Objective
The overall objective of this proposal is to develop a molecular diagnostic device that can continuously measure insulin in the intestinal fluid and demonstrate its accuracy and reliability prior to investigating its applicability to AP systems. To accomplish this, we will generate a fully reagentless and self-regenerating biomolecular detection system. This technology will be developed where all required elements are incorporated into a self-contained sensor modified with a biorecognition element (nanobody) that binds specifically to insulin to then produce an electrochemical signal. We will test this technology on the benchtop to ensure it tracks insulin levels accurately. We will demonstrate the limit of detection (LOD) < [biomarker]min in synthetic interstitial fluid (in vitro) and the ability of our system to quantify changes in insulin levels within a physiological dynamic range. Next, we will prototype a filamentary microneedle microwire (FMM) device which allows for the integration of multiple sensors and can accommodate simultaneous monitoring of multiple biomarkers if desired. In our study, the FMM will perform in vitro and in vivo studies to determine the feasibility of this sensing system and assess biocompatibility. These results will be benchmarked against gold standard assays. Real-time insulin data upon insulin challenges will be obtained from T1D rats and healthy rats for at least 6 hours, and the data will be validated with standard ELISA. In addition to this data, we will also collect continuous glucose measurements with a commercially available CGM. These data sets will then be used to build a machine learning model to accurately identify non-insulin drivers of glucose excursions. This feature will improve interpretability and allows for separation of pathological dysregulation from physiological variation, a critical step in precision diabetes care. If the sensors accurately track changes in insulin in these studies, then they will be ready for integration into an artificial pancreas system, where they will need to be tested for the purpose of controlling glucose levels more accurately based on that information.
Background Rationale
Insulin replacement therapy, along with advances in continuous glucose monitoring and insulin delivery technologies that are being used in AP systems, has revolutionized the treatment of diabetes. However, people with T1D still experience adverse outcomes from incorrect insulin dosing due to heterogenous individual insulin metabolism. In the single-hormone AP systems, the insulin pump can dose insulin in response to hyperglycemia but has no way to directly address a predicted low blood sugar other than lowering or stopping insulin delivery or suggesting external actions (such as eating carbs). Hence, the need for improved glycemic control gives rise to the necessity of a new technology for continuous monitoring of insulin concentrations in vivo.
Unlike glucose, which can be monitored continuously in the bloodstream, insulin levels are not yet accessible with a continuous monitor. Continuous detection of this protein hormone has provided a challenge due to the size of this molecule, the low concentration in plasma, and the selectivity against interferants in blood. Currently, insulin levels can only be tested in blood and require expensive laboratory equipment for analysis. Platforms that can measure multiple analytes in vivo at physiologically relevant concentrations in real-time without the use of sophisticated purification methods for plasma separation are not yet widely available. Herein, we aim to develop a first-of-its-kind molecular diagnostic platform that can continuously measure insulin in the interstitial fluid. Such system does not currently exist commercially or under development. This platform will enable a truly personalized approach to diabetes care.
Description of Project
People with type 1 diabetes (T1D) are unable to produce endogenous insulin to maintain their normal blood glucose levels (BGL). Therefore, they require frequent BGL measurements and multiple self-administrations of insulin injection every day to sustain their life. To tightly maintain a normal BGL (euglycemia), insulin dosing needs to be specific to each patient and can vary depending on many factors (e.g. physical activities and amount of food intake, etc.) The incorrect prediction of insulin dose can lead to deadly consequences. Currently, the most common modality of insulin administration is restricted to self-injection of using a syringe, pen or insulin pump, which inevitably limits the efficacy of insulin dose along with a potential risk of acute hypoglycemia. To this end, there have been ongoing efforts to develop a more advanced glucose-responsive insulin delivery system known as an artificial pancreas (AP). Although continued glucose monitoring systems significantly improved the performance of AP systems, glucose concentrations do not linearly correlate with insulin concentrations. The currently available single hormone AP systems do not take all types of dietary and lifestyle into consideration to avoid hyperglycemia or hypoglycemia. In these systems, the insulin pump can dose insulin in response to high blood sugars but has no way to directly address a predicted low blood sugar other than lowering or stopping insulin delivery. However, the lack of information regarding the concentration of circulating insulin already present in blood before the administration of insulin are still a challenge for these systems. This is even more challenging during exercise as exercise causes profound changes in insulin concentrations in T1D people. Therefore, to improve the performance of AP systems, this project aims to develop a real-time continuous monitoring technology measuring insulin concentration in a minimally invasive manner. This system integrates novel technology platforms including a microneedle device to access interstitial fluid (ISF) and a novel molecular pendulum-based sensor that changes conformation upon specific target binding and can regenerate sensors. The proposed device will solve the problem of incorrect insulin dosing and allow a more accurate diagnosis of people with T1D.
Anticipated Outcome
The proposed device will address the problem of incorrect insulin dosing in AP systems and allow a more accurate diagnosis of people with diabetes. This platform could also play a potentially crucial role in mapping the behavior of this hormone in vivo and provide a more comprehensive understanding of the highly complex interplay of insulin and glucose in diabetes. An immediate result of this project will be integrating a molecular sensing unit with a filamentary microneedle microwire (FMM) for monitoring insulin in the interstitial fluid noninvasively. Importantly, the micronnedle will be housed within an applicator which will provide structural integrity during insertion into the skin. Based on our previous works with molecular pendulum (MP) sensors, we expect that nanobodies will perform to achieve high sensitivity and selectivity for detection of insulin in interstitial fluid (ISF). Moreover, sensor regeneration with our active-reset strategy should reset the MP sensor within 1 minute thus providing exquisite resolution over several hours during in vivo experiments. We anticipate that these innocuous and painlessly worn microneedle diagnostic device will effectively monitor insulin levels in a diabetic rat model after an insulin challenge. Fundamentally, this project will result in robust technologies that enable sensor integration with microfabricated microneedle form factors for physiological biomarker monitoring.
Relevance to T1D
In the people with T1D, the level of insulin can change unexpectedly according to their daily life activities such as exercising, having meals, sleeping, and etc. Currently there is not a good way to continuously track insulin levels. Here, we plan to develop an in-vivo real-time, continuous insulin monitoring system and will demonstrate its operability in a T1D rat model. The introduced device will solve the problem of incorrect insulin dosing for artificial pancreas systems and allow a more accurate diagnosis of people with T1D. This platform is uniquely positioned to provide a more comprehensive understanding of the highly complex interplay between insulin and glucose levels in T1D patients. Understanding the complexity of insulin kinetics would enable further insights into how this hormone affects target organs in the body.