Contents
Contributors
Search

Type 1A diabetes mellitus is a chronic autoimmune disease characterized by a long preclinical phase, followed by an abrupt clinical onset. Individuals with type 1 diabetes are dependent on exogenous insulin to sustain-life and prevent diabetic ketoacidosis. As a consequence of clinical trials for the prevention of diabetes it is now possible to diagnose diabetes before the onset of clinical symptoms (1). Assays to detect islet antibodies have been developed that allow identification of individuals at high risk for type 1 diabetes (2). Insulin is an important autoantigen in the disease. Anti-insulin antibodies are detected prior to the initiation of exogenous insulin therapy (3) and they are a good predictive marker of the disease (4). In addition, insulin-specific T cells can induce diabetes in animal models (5). Many individuals with type 1 diabetes will develop a transient remission of their disease characterized by a reduction in insulin requirements. This so called "honeymoon phase", can last up to 12 months (6). Insulin therapy is often discontinued and then the disease recurs. Intensive insulin therapy at the time of diagnosis may help to preserve endogenous insulin secretion (7). In addition, in the Diabetes Control and Complication trial, patients treated with intensive insulin therapy had greater residual endogenous insulin secretion than patients treated with conventional therapy (8). Thus, insulin itself may be inducing this "honeymoon" phase, but the mechanism for this effect is not known. This remission may be related to correction of glucose toxicity, and reduction in insulin resistance, or it could be related to the immunological effects of insulin. Insulin therapy administer to relatives at risk to develop type 1 diabetes, failed to prevent the onset of diabetes (1). Therefore, outside of a clinical trial setting, it would not be advisable to administer insulin to individuals in the preclinical phase of the disease. However, intensive insulin therapy at the time of diagnosis may help to preserve endogenous insulin secretion. Numerous international studies are currently underway to develop therapies for the preservation of islet cell function and the prevention of type 1 diabetes.

Type 2 diabetes is felt to be a multigenic disorder characterized by both impairment in insulin secretion and insulin action. These abnormalities lead to excessive hepatic glucose production, decrease glucose uptake in peripheral tissue and abnormal lipid metabolism. In the past this form of diabetes was referred to as adult-onset diabetes, but that term is no longer used. In fact, there has been an epidemic of type 2 diabetes in children over the past decade, with up to 34% of children developing diabetes having type 2 diabetes. Multiple medications are currently available for the treatment of type 2 diabetes. These medications improve insulin secretion, reduce insulin resistance, and slow the absorption of carbohydrates. Individuals with type 2 diabetes often are treated with multiple medications, and some require insulin therapy. Interestingly, a subset of patients who have been diagnosed with type 2 diabetes have recently been found to have an immune-mediated disease. This form of diabetes has been called various names including: latent autoimmune diabetes of adults (LADA), type 1.5 diabetes, or slowly progressing type 1 diabetes (9-11). LADA has been characterized by the presence of islet cell antibodies (to glutamic acid decarboxylase (GAD), tyrosine phosphatase IA-2, or islet cells by immunofluorescence). In some studies up to eighty percent of individuals with previously characterized type 2 diabetes who develop islet antibodies will require insulin therapy (12).

This review will evaluate results of recent trials for the prevention of type 1 diabetes as well as provide an update on the current trials that are actively recruiting patients. Also, reviewed are the current guidelines for conducting a type 1 diabetes trial. It should be emphasized that since there are no proven therapies for the prevention of type 1 diabetes, the use of any immune therapy for the prevention of diabetes should be done in the context of a randomized, controlled clinical trial. Also, exciting data from recently completed trials on the prevention of type 2 diabetes will be presented along with guidelines of how some of these interventions can be implemented by the practicing physician. The final section will included currently available strategies which may be helpful for preserving islet cell function.

TYPE 1 DIABETES PREVENTION STUDIES

The demonstration that both type 1 diabetes and LADA are immune-forms of diabetes has lead to international efforts to develop immune therapies that would halt the progression of disease. The idea that such immune therapies might be possible was suggested over sixty years ago by C.H. Best and colleagues (13). One of the first attempts at immune therapy of diabetes was in 1983 when Stiller and Dupre placed children with new onset type 1 diabetes on the immununosuppressive drug cyclosporine (14). They were able to induce an insulin-free remission in over 40% of patients. However, long-term follow-up demonstrated the development of toxicity to cyclosporine including renal side effects(15). After cyclosporine was discontinued, the disease relapsed. Nevertheless, this was one of the first demonstrations that altering immune function could induce a remission in diabetes. As a consequence of these trials, international efforts were focused on developing immune therapies that would be either antigen specific directed against only the immune destruction of the islet or with less systemic toxicity. Current efforts to prevent immune-mediated diabetes focus on three major therapeutic categories: antigen specific immune therapy (16), immunosuppressive therapy (14;17-20) or immune modulation therapy (21) (Table 1).

Diabetes Prevention Trial, Dpt-1

Since Diabetes Prevention Trial, DPT-1 was one of the largest trials ever conducted for the prevention of type 1 diabetes, it will be considered in more detail. In contrast to previous trials which utilized broad spectrum immunosuppressive agents, DPT-1 used insulin as an antigen-specific immune therapy (1). This multicenter trial sponsored by the National Institutes of Health with sites throughout United States, as well as in Canada and Puerto Rico, had over 380 centers. The trial consisted of three stages (Figure 1): screening, staging level of risk, and administering insulin or observing a control group closely for the development of diabetes. Screening was completed in 89,827 first and second degree relatives of individuals for type 1 diabetes (including siblings, offspring and parents). Of the 84,594 that were eligible for further study, there were 3,152 (3.7%) positive for islet cell antibodies. The staging process involved confirmation of the presence of islet cell antibodies, measurement of insulin autoantibodies and assessment of islet cell function by measurement of the first-phase insulin response to intravenous glucose and an oral glucose tolerance test. The first-phase insulin secretion was determined by evaluating the serum insulin response one and three minutes after administration of intravenous glucose. For siblings, offspring and second degree relatives over eight years of age, a test was considered abnormal if the insulin value was below the tenth percentile ( < 100 uU per milliliter) or < 60 uUnits per milliliter for subjects less than eight years of age. If an individual had confirmed islet cell antibodies present and impaired first phase insulin response on two occasions, they were considered to have a high risk to develop diabetes (a five year risk greater than fifty percent). If the first phase insulin response was above the threshold, but the islet antibodies were present and the oral glucose tolerance test was normal, they were considered to have an intermediate risk to develop diabetes (with a five years risk to develop diabetes of 26-50%) and they were deemed eligible for the oral insulin study. To exclude individuals with decreased genetic risk for diabetes, HLA testing for the presence or absence of HLA DQA1*0102,DQB1*0602 was performed. If this protective HLA haplotype was detected, the subjects were excluded from the study.

Figure 1. DPT-1 Parenteral Insulin Trial: Study Design

The intervention for the high risk subjects was parenteral insulin, human ultralente, 0.125 U per kilogram subcutaneously twice a day as well as a continuous intravenous infusion of human regular insulin at a dose of 0.015 U per kilogram per hour x 4 days every twelve months. The development of diabetes was the primary study outcome and subjects were evaluated every six months by oral glucose tolerance test. The subjects were followed for a median of 3.7 years. Diabetes developed in 69 of 169 subjects in the intervention group (15.1 percent per year) compared to 70 of 170 subjects in the observation group (14.6 percent per year). The cumulative incidence of diabetes was similar in the two groups. Thus, low dose parenteral insulin administration failed to prevent diabetes in individuals at high risk to develop the disease.

In the second arm of DPT-1, individuals with an intermediate risk for diabetes (a five year risk of 26 to 50 percent) were eligible for the oral insulin trial. These individuals tested positive for islet cell antibodies insulin antibodies and had a first-phase insulin response in the intravenous glucose tolerance test of above the tenth percentile for the group (> 100 ?U per ml for subjects eight years of age or older, and > 60 for subjects less than 8 years of age). In this five year trial, subjects were given either oral insulin or placebo. However, as revealed at the American Diabetes Association meeting in New Orleans in June 2003, the oral insulin also failed to prevent type 1 diabetes in the nearly 200 subjects that were at risk for the disease (31).

The results of the DPT-1 were very disappointing since previous small pilot trials suggested that insulin could delay the onset of diabetes. However, there were a number of important lessons learned from the DPT-1 trial. One of these was the recognition of a different phenotype for the clinical presentation of type 1 diabetes. In DPT-1, 73% (102 of 139) of the individuals with newly diagnosed type 1 diabetes were asymptomatic at the time of diagnosis (1). Subjects with this new phenotype of type 1 diabetes have normal or impaired fasting glucose, with an elevated two hour glucoses, > 11.1 mmol/L by oral glucose tolerance testing, but no symptoms of diabetes. The early detection of diabetes in this protocol allowed for the early treatment of diabetes before they developed such complications such as life-threatening diabetic ketoacidosis (32).

Another fact emerging from the study is that giving insulin prior to the development of diabetes did not accelerate the onset of disease. This is important, since in some animal studies, giving insulin with complete Freund's adjuvant induced lymphocytic infiltration of the islets and diabetes (33). DPT-1 demonstrated that insulin administered in small doses can be given safely. Children had increased episodes of hypoglycemia but there was no adverse effect on cognitive function (1). The reasons for the lack of protection from the development of diabetes are not clear. One possibility is that the therapy was started too late in the course of autoimmune disease. The DPT-1 investigators suggest that a different dosing scheme or different treatment regimen, with insulin or insulin-like peptides might prevent the development of diabetes. The study demonstrated that a large cohort of individuals at high risk for diabetes can be identified and enrolled in a long-term intervention study (1).

TrialNet

The efforts initiated in DPT-1 now have evolved into a larger study group call TrialNet involving eighteen clinical centers throughout the United States, as well as sites in Canada, Europe and Australia. In the DPT-1, it was necessary to screen almost 90,000 relatives over a six year period to identify three hundred and forty subjects at high risk to develop type 1 diabetes. Therefore, many investigators have now focused intervention trials on individuals who have already developed type 1 diabetes since it is such a difficult process to identify sufficient numbers of individuals with prediabetes. The rationale would be to be to test a number of agents in individuals with newly diagnosed diabetes in a number of pilot and feasibility trials. Thus, only individuals who already have disease would be treated with agents which potentially could have some toxicity as opposed to giving a potentially toxic agent to individuals who have not yet developed or may never develop clinical disease. Once an agent has been demonstrated to have some effect at preserving islet function, it would be studied in larger trials in individuals with prediabetes. Data from these trials in individuals with new onset type 1 diabetes is now emerging (31). TrialNet is a collaborative network of clinical centers (http://www.diabetestrialnet.org/en/public/centers.html) and specialized core laboratories which will focus on developing new approaches for understanding, preventing and treating type 1 diabetes. TrialNet will conduct multiple clinical trials simultaneously throughout the clinical network. Studies will include the following three categories:

1. Natural History Studies to evaluate risk factors associated with the development of type 1 diabetes;
2. Diabetes Prevention Studies testing therapies that may delay or prevent the onset of type1 diabetes;
3. Diabetes Intervention Studies: to test therapies to preserve insulin in individuals with recently diagnosed type 1 diabetes.

Currently there are six trials under development for new onset diabetes treatment, four trials for prevention treatments, and three studies for DPT-1 follow-up, screening and natural history studies with risk assessment. Recruitment for new trials will begin in early 2004. TrialNet is funded by the National Insititute of Diabetes and Digestive and Kidney Disease, National Institute of Child Health and Human Development, the National Institute of Allergy and Infectious Diseases, as well as the Juvenile Diabetes Research Foundation and American Diabetes Association. People at risk for type 1 diabetes and those individuals who are newly diagnosed can join these clinical trials. Further information can be obtained at www.diabetestrialnet.org or (800) HALT-DM1 (800-425-8361).

Anti-Cd3 Antibody Therapy

Since type 1 diabetes is a T cell mediated disease, one approach to therapy would be to try to alter T cell function using immunosuppressant monoclonal antibodies that have been used to prevent rejection for organ transplantion. In preclinical studies, administration of these anti-CD3 antibodies, induce a reversal of diabetes. Treatment with monoclonal anti-CD3 antibodies in individuals with newly diagnosed diabetes showed promising results (34). The rationale for these studies was based on results in animal models of diabetes which suggested that an antibody against CD3 could reverse hyperglycemia and induce tolerance to recurrent disease (35;36). To reduce the toxicity from therapy with this mouse monoclonal antibody, a humanized antibody was developed and the structure of the antibodies was altered to reduced binding to the Fc receptor. In individuals with recently diagnosed diabetes, treatment with a two-week course of the modified anti-CD3 antibody, improved endogenous insulin secretion in nine of twelve patients one year after they received the therapy (34). Expanded studies evaluating this therapy are currently underway (Table 2, below).

Glutamic Acid Decarboxylase Vaccinations

An enzyme, glutamic acid decarboxylase, GAD65, has been identified as an islet and neuroendocrine antigen. In studies of animal models of diabetes, treatment with GAD or GAD peptides have been demonstrated to prevent diabetes. Recent clinical studies demonstrate that treatment of individuals with latent autoimmune diabetes with a synthetic form of the GAD65 (Diamyd), preserves islet cell function and has no obvious toxicity in the initial Phase I/II studies (37).

Heat Shock-Related Peptides

In studies in nonobese diabetic mice, an animal model of type 1 diabetes, it was demonstrated that residual islet beta cell function could be preserved even after the onset of hyperglycemia by treatment with a peptide of HSP60, p277. Based on this preclinical data, a randomized, double blind phase II trial was undertaken in individuals with type 1 diabetes of less than six month duration to determine the effect of treatment with DiaPep277 (n=17) versus placebo (n=18) on beta cell function (23). The peptide (2.5mg) or placebo was injected subcutaneously at baseline, 1 month and 6 months. The beta cell function was assessed by glucagon stimulation tests over a ten month period. The results indicated that the C-peptide levels fell in the placebo group but were maintained in the DiaPep277 treated group (p=0.039). In addition, the need for exogenous insulin was higher in the placebo group than the DiaPep277 group, while maintaining comparable levels of glycemic control (HbA1c 7% in both groups). Studies are currently underway evaluating this peptide in individuals with latent autoimmune diabetes of adults (Table 3, below).

Type 1 Diabetes Prevention Trial Guidelines:

In the past several years, our understanding on the etiology and cause of type 1 diabetes has advance sufficiently to prompt the American Diabetes Association to issued a position statement on the prevention of diabetes (38). This is based on the fact that several risk factors and immune-related markers are known that accurately identify first-degree relatives of individuals with diabetes that are at high risk to develop diabetes. This ability to predict the risk of development of diabetes in relatives of individuals with type 1 diabetes was clearly demonstrated in the Diabetes Prevention Trial, DPT-1 (1).

In an effort to standardize type 1 diabetes intervention trials, the Immunology of Diabetes Society has developed guideline for the design and implementation of these trials (39). This would allow for comparative and pooled data analysis, and facilitate the evaluation of potential preventative therapies. There suggestions for standardizing trial protocols include the following:

1. Define the onset of diabetes from the time of diagnosis of by a physician using the American Diabetes Association criteria.
2. Age of subjects in treatment and control groups should be matched as closely as possible.
   1. Phase I Trials Should Only Enroll Subjects > 18 Years of Age.
   2. Phase Ii and Iii Trials Limit Entry to < 35 Years of Age.
3. To define if diabetes is immune-mediated (type 1A) they should have at least one of four islet autoantibodies (GAD65, IA2, ICA or insulin if on insulin treatment for < two weeks).
4. The trial should contain sufficient number of subjects to enable stratification in Phase III trials.
5. The duration of the trial should be at least two years to allow for assessment of the efficacy of the treatment.
6. the Outcome Measures Include.
   1. metabolic testing assessing beta cell function by a two hour mixed meal tolerance test every three months.
   2. Standardized Autoantibody Measurements.
   3. primary outcome: difference in 2 hour area under curve C-peptide between treated and control groups.
   4. secondary outcome: HbA1c levels and insulin dose per kilogram (39).

Current Clinical Trials for the Prevention of Type 1 Diabetes

A synopsis of the current clinical trials (Tables 2, 3, 4) that are actively recruiting patients is provided based on listing provided through both the NIH ClinicalTrials.gov clinical research registry and the Immunology of Diabetes Society. These studies are divided into three major categories:

• Immune Modulation.
• Antigen Specific Therapies.
• Identification and Modification of Environmental Triggers.

The immune modulation category includes the use agents such as monoclonal antibodies, vitamins or cytokines that alter immune function, decrease inflammation or induce tolerance.

The antigen specific agents are peptides or proteins that in preclinical studies were able to prevent diabetes and would potentially have fewer systemic effects than immunosuppressive agents. These include insulin related peptides, heat shock related peptide (AVE077), and glutamic acid decarboxylase.

The environmental triggers trials evaluate infectious agents, dietary factors and other environmental factors that trigger diabetes.

In addition, there are a number of new protocols under development by the TrialNet research group (www.diabetestrialnet.org/en/public/centers.html ) and the Immune Tolerance Network (http://www.immunetolerance.org/) that should begin patient recruitment in the near future.

For further information follow links listed under contacts including the National Library of Medicine (NLM) identifiers for the NIH Clinical Trial web page (http://www.clinicaltrials.gov/) or the Immunology of Diabetes Society International Diabetes Immunotherapy Group (IDIG) at http://www.idsoc.org/idig.html.

PREVENTION OF TYPE 2 DIABETES

Diabetes Prevention Program

The Diabetes Prevention Program (DPP) was a multicenter randomized controlled trial of persons at risk for type 2 diabetes to determine if a lifestyle intervention or treatment with metformin would prevent or delay the onset of type 2 diabetes. Eligibility criteria included age of at least 25 years, body mass index of equal to or greater than 24 (22 or higher in Asians), and a fasting plasma glucose of 95 to 125 mg/dl and 140 to 199 mg/dl two hours after 75 grams of glucose. There were 158,177 subjects recruited at 27 centers who underwent screening, followed by oral glucose tolerance testing in 30,985, and 3,819 eligible for randomization (Figure 2). After the fourth intervention with troglitazone was discontinued in 1998, the final study involve 3234 subjects, 1082 randomized to placebo, 1073 to metformin, and 1079 to the intensive lifestyle group. The three interventions were standard lifestyle recommendations with placebo or metformin 850 mg po twice daily, or an intensive lifestyle program.

Figure 2. NIH Diabetes Prevention Program: Study Design

The standard lifestyle recommendations consisted of written information and an annual 20-30 minute individual education session emphasizing the importance of a healthy lifestyle. They were encouraged to follow the Food guide Pyramid and a diet equivalent to the National Cholesterol Education Program Step 1 diet. They were also advised to increase their physical activity. The goal of the intensive lifestyle group was to achieve and maintain a weight reduction of at least 7% by following a healthy low-calorie, low-fat diet and to engage in physical activity of moderate intensity such as brisk walking for at least 150 minutes per week. This was accomplished by covering a 16-lesson curriculum on diet, exercise and behavioral modification by one on one teaching during the first 24 weeks after enrollment. This was followed by monthly individual and group sessions designed to reinforce the behavioral changes. This lifestyle program has been described in detailed (40). To achieve the lifestyle goals, individual case managers served as "lifestyle coaches", had frequent contacts with participants as well as supervised physical activity sessions. They also had more flexible maintenance interventions with motivational campaigns and "restarts" and utilized an extensive network of training, feedback and clinical support(40).

The primary outcome was development of diabetes diagnosed on the basis of an annual oral glucose tolerance test or a semiannual fasting plasma glucose using the 1997 American Diabetes Association criteria (fasting plasma glucose equal to or greater than 126 mg/dl (7 mmol per liter) or 200 mg/dl (11.1 mmol per liter) two hours after a 75 gram glucose load.

The Diabetes Prevention Program demonstrated that type 2 diabetes could be prevented in persons at high risk for the disease by lifestyle intervention (58% reduction) and to a lesser extent with metformin (31% reduction).

There have been two additional type 2 diabetes prevention trials utilizing lifestyle programs which have had similar results:

1. Diabetes Prevention Program (Finland) n=522 patient characteristics: impaired glucose tolerance, BMI 31, mean age 55, followed for 3.2 years; results: lifestyle changes (5% weight loss, 210 minutes physical activity per week) resulted in 58% reduction in incidence of diabetes compared to the control group(41).
2. Da Quing IGT and Diabetes Study (China) n=577 patient characteristics: impaired glucose tolerance, BMI 26, mean age 45, followed for 6 years; results: diet (target BMI 23) reduction of diabetes incidence by 33%, exercise (210 minutes/week), reduction of diabetes incidence by 47%.

Diet and exercise, reduction of diabetes incidence by 38% (42). In addition, treatment of prediabetic individuals with acarbose or troglitazone (which has been taken off the market), have demonstrated that the incidence of diabetes can be reduce(42).

Type 2 Diabetes Mellitus Prevention Guidelines

It is estimated that 16 million Americans are at risk to develop type 2 diabetes and have a condition called pre-diabetes. There are definitive risk factors that identify individuals with this condition. This is different than the pre-diabetes for individuals at risk for type 1 diabetes who have an autoimmune disorder identified by the presence of islet cell antibodies. Perhaps a better terminology would be pre-diabetes Type 1 and pre-diabetes type 2 to distinguish these two distinct disorders. The U.S. Department of Health and Humans Services National Diabetes Education Program (NDEP) has issued guidelines for identifying individuals at risk for type 2 diabetes along with eleven steps to prevent the development of disease. These recommendations are based on the results of the NIH sponsored Diabetes Prevention Program which demonstrated that people with pre-diabetes lowered their risk for developing diabetes by 58% by implementing intensive lifestyle changes including getting thirty minutes of physical activity five days a week and losing five to seven percent of their body weight.

Screening Guidelines for Prediabetes Type 2

Any person over age 45, even if they are not overweight, should have a fasting plasma glucose tested every three years performed within a health care setting (43). A capillary blood glucose should not be used since glucose meters lack the accuracy to be used as diagnostic tool. Individuals who are younger than 45 but are overweight, with one or more risk factors for diabetes, should be screened for diabetes. In addition, since the incidence of type 2 diabetes in children is increasing, youth at substantial risk for type 2 diabetes should be tested. Overweight children with BMI > 85th percentile for age and sex or weight > 120% of ideal for height and any two of the above risk factors, should be tested for diabetes. Screening should be done beginning at age 10 and repeated every two years (43).

Type 2 Diabetes Prevention Guidelines

Individuals at risk for developing diabetes need to be made aware of the benefits of increasing regular exercise and modest weight loss. They need to make small changes in their lifestyle with realistic goals. Exercise should be started at a modest rate, such as 15 minutes per day. The duration of exercise should be increased 5 minutes each week with a goal of 30 minutes per day for most days of the week.

The diet should be altered to include more fruits and vegetables, beans and grains (five to nine servings per days. The fat content should be reduced and grilled or bake foods substituted for fried foods. A food diary should be kept to record progress on the diet. Forms can be downloaded for this at (http://www.ndep.nih.gov/diabetes/prev/prevention.htm). Since making these modest lifestyle changes can be challenging, it is recommend to make one healthy change per week. Education materials are available and are free upon request from the NDEP. These include the booklet: "Small Steps. Big Rewards. Your Game Plan for Preventing type 2 Diabetes," which can be requested at 1-800-438-5383 or visit the NDEP website at www.ndep.nih.gov.

Prevention of Gestational Exposure to Rubella

Based on careful follow-up of patients from the rubella epidemic of 1940, congenital rubella infections are the only infections clearly associated with type 1 diabetes (44). Congenital exposure to rubella leads to a twenty to thirty fold increased risk to develop diabetes (45). In contrast, rubella infections in early childhood to not appear to increase the incidence of diabetes (46). It is felt that rubella may cause diabetes as a result of molecular mimicry since antibodies to rubella capsid protein cross-react with insulin secretory granule proteins (47). Worldwide efforts to eradicate rubella infections have lead to widespread immunization programs with a progressive decline in the reported cases of rubella. However, rubella outbreaks continue to occur particularly among immigrants. In a recent outbreak of rubella in Nebraska, 83 cases of rubella were reported, and none of them had prior rubella vaccination or their vaccination status was unknown (48). In addition, surveys of immunization status of women of childbearing age in the United States revealed 23% were susceptible to rubella infection (i.e. they did not have adequate rubella antibody titers) (49). Since the rubella vaccine does not provide the lasting protection that was previously seen in individuals infected with rubella, repeat vaccination is necessary in individuals with low rubella antibody titers. This takes on even more importance because of the fact that a mild rubella infection late in pregnancy, often asymptomatic, is most likely to result in diabetes in the children born with congenital rubella infection. In contrast, a rubella infection early in pregnancy is more likely to induce congenital defects such as deafness, and cardiac abnormalities. It is estimated that there are up to 100,000 cases of congenital rubella worldwide (50). Asymptomatic rubella infections late in pregnancy in women with low titer rubella antibodies may be a factor contributing to the rising incidence of diabetes worldwide. In addition, the rubella antibody status of health care workers should be monitored since in some surveys up to 20% of obstetricians did not have adequate rubella immunity (51). The CDC recommends that women of childbearing age who do not have acceptable evidence of rubella immunity should be vaccinated (52). This should be done in the setting of general medical or gynecological care or in family-planning settings. The women should be counseled not to become pregnant for 3 months after the vaccination because of the theoretical risk of infection of the fetus with the vaccine virus. Currently there is no evidence that this has occurred. Prenatal serologic testing for rubella should be done in all women who lack acceptable evidence of rubella immunity (52).

Encourage Breast Feeding and Minimize Early Exposure to Cows Milk Protein

Epidemiological data shows a lower incidence of type 1 diabetes in children who are breast fed compare to formula fed infants. In families with a history of type 1 diabetes, the World Health Organization's in their 2001 guidelines recommends exclusively breastfeeding for their babies for at least the first six months of life (WHO_CAH_01_24.pdf) (http://www.who.int/child-adolescent-health/New_Publications/NUTRITION/ f)

• The relative importance of the protective effects of mother's milk compared to the avoidance of cows milk exposure for the protection from type 1diabetes is currently under investigation.
• International clinical trials are underway investigating these possibilities (www.TRIGR.org).

For individuals with newly diagnosed type 1 diabetes maintain intensive insulin therapy at the time of diagnosis.

Do Not Stop Insulin Therapy if a Spontaneous Remission Occurs.

• Insulin may have induced remission or honeymoon phase and the Diabetes Control and Complication Trial data suggest that intensive insulin therapy helps to preserve endogenous insulin secretion (8).

For individuals with relatives with type 2 diabetes, eat a low fat diet, maintain an active lifestyle, and keep body weight within ten percent of their ideal body weight to prevent the development of diabetes.

The NIH Diabetes Prevention Program showed a 58% reduction in progression to diabetes in relatives with impaired glucose tolerance. This was accomplished by exercising or walking five times a week for thirty minutes resulting in a 5 to 7% reduction in body weight. A new education program is now available to facilitate this process. Education materials are available and are free upon request from the NDEP. These include the booklet: "Small Steps. Big Rewards. Your Game Plan for Preventing type 2 Diabetes," which can be requested at 1-800-438-5383 or visit the NDEP website at www.ndep.nih.gov.

In conclusion, we are now able to identify individuals who are at risk to develop either type 1 or type 2 diabetes. Multiple international trials are underway trying to develop strategies for the prevention of type 1 diabetes and to preserve residual beta cell function in individuals who have already developed the disease. In addition, lifestyle programs and treatment with oral hypoglycemic agents have been demonstrated to prevent the development of type 2 diabetes. Oral agents are not yet approved to be used for the prevention of diabetes, so lifestyle programs should be the main strategy for the prevention of type 2 diabetes in individuals at high risk to develop the disease.