Pharmacological Primary Prevention of Diabetes Mellitus Type II: A Narrative Review

The evolving epidemic of type 2 diabetes mellitus has challenged health-care professionals. It stands among the leading causes of mortality in the present world. It warrants new and versatile approaches to improve mortality and the associated huge quality-adjusted life years lost to it once diagnosed. A possible venue to lower the incidence is to assess the safety and efficacy of various diabetes prevention strategies. Diet and exercise have a well-developed role in the prevention of weight gain and, ultimately, diabetes mellitus type II in high-risk individuals. However, high-risk individuals can also benefit from adjunct pharmacotherapy. In light of this information, we decided to conduct a systematic review of randomized controlled trials. This article summarizes the evidence in the literature on the pharmacological prevention of diabetes in high-risk individuals.

The review was done to study the pharmacological approaches used in the prevention of diabetes mellitus type II over the years. The literature on the primary prevention of diabetes mellitus using different medications is limited, and an inclusive review with all such approaches will be a useful read for endocrinologists and internists alike to devise mechanisms to control the incidence of diabetes mellitus type II.

Databases
The literature review for the research was performed on PubMed, Cochrane, and Clinicaltrials.gov [14]. MeSH terms type II diabetes mellitus, primary prevention, and pharmaceutical preparations were searched with all corresponding keywords, and relevant articles were imported into Endnote. Additionally, we searched databases using individual diabetic medications to make sure we do not miss any articles. All keywords are shown in Table 1  Described below are the inclusion criteria for our review: 1: All studies with nondiabetic and prediabetics participants 2: Observational and experimental studies 3: Good or fair-quality studies on the quality assessment questionnaire as shown in Table 2 4: Studies published in the English language

Exclusion Criteria
Described below are the exclusion criteria for our review: 1: Review studies 2: Poor-quality studies on the quality assessment questionnaire 3: Studies published in languages other than English 4: Studies that significantly include patients already diagnosed with type II diabetes mellitus 5: Editorials, letters to editors, and comments 6: Ongoing clinical trials

Study Selection
A total of 557 studies were imported into Endnote from the three databases. A total of 102 repeat articles were removed. Inclusion and exclusion criteria were applied and a total of 63 (13.8%) articles were selected after going through titles and abstracts. These 63 full articles were extracted and independently passed through the quality assessment questionnaire to finally select a total of 18 (30%) articles that have been included in the final review. Figure 1 shows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow chart for study selection [15].   Data extraction was done from selected studies in a tabulated form. The first author name, country of research, year of research, mean characteristics and size of the sample, the drug studied, and results were documented in the tabulated form shown in Table 3. The study concluded that participants on acarbose were 25% less likely to proceed to diabetes vs. placebo. 32.4% of participants in the acarbose arm developed diabetes at two years follow-up compared to 41.5% in the placebo arm.
Gastrointestinal side effects were noted in the treatment arm.
Nijpels et al. There was a statistically significant decrease in the blood glucose levels and incidence of diabetes in the treatment arm.
However, there was a 30% dropout in the treatment arm due to abdominal pain and other gastrointestinal side effects. The large randomized trial showed that there Is no statistically significant reduction in the incidence of diabetes mellitus in highrisk populations when they receive vitamin D. The study showed a mild increase in hypercalcemia, nephrolithiasis, and associated side effects of hypercalcemia but the findings were statistically insignificant.
R.J. Durbin The study concluded that thiazolidinediones were effective in reducing the incidence of diabetes in high-risk populations. It was estimated that treatment of 4.2 high-risk people with a glitazone for 3 years will prevent the occurrence of diabetes in one patient The study concluded that troglitazone reduces the incidence of diabetes high-risk women by 50%. Statistically significant reduction in the incidence of diabetes mellitus was seen in Hispanic women. Hepatotoxicity was seen in 8 patients who received troglitazone. The drug was later withdrawn from the market due to poor side-effect profile.

Metformin
Metformin is the drug most studied for the primary prevention of diabetes mellitus in high-risk patients. In our review, we considered four randomized control trials that studied the effect of metformin. The studies show an average absolute risk reduction of 16% among the participants who received metformin across all those trials. The Indian Diabetes Prevention Programme was a community-based trial that studied the influence of intervention in the incidence of diabetes [16]. The subjects were individuals with impaired glucose tolerance (IGT) who were treated with lifestyle modifications (LSM), metformin (MET), and a combination of both LSM and metformin. Lifestyle modifications and metformin showed a similar reduction in the incidence of diabetes mellitus type II over the 30 months of trial, however, the combination of lifestyle modifications and metformin did not show any added benefits as compared to lifestyle modifications and metformin alone. None of the side effects were statistically significant except for the gastrointestinal side effects that occurred in 30 patients taking metformin. On reporting of side effects, the metformin dose was reduced in those patients after which no side effects were reported.
The Canadian Normoglycemia Outcomes Evaluation (CANOE) trial investigated the effect of combination pharmacological therapy on the development of diabetes mellitus type II [17]. Participants included patients with IGT who were allocated into two groups. The first group received combination rosiglitazone (2 mg) and metformin (500 mg) twice daily while the second group received matching placebo for a median of 3·9 years. The incidence of diabetes was significantly reduced in the active treatment group (n=14 (14%)) as compared to the placebo group (n=41 (39%); p<0·0001). The relative risk reduction was 66% (95% CI 41-80) and the absolute risk reduction was 26% . Around 80% of patients in the treatment group regressed to normal glucose tolerance as compared with only 53% in the placebo group (p=0·0002). Furthermore, by study end, insulin sensitivity had decreased to a greater extent in the placebo group than in the rosiglitazone and metformin group. The study did not report any side effects experienced by the patients.
Sussmann et al. studied the occurrence of diabetes in subjects who were stratified by their risk of developing diabetes according to a diabetes risk prediction model [18]. In this study, 3081 participants with impaired glucose metabolism at baseline were taken. It was observed that the benefit of metformin was distributed quite unevenly across the study population. Participants who were at the highest likelihood of having diabetes in the future received far greater benefits from metformin therapy (21.5% absolute reduction in diabetes over three years of treatment) while patients who were in the lower risk group achieved little or no benefit. In the lowest risk quarter for progression to diabetes, the Metformin group had a slightly higher risk of developing diabetes (9.6%) than did the control group (8.3%). In the highest risk quarter, the metformin group had an absolute risk reduction of 21.4% while the control group had a 59.6% observed rate of developing diabetes. Therefore, patients with the greatest risk of developing diabetes had a statistically significant relative risk reduction from metformin use (p<0.001). Therefore, this type of benefit-based tailored treatment, using a multivariable risk prediction tool, could decrease drug overuse and help make the prevention of diabetes far more efficient, effective, and patient-centered. The study did not report the side effects experienced by the patients.
Alfawaz et al. studied the status of metabolic syndrome in individuals receiving the Intensive Lifestyle Modification Programme (ILMP) or low dose metformin as compared to the control group [19]. The occurrence of metabolic syndrome in the ILMP group decreased by 26% (p<0.001), in the metformin group by 22.4% (p=0.013), and the control group by 8.2% (p=0.281). Mean fasting glucose was significantly reduced in the ILMP and metformin groups while in the control group, this reduction was modest. Similarly, hyperglycemia also decreased significantly by 38.4% in the ILMP group and 39% in the metformin group, respectively. The mean reduction in fasting glucose in the ILMP group is less than that found in the metformin group. In the study, the mean weight loss from baseline to the end of the study was 4.15 kg in the metformin group as compared to 1.6 kg in the ILMP group. The study did not document the side effects that were experienced by the patients who had metformin.

Liraglutide
Ariel et al. were the first to study the effect of liraglutide on a small number of prediabetic patients [20].
Twenty-three patients out of 50 received the medication while 27 received a placebo. Weight loss was greater in the liraglutide group as compared to the placebo group (6.9 vs. 3.3 kg, p<0.001) and so was the decrease in fasting blood glucose level (9.9 mg/dL vs. 0.3 mg/dL, p<0.001). This study showed the potential for the use of liraglutide in the prevention of diabetes mellitus type II. A three-year study was designed to observe the effect of liraglutide in combination with diet and exercise to decrease the risk of developing diabetes mellitus Type II in prediabetic individuals [21]. Liraglutide was associated with a risk reduction in the onset of type 2 diabetes by 80% as compared to placebo (HR 0·21, 95% CI 0·13-0·34). By week 160, 2% of individuals in the liraglutide group versus 6% in the placebo group were diagnosed with diabetes while on treatment.
The time to onset of diabetes over 160 weeks among all groups was 2·7 times longer with liraglutide than with placebo (95% CI 1·9 to 3·9, p<0·0001). After 160 weeks, regression from prediabetes to normoglycemia was observed in 66% of individuals in the liraglutide group (odds ratio (OR) 3·6, 95% CI 3·0-4·4, p<0·0001). Liraglutide-induced greater weight loss was more significant compared to the placebo at week 160 (-6·1% for liraglutide vs -1·9% for placebo; estimated treatment difference -4·3%, 95% CI -4·9 to -3·7 (p<0.0001). Weight loss with liraglutide treatment was sustained over three years. Insulin resistance and β-cell function improved in the liraglutide group as compared with the placebo group at week 160 and glycated hemoglobin, fasting glucose, and fasting insulin concentrations were also lower with liraglutide. After 12-week treatment cessation, some individuals in the liraglutide group reverted to prediabetes but 50% of the treated individuals still had normoglycemia at week 172 as compared with 36% of the individuals in the placebo group (p<0·0001).
Liraglutide has a well-documented safety profile and is well-tolerated. Some gastrointestinal disorders, notably nausea, diarrhea, constipation, and vomiting, were observed in the liraglutide group as compared to the placebo group.

Linagliptin
The PRELLIM (Diabetes Prevention with Linagliptin, Lifestyle, and Metformin) project was designed to study the efficacy of a combination of linagliptin + metformin + lifestyle modification as compared to metformin + lifestyle modification only [22]. At 24 months, glucose levels remained improved in both groups but significantly more in the LM group mainly at 60 minutes in the oral glucose tolerance test (OGTT) (167±3 vs 155±3, in M and LM group, respectively, p<0.05). Insulin sensitivity measured from OGTT and during fasting had more pronounced improvement in the LM group as compared to the M group (p<0.05). Reduction in glycated hemoglobin (HbA1c) was observed only in the LM group (p<0.01). Improvements in β-cell function at 24 months persisted in both groups but they were significantly better in the LM group (p<0.05). The probability to regress to normoglycemia and normal glucose tolerance was significantly higher in the LM group than in the M group during the entire follow-up (OR adjusted per month: 3.31, 95% CI: 1.5 4 -7.09, p=0.00 2). Side effects: Diarrhea occurred in six patients from the M group and in five patients from the LM group. Nausea occurred in two patients from each group, cephalea in two from the M group, and one from the LM group.

Acarbose
Chiasson et al. (2002) (n=1429) studied the role of acarbose in delaying or preventing the progression of impaired glucose tolerance to full-blown type 2 diabetes mellitus [23]. In this multicenter placebo-controlled randomized trial, participants were randomized 1:1 to either arm and were followed up for 3.3 years. Participants on acarbose were 25% less likely to proceed to diabetes vs. placebo. In the acarbose arm, 32.4% of participants developed diabetes at the two-year follow-up as compared to 41.5% in the placebo arm. The study concluded that acarbose not only decreases progression from glucose impairment to diabetes but also brings the impairment back to normal. Ninety-eight percent (98%) of the acarbose group and 95% of the placebo group suffered from at least one adverse event. The most common adverse events were gastrointestinal and were significantly higher in the acarbose arm than the placebo (p<0.0001), but they were considered mild to moderate in intensity. The common adverse events were diarrhea, flatulence, and abdominal pain, all of which were higher in the acarbose group.  [26]. In this double-blinded, randomized, placebo-controlled trial, participants were to be followed for three years. The interim analysis showed significant improvement in the voglibose group, hence leading to early termination of the trial. At the end of the study, a significantly fewer number of participants had developed full-blown diabetes in the voglibose group vs. the placebo group (50 of 897 vs. 106 of 881; hazard ratio 0.595 (95% CI 0.433-0.818); p=0.0014). Similarly, more people reverted to normal glucose levels in the voglibose group than in the placebo group (599 of 897 vs. 454 of 881; hazard ratio 1.539 (95% CI 1·357-1·746); p<0.0001). Ninety percent (90%) of the experimental arm were reported to have suffered adverse events vs. only 85% in the control group (p=0.0009). Five percent of participants from the voglibose arm vs. 3% of participants from the control arm discontinued the therapy due to adverse events in which causality could not be denied (p=0.0092). Though only mild to moderate in severity, most common adverse events in voglibose vs. placebo were flatulence 17% vs. 7%, abdominal distention 13% vs. 5%, and diarrhea 13% vs. 5%, respectively. A total of six deaths were reported in the voglibose arm, none of which were related to drug treatment (two were attributed to accidents and one each to heart failure due to self-intoxication, myocardial infarction, lung cancer, suicide). No deaths were reported in the placebo group.

Orlistat
Heymsfied et al. (2000) (n=675) did a retrospective meta-analysis to study the effect of weight loss from orlistat on glucose tolerance and progression to type 2 diabetes in obese adults [27]. Results from three randomized, double-blinded, placebo-controlled, multicenter trials were pooled. The mean duration was 582 days. Participants in the orlistat group lost more weight than in the placebo, mean ± SEM; 6.81 ± 0.41 kg vs.
3.79 ± 0.38 kg, respectively (p<0.001). In participants with impaired glucose tolerance, the orlistat group achieved normoglycemia in 71.6% vs. 49.1% in the placebo group (p=0.04). Moreover, only 3% of participants from the orlistat arm converted to diabetes vs. 7.6% from the placebo arm. No adverse event data were reported. The study concluded that orlistat, through weight loss, significantly lowers the risk of deterioration of glucose tolerance and even reverts to normal levels in high-risk obese adults.

Torgerson et al. (2004) (n=3305) conducted XENDOS (XENical in the Prevention of Diabetes in Obese
Subjects), a randomized study of orlistat effects as an adjunct to lifestyle modification in the prevention of type 2 diabetes in obese adults [28]. In the intention-to-treat analysis, 52% of the orlistat group completed the treatment vs. 34% of the placebo (p=<0.0001). At the four-year follow-up, 9% of the placebo group participants vs. 6.2% of the orlistat group participants developed diabetes (risk reduction: 37.3%; p=0.0032). Subanalysis revealed that this difference was due to subjects with impaired glucose tolerance in the respective group. Mean weight loss with orlistat was 5.8 kg vs. 3 kg with placebo (p<0.001). The inclusion of dropouts into the analysis still revealed significant weight loss in the orlistat group 3.6 kg vs. 1.4 kg (p<0.001). Orlistat was associated with a significantly higher incidence of gastrointestinal adverse events vs. placebo. The incidence of all other adverse events was similar in both groups. Most of these adverse events were mild to moderate. Ninety-one percent (91%) of the orlistat group reported adverse events vs. 65% of the placebo group in the first year of treatment. With time, the incidence reduced to 35% vs. 23% in orlistat vs. placebo at the four-year follow-up. Overall, 8% vs. 4% of participants stopped treatment due to adverse events in the orlistat vs. placebo groups, respectively. No deaths were attributed to orlistat. The study concluded that orlistat significantly improves the outcome in obese adults who are at risk of developing diabetes when used in adjunct with lifestyle modifications.

Nateglinide
A study group called NAVIGATOR (Nateglinide and Valsartan in Impaired Glucose Tolerance Outcomes Research) conducted a large, double-blinded, randomized clinical trial (2010) (n=9306) where the participants with impaired glucose tolerance were randomized into 2 x 2 factorial design to either receive nateglinide vs. placebo or valsartan vs. placebo, both in adjunct with lifestyle modifications [29]. About 69.9% and 71% of the nateglinide and placebo groups, respectively, completed treatment for the whole five years. Thirteen point one percent (13.1%) from the nateglinide group vs. 12.9% from the placebo group were lost to follow-up or withdrew while 11.2% vs. 10.4% from the respective group stopped treatment due to adverse events. At the five-year follow-up, the study reported that nateglinide did not reduce the incidence of diabetes significantly (36% vs. 34%, respectively; hazard ratio, Rosiglitazone Durbin et al. (2004) (n=172) did a prospective analysis among individuals with impaired glucose tolerance who were initially taking troglitazone but later switched to rosiglitazone or pioglitazone when the troglitazone was withdrawn from the United States market due to toxicity reports [31]. One-hundred one (101) such individuals were identified. A control group of 71 individuals who had glucose impairment but were not taking any antidiabetic medication was included. At the three-year follow-up, 2.97% of participants in the rosiglitazone group progressed to diabetes while 26.6% of participants in the control group progressed to diabetes. The mean baseline HbA1c for rosiglitazone (6.12% ± 0.60) went down to 5.57% ± 0.37 after three years. Similarly, 6.23% ± 0.74 of pioglitazone went down to 5.65% ± 0.48. Paradoxically, the HbA1c in the placebo group went from a baseline of 6.18% ± 0.20 to 6.68% ± 0.19, an increase of 0.52% ± 0. 16  Troglitazone was discontinued in the US market in the year 2000 after significant hepatotoxicity-related deaths were reported. But before the withdrawal, a trial was conducted between 1995 and 1998 by Buchanan et al. (2002) (n=266) to evaluate the prevention of type 2 diabetes using troglitazone in high-risk Hispanic women with previous gestational diabetes history and impaired glucose tolerance at enrollment [33]. A total of 30 participants were lost to follow-up (19 in the intervention group vs. 11 in the control group) all of whom did not differ significantly at baseline. During blind treatment, the annual incidence of diabetes remained at 12.1% vs. 5.4% in the placebo vs. troglitazone group, respectively (hazard ratio 0.45 (95% CI 0.25-0.83) and remained unchanged after adjustment for baseline characteristics. The study concluded that troglitazone reduces the incidence of diabetes in high-risk women by 50% at the minimum and offers longterm protection from it even after stopping the drug. Eight patients discontinued the medication due to raised serum hepatic enzymes. They resumed the drug after the enzymes returned to the baseline. One patient left the trial citing personal reasons.

Discussion
Diabetes mellitus type II is one of the leading factors of cardiovascular morbidity and mortality, which is the most prevalent cause of death in the older population [34]. Diabetes mellitus also leads to multiple organ damage and is one of the leading causes of disability in vision, kidney function, and limbs in old age [35]. Diabetes mellitus and its complications have been a challenge for physicians all over the globe in the last century. After thousands of studies, it has been concluded that the best way to save mankind from the morbidity of diabetes is to prevent its occurrence [36].
Intense lifestyle modifications in high-risk individuals have always been presented as the best means to prevent diabetes mellitus type II. Numerous studies have talked about the benefits of exercise and diet restriction on not only diabetes mellitus but also overall better health of the individuals [37]. Watanabe et al. concluded that lifestyle modifications, diet, and exercise improve all the factors of metabolic syndrome [38]. However, from a practical point of view, things are not as simple as they look. Rise et al. conclude that the knowledge and awareness of the public about lifestyle modification have almost no effect at all on their change in diet and physical activity [39]. Numerous nutritionists and physicians have tried and failed to find ways and means to convince the public into following ways and means to change their lifestyle [40]. On the other hand, there is a steep rise in the use of off-the-label medications [41] both by physicians and online in the management of obesity, which brings with it a long list of side effects and complications.
In this current status quo, much neglected are the different studies done in the past to prevent diabetes mellitus in a high-risk population. Most of these studies have compelling evidence that the use of certain medications in the primary prevention of diabetes mellitus is not only effective but also associated with only mild side effects. Sussmann et al. showed effective primary prevention of diabetes mellitus in a big sample of high-risk individuals by the use of metformin [18]. Le Roux et al. studied a large sample of the high-risk population and proved with evidence that liraglutide is effective in the primary prevention of diabetes mellitus type II in the high-risk population with minimal side effects [21]. Multiple studies have shown the effectiveness of the thiazolidinedione group of anti-glycemic medications in the primary prevention of diabetes mellitus type II with a mild or minimal side-effect profile [17,[31][32]. In current settings, with the surge of diabetes diagnosis and more expected in the upcoming years and impracticality of lifestyle modifications in the current day and age, it is imperative that medical scientists look into the promising initial trials in the prevention of diabetes in at least high-risk populations and consider doing large-scale studies that lead to the formation of guidelines to prevent diabetes mellitus.

Conclusions
Our review of the literature has shown that even modest changes in lifestyle can decrease the progression of impaired glucose tolerance to diabetes by 50%-60%. However, the incidence of progression is still very high. The adjunct use of pharmacotherapy can further reduce the progression to full-blown diabetes. Antiglycemic drugs are effective in reducing the incidence of type 2 diabetes in high-risk populations. Large randomized controlled trials should be designed to further corroborate these findings.

Conflicts of interest:
In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.