Protective Effect of Pioglitazone on Diabetes Risk Is Reduced After Treatment Cessation
Commentary by senior author Ralph A. DeFronzo, MD and Leigh Perreault, MD
The protective effect of pioglitazone on risk for type 2 diabetes is reduced following treatment cessation, according to an extension of the ACT NOW trial. The findings were published by Tripathy et al in the May issue of The Journal of Clinical Endocrinology and Metabolism.
“In the ACT NOW study,1 pioglitazone decreased the conversion rate of impaired glucose tolerance (IGT) diabetes by 72%, and 48% of IGT subjects reverted to normal glucose tolerance,” said senior author Ralph A. DeFronzo, MD, Chief of the Diabetes Division, University of Texas Health Science Center and Deputy Director, Texas Diabetes Institute, San Antonio, Texas. “These are fantastic results.”
“However, following cessation of pioglitazone individuals with IGT started to convert to diabetes and, after one year, the conversion rate of IGT to diabetes was similar to that in the placebo group,” Dr. DeFronzo said. “This emphasizes that diabetes is a genetic disorder and requires continuous life-long therapy, with both lifestyle intervention and antidiabetic medications. The results also emphasize that ‘prediabetes’ is part of a continuum of diabetes and should be viewed as such.”
Patients Followed For a Median of 1 Year After Treatment Cessation
The present study followed 290 participants in the ACT NOW study who returned for at least one 6-month follow-up visit after cessation of pioglitazone (n=152) or placebo (n=138). The median follow-up after stopping the study mediation was 11.4 months.
Diabetes developed in 11.2% and 12.3% of the pioglitazone and placebo groups, respectively (P=NS), during the follow-up period. However, the cumulative incidence of diabetes from when patients started their study medication to the end of the follow-up period was significantly lower in the pioglitazone group than in the placebo group (10.7 vs 22.3%; P<0.005).
In addition, among patients who reverted to normal-glucose tolerance (NGT) during the study, patients who received pioglitazone were significantly more likely to remain at NGT at 11.4 months after treatment cessation (23.0% vs 13.8%; P=0.04). Low insulin secretion/insulin resistance index was determined to be a strong predictor of diabetes development following pioglitazone discontinuation, the study authors noted.
Preventive Strategies May Delay Rather Than Prevent Diabetes, Expert Says
“ACT NOW, similar to the Diabetes Prevention Program (DPP),2 demonstrated the ability of an intervention to delay the onset of diabetes,” commented Leigh Perreault, MD, who is a coinvestigator of the DPP. “The intention-to-treat analyses in both cases continue to show a delay in diabetes onset after cessation of the interventions, but the incidence rates are no longer different. Because the natural history of diabetes development was unaffected post-treatment, this implies that diabetes was delayed rather than truly prevented,” said Dr. Perreault, who also is Associate Professor of Medicine at the University of Colorado Anschutz Medical Campus, and Associate Professor of Epidemiology at the Colorado School of Public Health, both in Aurora, CO.
“Nevertheless, delay of diabetes—with all of the human and economic cost—is highly desirable,” Dr. Perreault continued. “Further, regression from prediabetes to normoglycemia was nearly twice as high for those who had been treated with pioglitazone nearly 1 year after the medication was stopped. This may signify a subpopulation in whom pioglitazone could have disproportionate benefit.”
Pioglitazone and Bladder Cancer Risk
Prescribers may need to consider the conflicting reports of an increased risk bladder cancer with pioglitazone when weighing the decision to use this medication. In one recent study involving data from more than 145,000 people who were newly treated with diabetes medications, pioglitazone was associated with an overall 63% increased risk of bladder cancer (121 per 100,000 person years vs 89 per 100,000 person years in patients not treated with thiazolidinediones), with the risk increasing with longer duration of use and higher dose.3
“Indeed, there have been conflicting reports on the link between thiazolidinediones and bladder cancer,” Dr. Perreault told EndocrineWeb. “Altogether, the association is uncommon and highly speculative. The decision to use pioglitazone for diabetes prevention should weigh the risk versus benefits of doing so, including consideration of bladder cancer, fracture, and weight gain. The ultimate decision to use medication to prevent/delay diabetes always comes down to a conversation between the patient and provider.”
Dr. DeFronzo pointed out that the large FDA-mandated study funded by Takeda showed no increased risk of bladder cancer in pioglitazone-treated subjects.4 In addition, “in a large cohort of 1.01 million individuals (over 5.9 million person-years) in 6 populations with reporting for all malignancies, the hazard ratio for bladder cancer with rosiglitazone was 1.01 in men and 1.00 in women and with pioglitazone was 1.01 in men and 1.04 in women,” Dr. DeFronzo said.5
Pioglitazone May Prevent Atherosclerosis
Dr. DeFronzo said that the risk:benefit assessment should also include the potential beneficial effect of pioglitazone on lowering the risk for atherosclerotic cardiovascular complications as demonstrated in 3 trials.1,6,7
“In the ACT NOW Study,1 pioglitazone reduced the progression of carotid intima media thickness (a surrogate measure of atherosclerosis) by 31.5% (P=0.047),” Dr. DeFronzo explained. “In PROactive,6 pioglitazone reduced the second principal MACE (Major Atherosclerotic Cardiovascular Events) endpoint of cardiovascular death, nonfatal MI [myocardial infarction], nonfatal stroke by 16% (P=0.027) in type 2 diabetic patients with a previous cardiovascular event; recurrent stroke was decreased by 47% (P=0.009). In the recent IRIS (Insulin Resistance Intervention After Stroke) study,7 carried out in patients without diabetes who had a recent stroke or TIA [transient ischemic attack], recurrent stroke or myocardial infarction was reduced by 24% (P=0.007). These results have important implications for the prevention of atherosclerotic cardiovascular complications.”
Dr. DeFronzo recommends initiating therapy at 15 mg/day and increasing the dose to 30 mg/day as necessary. “Rarely, should physicians exceed a dose of 30 mg/day since most of the efficacy is obtained with this dose and the incidence of side effects increases significantly at the 45 mg/day dose,” he said.
A Change in Thinking Toward Treatment of Prediabetes
“Currently, no medication has an FDA indication for the prevention of diabetes,” Dr. Perreault concluded. “In the 2008 FDA’s Guidance to Industry, the FDA stated that any medication aiming for such an indication must demonstrate an ability to change the natural history of the disease. Frankly, I find this nonsensical, as the FDA-approved medications we use for hypertension, dyslipidemia, and diabetes itself do not change the diseases we treat, but have powerful effects to lower morbidity and mortality anyway. There is increasing evidence that the same is true in the treatment of prediabetes. Hence, it may well be time to change our thinking toward treatment of prediabetes rather than prevention of diabetes.”
June 9, 2016