Diabetes Ther (2021) 12:1159–1174 https://doi.org/10.1007/s13300-021-01031-z ORIGINAL RESEARCH Glycaemic Control with Insulin Glargine 300 U/mL in Individuals with Type 2 Diabetes and Chronic Kidney Disease: A REALI European Pooled Data Analysis Didac Mauricio . Pierre Gourdy . Riccardo C. Bonadonna . Nick Freemantle . Gregory Bigot . Celine Mauquoi . Alice Ciocca . Mireille Bonnemaire . Dirk Müller-Wieland Received: January 12, 2021 / Accepted: February 10, 2021 / Published online: March 9, 2021 Ó The Author(s) 2021 ABSTRACT daily subcutaneous injections of Gla-300 were classified into four renal function subgroups, Introduction: Management of type 2 diabetes according to baseline eGFR: C 90 (N = 599), mellitus (T2DM) in patients with chronic kid- 60–89 (N = 786), 45–59 (N = 219), and ney disease is complex. Using the REALI Euro- 15–44 mL/min/1.73 m2 (N = 108). pean pooled database, we determined the Results: Compared to those with baseline impact of baseline renal function on the effec- eGFR C 60 mL/min/1.73 m2, patients with tiveness and safety of insulin glargine 300 U/mL lower eGFR values tended to be older, had a (Gla-300) initiated in adults with inadequately longer T2DM duration, and were more likely to controlled T2DM. present diabetic complications. After 24 weeks Methods: Data from 1712 patients with avail- of Gla-300 therapy, the least-squares mean able estimated glomerular filtration rate (eGFR) (95% confidence interval) decrease in hae- at baseline were pooled from six 24-week moglobin A1c (HbA1c) from baseline (- 1.14% prospective studies. Patients who received once- [- 1.28 to - 1.00], - 1.21% [- 1.34 to - 1.08], - 1.19% [- 1.36 to - 1.01], and - 0.99% R. C. Bonadonna D. Mauricio (&) Department of Medicine and Surgery, University of Department of Endocrinology and Nutrition, Parma, Parma, Italy CIBERDEM, Hospital de la Santa Creu i Sant Pau Institut de Recerca, Barcelona, Spain N. Freemantle e-mail: didacmauricio@gmail.com Institute of Clinical Trials and Methodology, University College London, London, UK P. Gourdy Endocrinology, Diabetology and Nutrition G. Bigot Department, Toulouse University Hospital, IVIDATA, Paris, France Toulouse, France C. Mauquoi P. Gourdy IDDI, Louvain-la-Neuve, Belgium Institute of Metabolic and Cardiovascular Diseases, UMR1048 INSERM/UPS, Toulouse University, A. Ciocca  M. Bonnemaire Toulouse, France General Medicines, Sanofi, Paris, France R. C. Bonadonna D. Müller-Wieland Azienda Ospedaliero-Universitaria di Parma, Parma, Department of Medicine I, University Hospital Italy Aachen, Aachen, Germany 1160 Diabetes Ther (2021) 12:1159–1174 [- 1.22 to - 0.76]) and the proportion of What was learned from the study? patients achieving HbA1c \ 7.5% (53.3%, 51.3%, 49.5%, and 51.5%) were comparable in Renal function does not seem to affect the the C 90, 60–89, 45–59, and 15–44 mL/min/ effectiveness and safety of Gla-300, as 1.73 m2 subgroups, respectively. Although the individuals with inadequately controlled incidence of hypoglycaemia was overall low, T2DM and renal impairment achieved a more patients in the eGFR 15–44 mL/min/ clinically important improvement in 1.73 m2 subgroup experienced hypoglycaemia glycaemic control with limited at night or at any time of the day compared hypoglycaemia concerns during 24-week with higher eGFR subgroups. There were no treatment with Gla-300 notable differences between the renal function The findings of the REALI CKD pooled subgroups in the changes in Gla-300 daily dose analysis support Gla-300 as a therapeutic and body weight from baseline to week 24. option for individuals with T2DM and Conclusion: Although an eGFR of 15–44 mL/ renal impairment min/1.73 m2 was associated with a slightly increased risk of hypoglycaemia among patients with inadequately controlled T2DM, Gla-300 provided glycaemic improvement with an overall favourable safety profile regardless of DIGITAL FEATURES baseline eGFR. This article is published with digital features, including a summary slide, to facilitate under- Keywords: Chronic kidney disease; Estimated standing of the article. To view digital features glomerular filtration rate; Europe; Insulin for this article go to https://doi.org/10.6084/ glargine 300 U/mL; Pooled analysis; Type 2 m9.figshare.13809482. diabetes Key Summary Points INTRODUCTION Type 2 diabetes mellitus (T2DM) and chronic Why carry out this study? kidney disease (CKD) are common chronic Chronic kidney disease (CKD) is a comorbid conditions, with the prevalence of common complication in individuals CKD among patients with diabetes estimated at with type 2 diabetes mellitus (T2DM), and 36% in the USA [1]. In Europe, the prevalence of the level of renal function impacts on the end-stage renal disease among patients with choice and dosage of glucose-lowering diabetes is projected to increase approximately therapies as well as on the individual’s 3% per year from 2012 to 2025 [2]. CKD is an glycaemic target and the risk of independent risk factor for hypoglycaemia, and hypoglycaemia increases the risk already present in patients with T2DM due to accumulation of uremic Since data on insulin glargine 300 U/mL toxins leading to lower hepatic and renal insu- (Gla-300) in patients with T2DM and CKD lin degradation, and as a result of decreased are limited in settings close to clinical renal gluconeogenesis as well as deficient cate- practice, the REALI CKD analysis aimed to cholamine release [3–5]. Unfortunately, the determine the impact of renal function on combination of hypoglycaemia and CKD is the effectiveness and safety of Gla-300 associated with increased morbidity and mor- using pooled data from six 24-week, open- tality, particularly from cardiovascular disease label, prospective studies including [4, 5], and T2DM was found to increase the risk European patients with inadequately of death from renal disease by threefold [6]. controlled T2DM who were initiated on or switched to Gla-300 Diabetes Ther (2021) 12:1159–1174 1161 The management of T2DM in patients with METHODS CKD is especially difficult, in part because of treatment complexity and due to insufficient Pooled Studies convincing data supporting the benefits of tight glycaemic control in this subset of patients [7]. The REALI CKD analysis was based on pooled In patients with T2DM and advanced CKD, data from six multicentre, open-label, prospec- basal insulin is considered the therapeutic tive, interventional or observational studies option of choice, as oral glucose-lowering conducted over a minimum period of 24 weeks agents such as metformin, sulfonylureas, and in European adult patients with inadequately sodium glucose co-transporter 2 inhibitors are controlled T2DM (i.e. in whom glycaemic tar- less used because their glycaemic efficacy is gets have not been achieved) who were initiated reduced and/or the risk of hypoglycaemia or on or switched to Gla-300 (Table 1) [10, 15–18]. other adverse reactions is increased as a conse- Patients included in the REALI CKD analysis quence of drug accumulation [4, 5, 7, 8]. Insulin were either insulin-naı̈ve or previously treated requirements are generally lower in patients with basal insulin with or without oral glucose- with T2DM and renal impairment due to a lowering drugs. Common exclusion criteria decreased insulin clearance; yet, there are no included the presence of type 1 diabetes, preg- guidelines regarding insulin dose adjustment in nant or breastfeeding women, stage 5 CKD (es- this specific population [5, 9]. timated glomerular filtration rate Insulin glargine 300 U/mL (Gla-300) is a [eGFR] \ 15 mL/min/1.73 m2), and any condi- second-generation basal insulin that was tion that could interfere with insulin require- developed to optimise glycaemic control while ments and/or haemoglobin A1c (HbA1c) levels. minimising the risk of hypoglycaemia [10]. Gla- In each study, Gla-300 was administered 300 provides a more stable and prolonged subcutaneously once daily, using a pre-filled pharmacokinetic/pharmacodynamic profile insulin pen, at the same time of the day ± 3 h if compared with the first-generation analogue needed. Two studies (Take Control [15] and insulin glargine 100 U/mL (Gla-100) owing to a ITAS [10]) were parallel-group trials in which more gradual and extended release of insulin patients were randomised (1:1) to a self- or glargine from the subcutaneous depot at the physician-managed titration of Gla-300, injection site, extending blood glucose control whereas the other studies included only one beyond 24 h [11]. study arm: Gla-300 prescribed according to the Gla-300 was compared to Gla-100 in the investigating physician’s choice. randomised phase III EDITION 1, 2 and 3 stud- All studies included in this analysis were ies [12] and to insulin degludec 100 U/mL conducted in accordance with the Declaration (IDeg-100) in the head-to-head randomised of Helsinki, Good Clinical Practice guidelines, BRIGHT trial [13], demonstrating equivalent and all applicable regulatory requirements. All glycaemic control with a lower hypoglycaemia participants provided written informed consent risk at night and at any time of the day in a prior to participation in the studies. Before data broad population of patients with T2DM, pooling, all patient information was de-identi- including patients with an impaired renal fied. Consequently, no ethical approval was function. To further investigate the effective- required for this pooled analysis. ness and safety of Gla-300 in individuals with T2DM and CKD, we performed analyses on the REALI pooled database [14] including patient- Outcomes level data from six prospective studies con- ducted in Europe. The main efficacy endpoint for this pooled analysis was the change in HbA1c from baseline to week 24 of Gla-300 treatment. Other efficacy outcomes included the change in HbA1c from baseline to week 12 and week 24 of Gla-300 1162 Diabetes Ther (2021) 12:1159–1174 Table 1 Characteristics of studies included in the pooled analysis Study Location(s) Study design Number Inclusion criteria of Previous treatment HbA1c (%) eGFR patients status (mL/min/ 1.73 m2) Take Control Greece, Spain, 24-week, 620 C 6 months on C 7.0 and C 15 [15] Czech multinational, treatment with B 10.0 for Republic, open-label, C 1 non-insulin patients Switzerland, randomised (1:1), anti- previously Poland, two-arm, parallel- hyperglycaemic treated with Denmark, group study drug, with or basal Slovenia, evaluating a self- without a basal insulin, Slovakia, versus physician- insulin or C 7.5 Croatia, UK managed titration and B 11.0 of Gla-300 for insulin- naı̈ve patients ITAS [10] Italy 24-week, 335 Insulin-naı̈ve C 7.5 and C 15 multicentre, open- patients B 10.0 label, pragmatic, randomised (1:1), parallel-group study evaluating a self- versus physician- managed titration of Gla-300 TRANSITION France 24-week, 154 C 6 months on [ 7.5 C 15 2 [16] interventional, treatment with multicentre, open- basal insulin with label, single-arm or without non- study to evaluate insulin switching from antidiabetic agents any basal insulin to Gla-300 TOPAZ [17] Czech 24-week, 232 C 3 months on [ 7.6 C 30 Republic multicentre, treatment with observational, basal insulin with single-arm study or without non- to evaluate insulin switching from antidiabetic agents any basal insulin to Gla-300 Diabetes Ther (2021) 12:1159–1174 1163 Table 1 continued Study Location(s) Study design Number Inclusion criteria of Previous treatment HbA1c (%) eGFR patients status (mL/min/ 1.73 m2) COBALTA Spain 26-week, 109 Hospitalised C 8.0 and C 30 [18] interventional, patients with B 10.0 multicentre, open- C 3 months on label, single-arm treatment with study to evaluate basal insulin with Gla-300 in a or without non- basal-bolus insulin regimen during antidiabetic agents hospitalisation To UPGRADE Bulgaria 24-week, 262 Previous treatment C 7.0 No eGFR (Data on file) multicentre, with limitation observational, NPH ± prandial single-arm study insulin or to evaluate premixed insulin switching from with or without NPH ± prandial oral antidiabetic insulin or agents premixed insulin to Gla-300 eGFR estimated glomerular filtration rate, Gla-300 insulin glargine 300 U/mL, HbA1c haemoglobin A1c, NPH Neutral Protamine Hagedorn, UK United Kingdom treatment; the change in fasting plasma glucose administer carbohydrates, glucagon, or take (FPG) from baseline to weeks 12 and 24 of Gla- other corrective actions. Symptomatic hypo- 300 treatment; and the percentage of patients glycaemia was defined as an event during which achieving HbA1c targets of \ 7.0% typical symptoms of hypoglycaemia occurred (53 mmol/mol), \ 7.5% (58.5 mmol/mol), and (e.g. sweating, hunger, shakiness, palpitations). \ 8.0% (63.9 mmol/mol) at weeks 12 and 24. With the exception of COBALTA [18] in which Safety endpoints included the percentage of capillary blood glucose monitoring was used to patients with at least one hypoglycaemic event; assess hypoglycaemia in hospitalised patients, the event rate of hypoglycaemic events; and the the other pooled studies [10, 15–17] relied on changes in daily insulin dose and body weight the presence of typical symptoms of hypogly- from baseline to weeks 12 and 24 of Gla-300 caemia, confirmed or not by documented self- treatment. Hypoglycaemic events were reported measured capillary glucose measurement. according to their time of occurrence, during the night and at any time of the day. The defi- Statistical Analyses nitions of hypoglycaemia were predetermined in the present pooled analysis. Severe hypogly- In the present analysis, study participants were caemia was defined as any event requiring pooled and results assessed across four renal assistance from another person to actively 1164 Diabetes Ther (2021) 12:1159–1174 function subgroups, according to baseline these 1712 patients, 1672 (97.7%) were treated eGFR: C 90 (indicating normal kidney func- with at least one dose of Gla-300 and were tion), 60–89 (mild impaired kidney function), included in both efficacy and safety analyses. 45–59 (mild to moderate impaired kidney Demographics and baseline characteristics of function), and 15–44 mL/min/1.73 m2 (moder- the four renal function subgroups are summa- ate to severe impaired kidney function). If not rized in Table 2. Overall, compared to those already determined in each individual study, with a baseline eGFR C 60 mL/min/1.73 m2, eGFR was calculated using the Modification of patients with lower eGFR values tended to be Diet in Renal Disease equation without older, more often female, and had a longer accounting for race, as this was not reported in duration of diabetes. With a decreasing baseline the pooled studies. eGFR, patients were more likely to be pretreated The changes in HbA1c and FPG from base- with insulin, and to suffer from diabetic com- line to weeks 12 and 24 of Gla-300 therapy were plications and from previous cardiovascular analysed using a mixed model for repeated events such as myocardial infarction and stroke. measures (MMRM), including fixed categorical With regards to previous non-insulin anti-hy- effects of visit and eGFR category as well as perglycaemic treatments, patients in the continuous fixed covariates of baseline HbA1c 15–44 mL/min/1.73 m2 subgroup were far less or FPG, baseline HbA1c or FPG value-by-visit likely to be treated with biguanides compared to interaction, and eGFR category-by-visit interac- those in the other subgroups. Mean baseline tion. The least-square (LS) mean change from HbA1c was slightly higher in patients with an baseline to weeks 12 and 24 in HbA1c and FPG eGFR of 15–44 mL/min/1.73 m2 compared to estimates and their corresponding 95% confi- the other eGFR subgroups. dence intervals (CIs) were provided for each eGFR category. All other efficacy and safety Glycaemic Control endpoints as well as baseline demographic and disease characteristics were assessed descrip- The LS mean decrease in HbA1c from baseline tively, with categorical variables presented as to weeks 12 and 24 of Gla-300 therapy was counts and percentages, and continuous vari- clinically meaningful in all four renal function ables as mean, standard deviation (SD), median, subgroups, with overall similar HbA1c reduc- and first and third quartiles. Hypoglycaemic tions across subgroups (Table 3). At 24 weeks, event rates were calculated as the number of the LS mean (95% CI) change in HbA1c from events per patient-year of exposure. baseline was - 1.14% (- 1.28 to - 1.00), No imputation of missing data was per- - 1.21% (- 1.34 to - 1.08), - 1.19% (- 1.36 to formed. All statistical tests were two-sided, with - 1.01), and - 0.99% (- 1.22 to - 0.76) in the a p value \ 0.05 considered statistically signifi- C 90, 60–89, 45–59, and 15–44 mL/min/1.73 m2 cant. All analyses were performed using SAS, subgroups, respectively. version 9.4 (SAS Institute Inc, Cary, North Car- Over a quarter of patients with an olina, USA). eGFR C 45 mL/min/1.73 m2 reached HbA1c levels \ 7.0% at 24 weeks of Gla-300 therapy RESULTS versus 21.6% of patients in the eGFR 15–44 mL/ min/1.73 m2 subgroup. The achievements of Study Population HbA1c targets \ 7.5% or \ 8.0% were fairly comparable across the renal function subgroups at week 24 (Fig. 1). A total of 1712 patients from 13 European In line with changes in HbA1c, the LS mean countries were included in the REALI CKD change in FPG from baseline to weeks 12 and 24 pooled analysis, of whom 108 (6.3%) had a of Gla-300 therapy was also similar across the baseline eGFR between 15 and 44, 219 (12.8%) four renal function subgroups (Table 4). between 45 and 59, 786 (45.9%) between 60 and 89, and 599 (35.0%) C 90 mL/min/1.73 m2. Of Diabetes Ther (2021) 12:1159–1174 1165 Table 2 Baseline characteristics, according to renal function Baseline characteristic by baseline eGFR, mL/min/ 15–44 45–59 60–89 ‡ 90 1.73 m2 (N = 108) (N = 219) (N = 786) (N = 599) Female, n (%) 59 (54.6) 123 (56.2) 358 (45.5) 239 (39.9) Age, years 69.8 ± 7.9 69.2 ± 7.3 64.4 ± 8.3 60.5 ± 8.6 Body weight, kg 90.9 ± 17.6 85.9 ± 14.3 87.5 ± 14.9 88.5 ± 16.8 2 Body mass index, kg/m 32.7 ± 5.4 31.6 ± 4.4 31.2 ± 4.8 31.5 ± 5.2 eGFR, mL/min/1.73 m2 38.0 ± 5.8 53.4 ± 4.1 76.0 ± 8.3 110.0 ± 42.4 Diabetes duration, years 14.5 13.0 11.0 11.0 (10.5–20.5) (8.0–17.0) (7.0–16.0) (6.0–15.0) Previous insulin use, n (%) 88 (81.5) 168 (76.7) 548 (69.7) 334 (55.8) Prior basal insulin use, n (%)a 69 (63.9) 145 (66.2) 484 (61.6) 312 (52.1) Prior insulin glargine 34 (49.3) 59 (40.7) 196 (40.5) 156 (50.0) Prior NPH insulin 25 (36.2) 65 (44.8) 202 (41.7) 97 (31.1) Prior insulin detemir 9 (13.0) 21 (14.5) 81 (16.7) 51 (16.3) Prior insulin degludec 1 (1.4) 0 1 (0.2) 8 (2.6) Duration of basal insulin therapy, years 1.5 (0.7–4.3) 2.0 (0.8–4.1) 1.5 (0.7–3.7) 1.2 (0.7–3.6) Prior basal insulin dose, U/day 33.9 ± 19.3 31.9 ± 16.8 33.3 ± 21.8 37.6 ± 26.2 Previous non-insulin anti-hyperglycaemic treatment, n 93 (86.1) 194 (88.6) 726 (92.4) 576 (96.2) (%)b Biguanides 51 (54.8) 161 (83.0) 669 (92.1) 552 (95.8) Sulfonylurea 28 (30.1) 84 (43.3) 312 (43.0) 230 (39.9) Dipeptidyl peptidase 4 inhibitors 33 (35.5) 71 (36.6) 223 (30.7) 178 (30.9) Glucagon-like peptide 1 receptor agonists 15 (16.1) 18 (9.3) 99 (13.6) 90 (15.6) SGLT2 inhibitors 2 (2.2) 16 (8.2) 95 (13.1) 68 (11.8) Patients with C 1 diabetic complication, n (%) 50 (46.3) 93 (42.5) 258 (32.8) 149 (24.9) Diabetic neuropathy 36 (33.3) 66 (30.1) 192 (24.4) 85 (14.2) Diabetic retinopathy 17 (15.7) 24 (11.0) 88 (11.2) 53 (8.8) Diabetic nephropathy 32 (29.6) 35 (16.0) 41 (5.2) 44 (7.3) Patients with C 1 cardiovascular event and/or risk 93 (86.1) 195 (89.0) 661 (84.1) 451 (75.3) factor, n (%) Hypertension 91 (84.3) 183 (83.6) 601 (76.5) 382 (63.8) Dyslipidaemia 61 (56.5) 146 (66.7) 479 (60.9) 336 (56.1) Peripheral arterial disease 16 (14.8) 25 (11.4) 97 (12.3) 57 (9.5) Previous myocardial infarction 21 (19.4) 21 (9.6) 67 (8.5) 40 (6.7) Previous stroke 17 (15.7) 22 (10.0) 59 (7.5) 30 (5.0) 1166 Diabetes Ther (2021) 12:1159–1174 Table 2 continued Baseline characteristic by baseline eGFR, mL/min/ 15–44 45–59 60–89 ‡ 90 1.73 m2 (N = 108) (N = 219) (N = 786) (N = 599) Haemoglobin A1c, % 8.89 ± 0.98 8.75 ± 0.98 8.77 ± 1.00 8.75 ± 0.96 Fasting plasma glucose, mg/dL 182.0 ± 61.0 187.3 ± 61.1 185.8 ± 61.4 178.4 ± 50.2 Data are expressed as mean ± standard deviation or median (first and third quartiles), unless otherwise indicated N refers to all patients from the pooled REALI database included in the eGFR subgroup mentioned; means and percentages are calculated on the basis of data available for each variable eGFR estimated glomerular filtration rate, NPH Neutral Protamine Hagedorn, SGLT2 sodium glucose co-transporter 2 a The total number of patients who were previously treated with basal insulin in each eGFR subgroup was used as the denominator to calculate the percentages of patients who received prior insulin glargine, NPH, detemir, or degludec b The total number of patients who were previously treated with non-insulin anti-hyperglycaemic agents in each eGFR subgroup was used as the denominator to calculate the percentages of patients in each drug class Safety and similar reductions from baseline to week 24 of Gla-300 therapy in HbA1c and FPG were The incidences and event rates of hypogly- recorded across the four eGFR subgroups, indi- caemic events were overall low across the four cating that the efficacy of Gla-300 was not renal function subgroups. However, compared impacted by renal function status. with the higher eGFR subgroups, more patients Our findings are in line with post hoc anal- in the eGFR 15–44 mL/min/1.73 m2 subgroup yses of the EDITION 1, 2 and 3 trials [9] and of experienced hypoglycaemia at night or at any the BRIGHT study [8]. In the post hoc patient- time of the day during the 24-week treatment level meta-analysis [9] of the EDITION trials period with a hypoglycaemia event rate any conducted among 1247 patients with T2DM time of the day of 4.4 versus 2.7, 2.5 and 2.8 treated with once-daily evening injections of episodes per patient-year in the 45–59, 60–89 Gla-300 and who were classified into two base- and C 90 mL/min/1.73 m2 subgroups, respec- line eGFR subgroups (208 patients with an tively (Table 5). eGFR \ 60 mL/min/1.73 m2 and 1039 with an The mean ± SD change in the daily dose of eGFR C 60 mL/min/1.73 m2), the LS mean Gla-300 from baseline to weeks 12 and 24 was decrease in HbA1c after 6 months (- 0.98% and overall comparable across subgroups, with a - 1.03% for those with an eGFR \ 60 and slightly higher mean daily insulin dose increase C 60 mL/min/1.73 m2, respectively) and the in the eGFR C 90 mL/min/1.73 m2 subgroup at proportion of patients achieving HbA1c tar- both time points (Table 6). The mean ± SD get \ 7.5% (54.9% and 54.0% for those with an change in body weight from baseline to week 12 eGFR \ 60 and C 60 mL/min/1.73 m2, respec- and week 24 of Gla-300 therapy was marginal tively) were similar in both renal function sub- across the four renal function subgroups groups [9]. The meta-analysis [9] also found that (Table 6). the risk of confirmed (B 3.9 mmol/L) or severe hypoglycaemic events at any time of the day or at night was lower in Gla-300-treated patients DISCUSSION (n = 1247) than in those who received Gla-100 therapy (n = 1249), regardless of baseline renal In the REALI CKD pooled analysis of six function. No heterogeneity of treatment effects prospective studies conducted among patients was observed across renal function subgroups with T2DM, Gla-300 provided glycaemic [9]. improvement with a favourable safety profile, In the prespecified subgroup analysis [8] of regardless of baseline eGFR. Indeed, consistent the randomised BRIGHT trial conducted among Diabetes Ther (2021) 12:1159–1174 1167 Table 3 Change in HbA1c (%) from baseline to weeks 12 and 24 of Gla-300 treatment, according to baseline eGFR (mL/ min/1.73 m2) 15–44 (N = 104) 45–59 (N = 214) 60–89 (N = 766) ‡ 90 (N = 588) Mean ± SD HbA1c at 8.90 ± 0.98 8.76 ± 0.98 8.79 ± 0.99 8.77 ± 0.95 baseline Mean ± SD HbA1c at 7.89 ± 1.09 7.84 ± 1.00 7.76 ± 0.98 7.69 ± 0.95 week 12 Change from baseline to n = 94 n = 192 n = 696 n = 543 week 12 LS mean ± SE (95% - 0.83 ± 0.11 - 0.90 ± 0.09 - 0.98 ± 0.06 - 1.00 ± 0.07 CI) (- 1.06 to - 0.61) (- 1.07 to - 0.73) (- 1.11 to - 0.86) (- 1.14 to - 0.86) LS mean ± SE 0.17 ± 0.11 (- 0.05 0.10 ± 0.08 (- 0.06 0.02 ± 0.06 (- 0.09 difference* (95% CI) to 0.38) to 0.26) to 0.13) p value 0.133 0.212 0.765 Mean ± SD HbA1c at 7.74 ± 1.17 7.59 ± 1.14 7.57 ± 0.96 7.58 ± 1.09 week 24 Change from baseline to n = 97 n = 205 n = 741 n = 559 week 24 LS mean ± SE (95% - 0.99 ± 0.12 - 1.19 ± 0.09 - 1.21 ± 0.07 - 1.14 ± 0.07 CI) (- 1.22 to - 0.76) (- 1.36 to - 1.01) (- 1.34 to - 1.08) (- 1.28 to - 1.00) LS mean ± SE 0.15 ± 0.11 (- 0.07 - 0.05 ± 0.08 - 0.07 ± 0.06 difference* (95% CI) to 0.37) (- 0.21 to 0.12) (- 0.18 to 0.04) p value 0.194 0.558 0.219 CI confidence interval, eGFR estimated glomerular filtration rate, Gla-300 insulin glargine 300 U/mL, HbA1c haemoglobin A1c, LS least-squares, SD standard deviation, SE standard error *For the difference between the subgroups, the reference is the C 90 mL/min/1.73 m2 subgroup 466 insulin-naı̈ve patients with T2DM who were no differences in hypoglycaemia incidence received evening Gla-300, HbA1c reductions or rates over 24 weeks in that subgroup [8]. over 24 weeks were also similar in those with With the exception of nocturnal severe eGFR C 90 (n = 246), between 60 and 89 hypoglycaemia which was rarely reported in (n = 172), and \ 60 mL/min/1.73 m2 (n = 47). REALI CKD (with an incidence ranging from 0% However, heterogeneity of treatment effect to 0.5% across the four eGFR subgroups), the across renal function subgroups was observed incidences and event rates of hypoglycaemic (p = 0.02), reflecting a significantly greater events were higher in the eGFR 15–44 mL/min/ mean HbA1c reduction from baseline to 1.73 m2 subgroup compared with the other week 24 (8.58% to 6.94%) among Gla-300-trea- eGFR subgroups, which is not surprising and in ted patients with eGFR \ 60 mL/min/1.73 m2 line with the existing literature [8, 9, 19]. The (n = 47) compared to patients with the same observed differences between the eGFR range of eGFR randomised to IDeg-100 therapy 15–44 mL/min/1.73 m2 subgroup and the other (n = 49) (8.30–7.28%) (LS mean difference of eGFR subgroups in the event rate and incidence - 0.43%; 95% CI - 0.74 to - 0.12), while there of hypoglycaemia are unlikely to be related to 1168 Diabetes Ther (2021) 12:1159–1174 Diabetes Ther (2021) 12:1159–1174 1169 b Fig. 1 Percentage of patients achieving HbA1c targets control benefit and its overall low risk of during Gla-300 treatment, according to baseline eGFR hypoglycaemia in this challenging-to-treat (mL/min/1.73 m2). eGFR estimated glomerular filtration population that is particularly vulnerable to rate, Gla-300 insulin glargine 300 U/mL, HbA1c hypoglycaemia. This vulnerability is due to haemoglobin A1c, W week multiple factors, including a reduced renal mass and therefore a reduced capacity for renal glu- Gla-300 dose increase throughout the 24-week cose release, especially when the eGFR falls treatment period, but are consistent with the below 45 mL/min/1.73 m2 [4, 7, 21]. Hypogly- higher risk for hypoglycaemia reported in peo- caemic events affect patients’ lives profoundly, ple with CKD [9, 20]. as they are associated with psychological dis- Overall, the findings of the REALI CKD tress, cardiovascular morbidity, and a signifi- pooled analysis, along with the data from the cantly increased risk of death within 1 day of a aforementioned post hoc analyses [8, 9], sup- hypoglycaemic event [4, 7, 20]. The overall low port the use of Gla-300 in patients with renal reported incidences of hypoglycaemia associ- impairment, given its sustained glycaemic ated with Gla-300 in the present pooled analysis Table 4 Change in fasting plasma glucose (mg/dL) from baseline to weeks 12 and 24 of Gla-300 treatment, according to baseline eGFR (mL/min/1.73 m2) 15–44 (N = 104) 45–59 (N = 214) 60–89 (N = 766) ‡ 90 (N = 588) Mean ± SD FPG at 185.6 ± 60.2 187.3 ± 60.1 186.7 ± 60.8 179.3 ± 49.4 baseline Mean ± SD FPG at 139.9 ± 44.7 144.5 ± 50.1 137.3 ± 43.5 132.4 ± 39.6 week 12 Change from baseline to n = 58 n = 125 n = 506 n = 443 week 12 LS mean ± SE (95% - 32.2 ± 5.8 - 31.2 ± 4.4 - 37.0 ± 3.1 - 37.6 ± 3.2 CI) (- 43.6 to - 20.9) (- 39.7 to - 22.6) (- 43.1 to - 30.9) (- 43.9 to - 31.3) LS mean ± SE 5.4 ± 5.6 (- 5.7 to 6.4 ± 4.1 (- 1.5 to 0.6 ± 2.6 (- 4.5 to difference* (95% CI) 16.4) 14.4) 5.8) p value 0.340 0.114 0.815 Mean ± SD FPG at 134.1 ± 55.0 140.7 ± 41.4 135.1 ± 42.5 133.1 ± 41.7 week 24 Change from baseline to n = 61 n = 143 n = 555 n = 452 week 24 LS mean ± SE (95% - 40.3 ± 5.8 - 33.0 ± 4.3 - 40.8 ± 3.1 - 38.4 ± 3.2 CI) (- 51.7 to - 28.9) (- 41.4 to - 24.7) (- 46.8 to - 34.8) (- 44.6 to - 32.1) LS mean ± SE - 1.9 ± 5.7 (- 13.0 5.3 ± 4.0 (- 2.5 to - 2.5 ± 2.6 (- 7.6 difference* (95% CI) to 9.2) 13.1) to 2.7) p value 0.733 0.182 0.347 CI confidence interval, eGFR estimated glomerular filtration rate, FPG fasting plasma glucose, Gla-300 insulin glargine 300 U/mL, LS least-squares, SD standard deviation, SE standard error *For the difference between the subgroups, the reference is the C 90 mL/min/1.73 m2 subgroup 1170 Diabetes Ther (2021) 12:1159–1174 Table 5 Incidence and event rate of hypoglycaemic events, according to baseline eGFR (mL/min/1.73 m2) 15–44 (N = 104) 45–59 (N = 214) 60–89 (N = 766) ‡ 90 (N = 588) Total patient-year exposure 46.92 99.39 355.58 267.60 Any time of the day hypoglycaemia Any hypoglycaemia Patients with C 1 event, n (%) 36 (34.6) 58 (27.1) 217 (28.3) 165 (28.1) Total number of events (event rate) 206 (4.39) 272 (2.74) 899 (2.53) 753 (2.81) Symptomatic hypoglycaemia Patients with C 1 event, n (%) 28 (26.9) 36 (16.8) 160 (20.9) 119 (20.2) Total number of events (event rate) 97 (2.07) 94 (0.95) 513 (1.44) 414 (1.55) Severe hypoglycaemia Patients with C 1 event, n (%) 3 (2.9) 4 (1.9) 7 (0.9) 6 (1.0) Total number of events (event rate) 5 (0.11) 5 (0.05) 11 (0.03) 13 (0.05) Nocturnal hypoglycaemia Any hypoglycaemia Patients with C 1 event, n (%) 10 (9.6) 11 (5.1) 54 (7.0) 39 (6.6) Total number of events (event rate) 14 (0.30) 22 (0.22) 127 (0.36) 125 (0.47) Symptomatic hypoglycaemia Patients with C 1 event, n (%) 9 (8.7) 7 (3.3) 43 (5.6) 34 (5.8) Total number of events (event rate) 12 (0.26) 9 (0.09) 94 (0.26) 76 (0.28) Severe hypoglycaemia Patients with C 1 event, n (%) 0 1 (0.5) 3 (0.4) 2 (0.3) Total number of events (event rate) 0 1 (0.01) 4 (0.01) 5 (0.02) Event rates, which are based on total patient-year exposure, are expressed as number of events per year Symptomatic hypoglycaemia was defined as an event during which typical symptoms of hypoglycaemia occurred (e.g. sweating, hunger, shakiness, palpitations). Severe hypoglycaemia was defined as any event requiring assistance from another person to actively administer carbohydrates, glucagon, or take other corrective actions eGFR estimated glomerular filtration rate in both patients with reduced and preserved Although not having issued specific guideli- renal function are thus encouraging and are nes on glycaemic goals for patients with T2DM particularly beneficial during the early weeks of and CKD, the American Diabetes Association treatment, as they enable patients to titrate (ADA) recommends individualised glycaemic their basal insulin dose as intended to improve targets according to age, comorbidities, and life glycaemic control, without under- or delayed expectancy [23]. For patients with a history of titration which often results from patients’ fear severe hypoglycaemia, limited life expectancy, of hypoglycaemia [13, 22]. advanced microvascular or macrovascular complications, extensive comorbid conditions, Diabetes Ther (2021) 12:1159–1174 1171 Table 6 Gla-300 daily dose (U/day and U/kg/day) and body weight (kg) changes from baseline to weeks 12 and 24 of treatment, according to baseline eGFR (mL/min/1.73 m2) 15–44 (N = 104) 45–59 (N = 214) 60–89 (N = 766) ‡ 90 (N = 588) Gla-300 daily dose (U/day) Baseline 26.0 ± 16.0 25.7 ± 13.1 25.9 ± 15.6 26.2 ± 18.2 Week 12 33.2 ± 16.0 34.2 ± 18.2 33.9 ± 18.8 37.3 ± 22.4 Change from baseline to week 12 7.8 ± 11.3 9.0 ± 12.7 8.7 ± 11.5 11.8 ± 13.7 Week 24 35.6 ± 18.0 35.6 ± 18.8 36.3 ± 20.3 39.4 ± 23.1 Change from baseline to week 24 10.4 ± 13.0 10.4 ± 13.0 10.5 ± 13.8 13.6 ± 16.2 Gla-300 daily dose (U/kg/day) Baseline 0.29 ± 0.16 0.29 ± 0.14 0.29 ± 0.16 0.28 ± 0.17 Week 12 0.37 ± 0.16 0.38 ± 0.17 0.38 ± 0.18 0.41 ± 0.21 Change from baseline to week 12 0.06 ± 0.11 0.09 ± 0.12 0.10 ± 0.13 0.13 ± 0.14 Week 24 0.40 ± 0.20 0.40 ± 0.18 0.41 ± 0.20 0.43 ± 0.21 Change from baseline to week 24 0.11 ± 0.16 0.11 ± 0.14 0.12 ± 0.15 0.15 ± 0.16 Body weight (kg) Baseline 91.0 ± 17.9 85.7 ± 14.3 87.5 ± 14.9 88.4 ± 16.9 Week 12 92.3 ± 16.6 86.1 ± 14.7 87.8 ± 15.0 88.8 ± 16.8 Change from baseline to week 12 - 0.5 ± 5.5 0.1 ± 2.0 - 0.08 ± 2.4 0.2 ± 2.6 Week 24 90.5 ± 17.5 85.6 ± 14.8 87.8 ± 14.8 88.8 ± 16.9 Change from baseline to week 24 - 0.7 ± 4.6 - 0.04 ± 2.3 0.1 ± 3.6 0.2 ± 3.7 All data are expressed as mean ± standard deviation eGFR estimated glomerular filtration rate, Gla-300 insulin glargine 300 U/mL or longstanding diabetes, HbA1c goals that are 1.73 m2 subgroup, occurred mainly in the first less stringent than the general target of \ 7.0% 12 weeks of Gla-300 treatment. This might be a (i.e. \ 8.0%) are suggested [23]. Similarly, the result of the usual dose escalation during the National Kidney Foundation (NKF) recom- early weeks of basal insulin therapy parallel to mends that in patients at risk for hypogly- titration according to fasting glucose targets. caemia, such as those with diabetes and The present pooled analysis has several impaired renal function, target HbA1c should strengths including the prospective nature and be extended above 7.0% [24]. In this pooled the high completion rate (C 90%) of the eval- analysis, approximately half of the patients in uated studies. This pooled analysis also inclu- the four eGFR subgroups achieved target HbA1c ded a large, diverse population of patients with values \ 7.5% at 24 weeks of Gla-300 therapy, T2DM, with a variety of baseline characteristics and approximately two-thirds achieved an that are typically encountered in a clinical set- HbA1c target \ 8.0%, in line with the recom- ting. Most importantly, the REALI CKD pooled mendations of the NKF [24] and the ADA [23]. analysis provides valuable information sup- Interestingly, the HbA1c targets achievement, porting the use of Gla-300 in individuals with particularly the achievement of HbA1c T2DM and renal impairment, a patient popu- levels \ 7.0% in the eGFR 15–44 mL/min/ lation that is generally understudied and for 1172 Diabetes Ther (2021) 12:1159–1174 which clinical study results are not necessarily Funding. This study was funded by Sanofi similar to those reported for the general T2DM (Paris, France). The article processing charges population. However, our findings should be were also funded by Sanofi (Paris, France). interpreted in the context of certain limitations. The number of patients in the evaluated renal Medical Writing and/or Editorial Assis- functions subgroups was not controlled, and tance. Medical writing support in accordance subsequently there were substantially fewer with Good Publication Practice (GPP3) guideli- patients in the 15–44 and 45–59 mL/min/ nes (https://www.ismpp.org/gpp3) was pro- 1.73 m2 groups than the 60–89 and C 90 mL/ vided by Thomas Rohban, M.D. and Magalie El min/1.73 m2 groups. Other limitations include Hajj, PharmD, of Partner 4 Health (Paris, the post hoc nature of the analysis, the exclu- France), and was funded by Sanofi. sion of data from participants with a baseline eGFR \ 15 mL/min/1.73 m2 (stage 5 CKD), and Authorship. All named authors meet the the lack of comparative data with another basal International Committee of Medical Journal insulin. Furthermore, the fact that race was not Editors (ICMJE) criteria for authorship for this taken into account in the calculations of eGFR article, take responsibility for the integrity of might affect the accuracy of renal function the work as a whole, and have given their assessment. However, given the geographical approval for this version to be published. coverage of the studies, we assumed that a very small number of patients of African descent Authorship Contributions. All authors con- were enrolled. Longer-term assessment (beyond tributed to the study design and the analysis 24 weeks) would have also been helpful in plan. Celine Mauquoi performed the statistical evaluating the long-term effectiveness and analysis of the data. All authors were involved safety of Gla-300 therapy according to renal in the interpretation of the data, writing and function. Lastly, the results of REALI could have reviewing drafts of the manuscript, and been potentially impacted by the concomitant approved the final version for submission. administration of Gla-300 with oral antidiabetic Prior Presentation. Preliminary results were drugs in some patients. Nevertheless, the find- presented at the American Diabetes Association ings of the REALI CKD pooled analysis support (ADA) 79th Scientific Sessions, 07–11 June 2019, Gla-300 as a therapeutic option for individuals San Francisco, CA. with T2DM and renal impairment. Disclosures. Didac Mauricio has received CONCLUSIONS funding for consulting or speaker fees from Almirall, Ascensia, AstraZeneca, Boehringer Kidney function does not seem to affect the Ingelheim, GlaxoSmithKline, Eli Lilly, Ferrer, glycaemic effectiveness and safety of Gla-300, as Janssen, Menarini, Merck Sharp & Dohme, individuals with inadequately controlled T2DM Novartis, Novo Nordisk, and Sanofi. Pierre and renal impairment who were initiated on or Gourdy has received advisory board and speaker switched to Gla-300 achieved a clinically honoraria from Abbott, Amgen, AstraZeneca, important improvement in glycaemic control Novo Nordisk, Boehringer Ingelheim, Eli Lilly, with a low hypoglycaemia risk, especially for Merck Sharp & Dohme, Mundipharma, Sanofi, severe hypoglycaemia. and Servier. Riccardo C. Bonadonna has received funding for consulting or speaker fees from Sanofi, Merck Sharp & Dohme, Bristol- Myers Squibb, AstraZeneca, Janssen, Eli Lilly, ACKNOWLEDGEMENTS and Johnson & Johnson. Nick Freemantle has received funding for consulting, research or The authors thank the participants, staff, and speaker fees from Sanofi, Novo Nordisk, Aller- investigators of all pooled studies. gan, Merck Sharp & Dohme, AstraZeneca, PCT, Diabetes Ther (2021) 12:1159–1174 1173 Ipsen, Takeda, Akcea, and Accelovance. Gregory REFERENCES Bigot is an IVIDATA employee for Sanofi. Alice Ciocca and Mireille Bonnemaire are Sanofi 1. United States Renal Data System. Chapter 1: CKD in employees. Celine Mauquoi is an IDDI the general population. Am J Kidney Dis. 2019;73(3 employee, and has acted as a biostatistics con- Suppl 1):S1–28. tractor for Sanofi. Dirk Müller-Wieland has 2. Kainz A, Hronsky M, Stel VS, et al. Prediction of received funding for consulting or speaker fees prevalence of chronic kidney disease in diabetic from Amarin, Amgen, Bayer AG, AstraZeneca, patients in countries of the European Union up to Boehringer Ingelheim, Daiichi-Sankyo, Lilly, 2025. Nephrol Dial Transplant. 2015;30(Suppl 4): Merck Sharp & Dohme, Novo Nordisk, and iv113–8. Sanofi. 3. Nogueira C, Souto SB, Vinha E, Braga DC, Carvalho D. Oral glucose lowering drugs in type 2 diabetic Compliance with Ethics Guidelines. This patients with chronic kidney disease. Hormones analysis did not involve primary data collection (Athens). 2013;12(4):483–94. by the authors, and all pooled analysed data 4. Alsahli M, Gerich JE. Hypoglycemia in patients with were de-identified; consequently, separate ethi- diabetes and renal disease. J Clin Med. 2015;4(5): cal approval was not required. Protocols for all 948–64. included studies were approved by the appro- priate ethics committees, and the studies were 5. Marı́n-Peñalver JJ, Martı́n-Timón I, Sevillano-Col- lantes C, Del Cañizo-Gómez FJ. Update on the conducted according to Good Clinical Practice treatment of type 2 diabetes mellitus. World J Dia- and the Declaration of Helsinki. All participants betes. 2016;7(17):354–95. provided written informed consent prior to participation in the studies. 6. Rao Kondapally Seshasai S, Kaptoge S, Thompson A, et al. Diabetes mellitus, fasting glucose, and risk of cause-specific death. N Engl J Med. 2011;364(9): Data Availability . The datasets generated 829–41. during and/or analysed during the current study are available from the corresponding 7. Betônico CC, Titan SM, Correa-Giannella ML, Nery M, Queiroz M. Management of diabetes mellitus in author on reasonable request. individuals with chronic kidney disease: therapeu- tic perspectives and glycemic control. Clinics (Sao Open Access. This article is licensed under a Paulo). 2016;71(1):47–53. Creative Commons Attribution-Non- Commercial 4.0 International License, which 8. Haluzı́k M, Cheng A, Müller-Wieland D, et al. Dif- ferential glycaemic control with basal insulin glar- permits any non-commercial use, sharing, gine 300 U/mL versus degludec 100 U/mL adaptation, distribution and reproduction in according to kidney function in type 2 diabetes: a any medium or format, as long as you give subanalysis from the BRIGHT trial. Diabetes Obes appropriate credit to the original author(s) and Metab. 2020;22(8):1369–77. the source, provide a link to the Creative 9. Javier Escalada F, Halimi S, Senior PA, et al. Gly- Commons licence, and indicate if changes were caemic control and hypoglycaemia benefits with made. The images or other third party material insulin glargine 300 U/mL extend to people with in this article are included in the article’s type 2 diabetes and mild-to-moderate renal impairment. Diabetes Obes Metab. 2018;20(12): Creative Commons licence, unless indicated 2860–8. otherwise in a credit line to the material. If material is not included in the article’s Creative 10. Bonadonna RC, Giaccari A, Buzzetti R, et al. Italian Commons licence and your intended use is not Titration Approach Study (ITAS) with insulin glar- gine 300 U/mL in insulin-naı̈ve type 2 diabetes: permitted by statutory regulation or exceeds the design and population. Nutr Metab Cardiovasc Dis. permitted use, you will need to obtain permis- 2019;29(5):496–503. sion directly from the copyright holder. To view a copy of this licence, visit http:// 11. Becker RH, Dahmen R, Bergmann K, Lehmann A, Jax T, Heise T. New insulin glargine 300 Units  mL-1 creativecommons.org/licenses/by-nc/4.0/. provides a more even activity profile and prolonged glycemic control at steady state compared with 1174 Diabetes Ther (2021) 12:1159–1174 insulin glargine 100 Units  mL-1. Diabetes Care. settings in the Czech Republic. J Diabetes Mellitus. 2015;38(4):637–43. 2020;10(3):109–23. 12. Ritzel R, Roussel R, Bolli GB, et al. Patient-level 18. Perez A, Carrasco-Sánchez FJ, González C, et al. meta-analysis of the EDITION 1, 2 and 3 studies: Efficacy and safety of insulin glargine 300 U/mL glycaemic control and hypoglycaemia with new (Gla-300) during hospitalization and therapy insulin glargine 300 U/ml versus glargine 100 U/ml intensification at discharge in patients with insuf- in people with type 2 diabetes. Diabetes Obes ficiently controlled type 2 diabetes: results of the Metab. 2015;17(9):859–67. phase IV COBALTA trial. BMJ Open Diabetes Res Care. 2020;8(1):e001518. 13. Rosenstock J, Cheng A, Ritzel R, et al. More simi- larities than differences testing insulin glargine 300 19. Majumder A, Roychaudhuri S, Sanyal D. A retro- Units/mL versus insulin degludec 100 Units/mL in spective observational study of insulin glargine in insulin-naive type 2 diabetes: the randomized head- type 2 diabetic patients with advanced chronic to-head BRIGHT trial. Diabetes Care. 2018;41(10): kidney disease. Cureus. 2019;11(11):e6191. 2147–54. 20. Moen MF, Zhan M, Hsu VD, et al. Frequency of 14. Freemantle N, Bonadonna RC, Gourdy P, et al. hypoglycemia and its significance in chronic kid- Rationale and methodology for a European pooled ney disease. Clin J Am Soc Nephrol. 2009;4(6): analysis of postmarketing interventional and 1121–7. observational studies of insulin glargine 300 U/mL in diabetes: protocol of REALI project. BMJ Open. 21. Cheng AYY, Wong J, Freemantle N, Acharya SH, 2020;10(4):e033659. Ekinci E. The safety and efficacy of second-genera- tion basal insulin analogues in adults with type 2 15. Russell-Jones D, Dauchy A, Delgado E, et al. Take diabetes at risk of hypoglycemia and use in other Control: a randomized trial evaluating the efficacy special populations: a narrative review. Diabetes and safety of self- versus physician-managed titra- Ther. 2020;11(11):2555–93. tion of insulin glargine 300 U/mL in patients with uncontrolled type 2 diabetes. Diabetes Obes Metab. 22. Ghosh S, Ghosh R. Glargine-300: an updated liter- 2019;21(7):1615–24. ature review on randomized controlled trials and real-world studies. World J Diabetes. 2020;11(4): 16. Gourdy P, Bahloul A, Boultif Z, Gouet D, Guerci B. 100–14. Efficacy and safety of switching patients inade- quately controlled on basal insulin to insulin glar- 23. American Diabetes Association. 2021. 6. Glycemic gine 300 U/mL: the transition 2 study. Diabetes targets: standards of medical care in diabetes— Ther. 2020;11(1):147–59. 2021. Diabetes Care. 2021;44(Suppl 1):S73–84. 17. Prázný M, Flekač M, Jelı́nek P, Mašková J. Insulin 24. National Kidney Foundation. KDOQI clinical prac- glargine 300 Units/mL effectiveness in patients tice guideline for diabetes and CKD: 2012 update. with T2DM uncontrolled by basal insulin in real-life Am J Kidney Dis. 2012;60(5):850–86.