Patients on metformin, pioglitazone, and exenatide showed greater and more durable reductions in HbA1c, less hypoglycemia, and less weight gain compared with conventional, stepwise add-on treatment with metformin, a sulfonylurea (glyburide), and basal insulin (glargine), in a new study reported at American Diabetes Association (ADA) 2013 Scientific Sessions.
The results of this so-called "triple therapy" were hailed as potentially revolutionary and possibly "paradigm shifting," although questions remain.
Muhammad A. Abdul-Ghani, MD, from the University of Texas Health Science Center at San Antonio, presented the findings. He stated that, "We think this might be clinically meaningful and, hopefully, in a larger study, it might show differences in the risk of microvascular complications, and that definitely will be [practice] changing, in terms of guidelines and the approach to treating patients."
With conventional therapy for diabetes, where individual drugs are added 1 at a time and a second is only started when it is deemed needed, "the HbA1c starts to slide over time," Dr. Abdul-Ghani observed. "But with the triple therapy, at 6 months and at 2 years, our patients had exactly the same HbA1c."
Those patients who have reached 3 years of therapy in the study, which is ongoing, have maintained the same HbA1c, "and we intend to extend the study beyond 3 years," he noted. "So far, there is no other therapy that has shown such a durable and stable reduction in HbA1c," but how long this will continue to be the case remains to be seen, he acknowledged.
Asked to comment on the findings, Lawrence A. Leiter, MD, professor in the departments of medicine and nutritional sciences at the University of Toronto, Ontario, said: "These are very provocative results. The fact that they were able to achieve a mean HbA1c of 6%, which was sustained over the duration of the trial, was very impressive. I think one might quibble with the comparators that were used, but nonetheless it certainly is a paradigm shift. I think there are other studies under way with similar kinds of design that will confirm it, but certainly we've long thought that initial combination therapy is good, and some of our guidelines would recommend initial combination therapy in patients with high baseline HbA1c. What they did, of course, was initiate triple therapy in everyone, so that is the paradigm shift. But it makes sense."
Carol Wysham, MD, clinical associate professor of medicine at the University of Washington School of Medicine, Seattle, stated, "I'm very excited by this, by any concept that will help to spare beta cells and that has extreme durability and will keep patients stable for longer.
"I've heard [senior author] Dr. [Ralph A.] DeFronzo [University of Texas Health Science Center at San Antonio] talking about this for at least 5 years, so I'm happy to see the data and it's not unexpected. We have to try to change the paradigm of how we treat diabetes, and when Dr. DeFronzo first presented this I loved the analogy he used -- that we don't treat breast cancer with 1 drug and wait for it to fail and hope we have a good outcome. So I think we really do have to look at more aggressive treatment."
However, this will take a long time to filter down, she stressed. "First of all, you can't take any single study of 100 patients and extrapolate that. I think it's going to be like anything else in medicine: it's going to be 10 years before it does get into common practice."
John Buse, MD, PhD, from the University of North Carolina, Chapel Hill, was a little more circumspect. "In my mind, the good news is that both approaches work quite well. The preconceived notion was that 3 drugs started at the same time would do very well, and they did exceptionally well, with essentially no weight gain, no hypoglycemia." But the control arm also performed well, he stressed.
And he agrees with Dr. Leiter that "the comparator drugs could have been better. People are not enthusiastic about glyburide -- many people think it is the sulfonylurea that no one should ever use, although I'm not sure if that's really true. And many people believe that you should never combine sulfonylureas with insulin, and there is some pretty good data to suggest that this may be true. I think when Dr. DeFronzo proposed doing this study, he was trying to make a point, so he didn't tee up the other arm as artfully as some might, but at the same time it's a fair comparison.
"To me, it's almost self-evident that if you use 3 drugs that work in different ways to control glucose, it's going to be better than using metformin and insulin. But is the difference enough to say that metformin followed by sulfonylurea followed by insulin is malpractice, that you should never do this? No, I think that sort of black-and-white look at things is inappropriate."
The results reported by Dr. Abdul-Ghani today were interim and came from an open-label, randomized controlled trial in 155 drug-naive, newly diagnosed (less than 2 years) type 2 diabetes patients. The average duration since diabetes diagnosis was 5 months, mean HbA1c was 8.6%, average age was 46 years, and body mass index (BMI) 36 to 37.
Patients were randomized to 1 of 2 groups: triple therapy starting with 1000 mg/day of metformin, 15 mg/day of pioglitazone, and 5 µg twice daily of exenatide. These doses could be titrated up to 2000 mg, 30 mg, and 10 µg, respectively for each agent, by the end of month 1.
Those randomly assigned to conventional therapy started on metformin 1000 mg/day in month 1, increased to a maximum of 2000 mg/day. If fasting plasma glucose (FPG) exceeded 100 mg/dL, then glyburide 5 mg/day was started. In month 2, glyburide was increased to 10 mg/day if FPG exceeded 100 mg/dL or HbA1c was greater than 6.5%. In month 3, glargine insulin (10 U/day) was started if FPG was greater than 100 mg/dL or HbA1c greater than 6.5%. The glargine dose was increased weekly by 10 units/day for FPG greater than 100 mg/dL.
After the first 3 months, subjects were seen in the clinic every 3 months and their FPG, HbA1c, weight, and home blood glucose monitoring were recorded. Medications could be adjusted down for blood glucose less than 60 mg/day or symptoms of hypoglycemia.
Treatment failure was defined as HbA1c greater than 6.5% on 2 consecutive visits 3 months apart, despite maximal therapy. Rescue therapy for the conventional treatment arm was rapid-acting insulin (Humalog, Lilly), 4 to 6 units before meals, with dose escalation to maintain 2-hour postprandial glucose at less than 140 mg/dL. For the triple-therapy group, rescue therapy was glargine insulin, 10 units in the morning with dose escalation to maintain FPG less than 110 mg/dL and HbA1c less than 6.5%, with a maximum dose of 60 units/per day before starting Humalog.
The primary end point of the study was difference in HbA1c between the triple-therapy and conventional-therapy groups. Secondary end points included time to treatment failure, change in body weight, rate of hypoglycemic events, and decrease in FPG and postprandial plasma glucose concentrations.
At the interim analysis, at 2 years, those in the triple-therapy group had a mean HbA1c of 6.0% compared with 6.6% in those on conventional treatment (P < .001).The curves started to diverge at 6 months, said Dr. Abdul-Ghani.
In the triple-therapy group, 60% had HbA1c of less than 6.0% after 2 years, compared with 27% of those in the conventional group (P < .001). For HbA1c less than 7.0%, these figures were 92% and 72%, respectively (P < .001).
Those in the triple-therapy group lost 1.2 kg, on average, compared with a weight gain in those on conventional treatment (P < .001).
Of those on conventional therapy, 42% were treatment failures by 2 years, compared with 17% in the triple-therapy group (P < .001). Hypoglycemia occurred in 46% of those in conventional treatment compared with 15% in triple therapy; there were no cases of severe hypoglycemia. Edema occurred more frequently, however, in the triple-therapy group (5.3% vs 1.3%), as did gastrointestinal side effects (33% vs 21%).
"We were able to normalize the HbA1c in 60% of the people [on triple therapy], without much hypoglycemia, which we know has a very big adverse effect," said Dr. Abdul-Ghani. "We actually have weight loss, on average [in the triple-therapy group], despite having a maximal or near-maximal dose of pioglitazone, because of the combination with the GLP-1 [agonist], which mitigates the weight gain. That doesn't mean that every patient lost weight, but on average we have a decrease."
And although there was a greater rate of peripheral edema in the people receiving pioglitazone, the absolute incidence was relatively small compared with other studies, he says. "I think that because GLP-1 agonists have diuretic effects, this mitigates part of the fluid-retention side effects that are associated with pioglitazone. So also in terms of side effects, we believe this combination is beneficial.
"The metformin and pioglitazone are 1 pill. [Now, with the advent of extended-release exenatide, you could] have a once-weekly GLP-1 [agonist], so you have 1 injection per week, 1 pill a day.
"Very simple, very effective. And metformin and pioglitazone are already generic, so they are cheap and hopefully soon the GLP-1 agonists [will fall in price]."
He stressed that this study is ongoing, and that this is only an in-treatment analysis. "It is planned to continue, so hopefully next time we will report the 3-year results."