This article summarizes the Diabetes Diabetes Program/Diabetes Prevention Program Outcomes Study (DPP/DPPOS) the metformin study for the prevention of larger and longer diabetes.
The DPP was a randomized, controlled study of 3234 participants with high risk of diabetes.The participants were assigned to placebo or metformin 850 mg twice a day, or intensive intervention in life habits.The metformin/placebo phase was ended a year earlier than expected due to its effectiveness.
At the end of the DPP, the participants continued the follow -up in the DPPOS.Participants originally assigned to Metformin continued to receive it, without masking.The DPP/DPPOS cohort already has a follow -up of more than 15 years.At 2.8 years, metformin decreased the incidence of diabetes by 31% in relation to placebo.
The DPPOS study analyzed the long -term effects of metformin and showed a reduction in the risk of diabetes from 18% to 10 and 15 years of randomization.At 15 years, the lack of diabetes progression was associated with 28% of less risk of microvascular complications.Recent data suggest that metformin can reduce the development of atherosclerosis in men.
The prolonged monitoring, which continues in the DPP/DPPOS, is currently evaluating the possible importance of metformin, started early in the spectrum of dysgucemia, in the morbidity of the latest stages, including cardiovascular disease and cancer.
► Introduction
Diabetes Program - Diabetes Prevention Program- (DPP for the acronym for English; 1996–2001), is a randomized, controlled study whose objective wasOne of the interventions, based on its mechanism of action and its acceptable safety and tolerability.
The other branches of treatment were intervention in life habits or placebo [1, 2].The possibility of preventing diabetes or delaying its beginning in high -risk adults was raised for decades.Small randomized studies were carried out with phenformin or tolbutamide in the 1960/70 decades, but were not conclusive [3–5].Larger subsequent studies carried out interventions in life habits that were effective [6, 7].The DPP was the first important study of diabetes prevention with metformin [2].
The DPP/Diabetes Prevention Program outcomes Study- DPP/Study of results of the Diabetes Prevention Program- (DPPOS for the follows of English) is the largest controlled clinical study of metformin in a population with a high risk of contracting diabetes and also theMore prolonged metformin study for any indication.
This article focuses on the effects of metformin on diabetes prevention, its glycemic and prolonged cardiomethabolic effects and its safety in DPP/DPPOS.
► DPP/DPPOS panorama
♦ Study design
The DPP incorporated 3234 participants of 25 years or more at high risk of contracting diabetes, defined as glucose intolerance, with fasting hyperglycemia.(5.3–6.9 mmol/l) and an BMI of 24 kg/m2 or greater.The participants were randomized to receive placebo (n = 1082), metformin (n = 1073) 850 mg twice a day or intensive intervention in life habits (n = 1079), with the aim of obtaining a weight decrease of 7% through a hypocaloric diet, hypogger and ≥150 min/week of physical activity of moderate intensity [2].
The diagnosis of diabetes was based on an annual test of oral glucose tolerance and semi -proof of fasting blood glucose, with the use of the diagnostic criteria of the ADA, where the diagnosis should be confirmed with the repetition of the test[11].Before the diagnosis of diabetes and fasting glycemia ≥7.8 mmol/l the study of the study was suspended and the participant referred to his doctor to continue the treatment [2].
The DPP ended in 2001, one year earlier than planned because the efficacy of both metformin and intervention in life habits [2] 88% (n = 2776) of the suitable participants were demonstratedDPP continued the follow -up in the DPPOS, in which the placebo was suspended, those previously assigned to Metformin received Metformin 850 mg twice a day (now without blind) and intermittently reinforced the messages about life habits.
The metformin provided by the study was suspended if diabetes was diagnosed and HBA1C was ≥7% (≥53 mmol/mol), therefore the treatment by the participant's doctor [8] was necessary.The results in the DPPOS from 2002 to 2013 focused on the long -term effects of the interventions for diabetes prevention, the microvascular complications associated with the same [9] and the risk factors of cardiovascular disease (ECV).
♦ Characteristics of participants
The DPP intentionally incorporated a heterogeneous population, with 45% of racial or ethnic minorities, 20% of 60 years or more and 68% of women, including 350 women with a history of gestational diabetes.The average age to randomization was 51 years and the average BMI was 34 kg/m2.Average medium glycemia average was 5.9 mmol/L and HBA1C at the beginning was 5.9% (41 mmol/mol).Sex, ethnic distribution and diabetes risk factors were similar between the branches of random treatment [2].
♦ Exposure to metformin
In order to know the effects of metformin over time it is essentialmetformin that did not have the study.
At 15 years of randomization, 37% of the original participants who had received placebo had been treated with metformin by their doctors, the vast majority with the diagnosis of diabetes.The average exhibition, including the metformin of the study and that did not belong to it, from 1996 to 2013, continued to be very different: 10.7 vs. 2.3 years- metformin in metformin vs. placebo groups [9].
During the DPP, the therapeutic adhesion to metformin, (taking at least 80% of the tablets of the assigned study), were 72% [2].Another 10–15% of the participants took some metformin, although less than 80% of the assigned tablets.Adhesion to metformin descended to an average of 49% during DPPOS (2002–2013) [9].
► Results
♦ Effects of metformin on diabetes prevention
In 2002, the DPP published its main data of the masked treatment phase and showed that intensive intervention groups on life habits and metformin groups had lower incidence of diabetes than the placebo group (58% and 31%, respectively)[2].Then, the DPPOS studied the longest effects of metformin and showed mmenor effectiveness to reduce the risk in relation to the placebo for 10 and 15 years after randomization.[8, 9].
The incidence of diabetes during the DPP was 7.8 cases per 100 years-person in the metformin group and 11.0 cases per 100 years- person in the placebo group [2], and decreased in the DPPOS (2002–2008)to 4.9 cases per 100 years - person for metformin and 5.6 cases per 100 years - spell for placebo [8], stabilizing from there.This reduced incidence of diabetes approaches the five cases per 100 years-person observed in the group of life habits during the DPP, whichIt remained almost constant through the DPP/DPPOS [9].
The average genetic risk score, derived from 34 genetic variants associated with type 2 diabetes, decreased over time among the participants who were still without diabetes in the DPP/DPPOS, in both groups, the metformin group and the placebo group [13].This happens because the lower annual incidence of diabetes observed in the DPPOS was not totally due to the effect of intervention in life habits during the transition to DPPOS, but, in part, due to the 'exhaustion of the susceptibles', or that theDiabetes manifested himself in the most susceptible people to it during the DPP and that the remaining participants of the DPPOS were less susceptible to diabetes [13].
♦ Effects of metformin on the prevention or delay of diabetes in subgroups of interest
The DPP had no power to evaluate the significance of the effects on subgroups.However, the examination of the effects of treatment in subgroups of cohorts revealed significant heterogeneity.For example, obese participants with BMI ≥35 kg/m2 responded better to metformin than to placebo, with reduction of diabetes risk of 53%, but only 3% reduction in those with BMI of 22 A & lt; 30 kg/m2.
In addition, those with the greatest fasting blood glucose (6.1–6.9 mmol/l) had a greater reduction in metformin risk (48%) than those with fasting blood glucose of 5.3–6.1 mmol/l (15 (15% risk reduction).Metformin seemed to be more effective in younger participants in relation to placebo, reducing the start of diabetes by 44% in those of 25–44 years vs. 11% in the ≥60 years when entering the study.
During the DPP, women with a history of random gestational diabetes had a 71% higher risk of diabetes than women who had children without this background, despite similar values of fasting blood glucose and fasting blood glucose at 2 hoursat the beginning [14].
Significant heterogeneity is observed in response to metformin with 50% decrease in the incidence of diabetes in women with a history of gestational diabetes and 14% in women who had children without these background.Ten years of follow -up in the DPPOS confirmed these effects, demonstrating a sustained and relatively greater risk of diabetes in women with a history of gestational diabetes, which decreased by 40% with metformin [15].
Part of the preventive effect of metformin diabetes attributes weight descent, which was durable
♦ Knowledge of the DPP/DPPOS on how metformin prevents or delays diabetes
≈ ¿Acute pharmacological effect or improvement of pathophysiology?
During the DPP, the evaluations were carried out without interrupting the medication of the study (placebo or metformin), except in the morning of the blood glucose tests.Therefore, some or all the effect of metformin could have been a transitory effect of pharmacological treatment that masked diabetes, instead of a true delay in the beginning of diabetes.
The DPP considered that issue and repeated the evidence with the participants who had not contracted diabetes at the end of the study, 1–2 weeks after suspending metformin.After this period of pharmacological purification, the incidence of diabetes was reduced by 25%, compared to the 31% reduction observed in the main analysis, which suggests a more lasting effect of metformin treatment on glucose metabolism [16].
≈ Explanation of the effects induced by metformin Part of the preventive effect of metformin diabetes is attributed the descent of weight, which was durable in the DPP/ DPPOS [2, 8, 9].This decrease explained 64% of the favorable effect of metformin on the risk of diabetes at the end of the DPP [17].Favorable changes were observedAlso in other measurements such as the circumference of the waist and the hip waist index, and in insulin and proinsulin in fasting [17].
No differences in physical activity or diet reported by the patient or in insulin secretion measured by the insulinogenic index between the metformin and placebo groups were observed.The combination of weight decrease, insulin and proinsulin concentrations and other metabolic factors explained 81% of favorable results with metformin [17].
The improvements of fasting glycemia and the estimated sensitivity to insulin with metformin may be due to the combination of weight descent and other direct effects on the liver and, perhaps, other tissues.
≈ Effects of metformin on glycemia measurements The effects of DPP interventions on fasting glycemia and HBA1C were examined in all participants, with or without diabetes.During the DPP, metformin and intensive interventions in life habits were equally effective in restoring normal fasting glymia values [2].
Although both had similar effects on blood glucose, the incidence of diabetes was more significantly reduced by interventions in life habits than by metformin, reflecting that most diagnostics of diabetes in DPP were made by glycemiaAt 2 hours more than fasting glycemia and that interventions in life habits were more effective than metformin to restore postprandial blood glucose of 2 normal 2 hrs.
This last observation was likely because, although both interventions improved the function of beta cells, this effect was greater with interventions on life habits [18].According to the ability of the metformin to suppress the production of liver glucose during fasting [19, 20], its reduction in the incidence of diabetes in relation to the placebo was much greater in those who incorporated into the study with an empty blood glucoseof 6.1–6.9 mmol/l that in those with fasting glycemia of 5.3–6.1 mmol/l [2].Metformin also lowered HBA1C in relation to placebo, but to a lesser extent than interventions on life habits [2].
After the completion of the DPP, an International Specialist Committee and the ADA expanded the diagnostic criteria for diabetes and included HBA1C ≥ 6.5% (≥ 48 mmol/mol) [21, 22].Metformin was as effective as interventions on life habits to prevent diabetes according to HBA1C, but not according to postprandial blood glucose of 2 hrs in the population of the DPP/DPPOS.
♦ Metformin interaction with genetic factors The DPP investigated several genetic variants previously associated with the risk of type 2 diabetes or the action of metformin.For example, homozygosis for the main risk variant RS7903146 diabetes in the TCF7L2 gene was associated with an incidence of diabetes the 81% higher in the placebo group, which decreased to 62% in the metformin group [24].
♦ Effects of metformin on microvascular complications.At the end of the DPP, the only microvascular result evaluated was microalbuminuria.There was no effect of therapeutic intervention on the proportion of participants with an increase in the albumin index: creatinine (IAC), although those who contracted diabetes had a 59% increase in the risk of having increased IAC (≥3.39 mg/mmol) [27]
One of the main objectives of the longest follow -up of DPPOS is to determine whether effective treatments to prevent diabetes also affect the development of microvascular complications, specifically retinopathy, nephropathy and neuropathy.
The combination of these microvascular results at 15 years in DPPOS was 28%Less frequent in those who did not progress to diabetes, but there was no difference in the original branches of treatment [9].
The very low difference in HBA1C concentrations between the therapeutic groups their limited power and early derivation for the treatment of hypertension and dyslipidemia were considered the reasons for the lack of effect of treatments on microvascular results so far, despiteof the decrease in the incidence of diabetes [9].It is still possible that the effects of treatment can arise with greater monitoring and longer duration of diabetes in cohorts.
♦ Effects of metformin on cardiovascular disease risk factors
In the DPP, metformin had favorable effects on several cardiovascular risk factors, including lipoprotein subfractions [28] ,, C-reactive protein and tissue plasminogen activator [29].The incidence of metabolic syndrome by 17% in relation to placebo [30] also decreased.
No significant effects on lipid concentrations or blood pressure were observed [31]. During the longest follow -up (10 years), no significant differences were observed in the risk factors of cardiovascular disease (ECV) between metformin and placebo groups[32].
An average of 14 years after randomization was evaluated subclinical atherosclerosis in 2029 participants through calcium measurements of the coronary arteries (CAC), according to the original randomization group.There was a significant interaction between sex and the effects of Metformin vs Placebo on the presence of CAC (P = 0.01) and the severity of the CAC (P = 0.08).
In relation to placebo, metformin significantly decreased the presence and severity of the CAC in men, without effect in women.It is interesting that there was no decrease in the frequency of clinically significant plates, which suggests the possibility that metformin affects more small plates, with more recent calcification.There was no difference in the CAC between the intervention groups in life habits and placebo groups.This suggests a possible long -term differentiation between metformin and interventions in life habits [33].
♦ Safety and long -term tolerability of metformin in DPP/DPPOS
Current recommendations advise the periodic determination of vitamin B12 values and administration according to need
The prolonged use of metformin in a closely controlled clinical study provided more information about its safety and tolerability.9.5% of metformin randomized pacienes referred minor gastrointestinal symptoms, while 1.1% of assigned patients the placebo suffered those symptoms, which were usually mild and tended to decrease over time. [34].In the DPP/DPPOS, there were no cases of lactic acidosis with the use of metformin in more than 15000 years of exposure to metformin.
The use of metformin was associated with the decrease in the intestinal absorption of vitamin B12 and greater risk of deficiency of this vitamin.In the DPP annual tests were performed to detect anemia as a possible manifestation of low values of B12.In addition, vitamin B12 values were measured directly into two temporal points of DPPOS.
The biochemical deficiency of vitamin B12 (& lt; 150 pmol/l) was more often observed in patients in the metformin group than in the 5 -year placebo group (4.3%vs 2, .3%; p = 0,02);Similar characteristics were observed, but not significant to 13 years (7.4% vs 5.4%; p = 0.12) [36].Low or bordering vitamin B12 (& lt; 220 pmol/l) were accepted by some as evidence of depositsInsufficient vitamin B12 and was more frequent in those of the 5 -year metformin group than in the placebo group (19.1% vs 9.5%; p = 0.01) as well as 13 years (20.3% vs. vs.15.6%; P = 0.02).Current recommendations advise the periodic determination of vitamin B12 values and administration as needed in patients treated with metformin [37].
► Looking to the future
The impact of prediabetes and diabetes worldwide is huge, with 415 million adults with diabetes and the projection of 642 million in 2040 [38].Both intervention on life habits and metformin are effective in preventing or delaying diabetes.Metformin can be an important instrument for the fight against the growing diabetes epidemic, due to its sustained effects to prevent or delay the start of diabetes for at least 15 years.
Although the intervention in life habits was uniformly effective and the subgroups [2], the DPP identified significant benefit of metformin in which they were more obese, had more fasting hyperglycemia or background of gestational diabetes, and a suggestion of greater effectthat the intervention on life habits in the youngest.Regardless of randomization, the high of diabetes progression was associated with less risk of microvascular complications [9] and, in man, metformin reduced the development of atherosclerosis [33].
In addition, over 10 years, metformin treatment saved costs [39].Recommendations advise intervention on life habits or metformin treatment to prevent diabetes [37, 40].Given the current knowledge of the beneficial effects of metformin to prevent or delay diabetes, the effort arranged to transfer this evidence to practice can contribute to re -encompassing the continuous increase in the frequency of type 2 diabetes.
The possible utility of metformin extends beyond the prevention of diabetes and represents the following phase of the study for DPPOS.As the largest and longest study on metformin treatment in a population initially without diabetes, DPP/DPPOS is now in good position to evaluate whether to start metformin early in high -risk people influences the development and risk ofOther concomitant diseases, even later, such as ECV and cancer.
Although it could be expected that by decreasing the incidence of diabetes, the risk of ECV would decrease, the effect of metformin and the prevention or delay of the appearance of diabetes on the CVD is not demonstrated.In addition, on the basis of experimental and epidemiological data, metformin could be a drug against cancer.Prospective interventionist studies with prolonged treatment and monitoring are necessary to address these important questions.
In conclusion, the DPP/DPPOS clearly demonstrated the importance of metformin to prevent diabetes.Looking at the future, it is essential to know whether to transfer this data to the usual clinical practice improves current trends in the development of diabetes.
Summary and Comment Objective: Dr. Ricardo Ferreira