European Association for the Study of Diabetes – 47th Annual Meeting

September 12-16, 2011; Lisbon, Portugal; Full Commentary – Insulin and Insulin Therapy – Draft

Executive Highlights

At this year’s meeting, we noticed a somewhat reduced focus on novel insulin products. Outside of some compelling data on the use of Halozyme’s PH20 in insulin pumps, we heard little regarding other ultra- rapid acting insulin formulations in development, including MannKind’s Afrezza and Biodel’s several early stage candidates (given that two of Biodel’s early-stage compounds were dropped from development after EASD, this was unsurprising in hindsight). On the basal insulin front, the spotlight again fell on Novo Nordisk’s degludec and degludecPlus, which were filed for approval in both the EU and US shortly after the conference. Dr. Stephen Atkin (Hull York Medical School, Hull, UK) presented data from a phase 3 trial (which was also presented in a poster at ADA) that demonstrated the ability of insulin degludec to be dosed flexibly daily without compromising safety and efficacy relative to insulin glargine dosed at a fixed time each day. He argued that this added benefit of degludec could help greatly improve adherence among certain patient populations (i.e. adolescents, nursing home residents, individuals reliant on caregivers). Dr. Irl Hirsch (University of Washington, Seattle, WA) and Dr. Allan Vaag (Rigshospitalet, Copenhagen, Denmark) presented separate posters for degludecPlus that highlighted the premixed product’s potential to reduce both nocturnal and overall hypoglycemia – qualities of the insulin that we think will be most compelling to regulators and payors. More generally, and more so than in the US, in Q&A sessions throughout the meeting, we heard repeated calls for early, more advanced treatment of diabetes, including the earlier use of insulin therapy to help preserve beta cells and endogenous insulin production. Of course, we eagerly anticipate learning much more about the benefits of early insulin initiation (as well as many other topics regarding insulin therapy) through the ORIGIN study (Lantus in patients with early type 2 diabetes or prediabetes); topline data from this trial are expected to report at next year’s ADA meeting and this is already becoming the buzz in conference hallways planet-wide – speculations abound regarding ORIGIN’s results and what these results may mean for the use of insulin therapy and oral drugs at all stages of disease.


Table of Contents 


Detailed Discussion and Commentary

Oral Presentations: Insulin Therapy


Stephen Atkin, PhD (Hull York Medical School, Hull, UK)

Dr. Atkin presented data from a phase 3 trial examining the safety and efficacy of a flexible dosing regimen of insulin degludec (given once daily in a rotating morning and evening schedule, creating 8- 40 hour dosing intervals) compared to insulin degludec and insulin glargine administered at the same time each day in people with type 2 diabetes (n=687). The same trial was presented in a poster at ADA in June (see page 163 of our ADA 2011 Full Report for our full coverage of this trial). As a reminder, at 26 weeks, both degludec arms provided similar improvements in glycemic control (approximately 1.1- 1.3%), rates of confirmed overall hypoglycemia (3.5-3.6 episodes/patient/year), rates of confirmed nocturnal hypoglycemia (although numerically lower with both degludec arms; 0.6 vs. 0.8 events/patient/year), and weight gain (approximately 2.9-3.5 lbs) compared to the fixed insulin glargine regimen. New in this presentation, Dr. Atkin offered his view on the patient populations that would benefit most from the opportunity to flexibly dose insulin degludec: 1) adolescent patients who have difficulty keeping to a strict insulin dosing regimen; 2) nursing home residents who are dosed at inconsistent times each day; and 3) patients who rely on caregivers to administer their insulin who have variable schedules themselves.



Andrea Palermo, MD (University Campus Bio-Medico, Rome, Italy)

Dr. Palermo presented results from an open-label pilot study on the ability of Oral-lyn, Generex’s buccal insulin spray, to reduce A1c in patients with impaired glucose tolerance. In the treatment group, Oral- lyn was used in combination with diet and exercise, while the control group was treated with diet and exercise alone. Patients in the study received 12 puffs of the spray, which is absorbed in the oral mucosa and hypopharynx, at meals split into two doses. After six months of treatment, A1c was statistically significantly reduced in the experimental group as compared to the control group. There was additionally a non-significant trend towards decreased BMI from baseline to six months in the treatment group and no hypoglycemic events were reported. According to Dr. Palermo, these results suggest that treatment with buccal insulin spray could be useful for the treatment of IGT. The study is still ongoing.

  • Oral-lyn is a human DNA insulin buccal spray that is absorbed in the oral mucosa and hypopharynx. It is dissolved in neutral pH buffer, has peak efficiency after one hour, and has 15% bioavailability. It is not absorbed in the lung. Its effects on oral glucose tolerance tests in people with impaired glucose tolerance have previously been reported in a 2011 Diabetes, Obesity, and Metabolism paper by Palermo et al.
  • This randomized, controlled pilot trial compared a regimen of buccal spray insulin plus diet and physical exercise versus diet and physical exercise alone in patients with impaired glucose tolerance (IGT). Patients receiving the buccal spray insulin received 12 puffs at meals split into two equal doses. The trial’s primary endpoint was difference in A1c of at least 0.3% at six months of treatment between the experimental and control groups. Generation of antibodies to insulin was a secondary endpoint. There were eleven patients in both the experimental and control groups. The trial was not double-blinded.
  • A1c was statistically significantly reduced in the treatment group as compared to the control group. There was a trend towards decreased BMI from baseline to six months in the treatment group, but the decrease was not statistically significant. No hypoglycemic or other adverse events were observed during the study period. Additionally, there were no reports of insulin antibody generation.
  • This study is still ongoing, but Dr. Palermo indicated that treatment with buccal insulin spray could be useful for patients with IGT. Moreover, the easy, pain-free administration of buccal insulin could increase compliance and diminish adverse events.

Questions and Answers

Q: Do you think it would not have been more robust to do a double-blind study? There is a question of whether you made their exercise more interesting by adding puffing of insulin. Is there a reason you didn’t do a double-blind study?

A: That is a good and important question. This is just a preliminary study, a pilot study. It aimed to compare Oral-lyn versus the standard of care, which is physical activity and diet. Once we finish the trial, if the reduction of A1c is confirmed, we plan to design a bigger and larger study that is placebo-controlled and double-blinded. We need a larger group of patients for long-term treatment.



Malgorzata Saruysz-Wolska, MD, PhD (Medical University of Lodz, Lodz, Poland)

Dr. Saruysz-Wolska described the findings of a randomized, prospective trial that assessed the glycemic effects and safety of splitting high doses of insulin and injecting them into two sites for severely insulin resistant patients who live in places where U500 insulin is not available. Patients enrolled in the study had type 2 diabetes and a baseline A1c of > 8%, were treated with a multiple insulin dose regimen, and used > 60 IU insulin as a once-daily single dose. In the study, participants were assigned to receive insulin injections either in two sites (a dose of > 60 IU split into two equal parts and given in symmetrical locations (n=16)) or one site (a usual injection of one > 60 IU dose at a usual site (n=15)). A1c, plasma lipids, and hypoglycemia occurrence were assessed and patient satisfaction was measured via a visual analogue scale every three months for twelve months. At baseline, mean A1c was 10.3% and 10% in the two-site and one-site groups, respectively. Mean A1c decreased to 9.1% in the two-site group and was stable in the one-site group after three months and decreased to 8.6% in the two-site group and 9.9% in the one-site group after six months. After twelve months, the mean A1c of the two-site group was 8.8% (a statistically significant decrease from baseline, p< 0.05), while that of the one-site group was 10.4%. Dr. Saruysz-Wolska suggested that the greater A1c improvements observed with the two- site injection regimen may be due to improved insulin absorption. Occurrence of minor hypoglycemic episodes was similar between the two groups and no major hypoglycemia occurred during the study. Mean weight remained stable over the course of the study in both groups. Moreover, treatment satisfaction increased in the two-site group from baseline to 12 months (53 to 72, P< 0.05), but stayed relatively constant in the one-site group (55 at baseline to 58 at 12 months).

Questions and Answers

Q: What you are doing in the intervention group is changing conditions. There is evidence that if you change lighting in a factory, you change results. Do you think that could have changed your results?

A: That is a disadvantage of the study. Another is that patients were not enrolled within six months. It took nine years to collect all of the data. You might also guess that we might change our performance over the nine years as well.

Q: What type of insulin did you use?

A: These were intensively treated patients treated with regular human insulin and NPH. In my country, long-acting analogs are not reimbursed. But the percentage of doses over 60 units in each patient varied over the years. Some were only 80 units at night. Some injected once daily before dinner. It was pretty different.



Beverley Balkau, PhD (INSERM CESP U1018, Villejuif, France)

Dr. Balkau presented a one-year interim analysis from the CREDIT study that aimed to identify factors predictive of weight gain in patients newly treated with insulin. CREDIT is a four-year observational study (n=2,442) designed to evaluate the relationship between blood glucose control and cardiovascular events in people with type 2 diabetes newly treated with insulin – with the available one-year data, Dr. Balkau analyzed two subgroups of patients gaining either greater or less than the mean weight gain for the population (1.6 kg [3.5 lbs]). As might be expected, patients gaining over 1.6 kg (3.5 lbs) of weight had a higher A1c at one year and had a higher A1c at baseline. Factors predictive of weight gain over 1.6 kg (3.5 lbs) included higher daily dose of insulin, BMI at insulin initiation (not intuitively, patients with a lower BMI at baseline were more likely to gain weight), utilization of three or more other glucose- lowering medications, and female gender – with such an important clinical issue, we look forward to further analyses for a more definitive breakdown of these factors, such as what background medications were used.

Questions and Answers

Q: I want to ask what this had to do with cardiovascular outcomes?

A: The CREDIT study is going for four years when we hope we have an adequate number of events. This is just a one-year analysis.

Q: I think we’re worried about those that gain 10 kg or so; have you made analysis of those patients?

A: I’m sorry I don’t have data on that, but that’s a very good point to look further into this data.

Q: I find the triple therapy at baseline consequences interesting. The question then is what happened to triple therapy in the first year? Was there any plan to discontinue it?

A: It’s an observational study, so it’s up to the physician to decide what occurs. The separate treatments will eventually be looked at.

Q: Could you explain why people with higher BMI gained less than people with lower BMI?

A: I’m sorry I don’t have an explanation.

Q: I have seen that patients under worse control end up losing weight through glucosuria, only to gain it back once under control – isn’t that what the data are reflecting?

A: That could be a reasonable explanation.



Antonio Pontiroli, MD (University of Milan, Milan, Italy)

Dr. Pontiroli presented the results of a meta-analysis evaluating the A1c-lowering efficacy and hypoglycemia of various insulin treatments. The meta-analysis included data for 21,347 patients (5,509 without concomitant oral antidiabetics [OADs], 15,838 with OADs from a total of 67 studies with targeted insulin treatments lasting up to one year, published between 1991 and 2010. The meta-analysis found that basal regimens caused less hypoglycemia than twice-a-day and prandial regimens; however, they had less A1c-lowering efficacy. It was determined that in basal regimens, new analogs caused less hypoglycemia, and in prandial regimens, new analogs brought about better glycemic control. Compared to NPH, insulin glargine and insulin detemir induced less hypoglycemia and had less of an effect on A1c. In closing, Dr. Pontiroli commented that it appears possible to adjust insulin treatment for each patient depending on their primary goal – A1c lowering for younger patients, and avoiding hypoglycemia for older patients.

Questions and Answers

Q: You’ve shown us a huge amount of data that is difficult to assimilate. How easy will it be to take all the data and make practical guidelines for the clinician?

A: This is something we have to think about. If we have young patients, and the main aim is to prevent complications, we should be strict and us more strict/intense regimens, but if we have a 65-year-old with diabetes who has established complications, we probably just want to avoid hypoglycemia.


Poster Presentations: Insulin Therapy in Type 1 Diabetes


M. Homiest, D. Much more, L. Morrow, and D. Vaughn

This poster highlighted new data from Halozyme’s study of PH20 in insulin pumps, which was previously presented in less detail and fewer study subjects at Day #2 of AACE (see April 15, 2011 Closer Look at Consistent with previous injection studies, the co-infusion of PH20 with aspart in an insulin pump resulted in improved pharmacokinetic and pharmacodynamic profiles over aspart alone. We were most impressed to see the mixed test meal results, which suggest substantial glycemic control improvements with PH20 plus aspart vs. aspart alone. Interestingly, this effect disappeared on day three due to the accelerated insulin absorption associated with an older infusion set. Adverse events were comparable between the aspart-only group and PH20 plus aspart group. Overall, we think this data bodes well for Halozyme, especially considering Biodel’s recent decision to drop BIOD-105 and BIOD-107 from further development. With this in mind, the super-rapid acting insulin landscape is looking better and better for Halozyme, although we’ll be interested to see if this data holds up in larger studies.

  • This study examined the coinfusion of aspart with Halozyme’s PH20 in insulin pumps over 72 hours. Twenty individuals with type 1 diabetes on insulin pumps participated (16 completers). Patients were admitted to the clinical research center at 5 pm on Day 1, and randomly assigned treatment with aspart alone or aspart coinfused with PH20. Basal rates were adjusted to target a fasting blood sugar of 110 mg/dl. Euglycemic clamps were performed on days two and four, with boluses of 0.15 units per kilogram administered. Test meals were provided throughout the study (45-50% carbohydrates, 18-22% protein, 30-34% fat), postprandial glucose was measured by YSI, safety and tolerability were assessed with skin biopsies, and participants were readmitted 5-14 days later for repeat procedures on the other study drug (either aspart or aspart plus PH20).
  • When delivered via an insulin pump, PH20 accelerated the exposure to aspart. PH20 plus aspart had a “faster in” profile, with PH20 plus insulin aspart exposure in the first hour measuring 191% of aspart alone (p < 0.0001). Additionally, PH20 coinfused with aspart had a “faster out,” demonstrating a 35% decreased exposure beyond two hours compared with aspart alone (p = 0.005).
  • Overall, insulin absorption accelerated as the infusion set aged. The coinfusion of PH20 with aspart reduced this variability over the course of the study.
  • Aspart coinfused with PH20 had a faster on, faster off profile than aspart alone. With PH20, the onset of insulin aspart action (Early t50%) was accelerated from 47 to 35 minutes (p=0.05). Additionally, the mean duration of insulin aspart action with PH20 was reduced from 164 to 147 minutes (p=0.003).
  • Mixed test meal results suggest the profile of PH20 leads to improved early glycemic control. This effect diminished as the infusion set aged, such that by day three, the difference between aspart alone and aspart plus PH20 was not statistically significantly different.

Aspart →→ Aspart plus PH20

Blood Glucose (mg/dl)

Day 1

Day 2

Day 3

1 hour

165 →→  128

(p = 0.0055)

136 →→ 123

(p = 0.056)

135 →→ 125

(p = 0.46)

2 hour

155 →→ 108


153 →→ 119


131 →→ 137

(p = 0.66)

  • Adverse events were similar between the aspart alone and aspart plus PH20 groups. Overall, 72% (13 out of 18) of aspart plus PH20 participants and 80% (16 out of 20) of aspart-only participants had an adverse event. In the aspart-PH20 group, five adverse events in three participants were judged related to the study drug, compared with seven events in four participants for aspart alone. Pump infusion site events (erythema, hemorrhage, pruritis, edema, inflammation) were comparable between the two groups: six events in three participants for aspart plus PH20 vs. five events in two participants for aspart alone. Finally, assorted unrelated adverse events were abnormally high in the control group, with 20 events in 12 participants in the aspart-only group, compared to five events in five participants in the aspart plus PH20 group. We are not clear on why the control group experienced such a high rate of unrelated adverse events, which were defined as: anemia, neck pain, retinal hemorrhage, cough, nasal congestion, pruritis, nausea, diarrhea, abdominal pain, headache, and dizziness.
  • Skin biopsy results demonstrated transient and local effects of PH20. Staining for hyaluronan demonstrated local depletion at time of cannula removal, with full restoration within 18-24 hours. H&E staining revealed comparable local mild inflammatory response to both treatments at the cannula insertion track.


Poster Presentations: Insulin Therapy in Type 1 Diabetes


Irl Hirsch, Edward Franek, Jean-Pierre Courreges, Henriette Mersebach, Patrik Dykiel, Bruce Bode

In a poster session largely focused on insulin degludec, Dr. Irl Hirsch presented new data on the use of premix insulin (70% degludec and 30% aspart), also called degludecPlus, in people with type 1 diabetes. Although Dr. Hirsch admitted during Q&A that using premix was atypical in type 1 diabetes, the data in this trial showed comparable glycemic efficacy to detemir, with the obvious advantage of at least one fewer injection per day. Notably, this trial reported a significantly reduced rate of nocturnal hypoglycemia compared to the comparator. Dr. Hirsch commented that this has been a consistent pattern in the degludec development program. It’s unclear to us if this will be a clinically meaningful differentiator for degludec given the market dominance of glargine and its impending patent expiry. Nevertheless, we’ll be interested to hear more about the practicality and flexible-dosing options of degludec and degludec plus as the insulin gets closer to market.

  • This 26-week trial compared the efficacy and safety of IDegAsp (a fixed premix of 70% degludec and 30% aspart) with insulin detemir, both in combination with mealtime aspart. IDegAsp was dosed once-daily at any meal with insulin aspart administered at the remaining meals. Insulin detemir was dosed according to the label, with insulin aspart at all meals. Doses of IDegAsp and detemir were adjusted weekly, on the basis of the mean pre- breakfast plasma glucose values measured on three consecutive days, to a fasting target of <5 mmol/l (90 mg/dl). Pre-breakfast and pre-lunch aspart doses were also adjusted weekly to target a similar pre-meal value, while pre-dinner doses were adjusted to target a bedtime level of <8 mmol/l (144 mg/dl).
  • This study in people with type 1 diabetes randomized 366 patients to IDegAsp plus aspart and 182 patients to detemir plus aspart. Mean A1c was 8.3% at baseline, mean BMI was 26.2 kg/m2 in the IDegAsp group and 26.7 kg/m2 in the detemir group. Mean duration of diabetes was 17.2 years for the IDegAsp group and 17.9 years in the detemir group. Of those on IDegAsp, 97% were between 18 and 65 years, compared to 92% in the detemir group.
  • At 26 weeks, no significant differences between the groups were observed in glycemic control as measured by A1c or fasting plasma glucose. In both groups, mean A1c declined to 7.6% from a baseline of 8.3%. Fasting plasma glucose declined to 8.7 mmol/l (156.6 mg/dl) in the IDegAsp group and 8.6 mmol/l (154.8 mg/dl) in the detemir group.
  • The rate of confirmed hypoglycemia was not significantly different between the two groups, although those in the IDegAsp group had 37% less nocturnal hypoglycemia than those in the detemir group. The rate of nocturnal confirmed hypoglycemia (onset between 12:01 AM and 5:59 AM) was 3.7 episodes per patient year in the IDegAsp group, compared to 5.7 episodes per patient year in the detemir group (p=0.0004). Very few severe hypoglycemic episodes were reported for IDegAsp and insulin detemir – the rate was 0.33 and 0.42 episodes per patient year respectively. Dr. Hirsch noted that although the difference was not significant, it trended in the direction favoring degludec.

Questions and Answers

Moderator: Dr. Geremia Bolli (University of Perugia, Perugia, Italy): I have a question about using premix insulin in type 1 diabetes. We tend to teach our patients flexibility, and premix doesn’t necessarily go along with that. When would you use this fixed combination in clinical practice?

A: I agree with you completely about using premix. Anytime you develop a new insulin these days and seek regulatory approval, you need to study it in both type 1 and type 2 diabetes. We occasionally use premix insulin in type 1 patients, although it’s very rare. Sometimes we use it in pediatrics or the homeless. But I agree, we don’t generally recommend this.


Poster Presentations: Insulin Therapy in Type 2 Diabetes II


Allan Vaag, MD (Rigshospitalet, Copenhagen, Denmark)

Dr. Vaag presented previously unreported data from a trial comparing the combination of insulin degludec/insulin aspart (Novo Nordisk’s DegludecPlus) against biphasic insulin aspart 30. In the trial, insulin-naïve type 2 diabetes patients uncontrolled on oral therapy were randomized to receive metformin and either twice-daily insulin degludec/aspart (n=61) or aspart 30 (n=62) for 16 weeks. As expected in the treat-to-target design, both groups experienced similar declines in A1c (~8.6% to 6.7%). However, patients treated with insulin degludec/aspart exhibited significantly greater declines in mean fasting plasma glucose (205.4 to 115.3 mg/dl vs. 210.8 to 135.1 mg/dl; p<0.05) with reduced mean rates of hypoglycemia (2.9 vs. 7.3 episodes/patient year; p<0.05); patients trended toward reduced rates of nocturnal hypoglycemia (0.4 vs. 1.1 episodes/patient year), though results were not significant. This continues to round out Novo Nordisk’s clinical package for degludecPlus; we will be interested to see results from East Asian patients, given the high use of premixed insulins in China (see Novo Nordisk’s Capital Markets Day from the May 11, 2011 Closer Look).

  • In this 16-week trial, insulin-naïve type 2 diabetes patients were randomized to receive metformin and either twice-daily insulin degludec/aspart or aspart 30. Patients were required to be uncontrolled on oral therapy (on 1-2 oral drugs for ≥2 months); and duration of diabetes was similar in both groups (9.0 vs. 8.6 years). Doses were administered before both breakfast and dinner; initial dosing was 6 units, with patients adjusting dosing once a week toward a pre-breakfast and pre-dinner goal of 72-108 mg/dl.
  • While both groups experienced similar substantial declines in A1c (~8.6% to 6.7%), patients treated with combination insulin degludec/aspart exhibited significantly greater declines in mean fasting plasma glucose (205.4 to 115.3 mg/dl vs. 210.8 to 135.1 mg/dl; p<0.05) with reduced mean rates of hypoglycemia (2.9 vs. 7.3episodes/patient year defined as plasma glucose <56 mg/dl; p<0.05). The percentage of patients achieving target A1c levels were similar (74% vs. 77% for A1c of <7.0%; 48% vs. 48% for A1c of <6.5%), although more patients were able to achieve target without hypoglycemia using combination degludec/aspart (defined as no confirmed hypoglycemia in the last four weeks of treatment; 67% vs. 40% for A1c of <7.0%; 40% vs. 25% for A1c of <6.5%). Patients trended toward reduced rates of nocturnal hypoglycemia (0.4 vs. 1.1 episodes/patient year) though results were not significant.
  • Similar increases in weight were observed in both groups (2.4 lbs with degludec/aspart vs. 3.1 lbs with biphasic aspart 30). Dr. Vaag suggested overall rates of adverse events were similar between the groups as well. Mean daily insulin dose in the degludec/aspart group was slightly lower (0.57 U/kg vs. 0.66 U/kg) at the end of the trial.

Questions and Answers

Q: What about severe hypoglycemia?

A: There were no episodes of severe hypoglycemia seen in either arm of the trial.


Corporate Symposium: A Step Towards Individualizing Care to Achieve Early and Sustained Glycemic Control (Sponsored by Sanofi)


Hertzel Gerstein, MD (McMaster University, Hamilton, Canada)

Dr. Gerstein reviewed the aims, study design, and baseline characteristics of the ORIGIN trial. As a reminder, ORIGIN aims to determine: 1) whether insulin glargine can improve cardiovascular outcomes in high-risk patients with IFG, IGT, or early type 2 diabetes; and 2) whether omega-3 fatty acids can reduce cardiovascular mortality in people with IFG, IGT, or early type 2 diabetes. (In addition, cancers are being prospectively measured in ORIGIN.) In the two-by-two factorial design study, patients were randomized to receive insulin glargine + omega-3 fatty acids, insulin glargine + placebo, omega-3 fatty acids, or placebo, and were followed up for an average duration of over six years. At baseline, 81.9% had established diabetes, 6.3% had newly diagnosed diabetes, and 11.5% had IFG and/or IGT. No preliminary data was disclosed; results will be available in mid-2012 at ADA.

  • Dr. Gerstein highlighted the benefits and concerns of insulin therapy. In terms of benefits: 1) insulin does not have a maximum or minimum dose; 2) it is “easily titrated” and virtually painless (Editor’s note – we do not think all patients would agree with this because it largely depends on the care in which they are under and the training they receive); 3) it has no contraindications or true drug interactions; 4) it is available with easy-to-use insulin delivery devices; 5) patients often only need one dose per day if they start early; 6) medicine has more experience with insulin than almost any other drug; and 7) “even a cardiologist can use insulin”. Meanwhile, there have been concerns that excess glucose lowering could be harmful, insulin- mediated hypoglycemia could be harmful, exogenous insulin could be atherogenic, weight gain could cause cardiovascular outcomes, and that exogenous insulin could be carcinogenic.

Questions and Answers

Q: Can you predict what the risk reduction is likely to be in ORIGIN based on your background in setting up this study?

A: I can’t. If I knew the answer, I wouldn’t do the study. The beauty of clinical trials is that we don’t know the answer. That’s why we do large outcomes trials. We can hope, but we can’t predict.

Q: You told us that insulin was pretty easy to use, especially with all the advances in technology. In your clinical experience, why is it that patients and physicians resist it?

A: I think patients resist insulin to a large extent because the medical community has poisoned people’s perceptions about insulin. We all know there are many physicians who threaten their patients with insulin. Insulin has been used, and is still being used, as the punishment you get for not taking care of your diabetes appropriately. If you’re a patient and your doctor says, “I think you have to go on insulin,” the patient thinks, “Oh my god, I’ve failed my diabetes.” I don’t think we’ve done anyone a service by promoting that idea.



Stefano Del Prato, MD (University of Pisa, Pisa, Italy)

Dr. Del Prato emphasized the importance of early diagnosis and treatment of diabetes, individualized treatment, and treatment of cardiovascular risk factors. Briefly reviewing the ADVANCE, ACCORD, and VADT trials, he noted that chronic exposure to glucose could have a negative legacy effect; that is, delayed treatment could increase the risk of complications and mortality. In terms of individualized treatment, Dr. Del Prato suggested that insulin could be appropriate for both patients with newly onset diabetes (insulin deficient, A1c >9%, severe symptoms) and longstanding diabetes (insulin resistant, free of complications, at target with alternate treatments). He mentioned that unfortunately, there is still a lot of clinical inertia that delays the treatment of diabetes, especially with insulin. Lastly, Dr. Del Prato stressed the importance of promptly treating all cardiovascular risk factors, as STENO-2 results suggested there could be a legacy effect on outcomes, even when each individual risk factor does not decrease significantly.

Questions and Answers

Q: Why do you think there is so much clinical inertia? Why don’t people get the message that treatment, whether early or late, is essential?

A: If the message is to be conveyed to the patient, someone has to convey it. There is still this idea that type 2 diabetes is a mild condition. There is no mild form of diabetes. It is a severe disease from the beginning, and we need to first of all convince ourselves. Then we can convey [this] much better to patients, and perhaps even improve adherence.



Hannele Yki-Järvinen, MD (University of Helsinki, Helsinki, Finland)

Dr. Yki-Järvinen discussed the potential benefits of early insulinization for people with diabetes, including the prevention/delaying of complications, the (slight) preservation of insulin secretion in newly diagnosed patients, the reduction of future insulin requirements, the reduction of weight gain, and the increased likelihood of achieving an A1c of less than 7.0%. To support her claim that insulin helps preserve insulin secretion, Dr. Yki-Järvinen highlighted the results from a study conducted in China (n=382), in which patients with newly diagnosed diabetes who received insulin treatment for two weeks had a significantly higher insulin secretion one year later than patients who received sulfonylurea for two weeks (Weng et al., Lancet 2008). In terms of insulin initiation, Dr. Yki-Järvinen proposed for patients to start with a simple basal regimen, since it causes less hypoglycemia and less weight gain than prandial insulin (Bretzel et al., Lancet 2008; Holman et al., NEJM 2009).

Questions and Answers

Q: If you use insulin and you suppress C-peptide by giving insulin as opposed to a sulfonylurea that might increase C-peptide and insulin, is there any advantage?

A: The C-peptide goes down I think mainly because you lower glucose and improve insulin sensitivity, because it is seldom that the insulin doses are such that exogenous insulin would suppress the endogenous. You would need quite high concentrations for that. I don’t think we have any study that would have addressed that question. It is an important reminder that insulin should come from to the liver from where it normally comes from and not the periphery.

Q: What is your approach to start dosing insulin for an individual – do you take weight and A1c into account?

A: It is automated – everyone gets 10 units to start. The only important thing the patient needs to learn is to measure fasting glucose every morning. If his or her fasting glucose is above 100 mg/dl per day for three consecutive days, then they are supposed to increase their insulin dose by two units. By this mechanism one could increase to 240 units per day in a year; certainly, most patients will be in good control before then. This is a safe way to do it – hypoglycemia risk is really very low when you allow rapid self-titration, but only in small increments.

Q: Do they stick with that, or do you have to continue to give them confidence?

A: That is an important point. One of the lessons we’ve learned over the years is that the one of the most important jobs of the nurse is to check that patients are actually self-titrating, because patients tend to forget. Whenever we give prescriptions for basal insulin, we write then out for 10-200 units, and tell the pharmacists to give them what the patients need, because we don’t know that. Even if we have nice predictive algorithms, it’s more of an academic than a practical exercise.


Corporate Symposium: Advancing Insulin – Designing for Better Therapy (Sponsored by Novo Nordisk)


Peter Kurtzhals, PhD (Novo Nordisk, Bagsvaerd, Denmark)

Dr. Kurtzhals provided an overview of the structural design of insulin degludec, highlighting in particular the use of subcutaneous multi-hexamer depots to extend the insulin product’s release into the body. Dr. Kurtzhals emphasized that because of insulin degludec’s ultra-long duration of action and minimal degree of variability, the insulin product more closely mimicked the action profile of endogenously secreted insulin.



Tim Heise, MD (Profil Institute for Metabolic Research, Neuss, Germany)

Dr. Heise reviewed the results from several pharmacokinetic and pharmacodynamic studies to highlight the longer half-life (25.4 hours for degludec vs. 12. 5 hours for glargine), lower within day variability, and flatter action profile of insulin degludec versus insulin glargine. Dr. Heise argued that degludec’s action profile may make it safer to target lower fasting glucose levels without risking the development of hypoglycemia.



Chantal Mathieu, MD, PhD (Catholic University of Leuven, Leuven, Belgium)

Dr. Mathieu reviewed the results from a phase 3 clinical trial that examined the safety and efficacy of a flexible dosing regimen of insulin degludec (given once daily in a rotating morning and evening schedule) versus a fixed dosing regimen of insulin glargine. This trial was presented as both a poster at ADA in June (see page 163 of the ADA 2011 Full Report) as well as in an oral presentation at this year’s EASD meeting. At 26 weeks, the degludec arm provided similar improvements in glycemic control, rates of hypoglycemia (with a trend toward reduced nocturnal hypoglycemia), and weight gain compared to the fixed insulin glargine regimen. To Dr. Mathieu, these results demonstrated that insulin Degludec could be administered at different times each day without compromising efficacy and safety. Dr. Mathieu stressed that this increased flexibility may help greatly improve insulin therapy adherence given the day-to-day variability in the lives of patients.



Miles Fisher, MD (University of Glasgow, Glasgow, UK)

Dr. Fisher reviewed the results from two treat-to-target phase 3 studies that examined the safety and efficacy of basal-bolus therapy with insulin degludec vs. insulin glargine in people with type 1 and type 2 diabetes. Dr. Fisher underscored that in both trials, the insulin degludec arm led to statistically significant reductions in the rates of nocturnal hypoglycemia (25% reduction, p<0.05). Moreover, a pre- specified meta-analysis of pooled data from seven phase 3 studies comparing once daily insulin Degludec and insulin glargine as a basal-only or a basal-bolus regimen in people with type 1 and type 2 diabetes found a statistically significant 9% reduction in overall confirmed hypoglycemia and 26% reduction in nocturnal hypoglycemia with insulin degludec. Over the maintenance period (from week 16 to the end of the study), there was a 16% reduction in overall hypoglycemia and 32% reduction in nocturnal hypoglycemia with insulin degludec (all reductions were statistically significant).



Bernard Zinman, MD (Mount Sinai Hospital, University of Toronto, Toronto, Canada), Peter Kurtzhals, PhD (Novo Nordisk, Bagsvaerd, Denmark), Tim Heise, MD (Profil Institute for Metabolic Research, Neuss, Germany), Chantal Mathieu, MD, PhD (Catholic University of Leuven, Leuven, Belgium), and Miles Fisher, MD (University of Glasgow, Glasgow, UK)

Q: Is there any cancer risk with degludec?

Dr. Kurtzhals: We did comprehensive battery of tests, including IGF-1 receptor assays, mitogenic potency assays, and tests in animal cancer models. Degludec has been through this entire battery. It has a low affinity for the IGF-1 receptor and insulin receptor. It has low mitogenic potency in cancer cell lines. Through all these assays, degludec has looked very favorable regarding these parameters.

Q: Degludec looks great in terms of variability. But patients have a lot of variability themselves. Is it really that beneficial to have such a stable insulin when the lives of patients are so variable?

Dr. Mathieu: I think that this stable profile will prove even more beneficial once you go into real life. If you look at longer studies, the stability of insulins really translate to reduced hypoglycemia in real life. Yes, patients will have variable lives, but having a more stable insulin will give them a more stable profile regardless.

Dr. Zinman: If patients have variable exercise patterns and a variable insulin, then you have variability squared. At least having a stable insulin eliminates one of those variables.

Q: What about postprandial glucose control?

Dr. Heise: We didn’t look at postprandial control, but I think there are some projects in the clinic. I would not expect, however, to see a large difference in postprandial control. DegludecPlus can significantly reduce both fasting and postprandial glucose.

Q: Insulin can have a cutaneous reaction. Was there anything specific with degludec?

Dr. Heise: That was one of our initial concerns in early trials. However, after very careful assessment, we really did not find a problem, even in the phase 3 program.

Q: Are there any problems with antibodies?

Dr. Kurtzhals: We monitored antibodies in some of the phase 3 trials. Consistently, we found that antibody levels are generally very low. The levels were actually lower than with comparator insulins.

Q: Is there any concern with the safety of degludec in pregnant women? Will it pass the placental barrier?

Dr. Mathieu: Its human insulin that gets into the blood stream. It’s still early, but from a logical point of view, it is human insulin so there should be low risk.

Q: You mentioned physiological insulin replacement. Of course, it’s not normal for insulin to be secreted under the skin. Achieving therapeutic insulin levels with subcutaneous administration is likely to be different than getting insulin portally. Are there any data examining this and is there any understanding of the consequence of this difference?

Dr. Heise: Yes, ideally insulin would work on the liver first, but we are not going to achieve that with subcutaneous insulin administration. But all insulins have to live with this drawback.

Dr. Zinman: Years ago, we were looking at portal versus peripheral insulin infusions in animals. You get euglycemia with lower levels of portally infused insulin. Insulin levels were about double with peripheral insulin administration. Whether it has any other consequences we don’t really know. But my guess is that it’s not that big of an issue.

Q: With such a long half-life, do physicians have to do anything different in terms of dosing and titrating degludec?

Dr. Heise: I don’t think so. For degludec, all you have to do is wait three days instead of two before making dose changes. I don’t think the strategy will be very different from what we have typically used for insulins.

Q: The reality is that we have a growing diabetes epidemic and increasing healthcare costs. How do we justify paying more for a product that carries only a small additional benefit compared to older and cheaper drugs?

Dr. Fisher. We are working within constricted healthcare systems. If degludec is better than other insulins, maybe focus first on patients who need it most, such as people who are struggling with recurring hypoglycemia. We will have to be selective at first, but there will be patients that will benefit.

Dr. Mathieu: I would agree. On the other hand, if patients lived the same day every day then I we could easily use NPH. But our patients live lives like we do every day. So you can’t just say it’s a small added value because it’s an important added value. Additionally, as you said, nocturnal hypoglycemia is extremely important to address. It impacts the behavior and health of people enormously. It has a high impact on healthcare costs. It has a high impact on the number of registered emergency room visits.

Dr. Zinman: We need to do analyses that look at the long term costs and short term costs associated with the drug. We need to look at CV outcomes and the effects on real world hypoglycemia rates. Lets look at all that carefully first.


Corporate Symposium: Diabetes and Cancer (Sponsored by Sanofi)


Larry Deeb, MD (Florida State University, Tallahassee, FL)

Discussing a number of recent studies, Dr. Deeb highlighted the benefits of basal-plus and basal-bolus therapy for people with type 2 diabetes in A1c lowering and achieving A1c targets. In a proof-of-concept study, insulin glargine/insulin glulisine basal-plus therapy (basal plus prandial insulin at one meal) further improved glycemic control beyond basal insulin glargine therapy alone (Owens et al., DOM 2011). In another study, intensification of insulin glargine treatment with insulin glulisine allowed more patients to achieve an A1c target of less than 7.0% (Davidson et al., Endoc Pract 2011). Dr. Deeb also emphasized the benefits of insulin glargine over NPH, and insulin glulisine over insulin aspart and insulin lispro. Insulin glargine has been shown to cause less hypoglycemia than NPH (Mullins et al., Clin Ther 2007), while glulisine has been demonstrated to have a more rapid onset of action than insulin lispro in non-diabetic volunteers (Heise et al., DOM 2007), and a more rapid onset of action than insulin aspart in people with type 2 diabetes (Bolli et al., DOM 2011).

Questions and Answers

Q: In your experience, is there a better time for glargine – AM versus PM? As a follow-up, how often do you split it into two doses? It is something that has caught popularity, but is it really necessary?

A: Most of the time, I tend to use it in the evening but you certainly can use it in the morning. It’s not totally up to patient preference, because for some you do tend to see that by 24 hours there is a waning effect of the glargine. I’ll split the dose for patients who need large doses, and want to go lower in the night but need insulin.

Q: In general, what are your thoughts on pre-meal versus post-meal bolusing for rapid- acting insulin?

A: I’m sorry to say that I published the first paper where we gave insulin to children after the meal. That was a mistake. Is that an answer?

-- by Adam Brown, Eric Chang, Benjamin Kozak, Lisa Rotenstein, Joseph Shivers, Sanjay Trehan,

Vincent Wu, and John and Kelly Close