WO2024133412A1 - Methods and systems for estimating fbg value from cgm data - Google Patents

Methods and systems for estimating fbg value from cgm data Download PDF

Info

Publication number
WO2024133412A1
WO2024133412A1 PCT/EP2023/086873 EP2023086873W WO2024133412A1 WO 2024133412 A1 WO2024133412 A1 WO 2024133412A1 EP 2023086873 W EP2023086873 W EP 2023086873W WO 2024133412 A1 WO2024133412 A1 WO 2024133412A1
Authority
WO
WIPO (PCT)
Prior art keywords
fbg
percentile
data set
given
value
Prior art date
Application number
PCT/EP2023/086873
Other languages
French (fr)
Inventor
Iman Hajizadeh
Dimitri BOIROUX
Mikel Murphy GOMES
Anuar IMANBAYEV
Henrik Bengtsson
Original Assignee
Novo Nordisk A/S
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Novo Nordisk A/S filed Critical Novo Nordisk A/S
Publication of WO2024133412A1 publication Critical patent/WO2024133412A1/en

Links

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
    • G16H20/10—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to drugs or medications, e.g. for ensuring correct administration to patients
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
    • G16H20/10—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to drugs or medications, e.g. for ensuring correct administration to patients
    • G16H20/17—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to drugs or medications, e.g. for ensuring correct administration to patients delivered via infusion or injection
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
    • G16H20/60—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to nutrition control, e.g. diets
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients

Definitions

  • the present disclosure generally relates to systems and methods for assisting patients and health care practitioners in managing insulin treatment to diabetics.
  • the present invention relates to systems and methods suitable for use in a diabetes management system in which a fasting blood glucose value FBG is to be determined for a patient in treatment with an insulin.
  • Diabetes mellitus is impaired insulin secretion and variable degrees of peripheral insulin resistance leading to hyperglycaemia.
  • Type 2 diabetes mellitus is characterized by progressive disruption of normal physiologic insulin secretion.
  • basal insulin secretion by pancreatic p cells occurs continuously to maintain steady glucose levels for extended periods between meals.
  • prandial secretion in which insulin is rapidly released in an initial first-phase spike in response to a meal, followed by prolonged insulin secretion that returns to basal levels after 2-3 hours.
  • Years of poorly controlled hyperglycaemia can lead to multiple health complications. Diabetes mellitus is one of the major causes of premature morbidity and mortality throughout the world.
  • the preferred titration method is the 3-FBG method, according to which the FBG (Fasting Blood Glucose, being a measurement of the blood glucose concentration after 8 hours of fasting) is measured on titration day and the prior two days.
  • the measurements are provided as SMBG (Self-Monitoring of Blood Glucose) values and should preferably be in a target range between 4.4 mmol/L - 7.2mmol/L.
  • SMBG Self-Monitoring of Blood Glucose
  • US 2021/187075 addresses treatment of paediatric type 2 diabetes (T2D) and discloses a clinical study for the treatment of pediatrics with T2D using lixisenatide at different doses. Percentiles are used as an inclusion/exclusion criterion for the clinical trials or to group patients based on their demographic information and are not used to adjust the dose.
  • T2D paediatric type 2 diabetes
  • US 2022/386965 describes a method to classify continuous glucose monitoring (CGM) data sets using some measure of similarity: (i) system is provided for developing a model to classify CGM data, (ii) the system matches two CGM profiles based on their shapes, designates them as a pair, and transforms them into a motif, and (iii) the system labels the motif based on a clinical characteristic and repeats the process until a finite set of motifs is created. Thus, CGM characteristics are used for classification, and are not used for dose adjustment.
  • CGM continuous glucose monitoring
  • WO 2022/036214 discloses a method for modeling the glucose homeostasis of a patient with type 1 diabetes (T1 D).
  • the method reconstructs data supporting a glucose time series for the patient, and then personalizes the model and generates a variability control (VC) signal.
  • the VC signal accounts for insulin sensitivity and allows the patient to learn the effect of adjusting the data.
  • T1 D type 1 diabetes
  • the method is not used for dose adjustment.
  • US 10,943,675 is concerned with methods for altering patient care and increasing the efficacy of medical devices, and discloses in an example on how to assess the risk of T2D or the need for therapy intensification based on yearly HbA1c measurements and by measuring the variation between year n-1 and year n.
  • the methods and examples do not involve CGM data.
  • CGM devices can measure glucose levels continuously (e.g., 5-minutes sampling periods), for the entire day.
  • APS automated insulin delivery systems called artificial pancreas systems
  • CGM devices have proven to be highly effective for patients managing T1 D.
  • Use of CGMs for management of T2D has been supported in multiple studies published in JAMA, e.g. Martens, Thomas, et al. "Effect of continuous glucose monitoring on glycaemic control in patients with type 2 diabetes treated with basal insulin: a randomized clinical trial.” JAMA 325.22 (2021): 2262-2272.
  • CGMs have been demonstrated for patients with T2D who use basal insulin. Specific benefits included A1C lowering, fewer episodes of hypo-glycemia (and hyper-glycemia), as well as the potential to reinforce the value of diet and lifestyle modifications.
  • CGM devices measure blood glucose values frequently, it is not straightforward to estimate FBG from the measured data.
  • CGM measured data resolution with 5-minutes sampling time equals 288 values per day as opposed to previously only 1 FBG value and perhaps a few calibrations or correction Blood Glucose Monitoring (BGM) values throughout a single day.
  • BGM Blood Glucose Monitoring
  • an object of the present invention to provide methods and systems allowing an equitable FBG value to be determined from a plurality of CGM measuring points, e.g., up to 288 points per 24 hours.
  • Such an algorithmic derivation of an applicable FBG value from a CGM data resolution trace may e.g. be used for a once-weekly Icodec insulin titration protocol.
  • a system for determining an identifier allowing a fasting blood glucose (FBG) value to be estimated from a CGM data set for a subject from a given demographic and for a given drug treatment regimen comprising one or more processors and a memory, the memory comprising instructions that, when executed by the one or more processors, performs a method responsive to receiving a request for the determination of an identifier.
  • FBG fasting blood glucose
  • the method comprises the steps of (i) obtaining a plurality of CGM data sets from a plurality of subjects and for a plurality of time periods, each subject belonging to the given demographic and being in treatment according to the given drug regimen, each CGM data set comprising a plurality of BG values determined over a time period, (ii) determining for each CGM data set a series of individual percentile BG values, (iii) obtaining from each subject and for each time period a corresponding FBG data set comprising P FBG values, P being at least one, (iv) calculating an P-FBG value as the mean FBG for each FBG data set, (v) for each corresponding data set calculating an individual difference between a given individual percentile BG value and the corresponding P-FBG value, (vi) for each percentile calculating a global mean difference for the plurality of individual differences, and (vii) determining a global percentile as the percentile having the minimum global mean difference.
  • the determined global percentile represents the identifier allowing
  • a series of individual percentile BG values is determined, this covers that a series of percentiles is determined, e.g. the series may comprise the 1 st percentile, the 2 nd percentile, the 3 rd percentile etc., or the 5 th percentile, the 10 th percentile, the 15 th percentile etc., where the X th percentile BG value is the BG value where X percent of the BG values falls below it.
  • Mean (a-b) Mean (a) - Mean (b).
  • the term corresponding refers to the same period of time, e.g. the same week for obtaining the two sets of data.
  • P is at least 2.
  • P may be 3 allowing a traditional 3- FBG value to be calculated.
  • the time period may be a week of 7 days, and the number P of FBG values may be determined on the same weekdays in each FBG data set.
  • a system for estimating a fasting blood glucose (FBG) value for a subject from a given demographic and for a given drug treatment regimen comprising one or more processors and a memory.
  • the memory comprises instructions that, when executed by the one or more processors, perform a method responsive to receiving a request for an estimated FBG value.
  • the method comprises the steps of (i) obtaining a CGM data set from the subject for a given time period, the CGM data set comprising a plurality of BG values determined over the time period, (ii) obtaining an identifier percentile representing the given demographic, the given drug treatment regimen, and the given time period, and (iii) estimating the FBG value as the identifier percentile of the CGM data set.
  • the identifier percentile may be obtained using the system of described above.
  • the basic calculations allowing the identifier to be determined can also be calculated alternatively by (i) obtaining a plurality of CGM data sets from a plurality of subjects and for a plurality of time periods, each subject belonging to the given demographic and being in treatment according to the given drug regimen, each CGM data set comprising a plurality of BG values determined over a time period, (ii) calculating for each CGM data set a series of individual percentile BG values, (iii) calculating a series of global percentile BG values based on the plurality of percentile BG values, (iv) obtaining from each subject and for each time period a FBG data set comprising P FBG values, P being at least one, (v) calculating an P-FBG value as the mean FBG for each FBG data set, (vi) calculating a global P-FBG value based on the plurality of P-FBG values, and (vii) determining the global percentile BG value closest to the global P-FBG value.
  • the global percentile BG values may be calculated as the mean value of the plurality of percentile BG values.
  • P may be at least 2, the global P-FBG value being calculated as the mean of the plurality of P-FBG values.
  • the present invention and aspects thereof provide a number of advantages over the SoC 3- FBG approach for the individual user.
  • no more finger prick measurements may be required since some CGMs are factory-calibrated and do not need any calibration, thus reducing the burden on the user.
  • some CGMs need to be calibrated either initially or daily.
  • the fasting time might not be right before breakfast (e.g.
  • fig. 1 shows in a diagram the mean difference between FBGs and the x th percentile.
  • the dashed lines show the 95% confidence interval
  • fig. 2 shows a histogram depicting differences between a plurality of 3 FBGs values and the corresponding 20 th percentile values from a study over 16 weeks with 100 subjects.
  • the present invention relates to an algorithm adapted to estimate a fasting blood glucose (FBG) value for a patient to be used in calculating a recommended dose of an insulin.
  • FBG fasting blood glucose
  • the algorithm of the present invention may be used as a stand-alone solution providing a user with information about his/her FBG, however, the algorithm may also be used as part of an overall diabetes dose guidance system that helps people with diabetes by generating recommended insulin doses based on estimated FPG values.
  • a given algorithm is used to generate recommended insulin doses and treatment advice for diabetes patients based on BG data, insulin dosing history and, in more advanced applications, other factors like meals, physical activity, stress, illness etc. may be taken into consideration.
  • such a system comprises a back-end engine (“the engine”) used in combination with an interacting system in the form of a client and an operating system.
  • the client from the engine’s perspective is the software component that requests dose guidance.
  • the client gathers the necessary data (e.g. FPG values) and requests dose guidance from the engine.
  • the client then receives the response from the engine.
  • the engine may run directly as an app on a given user’s smartphone or on a blood glucose meter (BGM) and thus be a self-contained application comprising both the client and the engine.
  • the system setup may be designed to be implemented as a back-end engine adapted to be used as part of a cloud-based large-scale diabetes management system.
  • a cloud-based system would allow the engine to always be up to date (in contrast to app-based systems running entirely on e.g., the patient’s smartphone), would allow advanced methods such as machine learning and artificial intelligence to be implemented, and would allow data to be used in combination with other services in a greater “digital health” setup.
  • Such a cloud-based system ideally would handle a large amount of patient requests for dose recommendations.
  • a “complete” engine may be designed to be responsible for all computing aspects, it may be desirable to divide the engine into a local and a cloud version to allow the patient-near day-to-day part of the dose guidance system to run independently of any reliance upon cloud computing. For example, when the user via the client app makes a request for dose guidance the request is transmitted to the engine which will return a dose recommendation. In case cloud access is not available the client app would run a dose-recommendation calculation using local data. Dependent upon the user’s app-settings the user may or may not be informed.
  • a set-up for determining an identifier from a given demographic and for a given drug treatment regimen is provided, this subsequently allowing a fasting blood glucose (FBG) value to be estimated from a CGM data set for a subject from a given demographic and for a given drug treatment regimen.
  • FBG fasting blood glucose
  • the invention is based on the concept that a plurality of subjects belonging to a given demographic follow a given drug treatment for a plurality of time periods. For each time period each subject provides (i) a CGM data set comprising a plurality of BG values, and (ii) a FBG data set comprising P FBG values, P being at least one, this allowing a percentile (or “identifier”) to be determined providing an estimated FBG value based on a CGM data set for a period of time for a subject from the given demographic and for the given drug treatment regimen. More specifically, for each CGM data set a series of individual percentile BG values can be calculated. Based on the plurality of percentile BG values a series of global percentile BG values can be calculated.
  • a P-FBG value as the mean FBG is calculated.
  • a global P-FBG value can be calculated.
  • the global percentile BG value closest to the global P-FBG value can be determined, the determined global percentile representing the identifier.
  • the CGM data is generated by N subjects participating in an M weeks long clinical trial for a given drug, e.g. Icodec insulin (a once-weekly insulin from Novo Nordisk, Denmark).
  • Icodec insulin a once-weekly insulin from Novo Nordisk, Denmark.
  • P FBGs values are provided by each subject, each measured on a different day. It is the object to identify the percentile of one-week CGM values which is an accurate representative of a P-FBG (i.e. mean of P FBGs), measured in that week.
  • Next step is to calculate the difference between (N x M) Mean (P FBGs) values and (N x M) X th percentile values based on different X percentile values (1 st percentile, 2 nd percentile, ... , 20 th percentile, ... , 50 th percentile). This step is necessary to choose the best X percentile value that has the minimum difference values with the Mean (P FBGs) values.
  • fig. 1 shows the mean difference between the FBGs and the X th percentile, the figure suggesting that the 20 th percentile is nearly equal to the FBGs.
  • the dashed lines show the 95% confidence interval.
  • the X percentile value having the minimum difference value with the Mean (P FBGs) value is the 20 th percentile.
  • an FBG value for a subject from a corresponding demographic and for a corresponding drug treatment regimen can be estimated as the 20 th percentile of the CGM data set for a corresponding period, e.g. a 3-FBG value based on a week of CGM data.
  • the Y axes unit is mmol/L.
  • 1 is calculated as the mean of these 1.600 differences and the dashed lines are the mean ⁇ 1.96 * STD where STD is the standard deviation of these 1.600 differences.
  • the 1.96 * STD coefficient provides an estimate of the 95% confidence interval assuming the 1 ,600 differences follow a normal distribution. For the further percentile values (1 st percentile, 2 nd percentile, ... , 20 th percentile, ... , 50 th percentile) the calculations are repeated allowing the mean, STD and 95% confidence interval to be calculated after computing the difference values.
  • the mean of the differences or mean (1.600 Mean (3 FBGs) values - 1.600 20 th percentile values) is zero with a 95% confidence interval of ⁇ 2.1 mmol/L (0 ⁇ 2.1).
  • the values are -1 .5 ⁇ 2.1 mmol/L.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Medical Informatics (AREA)
  • Primary Health Care (AREA)
  • General Health & Medical Sciences (AREA)
  • Epidemiology (AREA)
  • Chemical & Material Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Medicinal Chemistry (AREA)
  • Pathology (AREA)
  • Databases & Information Systems (AREA)
  • Biomedical Technology (AREA)
  • Nutrition Science (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

A method responsive to receiving a request for an estimated FBG value. The method com- prises the steps of (i) obtaining a CGM data set from a subject for a given time period, the CGM data set comprising a plurality of BG values determined over the time period, (ii) obtaining an identifier percentile representing the given demographic, the given drug treatment regimen, 5 and the given time period being representative of the subject, and (iii) estimating the FBG value as the identifier percentile of the CGM data set.

Description

METHODS AND SYSTEMS FOR ESTIMATING FBG VALUE FROM CGM DATA
The present disclosure generally relates to systems and methods for assisting patients and health care practitioners in managing insulin treatment to diabetics. In a specific aspect the present invention relates to systems and methods suitable for use in a diabetes management system in which a fasting blood glucose value FBG is to be determined for a patient in treatment with an insulin.
BACKGROUND
Diabetes mellitus (DM) is impaired insulin secretion and variable degrees of peripheral insulin resistance leading to hyperglycaemia. Type 2 diabetes mellitus is characterized by progressive disruption of normal physiologic insulin secretion. In healthy individuals, basal insulin secretion by pancreatic p cells occurs continuously to maintain steady glucose levels for extended periods between meals. Also in healthy individuals, there is prandial secretion in which insulin is rapidly released in an initial first-phase spike in response to a meal, followed by prolonged insulin secretion that returns to basal levels after 2-3 hours. Years of poorly controlled hyperglycaemia can lead to multiple health complications. Diabetes mellitus is one of the major causes of premature morbidity and mortality throughout the world.
People suffering from diabetes have been treated with insulin for more than a century. Diabetics are depending on insulin products to prolong their lives and increase their life quality. Injections of insulin are by far the most used method of diabetics to take their insulin. Different types of diabetes, different stages of each type and different life-styles and individual physiological differences, all contribute to complicate optimisation of treatment regimens and as a result modern treatment is based on highly personalised treatment regimens based on extensive research and development as well as carefully supervised introduction of treatment by titration.
Development of treatment is not only focussed on increasing the efficiency and prolonging the life of diabetics, but also on postponing and ultimately avoiding the serious adverse long-term effects of the condition, as well as making the treatment itself easier and less of a daily burden for diabetics.
It is known that the better the insulin management fit with the actual need, the better the efficiency and the longer adverse effects of the condition can be postponed. Furthermore, severe over- or under-dosing can be immediately life threatening and must be avoided. Thus, large fluctuations of insulin levels in the body should be limited to occur only when needed, which is most often related to ingestion of meals and exercising. The better the natural regulation of insulin in the body of a non-diabetic can be mimicked, the better, obviously. An obvious technical solution would be to measure the blood glucose level, determine the need for insulin, and inject the required amount at intervals of a few minutes. However, this would be extremely expensive and not improve the life quality of the diabetics, as it would become a full-time occupation to regulate blood glucose/insulin levels.
To decrease the number of required injections and thereby the interference with the diabetic’s life and routines, different types of insulin have been developed to allow treatment of different types of diabetes and to allow combination of products to optimise treatment regimens on an individual basis. In particular, products of long-lasting effect of slowly releasing insulin in the user’s body has been developed to provide a basic level of insulin, called basal treatment. This ensures insulin levels do not get too low during sleep and outside meal-situations.
As a new user is introduced to treatment with insulin, the user has to go through a period of titration. For insulins for use at fixed intervals, e.g. once daily or once weekly, the preferred titration method (also referred to as the Standard-of-Care (SoC) approach) is the 3-FBG method, according to which the FBG (Fasting Blood Glucose, being a measurement of the blood glucose concentration after 8 hours of fasting) is measured on titration day and the prior two days. The measurements are provided as SMBG (Self-Monitoring of Blood Glucose) values and should preferably be in a target range between 4.4 mmol/L - 7.2mmol/L. A mean value of these three measurements is used to avoid inefficient titration due to bad measurements and random variations in plasma Glucose variations caused by unusual activities of the user.
Thus, if Sunday is titration day for a new user of a once weekly insulin product, the user is required to obtain SMBG values Friday, Saturday and Sunday and calculate a mean value. If the value is above the target range, the dose should be increased and if it is below, the dose should be decreased. This process continues until the weekly measurements becomes stabile and within target range.
However, such measurements are uncomfortable for the users as they are not only required to prick a finger to obtain blood on which to measure, but also required to be fasting 8 hours each Friday, Saturday and Sunday. Furthermore, the current titration method and target range is a best compromise for a population based on a “one rule fits all” approach. As the BG values vary during the day, depending on the activities of the individual and the individual itself, it would be expected that users would experience increase stability of BG levels and better treatment outcomes if it was possible and feasible to provide a more personalised titration procedure and treatment.
US 2021/187075 addresses treatment of paediatric type 2 diabetes (T2D) and discloses a clinical study for the treatment of pediatrics with T2D using lixisenatide at different doses. Percentiles are used as an inclusion/exclusion criterion for the clinical trials or to group patients based on their demographic information and are not used to adjust the dose.
US 2022/386965 describes a method to classify continuous glucose monitoring (CGM) data sets using some measure of similarity: (i) system is provided for developing a model to classify CGM data, (ii) the system matches two CGM profiles based on their shapes, designates them as a pair, and transforms them into a motif, and (iii) the system labels the motif based on a clinical characteristic and repeats the process until a finite set of motifs is created. Thus, CGM characteristics are used for classification, and are not used for dose adjustment.
WO 2022/036214 discloses a method for modeling the glucose homeostasis of a patient with type 1 diabetes (T1 D). The method reconstructs data supporting a glucose time series for the patient, and then personalizes the model and generates a variability control (VC) signal. The VC signal accounts for insulin sensitivity and allows the patient to learn the effect of adjusting the data. Thus, a method for modeling the glucose homeostasis of a patient with T1 D is presented. The method is not used for dose adjustment.
US 10,943,675 is concerned with methods for altering patient care and increasing the efficacy of medical devices, and discloses in an example on how to assess the risk of T2D or the need for therapy intensification based on yearly HbA1c measurements and by measuring the variation between year n-1 and year n. The methods and examples do not involve CGM data.
As an alternative to SMBG it is possible to determine BG values using a CGM device. CGM devices can measure glucose levels continuously (e.g., 5-minutes sampling periods), for the entire day. Using CGM data in insulin titration algorithms as part of decision support systems, and automated insulin delivery systems called artificial pancreas systems (APS) have shown to be safe and effective. CGM devices have proven to be highly effective for patients managing T1 D. Use of CGMs for management of T2D has been supported in multiple studies published in JAMA, e.g. Martens, Thomas, et al. "Effect of continuous glucose monitoring on glycaemic control in patients with type 2 diabetes treated with basal insulin: a randomized clinical trial." JAMA 325.22 (2021): 2262-2272. The benefits of CGMs have been demonstrated for patients with T2D who use basal insulin. Specific benefits included A1C lowering, fewer episodes of hypo-glycemia (and hyper-glycemia), as well as the potential to reinforce the value of diet and lifestyle modifications.
Although CGM devices measure blood glucose values frequently, it is not straightforward to estimate FBG from the measured data. CGM measured data resolution with 5-minutes sampling time equals 288 values per day as opposed to previously only 1 FBG value and perhaps a few calibrations or correction Blood Glucose Monitoring (BGM) values throughout a single day.
Correspondingly, it is an object of the present invention to provide methods and systems allowing an equitable FBG value to be determined from a plurality of CGM measuring points, e.g., up to 288 points per 24 hours. Such an algorithmic derivation of an applicable FBG value from a CGM data resolution trace may e.g. be used for a once-weekly Icodec insulin titration protocol.
DISCLOSURE OF THE INVENTION
In the disclosure of the present invention, embodiments and aspects will be described which will address one or more of the above objects or which will address objects apparent from the below disclosure as well as from the description of exemplary embodiments.
In a first aspect of the invention a system for determining an identifier allowing a fasting blood glucose (FBG) value to be estimated from a CGM data set for a subject from a given demographic and for a given drug treatment regimen is provided, the system comprising one or more processors and a memory, the memory comprising instructions that, when executed by the one or more processors, performs a method responsive to receiving a request for the determination of an identifier. The method comprises the steps of (i) obtaining a plurality of CGM data sets from a plurality of subjects and for a plurality of time periods, each subject belonging to the given demographic and being in treatment according to the given drug regimen, each CGM data set comprising a plurality of BG values determined over a time period, (ii) determining for each CGM data set a series of individual percentile BG values, (iii) obtaining from each subject and for each time period a corresponding FBG data set comprising P FBG values, P being at least one, (iv) calculating an P-FBG value as the mean FBG for each FBG data set, (v) for each corresponding data set calculating an individual difference between a given individual percentile BG value and the corresponding P-FBG value, (vi) for each percentile calculating a global mean difference for the plurality of individual differences, and (vii) determining a global percentile as the percentile having the minimum global mean difference. In such a method the determined global percentile represents the identifier allowing a FBG value to be estimated from a CGM data set for a subject from the given demographic and for the given drug treatment regimen.
When it is defined that for each CGM data set a series of individual percentile BG values is determined, this covers that a series of percentiles is determined, e.g. the series may comprise the 1st percentile, the 2nd percentile, the 3rd percentile etc., or the 5th percentile, the 10th percentile, the 15th percentile etc., where the Xth percentile BG value is the BG value where X percent of the BG values falls below it.
As appears, the disclosed method comprises a calculation of the type Mean (a-b). Indeed, the defined method also covers equivalent calculations performed to reach the same result such as Mean (a-b) = Mean (a) - Mean (b). The term corresponding refers to the same period of time, e.g. the same week for obtaining the two sets of data.
By this method it is possible to determine an optimal CGM-derived BG percentile value for a subject from a given demographic and for a given drug treatment regimen, which percentile value can then be used to determine an individual CGM-derived FBG value for a person from the given demographic and for the given drug treatment regimen.
In an exemplary embodiment P is at least 2. For example, P may be 3 allowing a traditional 3- FBG value to be calculated.
The time period may be a week of 7 days, and the number P of FBG values may be determined on the same weekdays in each FBG data set.
In a second aspect of the invention a system for estimating a fasting blood glucose (FBG) value for a subject from a given demographic and for a given drug treatment regimen is provided, the system comprising one or more processors and a memory. The memory comprises instructions that, when executed by the one or more processors, perform a method responsive to receiving a request for an estimated FBG value. The method comprises the steps of (i) obtaining a CGM data set from the subject for a given time period, the CGM data set comprising a plurality of BG values determined over the time period, (ii) obtaining an identifier percentile representing the given demographic, the given drug treatment regimen, and the given time period, and (iii) estimating the FBG value as the identifier percentile of the CGM data set. The identifier percentile may be obtained using the system of described above.
As mentioned above, the basic calculations allowing the identifier to be determined can also be calculated alternatively by (i) obtaining a plurality of CGM data sets from a plurality of subjects and for a plurality of time periods, each subject belonging to the given demographic and being in treatment according to the given drug regimen, each CGM data set comprising a plurality of BG values determined over a time period, (ii) calculating for each CGM data set a series of individual percentile BG values, (iii) calculating a series of global percentile BG values based on the plurality of percentile BG values, (iv) obtaining from each subject and for each time period a FBG data set comprising P FBG values, P being at least one, (v) calculating an P-FBG value as the mean FBG for each FBG data set, (vi) calculating a global P-FBG value based on the plurality of P-FBG values, and (vii) determining the global percentile BG value closest to the global P-FBG value. In such a method the determined global percentile represents the identifier allowing a FBG value to be estimated from a CGM data set for a subject from the given demographic and for the given drug treatment regimen.
The global percentile BG values may be calculated as the mean value of the plurality of percentile BG values. P may be at least 2, the global P-FBG value being calculated as the mean of the plurality of P-FBG values.
The present invention and aspects thereof provide a number of advantages over the SoC 3- FBG approach for the individual user. Depending on the actual CGM used no more finger prick measurements may be required since some CGMs are factory-calibrated and do not need any calibration, thus reducing the burden on the user. This said, some CGMs need to be calibrated either initially or daily. Further, there is no need to take BG measurements prior to the titration day, as the percentiles are determined at the time of titration. In respect of safety, since all the data are available, it is possible to detect hypo-glycemia outside the usual measurements. In respect of robustness, the fasting time might not be right before breakfast (e.g. in case of night snacking, night shifts or long travel across time zones, these cases are handled by the percentile-based CGM titration concept. Further, using simple rules allow to build a universal titration algorithm, using either a CGM or a variable number of finger prick measurements for titration. The present invention and aspects thereof also provide a number of advantages over more advanced algorithms, such as model-based titration. For example, similar to 3-FBG, HCPs can easily compute the X-percentile for one week of CGM data, thus the decision-making process regarding manual changes in e.g. doses, target etc. would be similar. The titration process is based on simple rules thus it would be easy for HCPs to adapt these rules to the user. Since no model of the user is required, this approach is robust to inconsistent data, e.g. unannounced missed doses.
BRIEF DESCRIPTION OF THE DRAWINGS
In the following embodiments of the invention will be described with reference to the drawings, wherein: fig. 1 shows in a diagram the mean difference between FBGs and the xth percentile. The dashed lines show the 95% confidence interval, and fig. 2 shows a histogram depicting differences between a plurality of 3 FBGs values and the corresponding 20th percentile values from a study over 16 weeks with 100 subjects.
DESCRIPTION OF EXEMPLARY EMBODIMENTS
The present invention relates to an algorithm adapted to estimate a fasting blood glucose (FBG) value for a patient to be used in calculating a recommended dose of an insulin.
The algorithm of the present invention may be used as a stand-alone solution providing a user with information about his/her FBG, however, the algorithm may also be used as part of an overall diabetes dose guidance system that helps people with diabetes by generating recommended insulin doses based on estimated FPG values.
In such a system a given algorithm is used to generate recommended insulin doses and treatment advice for diabetes patients based on BG data, insulin dosing history and, in more advanced applications, other factors like meals, physical activity, stress, illness etc. may be taken into consideration.
Essentially, such a system comprises a back-end engine (“the engine”) used in combination with an interacting system in the form of a client and an operating system. The client from the engine’s perspective is the software component that requests dose guidance. The client gathers the necessary data (e.g. FPG values) and requests dose guidance from the engine. The client then receives the response from the engine.
On a small local scale, the engine may run directly as an app on a given user’s smartphone or on a blood glucose meter (BGM) and thus be a self-contained application comprising both the client and the engine. Alternatively, the system setup may be designed to be implemented as a back-end engine adapted to be used as part of a cloud-based large-scale diabetes management system. Such a cloud-based system would allow the engine to always be up to date (in contrast to app-based systems running entirely on e.g., the patient’s smartphone), would allow advanced methods such as machine learning and artificial intelligence to be implemented, and would allow data to be used in combination with other services in a greater “digital health” setup. Such a cloud-based system ideally would handle a large amount of patient requests for dose recommendations.
Although a “complete” engine may be designed to be responsible for all computing aspects, it may be desirable to divide the engine into a local and a cloud version to allow the patient-near day-to-day part of the dose guidance system to run independently of any reliance upon cloud computing. For example, when the user via the client app makes a request for dose guidance the request is transmitted to the engine which will return a dose recommendation. In case cloud access is not available the client app would run a dose-recommendation calculation using local data. Dependent upon the user’s app-settings the user may or may not be informed.
In a first aspect of the invention, a set-up for determining an identifier from a given demographic and for a given drug treatment regimen is provided, this subsequently allowing a fasting blood glucose (FBG) value to be estimated from a CGM data set for a subject from a given demographic and for a given drug treatment regimen.
The invention is based on the concept that a plurality of subjects belonging to a given demographic follow a given drug treatment for a plurality of time periods. For each time period each subject provides (i) a CGM data set comprising a plurality of BG values, and (ii) a FBG data set comprising P FBG values, P being at least one, this allowing a percentile (or “identifier”) to be determined providing an estimated FBG value based on a CGM data set for a period of time for a subject from the given demographic and for the given drug treatment regimen. More specifically, for each CGM data set a series of individual percentile BG values can be calculated. Based on the plurality of percentile BG values a series of global percentile BG values can be calculated. For each FBG data set a P-FBG value as the mean FBG is calculated. Based on the plurality of P-FBG values a global P-FBG value can be calculated. Finally, the global percentile BG value closest to the global P-FBG value can be determined, the determined global percentile representing the identifier.
Example
Task to be solved: Determine the percentile allowing a FBG value to be estimated from CGM data set for a subject from a given demographic and for a given drug treatment regimen based on a given number of subjects from the given demographic, for the given treatment regimen and for a given P FBG (P= 1 ,2,3... ). The CGM data is generated by N subjects participating in an M weeks long clinical trial for a given drug, e.g. Icodec insulin (a once-weekly insulin from Novo Nordisk, Denmark). For each week P FBGs values are provided by each subject, each measured on a different day. It is the object to identify the percentile of one-week CGM values which is an accurate representative of a P-FBG (i.e. mean of P FBGs), measured in that week.
First the mean of the P-FBGs for each M week and each subject is computed providing M P- FBGs, this resulting in N x M mean of P FBGs for all subjects.
Figure imgf000011_0001
For a CGM sampling rate of 5 minutes each day (ideally) generates (24 hours x 60 minutes) I 5 minutes = 288 samples. For each week, the Xth percentile of these 288 x 7 CGM values is computed providing for M weeks and N subjects M x N values of Xth percentile.
Figure imgf000012_0001
Next step is to calculate the difference between (N x M) Mean (P FBGs) values and (N x M) Xth percentile values based on different X percentile values (1st percentile, 2nd percentile, ... , 20th percentile, ... , 50th percentile). This step is necessary to choose the best X percentile value that has the minimum difference values with the Mean (P FBGs) values.
Corresponding to the above example fig. 1 shows the mean difference between the FBGs and the Xth percentile, the figure suggesting that the 20th percentile is nearly equal to the FBGs. The dashed lines show the 95% confidence interval.
In the example shown in fig. 1 the X percentile value having the minimum difference value with the Mean (P FBGs) value is the 20th percentile.
According to the second aspect of the present invention, an FBG value for a subject from a corresponding demographic and for a corresponding drug treatment regimen can be estimated as the 20th percentile of the CGM data set for a corresponding period, e.g. a 3-FBG value based on a week of CGM data. The histogram shown in fig. 2 is based on 1.600 difference values, i.e. the differences between 1.600 Mean (3 FBGs) values and 1.600 20th percentile values from a study over 16 weeks (M=16) with 100 subjects (N=100). The Y axes unit is mmol/L. Correspondingly, the central full line in fig. 1 is calculated as the mean of these 1.600 differences and the dashed lines are the mean ± 1.96 * STD where STD is the standard deviation of these 1.600 differences. The 1.96 * STD coefficient provides an estimate of the 95% confidence interval assuming the 1 ,600 differences follow a normal distribution. For the further percentile values (1st percentile, 2nd percentile, ... , 20th percentile, ... , 50th percentile) the calculations are repeated allowing the mean, STD and 95% confidence interval to be calculated after computing the difference values.
As shown in fig. 1 the mean of the differences or mean (1.600 Mean (3 FBGs) values - 1.600 20th percentile values) is zero with a 95% confidence interval of ~ 2.1 mmol/L (0 ± 2.1). Similarly for the 50th percentile the values are -1 .5 ± 2.1 mmol/L.
In the above description of exemplary embodiments, the different structures and means providing the described functionality for the different components have been described to a degree to which the concept of the present invention will be apparent to the skilled reader. The detailed construction and specification for the different components are considered the object of a normal design procedure performed by the skilled person along the lines set out in the present specification.
*****

Claims

1. A system for determining an identifier allowing a fasting blood glucose (FBG) value to be estimated from a continuous glucose monitoring (CGM) data set for a subject from a given demographic and for a given drug treatment regimen, the system comprising one or more processors and a memory, the memory comprising: instructions that, when executed by the one or more processors, perform a method responsive to receiving a request for an updated estimated FBG value, the method comprising the steps of: obtaining a plurality of CGM data sets from a plurality of subjects and for a plurality of time periods, each subject belonging to the given demographic and being in treatment according to the given drug regimen, each CGM data set comprising a plurality of BG values determined over a time period, determining for each CGM data set a series of individual percentile BG values, obtaining from each subject and for each time period a corresponding FBG data set comprising P FBG values, P being at least one, calculating a P-FBG value as the mean FBG for each FBG data set, for each corresponding data set calculating an individual difference between a given individual percentile BG value and the corresponding P-FBG value, for each percentile calculating a global mean difference for the plurality of individual differences, and determining a global percentile as the percentile having the minimum global mean difference, wherein the determined global percentile represents the identifier allowing a FBG value to be estimated from a CGM data set for a subject from the given demographic and for the given drug treatment regimen.
2. A system as in claim 1 , wherein P is at least 2.
3. A system as in claim 1 or 2, wherein: the time period is a week of 7 days, and the number P of FBG values is determined on the same weekday(s) in each FBG data set.
4. A system as in claim 3, wherein P=3.
5. A system for estimating a fasting blood glucose (FBG) value for a subject from a given demographic and for a given drug treatment regimen, the system comprising one or more processors and a memory, the memory comprising: instructions that, when executed by the one or more processors, perform a method responsive to receiving a request for an updated estimated FBG value, the method comprising the steps of: obtaining a continuous glucose monitoring (CGM) data set from the subject for a given time period, the CGM data set comprising a plurality of BG values determined over the time period, obtaining an identifier percentile representing the given demographic, the given drug treatment regimen, and the given time period, and estimating the FBG value as the identifier percentile of the CGM data set.
6. A system as in claim 5, wherein the identifier percentile is obtained using the system of any of claims 1-4.
7. A system as in claim 5 or 6, wherein the system is adapted to use the estimated FBG value to calculate a recommended dose of insulin for the subject being a patient in treatment with an insulin.
*****
PCT/EP2023/086873 2022-12-21 2023-12-20 Methods and systems for estimating fbg value from cgm data WO2024133412A1 (en)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US202263434102P 2022-12-21 2022-12-21
US63/434,102 2022-12-21
EP23157303 2023-02-17
EP23157303.1 2023-02-17

Publications (1)

Publication Number Publication Date
WO2024133412A1 true WO2024133412A1 (en) 2024-06-27

Family

ID=89507512

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/EP2023/086873 WO2024133412A1 (en) 2022-12-21 2023-12-20 Methods and systems for estimating fbg value from cgm data

Country Status (1)

Country Link
WO (1) WO2024133412A1 (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10943675B2 (en) 2017-07-28 2021-03-09 George S. Cembrowski Altering patient care based on long term SDD
US20210187075A1 (en) 2015-01-16 2021-06-24 Sanofi-Aventis Deutschland Gmbh Treatment of pediatric type 2 diabetes mellitus patients
WO2022036214A1 (en) 2020-08-14 2022-02-17 University Of Virginia Patent Foundation Method and system for generating a user tunable representation of glucose homeostasis in type 1 diabetes based on automated receipt of therapy profile data
US20220386965A1 (en) 2021-06-04 2022-12-08 University Of Virginia Patent Foundation Method for structuring and classification of continuous glucose monitoring (cgm) profiles

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20210187075A1 (en) 2015-01-16 2021-06-24 Sanofi-Aventis Deutschland Gmbh Treatment of pediatric type 2 diabetes mellitus patients
US10943675B2 (en) 2017-07-28 2021-03-09 George S. Cembrowski Altering patient care based on long term SDD
WO2022036214A1 (en) 2020-08-14 2022-02-17 University Of Virginia Patent Foundation Method and system for generating a user tunable representation of glucose homeostasis in type 1 diabetes based on automated receipt of therapy profile data
US20220386965A1 (en) 2021-06-04 2022-12-08 University Of Virginia Patent Foundation Method for structuring and classification of continuous glucose monitoring (cgm) profiles

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
MARTENS, THOMAS ET AL.: "Effect of continuous glucose monitoring on glycaemic control in patients with type 2 diabetes treated with basal insulin: a randomized clinical trial", JAMA, vol. 325.22, 2021, pages 2262 - 2272

Similar Documents

Publication Publication Date Title
Christiansen et al. Accuracy of a fourth-generation subcutaneous continuous glucose sensor
Shah et al. Performance of a factory-calibrated real-time continuous glucose monitoring system utilizing an automated sensor applicator
Bonora et al. Head-to-head comparison between flash and continuous glucose monitoring systems in outpatients with type 1 diabetes
Bailey et al. Accuracy and acceptability of the 6-day Enlite continuous subcutaneous glucose sensor
Harvey et al. Quest for the artificial pancreas: combining technology with treatment
US9962112B2 (en) Medical systems and methods of use
JP2007014751A (en) Method and device for evaluating a series of glucose concentration values concerning body fluid of diabetic in order to adjust insulin dosage
Forlenza et al. Factory-calibrated continuous glucose monitoring: how and why it works, and the dangers of reuse beyond approved duration of wear
US20200015760A1 (en) Method to determine individualized insulin sensitivity and optimal insulin dose by linear regression, and related systems
Benhamou et al. Real-time continuous glucose monitoring (CGM) integrated into the treatment of type 1 diabetes: consensus of experts from SFD, EVADIAC and SFE.
JPWO2020013230A5 (en)
US20150073754A1 (en) Method and system for calculating indices for diabetes control
Yeoh et al. A head‐to‐head comparison between Guardian Connect and FreeStyle Libre systems and an evaluation of user acceptability of sensors in patients with type 1 diabetes
Doyle et al. I. Glucose control strategies for treating type 1 diabetes mellitus
Tweden et al. Longitudinal analysis of real-world performance of an implantable continuous glucose sensor over multiple sensor insertion and removal cycles
Kapil et al. Artificial pancreas system for type 1 diabetes—challenges and advancements
Cescon et al. Linear modeling and prediction in diabetes physiology
Renard et al. Reduction of clinically important low glucose excursions with a long‐term implantable continuous glucose monitoring system in adults with type 1 diabetes prone to hypoglycaemia: the France Adoption Randomized Clinical Trial
WO2024133412A1 (en) Methods and systems for estimating fbg value from cgm data
JP5031762B2 (en) Analysis method and survey system
KR102028676B1 (en) A method, server and program for providing medical after case service
Nielsen et al. Glucose monitoring by microdialysis: performance in a multicentre study
US20220211306A1 (en) Glucose exposure diagnostics and therapeutics related thereto
Sun et al. Impact of errors in carbohydrate estimation on control of blood glucose in type 1 diabetes
Chlup et al. Effective assessment of diabetes control using personal glucometers (CONTOURLINK, Bayer, Germany; CALLA, Wellion, Austria; LINUS, Agamatrix, USA)