New ORIGIN Data Upstaged at EASD, as Discussant Warns of 'Trap'

Shelley Wood

October 09, 2012

October 8, 2012 (Berlin, Germany) — An investigator for the Outcome Reduction with an Initial Glargine Intervention (ORIGIN) trial presenting new subgroup data last week says the numbers show that "target-directed" interventions can keep hemoglobin A1c (HbA1c) levels within normal range for at least five years in both diabetic and nondiabetic patients. Moreover, said Dr Mathew Riddle (Oregon Health and Science University, Portland), maintenance of HbA1c levels <6.5% out to five years was independently associated with a lack of diabetes, lower baseline A1c, and use of a regimen that included daily injections of insulin glargine (Lantus, Sanofi).

But the upbeat tone of the European Association for the Study of Diabetes 2012 Annual Meeting session took an abrupt turn when the scheduled discussant for Riddle's presentation took the stage. In his talk, Dr Thomas Pieber (Medical University of Graz, Austria) took aim at what he called an inexplicable twofold higher risk of death in the ORIGINS trial as compared with other diabetes trials looking at cardiovascular outcomes.

"This truly and urgently needs additional analyses to understand why that is," Pieber said. "Insulin glargine causes a threefold increase in severe hypoglycemia--something I am taking as an important issue--and insulin glargine given in patients with impaired fasting glucose [IFG] cannot prevent type 2 diabetes. Please don't fall into this trap that they are showing us, that you can prevent diabetes with insulin."

Back to ORIGIN

The ORIGIN trial enrolled both diabetic patients and a smaller group of nondiabetics (12% of patients in ORIGIN) who had IFG or impaired glucose tolerance (IGT) at baseline. As previously reported by heartwire , the trial was negative, showing no impact of either omega-3 fatty acids or insulin glargine in reducing a composite end point of MI, stroke, or cardiovascular death.

In his presentation, Riddle noted that less than 50% of diabetic patients had HbA1c levels under 6.5% at baseline, but that this rose to 60% at five years among those randomized to insulin glargine (and dropped to 45% among those on standard therapy). In the no-diabetes group, 91% of patients had baseline A1c levels <6.5%, and this dipped only slightly, to 87% at five years in those randomized to daily insulin injections and to 79% among those in the standard-therapy group.

In other long-term studies of diabetics or people at risk of diabetes, A1c levels tend to rise over time. In ORIGIN, said Riddle, "our treatment algorithms prevented this rise."

These findings, he continued, point to "a moment of opportunity in the natural history of dysglycemia when intervention is relatively easy and can be relatively effective . . . with relatively limited adverse consequences. Second, the glargine-based regimen was able to do this."

So far, he added, investigators have not yet looked at whether levels of glucose control led "to either favorable or unfavorable medical consequences, and clearly further analyses are needed to do this."

Pieber Unconvinced

In his rousing counterpoint, however, Pieber suggested a simpler trial design that could have produced the same results. "Do this study as it has been done but instead of giving patients who don't have diabetes daily insulin injections for six years, with the risk of hypoglycemia, just leave them alone, because they don't have diabetes at this moment (and remember they don't get any benefit from this intervention). Then, at the end of the study, when you retest them, give them four units of rapid-acting insulin just before you do the test. This is just one injection in six years, and you might have the same lowering of glucose that you see here."

Pieber calculated that the overall mortality rate per year in ORIGIN was 2.57%, as compared with rates of 1.27% for ADDITION, 1.41% for PROactive and ACCORD, and 1.69% for ADVANCE. This higher mortality rate in ORIGIN, he added, could not be explained by patient risk factors, history of diabetes or CVD, or patient age.

Pieber also argued that the threefold increase in both nonsevere hypoglycemia and in severe hypoglycemia needs to be taken seriously. Endocrine agents are the second most common cause of emergency-room visits, Pieber reminded the audience, and almost 95% of endocrine emergencies are due to hypoglycemia. In this trial, 25 patients had to be treated with insulin glargine to cause one case of severe hypoglycemia, and 15 patients had to be treated to "prevent" one progression to diabetes. Yet 14 patients treated with insulin glargine for six years went on to develop type 2 diabetes anyhow.

Wrapping up his talk, Pieber quoted Berthold Brecht, saying, "The chief cause of poverty in science is imaginary wealth. The chief aim of science is not to open a door of infinite wisdom but to set a limit to infinite error."

Keeping the Findings in Context

Speaking with heartwire after the ORIGIN session, session moderator Dr Rury R Holman (Oxford Centre for Diabetes, Endocrinology, and Metabolism, UK) downplayed some of Pieber's comments, saying, "I think it's a little unfair to take the trial out of context."

The trial, he pointed out, was designed to test whether replacing insulin in an insulin-deficiency disease would bring benefit in terms of cardiovascular outcomes and to look for possible safety issues such as cancer and hypoglycemia. "So the fact that over the duration of the trial, with or without glargine, investigators could maintain near normal glucoses, is a fantastic testament to what can be achieved," Holman said. "And the fact that there were no additional cancer or CV deaths is extremely reassuring to all patients who get treated with insulin, because this is a longstanding fear."

Insulin-treated patients did have more hypoglycemic events and slight weight gain, he noted, but both increases were relatively small compared with the standard-therapy group.

"No one here in ORIGIN is suggesting we go out and give everyone insulin. They set up an experiment to test whether it would reduce CVD; it doesn't, but they've also shown that with modest doses you can maintain glucose levels, and we now know the extent of the risk for hypoglycemia."

Asked whether he would give insulin glargine to patients with impaired glucose tolerance or impaired fasting glucose, Holman was emphatic: no.

"People with IFG or prediabetes do not have a diagnosed condition, so to give a medication off-label we would first need a trial in that group to show long-term benefit. . . . ORIGIN was an experimental study to investigate issues; it wasn't an intention to treat IFG patients. If there had been a major CV reduction we wouldn't be having this conversation. But now we know insulin glargine is safe and that there are modest adverse effects: we have the information to put it in context."

Full results from ORIGIN were first presented by Dr Hertzel Gerstein (McMaster University, Hamilton, ON) earlier this year at the American Diabetes Association (ADA) 2012 Scientific Sessions; insulin glargine and omega-3 results were simultaneously published in two separate papers in the New England Journal of Medicine [1,2].

The ORIGIN trial is funded by Sanofi. Riddle disclosed consulting for Amylin, Eli Lilly, GlaxoSmithKline, Hoffman-La Roche, Sanofi, and Valeritas. Pieber disclosed receiving research funding from the EU Commission, European Foundation for the Study of Diabetes, Austrian Research Agency, Novo Nordisk, Menarini, DeQur, Lilly, Novartis, Roche, Sanofi, and Abbott, as well as consultancy, speaking, or advisory-board fees from Novo Nordisk, Roche, AstraZeneca/Bristol-Myers Squibb, Lilly, and Skye-Pharma. Holman has disclosed receiving grant/research support and consulting fees/honoraria from Amylin, Bayer, Lilly, Merck, Novartis, and Novo Nordisk.