This article is more than 5 years old. Information may no longer be current.
CONCLUDE: ‘Uncertainties’ complicate hypoglycemia findings comparing insulin degludec vs. glargine
Click Here to Manage Email Alerts
Adults with insulin-dependent type 2 diabetes assigned to insulin degludec for up to 88 weeks experienced fewer severe or confirmed symptomatic hypoglycemia episodes during a maintenance phase when compared with similar adults assigned to insulin glargine 300 u/mL; however, researchers cautioned that several factors complicate the data, according to analyses from the CONCLUDE trial presented at the European Association for the Study of Diabetes annual meeting.
In presenting findings from CONCLUDE, an open-label, randomized, treat-to-target study designed to assess the efficacy and safety of insulin degludec 200 U/mL (Tresiba, Novo Nordisk) vs. insulin glargine 300 U/mL (Toujeo, Sanofi) in type 2 diabetes, researchers noted that there are differences in the pharmacokinetics and pharmacodynamics between the two insulins that can make interpreting any findings difficult. Additionally, researchers discovered inconsistencies between participants’ blood glucose meters and the meters supplied by investigators during routine medical monitoring, leading to a post-amendment trial design that added on a variable maintenance period during which participants were switched to new glucose meters and paper-based diaries.
“There is no doubt that this was a difficult study to interpret,” Athena Philis-Tsimikas, MD, corporate vice president of the Scripps Whittier Diabetes Institute at Scripps Health and director of community engagement at Scripps Research Translational Institute in San Diego, said during a presentation. “The [trial] outcome was negative, and therefore, the remaining outcomes become hypothesis-generating, but we looked to the literature and found that ... consistency of the data across endpoints should be considered and the clinical interpretation of any trial should include the totality of the evidence.”
Study design
Researchers analyzed data from 1,609 adults with longstanding type 2 diabetes with an HbA1c of 9.5% or lower, prescribed basal insulin with or without oral diabetes agents (excluding sulfonylureas) and at increased risk for hypoglycemia, defined as a confirmed hypoglycemic event within 12 weeks of study enrollment (mean age, 63 years; 86% white; mean diabetes duration, 15 years, mean basal insulin use, 3.9 years). Researchers randomly assigned participants to daily insulin degludec 200 U/mL (n = 805) or daily insulin glargine 300 U/mL (n = 804), each titrated once weekly. Trial design initially included a 16-week titration period followed by a 36-week maintenance phase, later amended to include a variable maintenance phase for a total treatment period of up to 88 weeks.
Primary outcome was the number of severe or blood glucose-confirmed symptomatic hypoglycemic episodes during the maintenance period. Secondary outcomes were basal insulin dose at the end of treatment, number of nocturnal symptomatic hypoglycemic episodes during maintenance phase, number of symptomatic hypoglycemic episodes during total treatment period and the number of severe hypoglycemic episodes during the maintenance period.
Hypoglycemia findings
For overall symptomatic hypoglycemic episodes during the maintenance period, insulin degludec did not achieve the primary endpoint, Philis-Tsimikas said, with an RR of 0.88 (95% CI, 0.73-1.06). When the primary endpoint was not met, testing hierarchy stopped.
“Although the testing hierarchy is stopped, all statistical analyses specified in the protocol are still prespecified and can be analyzed,” Philis-Tsimikas said. “All the endpoints were still tested statistically outside of the hierarchical testing procedure, but no confirmatory results can be made from CONCLUDE from this point on.”
For the remaining prespecified endpoints, the RR for nocturnal hypoglycemia during the maintenance phase was 0.63 for insulin degludec vs. insulin glargine (95% CI, 0.48-0.84). The RR for severe hypoglycemic events during the maintenance phase was 0.2 (95% CI, 0.07-0.57) for insulin degludec vs. insulin glargine. For the total treatment period, participants assigned to insulin degludec experienced fewer overall symptomatic hypoglycemic events vs. patients assigned to insulin glargine (46.8% vs. 55.3%), as well as fewer nocturnal hypoglycemic events (20.2% vs. 28.9%) and fewer severe hypoglycemic events (2.2% vs. 5.5%), Philis-Tsimikas said.
“The remaining prespecified endpoints are all favoring degludec, but all [are] hypothesis-generating at this point,” Philis-Tsimikas said.
A post hoc analysis restricted only to participants who experienced hypoglycemic events during the study also favored insulin degludec, Philis-Tsimikas said, noting those findings were also hypothesis-generating.
"The good news is that both of these newer insulins have less hypoglycemia than all the current first-generation analogue insulins and therefore provide better protection against dangerous episodes of hypoglycemia," Philis-Tsimikas told Endocrine Today. "Insulin degludec has the added benefit of requiring lower doses to achieve its glucose control, so there may be some benefit in cost, although that will really depend on any price negotiations with health plans."
Interpret with caution
In commentary after the study findings were presented, Stefano Del Prato, MD, professor of endocrinology and metabolism at the School of Medicine, University of Pisa and chief of the section of diabetes at the University Hospital of Pisa, Italy, said there is much uncertainty with respect to the data.
“You would expect a CONCLUDE trial to be conclusive in its conclusions,” Del Prato said to audience laughter. “What I am trying to show is why it is not that conclusive.”
CONCLUDE data demonstrated that the primary endpoint was not significant, Del Prato said, with results suggesting similar, comparable hypoglycemia risk for the two insulins. Yet, there are differences in the pharmacokinetics and pharmacodynamics between the two insulins, he said, adding that the day-to-day glycemic variation is much lower for insulin degludec 200 U/mL vs. insulin glargine 300 U/mL.
“I warn you to interpret this data with a lot of caution,” Del Prato said. “As another expert pointed out, calculating pharmacokinetics and pharmacodynamics for insulin can be quite tricky and quite difficult for a number of reasons. ... I would suggest that how insulin degludec can account for lower risk for hypoglycemia remains uncertain.”
The protocol amendment and the inclusion of an additional maintenance period added more complexity to the trial, Del Prato said.
“There is one very important lesson of CONCLUDE,” Del Prato said. “Remember, CONCLUDE was very unfortunate in that there was a problem with the meters. If something like this can happen under the conditions of a randomized clinical trial, I am asking, what can happen in real life? There is a lot of variation [between blood glucose meters]. How much is this going to conceal the potential effect of treatment? What is the commitment of a patient in terms of using their meter?
“My conclusion is that too many uncertainties prevent us from drawing firm conclusions in the CONCLUDE trial,” Del Prato said.
Del Prato noted that understanding the differentiation among new basal insulin analogues may require better characterization of pharmacokinetics and pharmacodynamics, independent research and careful, real-world studies.
“Remember that translating potential benefit cannot just rely on the properties of the molecule,” Del Prato said. “Rather, it requires validation over time, the accuracy of glucose meters and adequate patient education and education reinforcement.” – by Regina Schaffer
Reference:
Pieber TR, et al. CONCLUDE: A trial comparing the efficacy and safety of insulin degludec and insulin glargine 300 units/mL in subjects with type 2 diabetes mellitus inadequately treated with basal insulin and oral antidiabetic drugs. Presented at: European Association for the Study of Diabetes Annual Meeting; Sept. 16-20, 2019; Barcelona, Spain.
Disclosures: Del Prato reports he has received research support from AstraZeneca, Boehringer Ingelheim and Merck, and serves on advisory boards for Abbott, AstraZeneca, Boehringer Ingelheim, Eli Lilly, GlaxoSmithKline, Laboratories Servier, Merck, Mundipharma, Novartis, Novo Nordisk, Sanofi and Takeda. Philis-Tsimikas reports she has received research and consultant fees from Dexcom, Eli Lilly, Merck, Novo Nordisk and Sanofi.
Editor's Note: This first paragraph of this article was updated on Sept. 23, 2019, to indicate that participants in the insulin degludec group experienced fewer severe or confirmed symptomatic hypoglycemia episodes than those assigned to insulin glargine. The Editors regret the error.