European Association for the Study of Diabetes (EASD) Full Report

September 14-19, 2014: Vienna, Austria – Full Report – Oral Diabetes Drugs

Executive Highlights

Regarding oral diabetes drugs, EASD provided cause for interest around new developments in the DPP-4 inhibitor class. We saw the first phase 3 results for Merck’s once-weekly DPP-4 inhibitor omarigliptin, which demonstrated a comparable safety and efficacy profile to the once-daily Januvia (sitagliptin); this clinical comparability should allow omarigliptin’s once-weekly convenience to win out, at least for some patients, as long as reimbursement emerges. Merck intends to file omarigliptin by the end of the year in Japan, with other geographies to follow afterwards. Perhaps an even more exciting application for DPP-4 inhibitors is in their combination with SGLT-2 inhibitors. Dr. Ralph DeFronzo (University of Texas Health Science Center, San Antonio, TX) presented new expanded phase 3 data on Lilly/BI’s empagliflozin/linagliptin fixed-dose combination – results confirmed the finding at 24 weeks of superior efficacy with the combination compared to either Jardiance (BI/Lilly’s SGLT-2 empagliflozin) or Tradjenta (BI/Lilly’s linagliptin) alone. Dr. Julio Rosenstock also presented 24-week data on AstraZeneca’s saxagliptin/dapagliflozin combination (previously presented as a poster at ADA); he emphasized the potential of this class as a second-line option after metformin rather than solely as a means to intensify therapy in patients on multiple oral medications. 

Below we outline the key themes that emerged from the conference, followed by a table of contents and our detailed reporting. Titles highlighted in blue are new additions that were not mentioned in our daily updates from Vienna, and those highlighted in yellow represent what we felt were the most notable talks of the meeting. 

Table of Contents 

Oral Diabetes Drugs

Oral Presentations: SGLT-2 Inhibitors – New Outcome Studies

Fixed Dose Combinations of Empagliflozin/Linagliptin for 52 Weeks as Add-On to Metformin in Subjects With Type 2 Diabetes

Ralph DeFronzo, MD (University of Texas Health Science Center, San Antonio, TX)

Although there were no real long-term outcomes studies presented in this block (as the title suggested), the session did feature some of the conference’s most exciting oral presentations, specifically those on SGLT-2 inhibitor/DPP-4 inhibitor fixed-dose combinations (FDCs). Dr. Ralph DeFronzo presented the full 52-week results of one of two pivotal phase 3 trials on Lilly/BI’s empagliflozin/linagliptin (“empa/lina”) – 24-week results were presented as a poster at ADA. The full year results were consistent with the half-year results: both doses of empa/lina achieved significantly greater A1c reductions from baseline (8.0%) than their component monotherapy (see table below). Also remarkable was how durable the reduction in A1c was for patients on the combination or empagliflozin, while the linagliptin group saw a slight upward creep during the second half of the study. A subgroup analysis found that empa/lina’s efficacy was close to additive in patients with lower baseline A1c (<8.5%), whereas its efficacy was largely driven by the empagliflozin component in patients with higher baseline A1c. As expected, there did seem to be a slight increase in genital mycotic infections with empa/lina, although the effect was not clearly dose-dependent and the incidence remained below 10% for all study groups. Dr. DeFronzo concluded by suggesting that in the future, all three therapies included in this study (empagliflozin, linagliptin, and metformin) could be combined into a triple-therapy tablet. We look forward to hearing more on formulation capabilities.  

Table: Efficacy results at 52 weeks

Study group

Number of patients

Change in A1c from baseline

Empagliflozin 25 mg / linagliptin 5 mg



Empagliflozin 10 mg / linagliptin 5 mg



Empagliflozin 25 mg



Empagliflozin 10 mg



Linagliptin 5 mg



  • The 52-week study randomized 677 type 2 diabetes patients to five dose groups (there were two doses of empagliflozin tested as both monotherapy and in combination with linagliptin). The majority of patients were male and white, although there was a fairly high representation of Asian patients (~15%). Nearly a quarter of patients had diabetes duration greater than 10 years at baseline, while around 10% were within one year of diagnosis. The remainder had diabetes duration between one and ten years. Mean A1c was 8%, while mean BMI was around 31 kg/m2.
  • Over the 52-week study, both empa/lina dose combinations achieved significantly greater A1c reductions than their component monotherapy doses (see table above). A time-plot showed that the A1c reductions in the empagliflozin and combination groups were stable throughout the 52-week study period, whereas the A1c reduction from baseline seen with linagliptin was attenuated from -0.70% at 24 weeks to -0.48% at 52 weeks. Dr. DeFronzo noted that the efficacy seen in the primary efficacy results with the combination was not fully additive, although they were meaningfully and statistically greater than the monotherapy efficacy values.
    • Dr. DeFronzo displayed an analysis of the 52-week efficacy by A1c status at baseline (>8.5% and <8.5%). In the low-baseline-A1c subgroup (comprising around 75% of the study patient population), empa/lina’s A1c reduction over 52 weeks was actually slightly greater than additive (-0.97% with empagliflozin 25 mg/linagliptin 5 mg, -0.44% with empagliflozin 25 mg, and -0.40% with linagliptin 5 mg). By comparison, in the high-baseline-A1c subgroup, the combination’s efficacy was not additive. Dr. DeFronzo suggested during Q&A that the difference might be due to the fact that empagliflozin’s efficacy is proportional to a patient’s level of hyperglycemia. This means that SGLT-2 inhibitors’ efficacy is likely more elastic (i.e.: they are particularly effective at achieving larger A1c reductions in patients with high baseline A1c), where as DPP-4 inhibitors’ efficacy is limited by limits in the body’s endogenous production of GLP-1.
    • Also during Q&A, Dr. DeFronzo warned against making too much of the paradoxical increase in glucagon secretion seen in his group’s study on AZ’s Forxiga (dapagliflozin) because there was also a decrease in insulin seen in patients on the drug. Dr. DeFronzo suggested that the reduction in insulin could be at least as important as the increase in glucagon.
  • Improvements in blood pressure (~3 mmHg SBP) and body weight (~3 lbs) were in line with expectations, and driven nearly entirely by the empagliflozin component of the combination.
  • There were no new or worrying imbalances in adverse events, including serious adverse events. Confirmed hypoglycemia was rare (1-4%) across groups, with most events occurring in patients on underlying SFU therapy, and there were no events requiring assistance.

Questions and Answers

Q: You showed that efficacy was additive in patients with low A1c at baseline, but not in patients with high A1c at baseline. Could this be related to glucagon secretion, which could be higher in patients with high A1c and less pronounced in those with low A1c?

A: That’s a possibility, although I do not have glucagon data to share with you today. Although everyone has jumped on this glucagon story, if you read our original paper, insulin levels fell inappropriately with SGLT-2 inhibitor use as well. I would be careful when ascribing the entirety of the paradoxical rise in hepatic glucose production to glucagon – I think the decrease in insulin is playing at least as much of a role. It may not just be an effect on the alpha cell, but one that is mediated by the central nervous system. Another obvious explanation is that these drugs lower the threshold for glucose excretion, and that the higher your glucose at baseline, the higher your drop in glucose and A1c will be. That’s probably the more likely explanation.

Q: Is there an explanation for the higher dropout rate in the linagliptin group?

A: I don’t think that the dropout rate was all that much higher in that group – the side effect profile with all of these drugs is quite benign, and most of the withdrawals were not attributed to significant increases in adverse reactions.

Randomized, Double-Blind Trial of Dual Add-On Saxagliptin Plus Dapagliflozin vs. Saxagliptin or Dapagliflozin Add-On Alone in Poorly Controlled Type 2 Diabetes on Metformin

Julio Rosenstock, MD (Dallas Diabetes and Endocrine Center, Dallas, TX)

Dr. Julio Rosenstock presented the 24-week results from a phase 3 study on AZ’s saxagliptin/dapagliflozin fixed-dose combination (FDC) – results from this study were first presented as a poster at this year’s ADA. The mean A1c reduction from baseline seen with the combination was quite striking (1.5% vs. 1.2% with dapagliflozin alone and 0.9% with saxagliptin alone), although baseline A1c was on the high side (8.9%).  

Questions and Answers

Q: I have the impression that the additive effect here was less compared to empa/lina, despite the fact that you started with a much higher A1c level.

A: You raise a very interesting point, but you know very well that you cannot make comparisons between two different studies, which looked at two different populations. It would an interesting study to do a head-to-head comparison, but I doubt the companies will do such a study.

Empagliflozin Compared With Glimepiride as Add-On to Metformin for 2 Years in Patients with Type 2 Diabetes

Martin Ridderstråle, MD, PhD (Steno Diabetes Center, Gentofte, Denmark)

Dr. Martin Ridderstråle presented results, also presented at ADA, from the two-year EMPA-REG H2H-SU trial, which compared Lilly/BI’s Jardiance (empagliflozin 25 mg) to the sulfonylurea (SFU) glimepiride (1-4 mg), both as an add-on to metformin. A1c reduction was modestly but significantly greater with Jardiance than glimepiride (-0.66% vs. -0.55% from a baseline of 7.92%; p=0.026), and Jardiance also came out ahead with regard to body weight (3.1 kg loss vs. 1.3 kg gain) and hypoglycemia (2.5% vs. 24.2% of patients experiencing at least one event). Treatment with Jardiance also led to improvements in blood pressure compared to glimepiride, but it was associated with a slight increase in cholesterol and the incidence of genital infections. Dr. Ridderstråle also presented graphs of the change in A1c over the course of the study; while glimepiride led to a sharper initial drop, the average A1c began to rise after ~12 weeks, whereas the reduction achieved with Jardiance proved to be more durable. The study’s efficacy conclusions have been questioned to some extent because only 40% of patients in the glimepiride arm were titrated up to the maximum dose of 4 mg, but during Q&A, Dr. Ridderstråle said that the high risk of hypoglycemia with glimepiride was responsible for the conservative dosing and pointed out that the mean dose of 2.7 mg is in line with the typical dose given in clinical practice. 

Questions and Answers

Q: Why weren’t patients in the glimepiride group up-titrated to the maximum dose of 4 mg?

A: 40% of patients were able to up-titrate. The titration scheme was based on need, and the barrier was hypoglycemia. If you look at the mean dose of 2.7 mg, that corresponds to large surveys of what dose people are actually on in clinical practice.

Q: Are you concerned about the amount of genital infections and chronic inflammation with SGLT-2 inhibitors, or the risk of genitourinary cancers?

A: When infections do happen, patients respond well to treatment, and there has been no signal for any sort of cancer so far. Of course, we don’t have long-term data to rule it completely out. But even though these drugs are on the market, they are closely surveyed and any of those instances should be reported.

Q: Do you have any data about the GFR difference at the end of the study?

A: I don’t have the exact numbers in my head, but there was an improvement of eGFR during the study. It was minor but significant.

Energy Balance Following SGLT-2 Inhibition

Giulia Ferrannini, MD (University of Modena & Reggio Emilia, City, Modena, Italy)

Dr. Giulia Ferrannini presented her team’s work on why weight loss in those treated with SGLT-2 inhibitors is lower than what is predicted by glycosuria, developing the hypothesis of a compensatory increase in energy intake. To quantify SGLT-2 inhibitors’ effects on body weight, fat mass, and fat-free mass, Dr. Ferrannini observed these measures in type 2 patients on empagliflozin (n=86), half of who were drug-naïve and the other half who were on metformin. By predicting body weight using data from a mathematical model simulator (based on changes in calories and weight), she presented how the observed weight loss feel short of what would be expected given the increase in urinary glucose excretion alone. The results suggest that energy intake increased as a compensatory mechanism, suggesting that a substantial increase in energy intake is part of the compensatory response to glycosuria. Notably, this increased energy intake was more significant in those with lower BMIs at baseline. Those previously on metformin also experienced greater weight loss (-4.5 kg vs. -2 kg, p <0.01) compared to the drug naïve group, indicating that the compensation may be partially offset by previous metformin treatment.

Questions and Answers

Q: Regarding the fact that there was not as much weight lost in the trial, could you attribute that to the glucagon increase of SGLT-2 inhibitors?

A: That’s a point that should be taken into account – we did not investigate glucagon in our patients. The compensatory mechanisms that are at play are complex.

Q: Did you measure energy consumption at the level of muscles?

A: The simulator took this into account and the simulator is validated. The fat-free weight loss is mainly water, but the simulator still takes it into account.

Q: As a suggestion for compensatory eating, do you have any idea of what it is? Is it carbs, fats, etc.?

A: Carbs have not been proven. We did not ask questions regarding dietary habits – the patients were free to eat whatever they wanted to. But we did see that the difference was consistent and so the idea popped into our minds and so looking at diet is something we’ll look into, if possible.

Long-Term Efficacy and Safety of Canagliflozin in Older Patients with Type 2 Diabetes Mellitus Over 104 Weeks

Kaj Stenlöf, MD, PhD (Sahlgrenska University Hospital, Gothenburg, Sweden)

Dr. Kaj Stenlöf presented results, previously presented at ADA, from a 78-week extension of a 26-week primary study intended to evaluate the efficacy and safety of J&J’s Invokana (canagliflozin; 100 mg and 300 mg doses) in older patients with type 2 diabetes. Over the course of the study, both groups treated with Invokana showed significant placebo-adjusted reductions in A1c (-0.49% and -0.60% with the 100 mg and 300 mg doses, respectively; baseline A1c was quite low at 7.7%), fasting plasma glucose, body weight, and blood pressure. Consistent with other studies, Invokana was associated with a slight increase in LDL and HDL cholesterol and a higher incidence of genital and urinary tract infections, though the overall incidence of adverse events was similar across all groups. The incidence of hypoglycemia and adverse events related to osmotic diuresis was also higher in the Invokana groups; elderly patients are at particularly high risk for these two classes of adverse events. However, Dr. Stenlöf cautioned that those findings were confounded by the fact many patients were also treated with agents like sulfonylureas or insulin that are known to be associated with hypoglycemia. Reassuringly from a renal safety perspective, eGFR levels remained stable throughout the study and were similar across all groups.

Questions and Answers

Q: Because SGLT-2 inhibitors cause glycosuria, do these agents cause polyuria and nocturia, and if so, is the effect more pronounced in older patients?

A: Yes, they do cause osmotic diuresis. That effect should be considered and patients should be informed about that possibility, but it was not related to any patients dropping out of this study.

Q: Why do SGLT-2 inhibitors raise LDL cholesterol? Shouldn’t we be worried about that?

A: The short answer is we don’t know. It’s reasonable to consider that it has to do with a change in diet, as was seen in previous presentations, but it needs to be clarified in specific studies. Of course this needs to be monitored. We have outcomes studies going on, which will give the final answer on the clinical importance of that effect.

Q: Do the lipid changes correlate with the changes in metabolic parameters?

A: We see no correlation between weight or metabolic changes and lipids.

Q: Can you comment on the magnitude of the systolic blood pressure drop compared to other SGLT-2 data? It seems to be greater in your study.

A: The size of the change was slightly higher than in other studies. That mostly relates to the fact that we saw an increase in the placebo group; the absolute decrease in both systolic and diastolic blood pressure was quite small.

Q: Can you explain the rise in A1c at the end of the study? Was it related to duration of diabetes?

A: We had an older population with a long duration of diabetes on multiple treatments, which probably explains the A1c values we saw and the increase.

Safety of Dapagliflozin in Patients with Type 2 Diabetes Mellitus and Hypertension Inadequately Controlled by a Renin-Angiotensin System Blocker with/without a Second Agent

Traci Mansfield, PhD (Bristol-Meyers Squibb, Princeton, NJ)

Dr. Traci Mansfield presented favorable data regarding the safety profile of AZ’s SGLT-2 inhibitor Forxiga (dapagliflozin) from two 12-week phase 3 placebo-controlled studies (n=613; n=449) of people with type 2 diabetes and inadequately controlled hypertension. Study 1 (n=613) included participants on either ACE inhibitors or ARBs and Study 2 (n=449) included participants on ACE inhibitors/ARBs along with a second antihypertensive drug. In both studies, participants were randomized to either placebo or dapagliflozin 10 mg/day. There were no significant differences in the proportion of participants who experienced one or more adverse events between the placebo and dapagliflozin groups (35% vs. 37%, respectively in Study 1; 42% vs. 44% in Study 2). Neither study had a significant number of volume-related adverse events, and there were no serious adverse events related to renal function. Dapagliflozin was shown to lower serum uric acid and notably did not lead to any changes in potassium and sodium (which are usually adversely affected by diuretics). In conclusion, these studies demonstrated that dapagliflozin was well tolerated in type 2 patients who also have hypertension and are on antihypertensive agents.

  • Both studies’ results showed that the dapagliflozin group demonstrated meaningful reductions in A1c as well as in systolic blood pressure. Compared to the placebo group, the dapagliflozin group had a 0.5% A1c reduction and a systolic blood pressure reduction of 3 mm Hg. At baseline, participants had a mean A1c of ~8%, a seated blood pressure of ~150/91 mm Hg, and a weight of 87 kg (192 lbs).

Oral Presentations: Novel Compounds on the Horizon

Effect of Omarigliptin, A Novel Once-Weekly DPP-4 Inhibitor, in Japanese Patients with Type 2 Diabetes: A Placebo- and Sitagliptin- Controlled Trial

Ira Gantz, MD (Merck Sharp & Dohme Corp, Whitehouse Station, NJ)

Dr. Ira Gantz presented the first phase 3 results on Merck’s omarigliptin, showing that the once-weekly DPP-4 inhibitor had a comparable safety and efficacy profile compared to the once-daily Januvia (sitagliptin). The 24-week, non-inferiority trial was conducted in Japanese patients (n=414) with type 2 diabetes and compared the safety, efficacy, and tolerability of omarigliptin 25 mg with both placebo and sitagliptin 50 mg. Omarigliptin met the 0.3% A1c margin for non-inferiority, with an efficacy profile comparable to that of sitagliptin – both achieved placebo-adjusted A1c reductions of ~0.8%. Both agents also significantly reduced two-hour post-meal and fasting plasma glucose levels relative to placebo and showed no significant difference in safety profile. Omarigliptin was weight neutral, similar to sitagliptin. Comparable clinical profiles allow the different administration frequencies to stand out as the primary differentiator between the products, and we imagine that once-weekly convenience will be valuable for many (though perhaps not all) patients – valuable meaning that this would result in higher adherence and ultimately better outcomes. As a reminder, Merck intends to file for approval by the end of 2014 in Japan, with other geographies to follow.

  • Treatment with omarigliptin resulted in significant reduction in postprandial glucose compared to placebo. The difference in LS means between omarigliptin vs. placebo was -36 mg/dl (p <0.001) and the difference between sitagliptin vs. placebo was -40 mg/dl (p <0.001). Omarigliptin and sitagliptin achieved very similar reductions with the difference in LS means between them being 3 mg/dl (p =0.555).
  • Similarly, treatment with omarigliptin resulted in a significant reduction in fasting plasma glucose compared to placebo, with a difference in LS means of -12 mg/dl (p <0.001). The difference for sitagliptin vs. placebo was -15 mg/dl (p <0.001) and the difference for omarigliptin vs. sitagliptin was 2 mg/dl (p=0.330).
  • A higher proportion of participants on omarigliptin achieved an A1c <7% and <6.5% compared with both sitagliptin and placebo. Almost half (47%) of those on omarigliptin achieved an A1c <7%, while 38% of those on sitagliptin and 7.3% of those on placebo did. The difference between omarigliptin and sitagliptin was less significant for the target of A1c <6.5%: 10.2% of the omarigliptin group achieved this, compared to 8.5% of the sitagliptin group and 1.2% of the placebo group.
  • There were no meaningful differences in the incidences of any adverse events compared to placebo and sitagliptin. The percentage of participants who experienced one or more adverse events in the omarigliptin, sitagliptin, and placebo groups were respectively 50%, 49%, and 66%. There were no deaths or serious drug-related adverse events and no reports of pancreatitis in any of the treatment groups.

Questions and Answers

Q: There’s a discrepancy between the fasting glucose values in the placebo group. The A1c values in the placebo group are going up during the 24-week period. The fasting plasma glucose values are actually reduced in the placebo group. Can you explain this discrepancy?

A:  We noticed that also and we really don’t have an explanation from the data.

Q: What is the half-life of the drug?

A: In the alpha phase, approximately 40 to 50 hours, which is most of the area under the curve as far as plasma exposure. It’s longer for the beta phase, approximately 120 hours.

Q: Do you have any data on drug adherence and compliance or any other data on quality of life?

A: For this specific study, there was no way to assess adherence because it’s a double dummy design – everyone is taking daily and weekly medications. From the literature, there is wide variation in those studies – anywhere from 30% to 80%. We didn’t do any quality of life measures in this study.

Q: If you look on a more granular level, is there any variation in the impact of the drug from day one to day seven?

A: We didn’t look on a daily level. These values were drawn at troth. But we know that the degree of DPP-4 inhibition is above a level that, depending on what assay you use, there shouldn’t be much variation. There’s a little bit of a dropout on that.

Q: Do you have any ongoing studies comparing omarigliptin in a higher dose of 15 mg or so with sitagliptin in the 100 mg dose?

A: Yes, we are presently conducting a study and we’re looking forward to seeing these results early next year. It’s looking at a 25 mg dose of omarigliptin against 100 mg dose of sitagliptin. That’s a global study and it’s posted on

Q: Did this not translate to more clinical efficacy?

A: No, it was very minor. We looked at that and did some modeling and simulation. It pretty much came to a plateau – it doesn’t really make sense to push it to get those couple more percentage points.

Q: Is kidney function an issue with the long-acting compound?

A: No, this is excreted through the kidney unchanged. We do have dose reductions for those subjects with severe renal insufficiencies and who are on dialysis. But this is based on exposures from clinical studies. As a principle, we don’t want to go over two-fold of the exposure that a normal person would have. So there’s a dose reduction and we have a renal study.

Oral Presentations: GLP-1 Analogs – Non-Glycemic Endpoints

Effect of Saxagliptin on Renal Outcome

Ofri Mosenzon, MD (Hadassah Medical Center, Jerusalem, Israel)

Dr. Ofri Mosenzon presented data from the SAVOR-TIMI cardiovascular outcomes trial (CVOT) demonstrating beneficial renal effects following treatment with AstraZeneca’s DPP-4 inhibitor Onglyza (saxagliptin). The trial enrolled 16,492 patients with type 2 diabetes, who were studied for a median of 2.1 years. Results showed that patients treated with Onglyza were more likely to experience improvement in their categorical albumin/creatinine ratio (ACR) than those treated with placebo (11% vs. 9%; p<0.01) and were less likely to see worsening in their categorical ACR (13% vs. 16%, p<0.01). The numerical difference was on the modest size, but the large size of the SAVOR trial allowed statistical significance to emerge. The ACR benefit held true for all degrees of baseline renal function (normal, microalbuminuria, and proteinuria) and ACR ranges (<30 mg/g, 30-300 mg/g, and <300 mg/g). Patients with normal renal function or mild renal impairment (eGFR >50 mL/min/1.73 m2; n = 13,916) experienced a mean reduction of 20 mg/g in ACR, patients with moderate renal dysfunction (eGFR 30-50 mL/min/1.73 m2; n = 2,240) saw a mean improvement of 100 mg/g, and those with severe renal impairment (eGFR <30 mL/min/1.73 m2; n = 336) experienced a 250 mg/g reduction, which Dr. Mosenzon characterized as having enormous clinical importance despite its borderline statistical significance. During Q&A, Dr. Mosenzon theorized that possible anti-inflammatory effects could explain the benefit seen in the trial, although he believes it is too early to speculate about mechanisms without first having more data from clinical trials.

  • Onglyza’s effect on albuminuria was independent of its effect on A1c, and no significant cardiovascular safety concerns emerged in any of the groups. Dr. Mosenzon said that longer-term studies will be required to elucidate any effects on eGFR and that more investigation on the basic science side will be necessary to understand the mechanism behind these effects.

Questions and Answers

Q: How many measurements of albuminuria were there? Was the conclusion of progression or regression reached after considering two out of three measurements or was it based on just one?

A: It was based on one measurement every year. It was measured at baseline, at one year, at two years, and at the end of the trial. It was measured only once per test.

Q: Could you speculate about a potential mechanism explaining why a DPP-4 inhibitor would improve renal function independent of glucose control?

A: That’s a great question. I’m a clinical researcher, and I think I should leave that to basic research to find out. There’s lots of literature talking about a possible anti-inflammatory effect, but that’s a whole other discussion. First we have to face facts that emerge from clinical research and then look for an explanation.

Saxagliptin in Patients With Prior Heart Failure: Observations From SAVOR-TIMI 53

Itamar Raz, MD (Hadassah University Hospital, Jerusalem, Israel)

SAVOR lead investigator Dr. Itamar Raz provided an overview of the primary results from the trial (which investigated cardiovascular outcomes with AZ’s DPP-4 inhibitor Onglyza [saxagliptin]) as well as secondary analyses relating to heart failure. Two main takeaways were: (i) in patients with a prior history of heart failure, saxagliptin did not alter the risk of MACE; and (ii) the increase in heart failure risk seen with saxagliptin was largely limited to the first six months of treatment. Generally, the discussion on heart failure with DPP-4 inhibitors has become a waiting game, as the results from Merck’s TECOS (for Januvia [sitagliptin]) are expected to arrive next year (2015).

Questions and Answers

Q: Do you recommend the use of saxagliptin in patients with heart failure?

A: You should know with nearly any other drug class, you don’t know much about the safety profile with regards to heart failure. We do know that TZDs are not recommended, and we don’t really know about insulin and sulfonylureas. You’re very limited in terms of what you can use. I think that what we learned is that with saxagliptin in heart failure patients, the benefit is worth the risk. However, when you use saxagliptin to treat patients with heart failure you should monitor them more carefully.

Q: Did you try and correlated hypoglycemia with hospitalization for heart failure?

A: We looked very carefully at whether hypoglycemia could explain the hospitalization for heart failure finding in SAVOR, but didn’t find any correlation. This is a very solitary finding.

Q: We give DPP-4 inhibitors to a large number of patients in India, especially in elderly patients with cardiovascular problems because the drug is perceived as being very safe. Is this drug really safe to start in that patient population?

A: We demonstrated in SAVOR that saxagliptin is not inferior to placebo with regard to the primary and secondary endpoints. In patients at high risk for heart failure, you should watch them carefully. SAVOR was conducted in patients with long disease duration. 

Posters: Clinical Studies with DPP-4 Inhibitors

Oral Glucose Lowering with Linagliptin Plus Metformin is a Viable Initial Treatment Strategy in Patients with Newly DIagnosied Type 2 Diabetes and Marked Hyperglycemia (Poster 894)

Baptist Gallwitz, Stuart A. Ross, A. Enrique Caballero, Stefano Del Prato, Diane Lewis-D’Agostino, Zelie Bailes, Sandra Thiemann, Sanjay Patel, Hans-Juergen Woerle, and Maximilian von Eynatten

Dr. Baptist Gallwitz and colleagues presented the subgroup analysis of a 24-week randomized double-blind trial that compared linagliptin plus metformin vs. linagliptin alone in recently-diagnosed type 2 patients with marked hyperglycemia. The study randomized 316 participants to linagliptin 5 mg daily plus placebo (n=157) or to linagliptin 5 mg daily plus metformin twice daily (n=159, metformin was uptitrated in the first 6 weeks to a maximal dose of 2,000 mg/day). Baseline characteristics were comparable between the two groups, with a mean age of 49 years, A1c of 9.8-9.9%, fasting plasma glucose of 196-198 mg/dl, and BMI of 30 kg/m2. The PPCC analysis (described below) showed that in the overall population, linagliptin plus metformin provided greater improvements in A1c (-2.8% from a baseline A1c of 9.73%) vs. linagliptin alone (-2.0% from a baseline A1c of 9.70%; p<0.0001). A similar trend was also observed in the subgroup analysis (further details below). In looking at the overall population, the investigators observed that a greater percentage of patients on the combination therapy achieved a target A1c <7.0% at 24 weeks vs. those on linagliptin plus placebo (61% vs. 40%, respectively; p=0.0008). This result was also found for the composite endpoint of A1c <7.0% with no hypoglycemia and no weight gain at 24 weeks (48% vs. 26%, respectively; p=0.0004). The authors conclude that in newly diagnosed type 2 patients with hyperglycemia, initial linagliptin plus metformin therapy provides consistent A1c reductions across different subgroups.

  • Dr. Gallwitz and colleagues note that while insulin therapy is often initiated in type 2 patients with marked hyperglycemia, its use poses several challenges, including an increased risk of hypoglycemia. The authors highlight that linagliptin and metformin have complementary mechanisms of action and a low risk of hypoglycemia, and posit whether a linagliptin/metformin combination therapy would be a beneficial initial therapy in this patient population.
  • The primary endpoint (A1c reduction at 24 weeks) was analyzed using the per-protocol completers cohort method: all randomized patients who received ³1 dose of study drug and had a baseline A1c measurement along with at least one on-treatment measurement were included, as well as patients who did not have important protocol violations, patients who completed 24 weeks of treatments without receiving glycemic rescue therapy, and patients who had an A1c measurement at week 24. The analysis also included patients with available data – i.e., observed cases.
  • For patients with an A1c <9.5%, the combination therapy provided greater reductions in A1c (-2.1% from a baseline of 8.73%) vs. linagliptin alone (-1.4% from a baseline of 8.76%; 95% CI: -1.23 to -0.15). Participants with an A1c ³9.5% also experienced a larger A1c reduction with the combination therapy (-3.4% from a baseline A1c of 10.46%) vs. linagliptin alone (-2.5% from a baseline A1c of 10.49%; 95% CI: -1.32 to -0.35).  Similarly, the combination therapy provided greater A1c improvements in all subgroups of age, BMI, renal function, race, and ethnicity (with the exception of Hispanic/Latino patients, in which the two treatment groups provided an equal A1c reduction of -3%).
  • The linagliptin plus metformin group and the linagliptin alone group had similar rates of adverse events (AEs; 56% vs. 61%, respectively), as well as comparable rates of drug-related AEs (9% vs. 6%, respectively), and serious AEs (2% vs. 1%, respectively). Both groups also had similar rates of hypoglycemia (2% vs. 3%, respectively), and no patient in either group experienced severe hypoglycemia. The authors note that the long-term safety of linagliptin on a background of metformin is currently being investigated in the CAROLINA Trial ( Identifier: NCT01243424).

Symposium: Contemporary T2DM Management – Focus on Safety and Efficacy (Sponsored by the Metabolic Endocrine Education Foundation, Worldwide Diabetes, and PESI Inc.)

The Role of GLP-1 Analogs, DPP-4 Inhibitors, and TZDs in the Management of T2DM

Ralph DeFronzo, MD (University of Texas Health Science Center, San Antonio, TX)

Out of the three drug classes that Dr. Ralph DeFronzo was tasked with discussing, he had some clear favorites. He firmly stated that GLP-1 agonists and TZDs (in addition to SGLT-2 inhibitors) are the best agents we have for type 2 diabetes, while he downplayed DPP-4 inhibitors’ efficacy as weak. He spent a great deal of time highlighting early evidence on GLP-1 agonists’ positive effect on beta cell function, remarking that no other drug class has a similar effect. In the latter half of his talk, he highlighted the TZD class’ very durable efficacy. Citing the results from the full PROactive trial as well as an eight-year Kaiser Permanente database study, Dr. DeFronzo categorically stated that Takeda’s Actos (pioglitazone) does not cause bladder cancer.

  • The points Dr. DeFronzo made in this presentation are in line with those we have heard at other meetings over the previous year – see our coverage of three of Dr. DeFronzo’s talks at CODHy Latin America earlier this year.

Case-Based Recommendations & Panel Discussion

Yehuda Handelsman, MD (Metabolic Institute of America, Tarzana, CA)

Using a series of case studies, Dr. Yehuda Handelsman moderated a broad panel discussion on therapy recommendations for type 2 diabetes. There was general enthusiasm for SGLT-2 inhibitors and agreement that lifestyle change recommendations on their own were not particularly effective. Below we include a selection of some of the most quotable quotes:

  • Dr. Vivian Fonseca (Tulane University Health Sciences Center, New Orleans, LA): “While lifestyle makes sense, when we look at studies, metformin did better than lifestyle. Lifestyle did make the numbers better, but it did not make the outcomes better.”
  • Dr. Robert Henry (University of California San Diego, San Diego, CA): “I think, SGLT-2 inhibitors – once we show they’re safe, which we will – can be used more. They certainly are getting an impressive record.”
  • Dr. Ralph DeFronzo (University of Texas Health Science Center, San Antonio, TX): “The longer you wait, the more beta-cell function you lose. If you keep picking drugs that don’t affect beta-cell function, you are eventually going to get to the point where drugs don’t work and you require insulin. At that point, I’m all for using insulin but I would not like to get there. I would use pioglitazone and GLP-1. And it’s not unreasonable to add a SGLT-2 inhibitor. Also, in my clinical experience, SGLT-2 inhibitors work in the 60-65 GFR range.”
  • Dr. Itamar Raz (Hadassah Medical Center, Jerusalem, Israel): “My only concern with pioglitazone is that in patients who you cannot follow up with for two to three months, you may have a huge weight gain. These drugs are very effective, but they need good follow-up in patients.”

Symposium: 50th Annual Meeting of EASD – Where We Came From and Where We Go

The Pathogenesis of Type 2 Diabetes: 50 Years of Cogitating and Digesting

Ele Ferrannini, MD, PhD (University of Pisa School of Medicine, Pisa, Italy)

Dr. Ele Ferrannini called for earlier and more aggressive treatment of type 2 diabetes – as early as when patients present with prediabetes. Dr. Ferrannini stressed that insulin resistance and beta cell failure manifest themselves early in the progression of the disease. In his view, beta cell dysfunction is underappreciated and concealed by the physiological increase in insulin production that offsets growing insulin resistance during prediabetes. However, given that beta cell dysfunction and insulin resistance are both “at least partly reversible,” he advocated for acute (i.e., early and aggressive) treatment of diabetes at the first sign of hyperglycemia. This was a powerful perspective especially considering the sizeable prevalence of prediabetes (86 million patients in the US) and silent nature of the disease (90% of individuals with prediabetes do not know they have it). There are presently no drug classes approved to treat prediabetes in the US or EU, but we hope that greater advocacy from well-renowned figures such as Dr. Ferrannini can raise the level of conversation about this urgent and unmet public health need.

Symposium: Rising Star Symposium

A New Approach for Personalised Medicine in Type 2 Diabetes: Integrating Multiple Effects of a Single Drug

Hiddo Lambers Heerspink, PharmD, PhD (University of Groningen, Groningen, Netherlands)

Dr. Hiddo Heerspink’s talk on personalized medicine in type 2 diabetes encouraged providers to consider more integrative approaches to assessing the benefit and risks of treatment paradigms. Current clinical practices, in Dr. Heerspink’s view, tend to target a single drug to a single risk factor, falsely assuming that improvements to the risk factor necessarily translate into improved cardiovascular (CV) or renal outcomes. However, drugs have multiple effects on various risk factors that are not necessarily consistent with a single outcome profile. To underscore this point, Dr. Heerspink showed a slide of recently published data in type 2 diabetes patients indicating that increases in A1c were discordant with reductions in blood pressure (Smink et al., Clinical Pharmacology and Therapeutics 2014). The main takeaway was that individual biomarkers are not insufficient to predict outcomes. Dr. Heerspink called for greater appreciation of this point, emphasizing the use of algorithms that integrate a network of disease and drug interactions will more successfully predict outcomes in the future. In his view, that “future” is nearly here already – he shared unpublished data from his lab demonstrating that feedback from metabolites can accurately predict urine albumin and creatinine responses to spironolactone therapy. This predictive capability, in his view, has the power to transform care, allowing us to individually understand the effect of treatments and more appropriately affect personalized regimens.

Questions and Answers

Q: What do you think about pleiotropic effects? Will these have any impact on how we should design and interpret randomized control trials?

A: Yes, I think so. There are examples of factors that have decreased some biomarkers but have not reduced cardiovascular risk. People were confused. Perhaps the answer is that there are off-targets effects that offset the effect of decreased albuminuria and blood pressure reductions. I think the integral of all these effects determines cardiovascular risk and predicts outcomes.

Michael Berger Debate: Metformin – Where is the Evidence?

The Evidence for Metformin is Overwhelming

Harold Lebovitz (State University of New York, Brooklyn, NY)

Dr. Harold Lebovitz presented compelling evidence in favor of metformin as an appropriate first-line therapy for type 2 diabetes. Drawing on evidence of multiple studies of the agent, his argument hinged on a number of key points: i) that metformin is effective and durable in reducing A1c; ii) that it is weight-loss inducing; iii) that there is little risk for significant hypoglycemia or other adverse events; iv) that the agent may decrease the incidence and progression of some cancers; and last v) that metformin reduces macrovascular complications when compared against placebo or sulfonylureas. We did not find this argument surprising, especially given the ubiquitous use of metformin as first-line therapy for type 2 diabetes (even some type 1s are now using it off-label!). However, Dr. Lebovitz did call for long-term studies of metformin in order fully assess the cardiovascular risks of the agent – he stressed that the limited profit associated with marketing a generic, such as metformin, has disincentivized industry from conducting the necessary randomized control trials that would inform clinical decisions. Dr. Lebovitz criticized this structure, calling for government-industry collaboration that prioritizes the needs of the diabetes community. We appreciate this call to action and, from a patient perspective, echo the need to align incentives between patients, industry, government, and – we would add – payers.

  • Dr. Lebovitz characterized metformin-induced reductions in A1c as “potent” and “very effective.” He highlighted longstanding evidence that metformin monotherapy (n=143) in patients poorly controlled on diet led to a reduction in A1c of 1.4% relative to placebo (n=146) (baseline in both groups: ~8.3%) (DeFronzo et al., New England Journal of Medicine 1995).
  • Expanding on the efficacy of metformin, Dr. Lebovitz called the agent “reasonably durable.” He noted that TZDs induce longer-lasting effects that are attributable to their beta-cell preservation properties. As a reminder, we heard Dr. Lebovitz defend this drug class at IDF 2013 as a highly useful therapeutic option, especially at lower doses that may shield patients from the non-cardiovascular side effects. That said, he was clearly positive on metformin, sharing no concerns regarding durability.
  • Dr. Lebovtiz characterized metformin as a weight-loss inducing agent. He shared results from the DPPOS, highlighting that the change in weight (-2 kg or 4.4 pounds) was preserved throughout the duration of the 10-year follow-up. Dr. Lebovitz cited this weight loss as the key mechanism responsible for the potentially cardioprotective effects of the drug. However, he emphasized that there has yet to be any good evidence that weight loss necessarily reduces cardiovascular risks.
  • Metformin is associated with little risk for significant hypoglycemia or other adverse eventsthe former is an especially notable safety concern, said Dr. Lebovitz, considering that severe hypoglycemia is a significant predictor of cardiovascular complications. He cited the findings of ADVANCE, noting that even a single episode of severe hypoglycemia was significantly associated with an increased risk of major micro- and macrovascular events, death, cardiovascular disease, and non-cardiovascular diseases.
    • Though the jury is still out regarding metformin’s affect on cancer risk, Dr. Lebovitz highlighted that the drug may decrease the incidence and progression of the disease.
  • Drawing from multiple studies, Dr. Lebovitz concluded that metformin reduces macrovascular complications when compared against placebo or sulfonylureas. He asserted that what is known regarding the cardiovascular risks associated with metformin comes from limited long-term studies of the agent (e.g., UKPDS, HOME studies). This problem arises, in his view, because of the limited profit associated with marketing a generic, such that industry is not incentivized to conduct randomized control trials. Though these studies would be particularly instructive for providers, Dr. Lebovitz is skeptical that these cardiovascular questions will ever be answered by clinical data.
  • Metformin is not a perfect first-line option; however, our main takeaway from Dr. Lebovitz’s lecture was that the agent is the best of many imperfect options. Dr. Lebovitz listed clinical relevant side effects for TZDs (bone fracture, bladder caner), sulfonylureas (hypoglycemia), alpha-glucosidase inhibitors (gastrointestinal concerns), DPP-4 inhibitors (increase in hospitalizations for patients with outstanding chronic heart failure), GLP-1 agonists (nausea, vomiting, acute pancreatitis), and SGLT-2 inhibitors (mycotic infections).
    • Dr. Lebovitz’s message of caution regarding DPP-4 inhibitors surprised us. This drug class is often touted for its robust safety profile with limited side effects. His assessment that we presently lack long-term follow-up data is valid, though based on our current knowledge of the class, we have been more encouraged by the findings.

The Evidence for Metformin is Unclear

Rury Holman, MD (University of Oxford, Oxford, UK)

Dr. Rury Holman followed by discussing the shortcomings of the evidence base for metformin. Although he acknowledged that metformin has extensive clinical experience and is correctly recommended as the foundation pharmacotherapy, Dr. Holman emphasized that we do not fully understand the drug’s benefits and risks, even fifty years after its introduction. For example, he referred to confusion regarding the magnitude of lactic acidosis risk as well as the lack of randomized controlled outcomes trials (since UKPDS in 1998) for cardiovascular risk. To resolve the current uncertainty, Dr. Holman introduced GLINT, Glucose Lowering in Non-Diabetic Hyperglycemia Trial (previously discussed at Diabetes UK 2013), of which Dr. Holman is a joint-chair. The double-blind comparison trial will take six years and will study metformin vs. placebo in individuals (n=11,834) over 40 years of age who are at increased risk of type 2 diabetes and cardiovascular disease (with a 10-year risk ≥20%). The primary endpoint will be time to first cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke while the secondary endpoints will include incident cancer and new-onset diabetes. Dr. Holman explained that the study is in pilot phase for an NIH-funded trial, with plans to commence in the fourth quarter of this year and with results expected in 2022. We are excited to see concrete plans for this trial as we heard Dr. Lars Ryden (Karolinska Institute, Stockholm, Sweden) discuss similar concerns at ESC on the need for outcomes trials for older drug classes.

Questions and Answers

Dr. Lebovitz: I think Professor Holman’s presentation is absolutely right on but I’m more skeptical than he is. Why is it that we’ve had metformin for so long and not had another trial? Because nobody can make money marketing metformin. One of our major problems in clinical diabetes is that very critical studies, for which companies will not make a profit, are unlikely to get done. And that’s really a shame. [Applause] We need to encourage both government and industry to work together to really try and answer questions that are very important for diabetes, but may not make someone rich. That’s my first comment. My second comment is that there’s also a big problem with long-term clinical trials. If a trial comes out next year which proves that metformin decreases the risk of cancer and decreases cancer mortality, then it is unethical to conduct or continue a study that looks at whether or not you’re going to get that endpoint. And that really is a big problem in a long-term ten-year study. It’s not against the individuals designing the study; it just points to the practical issues in the US that if there’s a proven benefit, the FDA will make you put everyone on the drug for which there is a benefit. Evidence-based medicine is important and we should all work on it but there are some questions that will unfortunately not be answered by evidence-based medicine. And we’ll have to take our best clinical data and best basic science data and put them together to make really reasonable judgments.

Q: I was surprised to see that none of you mentioned the second UKPDS metformin trial, which actually showed a significant increase in all costs and mortality. This is usual care. And also in that context, they’ve done DPP studies on phenformin, which also showed significantly increased cardiovascular mortality vs. placebo. Could you maybe comment a little bit on that?

Dr. Holman: The DPP was a very well-run trial but underpowered and stopped as a result. One of the reasons UKPDS was done was that there was a disbelief in Europe that all biguanides were not safe. Also, phenformin is a greater risk factor for lactic acidosis. Coming to the second trial, there was a post-hoc analysis of a subgroup in which metformin was added on top for patients who were previously on sulfonylureas. At the time, there were some publications that showed us a statistically doubling of risk for cardiovascular disease and myocardial infarction. That explained the opportunity to rebut that case, where we showed ten-year follow-up data and the separation that occurred in this post-hoc analysis had disappeared and that data will be published shortly. That’s the problem with small numbers because the analysis was not pre-planned. We were unable to claim that there was an unreal effect. Many analyses have been done on observational data trying to rebut it. For me now, the fact that metformin is added to almost every drug and is currently licensed to do so means that by and large, people have not accepted this. You know, there are combinations with sulfonylureas, SGLT-2 inhibitors, and others. People have read that data and have evaluated that this is not a risk but it would still be great to do this study.

Dr. Lebovitz: I think that phenformin is really a different drug from metformin. Phenformin is very lipid soluble and it stays in the body for a long period of time. Unless you’ve really done the comparative types of pharmacology, it’s very hard to say that these drugs are the same. Some mechanisms are the same but some make them very different. The compartments of the cell they get into are very different. I was not involved in the studies but I was using phenformin when it was available in the US. The question is that we saw a lot of side effects. That’s why it took 20 years to get metformin to the US. When we look at the combination of metformin with sulfonylureas, we now know from studies like ADVANCE, ACCORD, and ORIGIN, that you must take into consideration whether you get significant hypoglycemia. Obviously, metformin alone doesn’t cause hypoglycemia but if you lower glucose with metformin and you have sulfonylureas, you are going to see some cases of severe hypoglycemia. So I’m not so sure that the original data reflects some of that. It may have disappeared as people got wiser in using combination therapies. I agree with Rory that in order to make anything out of that, we would have to do a true randomized controlled trial.

Q: What do you think about metformin in pregnancy?

Dr. Lebovitz: Well, we are seeing metformin being used. In the US, we quote data from South Africa, where metformin has been used in pregnant women for a long time, with no evidence of abnoramlites in fetus. This is not what I do but I will tell you that the OBGYN people who are treating infertility with metformin show that if you give metformin, you can reestablish ovulation. If you stop metformin, there’s a high incidence of abortion. From what I gather from talking to people, metformin is widely used in pregnancies. It’s interesting that we have published studies on sulfonylureas, which I think is more dangerous in pregnancy than metformin. There are several papers supporting the use of glimepiride but there are some that say that they are not so safe. So to answer your question, I think metformin is used and we have no evidence that it is detrimental at the moment.

Q: I treat patients on metformin who have said that they have gastrointestinal side effects. Do we need more formulations to deal with that?

Dr. Holman: Titration helps offset that. If we acknowledge benefits, any dose is better than none. Once-daily preparations appear to be a little better tolerated. It’s true that there are people who are just metformin-intolerant. I don’t know much about formulations, but I’d be interested.

Dr. Lebovitz: This has to do with the absorption by the OCT transporters. I think GI side effects may have to do with how rapidly metformin gets into the body. And that’s genetically determined. There’s a San Diego company that presented modified metformin and presented it at ADA some years ago. It was much better tolerated. And it was a chemical difference, not just longer acting. So there are companies working on this. There are also companies working on AMP kinases.

Q: Didn’t the metformin arm in UKPDS not show any particular microvascular endpoints? Any comments on this?

Dr. Holman: The effect size for microvascular disease was identical but because of the smaller number of patients, there was not enough statistical power for significance. So we don’t say there was any microvascular benefit.

Dr. Lebovitz: Just analyzing insulin data alone and not combining with the sulfonylurea data, you will not prove any microvascular benefit.

Dr. Holman: Yeah, it’s a number issue. For metformin, the effect size was the same and you have to take it or leave it whether that’s a clinical effect or an underpowered study.

Q: Regarding the practice of dose reductions or stopping metformin at a GFR of 60 or 45, these patients most commonly end up on insulin. Is lactic acidosis related to tissue hypoxia? Are these guidelines overcautious?

Dr. Holman: I think these are safe cut-off points because at the end of the day, patients can have rapidly progressing renal disease that could drop to much lower GFRs. As you say, there are other treatments. I think it’s wise to have sensible cut-offs. I can tell you that despite these cut-offs in the US, there is a substantial minority in the US that prescribes cut-offs differently than we do.

Dr. Lebovitz: If you look at the number of patients with lactic acidosis in the US, it’s very rare. I’ve seen patients on dialysis who have gotten metformin and have not developed lactic acidosis. I think that you can get down to much lower levels than even in the UK. The point to be made is to remember that when people have severe complications, there is a question about how aggressive you want to treat glucose. Do you really benefit people with renal disease by lowering their A1c below 8 or 7.5? These are the questions that we’re dealing with now. If you look at people with congestive heart failure, there are studies now that show that if you lower their A1c below 7.5%, they have a higher mortality. So the goals need to be modulated by the clinical condition of the patient.

Q: I’m very interested in obesity medicines. I think they are important for part of metformin’s mechanism of action with regards to reducing mortality. I think in GLINT, there is a good opportunity to see whether there is an interaction between metformin and visceral fat.

Dr. Holman: Currently, we’re in the pilot phase. When we move to the full trial, we are very much hoping to put in more complex markers like that. It’s always nice to have a mechanistic reason why.

Q: My question is on metformin and pregnancy. How many months of gestation should they stay on metformin? And what dose?

Dr. Holman: There’s a consensus but no evidence that it’s used safely in pregnancy. It’s certainly used in the UK and elsewhere. It is used at conventional doses to achieve levels to deliver a normal fetus in a normal pregnancy. That’s the standard rule, but it’s not licensed for that purpose. If you don’t have an alternative, it’s a better choice than nothing. As we don’t have trials from this, it’s tough to have guidelines.

Q: What are your thoughts on a maximum dose of 3 g? Are there benefits at another dose or any other limitations?

Dr. Holman: The original UK data showed benefits up to 3 g per day but by and large, we now cap it off at 2.5. Americans cap it off at 2 g. This is limited by the side effects.

Dr. Lebovitz: It really depends on the absorption patterns of the individual patient and there are genetic differences. In the US, we don’t go beyond 2 g. I’m not sure if there’s good evidence that you get any better benefits. In India, I see most of the patients are on 1,000 mg per day and it’s rare that they go over 1,500 mg and they seem to do very well. Maybe we ought to dose it on body weight in addition to other factors.

Q: What is the effect of metformin on the thyroid profile? Are there concerns over hypothyroidism or over TSH? Should we increase the dose or decrease the dose?

Dr. Lebovitz: I’m not aware of any significant differences. There’s data on B12 absorption, but I’m not aware of any for the thyroid.

Dr. Holman: I’m also not aware of any clinically relevant data on this. 

Corporate Symposium: Modern Type 2 Diabetes Management – The Experts’ Guide to the Universe of Choices (Sponsored by AstraZeneca)

How Safe Are the Novel Therapies?

Juris Meier, MD (St. Josef Hospital, Bochum, Germany)

Dr. Juris Meier discussed the safety profiles of newer drug classes, namely GLP-1 receptor agonists, DPP-4 inhibitors, and SGLT-2 inhibitors. Though the majority of safety data look positive, Dr. Meier urged caution in interpreting these results – his underlying message was conservative, stressing that future trials will be instrumental in informing risk-benefit profiles. With respect to DPP-4 inhibitors, Dr. Meier noted that pooled analyses of phase 2/3 trials suggest no increased incidence of cardiovascular risk with this drug class. He did bring up the heart failure signal uncovered in the SAVOR trial, noting that this limited data does not inform our understanding a great deal. Dr. Meier addressed similar concerns regarding the moderate increase in heart rate and increased risk of pancreatitis that have been associated with the GLP-1 agonist class. With respect to pancreatitis specifically, he shared data from a meta-analysis of multiple GLP-1 agonist phase 3 trials that indicate no increased incidence of acute pancreatitis with incretin treatment (Meier and Nauck, Diabetologia 2014). Dr. Meier was cautiously optimistic regarding the cardioprotective potential of SGLT-2 inhibitors. Drawing on multiple studies, he highlighted the significantly reduced rate of hospitalization for heart failure in patients treated with an SGLT-2 inhibitor, but again qualified this conclusion by calling for future studies that expand upon these preliminary results.

  • Pooled analyses of phase 2/3 trials of DPP-4 inhibitors suggests no increased incidence of cardiovascular risk with this drug class, although these trials were not long enough to provide robust CV safety findings. Studies have been relatively short in duration (26-52 weeks). Dr. Meier discussed evidence of an increased risk in heart failure in the SAVOR trial, which he does not believe is clinically significant just yet, though more study is needed on the issue.
  • Dr. Meier discussed the concern regarding pancreatitis/pancreatic cancer associated with GLP-1 agonists. Given the rarity of this form of adverse event, he stressed that the majority of studies to date lack the power to inform our understanding of this relationship. However, he presented findings from a meta-analysis of adverse event data collected from a number of GLP-1 agonist phase III studies – results indicate no increased incidence of acute pancreatitis with incretin treatment (Meier and Nauck, Diabetologia 2014).
    • Despite this evidence, Dr. Meier was not ready to write off this risk just yet. He cited evidence, presented at ADA 2014, that lipase and amylase levels increase in response to treatment with lixisenatide (20 μg), liraglutide (1.2 mg), and liraglutide (1.8 mg). In his view, the fact that we cannot explain the cause of this relationship is cause for some concern.
  • Of the new drug classes, SGLT-2 inhibitors are most likely to show cardioprotection, in Dr. Meier’s view, although the increase in LDL is a legitimate concern. Dr. Meier shared pooled data from Phase 2b/3 trials of dapagliflozin that suggest no increase – perhaps even a decrease – in major cardiovascular events vs. control. Similar studies actually indicate a lower risk of heart-failure-related hospitalization with the agent. Nevertheless, Dr. Meier was unconvinced by these preliminary results and called for future studies that expand upon these findings.

Questions and Answers

Q: The audience [based on a poll] reported that cardiovascular disease is their major area of concern with respect to the safety of novel therapies. What are your thoughts on that?

A: I think it reflects the data we are discussing. I agree that cardiovascular safety should be first on their minds. We’ve also had an unpleasant experience in the past with rosiglitazone. We might want to broaden cardiovascular disease to include heart failure and heart rate, but I agree that this is the primary thing we should focus on. In terms of pancreatic cancer and pancreatitis, I think I am less worried than I was a few years ago. There doesn’t seem to be a vast increase in pancreatitis with GLP-1 agonists. I would still be interested to know why amylase and lipase levels increase.

We can also ask: what do FDA and EMA say about new therapies? Apparently, they do not have big concerns about cardiovascular disease and heart failure. That said, there has been a label about for saxagliptin that says patients who have a prior heart failure have an elevated risk of an adverse event. However, the label does not discourage use. I would suggest to use these drugs with caution in these patients.

I also think that SGLT-2 inhibitors might be promising in the future, and we need to see more data looking at that. In patients with advanced heart failure, insulin will be the best treatment choice. Regarding lipase, there’s a question of whether we should measure it or not. I would not measure it specifically. If you measure lipase in diabetes, then in 20% of cases, you will have elevated levels. And, if it is elevated, then you need to follow up on it with multiple examinations. My recommendation is only to measure lipase levels if patient reports clinical symptoms, such as abdominal pain, etc.

Is There a Role for Early Combination Therapy in the Management of Patients with Type 2 Diabetes?

Bernard Zinman, MD (Mount Sinai Hospital, Toronto, Canada)

The highly respected Dr. Bernard Zinman called on providers to develop a new paradigm for the treatment of type 2 diabetes, explaining that an aggressive approach involving early initiation of combination therapy (already the dominant paradigm for the treatment of other diseases like HIV/AIDS) is likely to be far more effective than the current “treat to failure model” in addressing such a complex, progressive disease. He cited clinical inertia as a major obstacle to more effective therapy for type 2 diabetes and urged the audience to accept the reality that the vast majority of patients will not achieve long-term success with metformin alone. Dr. Zinman reviewed the pros and cons of several options for combination therapy, suggesting that fixed dose combinations of metformin and newer drug classes (namely GLP-1 agonists, DPP-4 inhibitors, and SGLT-2 inhibitors) hold great promise due to their robust A1c-lowering efficacy, low risk of hypoglycemia and weight gain, and fairly reasonable cost. He strongly criticized the risk-benefit profile of sulfonylureas, saying that “they have three advantages: they’re cheap, they’re cheap, and they’re cheap,” and that it is no longer necessary to accept weight gain and hypoglycemia as inevitable consequences of aggressively treating type 2 diabetes.

  • Several polls suggested that the packed audience was very receptive to Dr. Zinman’s message: 83% of the packed audience said that they already used fixed dose combinations to treat patients with type 2 diabetes, and 84% selected either metformin + SGLT-2 inhibitor + DPP-4 inhibitor combinations or metformin + SGLT-2 inhibitor + GLP-1 agonist combinations when asked to predict which type 2 diabetes drug combinations would be most commonly prescribed ten years from now.

Questions and Answers

Dr. Chantal Mathieu (University of Leuven, Leuven, Belgium): Let’s be honest, the more pills we make patients take, the more strain we put on them. Is there an effect on adherence with FDCs?

Dr. Bernard Zinman (Mount Sinai Hospital, Toronto, Canada): It’s been clearly demonstrated in the context of therapeutics that pills taken once a day have much better adherence. That’s the advantage of FDCs; patients don’t care how many chemicals are in the pill, they just want to take one pill. If I prescribe multiple pills, it means there’s a problem, and that it’s serious. Cost is also a factor, and often the metformin component is free. (Editor’s note – in the US, of course, one co-pay vs. multiple co-pays will also affect many patients.)

Dr. Mathieu: How does SGLT-2 inhibitor and incretin therapy impact glucagon?

Dr. Zinman: As Dr. Drucker discussed earlier, glucagon biology is abnormal starting early in type 2 diabetes, so anything we can do to decrease it is helpful. Dapagliflozin/metformin combinations led to an increase in glucagon, likely a compensatory response to prevent hypoglycemia. But when you add saxagliptin to dapagliflozin, you get a nice complementary mechanism where saxagliptin suppresses the increased glucagon, and that translates into better glycemic reduction.

Dr. Mathieu: Why do you think FDCs are valuable?

Dr. Zinman: They improve adherence, they tend to be cheaper, and there are fewer side effects with lower doses. There’s a real barrier among diabetologists, who are very much fixated on the step-wise intensification approach.

Dr. Mathieu: What do the data say about combinations with insulin and novel therapies?

Dr. Zinman: That’s another area that’s exciting because of robust responses. With SGLT-2 inhibitors and insulin, you see dramatic changes in three areas: A1c, insulin dose, and weight loss. SGLT-2 inhibitors and insulin in type 1 patients has been studied; it’s not on-label right now, but there’s tremendous interest in pursuing it.

Dr. Jiten Vora (University of Liverpool, Liverpool, UK): [Directed at the audience] Why do you use FDCs?

Audience member: We use them because of better adherence, better glycemic control, and because they target different mechanisms.

Dr. Vora: When do you start?

Audience member: Early in the disease.

Audience member: We use DPP-4 inhibitor/metformin combinations because that’s what we have in Portugal. When we use them depends on patient. With some patients who have a high A1c or fasting glucose, we use them with very good results. Sometimes we use them early; it depends on patient.

Dr. Zinman: It’s obvious when someone presents with an A1c of 10% that you feel compelled to use combinations. I’m taking it one step further, saying it doesn’t matter what the A1c is; if you have type 2 diabetes, the beta cells will fail and it will progress.

2nd Line Therapy, When Metformin is No Longer Enough, What to Use and When: Introduction

John Buse, MD, PhD (University of North Carolina Chapel Hill, Chapel Hill, NC)

Dr. John Buse introduced the debate on second-line therapy by first providing background on the current state of the field. He discussed the ADA/EASD position statement, highlighting particular patient priorities such as avoiding hypoglycemia or weight gain. After providing clinical overviews of sulfonylureas and pioglitazone, Dr. Buse set the stage for the debate by pointing out that DPP-4 inhibitors, GLP-1 agonists, and SGLT-2 inhibitors can facilitate the recent ADA/EASD recommendation to individualize glycemic targets and therapies. 

2nd Line Therapy, When Metformin is No Longer Enough, What to Use and When: DPP-4 Inhibitors

Juris Meier, MD (Ruhr-Universität Bochum, Bochum, Germany)

Arguing in favor of DPP-4 inhibitors, Dr. Juris Meier began by highlighting the class’ long-term efficacy, showing maintained A1c reductions of saxagliptin with metformin over 104 weeks. In addition, Dr. Meier noted the low risk of hypoglycemia associated with long-term use of DPP-4 inhibitors. Regarding weight, DPP-4 inhibitors had a more favorable weight profile compared to sulfonylureas and pioglitazone – Dr. Meier admitted that while DPP-4 inhibitors do not lead to significant weight reduction, its weight neutrality gives it a substantial advantage over other drug classes. He emphasized that the class has proven cardiovascular safety, referring to data from SAVOR and EXAMINE that saxagliptin and alogliptin did not increase rates of cardiovascular events. Similarly, Dr. Meier pointed out the drug class’s established tolerability profile with limited adverse events. Dr. Meier added that the fixed-dose combinations available with the class also offer unique adherence benefits.

Questions and Answers

Q: The tolerability of DPP-4 inhibitors is very good. But this class seems somewhat weaker with glycemic lowering compared to other therapies.

A: These are early treatments and I think this efficacy is perfectly sufficient for early therapy. If it’s not sufficient, then we can move onto other drugs but if we’re using drugs early on, the efficacy of DPP-4 inhibitors is adequate for most cases.

2nd Line Therapy, When Metformin is No Longer Enough, What to Use and When: GLP-1 Receptor Agonists

Tina Vilsbøll, MD (Gentofte Hospital, Copenhagen, Denmark)

Dr. Tina Vilsbøll followed by arguing for GLP-1 receptor agonists as a stronger second-line therapy. Similar to Dr. Meier’s presentation on DPP-4 inhibitors, Dr. Vilsbøll highlighted GLP-1 agonists’ long-term efficacy in glycemic lowering (presenting data for up to six years of treatment) as well as the class’s low hypoglycemia rates over sustained treatment. Although she admitted that there is no long-term data regarding cardiovascular safety, she argued that much of the current data show no safety signals. In addition, she called the side effects “manageable,” especially with slow uptitration.

Questions and Answers

Q: Can you talk about the high cost of GLP-1 agonists?

A: We all know that treating type 2 diabetes is very expensive. I think you have so many beneficial effects with GLP-1 agonist that for me, the price is at an acceptable level. But of course, we hope that costs will go down.

Q: What do you think about GLP-1 agonists for obesity?

A: I think that’s great, but it’s not a cure for diabetes. The future treatment would have to be combined with exercise.

2nd Line Therapy, When Metformin is No Longer Enough, What to Use and When: SGLT-2 Inhibitors

Stephan Matthaei, MD (Quakenbrück Hospital, Quakenbrück, Germany)

To round out the mini-debate, Dr. Stephan Matthaei took on the case for SGLT-2 inhibitors as the optimal second-line therapy. While he acknowledged that the field is still waiting on long-term cardiovascular data from the DECLARE study, Dr. Matthaei said that the drug class seems to have a favorable effect on cardiovascular events thus far. Regarding side effects, he stated that genital infections are easy to treat, unlikely to reoccur, and occur at low frequencies (1 in 100). Concluding, Dr. Matthaei said he looks forward to the additional experience the field will have with these compounds, highlighting that one or more members of the class are now approved in many countries and have had “impressive uptake.”

Questions and Answers

Q: Regarding urogenital infections, do you think providers are less comfortable with evaluating these side effects?

A: General hygiene actually plays a role in this. In Japan, which is very hygienic, this is not as big of an issue.

Q: How about renal function monitoring? Is that a barrier?

A: You should not use these if the GFR is below 45 or 60. But it hasn’t been a barrier in my experience. The effect seems relatively stable, with no further decline. There are no safety concerns really, just a need for more education. 

Do Novel Therapies Have a Role in Type 1 Diabetes?

Thomas Pieber, MD (Medical University of Graz, Graz, Austria)

After celebrating the immense impact that intensive insulin therapy has had on the lives of patients with type 1 diabetes, Dr. Thomas Pieber suggested that the use of newer type 2 diabetes drug classes, namely GLP-1 agonists and SGLT-2 inhibitors, could help address the significant unmet needs that remain for type 1 diabetes management. While insulin is a lifesaving drug that has been proven capable of reducing the risk of long-term microvascular complications, its well-documented side effects of hypoglycemia and weight gain can lead to poor quality of life, reduced adherence to treatment, and potentially serious health consequences. Dr. Pieber pointed to several ways that adding a GLP-1 agonist or an SGLT-2 inhibitors as an adjunct to insulin therapy could help mitigate these issues: both classes have beneficial effects on body weight, and both could potentially help blunt postprandial glucose excursions and allow patients to reduce their daily insulin dose (due to decreased glucagon production and delayed gastric emptying in the case of GLP-1 agonists, and as a result of increased glucose excretion with SGLT-2 inhibitors).

  • Dr. Pieber cited several preliminary studies showing promising results with GLP-1 agonists and SGLT-2 inhibitors in patients with type 1 diabetes. In a 2011 study by Kielgast et al., treatment with liraglutide (Novo Nordisk’s Victoza) for four weeks led to improved glycemic control and a reduced insulin dose in 29 type 1 diabetes patients; in a 2014 study by Sarkar et al., six months of therapy with exenatide (AZ’s Byetta/Bydureon) led to improved insulin sensitivity and an average weight loss of four kilograms in 14 adults with type 1 diabetes. With regard to SGLT-2 inhibitors, Dr. Pieber cited an eight-week study involving 40 patients with type 1 diabetes in which treatment with empagliflozin (Lilly/BI’s Jardiance) reduced average A1c from 8.0% to 7.6% while decreasing the risk of hypoglycemia, and a two-week study presented at ADA 2013 in which treatment with dapagliflozin led to a reduction in daily insulin doses in type 1 diabetes patients. While the credibility of these results is limited by the trials’ short duration and small sample size, Dr. Pieber argued that they ought to provide an impetus for longer-term trials that can lead to more meaningful conclusions.
  • An audience poll preceding the presentation suggested that attendees would be receptive to incorporating type 2 diabetes drug classes into their treatment regimens for type 1 diabetes. A striking 56% of the audience said they would consider using SGLT-2 inhibitors as an adjunct to insulin for type 1 diabetes, 21% said the same for GLP-1 agonists, and only 5% said they would not consider adding any of the options listed (which also included older type 2 drug classes like TZDs).

Corporate Symposium: A Special Report on Type 2 Diabetes – Finding the Right Treatment Routes to Optimize Your Patients’ Journey (Sponsored by Lilly/BI)

A Special Report on Simplifying Combination Therapy for Patients: Fixed-Dose Combinations (FDCs)

Stuart Ross, MB, ChB (University of Calgary, Alberta, Canada)

Dr. Stuart Ross gave a brief but thorough talk on the potential of fixed-dose combinations (FDCs) to improve treatment outcomes. He set the stage by mentioning two key challenges that exist in type 2 diabetes care today: (i) Prescribing inertia means that patients are not treated aggressively enough, especially early in the course of the disease; and (ii) Increasingly complex treatment regimens pose challenges for adherence to therapy. Insulin is perhaps the most effective therapy available with regards to glucose lowering, but Dr. Ross acknowledged that physicians are reluctant to prescribe it early in the course of treatment. Moving on to the main subject of the presentation, Dr. Ross introduced the concept of FDCs, noting that they can directly address treatment inertia and poor adherence through their simplicity. Citing data on BI/Lilly’s empagliflozin/linagliptin and AZ’s saxagliptin/dapagliflozin, he suggested that the combination of new drug classes (in this case, SGLT-2 inhibitors and DPP-4 inhibitors) deliver strong results as well as a fairly benign safety and tolerability profile.

Panel Discussion

Dr. Merlin Thomas (University of Melbourne Melbourne, Australia): We know clearly that insulin-sensitizing agents have an effect on glycemic control, so there’s theoretical synergism with agents like DPP-4 inhibitors that increase insulin release – you would make yourself more sensitive to the insulin you released, which is good. Other effects are less useful, and there are clearly side effects that are quite real. The worst thing you can do to a patient with diabetes is send them to an orthopedic surgeon with a fracture. Those data make me concerned. But I know a man with diabetes who said he’d rather die than take insulin, so then they’re a reasonable alternative because insulin has its own drawbacks. I can only use TZDs as part of triple therapy.

Q: The ADA recommends glipizide for patients with chronic kidney disease.

Dr. Thomas: All agents have utility in that situation. The reason we use sulfonylureas in patients with CKD is that people are more concerned about metformin. Using appropriate doses of metformin can be relatively safe and effective. We use sulfonylureas in many patients with CKD, especially when metformin is contraindicated; they work, but the hypoglycemia risk is ten-fold higher. Why would you take that risk with your patients’ health when you have an alternative?

Dr. Stuart Ross (LMC Diabetes & Endocrinology, Calgary, Alberta, Canada): The guidelines came from the ADVANCE study, which looked at the use of sulfonylureas with heart disease and renal disease and saw fewer hypoglycemic events and less weight gain. It doesn’t rule out those risks, and we try to avoid many of these drugs, but that’s where it came from.

Q: How many patients with lactic acidosis have you seen in your practice?

Dr. Thomas: Three, and they all had extremely good reasons for it. One had sepsis, one had profound ischemic gout, and the other had severe heart failure. All of them stopped metformin and all of them did terribly – it wasn’t the metformin’s fault. I can imagine a state with profound reduction of lactic acid in which metformin could theoretically interfere with lactate clearance and increase exposure. I would sometimes rather deal with the devil I know than one I don’t. I’ve seen many more severe hypos than lactic acidosis.

Dr. Alexandra Kautzky-Willer (Medical University of Vienna, Vienna, Austria): I’ve only seen one, but the ER says they see them. They may be seeing patients that we don’t see, so there’s still a concern.

Q: Are you concerned about bladder cancer with SGLT-2 inhibitors and TZDs?

Dr. Ele Ferrannini (University of Pisa, Pisa, Italy): There was a recent statement from the FDA that cleared pioglitazone from suspicion. I don’t think any signal has emerged with SGLT-2 inhibitors, but when it comes to cancer, all you can do is wait and see. It’s complicated by the fact that cancer’s not one disease, and each one has a different set of risk factors. You monitor, you have surveillance plans for the use of drugs, and you see what happens.

Q: How many ongoing studies are testing empagliflozin vs. DPP-4 inhibitors as an add-on to metformin?

Dr. Ferrannini: There’s a study with a 2x2 design testing linagliptin, empagliflozin, and both, which is obvious, and it’s probably not the only one. Other studies are testing dapagliflozin and saxagliptin combined. The studies are being carried out.


-- by Melissa An, Adam Brown, Varun Iyengar, Emily Regier, Manu Venkat, and Kelly Close

-- The authors thank Eric Chang, Hannah Deming, Jessica Dong, Nina Ran, and Melissa Tjota for additional help on conference writing and editing