Toward aptamer-enabled continuous health monitoring: challenges and opportunities

1 Introduction

Continuous health monitoring is rapidly transforming modern medicine, enabling a shift from episodic testing to real-time insight. Traditional diagnostic approaches use intermittent sampling, capturing only isolated physiological snapshots and often missing transient, but clinically meaningful biomarker fluctuations. These gaps obscure early disease transition, delay detection of acute events, and limit the ability to track therapeutic response with precision. Continuous monitoring provides real-time molecular and physiological data, enabling timely intervention and more personalized healthcare (Flynn et al., 2023; Moghimikandelousi et al., 2025; Tu et al., 2025).

The most successful example of continuous molecular monitoring is the continuous glucose monitor. It has transformed diabetes care by enabling real-time tracking of glucose levels, detection of hyperglycemic and hypoglycemic events, more informed insulin dosing, and improved long-term glycemic control and quality of life (Burckhardt et al., 2019; Iscoe et al., 2006). The success of glucose monitoring is built on several favorable features of glucose as a target. Glucose is present at relatively high concentrations, typically in the millimolar range, and its levels change over clinically meaningful time scales. In addition, glucose oxidase and glucose dehydrogenase provide robust enzyme-based recognition elements that can be readily coupled to electrochemical detection. Together, these features have enabled reliable wearable glucose sensors that can operate continuously in complex biological environments.

However, extending continuous monitoring beyond glucose remains much more difficult. Many important biomarkers, including hormones, neurotransmitters, cytokines, therapeutic drugs, cardiac markers, and inflammatory proteins, are present at much lower concentrations than glucose (Flynn et al., 2023; Donnelly et al., 2025). Many of these targets also do not have natural enzymes that can be directly used for selective detection. Affinity-based biosensors address this gap by converting specific molecular binding events into measurable signals. Several classes of affinity reagents have been explored, including (Chen et al., 2023) dies, protein receptors, molecularly imprinted polymers, and nucleic acid aptamers (Flynn et al., 2023; Tu et al., 2020). Among these, aptamers are especially attractive for continuous health monitoring.

Aptamers are short single-stranded DNA or RNA molecules that fold into defined structures and bind specific molecular targets. They are typically generated through Systematic Evolution of Ligands by Exponential Enrichment (SELEX), or in vitro selection (Ellington and Szostak, 1990; Tuerk and Gold, 1990). Compared with many protein-based recognition elements, aptamers can be chemically synthesized, modified at defined positions, reproduced with low batch-to-batch variation, and integrated with optical, electrochemical, and microfluidic sensing platforms (Qian et al., 2022; Ji et al., 2023). Their small size and nucleic acid structure also make them suitable for rational engineering, chemical stabilization, and structure-switching sensor designs. These properties have led to rapid progress in aptamer-based sensors for continuous and real-time monitoring. In a Web of Science Core Collection search conducted on 23 May 2026, using the topic terms “aptamer” and “continuous monitoring,” we identified 193 papers published since 2020, with 60 papers published from 2025 onward. This growth reflects increasing interest in aptamer-based platforms for real-time biomarker detection, including electrochemical aptamer-based sensors, wearable sensors, microneedle sensors, implantable devices, and sensors designed for biofluids such as blood, interstitial fluid, and sweat (Biswas et al., 2024; Ferguson et al., 2013; Wang et al., 2022; Reynoso et al., 2024). Several recent reviews have summarized advances in aptamer-based sensing and continuous monitoring (Li X. et al., 2025; Downs and Plaxco, 2022; Garg et al., 2026; Chamorro-Garcia et al., 2025; Cao et al., 2024; Li and Zhong, 2024).

Despite this progress, aptamer-enabled continuous monitoring still faces major challenges. In this Perspective, we focused on the challenges and opportunities for developing aptamers as practical biorecognition elements for continuous health monitoring. We examined key barriers that limit translation, including the shortage of sensor-ready aptamers, difficulties in signal transduction and device integration, limited biofluid compatibility, long-term stability, and the need for reversible binding and regeneration. We also highlighted future directions, including improved selection strategies, kinetic engineering, chemical modification, and standardized validation under sensor-relevant conditions.

2 Challenges and opportunities

2.1 Need for sensor-ready aptamers

We define a sensor-ready aptamer as an aptamer that not only binds its target with the required affinity and specificity but also retains reliable performance under sensing conditions (Figure 1A). For health monitoring, this requires target recognition in physiologically relevant media, within clinically relevant concentration ranges, and in the presence of competing biomolecules.

Since the development of SELEX in 1990, thousands of aptamers have been reported for diverse targets, including metal ions, small molecules, proteins, and cells (Qian et al., 2022). However, many clinically important biomarkers still lack well-validated aptamers, including cytokines, hormones, cardiac markers, diabetes-related protein biomarkers, and small-molecule metabolites. One reason is that natural DNA and RNA have limited chemical diversity. They are built from four nucleobases and have a highly charged backbone, which can make it difficult to bind targets that are hydrophobic, highly charged, or lack strong chemical features. Modified or unnatural nucleotide libraries have been explored to address this limitation by expanding aptamer chemical diversity and improving recognition of difficult targets (Shen et al., 2020). However, modified libraries also require compatible polymerases, amplification workflows, and validation methods to ensure that improved chemical diversity translates into practical sensor performance.

The selection process itself can limit the discovery of sensor-ready aptamers. Traditional SELEX enriches sequences through repeated rounds of binding, partitioning, and amplification. Although this strategy has produced many useful aptamers, it can also introduce selection artifacts. The target used during selection may not reflect its native state. For example, target immobilization can alter target conformation, expose nonnative binding surfaces, or favor sequences that recognize an artificial target presentation (Le et al., 2026). Sequences may also bind nonspecifically to solid supports, beads, membranes, or other materials used during partitioning. Selection buffer conditions can further influence enrichment. For example, insufficient salt or blocking agents can promote nonspecific electrostatic interactions between negatively charged nucleic acids and positively charged targets. Ideally, selection conditions should reflect the intended sensing environment as closely as possible. Counter-selection should also be performed against relevant interfering targets and matrix components from the future sensing environment, rather than using arbitrary counter-selection targets. PCR amplification can further bias enrichment by favoring sequences that amplify efficiently rather than sequences with the strongest target recognition, while nonfunctional or parasitic sequences can accumulate during later rounds.

Incomplete post-selection validation creates another gap between reported aptamers and sensor-ready aptamers. Many aptamers are characterized only in simplified buffers or against purified targets. However, aptamer performance can change substantially in biological fluids, where ionic strength, pH, temperature, protein content, and nonspecific interactions can affect folding and target recognition. Some studies also report target binding without quantitative affinity measurements, kinetic analysis, or specificity tests against related molecules. Binding alone is therefore not sufficient to establish an aptamer as a useful recognition element for sensing.

Post-selection optimization is also important but often incomplete. Aptamers identified from SELEX are often relatively long (>60 nt), usually because they contain primer-binding regions and sequence segments that often do not directly contribute to target binding (Qian et al., 2022). For sensor development, these regions often need to be removed to identify the core binding domain and improve folding, stability, specificity, and signal generation. The core sequence can then be further optimized through truncation, mutation analysis, chemical modification, or re-selection using partially randomized sequences. For example, a recent study on a SARS-CoV-2 spike protein aptamer showed that re-selection and sequence optimization within a 40-nt core region improved binding affinity by approximately 100-fold, shifting affinity from the nanomolar to the picomolar range (Wang et al., 2025). Without this type of optimization, target binding alone may not be enough to make an aptamer suitable for sensor integration.

Future aptamer discovery should therefore move beyond identifying binders and focus on generating aptamers that are ready for sensing applications. Selection and validation workflows should include physiological buffers, relevant temperatures, clinically relevant concentration ranges, competing biomolecules, and complex biofluids. When possible, selection should also mimic the final sensor format, for example, by incorporating structure-switching designs or reporter-modified aptamer constructs (Yang et al., 2023; Yoshikawa et al., 2023; Nutiu and Li, 2005). More effort should also be placed on post-selection validation and optimization, including quantitative affinity and kinetic measurements, specificity testing, truncation, and mutation analysis. Advances in capture-SELEX (Yang et al., 2023), microfluidic SELEX (Chang et al., 2023), flow-cell-based SELEX (Yoshikawa et al., 2022; He et al., 2026), modified nucleotide libraries (Ji et al., 2023), and machine-learning-assisted analysis (Nature, 2026) provide new opportunities to reduce selection artifacts and identify aptamers that function under realistic sensing conditions.

2.2 Aptamer integration into sensors

Even when a high-quality aptamer is available, converting it into a working sensor is not straightforward (Figure 1B). For continuous monitoring, the aptamer often needs to operate in a reagent-free or minimally reagent-dependent format, remain functional after immobilization, and generate a measurable signal upon target binding.

A major challenge is signal transduction. Many aptamer-based continuous sensors rely on target-induced structural changes to convert binding into a measurable signal. Electrochemical aptamer-based (EAB) sensors are among the most widely explored platforms for continuous monitoring, and a recent clinical study reported their use in a wearable sensor for vancomycin monitoring (Booth et al., 2026; Arroyo-Currás et al., 2017). In EAB sensors, target binding changes the position of a redox reporter relative to the electrode surface, producing an electrochemical signal. Fluorescent aptamer sensors use a similar concept, in which target binding changes the distance between a fluorophore and quencher (Hariri et al., 2024; Gopalan et al., 2026). Although these platforms use different readout mechanisms, they both require the aptamer to couple target recognition to a measurable physical change. However, not all aptamers undergo a useful conformational change upon binding. Some aptamers bind through local contacts or small structural rearrangements that do not produce a predictable change in distance, charge distribution, fluorescence, or electron transfer. Therefore, structure-switching behavior often requires engineering through stem design, reporter placement, or sequence modification (Nutiu and Li, 2003; Feagin et al., 2018).

Other sensing formats can detect aptamer-target binding without relying on large structure switching. For example, field-effect transistor (FET)-based sensors can detect changes in local charge or surface potential near the sensor surface, and force- or motion-based designs can convert molecular binding into changes in molecular dynamics (Wang et al., 2022; Nakatsuka et al., 2018; Wu et al., 2025). These approaches may expand the range of aptamers that can be used for sensing because they do not always require a large target-induced conformational change. However, they introduce a different challenge: the sensor must distinguish specific aptamer-target binding from nonspecific adsorption and matrix-induced background changes. Structure-switching aptamers can provide an additional layer of specificity because signal generation depends on a target-induced conformational change. In contrast, nonspecific adsorption may change the local surface environment but is less likely to produce the same defined structural response. Therefore, binding-based sensors require careful surface passivation, reference channels, and controls to separate true target binding from nonspecific signals.

Reporter and linker design can further affect aptamer performance. Fluorophores, quenchers, redox reporters, nanoparticles, or other labels are commonly added to convert binding into a measurable signal. These modifications can alter aptamer folding, block target access, or reduce affinity and specificity (Li A. Z. et al., 2025). As a result, sensor development often requires iterative optimization of the aptamer sequence, reporter position, linker length, and readout format.

Surface immobilization adds another major constraint. Many aptamer sensors require attachment to an electrode, optical surface, nanoparticle, membrane, hydrogel, microneedle, or microfluidic interface (Reynoso et al., 2024; Li X. et al., 2025; Qin et al., 2024). Immobilization can restrict aptamer folding, reduce target accessibility, and change binding kinetics (Liu et al., 2021). Probe density also strongly affects performance. Low surface density can limit signal intensity, while high surface density can cause steric crowding, electrostatic repulsion, and reduced target access. Surface chemistry, spacer length, aptamer orientation, electrode material, and local ionic environment can all influence sensor response. Thus, an aptamer that performs well in solution may not perform well after immobilization on a device surface.

Different sensor platforms impose different design requirements. Electrochemical sensors require efficient electron transfer and appropriate reporter motion near the electrode. Optical sensors often require large conformational changes or optimized fluorophore–quencher distances. FET-based sensors are affected by charge screening and the distance between the binding event and the sensor surface. Microneedle and wearable sensors must maintain aptamer function during mechanical deformation, tissue contact, and exposure to complex biofluids. Microfluidic sensors introduce constraints related to flow, mass transport, residence time, and sample handling. These platform-specific requirements mean that aptamer integration cannot be treated as a simple final step after aptamer selection.

A key opportunity is to develop aptamer sensors through co-design of the aptamer, reporter, linker, surface, and device architecture. Instead of selecting an aptamer first and adapting it to a device later, future workflows should incorporate signal transduction and device compatibility earlier in the aptamer selection and engineering process. This could include selecting aptamers in immobilized formats, screening reporter-modified constructs, testing structure-switching designs during selection, and validating binding under surface-confined and flow conditions. These strategies would help identify aptamers that not only bind their targets but also generate reliable signals in the final sensor format.

2.3 Binding kinetics, reversibility, and regeneration

Binding kinetics are important for continuous monitoring (Figure 1C). In most aptamer studies, performance is mainly evaluated by equilibrium affinity or K _d. However, a low K _d alone does not ensure good sensor performance. A continuous sensor must respond when biomarker concentrations increase and decrease. If an aptamer-target complex dissociates too slowly, the sensor may show slow recovery, signal saturation, or poor temporal resolution (Flynn et al., 2025). In this case, strong binding can become a limitation rather than an advantage. At the same time, fast dissociation is not always ideal. Aptamers with rapid binding and release can support real-time sensing, but they may also have weaker apparent affinity and reduced sensitivity. This is especially important for low-abundance biomarkers, where the sensor must detect small concentration changes against background noise. For higher-abundance biomarkers, faster and more reversible aptamers may be sufficient. For low-concentration biomarkers, continuous sensing may require a balance between affinity, association rate, dissociation rate, signal gain, and background suppression.

Regeneration provides another route to maintain sensor responsiveness. In principle, regeneration can reset the aptamer sensor after target binding and allow repeated measurements over time. This can reduce the impact of slow target dissociation, but only if the regeneration process is efficient and compatible with the device. Many common regeneration methods, such as extreme temperature, extreme pH, or harsh chemical rinsing, are not suitable for wearable or implantable continuous monitoring devices. These conditions can damage the aptamer, disrupt the sensor surface, alter the biological sample, or reduce device lifetime. Therefore, regeneration strategies for continuous health monitoring must be mild, repeatable, and compatible with the final device environment. Recent electrochemical and optical approaches have begun to address this challenge (Zhang et al., 2018; Zargartalebi et al., 2024).

Future development should optimize binding affinity, binding kinetics, target dynamic range, and regeneration efficiency together. For continuous monitoring, the most useful aptamer is not necessarily the tightest binder, but the one whose kinetic profile matches the biomarker concentration range, required measurement frequency, and operating lifetime of the sensor.

2.4 Long-term stability, biofouling, and nonspecific binding

Long-term operation in biological environments remains a major challenge for aptamer-based continuous sensors (Figure 1D, left panel). Unmodified DNA and RNA aptamers can be degraded by nucleases in biological fluids, reducing the number of functional aptamers on the sensor surface and causing signal loss over time. Many unmodified aptamers have limited in vivo stability, often with usable lifetimes less than 24 h (Leung et al., 2021; Son et al., 2026), which restricts their use for longer-term health monitoring.

Biofouling and nonspecific binding create additional challenges (Figure 1D, right panel). Proteins, cells, lipids, and extracellular matrix components can adsorb to the sensor surface, block target access, change local surface properties, and generate false signals. Because aptamers have a negatively charged phosphate backbone, they may also interact nonspecifically with nucleic acid-binding proteins, positively charged molecules, or charged surface domains (Shi et al., 2023). These effects are especially important for continuous monitoring because small changes at the sensor interface can accumulate over time and appear as a baseline drift. Signal loss can also result from aptamer desorption, reporter instability, or changes in the local chemical environment. Matrix effects further complicate performance because blood, interstitial fluid, sweat, saliva, and urine differ in pH, ionic strength, protein content, and interfering molecules, all of which can affect aptamer folding, target binding, mass transport, and signal output.

Addressing these challenges will require both molecular stabilization and interface protection. Chemical modifications such as 2′-fluoro, 2′-O-methyl, LNA, XNA, spiegelmer designs, and circularization can improve nuclease resistance and extend aptamer lifetime (Ji et al., 2023; Leung et al., 2024). For example, a recent XNA-based study extended aptamer sensor operation in mice from typically half a day to 1 week (Son et al., 2026). At the device level, antifouling coatings, hydrogels, protective membranes, nanostructured electrodes, microfluidic barriers, and sample filtration can reduce fouling and drift (Wu et al., 2023; Chen et al., 2025; Trowbridge et al., 2025; Sen et al., 2026; Riordan et al., 2025). Calibration strategies and reference channels may also be needed to correct background changes during long-term operation.

2.5 Other challenges and opportunities for translation

Beyond aptamer design, sensor integration, binding kinetics, and long-term stability, several broader challenges remain for translating aptamer-based continuous sensors. Performance is often difficult to compare across studies because different groups use different buffers, temperatures, biofluid matrices, assay formats, and reporting metrics. Many studies report affinity but not association and dissociation rates, sensor drift, operational lifetime, or reproducibility across devices. Continuous sensors may also require calibration to account for patient-to-patient differences, biofluid composition, tissue transport, and device placement. Multiplexed sensing, data interpretation, manufacturing reproducibility, storage stability, and regulatory requirements will also become increasingly important as the field moves from proof-of-concept devices toward practical health-monitoring systems. Addressing these challenges will require shared reporting standards, clinically relevant validation models, robust calibration methods, and closer integration between aptamer engineering, device fabrication, data analysis, and clinical testing.

3 Conclusion

Aptamers have shown strong potential for continuous health monitoring. To realize this potential, the field needs to move from aptamers that simply bind targets to aptamers that work reliably in real sensors. Future aptamers should be validated in relevant biofluids, generate reliable signals, have suitable binding kinetics and dynamic range, and remain stable during long-term use. Addressing these challenges will be necessary before aptamer-based continuous sensors can move broadly beyond proof-of-concept studies. Nevertheless, as summarized in Table 1, with their wide target range, programmability, chemical flexibility, and compatibility with miniaturized devices, aptamers remain among the most promising recognition elements for next-generation continuous health monitoring.

TABLE 1. Comparing the viability of aptamers versus antibodies for use in continuous monitors.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Vuslat Juska, Northwestern University, United States

Reviewed by: Muhammad Sheraz, Northwestern Polytechnical University, China