Comparison of Insulin Glargine300 Units/mL and 100 Units/mLin Adults With Type 1 Diabetes:Continuous Glucose MonitoringProfiles and Variability UsingMorning or Evening InjectionsDiabetes Care 2017;40:554–560 | DOI: 10.2337/dc16-0684

OBJECTIVE

The objective of this study was to compare glucose control in participants with type 1diabetes receiving insulin glargine 300 units/mL (Gla-300) or glargine 100 units/mL(Gla-100) in the morning or evening, in combination with mealtime insulin.

RESEARCH DESIGN AND METHODS

In this 16-week, exploratory, open-label, parallel-group, two-period crossoverstudy (clinicaltrials.gov identifier NCT01658579), 59 adults with type 1 diabeteswere randomized (1:1:1:1) to once-daily Gla-300 or Gla-100 given in the morningor evening (with crossover in the injection schedule). The primary efficacyend point was the mean percentage of time in the target glucose range(80–140 mg/dL), as measured using continuous glucose monitoring (CGM), duringthe last 2 weeks of each 8-week period. Additional end points included other CGMglycemic control parameters, hypoglycemia (per self-monitored plasmaglucose [SMPG]), and adverse events.

RESULTS

The percentage of time within the target glucose range was comparable betweenthe Gla-300 and Gla-100 groups. There was significantly less increase in CGM-based glucose during the last 4 h of the 24-h injection interval for Gla-300 com-pared with Gla-100 (least squares mean difference 214.7 mg/dL [95% CI 226.9to 22.5]; P = 0.0192). Mean 24-h glucose curves for the Gla-300 group weresmoother (lower glycemic excursions), irrespective of morning or evening injec-tion. Four metrics of intrasubject interstitial glucose variability showed no differ-ence between Gla-300 and Gla-100. Nocturnal confirmed (<54mg/dL by SMPG) orsevere hypoglycemia rate was lower for Gla-300 participants than for Gla-100participants (4.0 vs. 9.0 events per participant-year; rate ratio 0.45 [95% CI0.24–0.82]).

CONCLUSIONS

Less increase in CGM-based glucose levels in the last 4 h of the 24-h injectioninterval, smoother average 24-h glucose profiles irrespective of injection time,and reduced nocturnal hypoglycemia were observed with Gla-300 versus Gla-100.

1Park Nicollet International Diabetes Center,Minneapolis, MN2AMCR Institute, Escondido, CA3Biomedical Informatics Consultants LLC, Potomac,MD4Sanofi-Aventis Deutschland GmbH, Frankfurtam Main, Germany5BDM Consulting Inc., Somerset, NJ6Oregon Health & Science University, Portland,OR

Corresponding author: Richard M. Bergenstal,richard.bergenstal@parknicollet.com.

Received 29 March 2016 and accepted 26December 2016.

Clinical trial reg. no.NCT01658579, clinicaltrials.gov.

This article contains Supplementary Data onlineat http://care.diabetesjournals.org/lookup/suppl/doi:10.2337/dc16-0684/-/DC1.

This article is featured in a podcast available athttp://www.diabetesjournals.org/content/diabetes-core-update-podcasts.

© 2017 by the American Diabetes Association.Readers may use this article as long as the workis properly cited, the use is educational and notfor profit, and the work is not altered. More infor-mation is available at http://www.diabetesjournals.org/content/license.

RichardM. Bergenstal,1 Timothy S. Bailey,2

David Rodbard,3 Monika Ziemen,4

Hailing Guo,5 Isabel Muehlen-Bartmer,4

and Andrew J. Ahmann6

554 Diabetes Care Volume 40, April 2017

EMER

GINGTECHNOLO

GIESANDTH

ERAPEU

TICS

Despite advances in basal insulin ther-apy, many people with type 1 diabetesstill experience marked day-to-day dif-ferences in glucose levels as well as glu-cose excursions within the same day.These often unpredictable fluctuationsin glucose levels make it difficult to op-timize insulin doses and reach desiredglycemic targets. It is also well recog-nized that hypoglycemia is a limiting fac-tor when intensifying insulin therapy(1). Therefore, a basal insulin that leadsto more stable glucose control with areduced risk for hypoglycemia wouldprovide a distinct clinical advantage.A pharmacokinetic (PK)/pharmacody-

namic (PD) euglycemic clamp study atsteady state in people with type 1 dia-betes showed that the basal insulinanalog insulin glargine 300 units/mL(Gla-300) provides more consistent andprolonged insulin exposure comparedwith insulin glargine 100 units/mL (Gla-100), resulting in blood glucose controlthat lasts well beyond 24 h (2). A secondeuglycemic clamp study demonstratedpredictable and stable 24-h glycemic cov-erage by Gla-300 as a result of low fluctu-ation and high reproducibility of insulinexposure (3). Continuous glucosemonitor-ing (CGM), which records interstitial fluidglucose levels every 5 min throughout theday and night, is a valuable way to confirmwhether the differences observed in thePK and PD properties of insulins under eu-glycemic clamp conditions translate toclinically relevant differences in their 24-hglucose profiles and other parameters ofglycemic control, including hypoglycemia.CGM was used in this study by 59 adult

participants with type 1 diabetes over twosuccessive8-weekperiods toassessglucosecontrol, safety, and tolerability of Gla-300comparedwithGla-100whenadministeredin the morning or evening.

RESEARCH DESIGN AND METHODS

Study DesignThis exploratory, 16-week, open-label,phase II, parallel-group, two-period cross-over study of participants with type 1 di-abetes was conducted in three studycenters in the U.S. from August 2012 toMay 2013. After a 4-week screeningphase, participants were randomized1:1:1:1, using a remote telephone system,to receive treatment with Gla-300 orGla-100 (both Sanofi, Paris, France) inthe morning or evening during treatmentperiod A (weeks 1–8); participants then

crossed over to the alternate injectionschedule (evening or morning) for treat-ment period B (weeks 9–16).

CGM (using the Dexcom Seven PlusCGM system [Dexcom, San Diego, CA])was performed throughout the 16-weektreatment period. Participants weremasked to their CGM data. Data fromthe last 2 weeks of each 8-week treatmentperiod (A andB)were analyzed (weeks 7–8andweeks 15–16 combined). The study de-sign is summarized in Supplementary Fig. 1.

The study protocol was approved bythe appropriate local or central indepen-dent ethics committees or institutional re-view boards, and the studywas conductedaccording to Good Clinical Practice guide-lines and the Declaration of Helsinki. Allparticipants provided written informedconsent.

Key Inclusion and Exclusion CriteriaAdult participants ($18 and ,70 yearsof age at screening) diagnosed withtype 1 diabetes and receiving any basalinsulin regimen and mealtime insulinanalog for at least 1 year were eligiblefor inclusion. Exclusion criteria includedHbA1c .9.0% at screening; not taking astable insulin dose in the 30 days beforescreening; use of an insulin pump within6 months before screening; use of pre-mixed insulin, human regular insulin asmealtime insulin, and/or any antihyper-glycemic drugs other than an insulin an-alog at mealtime and basal insulinwithin 3 months before screening; andany contraindication to insulin glargine.

InterventionsParticipants self-administered sub-cutaneous injections of Gla-300 or Gla-100 once daily, at the same time eachdaydeither in the morning (immediatelybefore breakfast until lunch) or evening(immediately before the eveningmeal un-til bedtime)daccording to their assignedschedule in each treatment period. Injec-tions were administered using commer-cially available insulin syringes because aninsulin pen that could deliver the smallvolumes of Gla-300 required was notavailable when the study was conducted.All information pertaining to the time,dose, and location of the injection of basalinsulin treatment as well as the time anddose of mealtime insulin were recordeddaily in participant diaries.

Participants receiving basal insulintwice daily before study entry were

changed to a once-daily regimen atscreening (week22). For treatment pe-riod A, the basal insulin dose on the daybefore randomization and the medianfasting self-measured plasma glucose(SMPG) value before breakfast duringthe 3 days before baseline were usedto calculate the starting dose. For treat-ment period B, the starting dose wascalculated using the basal insulin doseon the day before the first study visitof treatment period B (week 9) and theSMPG before breakfast during the3 days before week 9.

The basal insulin dose was titrated nomore often than every 3 to 4 days duringthe first 6weeks of each treatment period(A andB) to reach the target fasting SMPGof 80–130mg/dL (4.4–7.2mmol/L), and itwas optimized by the investigators usingCGM data (downloaded at the study vis-its). In each treatment period, the 6-weektitration phase was followed by a 2-weekmaintenance phase (weeks 7–8 andweeks 15–16) inwhich basal insulin doseswere to remain as constant as clinicallypossible (representing a steady-state sce-nario to better allow comparison of thetwo basal insulins), and in which mostglycemic corrections were made usingthe mealtime insulin.

Each participant continued to use thesame rapid-acting insulin analog used inthe 3 months before screening, with ad-justments to achieve a target 2-h postpran-dial SMPG of,160mg/dL (,8.9 mmol/L).Glycemic targets were adapted for individ-ual participants, if deemed necessary.

End PointsThe primary end point was the meanpercentage of time within the predefinedCGM glucose range of 80–140 mg/dL(4.4–7.8 mmol/L) during the last 2 weeksof each treatment period. Secondaryend points based on CGM included themean percentage of time with glucose,80 mg/dL, the mean percentage oftime with glucose .140 mg/dL, meanglucose levels, mean and variation in glu-cose profiles, and glucose variability met-rics. Mean change in HbA1c (NGSP units)from baseline to week 16 (analyzed by acentral laboratory) and insulin dose (unitsper kilogram; total daily dose, basal dose,mealtime/bolus dose) were assessed.

Participant-reported hypoglycemicevents were collected for the entireon-treatment study period and ana-lyzed by categories as defined by the

care.diabetesjournals.org Bergenstal and Associates 555

American Diabetes Association (4); theywere classified as occurring during thenight (0000–0559 h) and at any time ofday (24 h). The number of hypoglycemicevents per participant-year were calcu-lated using plasma glucose concentrationcutoffs of #70 mg/dL (#3.9 mmol/L)and ,54 mg/dL (,3.0 mmol/L). Toconfirm a hypoglycemic event, partici-pants were instructed to measure capil-lary plasma glucose levels (SMPG) beforethe administration of carbohydrateswhenever symptomatic hypoglycemiawas suspected, unless safety consider-ations required immediate glucose rescuebefore confirmation. Treatment-emergentadverse events (AEs), both nonserious andserious, were evaluated.

Data Analysis and StatisticsNo formal sample size estimation wasperformed for this first-of-its-kind head-to-head comparison of basal insulins usingCGM in a crossover exploratory study.Assuming a 15% withdrawal rate, weplanned to enroll approximately 56 partici-pants to achieve 48 evaluable participants.Unless otherwise specified, the effi-

cacy results were analyzed using CGMdata from the last 2 weeks of each treat-ment period (weeks 7–8 andweeks 15–16,when basal insulin doses were to remainas constant as possible and most glycemiccorrections were made using mealtime in-sulin) by treatment group overall (i.e.,pooled morning and evening injectionschedules); some end points were also an-alyzed by treatment group and injectionschedule (morning vs. evening).The modified intent-to-treat (mITT)

population included all randomized par-ticipants who received at least one doseof studymedication and had at least oneefficacy assessment after the baseline.The primary efficacy population (theCGM population) included all partici-pants from the mITT population whohad evaluable CGM data after the base-line. The safety populationwas defined asall randomized participants who receivedat least one dose of study medication.All summaries and statistical analyses

were generatedusing SAS software, version9.2. The efficacy analyses are presentedusing last-observation-carried-forwardimputation. The primary end point wasanalyzed using a mixed model with re-peated measurements, with treatmentand period as fixed effects and partici-pant as the random effect. The model

was fitted to all the data simultaneously,and from this model the relevant treat-ment differences were estimated asleast squares (LS) means with SEs. Statis-tical comparisons were performedusing a two-sided test with a nominal5% significance level. Mean glucoselevels and 24-h glucose profiles weregenerated by calculating the meanCGM-based glucose level pooled acrossall participants within each treatmentgroup, reported overall and (for 24-hglucose profiles) by injection schedule.CGM-detected hypoglycemic eventswere defined as one or more continu-ously measured interstitial glucose valueeither #70 or ,54 mg/dL.

Post Hoc Analyses

To confirm the more stable and pro-longed duration of the glucose-loweringactivity of Gla-300, the mean and SD ofthe change per participant (D) in CGM-based glucose level in the last 4 h of the24-h injection interval was calculated forthe last 2 weeks of each treatment pe-riod, compared between Gla-300 andGla-100 overall (using mixed model withrepeated measurements analysis), anddescribed by treatment group for morn-ing and evening injections (D = glucoselevel 0–5 min before injection 2 glucoselevel 4 h before injection).

Intrasubject glucose variability met-rics were analyzed by treatment group.Metrics assessed included total SD of allglucose values over all days and times(SDT), within-day SD (SDw; averagedover all days), SD of daily means (SDdm),and SD between days (SDb) for any spec-ified time of day (5). Each variability met-ric is presented by treatment group as themean of the SD for all participants in eachgroup.

RESULTS

Study PopulationOf 85 participants with type 1 diabetesenrolled in the study, 59were randomizedto treatment with Gla-300 (n = 30) or Gla-100 (n = 29) and included in the mITTpopulation; 29 (97%) in the Gla-300 groupand 26 (90%) in the Gla-100 group com-pleted the study (Supplementary Fig. 2).Of the four participants who discontinuedthe study, one (1.7%) in theGla-300 groupwas discontinued because of pregnancy(details are provided in the ADVERSE EVENTS

section), and three (5.1%) in the Gla-100group were discontinued because of“other” non-safety-related reasons.Baseline characteristics were similaracross treatment groups. Characteristicsof the entire study population weremean age 44.2 years; mean duration ofdiabetes 22.1 years; mean HbA1c 7.46%(58.0 mmol/mol); and prior total dailyinsulin dose 0.6 units/kg/day (Supple-mentary Tables 1 and 2).

Glycemic Control (as Measured byCGM)The mean percentage of time within thepredefined target glycemic range of 80–140 mg/dL during the last 2 weeks oftreatment (primary end point) was com-parable between the Gla-300 and Gla-100 groups (LS mean [SE] 31.8% [1.5]vs. 31.0% [1.6], respectively; LS meandifference 0.75% [95% CI 23.61 to5.12]; P = 0.73) (Fig. 1). There was alsono difference between the Gla-300 andGla-100 treatment groups in the meanpercentage of time in the target rangewhen the injection was in the morning(mean [SE] 31.6% [1.8] vs. 31.5% [1.7])or in the evening (32.0% [1.7] vs. 30.5%[1.8]). There was no meaningful differencebetween treatment groups in terms of

Figure 1—Mean percentage of time within glucose ranges during the last 2 weeks of eachtreatment period overall among the CGM population.

556 CGM Profiles of Gla-300 vs. Gla-100 Diabetes Care Volume 40, April 2017

the percentage of time spent at values,80mg/dL (LS mean difference 21.6% [95%CI24.61 to 1.36];P =0.28) or.140mg/dL(LS mean difference 0.87% [95%CI 25.22 to 6.96]; P = 0.78).CGM-detected low interstitial glucose

values during the entire on-treatmentperiod in the Gla-300 and the Gla-100groups are shown in SupplementaryFig. 3A (at either threshold [#70and ,54 mg/dL] and by injection time)and Supplementary Fig. 3B.

24-h Glucose Profile

Figure 2 shows the mean 24-h glucoseprofiles obtained by CGM during the last2 weeks of each treatment period for par-ticipants receiving Gla-300 and Gla-100,pooled across participants in each treat-ment group. The profiles were smootherwith Gla-300, showing smaller differencesin glucose levels throughout the 24-h

period than with Gla-100 (difference be-tween the daily minimum and maximumvalues: Gla-300, 14 mg/dL; Gla-100,28 mg/dL) (Fig. 2A). This was particu-larly evident when comparing by injec-tion time: profiles of the Gla-300morning and evening injection groupswere almost able to be superimposed(Fig. 2B), whereas there were largerglycemic excursions in the Gla-100morning injection group (minimum/maximum values, 149/189 mg/dL) com-pared with the evening injection group(minimum/maximum values, 162/183mg/dL) (Fig. 2C).

Glycemic Control in the Last 4 h of the

24-h Injection Interval

In a post hoc sensitivity analysis, themean (SD) change (D) in glucose levelduring the last 4 h of each participant’s24-h injection interval showed significantly

less increase for Gla-300 than for Gla-100(10.9 [24.5] vs. 26.5 [21.0] mg/dL, respec-tively; LS mean difference betweengroups, 214.7 mg/dL [95% CI 226.9to 22.5]; P = 0.0192). The difference infavor of Gla-300 was observed regardlessof injection time.

Mean Glucose

Mean interstitial glucose levels decreasedover the 16-week treatment period in theGla-300 group, from182.1mg/dL at base-line to 165.0 mg/dL at week 16 (meanchange from baseline 213.2 mg/dL),and in the Gla-100 group, from 172.6mg/dL at baseline to 169.3 mg/dL atweek 16 (mean change from base-line 22.7 mg/dL).

Glucose Variability Metrics

All metrics assessing the intrasubjectCGM glucose variability, whether withindays or between days (SDT, SDW, SDdm,and SDb), showed no statistical differ-ence in glucose measurements for thosereceiving Gla-300 compared with thosereceivingGla-100during the last 2weeksof each treatment period (Fig. 3).

HbA1c

At baseline,meanHbA1c was similar in bothtreatment groups (Gla-300, 7.51%; Gla-100,7.41%). The mean change from baselineto week 16 was 20.44% (95% CI 20.64to 20.24) in the Gla-300 group, a statisti-cally significant decrease, and 20.22%(20.45 to 0.01) in the Gla-100 group.

Daily Insulin DoseThe total (basal + mealtime) daily insulindose throughout the study remained rel-atively stable in both groups (Gla-300group: baseline, 0.68 units/kg and week16, 0.67 units/kg; Gla-100 group: base-line, 0.59 units/kg and week 16, 0.63units/kg). The daily basal insulin dose in-creased slightly in both groups from base-line to week 16 (Gla-300 group: baseline,0.30 units/kg andweek 16, 0.35 units/kg;Gla-100 group: baseline, 0.30 units/kg andweek 16, 0.33 units/kg) (SupplementaryTable 2).

Safety

Hypoglycemia (by Participant-Recorded

SMPG)

Throughout the 16-week study, the rateof confirmed (,54mg/dL) or severe hypo-glycemia was lower during the noctur-nal interval (0000–0559 h) with Gla-300versus Gla-100 (Gla-300: 4.0 events perparticipant-year, Gla-100: 9.0 events per

Figure 2—Mean glucose profile over 24 h during the last 2 weeks of each treatment period,pooled across participants in the CGM population: Gla-300 vs. Gla-100 overall (A); Gla-300 byinjection schedule (B); Gla-100 by injection schedule (C). Data displayed aremean hourly glucosevalues by time of day, pooled across all participants within each treatment group and time ofadministration. The postprandial SMPG target used in this study (160mg/dL) is represented by ahorizontal black line to better enable comparisons between panels. Arrows represent the meantime of morning and evening injections.

care.diabetesjournals.org Bergenstal and Associates 557

participant-year; rate ratio 0.45 [95% CI0.24–0.82]) (Fig. 4). Rates of such eventsat any time of day (24 h) were compa-rable between the Gla-300 and Gla-100groups. When assessing hypoglycemiaat the less stringent glycemic threshold(#70 mg/dL), rates of confirmed or se-vere events, both at any time of day(24 h) and during the nocturnal interval(0000–0559 h), were comparable be-tween the Gla-300 and Gla-100 groups.There were no consistent differencesbetween morning and evening injectionwithin each treatment group (data notshown).One participant (3.3%) receiving Gla-

300 and three participants (10.3%) re-ceiving Gla-100 each reported onesevere hypoglycemic event (definedas requiring assistance from anotherperson). The participant in the Gla-300 group experienced confusion andwas treated with glucagon; the three

participants in the Gla-100 group re-ceived oral carbohydrate.Adverse Events

In total, 24 participants (80%) receivingGla-300 and 19 participants (66%)receiving Gla-100 reported one or moretreatment-emergent AE. The most com-monly reported AEs were nasopharyngi-tis, headache, pyrexia, and influenza(Supplementary Table 3). One partici-pant in the Gla-300 group who had a his-tory of repeated abdominal obstructionexperienced a serious treatment-emergentAE (intestinal obstruction), which was notconsidered to be related to the study treat-mentand resolved following treatment.An-other participant in the Gla-300 group waswithdrawn from the study owing to a treat-ment-emergent AE of pregnancy. At thetime the database was locked, the preg-nancywas continuingwithout any reportedabnormalities. A healthy child was prema-turely delivered at 33 weeks’ gestation.

CONCLUSIONS

A fundamental goal of type 1 diabetesmanagement is to maintain glucose lev-els within a target range and tominimizevulnerability to hypoglycemic or hyper-glycemic events that can lead to short-and/or long-term health complications(6–8). To achieve this, the “ideal” basalinsulin should provide stable glucose-lowering activity over an entire 24-h pe-riod and help maintain target glucoselevels in the fasting state and beforemeals (9). The basal insulin can thenbe paired with a rapid-acting insulinbefore a meal that is adjusted to opti-mize glycemic control after the meal(10). The first generation of basal insulinanalogs have substantially improvedglucose management, offering a pro-longed duration of action and a lowerrisk of hypoglycemia, particularly over-night, compared with NPH insulin (11).However, only a small percentage of in-dividuals with type 1 diabetes who usethese first-generation analogs achieveglycemic targets, and excursions intoclinically dangerous hypoglycemic andhyperglycemic states are common (12).

The results of these CGM analysesconsistently demonstrated a moreeven distribution of glucose-loweringactivity throughout the entire 24-h in-jection interval with Gla-300 comparedwith Gla-100 based on glucose profilespooled across individuals. Althoughthere was no difference between Gla-300 and Gla-100 regarding the timespent in the predefined target glycemicrange of 80–140 mg/dL during weeks7–8 and weeks 15–16 combined (Gla-300: 31.8%; Gla-100: 31.0%), the relative-ly low percentage of time in the targetglycemic range in both treatment groupsmay be attributed to the upper limit be-ing defined as 140 mg/dL, which is lowerthan thepostprandial target of 160mg/dLused in this study and considerably lowerthan the 180 mg/dL after-meal upperlimit set by the American Diabetes Asso-ciation (7).

The mean 24-h glucose profile, aver-aged for all participants on CGM in eachgroup by time of day, showed a nar-rower range of daily interstitial glucoselevels for Gla-300 than for Gla-100. Thisdifference in profile was even more ev-ident when morning and evening injec-tion groups were compared; the glucoseprofiles of Gla-300 morning and evening

Figure 3—Glucose variability metrics during the last 2 weeks of each treatment period for theCGM population, showing the relative difference between Gla-300 and Gla-100.

Figure 4—Ratios of annualized rates (events per participant-year) of confirmed (#70and ,54 mg/dL) or severe hypoglycemia (based on case report form data) among the safetypopulation.

558 CGM Profiles of Gla-300 vs. Gla-100 Diabetes Care Volume 40, April 2017

injections are nearly superimposable,whereas larger excursions were seen inthe morning injection group comparedwith the evening injection group forGla-100. The increase in glucose levelsfrom 020020800 h in the Gla-100morn-ing injection group may be more pro-nounced than any glucose increasesseen in the evening injection group as aresult of the basal insulin waning towardthe end of the 24-h dosing interval, whichis not masked by the use of rapid-actinginsulin during this early morning period.While glucose variability metrics weresmaller for Gla-300 than Gla-100, consis-tent with the 24-h glucose profiles, therewere no statistically significant differ-ences in these metrics.Evidenceof themoreprolongedglucose-

lowering activity of Gla-300 versus Gla-100is most apparent toward the end of thedaily injection interval, as shown by a sig-nificantly smaller increase in glucosewithinthe last 4 h of the injection interval for Gla-300 than for Gla-100. These observationsindicate that Gla-300 has amore sustainedglucose-lowering action and provide clini-cal evidence of this prolonged glycemiccontrol over an entire 24-h period.These results are consistent with previ-ous findings from euglycemic clampstudies (2,3).Rates of hypoglycemia detected using

SMPG values were consistent with theinterstitial glucose values detected byCGM. The annualized event rates forSMPG-confirmed or severe hypoglyce-mia were lower in the Gla-300 thanGla-100 group during the nocturnalperiod (0000–0559 h), which showed arisk reduction of 55% for confirmed(,54 mg/dL) or severe hypoglycemia.During the nocturnal period, mealtimeinsulin would be expected to have min-imal effect as a contributory factor,allowing a more reliable comparison ofthe effects of basal insulin. There wereno consistent within-treatment differ-ences in terms of hypoglycemia riskbetween themorning and evening injec-tion groups for either of the insulinpreparations. The phase IIIa EDITION 4study also investigated Gla-300 versusGla-100 in people with type 1 diabetesand observed an increase in SMPG withGla-300 before breakfast in the initialweeks of the study that might haveaffected the risk of hypoglycemia. Inthe current study, however, mean 24-hglucose profiles do not provide any

indication of glucose levels being higherwith Gla-300 than with Gla-100 at a timethat might be considered “before break-fast.” In addition, the hypoglycemiaresults presented here could be influ-enced by the fact that mean interstitialglucose levels were higher with Gla-300than Gla-100 at baseline, although thiswas likely offset by the greater decreasein glucose over the study period withGla-300 and the lower final mean glu-cose in that group.

The similar findings in the morningand evening injection groups for Gla-300 suggest the potential for flexibilityto select an injection time (morning orevening) without compromising clinicalbenefits, and should enable people withdiabetes to select their insulin injectionschedule according to their lifestyle toreduce the burden of therapy (e.g., amorningdosing timewouldnot beempha-sized to avoid the nocturnal hypoglycemiathat occurred in some patients given Gla-100 at bedtime) (13). The consistent 24-hinsulin activitywithGla-300 suggests therewould be value in a clinical trial exploringwhether people currently using Gla-100twice daily could be adequately controlledwith Gla-300 once daily.

The safety profile of both treatmentsin this study was consistent with thatreported previously (13–16), with simi-lar numbers of participants in eachgroup experiencing AEs. Both treat-ments were well tolerated.

The phase IIIa EDITION 4 study alsocompared morning versus evening injec-tion times (13). This 6-month treat-to-target study demonstrated comparableglucose control with Gla-300 and Gla-100, in terms of HbA1c and fastingplasma glucose, which did not differ be-tween the morning and evening injec-tion groups. Hypoglycemia rates weregenerally similar between treatmentgroups, except for lower nocturnal hypo-glycemia with Gla-300 than Gla-100 inthe first 8 weeks of the study. Neitherhypoglycemia nor the AE profile differedby time of injection.

This exploratory CGM study has limi-tations: the open-label study design,which was unavoidable because of thedifferent injection volumes of Gla-300and Gla-100, and the use of commercialsyringes that are not approved for usewith Gla-300 in clinical practice and aresuboptimal for delivery of an insulinwith a concentration of 300 units/mL.

Future studies could be enhanced by us-ing the recently approved Gla-300 peninjector; the use of syringes with Gla-300 might be expected to increaseglucose variability in the Gla-300 group,although an increase in variability wasnot seen in this study. This study alsoused a short duration of treatment (two8-week periods), although a 2-week pe-riod of CGM (weeks 7–8 in this study) isconsidered to be representative of thelonger-term pattern of glucose levels(17). A modest number of participantswere enrolled, with crossover of the in-jection schedule (morning to evening orvice versa), so group size was limited,with a low power to detect differencesbetween treatmentsdthat is, the per-centage of time within the glucose rangeof 80–140mg/dL (the primary end point),glucose variability metrics, and post hocanalyses. This exploratory study involvedparticipants with type 1 diabetes withrelatively good glycemic control at base-line (HbA1c 7.5% in the Gla-300 groupand 7.4% in the Gla-100 group), whichmight have limited the improvement ofthe quality of glycemic control. Never-theless, consistent differences werenoted between Gla-300 and Gla-100 interms of the average glucose profilespooled across individuals, CGM-detectedlow and high interstitial glucose values,SMPG-documented nocturnal hypo-glycemia (000020559 h), and CGM-determined glucose levels during thelast 4 h of each participant’s injectioninterval.

Recording interstitial glucose levelscontinuously throughout the day andnight is a valuable tool for comparingnew therapies and an important way toconfirm whether observed PK/PD differ-ences between insulins can be translatedto clinically relevant differences in 24-hglucose profiles. CGM is also particularlyuseful for evaluating hypoglycemia andglucose variability, measures that arevery difficult to capture accurately or tocompare between treatmentsusing inter-mittent SMPG data.

In conclusion, inpeoplewithmoderatelywell-controlled type 1 diabetes, CGM find-ings are consistentwith previously reportedPK/PD analyses, demonstrating that Gla-300 offers improved glycemic controlover a full 24-hperiod,with lessfluctuationin glucose levels, compared with Gla-100.These results imply that Gla-300 shouldimprove the flexibility of the injection

care.diabetesjournals.org Bergenstal and Associates 559

schedule (morning or evening) withoutcompromising glycemic control.

Acknowledgments. The authors thank thestudy participants, trial staff, and investigatorsfor their participation. The authors also thankCassandra Pessina (Sanofi) for her critical reviewof the manuscript and for assistance with man-aging manuscript development. Editorial sup-port was provided by Rianne Stacey and FionaWoodward of Fishawack Communications,Abingdon, U.K., and was funded by Sanofi.Dualityof Interest.R.M.B.hasreceivedresearchsupport from Abbott Diabetes Care Division,Becton Dickinson, Boehringer Ingelheim, Bristol-Myers Squibb/AstraZeneca, Dexcom, Eli Lilly,Hygieia, Johnson & Johnson, Medtronic, Merck,Novo Nordisk, Roche, Sanofi, and Takeda; hasparticipated in advisory boards for Abbott Di-abetes Care Division, Medtronic, Novo Nordisk,Roche, Sanofi, and Takeda; and has acted as aconsultant for Abbott Diabetes Care Division,Becton Dickinson, Boehringer Ingelheim, Bristol-Myers Squibb/AstraZeneca, Dexcom, Hygieia,Johnson & Johnson, Medtronic, Roche, andSanofi. R.M.B. provides his services throughhis employer, the nonprofit Park Nicollet Insti-tute, and receives no personal income from theseservices. R.M.B. has inherited Merck stock.R.M.B. has also volunteered for the AmericanDiabetes Association and JDRF. T.S.B. has re-ceived research support from Abbott, Ascensia,Becton Dickinson, Boehringer Ingelheim, Compan-ion Medical, Dexcom, Elcelyx, GlySens, Janssen,Lexicon, Eli Lilly, Medtronic, Novo Nordisk,Sanofi, Senseonics, Versartis, and YOFiMeter;has acted as a consultant for AstraZeneca,Bayer, Becton Dickinson, Calibra, Eli Lilly,Medtronic, Novo Nordisk, and Sanofi; andhas received speaking honoraria from Abbott,Insulet, Medtronic, Novo Nordisk, and Sanofi.D.R. has been a consultant for Abbott Diabe-tes Care Division, Halozyme Therapeutics,Novo Nordisk, One Drop, and Sanofi. M.Z.and I.M.-B. are employees of Sanofi and holdshares of Sanofi stock. H.G. is an employeeof BDM Consulting, Inc., contracted to Sanofi.A.J.A. has been a consultant or has been amember of the scientific advisory board forAstraZeneca, Dexcom, Eli Lilly, Medtronic,and Novo Nordisk and has received researchsupport from Dexcom, Medtronic, Novo Nor-disk, and Sanofi. Sanofi sponsored this studyand was responsible for designing and coordi-nating the trial. Sanofi monitored the clinicalsites, collected and managed the data, and

performed all statistical analyses. No otherpotential conflicts of interest relevant to thisarticle were reported.Author Contributions. R.M.B. was the princi-pal investigator of this study. R.M.B., M.Z., andI.M.-B. designed the protocol, analyzed andinterpreted data, and wrote the manuscript.T.S.B. and A.J.A. were investigators, saw partic-ipants at designated visits, monitored partici-pants during the trial, and reviewed and revisedthe manuscript. D.R. guided several aspects ofthe statistical analysis of glucose metrics andreviewed and revised the manuscript. H.G. per-formed the statistical analysis of the data andreviewed and revised the manuscript. All authorshad access to relevant study data and inter-preted data, reviewed and commented on sev-eral drafts of the manuscript, and had the finalresponsibility to submit the article for publica-tion. R.M.B. is the guarantor of this work and, assuch, had full access to all the data in the studyand takes responsibility for the integrity of thedata and the accuracy of the data analysis.

References1. Cryer PE. Hypoglycaemia: the limiting factorin the glycaemic management of type I andtype II diabetes. Diabetologia 2002;45:937–9482. BeckerRH,DahmenR,BergmannK, LehmannA,Jax T, Heise T. New insulin glargine 300 units zmL21

provides a more even activity profile and pro-longed glycemic control at steady state comparedwith insulin glargine 100 units z mL21. DiabetesCare 2015;38:637–6433. Becker RH, Nowotny I, Teichert L, Bergmann K,Kapitza C. Low within- and between-day variabil-ity in exposure to new insulin glargine 300 U/ml.Diabetes Obes Metab 2015;17:261–2674. Workgroup on Hypoglycemia, American Dia-betes Association. Defining and reporting hypo-glycemia in diabetes: a report from the AmericanDiabetes Association Workgroup on Hypoglyce-mia. Diabetes Care 2005;28:1245–12495. Rodbard D. New and improved methods tocharacterize glycemic variability using continu-ous glucose monitoring. Diabetes Technol Ther2009;11:551–5656. The Diabetes Control and Complications TrialResearchGroup. The effect of intensive treatmentof diabetes on the development and progressionof long-term complications in insulin-dependentdiabetes mellitus. N Engl J Med 1993;329:977–9867. American Diabetes Association. Glycemictargets. Sec. 6. In Standards of Medical Carein Diabetesd2016. Diabetes Care 2015;38(Suppl. 1):S33–S40

8. Gubitosi-Klug RA; DCCT/EDIC ResearchGroup. The diabetes control and complicationstrial/epidemiology of diabetes interventionsand complications study at 30 years: summaryand future directions. Diabetes Care 2014;37:44–499. Garber AJ. Will the next generation of basalinsulins offer clinical advantages? DiabetesObes Metab 2014;16:483–49110. Bohannon NJ. Optimizing insulin regimensin type 1 diabetes. How to help patients getcontrol of their life. Postgrad Med 2003;113:39–42, 45–48, 54.11. Bolli GB, Andreoli AM, Lucidi P. Optimizingthe replacement of basal insulin in type 1 dia-betes mellitus: no longer an elusive goal in thepost-NPH era. Diabetes Technol Ther 2011;13(Suppl. 1):S43–S5212. Miller KM, Foster NC, Beck RW, et al.; T1DExchange Clinic Network. Current state of type 1diabetes treatment in the U.S.: updated datafrom the T1D Exchange clinic registry. DiabetesCare 2015;38:971–97813. Home PD, Bergenstal RM, Bolli GB, et al.New insulin glargine 300 units/mL versus glar-gine 100 units/mL in people with type 1 diabe-tes: a randomized, phase 3a, open-label clinicaltrial (EDITION 4). Diabetes Care 2015;38:2217–222514. Bolli GB, Riddle MC, Bergenstal RM, et al.;on behalf of the EDITION 3 study investiga-tors. New insulin glargine 300 U/ml comparedwith glargine 100 U/ml in insulin-naıve peoplewith type 2 diabetes on oral glucose-loweringdrugs: a randomized controlled trial (EDITION3). Diabetes Obes Metab 2015;17:386–39415. Riddle MC, Bolli GB, Ziemen M, Muehlen-Bartmer I, Bizet F, Home PD; EDITION 1 StudyInvestigators. New insulin glargine 300 units/mL versus glargine 100 units/mL in peoplewith type 2 diabetes using basal and mealtimeinsulin: glucose control and hypoglycemia in a6-month randomized controlled trial (EDITION1). Diabetes Care 2014;37:2755–276216. Yki-Jarvinen H, Bergenstal R, Ziemen M,et al.; EDITION 2 Study Investigators. New in-sulin glargine 300 units/mL versus glargine100 units/mL in people with type 2 diabetesusing oral agents and basal insulin: glucose con-trol and hypoglycemia in a 6-month randomizedcontrolled trial (EDITION 2). Diabetes Care 2014;37:3235–324317. Bergenstal RM, Ahmann AJ, Bailey T, et al.Recommendations for standardizing glucose re-porting and analysis to optimize clinical decisionmaking in diabetes: the Ambulatory GlucoseProfile (AGP). Diabetes Technol Ther 2013;15:198–211

560 CGM Profiles of Gla-300 vs. Gla-100 Diabetes Care Volume 40, April 2017