Akio Ohta, Kaori Arai, Ami Nishine, Yoshiyuki Sada, Hiroyuki Kato, Hisashi Fukuda, Shiko Asai, Yoshio Nagai, Takuyuki Katabami and Yasushi Tanaka

Endocrine Journal 2013, 60 (2), 173-177 Or i g i n a l Comparison of daily glucose excursion by continuous glucose monitoring between type 2 diabetic patients receiving preprandial insulin aspart or postprandial insulin glulisine Akio Ohta, Kaori Arai, Ami Nishine, Yoshiyuki Sada, Hiroyuki Kato, Hisashi Fukuda, Shiko Asai, Yoshio Nagai, Takuyuki Katabami and Yasushi Tanaka Department of Internal Medicine, Division of Metabolism and Endocrinology, St. Marianna University School of Medicine, Kawasaki 216-8511, Japan Abstract. Insulin glulisine (Glu) is a rapidly-acting insulin analog with a faster onset of action than the other insulin analogs of its class, which are insulin aspart (Asp) and insulin lispro (Lisp). While insulin Glu is usually injected just before meals, postprandial injection may help to avoid unexpected postprandial hypoglycemia or hyperglycemia by adjusting the insulin dosage according to food intake. However, the effect of postprandial insulin Glu on the glucose profile has not been evaluated. The aim of this study was to compare daily glucose excursion by continuous glucose monitoring (CGM) between multiple daily doses of preprandial insulin Asp or postprandial insulin Glu. In a randomized cross-over trial, we performed CGM to evaluate the 48-hour glucose profile during treatment with the same dosage of insulin Asp just before each meal in 12 hospitalized patients with type 2 diabetes. Patients also received the same dosage of long-acting insulin glargine at bedtime. The average glucose level, standard deviation of the glucose level, mean amplitude of glucose excursion, and daily glucose profile did not differ between preprandial Asp and postprandial Glu. The incidence of hypoglycemic episodes (glucose level<70 mg/dl with or without symptoms) and the area under the curve of glucose<70 mg/dl also did not differ between the two insulin regimens. Multiple daily injections of preprandial Asp and postprandial Glu achieved the same daily glucose excursion profile. Postprandial injection of Glu may provide greater flexibility for patients who require insulin therapy. Key words: Insulin aspart, Insulin glulisine, Continuous glucose monitoring, Multiple daily insulin injections A regimen of multiple daily insulin injections (MDI) that combines a once daily long-acting insulin analog with a rapidly-acting insulin analog before each meal is often required for type 2 diabetic patients with poor glycemic control. Such intensive insulin treatment can minimize postprandial glucose excursions and maintain a good preprandial glucose level, so it may lead to recovery from glucose toxicity and prevent the onset or progression of diabetic complications [1, 2]. The time-action profile of rapidly-acting insulin analogs more closely mimics that of endogenously secreted postprandial insulin compared with regular human insulin [3, 4]. Submitted Jul. 17, 2012; Accepted Sep. 20, 2012 as EJ12-0251 Released online in J-STAGE as advance publication Oct. 6, 2012 Correspondence to: Akio Ohta, Department of Internal Medicine, Division of Metabolism and Endocrinology, St. Marianna University School of Medicine, 2-16-1, Sugao, Miyamae-ku, Kawasaki, Kanagawa, 216-8511, Japan. E-mail: a2oota@marianna-u.ac.jp Injection of rapidly-acting insulin analogs just before meals is usually recommended in the majority of diabetic patients on MDI. However, unexpected postprandial hypoglycemia or hyperglycemia may occur if there is a mismatch between food intake and the dosage of preprandial insulin. If a rapidly-acting insulin analog is injected just after meals, abnormal glycemia due to such mismatch can be avoided by adjusting the insulin dose according to food intake. Insulin glulisine (Glu) is the newest rapidly-acting insulin analog. It was created by substitution of lysine for asparagine at position B3 and substitution of glutamic acid for lysine at position B29 on the B chain of human insulin, and is marketed as a zinc-free formulation. Because of these modifications, insulin Glu displays a more rapid onset of action, earlier peak effect, and shorter duration of action than both regular human insulin and other rapidly-acting insulin analogues [5-7]. Insulin Glu has been approved for use within 20 min after the start of a The Japan Endocrine Society

174 Ohta et al. meal in Europe and the United States [8]. Postprandial administration of insulin glulisine was previously studied in patients with type 1 diabetes, and was shown to provide glycemic control equivalent to that achieved with preprandial injection [9]. However, the daily glucose profile was not evaluated by continuous glucose monitoring (CGM) but by self-monitoring of blood glucose (SMBG), and a direct comparison between preprandial and postprandial injection of a rapidly-acting insulin analog by CGM has not been done in type 2 diabetic patients. In this study, we compared the influence of preprandial (just before each meal) injection of insulin aspart (Asp), which is the rapidly-acting insulin analog most commonly used [10], with postprandial (immediately after each meal) injection of insulin Glu on the 48-hour glucose profile by CGM in an open-label cross-over trial of hospitalized type 2 diabetic patients. Materials and Methods A total of 12 Japanese patients with type 2 diabetes (7 men and 5 women aged 60.2±16.9 years, mean±sd) were studied. They were recruited from the outpatient clinic of St. Marianna University Hospital (Kawasaki, Japan). Inclusion criteria were stable but inadequate glycemic control (glycated hemoglobin (HbA 1c ) >7.5% and variation of HbA 1c by <0.5% within 3 months before recruitment, National Glycohemoglobin Standardization Program (NGSP) values) without use of insulin or oral antidiabetic agents. The exclusion criteria were pregnancy, severe illness, anemia, renal failure (serum creatinine>2.0 mg/dl) and/or overt proteinuria, chronic liver disease, thyroid disease, malignancy, or an episode of severe hypoglycemia requiring assistance within the previous 6 months. All patients gave written informed consent and the study was approved by the Ethics Committee of St. Marianna University School of Medicine. They were admitted to St. Marianna University Hospital (Kawasaki, Japan), and were randomized to the insulin Glu group (n=6) or the insulin Asp group (n=6). Either insulin Glu or insulin Asp was injected just before (from -10 to 0 min) each meal, and a long-acting insulin analog (insulin glargine (Gla)) was injected once daily at bedtime. The duration of each meal was limited to 20 min. Total energy intake and the daily exercise level were fixed at 25 kcal/kg of ideal body weight and 150-200 kcal/day by the hospital dieticians and attending doctors, respectively. The insulin dosages were adjusted to obtain preprandial and 2-hour postprandial glucose levels<120 mg/dl and <180 mg/ dl, respectively. After reaching the target glycemic state and fixing the insulin dosages, the first CGM study was performed. The CGM device (Medtronic MiniMed, Northridge, CA, USA) was attached for 72 hours consecutively. All patients performed SMBG at least 4 times daily using a One touch Ultra (Life scan, Milpitas, CA, USA) and input data into the CGM recorder for calibration. During the first CGM study, the insulin Glu group changed the timing of insulin Glu injection from just before (-10 to 0 min) each meal to just after (0 to 5 min) each meal while maintaining the same dosage, but the insulin Asp group did not change the dosage or timing of insulin Asp injection. Insulin Gla was not altered in either group. After the first study, the rapidly-acting insulin analog was switched from insulin Asp just before each meal to insulin Glu just after each meal or vice versa without altering the dosage, and insulin Gla was continued at the same dose. Starting on the day after switching insulin therapy, the second CGM study was performed for 72 hours in the same way. Assessment of CGM Parameters and Data Analysis After downloading the recorded data, the following parameters were determined from the intermediate 48 hours of data: the average glucose level (AG), the standard deviation (SD) of the glucose level, the mean amplitude of glucose excursion (MAGE), and the area under the glucose concentration versus time curve (AUC) during the 30-min period before each meal, at 1-2 hours, 2-3 hours, and 3-4 hours after each meal, and during the night (10PM to 7AM). The pharmacodynamic effect of insulin Glu and insulin Asp on early postprandial glucose excursion was assessed by calculating the AUC ratio (%) as AUC 0-1h /AUC 0-4h. MAGE was calculated as the arithmetic mean of glucose increases and decreases that exceeded one SD [11]. Data are presented as the mean±sd. The 2-tailed unpaired Student s t-test was used for statistical analysis of differences of mean values between the groups and a probability (p) value of less than 0.05 was accepted as indicating statistical significance. Results Baseline characteristics of the patients on admission are listed in Table 1. Before the first CGM study,

Postprandial insulin glulisine 175 the patients were receiving an average dose of 24.2 U/ day of a rapidly-acting insulin analogue and an average dose of 10.7 U/day of insulin Gla. A comparison of segmental average glucose levels (30 min before each meal, 1-2 hours, 2-3 hours, and 3-4 hours after each meal, and during the night (10PM-7AM)) is shown in Fig. 1. No significant difference of the average glucose level was seen at all time points. A comparison of 48-hour AG, SD, and MAGE is shown in Table 2. Again, there were no differences between the two insulin regimens. The postprandial AUC ratio (%) also did not differ between the two insulins, as shown in Table 3. Hypoglycemic episodes (glucose (<70 mg/dl with or without symptoms)) occurred 4 and 7 times during Table 1 Patient characteristics Gender (Male : Female) (n) 12 (7:5) Age (years) 60.2 ±16.9 BMI (kg/m 2 ) 28.3 ± 4.6 Estimated duration of diabetes (years) 11.4 ±10.7 HbA 1c (NGSP) (%) 9.3 ± 1.9 Urinary C-peptide (μg/day) 32.4 ± 24.3 Diabetic complications: Retinopathy (n) 3 Nephropathy (n) 5 Neuropathy (n) 5 Insulin dose (U/day) Rapidly-acting insulin analogue 24.2 ± 8.8 (Min12- Max38) Long-acting insulin 10.7 ± 5.5 (Min 4- Max18) Data are shown as the mean ± SD or number. BMI, body mass index. Table 2 Comparison of the average and SD glucose level over 48 hours and the mean amplitude of glycemic excursion Insulin preparation Asp Glu p value Average glucose (mg/dl) 121.1 ± 27.2 123.0 ± 23.8 0.87 SD (mg/dl) 26.8 ± 14.7 27.8 ± 15.1 0.87 MAGE (mg/dl) 50.5 ± 28.3 52.1 ± 22.0 0.88 Data are shown as the mean ± SD. SD, standard deviation glucose over 48 hours; Asp, insulin aspart; Glu, insulin glulisine; MAGE, mean amplitude of glycemic excursion Table 3 Comparison of early postprandial glucose excursion with the two insulin preparations Meal Insulin preparation Asp Glu p value AUC 0-1h /AUC 0-4h (%) Breakfast 24.4 ± 4.4 23.8 ± 4.0 0.72 Lunch 25.8 ± 3.1 25.5 ± 3.4 0.83 Dinner 24.6 ± 2.9 25.8 ± 3.7 0.38 Data are shown as the mean ± SD. AUC 0-1h, AUC of glucose measured from 0 to 1 hour postprandially; AUC 0-4h, AUC of glucose measured from 0 to 4 hours postprandially. Fig. 1 Average glucose profile during MDI with postprandial insulin glulisine or preprandial insulin aspart BB, before breakfast (-0.5~0 h); BL, before lunch (-0.5~0 h); BD, before dinner (-0.5~0 h)

176 Ohta et al. lin Glu may be more suitable for postprandial use than either insulin Asp or insulin Lisp. Ratner et al. recently compared preprandial vs. postprandial insulin Glu in 344 type 2 diabetic patients [16], who received insulin Glu at 0-15 min before each meal or 20 min after the start of each meal together with once-daily insulin Gla. The dosages of both insulins were adjusted to obtain the target glucose level. In the postprandial group, the insulin Glu dose was adjusted according to food intake. The final daily insulin dosage, final HbA 1c, and incidence of hypoglycemic episodes did not differ between the two groups, but mean body weight was significantly lower in the postprandial group. The authors suggested that adjustment of the insulin Glu dose led to relatively precise matching of prandial insulin with actual food consumption, which in turn prevented unfavorable weight gain. In the present study, the average glucose levels after lunch and during the 30-min period before dinner tended to be higher in the postprandial insulin Glu group than in the preprandial insulin Asp group. With the same type of insulin preparation, preprandial administration will be more effective for glycemic control. We do not suggest that postprandial insulin glulisine is suitable for all patients. However, to avoid postprandial hypoglycemia or hyperglycemia when using a preprandial rapidly-acting insulin analog, patients need to adjust the size of their meals to the injected insulin dose, whereas postprandial administration allows adjustment of the dose to fit the actual meal. In the present study, CGM revealed no difference of daily glucose excursion between insulin Glu injection just after meals and insulin Asp injection just before meals. The present study had some limitations. First, the number of the patients was small, the observation period was short, and it was done under hospitalized conditions. Second, adjustment of the insulin dose according to food intake was not done. Third, we did not make a comparison with the effect of insulin Glu just before each meal. Fourth, we did not evaluate the influence of different injection timing on the quality of life. When we surveyed the opinions about insulin injection by questionnaire in 100 diabetic patients continuing MDI therapy for at least 1 year at the outpatient clinic of our hospital, about 50% of the patients preferred postprandial injection of a rapidly-acting insulin analog. These patients mentioned flexibility in various situations of daily life and safety (no postprandial hypoglycemia) as the reasons. Thus, a large-scale protreatment with insulin Glu and insulin Asp, respectively. There were no cases of severe hypoglycemia requiring medical intervention. There was also no significant difference of the glucose AUC<70 mg/dl between the two insulins (0.9 ± 1.2 mg*day/dl with preprandial insulin Asp and 0.3 ± 0.5 mg*day/dl with postprandial insulin Glu, p=0.07). Discussion The present study was the first to employ CGM to compare the daily glucose profile achieved with postprandial insulin Glu and preprandial insulin Asp in patients with type 2 diabetes. This study demonstrated no difference of the daily glucose profile between insulin Asp and insulin Glu. The glucose AUCs before and after each meal and during the night were similar with the two insulins. In addition, the SD of glucose excursion, MAGE, and postprandial AUC ratio (%) did not differ between the two insulin regimens. Previously, Bolli et al. compared the pharmacokinetics and pharmacodynamics of insulin Glu (0.2 U/kg) administered before a standard meal with the equivalent dose of insulin Asp by performing a double-blind two-way cross-over trial in type 2 diabetic patients. Plasma glucose and insulin concentrations were measured at -20 min, -10 min, and immediately before the test meal (0 min), every 10 min for 2 hours after the meal, and then every 15 min for the remaining 4-hour period. They observed that the 0-1 hour glucose AUC after insulin injection and the maximum glucose level were significantly lower with insulin Glu than with insulin Asp, while the peak plasma insulin concentration was significantly higher [12]. Similarly, 0.2 U/kg of insulin Glu showed a more rapid onset of action than the same dose of insulin Asp in healthy subjects under glucose clamp conditions [13]. In a cross-over comparison between insulin Glu and another rapidly-acting insulin analog, insulin Lispro (Lisp), Luzio et al. found that the mean of three maximal preprandial subtracted plasma glucose concentrations ( GLU max ) was significantly lower and the mean postprandial plasma insulin concentration was significantly higher when insulin Glu was injected before meals than when insulin Lisp was injected at the same time in type 2 diabetic patients [14]. Furthermore, evaluation by the glucose clamp method showed a faster onset of action for insulin Glu than insulin Lisp (0.2 U and 0.4 U/kg) in nondiabetic subjects [15]. These results suggest that insu-

Postprandial insulin glulisine 177 spective study will be required to confirm the usefulness of postprandial insulin Glu. In conclusion, the AG, SD, MAGE, and daily glucose profile were similar with preprandial insulin Asp and postprandial insulin Glu in patients with type 2 diabetes treated by MDI. Acknowledgments No sources of funding were used to assist in the preparation of this manuscript. The authors have no conflicts of interest that are relevant to the content of this manuscript. References 1. Hirsch IB (1990) Intensive insulin therapy for treatment of type I diabetes. Diabetes Care 13: 1265-1283. 2. Dailey G, Rosenstock J, Moses RG, et al. (2004) Insulin glulisine provides improved glycemic control in patients with type 2 diabetes. Diabetes Care 27: 2363-2368. 3. Freeman JS (2009) Insulin analog therapy: improving the match with physiologic insulin secretion. J Am Osteopath Assoc 109: 26-36. 4. Kato H, Ohta A, Kobayashi S, et al. (2012) patients receiving multiple daily insulin injections. Endocr J 59: 345-352. 5. Rave K, Klein O, Frick AD, et al. (2006) Advantage of premeal-injected insulin glulisine compared with regular human insulin in subjects with type 1 diabetes. Diabetes Care 29: 1812-1817. 6. Dailey G, Rosenstock J, Moses RG, et al. (2004) Insulin glulisine provides improved glycemic control in patients with type 2 diabetes. Daibetes Care 27:2363-2368. 7. Becker RH, Frick AD (2008) Clinical pharmacokinetics and pharmacodynamics of insulin glulisine. Clin Pharmacokinet 47: 7-20. 8. Apidra [prescribing information]. bridgewater NJ: Sanofi-Aventis U.S. LLC, 2009. Available from URL: http://products.sanofi.us/apidora/apidora.pdf.) 9. Garg SK, Rosenstock J, Ways K (2005) Optimized Basal-bolus insulin regimens in type 1 diabetes: insulin glulisine versus regular human insulin in combination with Basal insulin glargine. Endocr Pract 11: 11-17. 10. Heller S, McCance DR, Moghissi E, et al. (2012) Diversity in diabetes: the role of insulin aspart. Diabetes Metab Res Rev 28: 50-61. 11. Service F, Molnar GD, Rosevear JW, et al. (1970) Mean amplitude of glycemic excursions, a measure of diabetic instability. Diabetes 19: 644-655. 12. Bolli GB, Luzio S, Marzotti S, et al. (2011) Comparative pharmacodynamic and pharmacokinetic characteristics of subcutaneous insulin glulisine and insulin aspart prior to a standard meal in obese subjects with type 2 diabetes. Diabetes Obes Metab 13: 251-257. 13. Arnolds S, Rave K, Hövelmann U, et al. (2010) Insulin glulisine has a faster onset of action compared with insulin aspart in healthy volunteers. Exp Endocrinol Diabetes 118:662-664. 14. Luzio S, Peter R, Dunseath GJ, et al. (2008) A comparison of preprandial insulin glulisine versus insulin lispro in people with Type 2 diabetes over a 12-h period. Diabetes Res Clin Pract 79:269-275. 15. Heise T, Nosek L, Spitzer H, et al. (2007) Insulin glulisine: a faster onset of action compared with insulin lispro. Diabetes Obes Metab 9: 746-753. 16. Ratner R, Wynne A, Nakhle S, et al. (2011) Influence of preprandial vs. postprandial insulin glulisine on weight and glycaemic control in patients initiating basal-bolus regimen for type 2 diabetes: a multicenter, randomized, parallel, open-label study. Diabetes Obes Metab 13: 1142-1148.