A critical hurdle associated with long term implantable devices is the foreign body response (FBR). Macroencapsulation devices are a promising therapy for the treatment of Diabetes Mellitus (DM) as they improve cell retention and viability. However they have limited success due to a complex cascade of events, causing fibrous encapsulation and ultimately device failure. In 2017, 425 million individuals worldwide were diagnosed with DM, Type 1 DM accounting for 5-10% of all cases. Here we propose the incorporation of soft robotics into a long term implantable device to modulate the host’s response to the implant.
Non-porous dynamic soft robotic devices (DSR) were implanted subcutaneously in a rat model. Two devices were implanted per animal and subjected to actuation or no actuation. At day 14 the devices were explanted en bloc and stained with 1% phosphomolybdic acid (PMA) for 2 weeks. Imaging was performed using a Scanco Mediucal MicroCT 100 and analysed using Mimics 18 materialise software. Thickness analysis was performed using volumetric reconstructions.
A significant reduction in the mean fibrotic capsule thickness was seen when the device was actuated compared to the control (p=0.0005).
The use of soft robotics can modulate the foreign body response in vivo. This technology can be harnessed to improve the success of long term implantable medical devices, including macroencapsulation devices for diabetes treatment.
Technosphere® Insulin (TI) is an inhaled insulin with a rapid rate of onset and short duration of action that has been shown to improve glycemic control in adult patients with diabetes mellitus (DM). Unfamiliarity with TI dosing has limited use of TI among prescribers and patients. BluHale is a digital diabetes platform capable of enhancing user and prescriber experience with TI. A cartridge color detector embedded within BluHale V2.0 monitors usage of TI. A proof-of-concept single-patient study involving BluHale V2.0 was performed to provide insights into TI usage.
A 63-year-old male patient with type 1 DM who had been using mealtime TI for 3 years utilized BluHale V2.0 during TI self-administration for 21 days. Blood glucose levels were monitored via continuous glucose monitoring (CGM) and overlaid with dose monitoring output from BluHale V2.0.
The volunteer self-administered TI while using BluHale V2.0 for 21 days, during which the average number of TI administrations was 6 per day. Rapid reductions in glucose levels immediately following administration were observed. A dose-recommendation algorithm was developed and retroactively applied to the BluHale/CGM data based on blood glucose levels. The data suggested the volunteer often administered TI at a dose or time that was late or when glucose levels were already out of range.
BluHale V2.0 technology provides the ability to visualize dosing information overlaid with CGM data. This technology has the potential to improve dosing behaviors and reveal nonoptimal dosing behaviors, allowing patients to optimize their time in range.
Digital technologies are transforming how patients with diabetes mellitus (DM) monitor and manage their disease. Technosphere® Insulin (TI) is an inhaled rapid-acting insulin that can improve glycemic control in adult patients with DM. BluHale V1.0 is a device that can be mounted on the TI inhaler to enhance patient training on proper inhalation technique. Trainers responsible for teaching inhalation technique to patients newly prescribed TI were surveyed to evaluate the utility of BluHale as a training tool.
Certified diabetes educators (CDEs) and medical assistants (MAs) involved in patient training for use of TI evaluated the usefulness of BluHale as a training device. Trainers assessed their level of confidence in their own training skills, before and after using the device, on a scale of 1 (not confident at all) to 5 (very confident). Additionally, trainers evaluated the level of confidence of the individuals they trained before and after the trainees used BluHale.
Forty-five trainers (CDEs, 49%; MAs, 51%) were surveyed. On average, trainers reported increased training confidence after they used BluHale (pre-use, 3.7; post-use, 4.6) and believed that trained individuals were more confident performing proper inhalation of TI after the trainees used BluHale (pre-use, 3.4; post-use, 4.7). Additionally, most trainers (80%) believed that BluHale use may reduce the time needed to train patients.
BluHale V1.0 is a useful tool that can enhance insulin administration training for patients prescribed TI, which may help patients be more successful when using TI.
SC insulin is non-physiologic since the pancreas normally releases insulin directly to the liver. Using a liver-targeting lipid moiety (Hepatic Directed Vesicles: HDV), Diasome technology enables subcutaneous insulin to be delivered to the liver, restoring more physiologic hepatic glucose balance. We assessed safety and efficacy of HDV-Lispro (HDV-L), compared to lispro alone (LIS), in type 1 diabetes subjects (T1DM).
The ISLE-1 (InSulin Liver Effect-1) a 26-week randomized (2:1 HDV-L:LIS) double-blind study, enrolled 176 generally healthy T1DM with baseline A1C 7.0-10.5%. Recent severe hypoglycemia and impaired hypoglycemia awareness were exclusionary. Primary study endpoint was A1C noninferiority, with multiple secondary endpoints.
141 subjects (HDV-L=98;LIS=58) completed the study. Overall, mean change in A1C from baseline to endpoint was -0.09% for HDV-L and -0.16% for LIS (CI -0.18 to 0.35, confirming noninferiority of HDV-L). Overall, no statistically significant differences in measures of hypoglycemia or insulin dosing were observed. Prespecified subgroups demonstrated an inverse relationship between baseline A1C and measures of hypoglycemia, confirmed by interaction analysis (p<0.001). High baseline A1C (>8.5%) had less hypoglycemia [e.g. 73% reduced time <54 mg/dL by continuous glucose monitoring (CGM)] while using 25% less prandial insulin despite similar A1C response, and low A1C (<8.5%) had 3-fold increased hypoglycemia by CGM (p=0.16) with similar A1C and insulin dosing. Importantly, high A1C HDV-L subjects who achieved endpoint A1C<8.5% also had lower hypoglycemia and insulin dosing at endpoint.
HDV-L is noninferior to LIS, and provides insulin dosing and hypoglycemia benefits to poorly controlled T1DM.
Technosphere Insulin (TI), an ultra-rapid, short-acting inhaled insulin, provides glycemic control in patients with diabetes. This ongoing, 2-part, open-label, interventional study (NCT02527265) aims to assess efficacy and safety of TI in children aged 4 to 17 years with type 1 diabetes (T1D) who are receiving a stable regimen of basal-bolus insulin.
Part 1, a single-arm study, includes evaluation of insulin pharmacokinetics (PK) after a single prandial dose of TI (4, 8, or 12 U), followed by a 4-week titration period. Approximately 46 patients will be enrolled across 3 cohorts: cohorts 1 (13-17 years), 2 (8-12 years), and 3 (4-7 years).
Initial PK results from cohort 1 demonstrated insulin metabolism similar to that in adult patients receiving TI. Insulin concentrations rapidly increased in the first 30 minutes after TI treatment and returned to baseline by 120 minutes for the 4-, 8-, and 12-U doses. Mean postprandial glucose levels decreased within 1-hour postdose for the 8- and 12-U doses. Seven patients developed treatment-related cough of mild (6 events) or moderate (1 event) severity; 7 patients experienced a total of 41 hypoglycemic events; no patients required the assistance of another person to administer corrective carbohydrates. No clinically relevant declines in pulmonary function were reported. Two patients discontinued because of adverse events (cough, diabetic ketoacidosis).
These data will help determine the appropriate age range for inclusion and recommended dosing for part 2, which will be a 1-year efficacy and safety study.
Islet encapsulation devices can induce a Foreign Body Response (FBR) due to the formation of a hardened avascular fibrotic capsule. This FBR is heightened when the device features a smooth surface as fibrous tissue is unable to adhere to the device, causing friction and thus instigating a substantial immunological reaction. This limits nutrient and oxygen diffusion causing islet death and ultimately implant failure. In this study we examine whether additive manufactured multiscale porous coatings promote optimal tissue integration and vascularisation for long-term functional islet encapsulation devices.
Devices exhibiting progressively more complex surface architectures (quantity of pores, microtexture and macrotexture) were implanted subcutaneously in rodents. Upon explant, analysis of fibrous capsule, angiogenic and macrophage response was performed. To validate scalability and functionality, devices were implanted in an STZ-induced diabetes pig model for two weeks before the blood glucose levels were measured in response to the infusion of insulin through the device.
SEM and MicroCT imaging demonstrated no tissue attachment and a noticeable void between the smooth surface devices and surrounding tissue. A significant increase in capsule thickness, vessel density and maturity were associated with complex surface architecture with no difference in macrophage populations. Bioavailability was equal when the same dose of insulin is delivered via the device vs subcutaneously in diabetic pig model.
Additive manufactured multiscale porous coatings on macroencapsulation devices can increase tissue integration and vascularity. These findings demonstrate scalability and functionality and the ability to resolve the immunological and diffusion limitations of current encapsulation devices.
Technosphere Insulin (TI), an inhaled insulin with ultra-rapid onset and short duration of action, may improve patient outcomes. This study aimed to determine real-world dosing requirements, safety, and efficacy of TI use in patients with uncontrolled type 2 diabetes (T2D).
Twenty patients with T2D inadequately controlled with oral/injectable combination therapy in a real-world setting received TI added to current therapy. Dosing started at TI 4 U and was adjusted weekly on the basis of postprandial glucose levels. HbA1c and CGM profiles were collected at baseline and 12 weeks.
Final mealtime doses were 18, 17, and 19 U for breakfast, lunch, and dinner, respectively. Mealtime TI resulted in significant reductions from baseline at week 12 in HbA1c (-1.6%, P<0.0001; Table) and mean daily glucose (-41.0 mg/dL, P<0.0002) and increased time in range (TIR) by 23.5%, with a 1.5% increase in hypoglycemia and no severe hypoglycemia (Table). Patients averaging ≥20 U per meal had a 92% increase in TIR , with no substantial increase in hypoglycemia (1.9%). Weight loss of 0.91 kg occurred (not significant).
Primary endpoint (SD), %
Time in range
Time <70 mg/dL
Time >180 mg/dL
CGM, continuous glucose monitoring; HbA1c, glycated hemoglobin.
Addition of mealtime TI significantly improved overall glycemic control in patients previously uncontrolled with oral/injectable diabetes medications.
CSII set wear duration is typically limited to three-day best practice. Insulin preservatives may create local irritation contributing to altered insulin uptake and BG responses noted in literature, but limited paired PK/PD data exist for >3-day set wear. We clinically examined preservative impact on bolus insulin PK/PD during 7-day extended wear.
In a double-blind, randomized crossover study, n=25 T1D subjects received continuous SC infusions of insulin diluent (1.60/0.65 mg/mL, m-cresol/phenol) or 0.9% saline (negative control) via steel infusion sets during two 7-day inpatient periods. Insulin boli were administered via these same catheters on days 1, 3, 5, and 7, with accompanying standardized meals and blood sampling for insulin concentration PK and post-prandial blood glucose (PD). An additional set provided required insulin during non-test periods.
All sets survived the 7d wear period without occlusion alarm, undue irritation or required replacement. Diluent treatment accelerated early-phase absorption with Tmax and InsT50% max-rising (P<0.05) compared to saline across all days. Insulin AUC0-1hr and Cmax were not affected by treatment but showed significant increases with time at each successive dose (P<0.05). Despite faster absorption, there were few significant PD treatment differences; however, both showed strong trends of decreasing % time-in-range (70-180 mg/dL) and increasing hyperglycemia (% time>180 mg/dL) with increased wear time.
While inconclusive on phenolic preservatives impact, these data demonstrate potential for extended set survivability and provide paired PK/PD over extended wear time. Further investigation is required to understand the mechanisms affecting both insulin absorption and BG control.
The microbiological contamination of pen injectors during subcutaneous(SC) injections is relatively unexplored. As the industry moves toward more sustainable, integrated and automated devices, knowledge concerning device contamination and how to prevent this must be established. Contamination of the pen injector cartridge during injection has been previously documented, indicating potential transmition of various pathogens, including viruses, bacteria, and fungi. These studies, however, investigate only the drug cartridge and not needle, focusing on human cells and hemoglobin, rather than microorganisms. This paper will therefore establish the rate and nature of microbial contamination of the needle during typical SC injections. The risk of contamination will be assessed through biogeographical mapping of diabetic skin microflora through tape-stripping of abdomen and thigh of diabetic patients (type 1&2).
The microbiological identification and quantification of these specimens will be performed through microscopy and MALDI-TOF MS diagnosis. Secondly, needles and drug cartridges from used pen injectors will be collected and analyzed for contamination by microbiologic wash-outs followed by qPCR. The total number of contaminating microflora as well as viable microflora will be quantified. The identification of microorganisms will be performed using standard microbiologic assessment methods such as 16s NGS and analysis with CSLM.
The relation between the microbiological distribution of the skin and the contamination of the pen injector will be calculated, motivating the establishment of minimal antimicrobial efficacy levels required by materials for future device applications.
This work was supported by a research grant from the Danish Diabetes Academy, which is funded by the Novo Nordisk Foundation, grant number NNF17S0031406.