2084 Diabetes Care Volume 37, August 2014 Gregory R. Fulcher,1 Jens Sandahl Christiansen,2 Comparison of Insulin Ganapathi Bantwal,3 Miroslawa Polaszewska- Muszynska,4 Henriette Mersebach,5 Degludec/Insulin Aspart Thomas H. Andersen,5 and Leo K. Niskanen,6 on behalf of the BOOST: Intensify Premix I and Biphasic Insulin Aspart 30 Investigators in Uncontrolled, Insulin-Treated Type 2 Diabetes: A Phase 3a, Randomized, Treat-to-Target Trial Diabetes Care 2014;37:2084–2090 | DOI: 10.2337/dc13-2908 CLIN CARE/EDUCATION/NUTRITION/PSYCHOSOCIAL OBJECTIVE Insulin degludec/insulin aspart (IDegAsp) is the first combination of a basal insulin with an ultralong duration of action, and a rapid-acting insulin in a single injection. This trial compared IDegAsp with biphasic insulin aspart 30 (BIAsp 30) in adults with type 2 diabetes inadequately controlled with once- or twice-daily (OD or BID) pre- or self-mixed insulin with or without oral antidiabetic drugs. RESEARCH DESIGN AND METHODS In this 26-week, randomized, open-label, multinational, treat-to-target trial, par- ticipants (mean age 58.7 years, duration of diabetes 13 years, BMI 29.3 kg/m2,and 1University of Sydney, Royal North Shore Hospi- HbA 8.4% [68 mmol/mol]) were exposed (1:1) to BID injections of IDegAsp (n = tal, Australia 1c 2Aarhus University Hospital, Aarhus, Denmark 224) or BIAsp 30 (n = 222), administered with breakfast and the main evening meal 3St. Johns Medical College Hospital, Bangalore, and dose titrated to a self-measured premeal plasma glucose (PG) target of 4.0– India 5.0 mmol/L. 4Bydgoszcz Diabetes and Endocrinology Center, Bydgoszcz, Poland 5 RESULTS Novo Nordisk A/S, Søborg, Denmark 6School of Medicine, University of Eastern Fin- After 26 weeks, mean HbA1c was 7.1% (54 mmol/mol) for both groups, with land, Helsingfors, Finland fi IDegAsp achieving the prespeci ed noninferiority margin for mean change in Corresponding author: Gregory R. Fulcher, greg. HbA1c (estimated treatment difference [ETD] –0.03% points [95% CI –0.18 to [email protected]. 0.13]). Treatment with IDegAsp was superior in lowering fasting PG (ETD –1.14 Received 16 December 2013 and accepted 20 mmol/L [95% CI –1.53 to –0.76], P < 0.001) and had a significantly lower final mean March 2014. daily insulin dose (estimated rate ratio 0.89 [95% CI 0.83–0.96], P = 0.002). Fewer Clinical trial reg. no. NCT01009580, clinicaltrials confirmed, nocturnal confirmed, and severe hypoglycemia episodes were reported .gov. for IDegAsp compared with BIAsp 30. This article contains Supplementary Data online at http://care.diabetesjournals.org/lookup/ CONCLUSIONS suppl/doi:10.2337/dc13-2908/-/DC1. IDegAsp BID effectively improves HbA and fasting PG levels with fewer hypo- © 2014 by the American Diabetes Association. 1c Readers may use this article as long as the work glycemia episodes versus BIAsp 30 in patients with uncontrolled type 2 diabetes is properly cited, the use is educational and not previously treated with once- or twice-daily pre- or self-mixed insulin. for profit, and the work is not altered. care.diabetesjournals.org Fulcher and Associates 2085 Insulin is the most effective therapy for in the subcutaneous tissue that slowly RESEARCH DESIGN AND METHODS achieving glycemic control in patients dissociate to provide continuous IDeg Trial Design with type 2 diabetes (1). Owing to the absorption to meet basal needs. At the A total of 50 sites in 10 countries (Aus- progressive nature of type 2 diabetes, same time, IAsp hexamers immediately tralia, Denmark, Finland, India, Malay- many patients will become candidates dissociate into monomers that are rap- sia, Poland, Sweden, Taiwan, Thailand, for prescription of insulindfrequently idly absorbed into the circulation, pro- and Turkey) participated in this multina- basal insulin initially on a variable back- viding mealtime coverage (13,14). tional 26-week, phase 3a, open-label, ground of antidiabetic therapy. How- Pharmacodynamic studies have demon- randomized, treat-to-target trial. ever, as type 2 diabetes progresses, strated that the glucose-lowering effect The protocol, protocol amendments, further treatment intensification in the of IDegAsp is characterized by a distinct consent form, and subject information form of prandial insulin is commonly peak action (from IAsp) and a separate sheet were reviewed and approved by required to achieve HbA1c targets and basal action (from IDeg). This is a closer health authorities according to local reg- potentially to avoid the long-term approximation of physiological ac- ulations and by the local independent consequences of hyperglycemia (2). tion than seen with current biphasic ethics committees prior to trial initia- Currently, this approach is achieved formulations. The long-acting basal tion. This trial was performed in accor- through the addition of one or more component of IDegAsp allows for a dance with the Declaration of Helsinki separate prandial insulin injections to pharmacodynamic profile that is flat and Good Clinical Practice (19). All study an existing basal insulin regimen or and stable throughout the day, mimick- participants gave written consent prior by switching to premixed insulin ing that of IDeg alone (14). The IDeg to any trial-related activities, and the in- containing a fixed ratio of rapid- and component in IDegAsp has been vestigator retained the consent forms. intermediate-acting insulin (1). However, demonstrated to have a predictable several barriers remain to insulin intensi- glucose-loweringeffectanduptofour Trial Population fication. While separate basal and bolus times lower day-to-day variability com- Adults with a clinical diagnosis of type 2 injections offer the most precise and pared with insulin glargine at steady- diabetes for $6 months were enrolled if adaptable mealtime glycemic control state conditions (15). Clinical trials they had an HbA1c of 7–10% (53–86 (1), they can be perceived as complex have demonstrated a lower risk of mmol/mol), had a BMI of #40 kg/m2, (1,3–5). In some patients, this may result IDeg causing hypoglycemia compared and were $18 years of age. Patients in an increased burden of treatments and with insulin glargine (16) and the possi- were on premixed human or analog decreased adherence to injections (6). bility of flexibility in the timing of injec- insulin or self-mixed insulin regimens Currently, available premix insulins tions (17). containing 20–40% fast-/rapid-acting offer a more convenient alternative, An earlier proof-of-concept study component, administered OD or BID with fewer injections and the potential demonstrated favorable efficacy and with or without OADs (metformin, sul- use of a single insulin pen device (7), safety profiles of IDegAsp as an add-on fonylureas, glinides, a-glucosidase but are associated with an increased to metformin in insulin-naive patients inhibitors, DPP-4 inhibitors, or pioglita- rate of postmeal hypoglycemia, including with type 2 diabetes inadequately con- zone) for $3 months prior to trial nocturnal hypoglycemia (8). The fear of trolled on oral antidiabetic drugs (OADs) initiation. hypoglycemia represents a significant (18). IDegAsp twice daily (BID) plus Patients were excluded if they had re- barrier to treatment (9,10). As a result, metformin provided overall glycemic ceived treatment in the 3 months prior patients with type 2 diabetes may con- control at lower insulin doses and signif- to trial initiation with other insulin regi- tinue to have poor glucose control and icantly lower rates of hypoglycemia mens, rosiglitazone, or a glucagon-like an increased risk of developing diabetes- compared with the widely used pre- peptide 1 receptor agonist or a history related complications (1). mixed insulin, biphasic IAsp 30 (BIAsp of recurrent severe hypoglycemia (more Comprising 70% insulin degludec 30) (NovoMix 30, NovoLog Mix 70/30; than one severe hypoglycemic episode (IDeg) and 30% insulin aspart (IAsp), Novo Nordisk A/S) in combination with during the past 12 months) or hypogly- IDegAsp (Ryzodeg; Novo Nordisk A/S, metformin in insulin-naive patients with cemic unawareness. Patients were also Bagsværd, Denmark) is the first soluble type 2 diabetes with an HbA1c of 7–11% excluded if they had cardiovascular dis- combination product of a basal insulin (18). These phase 2 results, while prom- ease (heart failure: New York Heart As- with an ultralong duration of action ising, were limited by a relatively small sociation class III or IV, unstable angina and a rapid-acting insulin in a single sample size (n = 61 in the IDegAsp arm; pectoris, or a myocardial infarction) injection. As a soluble coformulation, n = 60 in BIAsp 30 arm) and short study within 6 months preceding trial and un- IDegAsp does not require resuspension, duration (16 weeks). controlled severe hypertension (systolic which eases administration (11) and, In this phase 3a randomized, multi- blood pressure $180 mmHg or sitting more importantly, eliminates the risk national clinical trial, the efficacy and diastolic blood pressure $100 mmHg). of incomplete mixing and therefore in- safety of IDegAsp were compared with creased hypoglycemia (12). In solution, those of BIAsp 30 in adults with type Randomization the individual components of IDegAsp 2 diabetes inadequately controlled on After randomization, eligible patients exist separately as di-hexamers (IDeg) once-daily (OD) or twice-daily (BID) discontinued their diabetic treatment and hexamers (IAsp) (13). Upon injec- premixed or self-mixed insulin regi- except for metformin, DPP-4 inhibitors, tion, the IDeg di-hexamers assemble to mens with or without concomitant and pioglitazone and were switched form soluble and stable multihexamers OADs. from their prior insulin to IDegAsp or 2086 IDegAsp in Insulin-Treated Type 2 Diabetes Diabetes Care Volume 37, August 2014 BIAsp 30. Patients were randomized baseline in fasting PG (FPG) and the maintenance period after stable glyce- (1:1) to BID injections of IDegAsp (100 changes in nine-point self-measured mic control and insulin dose had been units/mL Ryzodeg) or BIAsp 30 (100 PG (SMPG) profiles.