The Effects of CGM and Connected Pen in T2DM Treated With Multiple Daily Insulin Injections

The aim of the study is to evaluate the effect of a combination of 12 months treatment with CGM (Freestyle Libre 2) with hypo and hyper alert in combination with a smart pen for bolus insulin (NOVOpen6) on HbA1c and time in range (TIR) in T2DM patients with baseline HbA1c \>53 mmol/mol (7. 0%, eAG 8,6 mmol/l) as compared with standard care.

This study is a 12 months open-label randomized controlled trial. Subjects with basal/bolus insulin treated T2DM and HbA1c \>53 mmol/mol are randomized to: A (control): standard care with disposable insulin pens and fingerprick glucose measurement or B (intervention): Novopen6 for bolus insulin and glucose measurement with Freestyle Libre 2. Due to the nature of the treatments it is not possible to perform a blinded study. Randomization in ratio 1:1 will be formed electronically in a REDCap database. The investigators estimate that the mean HbA1c at baseline is 60 mmol/mol. Given a standard deviation of the difference of approximately 8.1 mmol/mol (based on in house data from 114 patients with T2DM and insulin), the investigators calculate that 47 patients are needed in each arm to detect a minimal relevant difference of 5 mmol/mol with a type I error of 5% and a type II error 15%. Allowing for 10 % dropouts, the investigators aim to recruit 52 patients in each arm. If an interim analysis 3 months after randomizing 40 patients suggest a higher drop-out rate, the study size will be increased accordingly to a drop-out rate of max. 20 % (56 in each arm). Study plan Pre-screening of eligible patients was performed by assessing data from the healthcare administrative electronic software platform Business Intelligence (BI) of the Central Denmark Region. In this portal, data on medication, lab results, time period for contact to outpatient clinic for persons followed in secondary healthcare are automatically registered by data capture. Data can be retrieved from the BI portal via access to an encrypted version of the civil registration numbers allocated to all Danish citizens. The investigators collected data on prescribed pharmacological treatment for all persons with Type 2 diabetes who have had a minimum of one contact to the study centers. The electronic patient records for potential study candidates were reviewed for in- and exclusion criteria. Patients still considered study candidates after pre-screening were contacted by e-mail and invited to a meeting. Visit 1 (- 4 weeks): Inclusion and exclusion criteria are reviewed. Orally information of the study plan provided with reflection-time up to 24 hours. Planning fasting blood sample, blinded CGM (Freestyle Libre Pro) is applied, medication reviewed. Questionnaires answered. Visit 2 (-2 weeks): Fasting blood samples, Libre Pro removed and uploaded. Visit 3 (0 months): Randomization if inclusion criteria are full-filled and no exclusion criteria. Otherwise considered a screening failure. Randomization to either: A (control): continues treatment with usual insulin pens and self measurements of blood glucose. B (intervention): changes to NovoPen 6 for bolus insulin and CGM (Abbott Freestyle Libre 2). Patients randomized to intervention will participate in a structured education programme about the clinical use of Libre 2 and Novopen 6. This also includes assistance to upload LibreLink app. on the mobile phone, to upload Novopen 6 data and to establish a Libre View account. The patients fill in the diabetes health related questionnaires. Visit 4 Intervention group only (1 month): Telephone call (by study nurse) or optional personal meeting as necessary. Technical support only as needed for the upload of Freestyle Libre 2 and Novopen6 to LibreView. Visit 5 (11½ months): Blinded Freestyle Libre Pro is applied to all subjects (arm A and B). Visit 6 (12 months): Blood sample for HbA1c. Discontinuation and upload of Freestyle Libre Pro IQ (Libre view) and blood glucose-meter (Glooko)) for the last 14 days for both the control and intervention arm. In addition upload of Libre 2 and Novopen6 data (Libre view) for the last 4 weeks is performed for the intervention arm. Fill in of diabetes health related questionnaires and use of the memory function of Novopen 6 In-between study visits patients follow standard care in terms of number of visits (physical or telemedicine), type of health care provider (nurse or doctor), measurement of HbA1c and other clinical activities at the health care providers discretion, to enhance extern validity. The glucose data is evaluated by the usual health care providers together with the patient in order to optimize insulin therapy, including an increase in the number of meals with bolus insulin if necessary. According to the study design all patients are treated with bolus insulin for at least one meal. Bolus insulin should be insulin aspart. If patients are treated with insulin aspart manufactured by other companies than Novo Nordisk, the patients are asked to change to the biosimilar product from Novo. This strategy ensures that all patients can undergo randomization, since only Novo products fits into the Novopen6. Fast acting insulin is provided to all the patients in the study period free of charge. Patients randomized to the intervention group B receives Novopen6 free of charge. Patients are included irrespective of the brand of basal insulin. Any other glucose lowering drug (oral or injection) can be added or discontinued during the study at the investigators discretion.