Multiplexed biosensors toward smart therapeutic drug management of antibiotics

Abstract: The increasing threat of antibiotic resistance jeopardizes the effectiveness of antibiotics, necessitating a re-evaluation of our approach to antibiotic utilization. The rate at which newly developed agents become ineffective is outpacing our ability to innovate, creating a critical bottleneck. Central to successful antibiotherapy is the necessity to keep antibiotic concentrations within personalized therapeutic ranges, precisely tailored to accommodate the unique pharmacokinetics/pharmacodynamics of individual patients. The therapeutic range represents the optimal concentration of a drug in the body to achieve the desired therapeutic effect while minimizing side effects. Currently, these therapeutic ranges are established based on generic models derived from experiments on animals and healthy individuals, leading to a one-size-fits-all approach. The inherent challenge lies in the intricate task of ensuring drug concentrations remain within the desired window through conventional dosage regimens. This generic approach often results in suboptimal drug levels, potentially leading to treatment failure or, conversely, increased toxicity.
Therapeutic drug monitoring (TDM) emerges as a potential solution, offering real-time assessment of patient drug concentrations. However, traditional TDM, reliant on single-point analysis and lacking routine monitoring, perpetuates the one-size-fits-all model. This limitation not only impedes the prompt adjustment of drug concentrations but also contributes to subtherapeutic conditions, fostering antibiotic resistance. Furthermore, therapeutic studies predominantly focus on blood-based analysis due to its extensive database, the practical challenges of collecting, transporting, processing, and analysing blood for personalized TDM underscore the need for alternative approaches. Recent advancements in blood-based studies aim to address these challenges by reducing sample volumes and eliminating the reliance on costly equipment and specialized expertise. Conversely, non-invasive matrices such as breath, interstitial fluid, tears, saliva, or sweat offer promising alternatives to invasive TDM. However, challenges arise from the complex transport mechanisms of antibiotics from blood to these sampling sites, necessitating cross-correlation studies.
This thesis addresses the gap of tailoring antibiotic treatment dynamics to suit individual patient needs to alleviate antibiotic resistance. A tailored multiplexed biosensor has been developed to realize the concept of continuous therapeutic drug management, ensuring rapid and accurate measurement of antibiotics. This biosensor embodies essential characteristics, including the capacity for frequent measurements, the ability to detect subtle changes in drug concentration, and the delivery of sensitive and consistent results. Design parameters have been structured to prioritize simplicity, user-friendliness, affordability, rapid sample-to-result time, and compactness. A crucial feature is its multiplexing capability, enabling simultaneous measurement of multiple analytes and/or multiple samples. This allows concurrent monitoring of various biomarkers or drugs, while its multi-sample analysis functionality facilitates the creation of a comprehensive cross-correlation database, incorporating both well-established blood-based (plasma) and less-explored non-invasive samples.
The first phase of the work involves developing technology that enables simultaneous multianalyte/sample measurements on the same chip. The current biosensor design does not allow simultaneous measurement and can only be used in serial mode, which increases both the sample-to-result time and the number of critical electrical and fluidic components. To come up with a robust design that can process multiple fluids at the same time with a minimum number of components, a new microfluidic sensor concept is introduced, consisting of multiple incubation areas along with their electrochemical cells separated by a stopping barrier in a single channel. In the design process, microfluidic channel configurations with up to eight incubation areas are tested with different readout strategies, the number of analytes/samples they can process, channel layouts, and the number of inlets/outlets. Designs are evaluated based on their flow characteristics (flow rate, bubble formation) and signal generation reproducibility between individual channels and chips. Three successful designs are further evaluated for their ease of use and ergonomics and their integration into the final point-of-care device. At the end, a four-channel vertical alignment is selected for TDM application
During the second phase, a bioassay for ß-lactam antibiotics, the most commonly prescribed antibiotic class in the clinical settings, is developed. This process includes the optimization of parameters such as assay component concentrations, incubation times, and blocking strategies. The on-chip integration of this bioassay facilitates the generation of a calibration curve, exhibiting a sensitivity of 56 ng/ml , a level unattainable through conventional chromatography-based methods. Subsequently, the system is tested within an animal study involving Landrace pigs subjected to under-, normal, and over-dosed ß lactam antibiotics. Biofluids including whole blood, plasma, urine, saliva, and exhaled breath condensate (EBC) are analysed for the measurement and tracking of time-dependent antibiotic concentration changes. The detection and temporal monitoring of ß-lactams in EBC are also demonstrated for the first time. The platform's performance is benchmarked against high-performance liquid chromatography measurements, revealing a substantial agreement in the measured drug concentrations.
In the last phase of the study, developed biosensor technology is further improved for point-of-care applications via redesigning the bioassay. Additionally, the introduction of an NFC-potentiostat and a microperistaltic pump is implemented to miniaturize the system. The enhanced assay, with a superior sensitivity of 176 pg/ml , is then tested for another ß-lactam antibiotic to successfully quantify the impact of lung injury on drug clearance. Finally, the potential of the proposed therapeutic drug monitoring concept is examined using data-driven methods and machine learning. The database for this investigation is sourced from a clinical trial involving 13 hospitals in Germany, where half of the patients underwent TDM. A state-space approach is proposed to quantify individual patient status based on a reference healthy state distribution. The results, for the first time, quantify the positive impact of dosage adjustment with TDM on patient recovery in a clinical study.
In summary, this thesis focuses on designing and fabricating a robust electrochemical microfluidic multiplexed biosensor for quantifying ß-lactam antibiotics. It contributes to advancements in sensor design, the development of a novel bioassay, and the conceptualization of TDM, aiming to make the technology more user-friendly for on-site antibiotherapy management while exploring data-driven methods for TDM in clinical settings. Future work involves deploying the developed sensing technology in a randomized, controlled clinical study to identify the impact of frequent concentration measurements on therapy efficacy. A subsequent advancement includes integrating the developed bioassay into a wearable format, coupled with a machine learning-assisted closed-loop drug delivery system

Object Details

• Location: Deutsche Nationalbibliothek Frankfurt am Main
• Extent: Online-Ressource
• Language: Englisch
• Notes: Universität Freiburg, Dissertation, 2024

Classification and Topics

• Keyword: Biosensor ; Therapie ; Antibiotikum

Contributors, Places and Time

• Event: Veröffentlichung
• (where): Freiburg
• (who): Universität
• (when): 2024
• Creator: Ates, Hatice Ceren
• Contributor: Dincer, Can ; Zengerle, Roland ; Weber, Wilfried ; Albert-Ludwigs-Universität Freiburg. Institut für Mikrosystemtechnik ; Albert-Ludwigs-Universität Freiburg. Fakultät für Angewandte Wissenschaften

Further information

• DOI: 10.6094/UNIFR/255465
• URN: urn:nbn:de:bsz:25-freidok-2554657
• Rights: Open Access; Der Zugriff auf das Objekt ist unbeschränkt möglich.
• Last update: 25.03.2025, 1:46 PM CET