EASD 2011

LISBON – Obese individuals who have already managed to lose weight via diet and exercise can keep the pounds off by taking the antidiabetic drug liraglutide, even it they do not have diabetes, data from a yearlong study suggest.

Data from the 56-week, phase III SCALE (Satiety and Clinical Adiposity Liraglutide Evidence in Nondiabetic and Diabetic Subjects) study showed that body weight was decreased by 6.11% in the 212 liraglutide-treated patients but by just 0.05% in the 210 who were given placebo in association with a low-calorie diet, a highly significant difference.

Novo Nordisk’s liraglutide (Victoza) is a long-acting glucagonlike peptide–1 (GLP-1) analogue that has been developed to treat people with type 2 diabetes. As such, it is not licensed for use in obese individuals who do not have diabetes.

However, these new data suggest that it could potentially be used to help nondiabetic people who have lost weight but are having trouble maintaining their reduced weight.

Reporting the findings of the final 12-week "off treatment" follow-up phase at the European Association for the Study of Diabetes annual meeting, Dr. Vincent Woo of the University of Manitoba, Winnipeg, said that the patients who were included in the trial were basically "unsuccessful dieters."

The subjects had to have a body mass index of 30 kg/m² or greater, or a body mass index greater than 27 with comorbidities, and have already achieved a stable weight loss of 5% or more with a low-calorie (1,200-1,400 kcal/day) diet and exercise plan. A total of 422 individuals (mean age, 46 years) took part in the trial.

The objective was to assess if up to 3 mg/day of liraglutide could help maintain this weight loss, Dr. Woo said, and to see if there was any additional weight loss that could be achieved with the drug, compared with diet and exercise alone.

Three coprimary end points at week 56 were the mean percent change in body weight from randomization; the percentage of individuals who maintained their weight loss after the 4- to 12-week run-in period; and the percentage of individuals who lost an additional 5% or more of their body weight after randomization.

For the drug to be proved better than diet and exercise alone, all three coprimary end points had to be met, and they were.

"If we look at the mean change in weight from randomization to week 56, there was a significant loss of weight in the liraglutide group of 6.1%," compared with minimal changes for the placebo group, Dr. Woo said, adding that the liraglutide group lost a mean 5.7 kg, compared with a slight increase of 0.16 kg in the placebo group.

A significantly higher percentage of the liraglutide group (51% vs. 22% of those on placebo) were able to achieve a 5% additional weight loss at 56 weeks, with 26% and 6.3%, respectively, achieving an additional 10% weight loss.

The majority (81%) of individuals treated with liraglutide were able to maintain their run-in weight loss, compared with about half (46%) of placebo-treated subjects. Only 1.9% of liraglutide – but 18% of placebo-treated individuals – regained 5% body weight during the trial. Both differences were highly significant.

Dr. Woo noted that liraglutide was generally well tolerated with "no major safety issues," compared with placebo. Both groups reported a similar percentage of any adverse events (91.5% vs. 88.8%, respectively). Likewise, there was no difference in the percentage of patients experiencing serious adverse events or withdrawing from the study as a result.

Dr. Woo said that the gastrointestinal adverse effects were to be expected with the GLP-1 agonist. They included nausea (48% vs. 17% in the placebo group), constipation (27% vs. 12%), diarrhea (18% vs. 12%), and vomiting (17% vs. 2%). Only 2.4% of individuals on liraglutide withdrew because of nausea, however, which Dr. Woo said occurred in the first 4-5 weeks and was generally mild in 64% of patients.

Unlike other drugs that have been used to manage obesity, there was no indication that liraglutide increased the risk of psychiatric disorders.

"The obesity indication is always a tough one for regulatory authorities," commented Dr. Michaela Diamant, professor of diabetology and director of the diabetes center at the Free University Medical Center in Amsterdam, in an interview. She chaired the session in which the SCALE findings were reported, but was not involved in the study.

Although liraglutide may become a preferred agent for an obese, diabetic patient and for those in whom an additional weight maintenance benefit would add to improved glycemic control, its use in someone who is simply obese is likely to be more controversial, especially because the drug must be administered via a daily subcutaneous injection.

"This is just for nondiabetic, obese individuals," Dr. Diamant explained, noting that around 80% of study participants were women who perhaps wanted cosmetic results.

Regulatory agencies must be sure that drugs are safe, Dr. Diamant observed, but even so, "what is obesity? It’s a lifestyle [issue], it’s [about] prevention. ... So should we treat that massively, with expensive drugs?" The 3-mg daily dose of liraglutide could work out to be a very expensive way to treat what is essentially a lifestyle-led condition, she suggested.

The study was funded by Novo Nordisk. Dr. Woo disclosed receiving honoraria for speaking and participating in advisory boards or clinical trials on behalf of Novo Nordisk, Eli Lilly, Sanofi-Aventis, Merck, Bristol-Myers Squibb, AstraZeneca, Boehringer Ingelheim, Roche, Abbott, and the government of Canada. Dr. Diamant has acted as a consultant, speaker, or both for Eli Lilly, Merck, Novo Nordisk, and Sanofi-Aventis. She has also received research support from Amylin, Lilly, Merck, Novartis, Novo Nordisk, and Takeda.

LISBON – Obese individuals who have already managed to lose weight via diet and exercise can keep the pounds off by taking the antidiabetic drug liraglutide, even it they do not have diabetes, data from a yearlong study suggest.

Data from the 56-week, phase III SCALE (Satiety and Clinical Adiposity Liraglutide Evidence in Nondiabetic and Diabetic Subjects) study showed that body weight was decreased by 6.11% in the 212 liraglutide-treated patients but by just 0.05% in the 210 who were given placebo in association with a low-calorie diet, a highly significant difference.

Novo Nordisk’s liraglutide (Victoza) is a long-acting glucagonlike peptide–1 (GLP-1) analogue that has been developed to treat people with type 2 diabetes. As such, it is not licensed for use in obese individuals who do not have diabetes.

However, these new data suggest that it could potentially be used to help nondiabetic people who have lost weight but are having trouble maintaining their reduced weight.

Reporting the findings of the final 12-week "off treatment" follow-up phase at the European Association for the Study of Diabetes annual meeting, Dr. Vincent Woo of the University of Manitoba, Winnipeg, said that the patients who were included in the trial were basically "unsuccessful dieters."

The subjects had to have a body mass index of 30 kg/m² or greater, or a body mass index greater than 27 with comorbidities, and have already achieved a stable weight loss of 5% or more with a low-calorie (1,200-1,400 kcal/day) diet and exercise plan. A total of 422 individuals (mean age, 46 years) took part in the trial.

The objective was to assess if up to 3 mg/day of liraglutide could help maintain this weight loss, Dr. Woo said, and to see if there was any additional weight loss that could be achieved with the drug, compared with diet and exercise alone.

Three coprimary end points at week 56 were the mean percent change in body weight from randomization; the percentage of individuals who maintained their weight loss after the 4- to 12-week run-in period; and the percentage of individuals who lost an additional 5% or more of their body weight after randomization.

For the drug to be proved better than diet and exercise alone, all three coprimary end points had to be met, and they were.

"If we look at the mean change in weight from randomization to week 56, there was a significant loss of weight in the liraglutide group of 6.1%," compared with minimal changes for the placebo group, Dr. Woo said, adding that the liraglutide group lost a mean 5.7 kg, compared with a slight increase of 0.16 kg in the placebo group.

A significantly higher percentage of the liraglutide group (51% vs. 22% of those on placebo) were able to achieve a 5% additional weight loss at 56 weeks, with 26% and 6.3%, respectively, achieving an additional 10% weight loss.

The majority (81%) of individuals treated with liraglutide were able to maintain their run-in weight loss, compared with about half (46%) of placebo-treated subjects. Only 1.9% of liraglutide – but 18% of placebo-treated individuals – regained 5% body weight during the trial. Both differences were highly significant.

Dr. Woo noted that liraglutide was generally well tolerated with "no major safety issues," compared with placebo. Both groups reported a similar percentage of any adverse events (91.5% vs. 88.8%, respectively). Likewise, there was no difference in the percentage of patients experiencing serious adverse events or withdrawing from the study as a result.

Dr. Woo said that the gastrointestinal adverse effects were to be expected with the GLP-1 agonist. They included nausea (48% vs. 17% in the placebo group), constipation (27% vs. 12%), diarrhea (18% vs. 12%), and vomiting (17% vs. 2%). Only 2.4% of individuals on liraglutide withdrew because of nausea, however, which Dr. Woo said occurred in the first 4-5 weeks and was generally mild in 64% of patients.

Unlike other drugs that have been used to manage obesity, there was no indication that liraglutide increased the risk of psychiatric disorders.

"The obesity indication is always a tough one for regulatory authorities," commented Dr. Michaela Diamant, professor of diabetology and director of the diabetes center at the Free University Medical Center in Amsterdam, in an interview. She chaired the session in which the SCALE findings were reported, but was not involved in the study.

Although liraglutide may become a preferred agent for an obese, diabetic patient and for those in whom an additional weight maintenance benefit would add to improved glycemic control, its use in someone who is simply obese is likely to be more controversial, especially because the drug must be administered via a daily subcutaneous injection.

"This is just for nondiabetic, obese individuals," Dr. Diamant explained, noting that around 80% of study participants were women who perhaps wanted cosmetic results.

Regulatory agencies must be sure that drugs are safe, Dr. Diamant observed, but even so, "what is obesity? It’s a lifestyle [issue], it’s [about] prevention. ... So should we treat that massively, with expensive drugs?" The 3-mg daily dose of liraglutide could work out to be a very expensive way to treat what is essentially a lifestyle-led condition, she suggested.

The study was funded by Novo Nordisk. Dr. Woo disclosed receiving honoraria for speaking and participating in advisory boards or clinical trials on behalf of Novo Nordisk, Eli Lilly, Sanofi-Aventis, Merck, Bristol-Myers Squibb, AstraZeneca, Boehringer Ingelheim, Roche, Abbott, and the government of Canada. Dr. Diamant has acted as a consultant, speaker, or both for Eli Lilly, Merck, Novo Nordisk, and Sanofi-Aventis. She has also received research support from Amylin, Lilly, Merck, Novartis, Novo Nordisk, and Takeda.

LISBON – Obese individuals who have already managed to lose weight via diet and exercise can keep the pounds off by taking the antidiabetic drug liraglutide, even it they do not have diabetes, data from a yearlong study suggest.

Data from the 56-week, phase III SCALE (Satiety and Clinical Adiposity Liraglutide Evidence in Nondiabetic and Diabetic Subjects) study showed that body weight was decreased by 6.11% in the 212 liraglutide-treated patients but by just 0.05% in the 210 who were given placebo in association with a low-calorie diet, a highly significant difference.

Novo Nordisk’s liraglutide (Victoza) is a long-acting glucagonlike peptide–1 (GLP-1) analogue that has been developed to treat people with type 2 diabetes. As such, it is not licensed for use in obese individuals who do not have diabetes.

However, these new data suggest that it could potentially be used to help nondiabetic people who have lost weight but are having trouble maintaining their reduced weight.

Reporting the findings of the final 12-week "off treatment" follow-up phase at the European Association for the Study of Diabetes annual meeting, Dr. Vincent Woo of the University of Manitoba, Winnipeg, said that the patients who were included in the trial were basically "unsuccessful dieters."

The subjects had to have a body mass index of 30 kg/m² or greater, or a body mass index greater than 27 with comorbidities, and have already achieved a stable weight loss of 5% or more with a low-calorie (1,200-1,400 kcal/day) diet and exercise plan. A total of 422 individuals (mean age, 46 years) took part in the trial.

The objective was to assess if up to 3 mg/day of liraglutide could help maintain this weight loss, Dr. Woo said, and to see if there was any additional weight loss that could be achieved with the drug, compared with diet and exercise alone.

Three coprimary end points at week 56 were the mean percent change in body weight from randomization; the percentage of individuals who maintained their weight loss after the 4- to 12-week run-in period; and the percentage of individuals who lost an additional 5% or more of their body weight after randomization.

For the drug to be proved better than diet and exercise alone, all three coprimary end points had to be met, and they were.

"If we look at the mean change in weight from randomization to week 56, there was a significant loss of weight in the liraglutide group of 6.1%," compared with minimal changes for the placebo group, Dr. Woo said, adding that the liraglutide group lost a mean 5.7 kg, compared with a slight increase of 0.16 kg in the placebo group.

A significantly higher percentage of the liraglutide group (51% vs. 22% of those on placebo) were able to achieve a 5% additional weight loss at 56 weeks, with 26% and 6.3%, respectively, achieving an additional 10% weight loss.

The majority (81%) of individuals treated with liraglutide were able to maintain their run-in weight loss, compared with about half (46%) of placebo-treated subjects. Only 1.9% of liraglutide – but 18% of placebo-treated individuals – regained 5% body weight during the trial. Both differences were highly significant.

Dr. Woo noted that liraglutide was generally well tolerated with "no major safety issues," compared with placebo. Both groups reported a similar percentage of any adverse events (91.5% vs. 88.8%, respectively). Likewise, there was no difference in the percentage of patients experiencing serious adverse events or withdrawing from the study as a result.

Dr. Woo said that the gastrointestinal adverse effects were to be expected with the GLP-1 agonist. They included nausea (48% vs. 17% in the placebo group), constipation (27% vs. 12%), diarrhea (18% vs. 12%), and vomiting (17% vs. 2%). Only 2.4% of individuals on liraglutide withdrew because of nausea, however, which Dr. Woo said occurred in the first 4-5 weeks and was generally mild in 64% of patients.

Unlike other drugs that have been used to manage obesity, there was no indication that liraglutide increased the risk of psychiatric disorders.

"The obesity indication is always a tough one for regulatory authorities," commented Dr. Michaela Diamant, professor of diabetology and director of the diabetes center at the Free University Medical Center in Amsterdam, in an interview. She chaired the session in which the SCALE findings were reported, but was not involved in the study.

Although liraglutide may become a preferred agent for an obese, diabetic patient and for those in whom an additional weight maintenance benefit would add to improved glycemic control, its use in someone who is simply obese is likely to be more controversial, especially because the drug must be administered via a daily subcutaneous injection.

"This is just for nondiabetic, obese individuals," Dr. Diamant explained, noting that around 80% of study participants were women who perhaps wanted cosmetic results.

Regulatory agencies must be sure that drugs are safe, Dr. Diamant observed, but even so, "what is obesity? It’s a lifestyle [issue], it’s [about] prevention. ... So should we treat that massively, with expensive drugs?" The 3-mg daily dose of liraglutide could work out to be a very expensive way to treat what is essentially a lifestyle-led condition, she suggested.

The study was funded by Novo Nordisk. Dr. Woo disclosed receiving honoraria for speaking and participating in advisory boards or clinical trials on behalf of Novo Nordisk, Eli Lilly, Sanofi-Aventis, Merck, Bristol-Myers Squibb, AstraZeneca, Boehringer Ingelheim, Roche, Abbott, and the government of Canada. Dr. Diamant has acted as a consultant, speaker, or both for Eli Lilly, Merck, Novo Nordisk, and Sanofi-Aventis. She has also received research support from Amylin, Lilly, Merck, Novartis, Novo Nordisk, and Takeda.

LISBON – The possibility that incretin-based therapies for type 2 diabetes can cause cancer warrants further investigation, a leading clinical diabetologist has said.

"Pancreatitis and pancreatic cancer are reported in excess with exenatide and sitagliptin therapy in the FDA AERS [Food and Drug Administration’s Adverse Events Reporting System database] in comparison to other diabetes medications," Dr. Peter Butler said at the annual meeting of the European Association for the Study of Diabetes.

Using the database, Dr. Butler and his associates recently found that treatment with exenatide (Byetta) and sitagliptin (Januvia) significantly increased the risk of pancreatitis more than sixfold, with odds ratios (ORs) of 10.7 and 6.7, respectively, compared with four control drugs – rosiglitazone (Avandia), nateglinide (Starlix), repaglinide (Prandin), and glipizide (Glucotrol) – (P less than 10^-16 for both comparisons) (Gastroenterology 2011;141:150-6). These data are somewhat alarming, as many experts now believe that pancreatitis may be a precursor to malignancy.

Furthermore, the risk of pancreatic cancer was more than doubled, compared with control treatments, with statistically significant odds ratios of 2.9 (P = 9 x 10^-5) for exenatide and 2.7 for sitagliptin (P = .008). Thyroid cancer risk was also significantly elevated by exenatide, with an odds ratio of 4.7, (P = 4 x 10^-16), but not by sitagliptin (OR = 1.5, P = .65).

The risk of other cancers did not seem to be elevated by the incretin-based therapies. Exenatide, a glucagon-like peptide–1 (GLP-1) receptor agonist, stimulates insulin secretion when glucose levels are high; sitagliptin, a dipeptidyl peptidase–4 (DPP-4) inhibitor, acts to lower glucose levels by prolonging the activity of GLP-1.

Dr. Butler, professor of medicine and chief of the division of endocrinology, diabetes, and hypertension at the University of California, Los Angeles, noted that there is also evidence that GLP-1–based therapy induces proliferative changes in the pancreatic ducts of rodents. GLP-1 receptors are also expressed in a very high percentage of premalignant and malignant pancreatic lesions in humans.

While he admitted that he would be happy to be proved wrong, Dr. Butler stated during a debate on GLP-1–based treatment and cancer that the association must be thoroughly investigated, as it was important to ensure the overall safety of patients using these drugs.

"For now, this analysis of the FDA database does not establish that pancreatitis, [and] pancreatic and thyroid cancer are caused by GLP-1–based therapy," Dr. Butler said, citing the concluding sentences of the recently published paper. "It simply raises the level of concern that they may be and that the appropriate prospective studies are required to rule them out."

Clinical Trials May Not Be Feasible

It might not be possible to conduct such studies because of the sheer number of patients that would need to be involved, suggested Dr. Michael Nuack of the Diabetes Center, Bad Lauterberg, Germany.

"In the case of pancreatitis ... we have around 5 cases per 1,000 patient years," Dr. Nuack observed. "If you want to exclude a rise in risk by 25% you would need to do a study with 80,000 patients per arm, and I think we are all pretty sure that this will never be tested in a prospective, randomized trial."

According to Dr. Nuack, there is not enough strong evidence for or against the notion that GLP-1–based therapy can cause cancer, and using the FDA AERS database to investigate any associated risk is significantly flawed for several reasons.

Severe adverse events are more likely to be reported than are less-severe events, he suggested, and newer drugs may receive more attention than older, more established medications. Overreporting could also occur because of rumors of a cancer link, and diagnoses may not be accurate or recorded in the same way by physicians.

More importantly, the FDA AERS review included years when exenatide and sitagliptin were not available and excluded data from 2010, when more adverse events associated with the two drugs would have been available.

"Talking about cancer and malignant disease in general, based on this analysis, some forms of cancer or malignant disease are reduced," Dr. Nuack observed. For exenatide this included reduced cases of lung cancer, prostate cancer, sarcoma, lymphoma, and multiple myeloma. For sitagliptin it included colon cancer and prostate cancer.

Change in Practice Not Necessary

There is no need to change current practice, Dr. Nuack advised. "I think we can say today that there is not a single case report of pancreatic cancer in patients treated with DPP-4 inhibitors or GLP-1 receptor agonists."

Reading a statement from the European Association for the Study of Diabetes on the matter, the society’s past president, Dr. Ulf Smith, agreed that for now, no changes to treatment should be made.

"There is no immediate need for action, and patients should under no circumstances stop taking any medication, and should consult their physicians to be advised on their diabetes therapy," observed Dr. Smith of Sahlgrenska University Hospital, Gothenburg, Sweden.

"Although there are important questions which currently need to be answered with respect to the safety of incretin-based medications, there is no definitive evidence pointing to an increased cancer risk. The only robust way of measuring comparative risk is with randomized, controlled trials that record adverse events."

Dr. Butler noted that because of a possibly protective effect of metformin in relation to the development of cancer, GLP-1–based therapies should always be prescribed in association with the drug, which is in accordance with current diabetes treatment guidelines.

The FDA AERS database review was supported by the Larry L. Hillblom Foundation. Dr. Butler and Dr. Smith stated they had no conflicts of interest Dr. Nuack has received research support via his institution and acted as an advisory board member, consultant, or speaker for multiple pharmaceutical companies including the manufacturers of incretin-based therapies.

LISBON – The possibility that incretin-based therapies for type 2 diabetes can cause cancer warrants further investigation, a leading clinical diabetologist has said.

"Pancreatitis and pancreatic cancer are reported in excess with exenatide and sitagliptin therapy in the FDA AERS [Food and Drug Administration’s Adverse Events Reporting System database] in comparison to other diabetes medications," Dr. Peter Butler said at the annual meeting of the European Association for the Study of Diabetes.

Using the database, Dr. Butler and his associates recently found that treatment with exenatide (Byetta) and sitagliptin (Januvia) significantly increased the risk of pancreatitis more than sixfold, with odds ratios (ORs) of 10.7 and 6.7, respectively, compared with four control drugs – rosiglitazone (Avandia), nateglinide (Starlix), repaglinide (Prandin), and glipizide (Glucotrol) – (P less than 10^-16 for both comparisons) (Gastroenterology 2011;141:150-6). These data are somewhat alarming, as many experts now believe that pancreatitis may be a precursor to malignancy.

Furthermore, the risk of pancreatic cancer was more than doubled, compared with control treatments, with statistically significant odds ratios of 2.9 (P = 9 x 10^-5) for exenatide and 2.7 for sitagliptin (P = .008). Thyroid cancer risk was also significantly elevated by exenatide, with an odds ratio of 4.7, (P = 4 x 10^-16), but not by sitagliptin (OR = 1.5, P = .65).

The risk of other cancers did not seem to be elevated by the incretin-based therapies. Exenatide, a glucagon-like peptide–1 (GLP-1) receptor agonist, stimulates insulin secretion when glucose levels are high; sitagliptin, a dipeptidyl peptidase–4 (DPP-4) inhibitor, acts to lower glucose levels by prolonging the activity of GLP-1.

Dr. Butler, professor of medicine and chief of the division of endocrinology, diabetes, and hypertension at the University of California, Los Angeles, noted that there is also evidence that GLP-1–based therapy induces proliferative changes in the pancreatic ducts of rodents. GLP-1 receptors are also expressed in a very high percentage of premalignant and malignant pancreatic lesions in humans.

While he admitted that he would be happy to be proved wrong, Dr. Butler stated during a debate on GLP-1–based treatment and cancer that the association must be thoroughly investigated, as it was important to ensure the overall safety of patients using these drugs.

"For now, this analysis of the FDA database does not establish that pancreatitis, [and] pancreatic and thyroid cancer are caused by GLP-1–based therapy," Dr. Butler said, citing the concluding sentences of the recently published paper. "It simply raises the level of concern that they may be and that the appropriate prospective studies are required to rule them out."

Clinical Trials May Not Be Feasible

It might not be possible to conduct such studies because of the sheer number of patients that would need to be involved, suggested Dr. Michael Nuack of the Diabetes Center, Bad Lauterberg, Germany.

"In the case of pancreatitis ... we have around 5 cases per 1,000 patient years," Dr. Nuack observed. "If you want to exclude a rise in risk by 25% you would need to do a study with 80,000 patients per arm, and I think we are all pretty sure that this will never be tested in a prospective, randomized trial."

According to Dr. Nuack, there is not enough strong evidence for or against the notion that GLP-1–based therapy can cause cancer, and using the FDA AERS database to investigate any associated risk is significantly flawed for several reasons.

Severe adverse events are more likely to be reported than are less-severe events, he suggested, and newer drugs may receive more attention than older, more established medications. Overreporting could also occur because of rumors of a cancer link, and diagnoses may not be accurate or recorded in the same way by physicians.

More importantly, the FDA AERS review included years when exenatide and sitagliptin were not available and excluded data from 2010, when more adverse events associated with the two drugs would have been available.

"Talking about cancer and malignant disease in general, based on this analysis, some forms of cancer or malignant disease are reduced," Dr. Nuack observed. For exenatide this included reduced cases of lung cancer, prostate cancer, sarcoma, lymphoma, and multiple myeloma. For sitagliptin it included colon cancer and prostate cancer.

Change in Practice Not Necessary

There is no need to change current practice, Dr. Nuack advised. "I think we can say today that there is not a single case report of pancreatic cancer in patients treated with DPP-4 inhibitors or GLP-1 receptor agonists."

Reading a statement from the European Association for the Study of Diabetes on the matter, the society’s past president, Dr. Ulf Smith, agreed that for now, no changes to treatment should be made.

"There is no immediate need for action, and patients should under no circumstances stop taking any medication, and should consult their physicians to be advised on their diabetes therapy," observed Dr. Smith of Sahlgrenska University Hospital, Gothenburg, Sweden.

"Although there are important questions which currently need to be answered with respect to the safety of incretin-based medications, there is no definitive evidence pointing to an increased cancer risk. The only robust way of measuring comparative risk is with randomized, controlled trials that record adverse events."

Dr. Butler noted that because of a possibly protective effect of metformin in relation to the development of cancer, GLP-1–based therapies should always be prescribed in association with the drug, which is in accordance with current diabetes treatment guidelines.

The FDA AERS database review was supported by the Larry L. Hillblom Foundation. Dr. Butler and Dr. Smith stated they had no conflicts of interest Dr. Nuack has received research support via his institution and acted as an advisory board member, consultant, or speaker for multiple pharmaceutical companies including the manufacturers of incretin-based therapies.

LISBON – The possibility that incretin-based therapies for type 2 diabetes can cause cancer warrants further investigation, a leading clinical diabetologist has said.

"Pancreatitis and pancreatic cancer are reported in excess with exenatide and sitagliptin therapy in the FDA AERS [Food and Drug Administration’s Adverse Events Reporting System database] in comparison to other diabetes medications," Dr. Peter Butler said at the annual meeting of the European Association for the Study of Diabetes.

Using the database, Dr. Butler and his associates recently found that treatment with exenatide (Byetta) and sitagliptin (Januvia) significantly increased the risk of pancreatitis more than sixfold, with odds ratios (ORs) of 10.7 and 6.7, respectively, compared with four control drugs – rosiglitazone (Avandia), nateglinide (Starlix), repaglinide (Prandin), and glipizide (Glucotrol) – (P less than 10^-16 for both comparisons) (Gastroenterology 2011;141:150-6). These data are somewhat alarming, as many experts now believe that pancreatitis may be a precursor to malignancy.

Furthermore, the risk of pancreatic cancer was more than doubled, compared with control treatments, with statistically significant odds ratios of 2.9 (P = 9 x 10^-5) for exenatide and 2.7 for sitagliptin (P = .008). Thyroid cancer risk was also significantly elevated by exenatide, with an odds ratio of 4.7, (P = 4 x 10^-16), but not by sitagliptin (OR = 1.5, P = .65).

The risk of other cancers did not seem to be elevated by the incretin-based therapies. Exenatide, a glucagon-like peptide–1 (GLP-1) receptor agonist, stimulates insulin secretion when glucose levels are high; sitagliptin, a dipeptidyl peptidase–4 (DPP-4) inhibitor, acts to lower glucose levels by prolonging the activity of GLP-1.

Dr. Butler, professor of medicine and chief of the division of endocrinology, diabetes, and hypertension at the University of California, Los Angeles, noted that there is also evidence that GLP-1–based therapy induces proliferative changes in the pancreatic ducts of rodents. GLP-1 receptors are also expressed in a very high percentage of premalignant and malignant pancreatic lesions in humans.

While he admitted that he would be happy to be proved wrong, Dr. Butler stated during a debate on GLP-1–based treatment and cancer that the association must be thoroughly investigated, as it was important to ensure the overall safety of patients using these drugs.

"For now, this analysis of the FDA database does not establish that pancreatitis, [and] pancreatic and thyroid cancer are caused by GLP-1–based therapy," Dr. Butler said, citing the concluding sentences of the recently published paper. "It simply raises the level of concern that they may be and that the appropriate prospective studies are required to rule them out."

Clinical Trials May Not Be Feasible

It might not be possible to conduct such studies because of the sheer number of patients that would need to be involved, suggested Dr. Michael Nuack of the Diabetes Center, Bad Lauterberg, Germany.

"In the case of pancreatitis ... we have around 5 cases per 1,000 patient years," Dr. Nuack observed. "If you want to exclude a rise in risk by 25% you would need to do a study with 80,000 patients per arm, and I think we are all pretty sure that this will never be tested in a prospective, randomized trial."

According to Dr. Nuack, there is not enough strong evidence for or against the notion that GLP-1–based therapy can cause cancer, and using the FDA AERS database to investigate any associated risk is significantly flawed for several reasons.

Severe adverse events are more likely to be reported than are less-severe events, he suggested, and newer drugs may receive more attention than older, more established medications. Overreporting could also occur because of rumors of a cancer link, and diagnoses may not be accurate or recorded in the same way by physicians.

More importantly, the FDA AERS review included years when exenatide and sitagliptin were not available and excluded data from 2010, when more adverse events associated with the two drugs would have been available.

"Talking about cancer and malignant disease in general, based on this analysis, some forms of cancer or malignant disease are reduced," Dr. Nuack observed. For exenatide this included reduced cases of lung cancer, prostate cancer, sarcoma, lymphoma, and multiple myeloma. For sitagliptin it included colon cancer and prostate cancer.

Change in Practice Not Necessary

There is no need to change current practice, Dr. Nuack advised. "I think we can say today that there is not a single case report of pancreatic cancer in patients treated with DPP-4 inhibitors or GLP-1 receptor agonists."

Reading a statement from the European Association for the Study of Diabetes on the matter, the society’s past president, Dr. Ulf Smith, agreed that for now, no changes to treatment should be made.

"There is no immediate need for action, and patients should under no circumstances stop taking any medication, and should consult their physicians to be advised on their diabetes therapy," observed Dr. Smith of Sahlgrenska University Hospital, Gothenburg, Sweden.

"Although there are important questions which currently need to be answered with respect to the safety of incretin-based medications, there is no definitive evidence pointing to an increased cancer risk. The only robust way of measuring comparative risk is with randomized, controlled trials that record adverse events."

Dr. Butler noted that because of a possibly protective effect of metformin in relation to the development of cancer, GLP-1–based therapies should always be prescribed in association with the drug, which is in accordance with current diabetes treatment guidelines.

The FDA AERS database review was supported by the Larry L. Hillblom Foundation. Dr. Butler and Dr. Smith stated they had no conflicts of interest Dr. Nuack has received research support via his institution and acted as an advisory board member, consultant, or speaker for multiple pharmaceutical companies including the manufacturers of incretin-based therapies.

LISBON – A novel investigational chemokine receptor antagonist produced a statistically significant decrease in hemoglobin A_1c and a higher percentage of HbA_1c responders, compared with placebo after 4 weeks of treatment in a phase II multinational study of 159 patients with type 2 diabetes who were already taking metformin.

The compound, ChemoCentryx’s CCX140-B, works by blocking the infiltration and activation of certain monocytes/macrophages and other cells bearing chemokine receptor 2 that occur during inflammation and are implicated in the development of insulin resistance in type 2 diabetes. The agent is designed to provide selective treatment without compromising other immune functions, said Dr. Markolf Hanefeld, director of the Centre for Clinical Studies, Technical University Dresden (Germany).

The subjects had been on stable metformin treatment for at least 8 weeks before study entry. They had a mean age of 59 years, were 64% male, had a mean body mass index of 32 kg/m², median diabetes duration of 5.8 years. At baseline, they had a mean hemoglobin A_1c value of 7.5% and mean fasting plasma glucose of 170 mg/dL. They were randomized to receive 5 mg/day CCX140-B, 10 mg/day CCX140-B, placebo, or open-label 30 mg/day pioglitazone as an active comparator.

Reductions in HbA_1c from baseline at week 4 were 0.09 percentage points with both placebo and the 5 mg CCX140-B group, 0.23 for the 10 mg CCX140-B group, and 0.13 with pioglitazone. The difference, compared with placebo, was significant for 10 mg CCX140-B (P = .045) but not for pioglitazone. The percentage of patients who had a decrease in HbA_1c from baseline to week 4 was 58% for placebo, 51% for 5mg CCX140-B, 82% for 10 mg CCX140-B, and 77% for pioglitazone. Again, the difference from placebo was significant for 10 mg CCX140-B (P = 0.45) but not for 5 mg CCX140-B or pioglitazone, Dr. Hanefeld reported.

Treatment with CCX140-B also showed a dose-dependent decrease in fasting glucose through week 4, he noted.

There were no safety concerns regarding laboratory hematology, chemistry, or urinalysis. In contrast to what has been seen with other CCR2 agonists in animal studies, there were no significant changes in plasma MCP-1 or blood monocyte count with CCX140-B. There were also no significant changes in body weight or any evidence of hemodilution or peripheral edema, and there was no effect on HDL cholesterol or LDL cholesterol. There was a slight decrease in triglycerides with pioglitazone but not with CCX140-B, he said.

In a separate rodent study presented at EASD, CCX140-B significantly improved metabolic/renal parameters including hyperglycemia, insulin sensitivity, serum adiponectin, glucosuria, albuminuria, and serum markers (creatine and BUN) of renal function. The metabolic improvements correlated with a significant reduction in adipose tissue macrophage numbers. ChemoCentryx will soon be initiating phase II clinical studies of CCX140-B for the treatment of diabetic nephropathy, according to a company statement.

Dr. Hanefeld has received lecture honoraria from Bayer AG, Takeda Pharmaceutical, GlaxoSmithKline, Sanofi-Aventis, and Boehringer-Ingelheim. He is also a consultant for Sanofi-Aventis and Eli Lilly.

LISBON – A novel investigational chemokine receptor antagonist produced a statistically significant decrease in hemoglobin A_1c and a higher percentage of HbA_1c responders, compared with placebo after 4 weeks of treatment in a phase II multinational study of 159 patients with type 2 diabetes who were already taking metformin.

The compound, ChemoCentryx’s CCX140-B, works by blocking the infiltration and activation of certain monocytes/macrophages and other cells bearing chemokine receptor 2 that occur during inflammation and are implicated in the development of insulin resistance in type 2 diabetes. The agent is designed to provide selective treatment without compromising other immune functions, said Dr. Markolf Hanefeld, director of the Centre for Clinical Studies, Technical University Dresden (Germany).

The subjects had been on stable metformin treatment for at least 8 weeks before study entry. They had a mean age of 59 years, were 64% male, had a mean body mass index of 32 kg/m², median diabetes duration of 5.8 years. At baseline, they had a mean hemoglobin A_1c value of 7.5% and mean fasting plasma glucose of 170 mg/dL. They were randomized to receive 5 mg/day CCX140-B, 10 mg/day CCX140-B, placebo, or open-label 30 mg/day pioglitazone as an active comparator.

Reductions in HbA_1c from baseline at week 4 were 0.09 percentage points with both placebo and the 5 mg CCX140-B group, 0.23 for the 10 mg CCX140-B group, and 0.13 with pioglitazone. The difference, compared with placebo, was significant for 10 mg CCX140-B (P = .045) but not for pioglitazone. The percentage of patients who had a decrease in HbA_1c from baseline to week 4 was 58% for placebo, 51% for 5mg CCX140-B, 82% for 10 mg CCX140-B, and 77% for pioglitazone. Again, the difference from placebo was significant for 10 mg CCX140-B (P = 0.45) but not for 5 mg CCX140-B or pioglitazone, Dr. Hanefeld reported.

Treatment with CCX140-B also showed a dose-dependent decrease in fasting glucose through week 4, he noted.

There were no safety concerns regarding laboratory hematology, chemistry, or urinalysis. In contrast to what has been seen with other CCR2 agonists in animal studies, there were no significant changes in plasma MCP-1 or blood monocyte count with CCX140-B. There were also no significant changes in body weight or any evidence of hemodilution or peripheral edema, and there was no effect on HDL cholesterol or LDL cholesterol. There was a slight decrease in triglycerides with pioglitazone but not with CCX140-B, he said.

In a separate rodent study presented at EASD, CCX140-B significantly improved metabolic/renal parameters including hyperglycemia, insulin sensitivity, serum adiponectin, glucosuria, albuminuria, and serum markers (creatine and BUN) of renal function. The metabolic improvements correlated with a significant reduction in adipose tissue macrophage numbers. ChemoCentryx will soon be initiating phase II clinical studies of CCX140-B for the treatment of diabetic nephropathy, according to a company statement.

Dr. Hanefeld has received lecture honoraria from Bayer AG, Takeda Pharmaceutical, GlaxoSmithKline, Sanofi-Aventis, and Boehringer-Ingelheim. He is also a consultant for Sanofi-Aventis and Eli Lilly.

LISBON – A novel investigational chemokine receptor antagonist produced a statistically significant decrease in hemoglobin A_1c and a higher percentage of HbA_1c responders, compared with placebo after 4 weeks of treatment in a phase II multinational study of 159 patients with type 2 diabetes who were already taking metformin.

The compound, ChemoCentryx’s CCX140-B, works by blocking the infiltration and activation of certain monocytes/macrophages and other cells bearing chemokine receptor 2 that occur during inflammation and are implicated in the development of insulin resistance in type 2 diabetes. The agent is designed to provide selective treatment without compromising other immune functions, said Dr. Markolf Hanefeld, director of the Centre for Clinical Studies, Technical University Dresden (Germany).

The subjects had been on stable metformin treatment for at least 8 weeks before study entry. They had a mean age of 59 years, were 64% male, had a mean body mass index of 32 kg/m², median diabetes duration of 5.8 years. At baseline, they had a mean hemoglobin A_1c value of 7.5% and mean fasting plasma glucose of 170 mg/dL. They were randomized to receive 5 mg/day CCX140-B, 10 mg/day CCX140-B, placebo, or open-label 30 mg/day pioglitazone as an active comparator.

Reductions in HbA_1c from baseline at week 4 were 0.09 percentage points with both placebo and the 5 mg CCX140-B group, 0.23 for the 10 mg CCX140-B group, and 0.13 with pioglitazone. The difference, compared with placebo, was significant for 10 mg CCX140-B (P = .045) but not for pioglitazone. The percentage of patients who had a decrease in HbA_1c from baseline to week 4 was 58% for placebo, 51% for 5mg CCX140-B, 82% for 10 mg CCX140-B, and 77% for pioglitazone. Again, the difference from placebo was significant for 10 mg CCX140-B (P = 0.45) but not for 5 mg CCX140-B or pioglitazone, Dr. Hanefeld reported.

Treatment with CCX140-B also showed a dose-dependent decrease in fasting glucose through week 4, he noted.

There were no safety concerns regarding laboratory hematology, chemistry, or urinalysis. In contrast to what has been seen with other CCR2 agonists in animal studies, there were no significant changes in plasma MCP-1 or blood monocyte count with CCX140-B. There were also no significant changes in body weight or any evidence of hemodilution or peripheral edema, and there was no effect on HDL cholesterol or LDL cholesterol. There was a slight decrease in triglycerides with pioglitazone but not with CCX140-B, he said.

In a separate rodent study presented at EASD, CCX140-B significantly improved metabolic/renal parameters including hyperglycemia, insulin sensitivity, serum adiponectin, glucosuria, albuminuria, and serum markers (creatine and BUN) of renal function. The metabolic improvements correlated with a significant reduction in adipose tissue macrophage numbers. ChemoCentryx will soon be initiating phase II clinical studies of CCX140-B for the treatment of diabetic nephropathy, according to a company statement.

Dr. Hanefeld has received lecture honoraria from Bayer AG, Takeda Pharmaceutical, GlaxoSmithKline, Sanofi-Aventis, and Boehringer-Ingelheim. He is also a consultant for Sanofi-Aventis and Eli Lilly.

LISBON – Use of the recently proposed International Association of Diabetes and Pregnancy Study Group criteria for identifying women with gestational diabetes would increase the diagnosis by 240%, compared with the World Health Organization’s criteria, according to the findings of a population-based multiethnic cohort study of 823 healthy pregnant women in Norway.

"Before endorsing the IADPSG criteria, the impact of diagnosing nearly one-third of all pregnant women should be considered," said Dr. Kjersti Morkrid of the department of endocrinology at Oslo University Hospital and the Institute of Clinical Medicine at the University of Oslo.

The IADPSG criteria were derived from an international workshop held in Pasadena, Calif., in which 225 conferees from 40 countries reviewed the results of the HAPO (Hyperglycemia and Adverse Pregnancy Outcome) study and other data suggesting that there is a risk for adverse fetal outcomes below current gestational diabetes mellitus (GDM) diagnostic thresholds. Indeed, currently used criteria were either chosen to identify women at risk for the later development of diabetes (Diabetes Care 2007;30:(Suppl. 2):S251-60), or were derived from nonpregnant individuals (World Health Org. Tech. Rep. Ser. 1980;646:1-80) and were not primarily aimed at identifying pregnancies with increased risk for adverse perinatal outcome, the IADPSG authors noted (Diabetes Care 2010;33:676-91).

The newly proposed criteria, which have been endorsed by the American Diabetes Association but not the American College of Obstetricians and Gynecologists (ACOG), were aimed at identifying women who have a 75% increased risk for adverse fetal outcomes, including birth weight greater than the 90th percentile, primary cesarean delivery, neonatal hypoglycemia, and cord C-peptide above the 90th percentile (the primary HAPO outcomes).

The IADPSG criteria call for an initial measurement of fasting plasma glucose (FPG), random glucose, or hemoglobin A_1c level in all or just high-risk women at the first prenatal visit. If the results do not detect overt diabetes but the FPG is 92-126 mg/dL, then GDM is diagnosed. If the FPG is less than 92 mg/dL, then a single 75-g, 2-hour oral glucose tolerance test (OGTT) is performed at 24-28 weeks’ gestation. The OGTT should be performed after an overnight fast, with overt diabetes diagnosed if the fasting plasma glucose level is at least 126 mg/dL. GDM is diagnosed if the FPG is greater than 92 mg/dL, if the 1-hour value exceeds 180 mg/dL, or if the 2-hour value exceeds 153 mg/dL.

The IADPSG authors acknowledged that the criteria would "substantially increase the frequency of hyperglycemic disorders in pregnancy," compared with other criteria such as those of the World Health Organization, which diagnose GDM at 24-28 weeks by either an FPG of at least 126 mg/dL or a 2-hour value of at least 140 mg/dL or higher following a 75-g glucose load.

The Norwegian Stork Groruddalen Research Program quantified that increase in a multiethnic population, including 41% of the 823 from Scandinavia, 24% from South Asia (primarily Pakistan and Sri Lanka), and 15% from the Middle East (mainly from Iraq, Morocco, and Turkey), as well as others from Eastern Europe, East Asia, Somalia, sub-Saharan Africa, and South America.

"Before endorsing the IADPSG criteria, the impact of diagnosing nearly one-third of all pregnant women should be considered."

Compared with women from Scandinavia, those of South Asian and Middle Eastern origin were significantly younger on average (aged 28 and 29 years, respectively, vs. 30 years for Scandinavian women). South Asian women had lower prepregnancy body mass indexes compared with Scandinavian women (23.7 vs. 24.6 kg/m²), whereas the Middle Eastern women had a higher average BMI (25.9). Both ethnic minority groups were more likely than Scandinavian women to have a first-degree relative with diabetes (47% of South Asian women and 39% of Middle Eastern, vs. 13% of Scandinavian). Both minority groups were also shorter on average, and had more previous pregnancies and lower educational levels, she reported.

Of the 759 women who completed the OGTT at visit 2, GDM prevalence rates were 32% by the IADPSG criteria, compared with just 13% by the WHO criteria. The overall increase in GDM diagnosis was by a factor of 2.4, ranging from 2.2 for Scandinavian and Middle Eastern women to a 2.8-fold increase among Southeast Asian women.

There was poor overlap between the two criteria, with just 9% identified by both, 4% by only the WHO criteria, and 22% by only the IADPSG criteria. More than two-thirds who were identified by the IADPSG were not identified by WHO criteria, Dr. Morkrid pointed out.

After adjustment for ethnicity, age, BMI, body height, parity, education level, and first-degree relatives with diabetes, factors significantly associated with GDM by the WHO criteria were body height per cm (odds ratio, 0.92), low education level (OR, 1.83), and having a first degree relative with diabetes (OR, 2.28). With the IADPSG criteria, prepregnancy BMI (OR, 1.09) and South Asian origin (OR, 2.48) were the most significant factors.

In response to a question from the audience about fetal outcome, Dr. Morkrid responded that those data are available but have not yet been analyzed.

The IADPSG criteria have been adopted by some countries, including Japan and Germany, but not widely in the United States. In September 2010, the ACOG Committee on Obstetric Practice issued a committee opinion restating its recommendation for a "two-step" method that differs from both the IADPSG and WHO criteria, utilizing a 50-g, 1-hour "loading" test at 24-28 weeks’ gestation, followed by a confirmatory 100-g, 3-hour OGTT.

In addition to concerns regarding the dramatic increase in the number of women diagnosed with GDM that would occur with adoption of the IADPSG criteria – ACOG estimates the proportion would be 18% of all pregnant women – the ACOG committee opinion also cited the lack of evidence regarding impact on outcomes.

"There is no evidence that the identification and treatment of women based on the new [IADPSG] recommendations will lead to clinically significant improvements in maternal and neonatal outcomes, and it would lead to a significant increase in health care costs," the ACOG authors noted, adding that "significant questions remain regarding the implications on health care costs, the effect of GDM diagnosis on the pregnant woman and her family, the effect of diagnosis on obstetric interventions in pregnancy, and whether the identification and treatment of GDM will improve meaningful perinatal, neonatal, and maternal outcomes."

ACOG and other organizations are looking ahead to the National Institutes of Health’s Consensus Development Conference on Diagnosing Gestational Diabetes Mellitus, scheduled for October 29-31, 2012, to develop a uniform diagnostic standard. "Consensus regarding optimal diagnostic criteria among the many groups and professional organizations will further much needed research regarding the benefits and harms of screening and diagnosis of GDM," the ACOG opinion statement said.

The Norwegian study was funded by the Research Council of Norway, the South-Eastern Norway Regional Health Authority, and the Norwegian Directorate of Health. Neither Dr. Morkrid nor Dr. Diamant had any financial disclosures.

LISBON – Use of the recently proposed International Association of Diabetes and Pregnancy Study Group criteria for identifying women with gestational diabetes would increase the diagnosis by 240%, compared with the World Health Organization’s criteria, according to the findings of a population-based multiethnic cohort study of 823 healthy pregnant women in Norway.

"Before endorsing the IADPSG criteria, the impact of diagnosing nearly one-third of all pregnant women should be considered," said Dr. Kjersti Morkrid of the department of endocrinology at Oslo University Hospital and the Institute of Clinical Medicine at the University of Oslo.

The IADPSG criteria were derived from an international workshop held in Pasadena, Calif., in which 225 conferees from 40 countries reviewed the results of the HAPO (Hyperglycemia and Adverse Pregnancy Outcome) study and other data suggesting that there is a risk for adverse fetal outcomes below current gestational diabetes mellitus (GDM) diagnostic thresholds. Indeed, currently used criteria were either chosen to identify women at risk for the later development of diabetes (Diabetes Care 2007;30:(Suppl. 2):S251-60), or were derived from nonpregnant individuals (World Health Org. Tech. Rep. Ser. 1980;646:1-80) and were not primarily aimed at identifying pregnancies with increased risk for adverse perinatal outcome, the IADPSG authors noted (Diabetes Care 2010;33:676-91).

The newly proposed criteria, which have been endorsed by the American Diabetes Association but not the American College of Obstetricians and Gynecologists (ACOG), were aimed at identifying women who have a 75% increased risk for adverse fetal outcomes, including birth weight greater than the 90th percentile, primary cesarean delivery, neonatal hypoglycemia, and cord C-peptide above the 90th percentile (the primary HAPO outcomes).

The IADPSG criteria call for an initial measurement of fasting plasma glucose (FPG), random glucose, or hemoglobin A_1c level in all or just high-risk women at the first prenatal visit. If the results do not detect overt diabetes but the FPG is 92-126 mg/dL, then GDM is diagnosed. If the FPG is less than 92 mg/dL, then a single 75-g, 2-hour oral glucose tolerance test (OGTT) is performed at 24-28 weeks’ gestation. The OGTT should be performed after an overnight fast, with overt diabetes diagnosed if the fasting plasma glucose level is at least 126 mg/dL. GDM is diagnosed if the FPG is greater than 92 mg/dL, if the 1-hour value exceeds 180 mg/dL, or if the 2-hour value exceeds 153 mg/dL.

The IADPSG authors acknowledged that the criteria would "substantially increase the frequency of hyperglycemic disorders in pregnancy," compared with other criteria such as those of the World Health Organization, which diagnose GDM at 24-28 weeks by either an FPG of at least 126 mg/dL or a 2-hour value of at least 140 mg/dL or higher following a 75-g glucose load.

The Norwegian Stork Groruddalen Research Program quantified that increase in a multiethnic population, including 41% of the 823 from Scandinavia, 24% from South Asia (primarily Pakistan and Sri Lanka), and 15% from the Middle East (mainly from Iraq, Morocco, and Turkey), as well as others from Eastern Europe, East Asia, Somalia, sub-Saharan Africa, and South America.

"Before endorsing the IADPSG criteria, the impact of diagnosing nearly one-third of all pregnant women should be considered."

Compared with women from Scandinavia, those of South Asian and Middle Eastern origin were significantly younger on average (aged 28 and 29 years, respectively, vs. 30 years for Scandinavian women). South Asian women had lower prepregnancy body mass indexes compared with Scandinavian women (23.7 vs. 24.6 kg/m²), whereas the Middle Eastern women had a higher average BMI (25.9). Both ethnic minority groups were more likely than Scandinavian women to have a first-degree relative with diabetes (47% of South Asian women and 39% of Middle Eastern, vs. 13% of Scandinavian). Both minority groups were also shorter on average, and had more previous pregnancies and lower educational levels, she reported.

Of the 759 women who completed the OGTT at visit 2, GDM prevalence rates were 32% by the IADPSG criteria, compared with just 13% by the WHO criteria. The overall increase in GDM diagnosis was by a factor of 2.4, ranging from 2.2 for Scandinavian and Middle Eastern women to a 2.8-fold increase among Southeast Asian women.

There was poor overlap between the two criteria, with just 9% identified by both, 4% by only the WHO criteria, and 22% by only the IADPSG criteria. More than two-thirds who were identified by the IADPSG were not identified by WHO criteria, Dr. Morkrid pointed out.

After adjustment for ethnicity, age, BMI, body height, parity, education level, and first-degree relatives with diabetes, factors significantly associated with GDM by the WHO criteria were body height per cm (odds ratio, 0.92), low education level (OR, 1.83), and having a first degree relative with diabetes (OR, 2.28). With the IADPSG criteria, prepregnancy BMI (OR, 1.09) and South Asian origin (OR, 2.48) were the most significant factors.

In response to a question from the audience about fetal outcome, Dr. Morkrid responded that those data are available but have not yet been analyzed.

The IADPSG criteria have been adopted by some countries, including Japan and Germany, but not widely in the United States. In September 2010, the ACOG Committee on Obstetric Practice issued a committee opinion restating its recommendation for a "two-step" method that differs from both the IADPSG and WHO criteria, utilizing a 50-g, 1-hour "loading" test at 24-28 weeks’ gestation, followed by a confirmatory 100-g, 3-hour OGTT.

In addition to concerns regarding the dramatic increase in the number of women diagnosed with GDM that would occur with adoption of the IADPSG criteria – ACOG estimates the proportion would be 18% of all pregnant women – the ACOG committee opinion also cited the lack of evidence regarding impact on outcomes.

"There is no evidence that the identification and treatment of women based on the new [IADPSG] recommendations will lead to clinically significant improvements in maternal and neonatal outcomes, and it would lead to a significant increase in health care costs," the ACOG authors noted, adding that "significant questions remain regarding the implications on health care costs, the effect of GDM diagnosis on the pregnant woman and her family, the effect of diagnosis on obstetric interventions in pregnancy, and whether the identification and treatment of GDM will improve meaningful perinatal, neonatal, and maternal outcomes."

ACOG and other organizations are looking ahead to the National Institutes of Health’s Consensus Development Conference on Diagnosing Gestational Diabetes Mellitus, scheduled for October 29-31, 2012, to develop a uniform diagnostic standard. "Consensus regarding optimal diagnostic criteria among the many groups and professional organizations will further much needed research regarding the benefits and harms of screening and diagnosis of GDM," the ACOG opinion statement said.

The Norwegian study was funded by the Research Council of Norway, the South-Eastern Norway Regional Health Authority, and the Norwegian Directorate of Health. Neither Dr. Morkrid nor Dr. Diamant had any financial disclosures.

LISBON – Use of the recently proposed International Association of Diabetes and Pregnancy Study Group criteria for identifying women with gestational diabetes would increase the diagnosis by 240%, compared with the World Health Organization’s criteria, according to the findings of a population-based multiethnic cohort study of 823 healthy pregnant women in Norway.

"Before endorsing the IADPSG criteria, the impact of diagnosing nearly one-third of all pregnant women should be considered," said Dr. Kjersti Morkrid of the department of endocrinology at Oslo University Hospital and the Institute of Clinical Medicine at the University of Oslo.

The IADPSG criteria were derived from an international workshop held in Pasadena, Calif., in which 225 conferees from 40 countries reviewed the results of the HAPO (Hyperglycemia and Adverse Pregnancy Outcome) study and other data suggesting that there is a risk for adverse fetal outcomes below current gestational diabetes mellitus (GDM) diagnostic thresholds. Indeed, currently used criteria were either chosen to identify women at risk for the later development of diabetes (Diabetes Care 2007;30:(Suppl. 2):S251-60), or were derived from nonpregnant individuals (World Health Org. Tech. Rep. Ser. 1980;646:1-80) and were not primarily aimed at identifying pregnancies with increased risk for adverse perinatal outcome, the IADPSG authors noted (Diabetes Care 2010;33:676-91).

The newly proposed criteria, which have been endorsed by the American Diabetes Association but not the American College of Obstetricians and Gynecologists (ACOG), were aimed at identifying women who have a 75% increased risk for adverse fetal outcomes, including birth weight greater than the 90th percentile, primary cesarean delivery, neonatal hypoglycemia, and cord C-peptide above the 90th percentile (the primary HAPO outcomes).

The IADPSG criteria call for an initial measurement of fasting plasma glucose (FPG), random glucose, or hemoglobin A_1c level in all or just high-risk women at the first prenatal visit. If the results do not detect overt diabetes but the FPG is 92-126 mg/dL, then GDM is diagnosed. If the FPG is less than 92 mg/dL, then a single 75-g, 2-hour oral glucose tolerance test (OGTT) is performed at 24-28 weeks’ gestation. The OGTT should be performed after an overnight fast, with overt diabetes diagnosed if the fasting plasma glucose level is at least 126 mg/dL. GDM is diagnosed if the FPG is greater than 92 mg/dL, if the 1-hour value exceeds 180 mg/dL, or if the 2-hour value exceeds 153 mg/dL.

The IADPSG authors acknowledged that the criteria would "substantially increase the frequency of hyperglycemic disorders in pregnancy," compared with other criteria such as those of the World Health Organization, which diagnose GDM at 24-28 weeks by either an FPG of at least 126 mg/dL or a 2-hour value of at least 140 mg/dL or higher following a 75-g glucose load.

The Norwegian Stork Groruddalen Research Program quantified that increase in a multiethnic population, including 41% of the 823 from Scandinavia, 24% from South Asia (primarily Pakistan and Sri Lanka), and 15% from the Middle East (mainly from Iraq, Morocco, and Turkey), as well as others from Eastern Europe, East Asia, Somalia, sub-Saharan Africa, and South America.

"Before endorsing the IADPSG criteria, the impact of diagnosing nearly one-third of all pregnant women should be considered."

Compared with women from Scandinavia, those of South Asian and Middle Eastern origin were significantly younger on average (aged 28 and 29 years, respectively, vs. 30 years for Scandinavian women). South Asian women had lower prepregnancy body mass indexes compared with Scandinavian women (23.7 vs. 24.6 kg/m²), whereas the Middle Eastern women had a higher average BMI (25.9). Both ethnic minority groups were more likely than Scandinavian women to have a first-degree relative with diabetes (47% of South Asian women and 39% of Middle Eastern, vs. 13% of Scandinavian). Both minority groups were also shorter on average, and had more previous pregnancies and lower educational levels, she reported.

Of the 759 women who completed the OGTT at visit 2, GDM prevalence rates were 32% by the IADPSG criteria, compared with just 13% by the WHO criteria. The overall increase in GDM diagnosis was by a factor of 2.4, ranging from 2.2 for Scandinavian and Middle Eastern women to a 2.8-fold increase among Southeast Asian women.

There was poor overlap between the two criteria, with just 9% identified by both, 4% by only the WHO criteria, and 22% by only the IADPSG criteria. More than two-thirds who were identified by the IADPSG were not identified by WHO criteria, Dr. Morkrid pointed out.

After adjustment for ethnicity, age, BMI, body height, parity, education level, and first-degree relatives with diabetes, factors significantly associated with GDM by the WHO criteria were body height per cm (odds ratio, 0.92), low education level (OR, 1.83), and having a first degree relative with diabetes (OR, 2.28). With the IADPSG criteria, prepregnancy BMI (OR, 1.09) and South Asian origin (OR, 2.48) were the most significant factors.

In response to a question from the audience about fetal outcome, Dr. Morkrid responded that those data are available but have not yet been analyzed.

The IADPSG criteria have been adopted by some countries, including Japan and Germany, but not widely in the United States. In September 2010, the ACOG Committee on Obstetric Practice issued a committee opinion restating its recommendation for a "two-step" method that differs from both the IADPSG and WHO criteria, utilizing a 50-g, 1-hour "loading" test at 24-28 weeks’ gestation, followed by a confirmatory 100-g, 3-hour OGTT.

In addition to concerns regarding the dramatic increase in the number of women diagnosed with GDM that would occur with adoption of the IADPSG criteria – ACOG estimates the proportion would be 18% of all pregnant women – the ACOG committee opinion also cited the lack of evidence regarding impact on outcomes.

"There is no evidence that the identification and treatment of women based on the new [IADPSG] recommendations will lead to clinically significant improvements in maternal and neonatal outcomes, and it would lead to a significant increase in health care costs," the ACOG authors noted, adding that "significant questions remain regarding the implications on health care costs, the effect of GDM diagnosis on the pregnant woman and her family, the effect of diagnosis on obstetric interventions in pregnancy, and whether the identification and treatment of GDM will improve meaningful perinatal, neonatal, and maternal outcomes."

ACOG and other organizations are looking ahead to the National Institutes of Health’s Consensus Development Conference on Diagnosing Gestational Diabetes Mellitus, scheduled for October 29-31, 2012, to develop a uniform diagnostic standard. "Consensus regarding optimal diagnostic criteria among the many groups and professional organizations will further much needed research regarding the benefits and harms of screening and diagnosis of GDM," the ACOG opinion statement said.

The Norwegian study was funded by the Research Council of Norway, the South-Eastern Norway Regional Health Authority, and the Norwegian Directorate of Health. Neither Dr. Morkrid nor Dr. Diamant had any financial disclosures.

LISBON – Increasing accumulation of fat, inflammation, and fibrosis of the liver appears tied to corresponding increases in the risk of cardiovascular disease, especially in patients with diabetes, according to the findings of a small retrospective study presented at the annual meeting of the European Society for the Study of Diabetes.

"What we are realizing is that [nonalcoholic fatty liver disease] is adding extra cardiovascular risk to people with diabetes, and to those without, on top of that which is already existing," Dr. Christopher Byrne, one of the lead study investigators, said in an interview.

Dr. Byrne, professor of endocrinology and metabolism at the University of Southampton (England), suggested that patients with documented liver disease perhaps require more aggressive therapies to address the additional risk. Such therapies need to target the liver as much as the heart.

In the study of 112 patients with biopsy-proven nonalcoholic fatty liver disease (NAFLD), Kleiner scores – a histologic measure of NAFLD severity – were shown to be highly correlated with both Framingham Risk Score (FRS) and QRISK2, two cardiovascular risk calculators.

"Nonalcoholic fatty liver disease represents a spectrum of fat-mediated liver conditions causing progressive hepatocellular damage," said Sarah Hudson, a 5th-year medical student at the University of Southampton.

"There is increasing evidence of an increased cardiovascular risk associated with progression of nonalcoholic fatty liver disease," Ms. Hudson explained, noting that the preferred method of determining NAFLD severity was via histopathologic assessment.

The aim of the study was to see if a histopathologic marker – Kleiner score – correlated with cardiovascular risk, and if scores were higher in people already known to have a high cardiovascular risk, namely those with diabetes.

Kleiner scores assess the degree of steatosis, lobular inflammation, hepatocyte "ballooning," and fibrosis, with higher scores indicating more severe liver disease (Hepatology 2005;41:1313-21).

The mean age of the study cohort was 48 years and the mean Kleiner score was 5.3. The median FRS was 13 and the median QRISK2 score was 8. The mean body mass index of participants was approximately 34 kg/m².

Kleiner scores were not only found to be highly correlated with both cardiovascular risk models used, but they were also higher in a subgroup of 32 patients with diabetes when compared with those without diabetes (6.4 vs. 4.7, P less than .001).

The increased risk of cardiovascular disease in correlation with increasing NAFLD severity was found to be independent of both hyperglycemia and increasing body weight.

"We need more prospective studies to see what markers may be used to help stratify who requires biopsy and how best to manage people who have got NAFLD." Nonalcoholic steatohepatitis was associated with the highest cardiovascular risk estimates in the study.

"Up until now we’ve been very poor in providing cardiovascular risk reduction treatments for patients with NAFLD," Dr. Byrne said. Currently the only treatment strategy proven to work for NAFLD is lifestyle changes.

"We know that losing weight and increasing activity levels are very effective at decreasing liver fat," Dr. Byrne explained. "But what we don’t know is whether those lifestyle changes are good at decreasing liver inflammation, or decreasing liver fibrosis."

Treatment to decrease liver fat and prevent progression to fibrosis is thus urgently needed, and Dr. Byrne is part of a team now looking at the use of a high concentration of highly purified omega-3 fatty acid ethyl esters in the treatment of NAFLD.

The highly purified fish oil being used in the trial has been available commercially in Europe for at least a decade (Omacor) and in the United States since 2004 (Lovaza), and is currently licensed to treat hypertriglyceridemia.

The study is not due to be completed until December 2012. Until then, clinicians need to optimize the treatments they have at their disposal, including antihypertensive and lipid-lowering medicines commonly used to lower cardiovascular risk in conjunction with lifestyle modifications.

The study was funded by the National Institute for Health Research and Diabetes UK. Dr. Byrne and Ms. Hudson said they had no relevant financial disclosures. Dr. Byrne said he has given lectures on behalf of pharmaceutical companies in the past, including Pfizer.

LISBON – Increasing accumulation of fat, inflammation, and fibrosis of the liver appears tied to corresponding increases in the risk of cardiovascular disease, especially in patients with diabetes, according to the findings of a small retrospective study presented at the annual meeting of the European Society for the Study of Diabetes.

"What we are realizing is that [nonalcoholic fatty liver disease] is adding extra cardiovascular risk to people with diabetes, and to those without, on top of that which is already existing," Dr. Christopher Byrne, one of the lead study investigators, said in an interview.

Dr. Byrne, professor of endocrinology and metabolism at the University of Southampton (England), suggested that patients with documented liver disease perhaps require more aggressive therapies to address the additional risk. Such therapies need to target the liver as much as the heart.

In the study of 112 patients with biopsy-proven nonalcoholic fatty liver disease (NAFLD), Kleiner scores – a histologic measure of NAFLD severity – were shown to be highly correlated with both Framingham Risk Score (FRS) and QRISK2, two cardiovascular risk calculators.

"Nonalcoholic fatty liver disease represents a spectrum of fat-mediated liver conditions causing progressive hepatocellular damage," said Sarah Hudson, a 5th-year medical student at the University of Southampton.

"There is increasing evidence of an increased cardiovascular risk associated with progression of nonalcoholic fatty liver disease," Ms. Hudson explained, noting that the preferred method of determining NAFLD severity was via histopathologic assessment.

The aim of the study was to see if a histopathologic marker – Kleiner score – correlated with cardiovascular risk, and if scores were higher in people already known to have a high cardiovascular risk, namely those with diabetes.

Kleiner scores assess the degree of steatosis, lobular inflammation, hepatocyte "ballooning," and fibrosis, with higher scores indicating more severe liver disease (Hepatology 2005;41:1313-21).

The mean age of the study cohort was 48 years and the mean Kleiner score was 5.3. The median FRS was 13 and the median QRISK2 score was 8. The mean body mass index of participants was approximately 34 kg/m².

Kleiner scores were not only found to be highly correlated with both cardiovascular risk models used, but they were also higher in a subgroup of 32 patients with diabetes when compared with those without diabetes (6.4 vs. 4.7, P less than .001).

The increased risk of cardiovascular disease in correlation with increasing NAFLD severity was found to be independent of both hyperglycemia and increasing body weight.

"We need more prospective studies to see what markers may be used to help stratify who requires biopsy and how best to manage people who have got NAFLD." Nonalcoholic steatohepatitis was associated with the highest cardiovascular risk estimates in the study.

"Up until now we’ve been very poor in providing cardiovascular risk reduction treatments for patients with NAFLD," Dr. Byrne said. Currently the only treatment strategy proven to work for NAFLD is lifestyle changes.

"We know that losing weight and increasing activity levels are very effective at decreasing liver fat," Dr. Byrne explained. "But what we don’t know is whether those lifestyle changes are good at decreasing liver inflammation, or decreasing liver fibrosis."

Treatment to decrease liver fat and prevent progression to fibrosis is thus urgently needed, and Dr. Byrne is part of a team now looking at the use of a high concentration of highly purified omega-3 fatty acid ethyl esters in the treatment of NAFLD.

The highly purified fish oil being used in the trial has been available commercially in Europe for at least a decade (Omacor) and in the United States since 2004 (Lovaza), and is currently licensed to treat hypertriglyceridemia.

The study is not due to be completed until December 2012. Until then, clinicians need to optimize the treatments they have at their disposal, including antihypertensive and lipid-lowering medicines commonly used to lower cardiovascular risk in conjunction with lifestyle modifications.

The study was funded by the National Institute for Health Research and Diabetes UK. Dr. Byrne and Ms. Hudson said they had no relevant financial disclosures. Dr. Byrne said he has given lectures on behalf of pharmaceutical companies in the past, including Pfizer.

LISBON – Increasing accumulation of fat, inflammation, and fibrosis of the liver appears tied to corresponding increases in the risk of cardiovascular disease, especially in patients with diabetes, according to the findings of a small retrospective study presented at the annual meeting of the European Society for the Study of Diabetes.

"What we are realizing is that [nonalcoholic fatty liver disease] is adding extra cardiovascular risk to people with diabetes, and to those without, on top of that which is already existing," Dr. Christopher Byrne, one of the lead study investigators, said in an interview.

Dr. Byrne, professor of endocrinology and metabolism at the University of Southampton (England), suggested that patients with documented liver disease perhaps require more aggressive therapies to address the additional risk. Such therapies need to target the liver as much as the heart.

In the study of 112 patients with biopsy-proven nonalcoholic fatty liver disease (NAFLD), Kleiner scores – a histologic measure of NAFLD severity – were shown to be highly correlated with both Framingham Risk Score (FRS) and QRISK2, two cardiovascular risk calculators.

"Nonalcoholic fatty liver disease represents a spectrum of fat-mediated liver conditions causing progressive hepatocellular damage," said Sarah Hudson, a 5th-year medical student at the University of Southampton.

"There is increasing evidence of an increased cardiovascular risk associated with progression of nonalcoholic fatty liver disease," Ms. Hudson explained, noting that the preferred method of determining NAFLD severity was via histopathologic assessment.

The aim of the study was to see if a histopathologic marker – Kleiner score – correlated with cardiovascular risk, and if scores were higher in people already known to have a high cardiovascular risk, namely those with diabetes.

Kleiner scores assess the degree of steatosis, lobular inflammation, hepatocyte "ballooning," and fibrosis, with higher scores indicating more severe liver disease (Hepatology 2005;41:1313-21).

The mean age of the study cohort was 48 years and the mean Kleiner score was 5.3. The median FRS was 13 and the median QRISK2 score was 8. The mean body mass index of participants was approximately 34 kg/m².

Kleiner scores were not only found to be highly correlated with both cardiovascular risk models used, but they were also higher in a subgroup of 32 patients with diabetes when compared with those without diabetes (6.4 vs. 4.7, P less than .001).

The increased risk of cardiovascular disease in correlation with increasing NAFLD severity was found to be independent of both hyperglycemia and increasing body weight.

"We need more prospective studies to see what markers may be used to help stratify who requires biopsy and how best to manage people who have got NAFLD." Nonalcoholic steatohepatitis was associated with the highest cardiovascular risk estimates in the study.

"Up until now we’ve been very poor in providing cardiovascular risk reduction treatments for patients with NAFLD," Dr. Byrne said. Currently the only treatment strategy proven to work for NAFLD is lifestyle changes.

"We know that losing weight and increasing activity levels are very effective at decreasing liver fat," Dr. Byrne explained. "But what we don’t know is whether those lifestyle changes are good at decreasing liver inflammation, or decreasing liver fibrosis."

Treatment to decrease liver fat and prevent progression to fibrosis is thus urgently needed, and Dr. Byrne is part of a team now looking at the use of a high concentration of highly purified omega-3 fatty acid ethyl esters in the treatment of NAFLD.

The highly purified fish oil being used in the trial has been available commercially in Europe for at least a decade (Omacor) and in the United States since 2004 (Lovaza), and is currently licensed to treat hypertriglyceridemia.

The study is not due to be completed until December 2012. Until then, clinicians need to optimize the treatments they have at their disposal, including antihypertensive and lipid-lowering medicines commonly used to lower cardiovascular risk in conjunction with lifestyle modifications.

The study was funded by the National Institute for Health Research and Diabetes UK. Dr. Byrne and Ms. Hudson said they had no relevant financial disclosures. Dr. Byrne said he has given lectures on behalf of pharmaceutical companies in the past, including Pfizer.

LISBON – Improved glucose parameters, substantial weight loss, and increased incretin hormone levels can be achieved by the insertion of a novel, minimally invasive, intestinal device in obese patients with type 2 diabetes.

The use of a duodenal-jejunal bypass liner (DJBL) not only improves hemoglobin A_1c and aids weight loss, but also appears to increase levels of glucagon-like peptide (GLP)-1 and peptide YY while in place, according to the findings of a small study presented at the annual meeting of the European Society for the Study of Diabetes.

Although the effects may be temporary, they could offer patients a reversible alternative to bariatric surgery that helps to kick-start weight loss and self-management of diabetes, said study author Dr. Charlotte de Jonge of Maastricht University Medical Centre in the Netherlands. "Not all patients want [bariatric] surgery, as it is permanent."

She added that the DJBL was perhaps "an easy first step" and that patients could perhaps still opt for weight-loss surgery later on or be retreated with the device to the point that medication was again sufficient to help manage their diabetes.

The DJBL (EndoBarrier) is a 60-cm impermeable sleeve that is inserted and removed endoscopically, and which effectively blocks the duodenum and proximal jejunum in a manner similar to the Roux-en-Y-gastric bypass procedure. It is thought to work by creating a physical barrier between ingested food and the intestinal wall, and perhaps alters the activation of incretin hormones in the gut. On average, the device can be inserted in 20 minutes and removed in 10 minutes, under conscious sedation, which allows the patient to go home the same day as the procedure.

The aim of the 17-patient study was to investigate the possible mechanisms for the improvement in diabetic parameters after insertion of the DJBL seen in previous studies. Fourteen men and three women participated, all of whom had type 2 diabetes and a body mass index in excess of 30 kg/m². All ate a low-calorie diet during the study, which restricted their intake to around 1,200-1,500 kcal per day. Participants consulted a nutritionist every month, but the diet was not prescriptive.

Before implantation of the device, subjects underwent a meal tolerance test that involved a 12-hour fast, then ingestion of a 500-kcal liquid meal and blood sampling at baseline and at 10-, 20-, 30-, 60-, 90-, and 120-minute intervals. HbA_1c, glucose, insulin, GLP-1, and PYY concentrations were measured. Measurements were repeated before removal of the device, and again 1 week after removal of the device.

The DJBL was left in place for 24 weeks, although Dr. de Jonge noted that the device could be used for up to 2 years in some patients. Around 500 patients have received the device in clinical studies, she said in an interview. It has received approval for use in a few European countries and Australia. It remains investigational in the United States.

Within 1 week after implantation, fasting and area under the curve (AUC) glucose concentrations were improved (11.4±0.5 mmol/L vs. 8.9±0.4 mmol/L and 1,999±88 vs. 1,535±53), respectively. In addition, AUC concentrations of GLP-1 increased from 2,584 at baseline to 4,112 at removal and PYY from 4,440 to 6,448 (P less than .01 for all comparisons with baseline).

When the device was removed at 6 months, a significant mean weight loss of 13 kg (P less than .001) had been recorded, with a mean loss of excess weight of 30% (P less than .001), said Dr. de Jonge. Importantly, mean HbA_1c decreased from 8.4% at baseline to 7.0% at removal (P less than .001) and there was a reduction in the use of antidiabetic medication in all but one of the study participants.

"Interestingly, GLP-1 and GIP [glucose-dependent insulinotropic peptide] not only have an effect on insulin, but they also affect glucagon as well," Dr. de Jonge reported during her presentation. There was a normalization of the glucagon response during treatment with the DJBL to a more physiological response.

Almost all patients reported increased satiety, she added.

Commenting on the presentation, Dr. Roy Taylor, professor of medicine and metabolism at the University of Newcastle in England, noted that it would be useful to know what the effects of diet alone were and to see the relationship to the other changes in parameters shown.

In an interview, Dr. de Jonge noted that other data had suggested the weight loss achieved by diet alone was around 4-5 kg, so there did appear to be a greater weight loss effect when the DJBL was inserted.

With regard to side effects, the most common adverse events were abdominal discomfort, including epigastric pain and nausea. Such events were more common in the first 1-2 weeks, but tended to resolve with longer duration of use. There was no withdrawal of the device, and Dr. de Jonge noted that even when patients reported side effects, they were loath to have it removed.

"What we hope is that we can encourage lifestyle changes and we can motivate patients to remain at a lower weight," Dr. de Jonge observed. "Maybe people still need a little bit of medication, but at least their diabetes is better [controlled] than before they had the procedure."

GI Dynamics funded the study. Dr. de Jonge and Dr. Taylor reported having no conflicts of interest.

LISBON – Improved glucose parameters, substantial weight loss, and increased incretin hormone levels can be achieved by the insertion of a novel, minimally invasive, intestinal device in obese patients with type 2 diabetes.

The use of a duodenal-jejunal bypass liner (DJBL) not only improves hemoglobin A_1c and aids weight loss, but also appears to increase levels of glucagon-like peptide (GLP)-1 and peptide YY while in place, according to the findings of a small study presented at the annual meeting of the European Society for the Study of Diabetes.

Although the effects may be temporary, they could offer patients a reversible alternative to bariatric surgery that helps to kick-start weight loss and self-management of diabetes, said study author Dr. Charlotte de Jonge of Maastricht University Medical Centre in the Netherlands. "Not all patients want [bariatric] surgery, as it is permanent."

She added that the DJBL was perhaps "an easy first step" and that patients could perhaps still opt for weight-loss surgery later on or be retreated with the device to the point that medication was again sufficient to help manage their diabetes.

The DJBL (EndoBarrier) is a 60-cm impermeable sleeve that is inserted and removed endoscopically, and which effectively blocks the duodenum and proximal jejunum in a manner similar to the Roux-en-Y-gastric bypass procedure. It is thought to work by creating a physical barrier between ingested food and the intestinal wall, and perhaps alters the activation of incretin hormones in the gut. On average, the device can be inserted in 20 minutes and removed in 10 minutes, under conscious sedation, which allows the patient to go home the same day as the procedure.

The aim of the 17-patient study was to investigate the possible mechanisms for the improvement in diabetic parameters after insertion of the DJBL seen in previous studies. Fourteen men and three women participated, all of whom had type 2 diabetes and a body mass index in excess of 30 kg/m². All ate a low-calorie diet during the study, which restricted their intake to around 1,200-1,500 kcal per day. Participants consulted a nutritionist every month, but the diet was not prescriptive.

Before implantation of the device, subjects underwent a meal tolerance test that involved a 12-hour fast, then ingestion of a 500-kcal liquid meal and blood sampling at baseline and at 10-, 20-, 30-, 60-, 90-, and 120-minute intervals. HbA_1c, glucose, insulin, GLP-1, and PYY concentrations were measured. Measurements were repeated before removal of the device, and again 1 week after removal of the device.

The DJBL was left in place for 24 weeks, although Dr. de Jonge noted that the device could be used for up to 2 years in some patients. Around 500 patients have received the device in clinical studies, she said in an interview. It has received approval for use in a few European countries and Australia. It remains investigational in the United States.

Within 1 week after implantation, fasting and area under the curve (AUC) glucose concentrations were improved (11.4±0.5 mmol/L vs. 8.9±0.4 mmol/L and 1,999±88 vs. 1,535±53), respectively. In addition, AUC concentrations of GLP-1 increased from 2,584 at baseline to 4,112 at removal and PYY from 4,440 to 6,448 (P less than .01 for all comparisons with baseline).

When the device was removed at 6 months, a significant mean weight loss of 13 kg (P less than .001) had been recorded, with a mean loss of excess weight of 30% (P less than .001), said Dr. de Jonge. Importantly, mean HbA_1c decreased from 8.4% at baseline to 7.0% at removal (P less than .001) and there was a reduction in the use of antidiabetic medication in all but one of the study participants.

"Interestingly, GLP-1 and GIP [glucose-dependent insulinotropic peptide] not only have an effect on insulin, but they also affect glucagon as well," Dr. de Jonge reported during her presentation. There was a normalization of the glucagon response during treatment with the DJBL to a more physiological response.

Almost all patients reported increased satiety, she added.

Commenting on the presentation, Dr. Roy Taylor, professor of medicine and metabolism at the University of Newcastle in England, noted that it would be useful to know what the effects of diet alone were and to see the relationship to the other changes in parameters shown.

In an interview, Dr. de Jonge noted that other data had suggested the weight loss achieved by diet alone was around 4-5 kg, so there did appear to be a greater weight loss effect when the DJBL was inserted.

With regard to side effects, the most common adverse events were abdominal discomfort, including epigastric pain and nausea. Such events were more common in the first 1-2 weeks, but tended to resolve with longer duration of use. There was no withdrawal of the device, and Dr. de Jonge noted that even when patients reported side effects, they were loath to have it removed.

"What we hope is that we can encourage lifestyle changes and we can motivate patients to remain at a lower weight," Dr. de Jonge observed. "Maybe people still need a little bit of medication, but at least their diabetes is better [controlled] than before they had the procedure."

GI Dynamics funded the study. Dr. de Jonge and Dr. Taylor reported having no conflicts of interest.

LISBON – Improved glucose parameters, substantial weight loss, and increased incretin hormone levels can be achieved by the insertion of a novel, minimally invasive, intestinal device in obese patients with type 2 diabetes.

The use of a duodenal-jejunal bypass liner (DJBL) not only improves hemoglobin A_1c and aids weight loss, but also appears to increase levels of glucagon-like peptide (GLP)-1 and peptide YY while in place, according to the findings of a small study presented at the annual meeting of the European Society for the Study of Diabetes.

Although the effects may be temporary, they could offer patients a reversible alternative to bariatric surgery that helps to kick-start weight loss and self-management of diabetes, said study author Dr. Charlotte de Jonge of Maastricht University Medical Centre in the Netherlands. "Not all patients want [bariatric] surgery, as it is permanent."

She added that the DJBL was perhaps "an easy first step" and that patients could perhaps still opt for weight-loss surgery later on or be retreated with the device to the point that medication was again sufficient to help manage their diabetes.

The DJBL (EndoBarrier) is a 60-cm impermeable sleeve that is inserted and removed endoscopically, and which effectively blocks the duodenum and proximal jejunum in a manner similar to the Roux-en-Y-gastric bypass procedure. It is thought to work by creating a physical barrier between ingested food and the intestinal wall, and perhaps alters the activation of incretin hormones in the gut. On average, the device can be inserted in 20 minutes and removed in 10 minutes, under conscious sedation, which allows the patient to go home the same day as the procedure.

The aim of the 17-patient study was to investigate the possible mechanisms for the improvement in diabetic parameters after insertion of the DJBL seen in previous studies. Fourteen men and three women participated, all of whom had type 2 diabetes and a body mass index in excess of 30 kg/m². All ate a low-calorie diet during the study, which restricted their intake to around 1,200-1,500 kcal per day. Participants consulted a nutritionist every month, but the diet was not prescriptive.

Before implantation of the device, subjects underwent a meal tolerance test that involved a 12-hour fast, then ingestion of a 500-kcal liquid meal and blood sampling at baseline and at 10-, 20-, 30-, 60-, 90-, and 120-minute intervals. HbA_1c, glucose, insulin, GLP-1, and PYY concentrations were measured. Measurements were repeated before removal of the device, and again 1 week after removal of the device.

The DJBL was left in place for 24 weeks, although Dr. de Jonge noted that the device could be used for up to 2 years in some patients. Around 500 patients have received the device in clinical studies, she said in an interview. It has received approval for use in a few European countries and Australia. It remains investigational in the United States.

Within 1 week after implantation, fasting and area under the curve (AUC) glucose concentrations were improved (11.4±0.5 mmol/L vs. 8.9±0.4 mmol/L and 1,999±88 vs. 1,535±53), respectively. In addition, AUC concentrations of GLP-1 increased from 2,584 at baseline to 4,112 at removal and PYY from 4,440 to 6,448 (P less than .01 for all comparisons with baseline).

When the device was removed at 6 months, a significant mean weight loss of 13 kg (P less than .001) had been recorded, with a mean loss of excess weight of 30% (P less than .001), said Dr. de Jonge. Importantly, mean HbA_1c decreased from 8.4% at baseline to 7.0% at removal (P less than .001) and there was a reduction in the use of antidiabetic medication in all but one of the study participants.

"Interestingly, GLP-1 and GIP [glucose-dependent insulinotropic peptide] not only have an effect on insulin, but they also affect glucagon as well," Dr. de Jonge reported during her presentation. There was a normalization of the glucagon response during treatment with the DJBL to a more physiological response.

Almost all patients reported increased satiety, she added.

Commenting on the presentation, Dr. Roy Taylor, professor of medicine and metabolism at the University of Newcastle in England, noted that it would be useful to know what the effects of diet alone were and to see the relationship to the other changes in parameters shown.

In an interview, Dr. de Jonge noted that other data had suggested the weight loss achieved by diet alone was around 4-5 kg, so there did appear to be a greater weight loss effect when the DJBL was inserted.

With regard to side effects, the most common adverse events were abdominal discomfort, including epigastric pain and nausea. Such events were more common in the first 1-2 weeks, but tended to resolve with longer duration of use. There was no withdrawal of the device, and Dr. de Jonge noted that even when patients reported side effects, they were loath to have it removed.

"What we hope is that we can encourage lifestyle changes and we can motivate patients to remain at a lower weight," Dr. de Jonge observed. "Maybe people still need a little bit of medication, but at least their diabetes is better [controlled] than before they had the procedure."

GI Dynamics funded the study. Dr. de Jonge and Dr. Taylor reported having no conflicts of interest.

LISBON – New data on two investigational glucagon receptor antagonists suggest that this novel class of drugs has potential to be an effective treatment for type 2 diabetes, but only if safety can be established, according to two presentations at the annual meeting of the European Society for the Study of Diabetes.

Both Merck’s MK-0893 and Eli Lilly’s LY2409021 work by suppressing glucagon action via receptor blockade. Dr. Samuel S. Engel, senior director of clinical research, diabetes and endocrinology, at Merck presented phase II study data on the efficacy and safety of MK-0893 in combination with metformin or sitagliptin in 146 patients with type 2 diabetes, and Dr. Melvin J. Prince, senior director of medical diabetes and endocrinology at Eli Lilly, presented phase I dose-finding data on LY2409021 in 47 patients.

Both agents produced statistically significant reductions in hemoglobin A_1c. However, both also produced small but significant elevations in liver enzymes, and both carry the theoretical potential for hypoglycemia and hyperglucagonemia. "I think there is definitely potential for the glucagon receptor antagonists in diabetes, but there is also a risk of hypoglycemia associated with them. Also, it has been noted in the past that the hyperglucagonemia that results when the alpha cells overproduce glucagon has led to pancreas tumors in animal studies, so that’s a risk," session moderator Dr. Finbarr O’Harte, professor of endocrinology and metabolism at the University of Ulster, Derry (Northern Ireland), said in an interview.

In June at the American Diabetes Association’s annual scientific sessions, Dr. Engel presented data from a phase II study in which 342 type 2 diabetes patients were randomized to once-daily MK-0893 in four different dosages, metformin 1,000 mg twice daily, or placebo. At 12 weeks, treatment with MK-0893 resulted in significant, dose-dependent reductions in fasting plasma glucose, ranging from 32 mg/dL with a 20-mg dose to 63 mg/dL with 80 mg, from a baseline of 180-193 mg/dL. Metformin reduced FPG by 37 mg/dL, and placebo by just 2 mg/dL from baseline. For HbA_1c, reductions at 12 weeks ranged from 0.6 to 1.5 percentage points, compared with 0.8 percentage points with metformin and 0.5 with placebo.

The current study evaluated 40-mg/day MK-0893 in combination with 2,000-mg/day metformin or 100-mg/day sitagliptin, as well as 100 mg of sitagliptin plus 2,000 mg of metformin. The 146 patients had a mean age of 53 years, HbA_1c of 8.6%, and mean diabetes duration of 7 years. The MK-0893/metformin combination was superior to sitagliptin/metformin in lowering 24-hour weighted mean glucose, with a reduction of 117 mg/dL compared with 85 mg/dL. However, MK-0893/sitagliptin was significantly less effective than sitagliptin/metformin, which produced a 24-hour WMG reduction of 100 mg/dL. This could be because sitagliptin, a DPP-4 inhibitor, also suppresses glucagon and so there could be a threshold effect, Dr. Engel noted.

All treatments were equally well tolerated. However, there was a significantly higher incidence of diarrhea in the two MK-0893 groups (10% vs. 0%). In addition, the liver enzymes alanine transaminase (ALT) and aspartate aminotransferase (AST) were elevated in the MK-0893/metformin group, and total cholesterol and LDL-cholesterol were increased from baseline with MK-0893/sitagliptin, relative to reductions with MK-0893/metformin and sitagliptin/metformin, he reported.

In the phase I dose-ranging study of LY-2409021, 47 patients were randomized to one of four doses or metformin. Patients had a mean baseline fasting blood glucose of 148 mg/dL, and a baseline HbA_1c of 8.0%. By day 28, mean reductions in HbA_1c were statistically significant compared with baseline in all treatment groups, ranging from 1.02 percentage points with 60 mg to 0.69 percentage points with 5 mg. In the placebo group, HbA_1c dropped by 0.49 percentage points, Dr. Prince reported.

Fasting glucagon significantly increased by 0.6- to 4.2-fold compared with baseline across the LY dose levels, and fasting active glucagon-like peptide-1 (GLP-1) rose by 59% at the highest dose. Glucagon and GLP-1 returned to baseline levels during follow-up, he said.

The agent was generally well tolerated. Reversible elevations in hepatic transaminases were seen in five of nine patients in the highest-dose group, with no clinical signs or significant elevations in bilirubin or alkaline phosphatase, he reported. In the interview, Dr. O’Harte said that these two agents seem to have already exceeded expectations based on previous experience with other small-molecule glucagon antagonists that have been tested and have failed due to toxicity issues. "A lot of these small-molecule drugs have fallen by the wayside, and these two are still in there. We need toxicity studies, but there’s a possibility for a breakthrough, I hope."

Dr. Engel is an employee of Merck, and Dr. Prince is an employee of Eli Lilly. Dr. O’Harte stated that he had no disclosures.

LISBON – New data on two investigational glucagon receptor antagonists suggest that this novel class of drugs has potential to be an effective treatment for type 2 diabetes, but only if safety can be established, according to two presentations at the annual meeting of the European Society for the Study of Diabetes.

Both Merck’s MK-0893 and Eli Lilly’s LY2409021 work by suppressing glucagon action via receptor blockade. Dr. Samuel S. Engel, senior director of clinical research, diabetes and endocrinology, at Merck presented phase II study data on the efficacy and safety of MK-0893 in combination with metformin or sitagliptin in 146 patients with type 2 diabetes, and Dr. Melvin J. Prince, senior director of medical diabetes and endocrinology at Eli Lilly, presented phase I dose-finding data on LY2409021 in 47 patients.

Both agents produced statistically significant reductions in hemoglobin A_1c. However, both also produced small but significant elevations in liver enzymes, and both carry the theoretical potential for hypoglycemia and hyperglucagonemia. "I think there is definitely potential for the glucagon receptor antagonists in diabetes, but there is also a risk of hypoglycemia associated with them. Also, it has been noted in the past that the hyperglucagonemia that results when the alpha cells overproduce glucagon has led to pancreas tumors in animal studies, so that’s a risk," session moderator Dr. Finbarr O’Harte, professor of endocrinology and metabolism at the University of Ulster, Derry (Northern Ireland), said in an interview.

In June at the American Diabetes Association’s annual scientific sessions, Dr. Engel presented data from a phase II study in which 342 type 2 diabetes patients were randomized to once-daily MK-0893 in four different dosages, metformin 1,000 mg twice daily, or placebo. At 12 weeks, treatment with MK-0893 resulted in significant, dose-dependent reductions in fasting plasma glucose, ranging from 32 mg/dL with a 20-mg dose to 63 mg/dL with 80 mg, from a baseline of 180-193 mg/dL. Metformin reduced FPG by 37 mg/dL, and placebo by just 2 mg/dL from baseline. For HbA_1c, reductions at 12 weeks ranged from 0.6 to 1.5 percentage points, compared with 0.8 percentage points with metformin and 0.5 with placebo.

The current study evaluated 40-mg/day MK-0893 in combination with 2,000-mg/day metformin or 100-mg/day sitagliptin, as well as 100 mg of sitagliptin plus 2,000 mg of metformin. The 146 patients had a mean age of 53 years, HbA_1c of 8.6%, and mean diabetes duration of 7 years. The MK-0893/metformin combination was superior to sitagliptin/metformin in lowering 24-hour weighted mean glucose, with a reduction of 117 mg/dL compared with 85 mg/dL. However, MK-0893/sitagliptin was significantly less effective than sitagliptin/metformin, which produced a 24-hour WMG reduction of 100 mg/dL. This could be because sitagliptin, a DPP-4 inhibitor, also suppresses glucagon and so there could be a threshold effect, Dr. Engel noted.

All treatments were equally well tolerated. However, there was a significantly higher incidence of diarrhea in the two MK-0893 groups (10% vs. 0%). In addition, the liver enzymes alanine transaminase (ALT) and aspartate aminotransferase (AST) were elevated in the MK-0893/metformin group, and total cholesterol and LDL-cholesterol were increased from baseline with MK-0893/sitagliptin, relative to reductions with MK-0893/metformin and sitagliptin/metformin, he reported.

In the phase I dose-ranging study of LY-2409021, 47 patients were randomized to one of four doses or metformin. Patients had a mean baseline fasting blood glucose of 148 mg/dL, and a baseline HbA_1c of 8.0%. By day 28, mean reductions in HbA_1c were statistically significant compared with baseline in all treatment groups, ranging from 1.02 percentage points with 60 mg to 0.69 percentage points with 5 mg. In the placebo group, HbA_1c dropped by 0.49 percentage points, Dr. Prince reported.

Fasting glucagon significantly increased by 0.6- to 4.2-fold compared with baseline across the LY dose levels, and fasting active glucagon-like peptide-1 (GLP-1) rose by 59% at the highest dose. Glucagon and GLP-1 returned to baseline levels during follow-up, he said.

The agent was generally well tolerated. Reversible elevations in hepatic transaminases were seen in five of nine patients in the highest-dose group, with no clinical signs or significant elevations in bilirubin or alkaline phosphatase, he reported. In the interview, Dr. O’Harte said that these two agents seem to have already exceeded expectations based on previous experience with other small-molecule glucagon antagonists that have been tested and have failed due to toxicity issues. "A lot of these small-molecule drugs have fallen by the wayside, and these two are still in there. We need toxicity studies, but there’s a possibility for a breakthrough, I hope."

Dr. Engel is an employee of Merck, and Dr. Prince is an employee of Eli Lilly. Dr. O’Harte stated that he had no disclosures.

LISBON – New data on two investigational glucagon receptor antagonists suggest that this novel class of drugs has potential to be an effective treatment for type 2 diabetes, but only if safety can be established, according to two presentations at the annual meeting of the European Society for the Study of Diabetes.

Both Merck’s MK-0893 and Eli Lilly’s LY2409021 work by suppressing glucagon action via receptor blockade. Dr. Samuel S. Engel, senior director of clinical research, diabetes and endocrinology, at Merck presented phase II study data on the efficacy and safety of MK-0893 in combination with metformin or sitagliptin in 146 patients with type 2 diabetes, and Dr. Melvin J. Prince, senior director of medical diabetes and endocrinology at Eli Lilly, presented phase I dose-finding data on LY2409021 in 47 patients.

Both agents produced statistically significant reductions in hemoglobin A_1c. However, both also produced small but significant elevations in liver enzymes, and both carry the theoretical potential for hypoglycemia and hyperglucagonemia. "I think there is definitely potential for the glucagon receptor antagonists in diabetes, but there is also a risk of hypoglycemia associated with them. Also, it has been noted in the past that the hyperglucagonemia that results when the alpha cells overproduce glucagon has led to pancreas tumors in animal studies, so that’s a risk," session moderator Dr. Finbarr O’Harte, professor of endocrinology and metabolism at the University of Ulster, Derry (Northern Ireland), said in an interview.

In June at the American Diabetes Association’s annual scientific sessions, Dr. Engel presented data from a phase II study in which 342 type 2 diabetes patients were randomized to once-daily MK-0893 in four different dosages, metformin 1,000 mg twice daily, or placebo. At 12 weeks, treatment with MK-0893 resulted in significant, dose-dependent reductions in fasting plasma glucose, ranging from 32 mg/dL with a 20-mg dose to 63 mg/dL with 80 mg, from a baseline of 180-193 mg/dL. Metformin reduced FPG by 37 mg/dL, and placebo by just 2 mg/dL from baseline. For HbA_1c, reductions at 12 weeks ranged from 0.6 to 1.5 percentage points, compared with 0.8 percentage points with metformin and 0.5 with placebo.

The current study evaluated 40-mg/day MK-0893 in combination with 2,000-mg/day metformin or 100-mg/day sitagliptin, as well as 100 mg of sitagliptin plus 2,000 mg of metformin. The 146 patients had a mean age of 53 years, HbA_1c of 8.6%, and mean diabetes duration of 7 years. The MK-0893/metformin combination was superior to sitagliptin/metformin in lowering 24-hour weighted mean glucose, with a reduction of 117 mg/dL compared with 85 mg/dL. However, MK-0893/sitagliptin was significantly less effective than sitagliptin/metformin, which produced a 24-hour WMG reduction of 100 mg/dL. This could be because sitagliptin, a DPP-4 inhibitor, also suppresses glucagon and so there could be a threshold effect, Dr. Engel noted.

All treatments were equally well tolerated. However, there was a significantly higher incidence of diarrhea in the two MK-0893 groups (10% vs. 0%). In addition, the liver enzymes alanine transaminase (ALT) and aspartate aminotransferase (AST) were elevated in the MK-0893/metformin group, and total cholesterol and LDL-cholesterol were increased from baseline with MK-0893/sitagliptin, relative to reductions with MK-0893/metformin and sitagliptin/metformin, he reported.

In the phase I dose-ranging study of LY-2409021, 47 patients were randomized to one of four doses or metformin. Patients had a mean baseline fasting blood glucose of 148 mg/dL, and a baseline HbA_1c of 8.0%. By day 28, mean reductions in HbA_1c were statistically significant compared with baseline in all treatment groups, ranging from 1.02 percentage points with 60 mg to 0.69 percentage points with 5 mg. In the placebo group, HbA_1c dropped by 0.49 percentage points, Dr. Prince reported.

Fasting glucagon significantly increased by 0.6- to 4.2-fold compared with baseline across the LY dose levels, and fasting active glucagon-like peptide-1 (GLP-1) rose by 59% at the highest dose. Glucagon and GLP-1 returned to baseline levels during follow-up, he said.

The agent was generally well tolerated. Reversible elevations in hepatic transaminases were seen in five of nine patients in the highest-dose group, with no clinical signs or significant elevations in bilirubin or alkaline phosphatase, he reported. In the interview, Dr. O’Harte said that these two agents seem to have already exceeded expectations based on previous experience with other small-molecule glucagon antagonists that have been tested and have failed due to toxicity issues. "A lot of these small-molecule drugs have fallen by the wayside, and these two are still in there. We need toxicity studies, but there’s a possibility for a breakthrough, I hope."

Dr. Engel is an employee of Merck, and Dr. Prince is an employee of Eli Lilly. Dr. O’Harte stated that he had no disclosures.