FormalPara Key Summary Points

Why carry out this study?

Hypoglycemia is a concern for insulin-treated patients with diabetes.

The hypoglycemia risk of patients treated with a once-weekly insulin needs to be examined.

This study sought to compare the hypoglycemia risk following efsitora dosing to that following glargine dosing under the experimental conditions of fasting, exercise, and double dosing in participants with T2D.

What was learned from the study?

Weekly efsitora did not increase the incidence, duration, or severity of hypoglycemia compared to daily glargine in participants with T2D.

Patients with diabetes, caregivers, and healthcare providers may be reassured that real-world hypoglycemia provocation scenarios elicited similar incidences of hypoglycemia with weekly efsitora treatment and daily glargine treatment.

Introduction

As T2D advances, patients often must incorporate basal insulin therapy into their treatment regimen to help improve glycemic control. However, currently available basal insulin formulations require daily injections, which places a substantial physical burden on patients and caregivers [1]. An efficacious basal insulin therapy requiring once-weekly injections could result in earlier adoption of insulin therapy and improve adherence, potentially leading to better real-world outcomes.

Efsitora is a novel once-weekly insulin receptor agonist undergoing phase 3 development. Efsitora has a flat pharmacokinetic (PK) profile and long half-life (17 days), allowing for weekly dosing [2, 3]. These attributes have the potential to provide stable glucose levels and improve glycemic control. Previous phase 2 clinical trials have demonstrated similar efficacy of efsitora compared to daily insulin degludec [4,5,6].

A potential concern for a basal insulin with such a long half-life is the risk of hypoglycemia and the duration of hypoglycemic events. However, two phase 2 studies of efsitora in adults with T2D demonstrated similar rates of hypoglycemia compared to once-daily degludec [4, 5]. Data from previous basal insulin formulations indicate that basal insulins with longer half-lives may lead to less of a peak effect, thereby reducing hypoglycemia or improving glycemic control [7]. For example, degludec, which has a half-life greater than 25 h, demonstrated a lower rate of hypoglycemia compared to insulin glargine U100, which has a half-life of 12.5 h [8, 9]. Nevertheless, a longer half-life will also reduce the flexibility to quickly reduce the basal insulin dose in situations that might provoke hypoglycemia.

This exploratory study aimed to characterize the glycemic responses of patients with T2D previously treated with basal insulin under experimental conditions that simulate real-life situations where hypoglycemic risk is increased, such as during a prolonged 24-h fast, a prolonged 24-h fast plus exercise, and following a double dose. The sequential design first evaluates the hypoglycemic risk with glargine U100 before converting the patient to efsitora and then re-evaluating the hypoglycemic risk under the same conditions.

Research Design and Methods

Study Design

This was a single-site (Profil, Neuss, Germany), phase 1, exploratory, open-label, two-period, fixed-sequence, multiple-dose study (Supplemental Fig. 1). The study consisted of four study periods: a screening and run-in period, a glargine U100 treatment period, an efsitora treatment period, and a safety follow-up.

During the glargine and efsitora treatment periods, there were three hypoglycemia provocation periods: a 24-h prolonged fast, a 24-h prolonged fast and exercise, and following a double dose. The trial is registered on clinicaltrials.gov (NCT04957914).

The trial was conducted in accordance with the principles of the Declaration of Helsinki and Good Clinical Practice guidelines and was approved by the Ethics Committee of the North Rhine Medical Association (protocol reference number: 202051). All participants provided written informed consent prior to participation.

Study Population

Eligible participants included adults (18–70 years) with T2D on a stable regimen of daily basal insulin (NPH insulin, glargine U100 or U300, insulin detemir, or degludec) with or without metformin, dipeptidyl peptidase IV inhibitors, sodium-glucose cotransporter 2 inhibitors, or glucagon-like peptide 1 agonists for at least 3 months. Participants had a BMI between 18.5 and 40 kg/m2 and HbA1c levels ≥ 6.5 and ≤ 9.5% and had not experienced any episodes of severe hypoglycemia or hypoglycemia unawareness in the 6 months prior to screening.

Study Procedures

Screening and Run-In

During the run-in period, self-monitored plasma glucose (SMPG) assessments were conducted on 4 of the 7 days prior to the first treatment period visit. On day −1, participants were admitted to the clinical research unit (CRU) and received a continuous glucose monitoring (CGM) system (Dexcom G6) to aid in the safe conduct of the study; CGM was not used for any safety endpoints. Participants were unblinded to the CGM, with preset alarms used for ‘urgent low’ (plasma glucose (PG) ≤ 55 mg/dL), ‘urgent low soon’ (PG ≤ 55 mg/dL within 20 min), ‘low’ (PG ≤ 70 mg/dL), and ‘fall rate’ (PG dropping at −3 mg/dL per minute) alerts.

Treatment Period 1—Glargine

Participants transitioned from their pre-study basal insulin to glargine U100 and received the first dose of glargine in the evening of day 1. Participants were discharged from the CRU the following day. Glargine was self-administered daily while outside the CRU. Dose assessments, including safety, CGM, and SMPG reviews, were conducted weekly. Glargine was titrated based on the median fasting plasma glucose (FPG) from SMPG assessed during the prior week to achieve a target fasting glucose of ≤ 100 mg/dL (Supplemental Table 1). On day 22, participants were required to have a stable FPG between 80 and 150 mg/dL based on three median FPG readings from the prior week in order to continue in the study.

The three hypoglycemia provocation tests occurred while in the CRU. The 24-h prolonged fast began on day 27 after the participants received a dose of glargine and ate an evening meal. Participants were discharged 2 days later. The prolonged fast followed by exercise provocation began on day 33. Again, participants received a dose of glargine and ate a meal prior to the start of a 24-h prolonged fast. The following morning, participants exercised for 40 min (on a stationary bicycle, with a target heart rate of 40% of heart rate reserve (HRR)). On days 35 through 37, participants transitioned from evening to morning dosing of glargine. Specifically, participants received 50% of the daily insulin glargine dose in the evening of day 35, in the morning on day 36, and in the evening on day 36. Finally, participants received a full daily glargine dose in the morning of day 37. The double-dosing provocation began with an overnight fast in the evening of day 37. The following morning, participants received a double dose of glargine and then ate a predefined meal. The daily dose of glargine was continued the next day. Participants were discharged from the CRU 2 days later.

On days 42 and 45, participants had outpatient visits to complete safety assessments and to review the CGM and SMPG data.

Treatment Period 2—Efsitora

On day 50, participants re-entered the CRU to begin the transition from glargine to efsitora. For the first efsitora injection, a one-time starting dose, calculated based on the glargine daily dose and the median baseline FPG (Supplemental Table 2), was administered on day 51. Participants were discharged the following day. Efsitora was administered weekly at the CRU by site personnel. Outpatient dose assessments, including safety, CGM, and SMPG reviews, were conducted weekly from day 58 through 100. Efsitora dose titrations were performed based on the median FPG from the previous week to achieve a fasting glucose target range of 80–130 mg/dL (Supplemental Table 3).

Similar to the glargine treatment period, the hypoglycemia provocation tests occurred while the participants were admitted to the CRU. The 24-h prolonged fast began on day 105 after participants ate an evening meal. The participants were discharged 2 days later. The prolonged fast with exercise provocation began on day 111 after the participants ate an evening meal. The following morning, the participants exercised for 40 min. On day 113, for the double-dosing provocation test, the participants ate an evening meal and began an overnight fast. Participants then received a double dose of efsitora the following morning. Participants were discharged on day 118.

On day 121, participants reviewed their CGM and SMPG data, completed safety assessments, and began their transition back to pre-study basal insulin based on the SMPG, hypoglycemia data, and investigator discretion.

Follow-up

At least 75 days after the last dose of efsitora, participants returned to the CRU for a follow-up visit.

Hypoglycemia treatment

Since one main goal of the study was to understand potential differences in hypoglycemia in response to the provocations, the usual management and intervention practices for hypoglycemia were not followed. Instead, during the hypoglycemia provocation periods for both efsitora and glargine, participants were monitored for signs and symptoms of hypoglycemia throughout the fasting period. PG was monitored approximately every 6 h, avoiding postprandial measurements, using the Super GL glucose analyzer. PG was monitored more frequently after the first PG reading below 100 mg/dL, a CGM hypoglycemia alarm was triggered, or the participant reported symptoms of hypoglycemia. PG was monitored every 15 min following a glucose reading below 80 mg/dL and monitored every 10 to 15 min if PG fell below 70 mg/dL. PG was monitored more frequently if the PG dropped by 3 mg/dL per min or more and triggered the CGM alarm. It was only when the participant experienced intolerable symptoms or PG was < 54 mg/dL that 15 g of oral carbohydrates was administered, and the participant was monitored for 90 min.

Study Assessments and Outcome Measures

The PD parameter assessments included nadir PG, time to hypoglycemia, the duration of hypoglycemia events occurring during the provocation monitoring periods, and FPG from SMPG. The hypoglycemia incidence and event rates during provocation periods were calculated. The hypoglycemic events were categorized according to the American Diabetes Association (ADA) descriptions as either level 1 hypoglycemia (glucose < 70 mg/dL (3.9 mmol/L) and ≥ 54 mg/dL (3.0 mmol/L)), level 2 hypoglycemia (glucose < 54 mg/dL (3.0 mmol/L)), and level 3 hypoglycemia (severe hypoglycemia characterized by altered mental and/or physical status requiring assistance). Nocturnal hypoglycemia was defined as a hypoglycemia event that occurred at night and presumably during sleep. All hypoglycemic events with a confirmed glucose reading of < 70 mg/dL were recorded. The end time of a hypoglycemic event was recorded when the PG value returned to 70 mg/dL and remained at or above 70 mg/dL for at least 15 min.

Because treatment of hypoglycemia could have included glucose administration, which disrupts the fasting status during the fasting and fasting with exercise provocation tests, the summary of the nadir PG and duration of hypoglycemia for each of these two provocation periods included data for the hypoglycemic events that occurred prior to the first hypoglycemic event that required glucose treatment. For the double-dosing period, during which participants received regular meals, a summary of all events, regardless of the time relative to glucose treatment, was included.

Two sensitivity analyses were conducted. The first assessed the incidence of hypoglycemia in a subgroup of participants who achieved a stable FPG of 80 to 120 mg/dL prior to the provocation periods. The second sensitivity analysis was conducted on a subgroup of participants who had similar FPG values (within 15% of the mean of the two FPG values) in their glargine and efsitora treatments prior to the double-dosing provocation period.

Adverse events (AEs), serious AEs (SAEs), and product complaints (PCs) were reported by participants throughout the study. Additionally, laboratory tests, vital signs, and electrocardiograms were documented.

Statistical Analysis

Sample Size

The sample size was determined such that we needed ~ 50 participants to enroll and ~ 40 participants would complete the study. Forty completers have at least 90% power to detect a treatment difference in the incidence of hypoglycemia (< 70 mg/dL) during the double-dose provocation period, assuming an incidence of hypoglycemia of 43% for efsitora, an incidence of hypoglycemia of 91% for glargine, a two-sided significance level of 0.05, and a within-patient correlation of zero between two treatment visits.

The primary objective—the incidence of hypoglycemia during each provocation period—was reported for each treatment, the difference between treatments was calculated, and the 95% confidence intervals were determined using the Wald method with continuity correction. Hypoglycemic provocation monitoring periods were defined as follows:

  1. 1.

    The prolonged fasting period started when the fast began and ended 24 h later.

  2. 2.

    The prolonged fasting with exercise period started when the fast began and ended 24 h later.

  3. 3.

    The double-dosing period started when the double dose was administered and ended 96 h later for efsitora and 48 h later for glargine.

AE safety data were summarized using descriptive statistics.

Results

Demographic and Baseline Characteristics

All 54 participants who were enrolled in the study received at least one dose of glargine in period 1 of the study. Forty-seven participants completed the study. Discontinuations were due to the physician’s decision (n = 4; the participants did not reach a stable FPG between 80 and 150 mg/dL), treatment-emergent AEs (TEAEs; n = 1; the participant reported a limb abscess), withdrawal by the participant (n = 1), and deviation from the protocol (n = 1).

The 54 participants enrolled in the study were between 37 and 71 years old, had a BMI between 21.8 and 39.7 kg/m2, and had HbA1c of between 6.5% and 9.4% (Table 1).

Table 1 Demographic and clinical characteristics
Full size table

Pharmacodynamics

Fasting Plasma Glucose

The mean FPG levels, as measured by SMPG assessment, are presented in Fig. 1a. PG levels decreased in participants following the start of glargine or efsitora during the dose adjustment periods. The mean PG levels prior to the prolonged fasting and prolonged fasting with exercise provocation periods for efsitora and glargine were within the target range, but the mean FPG level was slightly lower (111 vs. 117 mg/dl; ∆ ~ 6 mg/dL) during efsitora treatment compared to glargine treatment (Fig. 1b). The glucose level prior to the double-dosing provocation was also lower (97 vs. 107 mg/dl; ∆ ~ 10 mg/dL) during efsitora treatment compared to glargine treatment. The mean dose administered for the double-dosing provocation period was 68 U/day for glargine and was 574 U/week (approximately equivalent to 82 U/day) for efsitora.

Fig. 1
figure 1

Mean fasting plasma glucose A over the study duration and B during the prolonged fasting, prolonged fasting with exercise, and prior to the double-dosing provocation period. Data are presented as mean ± SE. In A, the vertical lines represent the start of each treatment period. The shaded boxes represent the provocation periods. SE standard error

Full size image

Hypoglycemia Incidence

No severe hypoglycemic events were observed throughout the study. The majority of the observed hypoglycemic events were level 1, which participants recovered from spontaneously or with 15 g of carbohydrates. The incidences of level 1 hypoglycemia during the prolonged fasting provocation periods were 44.7% for efsitora and 42.6% for glargine (Table 2; Supplemental Fig. 2a). There was no statistically significant difference between the incidence with efsitora and the incidence with glargine (difference in proportion [95% confidence interval (CI)], 2.1 [– 17.2, 21.4]). Level 2 hypoglycemia was reported infrequently during both treatment periods, with only 6 participants in the efsitora period and none in the glargine period experiencing events during prolonged fasting provocation; this difference was statistically significant (efsitora = 12.8%, glargine = 0; 12.8 [1.1, 24.4]; Table 2; Supplemental Fig. 2b).

Table 2 Hypoglycemia results
Full size table

During the prolonged fasting with exercise provocation period, the incidences of level 1 hypoglycemia were 65.9% and 50.0% for the efsitora and glargine treatments, respectively; no statistically significant difference was observed (15.9 [– 3.0, 34.8]). Similarly, there was no statistically significant difference in the incidence of level 2 hypoglycemia for efsitora (9.1%) and glargine (0%) during the prolonged fasting with exercise period (9.1 [– 1.7, 19.9]).

The highest incidences of level 1 and level 2 hypoglycemia were observed during the double-dosing provocation for efsitora (level 1, 55.3%; level 2, 8.5%) and glargine (level 1, 61.7%; level 2, 12.8%). There were no statistically significant differences in the incidences of level 1 (– 6.4 [– 28.4, 15.7]) or level 2 (– 4.3 [– 18.1, 9.6]) hypoglycemia for this provocation period.

The first sensitivity analysis, which attempted to address differences in fasting glucose at the start of the provocation periods by only including participants with a stable FPG between 80 and 120 mg/dL, demonstrated that treatment with efsitora resulted in comparable incidences of level 1 and 2 hypoglycemia compared to glargine in all provocation periods for this subgroup (Supplemental Table 4).

Number of Hypoglycemic Events

The number of level 1 hypoglycemic events was the same (n = 38) for efsitora and glargine during prolonged fasting provocation (Table 2; Supplemental Fig. 2a). During this same provocation period, only six level 2 events were reported during efsitora treatment and zero were reported during glargine treatment (Table 2; Supplemental Fig. 2b).

During the prolonged fasting with exercise provocation test, the number of level 1 hypoglycemic events was higher during efsitora treatment (n = 53) than during glargine treatment (n = 37). Only four level 2 hypoglycemic events were reported during efsitora treatment in the fasting with exercise provocation period. No level 2 events were reported during glargine treatment.

The number of level 1 hypoglycemic events up to 48 h after a double dose of glargine (n = 62) was lower than the number of events after efsitora double dosing (n = 79). The number of level 1 hypoglycemic events reported between 48 and 96 h after efsitora double dosing was 136 events. Up to 48 h after double dosing, four and six level 2 hypoglycemic events were reported for efsitora and glargine, respectively. Another eight level 2 hypoglycemic events were observed between 48 and 96 h after efsitora double dosing.

The second sensitivity analysis, which assessed the impact of the lower FPG exhibited by efsitora prior to the double dosing, demonstrated more comparable results than the primary analysis (Supplemental Table 4). Participants with similar FPG values prior to double dosing reported 44 and 61 level 1 events between 0 and 48 h and between 48 and 96 h after efsitora treatment, respectively, and 34 events up to 48 h after glargine treatment. One level 2 hypoglycemic event occurred between 0 and 48 h and five events occurred between 48 and 96 h after efsitora treatment. Four level 2 hypoglycemic events were reported up to 48 h after glargine double dosing.

Nadir Plasma Glucose

The nadir PG values were similar during the efsitora and the glargine treatment periods for all provocations. The mean ± SD (minimum, maximum) nadir PG during hypoglycemic events occurring prior to the first event that required glucose treatment in the prolonged fasting was 63.4 ± 5.1 (51, 69) mg/dL for efsitora and 65.8 ± 3.7 (57, 69) mg/dL for glargine, and during the fasting with exercise provocation periods, the nadir PG was 66.3 ± 3.1 (55, 69) mg/dL for efsitora and 65.9 ± 3.2 (59, 69) mg/dL for glargine. During all events in the double-dosing provocation periods, the nadir PG was 63.0 ± 4.9 (47, 69) for efsitora and 62.8 ± 5.2 (48, 69) mg/dL for glargine.

Time to Hypoglycemia

Kaplan–Meier plots suggest that participants were more likely to have a level 1 hypoglycemic event earlier during the prolonged fasting and prolonged fasting with exercise provocation periods after efsitora treatment compared to glargine (Supplemental Fig. 3). After double dosing, participants were more likely to have a level 1 hypoglycemic event earlier during glargine treatment compared to efsitora. There was no statistically significant difference in time to level 1 hypoglycemia between efsitora and glargine in the fasting and double-dosing provocation periods. During the prolonged fasting with exercise provocation period, participants were statistically significantly more likely to experience level 1 hypoglycemia earlier during efsitora treatment compared to glargine (p = 0.0103).

There were no level 2 hypoglycemic events during the prolonged fasting and prolonged fasting with exercise provocation periods after glargine treatment. The Kaplan–Meier plots suggest that the time to level 2 hypoglycemia was similar for efsitora treatment and glargine treatment during the double-dosing provocation periods (Supplemental Fig. 4); no statistically significant difference was observed.

Duration of Hypoglycemia

The mean duration of hypoglycemia occurring prior to the first event that required glucose treatment was approximately 14 min longer in the fasting provocation period during treatment with efsitora (115.2 min) compared to glargine (101.0 min). The mean duration of hypoglycemia occurring prior to the first event that required glucose treatment in the fasting with exercise provocation period during treatment with efsitora (90.0 min) was similar to that during treatment with glargine (92.6 min). For all hypoglycemic events during the double-dosing provocation period, the mean duration of hypoglycemia was similar during treatment with efsitora (76.6 min) compared to glargine (79.3 min).

Safety and Tolerability

Both treatments were well tolerated during the study. Overall, the incidence of TEAEs was higher with efsitora treatment (77.6%) compared to glargine (57.4%). Most TEAEs (82.5%) were considered mild in severity. Headache (efsitora n = 5, glargine n = 13), nasopharyngitis (efsitora n = 6, glargine n = 8), and Covid-19 (efsitora n = 6, glargine n = 0) were reported by more than 10% of participants. No deaths were reported during the study. Two SAEs, a limb abscess and obstructive pancreatitis, were reported but were not considered to be related to the study treatment. Injection-site reactions (erythema, pruritus, and induration) were reported in 6.1% of participants during efsitora treatment and in no participants during glargine treatment. No clinically significant trends or changes in clinical laboratory evaluations, vital signs, and electrocardiograms were observed.

Discussion

This phase 1 exploratory study was the first to evaluate the impact of once-weekly doses of efsitora on hypoglycemia in participants with T2D under three experimental conditions: fasting, exercise during fasting, and after receiving a double dose. Overall, the incidence of hypoglycemia was low during the once-weekly efsitora treatment period and similar to once-daily glargine, despite the lower FPG with efsitora prior to entering the provocation periods. While more participants reported level 2 hypoglycemia with efsitora, the incidence was still low, and the finding was not confirmed within the more provocative setting of prolonged fasting with exercise. Further, the duration and severity of hypoglycemic events and the nadir PG were similar for the two insulins. Despite the high-risk provocation scenarios, most of the observed hypoglycemic events were level 1; no severe hypoglycemic events were reported.

A recently published two-period crossover study in patients with T2D previously treated with basal insulin assessed the hypoglycemia frequency after double or triple doses of once-weekly icodec or once-daily glargine in a well-controlled inpatient setting [10]. Results showed similar incidences of level 2 hypoglycemia and that the symptomatic and counterregulatory response to hypoglycemia was similar for the weekly and daily basal insulin treatment groups.

A phase 2 study in patients with T2D previously treated with daily basal insulin demonstrated similar incidences of level 1 and 2 hypoglycemia when the patients were treated with efsitora or degludec [5]. However, the target glucose levels were different for the efsitora and degludec treatment groups, the study was conducted in an outpatient setting, and the usual standard of care for the management of hypoglycemia was implemented. These reassuring data from both an outpatient phase 2 study and the current well-controlled study using plausible, real-world hypoglycemia provocation scenarios may ease some concerns about the risk and features of hypoglycemia with a once-weekly basal insulin, but they require confirmation from the ongoing efsitora phase 3 studies.

This study had several limitations. Namely, this study was designed in an attempt to track the natural course of and recovery from hypoglycemia; hence, only level 2 and symptomatic level 1 events were treated. This may account for some of the increased frequency of level 1 hypoglycemia, because these events were not proactively treated until they became symptomatic. Importantly, during the efsitora and glargine treatment periods, the FPG levels leading into the provocation periods were lower with efsitora, which confounds the interpretation of the hypoglycemia results. Sensitivity analyses were conducted to better understand the impact of the differences in FPG, and the results for efsitora and glargine were more comparable when adjusting for FPG than the results of the primary analyses were. Additionally, there was an increase in the FPG prior to the transition from glargine to efsitora. Participants continued to receive the individualized dose assigned at day 22 throughout the glargine treatment period, without additional dose titrations, unless there were safety concerns.

Three separate provocation periods that mimic real-life scenarios strengthened the utility of this study. Additionally, the use of unblinded CGM with alarms improved the detection of hypoglycemia, which was necessary to address valid clinical concerns with weekly basal insulins. The use of a fixed sequence design, rather than a randomized crossover, for this exploratory study allowed basal insulin dose optimization to be achieved more efficiently and provided a shorter overall study, as the washing out of efsitora between periods was not required. Finally, investigations in larger, more diverse populations with the dosing used in the phase 3 program could strengthen the generalizability of these reassuring results [11].

Conclusion

This study was crucial for understanding how treatment with efsitora impacts the incidence of hypoglycemia during periods of provocation that simulate real-life situations. Overall, the results show that once-weekly efsitora did not increase the incidence, duration, or severity of hypoglycemia compared to once-daily glargine during periods of provocation in patients with T2D previously treated with basal insulin.