid electron transfer; large surface-to-volume ratio; strong adsorption of nucleic acids; carbon nanotubes provide characteristic Raman signal; graphene and carbon materials can be integrated into the flexible miniaturised and low-cost electrodes.	Hydrophobicity and aggregation in aqueous samples; non-specific adsorption of DNA and plasma proteins; variability in nanotube diameter, length, chirality and defect density; potential for signal drift and fouling at the electrode; some formulations of nanotubes have biocompatibility issues that have yet to be addressed.	A DNA mediated SERS assay with single-walled carbon nanotubes, a triple-helix molecular switch and RNase HII amplification was able to detect the KRAS G12D mutated ctDNA at 0.3 fM with a 5-µL blood sample.	[75–77]

Clinical Applications of the ctDNA in Monitoring Neoadjuvant Chemotherapy

Increasingly, ctDNA is considered the most time-resolved tumor biomarker available in routine blood samples, as it can be repeatedly sampled within a neoadjuvant window of opportunity, and reflects changes in tumor burden and clonal composition that are often quicker than radiographic or surgical endpoints.¹⁷ One of the challenges for clinical use is the technical challenge of detecting very low fractions of ctDNA during active treatment or in MRD. It is common to see variant allele frequencies (VAFs) as low as 0.01% in contemporary assays that are designed for tumors; however, low shedding, loss of fragments, and preanalytical variation can limit sensitivity.⁷⁸ This is where nanotechnology has a translationally relevant impact. The nano‑magnetic size selection can aid in recovery of cfDNA and decrease genomic DNA contamination; nanopore consensus sequencing can allow for detection of mutations at very low fractions; and nano‑enabled CRISPR biosensors can offer a very low limit of detection for ctDNA targets in the femtomolar range.⁵⁵

Early Prediction of Chemotherapy Response

The greatest near term clinical utility of ctDNA in neoadjuvant treatment may be in prediction of near-term response rather than simply the shrinkage of the tumor, but rather the biological effectiveness of the regimen being used. Overall ctDNA positivity has decreased with treatment in I‑SPY 2: 73% at baseline, 14% at inter regimen time point and 9% at the end of neoadjuvant chemotherapy.¹⁷ Failure to achieve clearance of ctDNA at T1 was also highly correlated with surgery at T1: 83% of patients who failed to do so at T1 were non‑pCR compared with 52% of patients who achieved clearance by T1 (odds ratio 4.33).¹⁷ This illustrates that blood collected prior to surgery can be used to stratify the risk of pathological nonresponse, and offers a potential rationale for adaptive strategies such as escalating, changing agents, or adding agents within the neoadjuvant window.¹⁷

Recent studies on neoadjuvant breast cancer also support this idea with more analytical sensitivity. In a study conducted by Cabel et al, all evaluable patients had baseline detection, and among baseline-positive patients, 94% were negative by pre‑surgery and those that were still positive or became positive during follow-up were at risk of recurrence after surgery.⁷⁹ Similarly, Marra et al note that baseline detectability and complete clearance by ddPCR are seen in patients who undergo neoadjuvant systemic therapy and have longitudinal follow-up, and also highlight that detectability may provide prognostic information.⁸⁰

Monitoring Dynamic Tumor Burden During Treatment

In addition to early binary prediction, ctDNA can also be used to continuously monitor tumour burden. The evidence is particularly strong when quantified and interpreted over time (longitudinal analysis), such as tumor fraction, VAF, or tumor molecules. For triple-negative breast cancer (TBCRC 030), Parsons et al use an ultrasensitive mutation enrichment approach (MAESTRO) and find dramatically reduced levels of ctDNA tumor fraction at week 3 of neoadjuvant chemotherapy (NAC): 285-fold decrease from baseline in responders versus 24-fold decrease in non-responders.⁸¹ Interestingly, the ctDNA levels went below the threshold of detection of a commercially available test in 58% of patients in this study, which could have important implications for future guidelines and assay selection to monitor patients on treatment as opposed to use as a baseline for genotyping purposes.⁸¹ In esophageal cancer treated with neoadjuvant chemoradiotherapy, the proportion of patients with detectable ctDNA also fell significantly after chemoradiotherapy, further suggesting that the decline of ctDNA is likely to be a biological marker of treatment response.⁸² So the absolute level of ctDNA is the most informative at baseline, and the slope of decrease and persistence is most informative after the initiation of therapy.¹⁷

Detection of Resistant Clones and Treatment Failure

Neoadjuvant therapy is not just a way to reduce tumour size, it also introduces selective pressure, which can identify or select resistant subclones. In theory, serial ctDNA can be used to identify failure to clearance, subsequent rise, new variants, and expanding variants that are consistent with the resistance biology. The most meaningful signal in clinical decision making is sustained detection of ctDNA after treatment. This is associated with poor outcomes across all types of tumor and appears to precede clinical failure. Patients who did not get rid of ctDNA by the end of neoadjuvant therapy (T3) in I‑SPY 2 had a significantly higher distant recurrence risk compared to those who cleared earlier and to those that were negative at baseline.¹⁷ Just like with TBCRC 030, most of the patients with a known recurrence had detectable ctDNA at the end of therapy, which again aligns with the concept of persistence as a proxy for resistance.⁸¹ However, new mutations in plasma during therapy need to be guarded against, especially as clonal hematopoiesis can provide tumour-unrelated variants in cfDNA, which may be mistaken for tumour evolution if white-blood cell sequencing is not undertaken.⁸³ This is not a hypothetical issue as clonal hematopoiesis is specifically listed as a potential confounder which can result in faulty therapeutic management if not recognized as such.⁸³

Assessment of Minimal Residual Disease (MRD)

Theoretical clinical value of ctDNA is highest and technical sensitivity requirements are most stringent in MRD. In both neoadjuvant paradigms, the idea of MRD assessment is usually to look to see if there is residual systemic risk that is not detected by the local pathological assessment, either after neoadjuvant therapy, after surgery, or both. Post‑neoadjuvant and post‑surgery positivity for ctDNA is a very strong risk factor for recurrence in rectal cancer. A study demonstrates that ctDNA positive within the defined MRD window (2–12 weeks after surgery) results in a 100% recurrence rate compared with much lower event rates in the ctDNA negative group, and a large hazard ratio (HR) of 19.65 for inferior disease‑free survival.⁴⁵ Importantly, the same report points out that the same MRD time point in the cohort found to be ctDNA positive had a positive predictive value and specificity of 100%, with the sensitivity being low, as is often the case in MRD states.⁴⁵

When evaluating follow‑up samples in breast cancer, Cabel et al report that in single patients, ctDNA positivity could occur months before clinical recurrence (4.4 and 9.8 months before, respectively), while in that data set all patients not showing clinical recurrence remained negative at follow‑up, resulting in 100% specificity.⁷⁹ Although these studies also demonstrate that there are some recurrences that may occur without detection of ctDNA (and thus represent biology, particularly the low shedding and sanctuary sites), this is a primary reason for the rationale behind MRD-guided escalation trials.⁷⁹ MRD is particularly relevant in esophageal cancer as there is a strong clinical need because residual disease can be missed by conventional clinical response evaluation. One recent study showed that a standard (nontargeted) diagnostic approach had significant false-negative rates, especially for detection of residual disease (including a cited 23% false-negative rate for detection of any residual disease in an established European cohort), thus encouraging the use of ctDNA in conjunction with clinical diagnostics for post-neoadjuvant decision-making and risk stratification following surgery.⁸⁴

Prediction of Pathological Complete Response (pCR)

Although widely used as a surrogate endpoint in several solid tumors, pCR is flawed; some patients have a recurrence despite a pCR, and many patients with any residual disease do not have a recurrence. To improve the prognostic resolution of pCR, the use of ctDNA has been tested.⁷⁹ All pCR patients were also ctDNA negative after neoadjuvant chemotherapy and the majority of non pCR patients were also ctDNA positive after neoadjuvant chemotherapy and this strongly discriminated patients with higher risk from those with similar risk as pCR.¹⁷ As an important clinical nuance, ctDNA negativity can in some instances be used to define a low risk subset in R‑ID that might be eligible to de‑escalation or for non‑aggressive treatment. However, such application must be prospectively validated. Similarly, it is evident that in rectal cancer after TNT, the sensitivity of post‑TNT ctDNA was low (23%) and is not the same as complete response, despite the excellent specificity.¹⁷ Hence ctDNA is best used as an adjunctive, and not as a surrogate tumor marker especially in organ preservation approach, where the cost of false reassurance is significant.⁸⁵ Table 2 summarizes different peer-reviewed clinical cohort studies for monitoring ctDNA in NACT.

Table 2.

A Summary of Different Peer Reviewed Clinical Cohort Studies for Monitoring ctDNA in NACT

Tumor Type	Assay Type	Key Findings	Study Reference
Breast	Personalized tumor informed ctDNA	ctDNA positivity decreased from 73% to 9%, no clearance at T1 associated with non pCR, ctDNA negative non‑pCR had outcomes similar to pCR	Magbanua et al17
Breast	Tumor informed ctDNA	Clearance of ctDNA predicts favorable response across the receptor subtypes, less than 20% of resistant ctDNA positive tumours after NACT, ctDNA refined risk stratification	Magbanua et al8
Breast	Tumour informed assay tracking many variants	Baseline ctDNA detected 29/29, 94% cleared during treatment, post surgery follow up of ctDNA predicted recurrence	Cabel et al79
Breast	ddPCR baseline monitoring	Baseline detectable in 83% by ddPCR, 100% clearance in longitudinal subset, baseline detectability was associated with relapse proportion in cohort	Marra et al80
Triple-negative breast cancer	MAESTRO mutation enrichment	Responders showed 285 times drop in ctDNA TFx by week 3 vs 24 times in non responders, the end therapy clearance correlated with lower residual cancer burden	Parsons et al81
Rectal	Commercial ctDNA assay	Post TNT ctDNA was comparable with imaging and pathology	Alden et al85
Esophageal	Tumor informed hybrid capture sequencing	Detection of ctDNA decreased from 56% to 10%, preoperative ctDNA detection was associated with recurrence	Hofste et al82
Gastric	Cancer‑specific panel, serial ctDNA	Post NACT ctDNA negative patients had markedly better 3 year overall survival than ctDNA positive, persistent undetectable ctDNA prognosis	Zhang et al86
Rectal	Tumor‑informed personalized assay	Post‑NAT ctDNA positivity associates with higher recurrence, and MRD‑window ctDNA positivity	Molinari et al45
HER2-positive early breast cancer	Tumour-informed digital PCR targeting PIK3CA and TP53 mutations	ctDNA detection decreased from 41% to 20% at week 2 and 5% before surgery	Rothé et al87

CRISPR Biosensing Platforms Based on Nanotechnology

CRISPR-based diagnostics have developed quickly from concept to widespread biosensing architecture, since programmable guide RNAs can lead to accurate target recognition and in collateral cleavage systems (Cas12/Cas13), target recognition can result in a signal amplification through the cleavage of reporter molecules.⁸⁸ Cleavage-based amplification offers the prospect of faster, more portable assays and could be of particular interest for ctDNA, many clinically relevant mutations of which are single nucleotide variants, and because CRISPR readouts could overcome the need for thermal cycling.⁸⁹ One example is an electrochemiluminescence resonance energy transfer (ECL-RET) biosensor based on CRISPR-Cas12a enzyme to detect EGFR L858R ctDNA in the plasma of NSCLC patient samples, with the report covering a wide dynamic range and femtomolar-level sensitivity down to LOD 3.0 fM, demonstrating how CRISPR amplification can be combined with nano-enabled ECL transduction.⁹⁰ Additionally, plasmonic CRISPR formats are coming out. The trend towards user-friendly optical outputs of AuNP-based nucleic acid biosensors was illustrated by describing a CRISPR-associated plasmonic colorimeter method (Ca-PCM) as a real-time RGB detection system.⁹¹

CRISPR readouts have been developed using optical and QD reporters to enhance brightness, multiplexing, and portability of the readouts.⁹² However, this larger CRISPR biosensing literature also identifies persistent challenges with translation to clinical use, the need for PAM in some systems, the risk of off-target activation, background cleavage and carryover contamination, indicating that clinically deployable ctDNA CRISPR platforms will likely be limited by careful design of the guides and by built-in sample handling and orthogonal controls.⁸⁹ The nano-enabled CRISPR assays could be particularly interesting when the clinical question is quite specific, eg monitor a known driver mutation, and if the turnaround time needs to be sufficiently fast to impact therapy decision making during the course of therapy.⁸⁹

Translational Challenges and Future Trends

Translational Challenges

While clinical interest in ctDNA monitoring during neoadjuvant therapy is increasing, challenges include the lack of uniformity in laboratories and platforms, preanalytical, reporting and reference materials, prospective validation to prove clinical value, and barriers to implementation in real‑world practice such as regulation.⁹³ First translational gate is analytical reproducibility. If serial ctDNA measurements are not comparable over time and across labs, then neoadjuvant kinetics can become a measurement artifact rather than a biological signal. A large, multi-site, multi-platform study assessed a variety of industry ctDNA assays and underscored the need for proficiency testing and standardized reference materials as the accuracy, sensitivity, and reproducibility of the entire workflow are not consistent.⁹³ In addition, the inter‑assay variability increases in complexity when comparing families of platforms. Various trade offs in terms of LOD, breadth and artefact sensitivity exist between targeted NGS, ddPCR, fragmentomics, and (in emerging settings) nano enabled CRISPR assays. Recent studies make the distinction between analytical validity and clinical utility and highlight that the choice of assay affects sensitivity and interpretability in various clinical settings.^94,95

In the neoadjuvant setting, the validation requirements are increased as the use of ctDNA relates to dynamic response and maybe even influencing adaptive therapy decisions during treatment. The evidentiary bar is consequently raised compared to post‑surgery MRD purely for prognostic purposes.⁹⁵ For neoadjuvant therapy, a test can directly influence the choice of treatment and could therefore affect the outcome, toxicity and timing of surgery. Prospective validation requirements include clearly defined endpoints, standardized timepoints and decision rules that have been pre‑defined.⁹⁶ Selecting the timepoints from baseline to early on‑treatment and to pre‑surgery is biologically motivated but requires protocolization since the time of sampling relative to the cycle, surgery, and perioperative inflammation can affect detectability.⁹⁶ Due to the utility, prospective studies are usually required that involve correlating ctDNA results with a pre‑specified action (eg, escalate therapy if ctDNA persists, and de‑escalate if ctDNA clears) and outcomes are compared to standard‑of‑care strategies. If there are no large neoadjuvant utility trials in the field to guide specific decision making, the evidence should be reported as clinical validity, but actionability is not specified, rather than a clear benefit.⁹⁵

In addition to the risks associated with sequencing assays, the commercialization of nano‑platforms brings hardware associated risks. Biosensors based on nanomaterials are prone to batch to batch variability and there are also issues of stability and shelf life and matrix effects affecting specificity in plasma. When translating an idea for a biosensor to a clinical product, the key challenges that have been identified in the literature on biosensor translation and nanotechnology regulation are: scale‑up of manufacturing, quality systems, design of the supply chain and regulatory‑quality alignment.⁹⁷ Where critical reagents are nanomaterials that must be tightly physically and chemically characterized and sourced, these issues are magnified.⁹⁷

Cost‑effectiveness and accessibility are other major challenges. The per‑test price is not the only cost driver for neoadjuvant ctDNA monitoring. The expenses will vary depending on sampling frequency, complexity of the assays, and changes required in downstream management.⁹⁸ A 2024 modelling study suggests that even when good separation of prognosis is achieved, the use of a ctDNA‑guided strategy may not be cost‑effective, depending on the assumptions made regarding the performance of the ctDNA test and the benefit of chemotherapy.⁹⁸

These high‑complexity NGS‑based ctDNA tests need to be centralized in high‑resource centers. Nano‑enabled concepts for POCs could render the instrumentation requirements less complex, but they need to satisfy the stringent criteria for analytical validity and EQA and have proven clinical utility in the scenarios in which they are used in the neoadjuvant workflow.⁹⁹ Thus, the equity challenge is two‑fold, do not exacerbate inequities with tests that are too costly, and do not miss out on tests that are less costly and not validated.

Future Trends

The development of closed‑loop clinical systems, including standardized sampling in the neoadjuvant window, assays with sufficient sensitivity to identify deep declines during treatment and prospectively tested decision rules, will be more important than any one assay for future progress in the integration of ctDNA and nanotechnology for neoadjuvant monitoring. Rapid clearance of ctDNA has been shown to be repeatedly associated with good outcomes and persistent ctDNA reflects the presence of higher risk biology, establishing the biological plausibility for response directed escalation and de‑escalation strategies.⁴³ Multi‑omics should help to stabilize low abundance signals and to enhance predictive modeling.¹⁰⁰ In the meantime, theranostic nanoplatforms and chip based biosensing offer real-time or near-real-time monitoring. However, majority of such platforms are in early translational stages with clinical readiness limited by reference standards, reproducibility, regulatory evidence and workflow integration.¹⁰¹

The near term future of ctDNA in the neoadjuvant setting is not just prognostication but response‑adaptive therapy, with early molecular kinetics (clearance vs persistence) serving as a trigger to modify therapy prior to surgery. A complementary hypothesis is that early clearance may be a marker of de‑escalation in patients, while persistent ctDNA may identify patients who are molecular nonresponders, who may benefit from intensification or early switching, and studies have summarized a considerable body of evidence suggesting that on‑treatment ctDNA dynamics predict pathological response and long‑term endpoints in many different solid tumors.⁴³ An important translational take‑away is that neoadjuvant decision rules should be time‑anchored. Only if it is clinically meaningful before irreversible commitments are done, eg, if it is done before the completion of an ineffective regimen or before surgery is delayed is early clinically meaningful. This is directly reflected in studies on rectal cancer, where ctDNA is being tested as part of selection logic for organ preservation and tailoring of therapy, but caution is warranted because of the risk of false‑negative results with low‑shedding ctDNA scenarios, which could lead to de‑escalation.¹⁰² Small neoadjuvant colorectal studies have also suggested that ctDNA conversion may also be a marker to guide the length of treatment, and that this should be tested prospectively rather than inferred from the data.¹⁰³

In future, it will be crucial to develop platforms involving nanoparticles that allow the enrichment, amplification and detection of ctDNA in a single closed and clinically practical system. MNPs in particular are of great promise since they have the ability to capture rare fragments, amplify them locally and enable rapid electrochemical readout. Portable and inexpensive monitoring could be achieved via nanostructured electrochemical sensors that are based on AuNPs, graphene-related materials and nanozymes; highly sensitive, multiplexed mutation analysis could be accomplished using plasmonic AuNPs and surface-enhanced Raman scattering probes.²⁶ Other kinds of emerging two-dimensional materials, such as MBene nanosheets combined with paper-based microfluidics, also offer a possible path to point-of-care testing.¹⁰⁴

Future platforms will need to make improvements in batch-to-batch uniformity of nanoparticles, probe orientation, antifouling, recovery of ctDNA, and long-term stability of reagents and discrimination of low-frequency variants from background cfDNA. These systems should be assessed with serial samples before, after early chemotherapy and prior to surgery for neoadjuvant chemotherapy. Finally, multicentre prospective trials are needed to confirm the feasibility of obtaining nanoparticle-assisted ctDNA within clinically relevant timeframes and to ensure it is safe to support escalation or de-escalation or early surgical planning.²⁶ Additionally, the future direction will be to establish ctDNA checkpoints that correspond to the typical NACT timing, and then randomize patients with kinetics that cross a defined boundary, such as persistently positive at an early checkpoint, into escalation vs standard arms. This is in line with the clinical‑utility focus of ESMO guidance which differentiates between analytical validity and evidence that a particular action taken based on a result leads to better outcomes.⁹⁵

Conclusion

NACT has now evolved as a major strategy in all solid tumours as it can downstage the disease, treat the micrometastases early and provide an in vivo assessment of sensitivity. However, there are limitations in the clinical tools currently available for NACT monitoring, including imaging and post-operative pathological analysis, which are not optimal for real time decision making. Imaging may be slower and may not reflect the molecular response, and pathology is inherently delayed and does not reflect MRD completely. ctDNA has unique potential for monitoring NACT because its short half life and accessibility can allow for near real time monitoring of tumor dynamics, with early clearance correlating with treatment sensitivity. The persistence of ctDNA indicates the presence of persistent disease, resistance to treatment and increased risk of recurrence. Yet biology of ctDNA also poses its primary analytical challenges since its low abundance during effective therapy, fragmentation and background cfDNA requires very low limits of detection to reach 0.01% VAF ranges for MRD applications. Nanotechnology provides a viable pathway to achieve these requirements, with a benefit of increased sensitivity and workflow. The importance of nanotechnology platforms is not only that they make increasingly low analytical detection limits possible. The most promise is to produce clinically relevant data at a time point where a treatment can be assessed or adjusted. Residual disease during NACT might be associated with ctDNA persistence or increase, while early clearance or reductions might be a sign of favourable molecular and pathological response. There are several other complementary capabilities such as MNPs, AuNP-based sensors, fluorescent nanoprobes, QDs and carbon-based nanomaterials, which also vary in their clinical readiness. The nano-enabled CRISPR platforms introduce programmable specificity and scalable signal amplification, both of which suggest realistic model development to monitor kinetics of ctDNA on the same day that could be used to inform adaptive therapy and MRD-informed post operative decisions.

However translation of nanotechnology integrated ctDNA platform depends on overcoming challenges such as preanalytical variability, inter lab reproducibility and incomplete assay standardization. Furthermore, the regulatory and commercialization pathways, reimbursement uncertainty, manufacturing limitations of nanomaterials as well as cost and equity issues need to be addressed. Future research should standardize the method of nanoparticle synthesis, blood collection, ctDNA processing, and reporting threshold and calibration of the assays. Direct comparisons of nanotechnology-assisted platforms with digital PCR and next-generation sequencing should be made in serial samples collected within the context of neoadjuvant prospective, multicentre studies. Interventional trials can then help clarify if intervention based on ctDNA persistence or clearance is safe and efficacy can be demonstrated in treatment selection, surgical planning, and long-term outcomes. Also, the integration of multi-omics with transparent computational models, the early involvement of regulatory agencies and robust scalable manufacturing and quality control will boost the potential of ctDNA integrated nanotechnology platforms for NACT.

Funding Statement

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Data Sharing Statement

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

Ethical Statement

Ethical statement is not applicable.

Disclosure

The author(s) report no conflicts of interest in this work.