Objective
The goal of this project is to advance the development of cell-based therapies for people living with Type 1 Diabetes (T1D). In T1D, the body’s immune system mistakenly destroys insulin-producing cells in the pancreas, forcing patients to rely on daily insulin injections to control their blood sugar. A promising alternative is to transplant insulin-producing cells, made from stem cells, to provide patients with the ability to naturally control their own blood sugar levels.
Current methods for creating these stem cell-derived islets still face major challenges. A critical problem is that artificial islets don’t yet produce insulin as effectively as natural islets from donors. There is also significant variability among the cells: some cells work well, while many others do not, making it difficult to predict how effective a batch of artificial islets will be.
This project addresses these issues by developing advanced imaging and artificial intelligence (AI) tools to quickly and accurately measure the insulin-producing capabilities of individual cells and islets. Our approach involves isolating single cells and islets into tiny containers (nanowells), along with specialized beads that capture insulin and other hormones as they are secreted. We then apply a fluorescent dye that causes these beads to glow under a microscope, allowing us to directly visualize and measure how much hormone each cell or islet releases.
Using the microscope images collected, we will train an AI algorithm to recognize subtle visual patterns linked to effective hormone production. Cells that release insulin, glucagon, or somatostatin (three hormones essential for regulating blood sugar) have characteristic shapes and internal structures. By learning these features, the AI will eventually predict how well individual cells or entire islets function based solely on their microscopic appearance, eliminating the need for slower, more complex tests.
Our integrated microscopy and AI platform provides a powerful and non-invasive way to continuously monitor the quality of islets made from stem cells throughout the manufacturing process. Because microscopic imaging is already a routine step in producing stem cell therapies, this new approach can seamlessly offer real-time feedback, allowing researchers and manufacturers to quickly identify and select the highest-quality islets for transplantation.
Ultimately, these tools will enable scientists to produce safer, more consistent, and effective cell therapies for T1D. By ensuring each batch of manufactured islets is robust and functional before transplantation, our research brings us closer to achieving a reliable, long-lasting treatment option—potentially reducing or even eliminating the need for daily insulin injections and significantly improving quality of life for people with Type 1 Diabetes.
Background Rationale
Type 1 Diabetes (T1D) is an autoimmune condition where the body’s own immune system mistakenly destroys specialized cell clusters in the pancreas, called islets, which produce insulin to control blood sugar levels. Without functioning islets, T1D patients must rely on daily insulin injections or insulin pump therapy to survive. Although life-saving, these treatments require constant monitoring and do not fully replicate the body’s natural control of blood sugar, often resulting in complications.
A potential cure for T1D is to transplant islets from deceased donors into T1D patients. These transplanted cells can produce insulin naturally, offering the possibility of reducing or even eliminating the need for daily insulin injections. Unfortunately, there is a critical shortage of donor organs, meaning that only a small fraction of those who could benefit from this therapy actually receive it.
To address this shortage, researchers have been exploring ways to produce islets from stem cells, which could create a limitless supply of artificial islets. These artificial islets have enormous potential, but currently, they do not perform as well as natural donor islets. Specifically, artificial islets produce lower levels of insulin and do not respond effectively to changing blood sugar levels, making them less reliable as a treatment option.
One significant barrier is that the methods used today to measure insulin secretion only assess the overall output from large groups of cells. As a result, researchers cannot detect important differences in how individual cells or single islets function. This limitation prevents scientists from fully understanding what makes certain artificial islets more effective than others, slowing the development of consistent, high-quality cell therapies.
To overcome these challenges, scientists urgently need better tools to evaluate insulin production at the level of single cells and individual artificial islets. Precise measurements at this scale would help researchers identify exactly why some artificial islets function well, while others do not. Improved testing methods would also allow for continuous monitoring of islet quality during manufacturing, ensuring that only highly functional artificial islets are selected for transplantation into patients.
This project aims to solve these critical challenges by combining advanced microscopic imaging techniques with powerful artificial intelligence (AI). Our innovative approach will enable rapid, accurate measurement of hormone secretion from individual stem cell-derived islets. By accurately linking detailed microscopic images with precise hormone secretion data, our method can identify the best-performing artificial islets quickly and reliably.
Ultimately, this research will greatly accelerate the development of reliable, high-quality artificial islets, bringing us closer to a world where people with T1D could rely on long-term cell-based therapies instead of daily insulin injections. This advance could dramatically improve the lives and health outcomes of millions living with Type 1 Diabetes.
Description of Project
This project aims to improve treatment options for people with Type 1 Diabetes (T1D) by developing innovative technologies to enhance the effectiveness of cell-based therapies. In T1D, the immune system mistakenly destroys specialized insulin-producing cell clusters in the pancreas, called islets. Without insulin, the body cannot properly regular blood sugar, requiring individuals with T1D to depend on daily insulin injections. A potential cure for T1D is to transplant islet cells from recently deceased donors into people with diabetes to help them produce their own insulin. However, there are an insufficient number of donors to meet the demand of all patients in need.
Scientists are currently developing ways to produce a limitless supply of insulin secreting cells from stem cells that are grouped together to form artificial islets that could be transplanted into diabetes patients. These artificial islets currently do not have the same level of function as donor islets because they release less insulin compared to donor islets.
Our project aims to overcome these challenges by creating advanced microscope-based tests combined with artificial intelligence (AI) technology to evaluate and improve the function of these artificial islets. Specifically, we will develop methods that allow us to measure precisely how much insulin and other critical hormones (like glucagon and somatostatin) individual cells and islets release. To do this, single cells and islets are placed into tiny containers, called nanowells, along with microscopic beads designed to capture hormones secreted by the cells. When cells produce hormones, the beads capture them, and we apply a fluorescent dye that causes the beads to glow under a microscope. By measuring this glow, we can accurately quantify how well each individual cell or islet functions.
Next, we will use AI methods to analyze microscope images of these cells and islets. The AI system will be trained to recognize subtle visual features that are associated with high levels of hormone secretion. Because insulin-producing cells and islets have distinctive shapes and internal structures, the AI can learn to identify the best-performing cells from images alone—without requiring lengthy chemical tests.
The same approach will be used to evaluate larger clusters of cells, known as islets, to determine how their structure and composition affect their ability to regulate blood sugar effectively. Our technology will help scientists quickly identify the healthiest, most functional stem cell-derived islets and track their quality during manufacturing. Imaging with microscopes is already standard during cell production, so integrating our AI-based evaluation method can provide immediate, real-time feedback on the quality and performance of each batch of islets.
Overall, these advanced imaging and AI-based technologies have the potential to transform the way cell therapies for T1D are developed, tested, and produced. By quickly identifying the most promising stem cell-derived islets, we can help ensure that patients receive the highest-quality treatment. This innovative approach brings us closer to a future where individuals with Type 1 Diabetes might rely on long-lasting cell therapies instead of daily insulin injections, greatly improving their quality of life.
Anticipated Outcome
This project will create new, powerful technologies to measure how effectively stem cell derived insulin-producing cells and artificial islets function, which will provide essential tools for advancing cell therapies for Type 1 Diabetes (T1D). By leveraging advanced microscopy combined with artificial intelligence (AI), we will significantly improve the ability to assess insulin secretion at the individual cell and islet level, something not possible with existing approaches.
Specifically, the project will produce an innovative microscopy-based test that precisely measures insulin secretion from thousands of single cells or individual islets. Our method involves placing cells or islets into extremely small containers, called nanowells, along with microscopic beads that capture insulin and other hormones as they are released. Using a special fluorescent dye, these beads will glow under a microscope when hormones bind to them, allowing us to directly and accurately quantify hormone secretion from each cell or islet. Unlike current testing methods, which only measure the average hormone production of large batches of cells, this new assay provides detailed, cell-level data, enabling researchers to pinpoint exactly which cells or islets function best.
Additionally, our project will develop an AI-based system that quickly analyzes microscope images to determine the quality of cells and islets based solely on their appearance. By training the AI to recognize subtle visual features associated with effective hormone secretion, it can accurately predict how well cells or islets will function without performing complex biochemical tests. This tool will enable rapid, non-invasive assessment of large numbers of cells, dramatically accelerating research and quality control in manufacturing.
The outcome of this project can potentially transform how artificial islets are developed and manufactured. Researchers and manufacturers can quickly identify the healthiest and most functional cells and islets, significantly improving the quality and consistency of cell therapy products. Ultimately, these innovations have the potential to speed the development of reliable, long-lasting cell-based therapies for T1D, moving us closer to a future where people with Type 1 Diabetes can rely on artificial islets instead of daily insulin injections.
Relevance to T1D
This project has the potential to make transformative advances in cell therapy for Type 1 Diabetes (T1D) by providing new tools to support development of more effective and reliable cell therapies. Currently, T1D is treated with lifelong insulin injections, but this treatment must be carefully matched to blood glucose levels to avoid serious complications, such as low blood sugar or long-term organ damage. Cell therapies offer a potential cure for T1D by transplanting insulin-secreting pancreatic islets into affected individuals. These transplanted islets can monitor changes in blood sugar levels and secrete insulin as needed, so that the transplant recipient no longer needs regular insulin injections.
A major limitation to the widespread use of cell therapies for T1D is that pancreatic islets must be collected from deceased donors, and there are not enough donor islets available to meet the demand for transplantation. Artificial islets produced from stem cells could potentially serve as a limitless supply of transplant islets. However, artificial islets produced to date secrete insulin at much lower levels compared to donor islets. Significant efforts are currently underway to develop artificial islets that better mimic the function of natural pancreatic islets. However, these efforts are hampered by the fact that current methods for measuring islet function can only measure insulin secretion on a large mass of islets and cannot measure the insulin secretion from individual islets.
This project will advance T1D cell therapies by developing a reliable method to measure insulin secretion from individual cells and islets. Unlike current methods, which cannot distinguish between whether insulin secretion comes from a few high-performing islets or a larger number of moderately secreting ones, our proposed approach will rapidly measure insulin output from thousands of individual cells or individual islets. Furthermore, our platform allows for the simultaneous measurement of other pancreatic hormones in order to develop islet cell therapies that better mimic complex glucose regulation achieved by natural islets.
This new approach provides researchers with detailed, single-cell-level data to help them better understand how artificial islets function. It will also allow scientists and manufacturers to carefully monitor the quality of stem-cell-derived islets during every step of the production process, ensuring only the most functional cells are selected for transplantation into patients.
Ultimately, this project directly supports the goal of creating reliable cell therapies that closely mimic natural islet function. For people with T1D, this means potentially reducing or even eliminating the daily burden of insulin injections, lowering the risk of complications, and dramatically improving their overall quality of life.