Advantages
- Ease of Use: Quantitative evaluation of metabolic memory with only one blood sample, making it possible to use the method in actual clinical practice.
- Early Diagnosis: Even for diabetic patients who are diagnosed for the first time, the hyperglycemia accumulated in the body in the past can be estimated, enabling early prediction of the risk of complications.
Technology Overview & Background
In diabetics and other patients, prolonged hyperglycemia causes increased oxidative stress in various cells and tissues in the body, resulting in damage and degeneration of biomolecules. These changes at the cellular level are called “metabolic memory” and are said to continue to increase the risk of diabetes-related complications over the long term, since they are stored in the body without being completely reversed even if blood glucose levels are later restored to normal. This implies the importance of initial treatment upon diagnosis of diabetes, and it is believed that the effects of metabolic memory can be minimized through appropriate early blood glucose control. In addition, early risk prediction and appropriate initial treatment can be expected to reduce the risk of long-term complications in diabetic patients, so understanding metabolic memory in the treatment of diabetes is important.
The researchers’ group has previously devised a calculation model for estimating metabolic memory based on the transition of HbA1c levels over time in subjects/patients over the past 10 years. That is, the quantitative relationship between the accumulation of hyperglycemia and the decline in renal function based on the area under the transition curve for the period when the HbA1c value in the subject exceeds the standard value of 6% (AUCHbA1c≥6%) is formulated by a nonlinear mixed effects model. However, this conventional method of evaluation requires information on detailed historical trends in HbA1c levels over a long period of time, making it impractical for clinical application.
Against this background, the researchers devised a new, simple method for quantitative evaluation of metabolic memory that utilizes the concentration of amino acids in the patient’s blood. Specifically, by measuring the concentration of amino acids in the patient’s blood and focusing on the weight ratio of each amino acid in a specific amino acid group, metabolic memory, which is the accumulation of past hyperglycemia, can be accurately predicted with only a single blood sample. The results can also be used to predict the risk of complications that may occur in the future.
Data
Using capillary electrophoresis/mass spectrometry (CE/MS) for blood amino acids, the researchers collected data from more than 180 patients with type 2 diabetes and confirmed that AUCHbA1c≥6% can be estimated by machine learning (neural network/multilayer perceptron) from amino acid weight ratio information.
Publication(s)
Currently in the pre-publication stage.
Patent(s)
A patent has been filed in Japan and is pending publication.
Principal Investigator & Academic Institution
Kentaro Oniki, PhD (Associate Professor, Kumamoto University, Japan)
Development Stage & Future Research Plans
Currently, the following are under investigation to improve the accuracy of estimation of metabolic memory for clinical use.
- The model equation is being improved to more accurately assess the relationship between the accumulation of the reference hyperglycemia state and diabetic complications. More specifically, the researchers are attempting to construct an accurate mathematical model by changing the weighting depending on the time of estimation of AUCHbA1c≥6%, etc., because it is assumed that the effect of metabolic memory is stored in the body, but that the effect gradually becomes weaker.
- In the estimation of AUCHbA1c≥6% using amino acids, they are reexamining the amino acids used for evaluation and selecting the best predictor for more accurate estimation.
Expectations
TECH MANAGE is now looking for companies that are interested in licensing this invention for commercialization of their products/services, etc. on behalf of Kumamoto University. It is possible to have a direct meeting with the researchers regarding the invention/project.
In addition, it is possible to disclose unpublished data, according to the non-disclosure agreement with Kumamoto University, so please consider this as well. It is also possible for companies to consider joint research on this topic or paid evaluation (MTA contract/preferential negotiation rights and other options).
Project. JT-04984