Pharmacokinetics and Pharmacodynamics of Insulin Glargine Given in the Evening As Compared with in the Morning in Type 2 Diabete

Diabetes Care Volume 38, March 2015 503

Francesca Porcellati, Paola Lucidi, Pharmacokinetics and Patrizia Cioli, Paola Candeloro, Anna Marinelli Andreoli, Pharmacodynamics of Insulin Stefania Marzotti, Maura Ambrogi, Glargine Given in the Evening as Geremia B. Bolli, and Carmine G. Fanelli Compared With in the Morning in Type 2 Diabetes Diabetes Care 2015;38:503–512 | DOI: 10.2337/dc14-0649 MRIGTCNLGE N THERAPEUTICS AND TECHNOLOGIES EMERGING

OBJECTIVE To compare pharmacokinetics (PK) and pharmacodynamics (PD) of insulin glargine in type 2 diabetes mellitus (T2DM) after evening versus morning administration.

RESEARCH DESIGN AND METHODS Ten T2DM insulin-treated persons were studied during 24-h euglycemic glucose clamp, after glargine injection (0.4 units/kg s.c.), either in the evening (2200 h) or the morning (1000 h).

RESULTS

The 24-h glucose infusion rate area under the curve (AUC0–24h) was similar in the evening and morning studies (1,058 6 571 and 995 6 691 mg/kg 3 24 h, P = 0.503), but the first 12 h (AUC0–12h) was lower with evening versus morning glargine (357 6 244 vs. 593 6 374 mg/kg 3 12 h, P = 0.004), whereas the opposite occurred for the second 12 h (AUC12–24h 700 6 396 vs. 403 6 343 mg/kg 3 24 h, P =0.002). The glucose infusion rate differences were totally accounted for by different rates of endogenous glucose production, not utilization. Plasma insulin and C-peptide Department of Medicine, University of Perugia levels did not differ in evening versus morning studies. Plasma glucagon levels Medical School, Perugia, Italy Corresponding author: Geremia B. Bolli, geremia. (AUC – 1,533 6 656 vs. 1,120 6 344 ng/L/h, P = 0.027) and lipolysis (free fatty 0 24h [email protected]. acid AUC0–24h 7.5 6 1.6 vs. 8.9 6 1.9 mmol/L/h, P = 0.005; b-OH-butyrate AUC0–24h 6 6 P P < Received 12 March 2014 and accepted 10 6.8 4.7 vs. 17.0 11.9 mmol/L/h, = 0.005; glycerol, 0.020) were overall November 2014. more suppressed after evening versus morning glargine administration. Clinical trial reg. no. EudraCT2010-019368-35, http://eudract.ema.europa.eu/. CONCLUSIONS This article contains Supplementary Data online The PD of insulin glargine differs depending on time of administration. With at http://care.diabetesjournals.org/lookup/ morning administration insulin activity is greater in the first 0–12 h, while with suppl/doi:10.2337/dc14-0649/-/DC1. evening administration the activity is greater in the 12–24 h period following This is an independent, investigator-designed dosing. However, glargine PK and plasma C-peptide levels were similar, as well project, neither shared with nor supported by as glargine PD when analyzed by 24-h clock time independent of the time of any pharmaceutical company. administration. Thus, the results reflect the impact of circadian changes in insulin © 2015 by the American Diabetes Association. Readers may use this article as long as the work sensitivity in T2DM (lower in the night-early morning vs. afternoon hours) rather is properly cited, the use is educational and not than glargine per se. for profit, and the work is not altered. 504 Glargine Evening vs. Morning in T2DM Diabetes Care Volume 38, March 2015

The insulin dynamic of the postabsorptive was obtained, 10 T2DM persons receiv- dinnertime to either morning (1000 h, and interprandial state of nondiabetic ing insulin therapy (all glargine as basal N = 5) or evening (2200 h, N = 5) glargine subjects (basal insulin) is best repro- with/without mealtime rapid-acting in- administration 9 days (day 29) prior to duced in people with diabetes requiring sulin and oral hypoglycemic agents) the study day (day 0). For morning dos- insulin administration by the long-acting were recruited (Table 1) and studied ac- ing, on day 29, subjects injected 50% of insulin analogs compared with NPH insu- cording to the guidelines of the Helsinki the usual glargine dose at dinnertime, lin (1). This more physiological substitu- Declaration and good clinical practice followed by the remaining 50% the tion translates into clinical benefits requirements. Subjects were free of next morning at 1000 h (day 28). On compared with NPH, primarily a reduced any detectable microangiopathic and day 27 and thereafter until the study risk of nocturnal hypoglycemia for similar macroangiopathic complications and of day, the full glargine dose was injected A1C levels (1–7). any major illness other than diabetes, at 1000 h every day. For evening dosing, NPH insulin is most often injected at as indicated by medical history, physical subjects injected glargine at 2200 h ev- bedtime to limit the risk of nocturnal examination, electrocardiogram, or rou- ery day from day 29 until the study day. hypoglycemia due to its peak of action tine laboratory tests. All of the subjects With either morning or evening dosing, 4–6 h postinjection (1–3). In contrast, the were receiving insulin glargine with the basal insulin concentration was titrated smoother activity of the long-acting insulin evening meal. as in the run-in period, and treatment analog glargine allows more flexibility in The study has a randomized, single- with mealtime rapid-acting insulin was dosing, and its administration is less strictly dose, open-label, crossover design (Sup- continued, if patients had been receiv- bound to the time of injection. Hence, in- plementary Fig. 1). After a 2-week run-in ing such treatment. sulin glargine is approved for administra- period, during which subjects with dia- On both evening and morning study tion at any time of day, provided it is at the betes continued their treatment, and occasions, subjects had the last subcu- same time each day (8). the dose of basal insulin glargine was taneous glargine injection 24 h before However, only a few studies have com- optimized to achieve a fasting plasma the time they received glargine on the pared the effects of glargine given in the glucose (PG) concentration of #100 study day. An equivalent interval of fast- evening compared with other times of day mg/dL while avoiding nocturnal hypo- ing was observed before initiation of both in type 1 diabetes mellitus (9–11) as glycemia (PG ,72 mg/dL), subjects each clamp study (14 h), with the last well as in type 2 diabetes mellitus (T2DM) were randomized to two 24-h euglyce- meal (688 kcal, 54% carbohydrate, 30% (12,13), with results basically consistent mic clamp studies, after either evening protein, and 16% lipids) consumed ei- in the noninferiority of morning versus (2200 h) or morning (1000 h) subcuta- ther at 0800 h the same day of the study evening dosing. Yet, the large majority of neous injection of insulin glargine (0.4 (glargineevening)orat2000htheday people with diabetes inject glargine in the units/kg). Each occasion was separated before the study (glargine morning). evening, often at bedtime, in keeping with by an interval of 3–4 weeks, during Care was taken by subjects to avoid hy- the tradition of NPH administration. which the usual insulin regimen of the poglycemia (PG ,72 mg/dL) with blood One reason why the potentially greater run-in period was continued (glargine at glucose self-monitoring in the 48 h be- flexibility of glargine in the treatment dinnertime). fore the clamp studies. For the entire of T2DM has not translated into a rec- After randomization, subjects were duration of the studies, all persons mon- ognized option of administration at a switched in a timely manner from itored blood glucose levels by means different time of day compared with NPH, might be the lack of head-to- Table 1—Demographic and baseline clinical characteristics of the subjects head studies of pharmacokinetics (PK) Gla at 1000–2200 ha Gla at 2200–1000 ha and pharmacodynamics (PD), compar- b,c ing glargine given in the evening versus N =10 N =5 N =5 P value the morning. Yet, an indirect compari- Sex son of PK and PD between studies, Male 7 3 4 where glargine was given in the evening Female 3 2 1 (14) compared with the morning (15– Age (years) 61.6 6 4.5 62.2 6 3.4 60.8 6 4.1 0.573 2 17), suggests important differences in BMI (kg/m )28.06 3.6 28.5 6 4.3 27.5 6 4.0 0.713 action profiles, although in such a com- Known diabetes 6 6 6 parison the contribution of circadian duration (years) 16.5 8.0 16.4 3.2 16.6 11.5 0.971 changes in insulin sensitivity should be A1C 0.801 6 6 6 taken into account (18,19). % 7.2 0.6 7.1 0.5 7.2 0.7 mmol/mol 55 6 8.1 54 6 7.9 55 6 9.0 The aim of the current study was to Daily insulin dose perform a head-to-head comparison of (units/kg)c the PK and PD of insulin glargine in per- Total 0.53 6 0.31 0.46 6 0.29 0.60 6 0.34 0.503 sons with T2DM, after either evening or Basal 0.28 6 0.11 0.27 6 0.12 0.30 6 0.12 0.703 morning administration. Data are mean 6 SD, unless otherwise indicated. aSubjects split by sequence of treatments. bAll 10 subjects were receiving basal insulin glargine, 7 on basal bolus (glargine plus prandial rapid-acting RESEARCH DESIGN AND METHODS analog); 8 subjects were also receiving oral agents (metformin, 5 subjects on basal bolus and 2 subjects After approval by the local ethics commit- receiving a basal insulin regimen; 1 subject was receiving a basal insulin regimen plus 1 mg glimepiride). cPrandial (if any, rapid-acting insulin analog) plus basal (glargine insulin) regimen. tee and after written informed consent care.diabetesjournals.org Porcellati and Associates 505

of a reflectometer (OneTouch Vita; Life- free fatty acid (FFA), glycerol, b-OH- (AUC12–24h). Secondary end points Scan Italy [part of Johnson & Johnson butyrate, C-peptide, and glucagon were PG concentration; glucose fluxes Medical S.p.A.], Milan, Italy). concentrations were measured by (AUC7–24h); plasma C-peptide, glucagon, previously described methods (14). and nonglucose substrate concentra- Euglycemic Clamp Procedure Glucose enrichment was determined tions; intravenous insulin infusion rates Persons with diabetes were admitted in on its penta-acetate (penta-O-acetyl- prior to and immediately after subcu- the fasting state to the Clinical Research b-D-glucopyranose) derivative by gas taneous glargine insulin injection; Unit of the Department of Medicine, chromatography–mass spectrometry and plasma insulin concentrations. A Section of Internal Medicine, Endocri- (HP 6890 II GC, HP 5973A GC/MS; sample size of 10 subjects was chosen nology and Metabolism of the Univer- Hewlett-Packard, Palo Alto, CA) in in order to achieve 80% power to sity of Perugia Medical School, 4 h electron impact ionization mode mon- detectameanofpaireddifferences prior to study initiation (i.e., at either itoring the ions 200 and 202 for the of 40% with an estimated SD of differ- 0600 or 1800 h, respectively, for the unlabeled and [6,6-2H ]-glucose, respec- ences of 40% between the GIR for glar- morning and evening studies), and re- 2 tively (24). gine infused during both the first 12 h mained fasting for the subsequent 28 h (AUC – for glargine at 1000 h vs. glar- (Supplementary Fig. 1). They were put at 0 12h Calculations and Statistical Analysis gine at 2200 h) and the second 12 h rest in a bed and studied with the eugly- Calculations of glucose fluxes are based (AUC – for glargine at 1000 h vs. glar- cemic clamp technique (20). In brief, 12 24h on a non–steady-state assumption and gine at 2200 h). All tests of statistical two venous lines were inserted, an ante- the total R and R values were calcu- hypothesis were carried out at the 5% cubital vein on one arm (for infusions) a d lated using a modified form of the Steele level of significance and comparisons and a dorsal vein (cannulated retro- equation to account for the addition of were two sided. Data in the text are ex- gradely with a 20-gauge butterflynee- stable labeled tracer to the exogenous pressed as the mean 6 SD and median dle) on the contralateral hand (kept at glucose infusate (24). Endogenous glu- with 25th and 75th percentiles as appro- 658C in a hot box) for intermittent sam- cose production (EGP) was obtained as priate, and in figures as the mean 6 SE. pling of arterialized venous blood. An the difference between R and the ex- Statistical analysis was usually per- intravenous feedback insulin infusion a ogenous glucose infusion rate (GIR) dur- formed using PASS (NCSS Statistical was initiated to achieve and maintain ing the clamp. Because the isotopic Software, Kaysville, UT). PG at 100 6 5 mg/dL until time 0 (i.e., steady state might have not been 1000 or 2200 h), as previously described achieved during the first hours of the RESULTS (21). At 2120 min, a primed sterile, study (25), in order to have reliable val- All 10 subjects enrolled completed the pyrogen-free constant infusion (0.222 ues of glucose fluxes, calculations were two studies. mmol/kg/min)of[6,6-2H ]-glucose 2 performed by excluding the first 6 h of (Cambridge Isotope Laboratories, Cam- Glycemic Control and Insulin Doses in the study. bridge, MA) was started and maintained the Week Prior to Studies The linear trapezoidal rule was used throughout the experiment to deter- After randomization, home daily glycemic to calculate the concentration-time mine glucose kinetics. A variable rate control (mean PG from self-monitoring area under the curve (AUC) between of a 20% dextrose solution enriched to data) during the 7 days before the 0and24h(AUC– )forGIRs,plasma 4% with [6,6-2H ]-glucose was used to 0 24h clamp studies was not different in the 2 free insulin, and nonglucose substrates. clamp the PG concentration at the de- morning study (139 6 11 mg/dL) versus The maximum plasma concentration sired target of 100 mg/dL for 24 h (14) the evening study (142 6 7 mg/dL, P = (C )andthetimetoreachC and to avoid non–steady-state errors in max max 0.221). The dose of glargine over the 7 (T )forthesamevariableswereread the measurement of glucose turnover, max days prior to the studies was slightly, but directly from the plasma concentration- as described previously (22). At 1000 or significantly, higher in the morning ver- time data for each subject. The determi- 2200 h, a subcutaneous injection of 0.4 sus the evening study (0.31 6 0.03 vs. nations of C and T for the GIR units/kg insulin glargine (Lantus 100 max max 0.29 6 0.03 units/kg, P = 0.002). Also, were derived from a smoothed 3-point units/mL) was administered by means total daily insulin doses were greater in running average GIR curve for each of injection with a 0.5-mL insulin syringe the morning (0.56 6 0.3 units/kg) com- subject in order to provide reliable into the abdominal area. On both occa- pared with the evening study (0.55 6 data for calculation. sions, the clamp procedure was termi- 0.3 units/kg), although the difference The primary analysis of the PK and nated at 24 h after the subcutaneous did not reach statistical significance PD parameters was performed using injection of glargine. (P = 0.070). ANOVA, which allowed for variation Analytical Methods of subjects nested within sequence, pe- Rates of Intravenous Insulin Infusion PG concentration was measured at riod, and treatment. Tmax variables were Prior to Subcutaneous Insulin the bedside using the YSI 2300 glucose analyzed nonparametrically (Wilcoxon Injection, Plasma Insulin and Glucose analyzer (YSI 2300 STAT; Yellow Springs rank sum test and Hodges-Lehmann Concentrations, GIRs, Rates of Instruments, Yellow Springs, OH). The estimates of the treatment with 95% Endogenous Glucose Production, and plasma free insulin concentration was confidence limits). Peripheral Glucose Utilization measured by radioimmunoassay after The primary end point of the study was The amount of insulin infused to achieve polyethylene glycol extraction of anti- the GIR over the first 12 h (AUC0–12h) and maintain euglycemia during the bodies from the plasma (23). Plasma and the GIR over the second 12 h feedback period (time 22to2hbefore 506 Glargine Evening vs. Morning in T2DM Diabetes Care Volume 38, March 2015

Figure 1—Rates of insulin and glucose infusion, production and utilization of glucose, and PG and insulin concentrations after subcutaneous injection of insulin glargine (Gla) at either 2200 or 1000 h in 10 subjects with T2DM. Because of an insufﬁcient isotopic steady state, the fragment of the curves corresponding to the initial 6 h of the studies has been deleted for both production and utilization of glucose.

and after time 0 h) was greater with signiﬁcance (P = 0.190). After the of intravenous insulin infusion tended glargine given in the morning (0.3 6 subcutaneous injection of glargine to be faster with evening (time 21 6 0.2 mU/kg/min) compared with the at time 0 h, the rate of intravenous 1 h) compared with morning glargine evening (0.2 6 0.2 mU/kg/min), but insulin infusion decreased in both dosing (time 0.3 6 1.3 h) (P , 0.065) the difference did not reach statistical treatments. However, the withdrawal (Fig. 1 and Table 2). care.diabetesjournals.org Porcellati and Associates 507

Table 2—Summary of the main study results state, was not different with evening Morning Evening compared with morning dosing. How- injection injection P value ever, the EGP exhibited opposite pro- ﬁles between day and night, indicating PG (mg/dL) 0–24 h 101 6 3.0 101 6 2.0 0.551 greater insulin sensitivity during the day 0–12 h 100 6 1.0 102 6 3.0 0.104 compared with the night. In fact, EGP 12–24 h 103 6 4.0 100 6 1.0 0.028 decreased from a peak value of 1.14 6 FIRI (mmol/L/h) 0.57 mg/kg/min at 0800 h to the lowest

AUC0–24h 1,636 6 440 1,605 6 504 0.836 value of 0.14 6 0.18 mg/kg/min at 1800 h 6 6 AUC0–12h 898 241 828 272 0.367 (evening insulin injection), whereas it in- 6 6 AUC12–24h 739 204 777 245 0.625 creased steadily from 2200 h (0.20 6 PD 0.21 mg/kg/min) over the night hours 3 6 6 GIR-AUC0–24h (mg/kg 24 h) 995 691 1,058 571 0.503 to achieve a peak value of 0.96 6 0.30 GIR-AUC – (mg/kg 3 12 h) 593 6 374 357 6 244 0.004 0 12h mg/kg/min between 0800 and 1000 h GIR-AUC12–24h (mg/kg 3 12 h) 403 6 343 700 6 396 0.002 Maximum GIR (mg/kg/min) 1.4 6 0.8 1.5 6 0.6 0.533 (morning insulin injection). a GIR Tmax (h) 9 (8.1, 10.7) 18 (16.5, 20.4) 0.002 Glucose utilization (GU) was not stim- Glucose ﬂuxes ulated and overall was not different

EGP-AUC7–24h (mg/kg 3 17 h) 595 6 348 495 6 267 0.351 among insulin treatments (Fig. 1 and GU-AUC7–24h (mg/kg 3 17 h) 1,260 6 349 1,395 6 368 0.141 Table 2). Glucagon (ng/L/h) 6 6 AUC0–24h 1,533 656 1,120 344 0.027 Plasma C-Peptide, Glucagon, FFA, and 6 6 AUC0–12h 739 322 576 160 0.067 b-OH-Butyrate Concentrations 6 6 AUC12–24h 794 342 544 200 0.012 Overall, the plasma C-peptide concen- FFA (mmol/L/h) tration was similarly suppressed with AUC – 8.9 6 1.9 7.5 6 1.6 0.005 0 24h both treatments (P = 0.104) (Fig. 2 and AUC0–12h 4.0 6 0.7 3.6 6 0.7 0.042 Table 2). AUC12–24h 4.8 6 1.2 3.8 6 1.0 0.006 b-OH-butyrate (mmol/L/h) Plasma glucagon concentrations were

AUC0–24h 17.0 6 11.9 6.8 6 4.7 0.005 more suppressed with evening compared AUC0–12h 7.1 6 6.4 2.7 6 1.6 0.035 with morning dosing. However, over the 6 6 AUC12–24h 9.9 6.4 4.1 3.2 0.001 initial 12-h period (AUC0–12h), the degree Data are mean 6 SD, unless otherwise indicated. FIRI, free immunoreactive insulin. aMedian of suppression was equivalent, whereas with 25 and 75 percentiles. in the second half of observation period it was less suppressed with morning glargine dosing (Fig. 2). Plasma FFA concentrations were At time 0 min, the plasma insulin with evening glargine dosing. The PG more suppressed with evening versus concentrationtendedtobehigherin concentration at the end of the clamp morning glargine dosing over the 24-h the morning versus the evening study was higher with morning glargine com- study period. During the last 12 h with (P = 0.082). The overall plasma free in- pared with evening injection (108 6 8vs. both treatments, FFA plasma concentra- sulin concentration after subcutaneous 100 6 2 mg/dL, P = 0.033) (Fig. 1). tions tended to increase compared with glargine injection did not differ be- The GIR needed to maintain euglyce- the initial 12 h of the clamp; however, tween the two treatments (Fig. 1 and mia was significantly different between this increase was greater with morning Table 2). the two studies at time 0 h: positive glargine dosing (Fig. 2). PG concentration decreased similarly (0.45 6 0.55 mg/kg/min) after the Similarly, plasma b-OH-butyrate con- during intravenous insulin infusion prior evening compared with morning dos- centrations were lower with evening glar- to glargine injection on the two occa- ing, where it was basically turned off gine dosing over the whole study period sions (data not shown), and at time (P = 0.038). The mean GIR for the 24-h (0–24 h) (P = 0.005). Indeed, the plasma 0 h it was not different in the two stud- study period (AUC0–24h) was equivalent b-OH-butyrate concentration increased ies (102 6 4.3 and 100 6 2.2 mg/dL, in the two studies. However, over the to 1.43 6 0.99 mmol/L by the end of respectively, for morning vs. evening initial 12-h period, the GIR was greater study with the morning dosing, but only study, P = 0.061). After glargine injec- in the morning compared with evening to 0.42 6 0.32 mmol/L with evening glar- tion, the mean PG over the 24-h study study, whereas the opposite was ob- gine dosing (P = 0.003) (Fig. 2). period was superimposable in the served for the second 12-h period. Al- After initial suppression, plasma glyc- morning and evening studies. During though the maximum GIR values were erol concentrations increased in both the first 12 h of the study, the two treat- similar with both treatments, the GIR treatments, and remained significantly ment groups had similar mean PG con- Tmax occurred earlier after morning in- lower after evening compared with morn- centrations, in the second half of the jection than after evening dosing (Fig. 1 ingdosing(P = 0.020) (data not shown). study period (time 12–24 h), the mean and Table 2). Lactate and alanine concentrations PG concentration was slightly, but signifi- EGP AUC7–24h, calculated during the were not different between treatments cantly, higher with morning compared last 17 h of the study at isotopic steady (data not shown). 508 Glargine Evening vs. Morning in T2DM Diabetes Care Volume 38, March 2015

Figure 2—Plasma C-peptide, glucagon, FFAs, and b-OH-butyrate concentrations after subcutaneous injection of 0.4 units/kg insulin glargine (Gla) at either 2200 or 1000 h in 10 subjects with T2DM.

Insulin Activity Analyzed by 24-h Clock for time and treatment 3 time interac- glargine was injected in the evening or in Time, Independently From the Time of tion, from repeated-measures ANOVA) the morning. However, relevant differences Glargine Injection (Fig. 3). emerged when considering time action When insulin activity (as indicated by profiles and distribution of metabolic ef- GIR, EGP, and GU profiles) was analyzed fect in the first12-hperiodcompared by 24-h clock time, comparing daytime CONCLUSIONS with the second 12-h period. Glargine (1000–2200 h) versus nighttime (2200– This is the first study directly comparing administration in the morning exerted 1000 h) periods from the last 12 h of in T2DM subjects the PK and PD of basal greater activity in the first 12 h, in con- each study, the GIR was greater during insulin glargine administered at differ- trast to evening dosing, which exhibited the day versus the night independent of ent times of the day (i.e., in the evening higher potency in the second 12 h. In the time of glargine injection (AUC12–24h and in the morning). The question is rel- addition, the latter was more effective 701 6 396 vs. 412 6 331 mg/kg 3 12 h evant not only in light of the potential in suppressing overall lipolysis and day vs. night, respectively, P = 0.002) clinical translation of the findings, but plasma glucagon concentrations, pri- (Fig. 3). EGP changes were consistent also considering the great emphasis re- marily over the last 12 h of the studies. with differences in GIR requirements cently given to the concept of flexibility In the current study, morning glargine (AUC12–24h 290 6 204 vs. 412 6 202 of dosing time, which has been advo- administration induced a robust increase mg/kg 3 12 h day vs. night, respectively, cated mainly for the second-generation in GIR during the first 12 h, whereas in the P = 0.041) (Fig. 3). GU did not change and long-acting insulin analogs (26). second 12 h of the study a progressive was not different between treatments The results indicate that overall 24-h decline in GIR occurred, as the result (P = 0.079 and P = 0.406, respectively, insulin activity was not different whether of a rebound in EGP, and PG tended to care.diabetesjournals.org Porcellati and Associates 509

Figure 3—Rates of glucose infusion, production, and utilization presented by time of day (24-h clock time) as result of the last 12 h of the evening study (left panel, 1000–2200 h) and the last 12 h of the morning study (right panel, 2200–1000 h). Gla, glargine. increase (Fig. 1). This action profile is re- by the lower rate of intravenous insulin and adipose tissue level (suppression of markably in line with previous reports infusion and greater GIR required to glucose production and lipolysis) with- (15–17) in which glargine PD was studied reach euglycemia prior to clamp initiation out promoting muscle glucose uptake after morning injection. In contrast, with before insulin injection at 2200 h, com- (17,27). evening glargine the distribution of the pared with the rates infused before the Morning and evening glargine admin- metabolic effect was quite different. In morning clamp. istration had differential effects on 24-h the first 12 h, GIR was nearly flat, whereas None of the two glargine dosing regi- lipolysis and plasma glucagon concen- in the second 12 h, GIR increased notably mens increased glucose disposal. This is trations. The evening glargine adminis- as the result of the suppression of EGP. in line with the physiological concept tration, more effectively than morning, The superior insulin activity at the end of that when basal insulin is replaced at restrained lipolysis and ketogenesis, the 24-h period after evening glargine ad- therapeutic doses to target euglycemia, and suppressed plasma glucagon ministration is also indirectly suggested it exerts its effects primarily at the liver concentrations. 510 Glargine Evening vs. Morning in T2DM Diabetes Care Volume 38, March 2015

It is tempting to speculate about the The existence of diurnal rhythmicity lesser effects until 1000 h and greater mechanisms by which glargine when in insulin sensitivity driving changes in effects thereafter. giveninthemorninghasdifferential PG concentration in normal subjects Since the PD results of the present metabolic effects compared with even- and in T2DM patients has been reported studies observed with glargine might ing administration. One explanation by older (32,33) and more recent studies be interpreted not on the basis of PK, might be different PK in the first 12-h (18,19,34,35). Boden et al. (19) de- but rather on the circadian rhythm of period compared with the second 12-h scribed a robust and well-defined insulin sensitivity, it is likely that simi- period (Fig. 1). However, plasma insulin rhythm in insulin sensitivity in T2DM pa- lar effects may occur also with other concentrations were relatively stable tients, with lower sensitivity in the morn- long-acting insulin formulations. In a throughout the 24-h study, without ing and higher sensitivity in the evening. recent study (14), NPH and detemir ex- any difference between first and second Similar to the current study, the circa- hibited greater activity in the after- 12-h periods, after both morning and dian changes in GIR were completely ac- noon compared with morning hours evening dosing. This applies to the sup- counted for by changes in EGP, whereas after evening injection in T2DM pa- pression of plasma C-peptide as well the Rd did not change (19). The elegant tients. However, a direct comparison (Fig. 2). work of Radziuk and Pye (28) provided a with morning and evening dosing of The alternative hypothesis might be deeper and fascinating insight with the other long-acting insulins (degludec, that the results are primarily related to hypothesis that the observed diurnal already on market outside the US, and differential responses to insulin at dif- temporal pattern in EGP might result glargine 300 units as well as pegylated ferent times of day (i.e., insulin sensi- from factors counteracting insulin action lispro, which has not yet been approved) tivity might be greater during daytime and triggered by the brain through cen- is needed. hours compared with nighttime hours). tral regulatory pathways. Nowadays, the A limitation of our study is the rela- In fact, when the data are analyzed by concept of diurnal rhythmicity in insulin tively small number of subjects studied. 24-h clock time (Fig. 3), it appears that sensitivity finds novel relevance in light However, the homogeneity of the data EGP, the driver of basal insulin effects of the more extensive concept of a “mo- observed and the crossover design rein- on glucose metabolism (i.e., GIR), is less lecular clock,” whichhasbeenshownto force the findings. suppressed during the night (2200– regulate the expression and function A strength of our study is its design, 1000 h) than during the day (1000– of a variety of enzymes, transport sys- which mimicked real-life conditions as 2200 h), and to a similar extent in morn- tems, and nuclear receptors involved closely as possible: an insulin dose close ing and evening dosing studies. At the in lipid and carbohydrate metabolism to that used by the insulin-treated sub- end of the night, at 1000 h, EGP surpris- (36–38). jects was administered, and the study ingly reached similar values after both Regardless of the mechanisms, the was performed after multiple dosing of morning and evening dosing. However, differential insulin activity observed in glargine (steady state). the incremental increase in EGP was the current study after morning glar- Finally, the results of the current greater with morning glargine dos- gine administration compared with study are relevant to the understanding ing compared with evening glargine evening glargine administration is rele- of PD in clamp studies after subcutane- dosing, possibly due to more acceler- vant to clinical practice. Indeed, insulin ous injection of long-acting insulin ana- ated lipolysis and greater plasma gluca- glargine provides a rather flat activity logs in T2DM (39,40). Distribution of the gon concentrations. Conversely, during profile at night whether given in the metabolic effect may be quite different the day (1000–2200 h), insulin activity morning or in the evening. This explains if the insulin to be studied is injected increases progressively as a result the reported lower risk of nocturnal hy- in the morning compared with the of the suppression of EGP. The latter, poglycemia with glargine compared evening. Therefore, it would not be cor- again surprisingly, reaches similar with “peak insulin” such as NPH (4–7). rect to assume that the greater meta- values at 2200 h after either morning On the other hand, the present studies bolic activity of glargine, observed 6–8h or evening glargine administration. indicate similarly greater insulin activity after morning dosing (15–17), is repre- Taken together, these differences during the daytime after glargine ad- sentative of the true action profile since speak in favor of the circadian rhythm ministration in either the morning or in the current study this appears quite of insulin sensitivity, with greater the evening. This may predict potentia- different when glargine is administered sensitivity during the day compared tion of the risk of hypoglycemia in the in the evening. with the night (19). Thus, the effects afternoon, for example, when prandial In conclusion, total insulin activity on of glargine administered in the morn- glucose-lowering drugs are combined glucose metabolism is similar with evening or evening only appear to differ. with basal insulin. Whether dosing ing or morning glargine administration. They are remarkably reproducible at a more usual breakfast time (0700– However, with evening glargine adminis- when analyzed by absolute time of 0800 h) would result in circadian PD tration, the suppression of nocturnal EGP, day, not the relative number of hours effects similar to those observed in lipolysis, and glucagon concentration are postinjection. the current study cannot be directly more consistent. As result, targeting fast- The results of the current study derived from our results. However, ing euglycemia appears more convenient closely reproduce those observed by the time curve of GIR indicates lower with evening glargine dosing compared Radziuk and Pye (28), and represent ad- insulin sensitivity in the interval 0200– with morning glargine dosing. Con- ditional evidence of the dawn phenom- 1000 h, suggesting that insulin dosing versely, morning dosing may be prefera- enon in T2DM (29–31). within this time frame would produce ble whenever greater protection against care.diabetesjournals.org Porcellati and Associates 511

the risk for nocturnal hypoglycemia is basal insulin glargine as compared with NPH in analogues (insulin detemir and NN344): compa- needed. patients with Type 1 diabetes mellitus given rable time-action profiles but less variability meal-time lispro insulin. Diabet Med 2004;21: than insulin glargine in type 2 diabetes. Diabe- 1213–1220 tes Obes Metab 2007;9:290–299 Acknowledgments. The authors thank Dr. 5. Riddle MC, Rosenstock J, Gerich J; Insulin 16. Luzio S, Dunseath G, Peter R, Pauvaday V, Gianluca Curti, Department of Medicine, Uni- Glargine 4002 Study Investigators. The treat- Owens DR. Comparison of the pharmacokinetics versity of Perugia Medical School, Perugia, Italy, to-target trial: randomized addition of glargine and pharmacodynamics of biphasic insulin as- for his skillful technical assistance in performing or human NPH insulin to oral therapy of type 2 part and insulin glargine in people with type 2 the gas chromatography–mass spectrometry diabetic patients. Diabetes Care 2003;26:3080– diabetes. Diabetologia 2006;49:1163–1168 assay for glucose kinetics. This article is dedi- 3086 17. Wang Z, Hedrington MS, Gogitidze Joy N, cated to the subjects with type 2 diabetes who 6. Hermansen K, Davies M, Derezinski T, et al. Dose-response effects of insulin glargine have volunteered for the studies. Martinez Ravn G, Clauson P, Home P. A 26- in type 2 diabetes. Diabetes Care 2010;33: Duality of Interest. F.P. has received honoraria week, randomized, parallel, treat-to-target trial 1555–1560 for speaker fees and/or travel grants from comparing insulin detemir with NPH insulin as 18. Jarrett RJ. Rhythms in insulin and glucose. Sanofi, Eli Lilly & Co, Bristol-Myers Squibb, and add-on therapy to oral glucose-lowering drugs In Endocrine Rhythms. Krieger DT, Ed. New York, Merck & Co. P.L. has received travel grants for in insulin-naive people with type 2 diabetes. Raven, 1979, p. 247–258 scientific meetings from Sanofi and Menarini. Diabetes Care 2006;29:1269–1274 19. Boden G, Chen X, Urbain JL. Evidence for a G.B.B. has received honoraria from Sanofi, 7. Vague P, Selam JL, Skeie S, et al. Insulin circadian rhythm of insulin sensitivity in pa- MannKind, and Eli Lilly & Co. for scientific advis- detemir is associated with more predictable tients with NIDDM caused by cyclic changes in ing and consulting. C.G.F. has served on scien- glycemic control and reduced risk of hypogly- hepatic glucose production. Diabetes 1996;45: tific advisory panels for Sanofi and has received cemia than NPH insulin in patients with type 1 1044–1050 honoraria for speaker fees and/or travel grants diabetes on a basal-bolus regimen with pre- 20. Andres R, Swerdoff T, Pozefsky T, from Bristol-Myers Squibb, Merck & Co., and meal insulin aspart. Diabetes Care 2003;26: Coleman D. Manual feedback technique for Menarini. No other potential conflicts of inter- 590–596 the control of blood glucose concentration. est relevant to this article were reported. 8. European Medicines Agency: Find medicine/ In Automation in Analytical Chemistry. Author Contributions. F.P. enrolled patients, Human medicine/Lantus, insulin glargine. Skeggs LT Jr, Ed. New York, Mediad, 1966, p. performed clamps, analyzed data, and wrote Available from http://www.ema.europa.eu/ 486–491 the manuscript. P.L. performed clamps and ema/index.jsp?curl=pages/medicines/human/ 21. Porcellati F, Rossetti P, Busciantella NR, glucose turnover measurements, analyzed medicines/000284/human_med_000882 et al. Comparison of pharmacokinetics and data, and reviewed and edited the manuscript. .jsp&mid=WC0b01ac058001d124. Accessed 1 dynamics of the long-acting insulin analogs P.Ci. performed clamps and laboratory assays December 2014 glargine and detemir at steady state in type 1 and reviewed and edited the manuscript. P.Ca., 9. Hamann A, Matthaei S, Rosak C, Silvestre L; diabetes: a double-blind, randomized, crossover A.M.A., and S.M. performed clamps and re- HOE901/4007 Study Group. A randomized clin- study. Diabetes Care 2007;30:2447–2452 viewed and edited the manuscript. M.A. per- ical trial comparing breakfast, dinner, or bed- 22. Gastaldelli A, Coggan AR, Wolfe RR. Assess- formed glucose turnover measurements and time administration of insulin glargine in ment of methods for improving tracer estima- reviewed and edited the manuscript. G.B.B. patients with type 1 diabetes. Diabetes Care tion of non-steady-state rate of appearance. J provided the study concept and design, super- 2003;26:1738–1744 Appl Physiol (1985) 1999;87:1813–1822 vised the protocol development and the re- 10. Rossetti P, Pampanelli S, Fanelli C, et al. In- 23. Kuzuya H, Blix PM, Horwitz DL, Steiner DF, search, contributed to discussion, and reviewed tensive replacement of basal insulin in patients Rubenstein AH. Determination of free and total and edited the manuscript. C.G.F. performed with type 1 diabetes given rapid-acting insulin insulin and C-peptide in insulin-treated dia- clamps, analyzed data, performed the statistical analog at mealtime: a 3-month comparison be- betics. Diabetes 1977;26:22–29 analysis,contributedtodiscussion,andre- tween administration of NPH insulin four times 24. Wolfe RR. Radioactive and Stable Isotope viewed and edited the manuscript. F.P. is the daily and glargine insulin at dinner or bedtime. Tracers in Biomedicine: Principles and Practice guarantor of this work and, as such, had full Diabetes Care 2003;26:1490–1496 of Kinetic Analysis. New York, Wiley-Liss, 1992, access to all the data in the study and takes 11. Ashwell SG, Gebbie J, Home PD. Optimal tim- p. 425–426 responsibility for the integrity of the data and ing of injection of once-daily insulin glargine in 25. Radziuk J, Pye S. Quantitation of basal en- the accuracy of the data analysis. people with Type 1 diabetes using insulin lispro dogenous glucose production in Type II diabe- Prior Presentation. Parts of this study were at meal-times. Diabet Med 2006;23:46–52 tes: importance of the volume of distribution. presented in abstract form at the 73rd Scientific 12. Fritsche A, Schweitzer MA, Haring¨ HU; 4001 Diabetologia 2002;45:1053–1084 Sessions of the American Diabetes Association, Study Group. Glimepiride combined with morn- 26. Meneghini L, Atkin SL, Gough SC, et al.; Chicago, IL, 21–25 June 2013. ing insulin glargine, bedtime neutral protamine NN1250-3668 (BEGIN FLEX) Trial Investigators. hagedorn insulin, or bedtime insulin glargine in The efficacy and safety of insulin degludec References patients with type 2 diabetes. A randomized, given in variable once-daily dosing intervals 1. Owens DR, Bolli GB. Beyond the era of NPH controlled trial. Ann Intern Med 2003;138: compared with insulin glargine and insulin insulin–long-acting insulin analogs: chemistry, 952–959 degludec dosed at the same time daily: comparative pharmacology, and clinical appli- 13.StandlE,MaxeinerS,RaptisS,Karimi- a 26-week, randomized, open-label, parallel- cation. Diabetes Technol Ther 2008;10:333– Anderesi Z, Schweitzer MA; HOE901/4009 Study group, treat-to-target trial in individuals with 349 Group. Good glycemic control with flexibility in type 2 diabetes. Diabetes Care 2013;36:858– 2. Lepore M, Pampanelli S, Fanelli C, et al. Phar- timing of basal insulin supply: a 24-week com- 864 macokinetics and pharmacodynamics of subcu- parison of insulin glargine given once daily in the 27. Gerich JE. Physiology of glucose ho- taneous injection of long-acting human insulin morning or at bedtime in combination with meostasis. Diabetes Obes Metab 2000;2: analog glargine, NPH insulin, and ultralente hu- morning glimepiride. Diabetes Care 2005;28: 345–350 man insulin and continuous subcutaneous in- 419–420 28. Radziuk J, Pye S. Diurnal rhythm in en- fusion of insulin lispro. Diabetes 2000;49: 14. Lucidi P, Porcellati F, Rossetti P, et al. Phar- dogenous glucose production is a major con- 2142–2148 macokinetics and pharmacodynamics of thera- tributor to fasting hyperglycaemia in type 3. Heise T, Nosek L, Rønn BB, et al. Lower peutic doses of basal insulins NPH, glargine, and 2 diabetes. Suprachiasmatic deficit or limit cy- within-subject variability of insulin detemir in detemir after 1 week of daily administration at cle behaviour? Diabetologia 2006;49:1619– comparison to NPH insulin and insulin glargine bedtime in type 2 diabetic subjects: a random- 1628 in people with type 1 diabetes. Diabetes 2004; ized cross-over study. Diabetes Care 2011;34: 29. Bolli GB, Gerich JE. The “dawn phenome- 53:1614–1620 1312–1314 non”dacommonoccurrenceinbothnon- 4. Porcellati F, Rossetti P, Pampanelli S, et al. 15. Klein O, Lynge J, Endahl L, Damholt B, Nosek insulin-dependent and insulin-dependent diabetes Better long-term glycaemic control with the L, Heise T. Albumin-bound basal insulin mellitus. N Engl J Med 1984;310:746–750 512 Glargine Evening vs. Morning in T2DM Diabetes Care Volume 38, March 2015

30. Porcellati F, Lucidi P, Bolli GB, Fanelli CG. glucose in blood and urine. Proc Am Diabetes 37. Yang X, Downes M, Yu RT, et al. Nuclear re- Thirty years of research on the dawn phenome- Assoc 1949;9:247–273 ceptor expression links the circadian clock to me- non: lessons to optimize blood glucose control in 34. Schmidt MI, Hadji-Georgopoulos A, tabolism. Cell 2006;126:801–810 diabetes. Diabetes Care 2013;36:3860–3862 Rendell M, Margolis S, Kowarski A. The dawn 38. Prasai MJ, George JT, Scott EM. Molecular 31. Monnier L, Colette C, Dejager S, Owens D. phenomenon, an early morning glucose rise: clocks, type 2 diabetes and cardiovascular dis- Magnitude of the dawn phenomenon and its implications for diabetic intraday blood glucose ease. Diab Vasc Dis Res 2008;5:89–95 impact on the overall glucose exposure in type 2 variation. Diabetes Care 1981;4:579–585 39. Porcellati F, Bolli GB, Fanelli CG. Pharmaco- diabetes: is this of concern? Diabetes Care 2013; 35. Van Cauter E, Polonsky KS, Scheen AJ. Roles kinetics and pharmacodynamics of basal insu- 36:4057–4062 of circadian rhythmicity and sleep in human lins. Diabetes Technol Ther 2011;13(Suppl. 1): 32. Hatlehol R. Paradoxical rise of the blood glucose regulation. Endocr Rev 1997;18:716– S15–S24 sugar concentration in diabetes mellitus. Acta 738 40. Heinemann L, Anderson JH Jr. Mea- Med Scand Suppl 1924;8:211–266 36. Staels B. When the clock stops ticking, met- surement of insulin absorption and insulin 33. Izzo JL. Diurnal (24-hour) rhythm in diabe- abolic syndrome explodes. Nat Med 2006;12: action. Diabetes Technol Ther 2004;6:698– tes mellitus; diurnal variations in levels of 54–55 718