COVER ARTICLE
`
`O A patient informa-
`tion handout on type
`2 diabetes mellitus,
`written by the authors
`of this article, is pro-
`vided on page 1759.
`
`Oral Agents in the Management
`of Type 2 Diabetes Mellitus
`
`BEATRIZ LUNA, PHARM.D., BCPS, and MARK N. FEINGLOS, M.D., C.M.
`Duke University Medical Center, Durham, North Carolina
`
`Despite exhaustive efforts to better manage patients with type 2 diabetes mellitus (for-
`merly known as non–insulin-dependent diabetes mellitus), attempts at maintaining
`near normal blood glucose levels in these patients remains unsatisfactory. This contin-
`ues to pose a real challenge to physicians as the prevalence of this disease in the United
`States continues to rise. Type 2 diabetes is defined as a syndrome characterized by
`insulin deficiency, insulin resistance and increased hepatic glucose output. Medications
`used to treat type 2 diabetes are designed to correct one or more of these metabolic
`abnormalities. Currently, there are five distinct classes of hypoglycemic agents avail-
`able, each class displaying unique pharmacologic properties. These classes are the sul-
`fonylureas, meglitinides, biguanides, thiazolidinediones and alpha-glucosidase
`inhibitors. In patients for whom diet and exercise do not provide adequate glucose con-
`trol, therapy with a single oral agent can be tried. When choosing an agent, it is pru-
`dent to consider both patient- and drug-specific characteristics. If adequate blood glu-
`cose control is not attained using a single oral agent, a combination of agents with
`different mechanisms of action may have additive therapeutic effects and result in bet-
`ter glycemic control. (Am Fam Physician 2001;63:1747-56,1759-80.)
`
`T he prevalence of type 2 dia-
`
`betes mellitus (formerly
`known as non–insulin-
`dependent diabetes melli-
`tus) in the United States has
`increased dramatically over the past two
`decades and continues to rise.1 Despite
`the introduction of new agents to the
`armamentarium of hypoglycemic agents,
`efforts for better management of this dis-
`ease have been disappointing and the
`control of blood glucose levels remains
`unsatisfactory.2 Recently, the results of
`the United Kingdom Prospective Dia-
`betes Study (UKPDS) were released.3
`This study, the largest and longest study
`of patients with type 2 diabetes, has rein-
`forced the belief that improved control of
`blood glucose levels can substantially
`
`See editorial
`on page 1687.
`
`In patients who have type 2 diabetes, pharmacologic interven-
`tion is required if the desired level of glycemic control is not
`achieved with diet and exercise within a three-month period.
`
`lower the overall morbidity associated
`with this disease, underscoring the
`urgency to obtain better glucose control
`in these patients. The focus of this review
`will be the management of patients with
`type 2 diabetes using one or more of the
`five available classes of oral hypoglycemic
`agents: sulfonylureas, meglitinides, bi-
`guanides, thiazolidinediones and alpha-
`glucosidase inhibitors (Table 1). Options
`for monotherapy and combination ther-
`apy, efficacy of specific agents, adverse
`effects and special populations are some
`issues addressed in this review.
`Type 2 diabetes can be described as a
`syndrome characterized by insulin defi-
`ciency, insulin resistance and increased
`hepatic glucose output.4,5 With this in
`mind, therapies used to treat patients
`with this disease are aimed at correcting
`one or more of these physiologic abnor-
`malities. Current recommendations of
`the American Diabetes Association
`include a trial of diet and exercise as first-
`line therapy for the treatment of patients
`with type 2 diabetes.6 If the desired level
`
`MAY 1, 2001 / VOLUME 63, NUMBER 9
`
`www.aafp.org/afp
`
`AMERICAN FAMILY PHYSICIAN
`
`1747
`
`MPI EXHIBIT 1070 PAGE 1
`
`MPI EXHIBIT 1070 PAGE 1
`
`

`

`TABLE 1
`Classes of Oral Hypoglycemic Agents
`
`Drug class
`
`Agent
`
`Drug class
`
`Agent
`
`Sulfonylureas
`
`First generation
`Acetohexamide (Dymelor)
`Chlorpropamide (Diabinese)
`Tolazamide (Tolinase)
`Tolbutamide (Orinase)
`Second generation
`Glyburide (Micronase)
`Glipizide (Glucotrol)
`Glimepiride (Amaryl)
`
`Meglitinides
`
`Biguanides
`Thiazolidinediones
`
`Alpha-glucosidase
`inhibitors
`
`Repaglinide (Prandin)
`Nateglinide (Starlix)
`Metformin (Glucophage)
`Pioglitazone (Actos)
`Rosiglitazone (Avandia)
`Acarbose (Precose)
`Miglitol (Glycet)
`
`of glycemic control is not achieved with diet
`and exercise within a three-month period,
`pharmacologic intervention is required.
`Criteria for initiation of therapy with an
`oral agent versus insulin are debated among
`diabetologists, but the decision should be
`made jointly by the physician and patient to
`obtain the best results.7 (Because of the appar-
`ently progressive nature of the beta cell defect
`in type 2 diabetes, current oral therapies may
`not prevent an eventual decline in glycemic
`control, and it is likely that many patients will
`ultimately require insulin therapy.) Once the
`decision is made to initiate therapy with an
`oral agent, it is prudent to consider patient-
`specific (age, weight, level of glycemic control)
`
`TABLE 2
`Clinical Efficacy of Oral Hypoglycemic Agents
`
`Class of hypoglycemic agents
`
`Sulfonylureas
`Meglitinides
`Biguanides
`Thiazolidinediones
`Alpha-glucosidase inhibitors
`
`Reduction
`in HbA1c (%)
`
`Reduction in FPG
`(mg per dL [mmol per L])
`
`0.8 to 2.0
`0.5 to 2.0
`1.5 to 2.0
`0.5 to 1.5
`0.7 to 1.0
`
`60 to 70 [3.3 to 3.9]
`65 to 75 [3.6 to 4.2]
`50 to 70 [2.8 to 3.9]
`25 to 50 [1.4 to 2.8]
`35 to 40 [1.9 to 2.2]
`
`HbA1c = glycosylated hemoglobin A1c; FPG = fasting plasma glucose.
`
`and agent-specific characteristics (relative
`potencies, duration of action, side-effect pro-
`files, cost) to make the most appropriate
`choice (Tables 2 and 38). Figure 1 illustrates a
`reasonable stepwise approach for initiating
`oral therapy in patients with type 2 diabetes
`and is consistent with the recommendations
`put forth by several expert committees and
`diabetes subspecialists.4-6,9
`
`Sulfonylureas
`Sulfonylureas have remained the mainstay
`of antidiabetic therapy since the early 1950s.
`Following the release of the University Group
`Diabetes Program (UGDP) study,10 which
`implicated tolbutamide in increased mortal-
`ity secondary to cardiovascular events, the use
`of the first generation sulfonylureas (aceto-
`hexamide, chlorpropamide, tolbutamide and
`tolazamide) quickly fell out of favor.11 Recent
`data, as summarized in an earlier review, sup-
`porting the benefits of the sulfonylureas as
`well as the availability of newer generation
`sulfonylureas with more favorable side-effect
`profiles (glyburide [Micronase], glipizide
`[Glucotrol] and glimepiride [Amaryl]), have
`contributed to their renewed popularity.3,8
`Sulfonylureas work by stimulating insulin
`release from the beta cells of the pancreas and
`may slightly improve insulin resistance in
`peripheral target tissues (muscle, fat). On
`average, this class reduces glycosylated hemo-
`
`1748 AMERICAN FAMILY PHYSICIAN
`
`www.aafp.org/afp
`
`VOLUME 63, NUMBER 9 / MAY 1, 2001
`
`MPI EXHIBIT 1070 PAGE 2
`
`MPI EXHIBIT 1070 PAGE 2
`
`

`

`Type 2 Diabetes
`
`TABLE 3
`Average Dose and Cost Comparison of Hypoglycemic Agents
`
`Drug class
`
`Brand name Generic
`
`Sulfonylureas
`
`DiaBeta
`
`Glyburide
`
`Available
`strengths
`
`Brand name
`dose range
`(mg/day) and cost*†
`
`Generic name
`dose range
`(mg/day) and cost*†
`
`1.25, 2.5,
`5.0
`
`5 ($21.24)
`to 20 ($84.96)
`
`5 ($16.06 to 20.40)
`to 20 ($64.26 to
`$81.73)
`
`5 ($16.06 to 20.40)
`to 20 ($64.26 to
`$81.73)
`
`1.5 ($8.86 to 11.33)
`to 12 ($50.45 to
`$60.96)
`
`10 ($18.21 to 20.82)
`to 40 ($66.84 to
`$76.62)
`
`‡
`
`‡
`
`‡
`
`‡
`
`‡
`
`‡
`
`‡
`
`‡
`
`‡
`
`‡
`
`Micronase
`
`Glyburide
`
`1.25, 2.5,
`5.0
`
`5 ($26.49)
`to 20 ($105.96)
`
`Glynase
`
`Glyburide
`(micronized)
`
`1.5, 3.0,
`4.5, 6.0
`
`0.75 ($13.73)
`to 12 ($73.18)
`
`Glucotrol
`
`Glipizide
`
`5.0, 10
`
`Glucotrol XL Glipizide
`
`5.0, 10
`
`Amaryl
`
`Glimepiride
`
`Meglitinides
`
`Prandin
`
`Repaglinide
`
`1.0, 2.0,
`4.0
`
`0.5, 1.0,
`2.0
`
`Starlix
`
`Nateglinide
`
`60, 120
`
`10 ($24.63)
`to 40 ($90.46)
`
`5 ($10.65)
`to 20 ($42.17)
`
`1 ($7.34)
`to 8 ($44.86)
`
`1.5 ($51.61)
`to 16 ($199.20)
`
`180 ($83.00) to
`360 ($86.50)
`
`Biguanides
`
`Glucophage Metformin
`
`500, 850,
`1,000
`
`1,500 ($58.16) to
`2,550 ($98.87)
`
`Glucophage Metformin
`XR
`
`500
`
`Thiazolidinedione Actos
`
`Pioglitazone
`
`15, 30, 45
`
`1,000 ($39.24)
`to 2,000 ($74.48)
`
`15 ($85.50)
`to 45 ($148.50)
`
`Avandia
`
`Rosiglitazone
`
`2.0, 4.0,
`8.0
`
`4 ($75.00)
`to 8 ($136.90)
`
`Alpha-glucosidase Precose
`inhibitor
`
`Acarbose
`
`Glyset
`
`Miglitol
`
`Combination
`
`Glucovance Glyburide/
`Metformin
`
`25, 50,
`100
`
`25, 50,
`100
`
`1.25/250
`2.5/500
`5.0/500
`
`150 ($46.54)
`to 300 ($60.01)
`
`150 ($51.75)
`to 300 ($59.26)
`
`2.5/500 ($23.50)
`to 20/2,000
`($94.00)
`
`*—Estimated cost to the pharmacist based on average wholesale prices in Red book, Montvale, NJ; Medical
`Economics Data, 2000. Cost to the patient will be higher based on prescription filling fee.
`
`†—Prices are for a 30-day supply.
`‡—Generic formulation not available.
`Information from Luna B, Hughes AT, Feinglos MN. The use of insulin secretagogues in the treatment of type
`2 diabetes. Prim Care 1999;26:895-915.
`
`MAY 1, 2001 / VOLUME 63, NUMBER 9
`
`www.aafp.org/afp
`
`AMERICAN FAMILY PHYSICIAN
`
`1749
`
`MPI EXHIBIT 1070 PAGE 3
`
`MPI EXHIBIT 1070 PAGE 3
`
`

`

`Management of Type 2 Diabetes
`
`Diagnosis of type 2 DM: use one of three tests (results should be
`confirmed on a subsequent day)
`• RPG (cid:42) 200 mg per dL (11.1 mmol per L) + symptoms
`• FPG (cid:42) 126 mg per dL (7.0 mmol per L)
`• OGT T (75 g) with 2 hr PG (cid:42) 200 mg per dL (11.1 mmol per L)
`
`Patient education/diet and exercise/HBGM
`Goals: FPG < 126 mg per dl (7.0 mmol per L),
`HbA1c < 7 percent; evaluate in three months
`
`Initiate monotherapy if diet and exercise alone are inadequate.
`
`Options for monotherapy
`
`Sulfonylureas
`Target population
`• Recent type 2 DM diagnosis
`• Type 2 DM <5 years’ duration
`
`Meglitinides
`Target population
`• Recent type 2 DM diagnosis
`• Elevated PPG
`
`Biguanides
`Target population
`• Overweight /obese
`• Insulin resistant
`
`Advantages
`• Rapid FPG reduction
`• Low cost
`
`Disadvantages
`• Weight gain
`• (cid:66) Risk of hypoglycemia
`
`Advantages
`• (cid:63) Risk of hypoglycemia
`• Short-acting
`• Meal-adjusted dosing
`
`Disadvantages
`• High cost
`
`Advantages
`• No weight gain
`• (cid:63) Risk of hypoglycemia
`Disadvantages
`• GI side effects
`• High cost
`• Rare lactic acidosis
`
`Thiazolidinediones
`Target population
`• Insulin resistant
`• Overweight /obese
`
`Advantages
`• (cid:63) Amount of insulin
`• (cid:63) Risk of hypoglycemia
`Disadvantages
`• High cost
`• Weight gain
`• Slow onset of action
`• Issue of liver toxicity
`
`Alpha-glucosidase
`inhibitors
`Target population
`• Elevated PPG
`
`Advantages:
`• (cid:63) Risk of hypoglycemia
`Disadvantages
`• High cost
`• GI side effects
`
`Initiate combination therapy if a single agent is inadequate.
`
`Options for combination therapy
`
`Sulfonylurea
`+
`biguanide
`or
`thiazolidinedione
`or
`alpha-glucosidase inhibitor
`
`Biguanide + meglitinide
`
`Biguanide + thiazolidinedione
`
`Biguanide + alpha-
`glucosidase inhibitor
`
`Triple combination therapy
`Sulfonylurea + biguanide +
`thiazolidinedione
`or
`Sulfonylurea + biguanide +
`alpha-glucosidase inhibitor
`
`If therapeutic goals are not met using the above combinations, switch to insulin +/- oral agent.
`
`FIGURE 1. Stepwise approach for the management of type 2 diabetes in patients inadequately controlled with diet and
`exercise. (RPG = random plasma glucose; FPG = fasting plasma glucose; OGTT = oral glucose tolerance test; PG = plasma
`glucose; HbA1c = glycosylated hemoglobin A1c; HBGM = home blood glucose monitoring; DM = diabetes mellitus; GI =
`gastrointestinal; PPG = postprandial glucose)
`
`1750 AMERICAN FAMILY PHYSICIAN
`
`www.aafp.org/afp
`
`VOLUME 63, NUMBER 9 / MAY 1, 2001
`
`MPI EXHIBIT 1070 PAGE 4
`
`MPI EXHIBIT 1070 PAGE 4
`
`

`

`Type 2 Diabetes
`
`When most of the hypoglycemic effects are not observed at
`one half the maximum dose of the sulfonylureas in patients
`who have type 2 diabetes, an alternative agent or combina-
`tion therapy should be considered.
`
`responding at one half the maximum dose, an
`alternative agent or combination therapy
`should be considered.
`
`Meglitinides
`Repaglinide (Prandin) is a new non-sul-
`fonylurea insulin secretagogue agent, the first
`available from the meglitinide class. Nateglin-
`ide (Starlix), the newest member of the class,
`has recently become available. The mecha-
`nism of action of the meglitinides closely
`resembles that of the sulfonylureas. The
`meglitinides stimulate the release of insulin
`from the pancreatic beta cells. However, this
`action is mediated through a different binding
`site on the “sulfonylurea receptor” of the beta
`cell, and the drug has somewhat different
`characteristics when compared with the sul-
`fonylureas. Unlike the commonly used sul-
`fonylureas, the meglitinides have a very short
`onset of action and a short half-life. Repaglin-
`ide has shown similar effects on HbA1c and
`FPG levels when compared with glyburide,
`0.5 to 2 percent and 65 to 75 mg per dL (3.6 to
`4.2 mmol per L), respectively.8 Some potential
`advantages of this class of agents include a
`greater decrease in postprandial glucose and a
`decreased risk of hypoglycemia.
`Because of the short onset of action of the
`meglitinides (15 to 30 minutes), patients
`should be instructed to administer a dose
`immediately before a meal. If a meal is omit-
`ted throughout the day, patients should be
`instructed to skip the corresponding dose to
`prevent hypoglycemia. Likewise, if an extra
`meal is added throughout the day, the patient
`should add a dose to cover that meal.
`Repaglinide can be titrated to a dosage of
`4 mg before each meal (maximum dosage of
`
`globin A1c (HbA1c) levels by 0.8 to 2.0 percent
`and fasting plasma glucose (FPG) concentra-
`tions by 60 to 70 mg per dL (3.3 to 3.9 mmol
`per L), with the greatest reductions observed
`in patients with the highest FPG concentra-
`tions at the initiation of therapy.4,5,8 Hypo-
`glycemia is the most worrisome side effect of
`the sulfonylureas. It is of particular concern
`with agents that are metabolized to an active
`metabolite with significant renal excretion.
`These agents include chlorpropamide (Diabi-
`nese) and glyburide, both of which should be
`avoided in the setting of impaired renal func-
`tion and used with caution in elderly patients.
`Glipizide and glimepiride are associated with
`a lower incidence of hypoglycemia. All sul-
`fonylureas have been associated with weight
`gain and thus, may not be the optimal first
`choice for obese patients.
`Unfortunately, not all patients treated with
`a sulfonylurea will have an adequate response.
`Treatment failure with sulfonylurea therapy
`can be divided into two categories: primary
`and secondary. Primary failure results when a
`patient exhibits an initial poor response to
`sulfonylurea therapy (a decrease in FPG levels
`of less than 20 mg per dL [1.1 mmol per L]).
`Approximately 20 to 25 percent of patients
`with type 2 diabetes will demonstrate pri-
`mary failure to sulfonylurea therapy.12 Sec-
`ondary failure results when the patient
`responds well to treatment initially (a
`decrease in FPG of greater than 30 mg per dL
`[1.7 mmol per L]), but eventually the treat-
`ment fails to maintain adequate control. This
`phenomenon is reported to occur in approx-
`imately 5 to 10 percent of patients per year.12
`Despite these drawbacks, sulfonylureas have
`been shown to be potent and cost-effective
`glucose-lowering agents.
`When initiating sulfonylurea therapy, the
`lowest effective dose should be used and
`titrated to the desired effect at one- to two-
`week intervals. Most of the hypoglycemic
`effects of the sulfonylureas will be observed at
`one half of the maximum dose recommended
`for a specific agent. In patients who are not
`
`MAY 1, 2001 / VOLUME 63, NUMBER 9
`
`www.aafp.org/afp
`
`AMERICAN FAMILY PHYSICIAN
`
`1751
`
`MPI EXHIBIT 1070 PAGE 5
`
`MPI EXHIBIT 1070 PAGE 5
`
`

`

`16 mg per day). Nateglinide can be titrated to
`a dosage of 120 mg before each meal. At least
`one week should be allowed between dosage
`adjustments to adequately assess blood glu-
`cose response. This unique dosing regimen
`may allow greater flexibility for patients who
`have difficulty maintaining a regular meal
`schedule.
`
`Biguanides
`Although biguanides have been in use for
`many years outside the United States, their
`reintroduction in this country has been rela-
`tively recent. Metformin (Glucophage) is cur-
`rently the only agent in this antidiabetic class
`available in this country. Metformin works by
`reducing hepatic glucose output and, to a
`lesser extent, enhancing insulin sensitivity in
`hepatic and peripheral tissues. Metformin has
`been shown to reduce HbA1c levels by approx-
`imately 1.5 to 2.0 percent and FPG levels by 50
`to 70 mg per dL (2.8 to 3.9 mmol per L).4,5,13
`Other effects include a reduction in plasma
`triglyceride levels and low-density lipoprotein
`(LDL) cholesterol levels.
`On the whole, metformin has a favorable
`side effect profile. Most of the related side
`effects (including metallic taste, gastrointesti-
`nal discomfort and nausea) are transient and
`commonly reported only during initiation of
`therapy. Slow-dosage titration is recom-
`
`The Authors
`BEATRIZ LUNA, PHARM.D., BCPS, is assistant professor in the Department of Pharmacy
`Practice at Campbell University School of Pharmacy, Buies Creek, N.C., and clinical phar-
`macist with the Duke Division of Internal Medicine at Durham Regional Hospital, Durham,
`N.C. She received her doctor of pharmacy degree at the University of Illinois, Chicago, and
`subsequently completed a pharmacy practice residency and specialty residency in internal
`medicine and infectious diseases at Duke University Medical Center, Durham, N.C.
`
`MARK FEINGLOS, M.D., C.M., is professor of medicine and chief of the Division of
`Endocrinology, Metabolism and Nutrition, associate professor of psychiatry and behav-
`ioral sciences, and associate professor of pathology at Duke University Medical Center.
`He received his medical degree and CM degree from McGill University, Montreal, Que-
`bec, Canada, and subsequently completed a residency in internal medicine and fel-
`lowship in endocrinology and metabolism at Duke University Medical Center.
`
`Address correspondence to Mark N. Feinglos, M.D., C.M., Box 3921, Duke University
`Medical Center, Durham, NC 27710 (e-mail: feing002@mc.duke.edu). Reprints are not
`available from the authors.
`
`mended to lessen these effects. Taking the
`drug with meals may also lessen the severity
`of the gastrointestinal side effects. Because
`metformin does not affect insulin secretion, it
`is not associated with hypoglycemia when
`used as monotherapy, but can potentiate
`hypoglycemia when used in combination
`with a sulfonylurea or insulin. A rare, but
`more worrisome potential adverse effect is
`that of lactic acidosis. As opposed to phen-
`formin, which was withdrawn from the U.S.
`market during the 1970s, this event is almost
`unknown with metformin in the absence of
`other underlying diseases, particularly renal
`insufficiency. The reported frequency of lactic
`acidosis in all patients is three cases per
`100,000 patient-years.14
`Metformin should not be used in patients
`with elevated serum creatinine levels, specifi-
`cally, 1.4 mg per dL (120 µmol per L) or more
`in women and 1.5 mg per dL (130 µmol per L)
`or more in men. In patients undergoing con-
`trast studies, metformin therapy should be
`withheld for approximately 48 hours follow-
`ing the procedure or until it has been deter-
`mined that renal function has returned to
`baseline. Other situations in which metformin
`therapy should be avoided include cardio-
`genic or septic shock, congestive heart failure
`that requires pharmacologic therapy, severe
`liver disease, pulmonary insufficiency with
`hypoxemia or severe tissue hypoperfusion.13-15
`Metformin is unusual among the oral
`antidiabetic drugs in that therapy has been
`associated with a lack of weight gain and even
`weight loss in some overweight patients.16-18
`Although these observed effects make it an
`ideal first-line agent in overweight patients,
`results from studies have shown that met-
`formin also improves glycemic control in
`patients who are not overweight.16-18 Recent
`reports from the UKPDS group indicate that
`metformin used as monotherapy in obese
`patients may significantly reduce the risk of
`macrovascular complications, although the
`validity of this finding has been questioned.19
`Metformin therapy should be initiated at
`
`1752 AMERICAN FAMILY PHYSICIAN
`
`www.aafp.org/afp
`
`VOLUME 63, NUMBER 9 / MAY 1, 2001
`
`MPI EXHIBIT 1070 PAGE 6
`
`MPI EXHIBIT 1070 PAGE 6
`
`

`

`Type 2 Diabetes
`
`500 mg twice daily with meals and can be
`increased by 500 mg (maximum dosage of
`2,000 mg daily) at two-week intervals to min-
`imize gastrointestinal side effects. A new
`extended release formulation of metformin
`was recently introduced to the market (Glu-
`cophage XR), allowing a more convenient
`once-daily dosing regimen.
`
`Thiazolidinediones
`Until recently, the agents in this drug class
`approved by the U.S. Food and Drug Admin-
`istration (FDA) included troglitazone (Rezu-
`lin), rosiglitazone (Avandia) and pioglitazone
`(Actos). Subsequently, in March 2000, the
`FDA asked the manufacturer (Parke-Davis,
`Warner-Lambert) of troglitazone, the first
`agent in this class to receive labeling approval,
`to remove the product from the market. This
`request occurred following more than 60
`reports of severe liver toxicity in patients tak-
`ing this agent.
`A majority of the data reporting the effi-
`cacy of this class comes from studies with
`troglitazone, although results from more
`recent studies with the newer agents (rosigli-
`tazone and pioglitazone) demonstrate similar
`properties.20-23 The thiazolidinediones work
`by enhancing insulin sensitivity in both mus-
`cle and adipose tissue and to a lesser extent by
`inhibiting hepatic glucose production. These
`agents have a notable effect on improving
`insulin resistance, particularly when used in
`combination with other antidiabetic drugs,
`but have no effect on insulin secretion.
`Monotherapy with these agents has been
`associated with a 0.5 to 1.5 percent reduction
`in HbA1c levels and 25 to 50 mg per dL (1.4 to
`2.8 mmol per L) reduction in FPG levels.4,5 As
`a class, the thiazolidinediones have also been
`shown to alter lipid profiles in patients with
`type 2 diabetes. Results from studies with
`troglitazone consistently show a decrease in
`triglyceride levels—in some cases by as much
`as 33 percent.24,25 The effects on high density
`lipoprotein (HDL) cholesterol levels have
`been either favorable or neutral, while some
`
`studies report an increase in total and LDL
`cholesterol levels.24,25 Newer data reveal that as
`monotherapy, rosiglitazone is associated with
`increases in total, LDL and HDL cholesterol
`levels and either no change or increases in
`triglyceride levels. Patients treated with piogli-
`tazone have displayed mean decreases in
`triglyceride levels, mean increases in HDL
`cholesterol levels, and no consistent mean
`changes in LDL and total cholesterol levels.20,21
`Because these agents do not increase
`insulin secretion, hypoglycemia does not pose
`a risk when thiazolidinediones are taken as
`monotherapy. Significant weight gain has
`been reported with all three agents. The thia-
`zolidinediones are relatively safe in patients
`with impaired renal function because they are
`highly metabolized by the liver and excreted
`in the feces; however, caution should be used
`in patients with hepatic dysfunction because
`troglitazone and its metabolites have been
`shown to accumulate in this setting.
`The manufacturers recommend that these
`agents not be prescribed for patients with
`serum transaminase levels that exceed 2.5
`times the upper limit of normal. Mild to
`moderate edema has been reported in 5 to 7
`percent of patients treated with rosiglitazone
`and pioglitazone.20,21 The increase in plasma
`volume is of concern in patients with conges-
`tive heart failure—particularly those with
`New York Heart Association class III or IV
`functional status. The use of thiazolidine-
`diones should be avoided in these patients.
`As referred to earlier, of greatest concern are
`the reports of an idiosyncratic drug reaction
`with troglitazone. This reaction is initially char-
`acterized by increased serum transaminase lev-
`els, which in some cases progressed to hepatitis,
`hepatic failure and death. Preliminary attempts
`(before troglitazone was withdrawn from the
`market in March 2000) to prevent such inci-
`dents included a request by the FDA that Parke-
`Davis strengthen the drug’s labeling and
`require stringent monitoring of transaminase
`levels in patients taking this agent. In March
`1999, the FDA’s Endocrine and Metabolic
`
`MAY 1, 2001 / VOLUME 63, NUMBER 9
`
`www.aafp.org/afp
`
`AMERICAN FAMILY PHYSICIAN
`
`1753
`
`MPI EXHIBIT 1070 PAGE 7
`
`MPI EXHIBIT 1070 PAGE 7
`
`

`

`Metformin (Glucophage) should not be used to treat
`patients with type 2 diabetes who have elevated serum cre-
`atinine levels, specifically, 1.4 mg per dL [106.8 µmol per L]
`in women and 1.5 mg per dL [114.4 µmol per L] in men.
`
`Drugs Advisory Committee reviewed the status
`of troglitazone and the potential toxicities and
`recommended continued availability in a select
`group of patients: those who are not well con-
`trolled with other antidiabetic agents. Since
`then, it has been determined that patients
`requiring the use of an insulin sensitizer should
`be treated with one of the alternative agents.
`Although results from pre-marketing trials
`revealed no evidence of hepatotoxicity with the
`newer agents (rosiglitazone and pioglitazone),
`two recent case reports demonstrated that
`rosiglitazone may be associated with hepatic
`failure following just 14 days of therapy,
`although a true cause-and-effect relationship
`has not been established.26,27
`The FDA recommends that serum trans-
`aminase levels be monitored every other
`month for the first year in all patients receiv-
`ing a thiazolidinedione. Following one year of
`therapy with the newer agents, the incidence
`of serum transaminase elevations has been
`reported to be similar to placebo.
`The time to achieve a desired effect with
`the thiazolidinediones is somewhat longer
`than the other classes of hypoglycemic agents
`discussed thus far. Intervals of at least three to
`four weeks should be allowed before increas-
`ing the dosage of
`these agents. Smaller
`dosages can be initiated if used as part of a
`combination regimen with a sulfonylurea or a
`sulfonylurea plus metformin. In patients
`receiving insulin therapy, the addition of a
`thiazolidinedione has resulted in significant
`reductions in daily insulin requirements.28,29
`
`Alpha-glucosidase inhibitors
`Acarbose (Precose) and miglitol (Glycet)
`are the two agents available in this class.
`
`Alpha-glucosidase inhibitors act by inhibiting
`the enzyme alpha-glucosidase found in the
`brush border cells that line the small intestine,
`which cleaves more complex carbohydrates
`into sugars. Because they inhibit the break-
`down and subsequent absorption of carbohy-
`drates (dextrins, maltose, sucrose and starch;
`no effect on glucose) from the gut following
`meals, the largest impact of these drugs is on
`postprandial hyperglycemia. Their effect on
`FPG levels is modest. They have been associ-
`ated with a reduction in HbA1c by 0.7 to 1.0
`percent and FPG levels by 35 to 40 mg per dL
`(1.9 to 2.2 mmol per L).4,5,30,31 Thus, these
`agents are most useful in patients who have
`mild FPG elevations or in patients with pre-
`dominant postprandial hyperglycemia.
`The most bothersome side effects observed
`with these agents are gastrointestinal, includ-
`ing abdominal discomfort, bloating, flatulence
`and diarrhea but are reversible with discontin-
`uation. Therapy with acarbose has been linked
`to elevations in serum transaminase levels and
`the use of this agent is contraindicated in
`patients with liver cirrhosis. Likewise, concen-
`trations of the alpha-glucosidase inhibitors
`have been shown to increase proportionally to
`the degree of renal dysfunction and their use in
`patients with a serum creatinine level more
`than 2.0 mg per dL (180 µmol per L) is not
`recommended. Other contraindications in-
`clude patients with inflammatory bowel dis-
`ease or a history of bowel obstruction.32
`Therapy should be initiated with the lowest
`effective dose and titrated slowly over inter-
`vals of two to four weeks. Patients should be
`instructed to take this medication with food.
`For maximum efficacy, the dietary carbohy-
`drate intake should exceed 50 percent.
`Although hypoglycemia is not typically asso-
`ciated with monotherapy with the alpha-glu-
`cosidase inhibitors, it can occur in combina-
`tion with other drugs. It is important,
`therefore, to inform patients that the tradi-
`tional treatment for hypoglycemia may be
`blocked while using this therapy and to con-
`sume only glucose.
`
`1754 AMERICAN FAMILY PHYSICIAN
`
`www.aafp.org/afp
`
`VOLUME 63, NUMBER 9 / MAY 1, 2001
`
`MPI EXHIBIT 1070 PAGE 8
`
`MPI EXHIBIT 1070 PAGE 8
`
`

`

`Type 2 Diabetes
`
`Combination Therapy
`If adequate control is not obtained with the
`use of a single agent, combination therapy is
`an option (Figure 1). Several of the available
`oral agents have been studied in combination
`and have been shown to further improve
`glycemic control when compared to mono-
`therapy.33 As with monotherapy, the choice of
`a second agent should be based on individual
`characteristics. Reasonable combinations of
`agents include a sulfonylurea plus metformin,
`a sulfonylurea plus an alpha-glucosidase
`inhibitor, a sulfonylurea plus a thiazolidine-
`dione, metformin plus repaglinide, biguanide
`plus alpha-glucosidase inhibitor, and met-
`formin plus a thiazolidinedione. A newer for-
`mulation (Glucovance) combines glyburide
`and metformin in one tablet, allowing a more
`convenient dosing schedule.
`Some physicians advocate therapy combin-
`ing three oral agents, (sulfonylurea, met-
`formin, thiazolidinedione or sulfonylurea,
`metformin, alpha-glucosidase inhibitor),
`although this approach has not been exten-
`sively studied.34 Significant data support the
`combination of bedtime insulin with daytime
`sulfonylurea therapy.35,36 Although beyond
`the scope of this review, this combination can
`be quite effective in reducing FPG. Several
`other combinations have recently been
`reviewed.37
`
`Dr. Feinglos has received research support from
`Novartis Pharmaceutical Corp., Novo Nordisk Phar-
`maceutical Industries, Aventis, Eli Lilly and Co., Pfizer
`Pharmaceuticals, Inc., Smithkline Beecham and
`Takeda-Abbott Pharmaceuticals.
`
`REFERENCES
`
`1. Harris MI, Flegal KM, Cowie CC, Eberhardt MS,
`Goldstein DE, Little RR, et al. Prevalence of dia-
`betes, impaired fasting glucose, and impaired glu-
`cose tolerance in U.S. adults. The Third National
`Health and Nutrition Examination Survey, 1988-
`1994. Diabetes Care 1998;21:518-24.
`2. Nathan DM, Kitrick C, Larkin M, Schaffran R,
`Singer DE. Glycemic control in diabetes mellitus:
`have changes in therapy made a difference? Am J
`Med 1996;100:157-63.
`
`3. United Kingdom Prospective Diabetes Study
`(UKPDS) Group. Intensive blood-glucose control
`with sulphonylureas or insulin compared with con-
`ventional treatment and risk of complications in
`patients with type 2 diabetes (UKPDS 33). Lancet
`1998;352:837-53 [Published erratum appears in
`Lancet 1999;354:602].
`4. DeFronzo RA. Pharmacologic therapy for type 2
`diabetes mellitus. Ann Intern Med 1999;131:281-
`303.
`5. Feinglos MN, Bethel MA. Treatment of type 2 dia-
`betes mellitus. Med Clin North Am 1998;82:757-90.
`6. American Diabetes Association. The pharmacolog-
`ical treatment of hyperglycemia in NIDDM. Dia-
`betes Care 1995;18:1510-8.
`7. Berger M, Jorgens V, Muhlhauser I. Rationale for
`the use of insulin therapy alone as the pharmaco-
`logical treatment of type 2 diabetes. Diabetes Care
`1999;22(suppl 3):C71-5.
`8. Luna B, Hughes AT, Feinglos MN. The use of insulin
`secretagogues in the treatment of type 2 diabetes.
`Prim Care 1999;26:895-915.
`9. Report of the Expert Committee on the Diagnosis
`and Classification of Diabetes Mellitus. Diabetes
`Care 1997;20:1183-97.
`10. Feinglos MN, Bethel MA. Therapy of type 2 dia-
`betes, cardiovascular death, and the UGDP. Am
`Heart J 1999;138(5 pt 1):346-52.
`11. Meinert CL, Knatterud GL, Prout TE, Klimt CR. A
`study of the effects of hypoglycemic agents on vas-
`cular complications in patients with adult-onset
`diabetes. II. Mortality results. Diabetes 1970;19
`(suppl):789-830.
`12. Gerich, JE. Oral hypoglycemic agents. N Engl J Med
`1989;321:1231-45 [Published erratum appears in
`N Engl J Med 1990;322:71].
`13. DeFronzo RA, Goodman AM. Efficacy of met-
`formin in patients with non-insulin-dependent dia-
`betes mellitus. The Multicenter Metformin Study
`Group. N Engl J Med 1995;333:541-9.
`14. Stumvoll M, Nurjhan N, Perriello G, Dailey G,
`Gerich JE. The metabolic effects of metformin in
`non-insulin-dependent diabetes mellitus. N Engl J
`Med 1995;333:550-4.
`15. Bailey CJ, Turner RC. Metformin. N Engl J Med
`1996;334:574-9.
`16. Hermann LS, Schersten B, Bitze