1 Drug Class Review Newer Diabetes Medications and Combinations Final Streamlined Update 1 Report June 2014 The purpose of Drug Effectiveness Review Project reports is to make available information regarding the comparative clinical effectiveness and harms of different drugs. Reports are not usage guidelines, nor should they be read as an endorsement of or recommendation for any particular drug, use, or approach. Oregon Health & Science University does not recommend or endorse any guideline or recommendation developed by users of these reports. Agency for Healthcare Research and Quality does not commission, review, or approve reports conducted by Evidence-based Practice Centers for the Drug Effectiveness Review Project. Original Report: February 2011 The literature on this topic is scanned periodically Cynthia Feltner, MD, MPH Roberta Wines, MPH Jennifer Simon, PharmD Cassandra Rowe, BA Rachael E. Posey, MSLS Daniel E. Jonas, MD, MPH Produced by RTI-UNC Evidence-based Practice Center Cecil G. Sheps Center for Health Services Research University of North Carolina at Chapel Hill Daniel E. Jonas, M.D., M.P.H., Director Drug Effectiveness Review Project Marian McDonagh, PharmD, Principal Investigator Pacific Northwest Evidence-based Practice Center Roger Chou, MD, Director Marian McDonagh, PharmD, Associate Director Copyright 2014 by Oregon Health & Science University Portland, Oregon All rights reserved.
2 STRUCTURED ABSTRACT Purpose To compare the effectiveness and adverse event profiles of amylin agonists, DPP-4 inhibitors, GLP-1 analogs, SGLT2 inhibitors, and certain drug combinations (administered as dual therapy or fixed dose combination products) for people with type 2 diabetes and for people with type 1 diabetes for pramlintide only. Data Sources To identify published studies, we searched MEDLINE, The Cochrane Library, Embase, and International Pharmaceutical Abstracts through November 201. We also reviewed reference lists of recent systematic reviews for studies that our electronic searches may have missed, and we requested relevant information from pharmaceutical manufacturers. Review Methods Study selection, data abstraction, validity assessment, grading the strength of the evidence (strength of evidence), and data synthesis were all carried out according to standard Drug Effectiveness Review Project methods. For this streamlined update report, we included only head-to-head comparisons of newer diabetes drugs (and combinations) and comparisons of newer diabetes drugs with metformin. Results Thirty articles met inclusion criteria for this streamlined update report; all were randomized controlled trials focused on adults with type 2 diabetes. All included trials evaluated intermediate outcomes (HbA1c and weight). Very few trials reported health outcomes, and only seven of the 0 included trials were longer than 6 months. Some trials reported some health outcomes (e.g., all-cause mortality or number of people with macrovascular disease) among secondary outcomes or adverse events, but overall evidence was generally insufficient to determine the comparative efficacy of newer diabetes drugs for improving health outcomes. Five trials assessed within-class comparisons of newer diabetes medications. Our metaanalysis ( trials, N=1225) found a greater reduction in mean HbA1c with exenatide XR than exenatide administered twice daily over 24 to 0 weeks: WMD -0.46%; 95% CI, to -0.2 (low strength of evidence); these drugs did not differ in their effect on weight (low strength of evidence). However, injection site reactions were higher in the group receiving exenatide XR weekly compared with exenatide twice daily (moderate strength of evidence). In one trial (N=464), liraglutide 1.8 mg once daily reduced mean HbA1c more than exenatide 10 µg twice daily (between-group difference: 0.%; 95% CI, 0.47 to 0.18); no difference was noted for weight change (low strength of evidence for both outcomes). Those receiving liraglutide had a lower rate of minor hypoglycemia compared with those receiving exenatide 10 µg twice daily (RR 0.55; 95% CI, 0.4 to 0.88) (low strength of evidence). We found insufficient evidence to determine the comparative efficacy or harms of sitagliptin and saxagliptin (1 trial, unknown consistency).
3 Seven trials compared a drug from one class of newer diabetes medications with a drug from a different class, 4 compared a DPP-4 inhibitor with a GLP-1 analog, and compared a DPP-4 inhibitor with an SGLT2 inhibitor. Pooled data from 2 trials (N=75) found that exenatide XR was more efficacious than sitagliptin 100 mg once daily in reducing mean HbA1c over 12 to 26 weeks (WMD -0.48; 95% CI, to -0.26) and weight (WMD -1.2; 95% CI, to -0.76) (low strength of evidence for both outcomes). Exenatide XR was associated with significantly higher rates of nausea, vomiting, and diarrhea compared with sitagliptin (moderate strength of evidence). One trial (N=665) found liraglutide at both dosages (1.2mg and 1.8mg) to be more efficacious than sitagliptin 100 mg in reducing mean HbA1c; the differences in mean HbA1c were -0.4% (95% CI, to -0.16) and -0.60% (95% CI, to -0.4) for both doses of liraglutide compared with sitagliptin, respectively (low strength of evidence). Rates of nausea were higher in those receiving liraglutide (both doses) than in those receiving sitagliptin (low strength of evidence). We found no difference between canagliflozin at both dosages (100 mg and 00 mg) and sitagliptin in reducing mean HbA1c (low strength of evidence); greater reduction in weight was seen with canagliflozin (at both doses) compared with sitagliptin (low strength of evidence). In one trial, women receiving canagliflozin (at both dosages) experienced a higher rate of genital mycotic infections compared to those receiving sitagliptin (low strength of evidence). Eleven trials compared a newer diabetes drug with metformin. Eight trials compared a DPP-4 inhibitor with metformin: linagliptin (1 trial), alogliptin (1 trial), sitagliptin (4 trials) and saxagliptin (2 trials). Moderate strength of evidence supports the conclusion that metformin in doses greater than1000 mg per day is more efficacious than linagliptin 5 mg, alogliptin 25 mg, and sitagliptin 100 mg in reducing HbA1c (between-group differences range from -0.0% to %). For comparisons between metformin and DPP-4 inhibitors, the rates of adverse events were similar (and low) for the vast majority of adverse events; however, metformin was associated with a higher rate of diarrhea compared with alogliptin and sitagliptin (low strength of evidence for both comparisons). We included 2 trials that compared the addition of saxagliptin to uptitration of metformin (in a population not at goal on submaximal doses of metformin); we found no difference between groups for reducing HbA1c (low strength of evidence). However, the addition of saxagliptin was associated with higher rates of hypoglycemia (low strength of evidence). Metformin 1000 mg per day was associated with a greater reduction in weight in all head-to-head trials of DPP-4 inhibitors. Two trials compared a GLP-1 analog with metformin (exenatide twice daily and exenatide XR); we found insufficient evidence to determine the comparative efficacy of metformin and GLP-1 analogs (1 trial each, unknown consistency). We found no difference between dapagliflozin and metformin for reducing HbA1c at either dose of dapagliflozin; the direction of effect favored dapagliflozin, but overall magnitude of effect was small (-0.11% and %) and not within a range that is considered clinically significant. Dapagliflozin was associated with a greater weight reduction over 24 weeks compared with metformin (low strength of evidence). There was no difference between dapagliflozin and metformin in rates of hypoglycemia or nausea (low strength of evidence). Four trials compared either an FDCP or dual therapy with a DPP-4 inhibitor and metformin: alogliptin (1 trial), linagliptin (1 trial), or sitagliptin (2 trials). Alogliptin plus metformin dual therapy was more efficacious than monotherapy for all dose comparisons; between-group differences ranged from -0.44% to -0.99% for all comparisons (moderate strength of evidence); weight changes were similar across all other groups (moderate strength of evidence
4 for no difference). One trial (N=791) found greater reduction in HbA1c with linagliptin plus metformin dual therapy than with component monotherapy over 24 weeks; between-group differences ranged from 0.50% to 1.10% (moderate strength of evidence). Linagliptin 2.5 mg plus metformin 1000 mg twice daily was associated with greater reduction in weight compared with linagliptin 5 mg daily ( 1.00 kg; 95% CI, 1.59 to 0.41) (low strength of evidence). No other groups experienced a significant change in weight (low strength of evidence). Our metaanalysis (2 trials; N=1478) found greater reduction in HbA1c with sitagliptin100 mg plus metformin 2000 mg daily over 18 to 24 weeks compared with metformin monotherapy (WMD %; 95% CI, to -0.45); there was no difference in weight reduction between sitagliptin plus metformin and component monotherapy. Conclusion We found few head-to-head trials within each category of newer diabetes medications; strength of evidence for conclusions related to head to head evidence is generally low for HbA1c, weight, and harms outcomes. Evidence was insufficient to assess the comparative efficacy of included medications for improving health outcomes. Few trials assessed within-class comparisons; we found exenatide XR to be more efficacious for reducing mean HbA1c than exenatide given twice daily, and also associated with more injection site reactions. Liraglutide was more efficacious than exenatide twice daily for reducing mean HbA1c. Metformin (at doses >1500 mg per day) is more efficacious than most DPP-4 inhibitors for reducing HbA1c and is associated with greater weight loss; in some cases, metformin is associated with higher rates of diarrhea. Exenatide XR and liraglutide are both more efficacious than sitagliptin. We found no difference between canagliflozin and sitagliptin for reducing mean HbA1c; canagliflozin is associated with higher rates of genital mycotic infections in women. We found no trials evaluating the comparative efficacy of included medications for subgroups based on age, gender, comorbidity, or drug-drug interaction.
5 TABLE OF CONTENTS INTRODUCTION... 9 Newer Diabetes Medications... 9 Dual Therapy and Fixed-dose Combination Products Purpose and Limitations of Systematic Reviews Scope and Key Questions METHODS Literature Search Study Selection Data Abstraction Validity Assessment Grading the Strength of Evidence Data Synthesis Public Comment RESULTS Overview Key Question 1. What is the comparative efficacy and effectiveness of newer diabetes medications and drug combinations (administered as fixed dose combination products or dual therapy) for children and adults with diabetes mellitus? I. Intra-class Comparisons (within a class) Key Findings Characteristics of Included Studies Detailed Assessment DPP-4 Inhibitors Study Characteristics GLP-1 analogs II. Between-Class Comparisons Key Findings DPP-4 inhibitors compared with GLP-1 analogs DPP-4 inhibitors compared with SGLT2 inhibitors... 2 Characteristics of Included Studies... 2 Detailed Assessment DPP-4 inhibitors compared with GLP-1 analogs DPP-4 inhibitors compared with SGLT2 inhibitors III. Newer Diabetes Medications compared with Metformin Key Findings DPP-4 inibitors compared with metformin GLP-1 analogs compared with metformin SGLT2 inhibitors compared with metformin Detailed Assessment... DPP-4 inibitors compared with metformin... GLP-1 analogs compared with metformin... 5 SGLT2 inhibitors compared with metformin... 5 IV. Fixed-dose Combination Products (FDCPs) or Dual Therapy... 6 Key Findings... 6 Characteristics of Included Studies... 7 Detailed Assessment Kazano or dual therapy with alogliptin plus metformin Jentadueto or dual therapy with linagliptin plus metformin Janumet or dual therapy with sitagliptin plus metformin Key Question 2. What is the comparative tolerability and frequency of adverse events for newer diabetes medications and drug combinations (administered as fixed dose combination products or dual therapy) for children and adults with diabetes mellitus?... 42
6 I. Intra-class Comparisons (within a class) Key Findings: Harms DPP-4 inhibitors: Sitagliptin compared with saxagliptin GLP1-analogs: Exenatide 10 µg compared with liraglutide 1.8 mg GLP-1 analogs: Exenatide XR compared with exenatide 10 µg Characteristics of Included Studies... 4 Detailed Assessment of for Intra-class Comparisons: Harms... 4 Sitagliptin compared with saxagliptin... 4 Exenatide 10 µg twice daily compared with liraglutide 1.8 mg once daily... 4 Exenatide XR 2 mg once weekly compared with exenatide 10 µg twice daily II. Between-Class Comparisons: Harms Key Findings: Harms DPP-4 inhibitors compared with GLP-1 analogs DPP-4 inhibitors compared with SGLT2 inhibitors Study Characteristics Detailed Assessment for Between-class Comparisons: Harms DPP-4 inhibitors compared with GLP-1 analogs DPP-4 inhibitors compared with SGLT2 inhibitors III. Newer Diabetes Medications compared with metformin: Harms Key Findings: Harms DPP-4 inhibitors compared with metformin GLP-1 analogs compared with metformin SGLT2 inhibitors compared with metformin Study Characteristics Detailed Assessment for Newer Diabetes Medications compared with Metformin: Harms DPP-4 inhibitors compared with metformin Abbreviations: CI, confidence interval; RR, relative risk GLP-1 analogs compared with metformin SGLT2 inhibitor compared with metformin IV. Fixed-dose Combination Products (FDCPs) or Dual Therapy Key Findings: Harms DPP-4 inhibitors in combination with metformin Study Characteristics Detailed Assessment for FDCPs and Dual Therapy Kazano or dual therapy with alogliptin plus metformin Jentadueto or dual therapy with linagliptin plus metformin Janumet or dual therapy with sitagliptin plus metformin Key Question. Are there subgroups of patients based on demographics (age, racial groups, gender), comorbidities (drug-disease interactions, obesity), or other medications (drug-drug interactions) for which newer diabetes medications and drug combinations (administered as fixed dose combination products or dual therapy) differ in efficacy/effectiveness or tolerability and frequency of adverse events? SUMMARY Strength of Evidence Limitations of this Report Applicability Trials in Progress REFERENCES... 69
7 TABLES Table 1. Characteristics of included drugs Table 2. Eligible drugs and comparators... 1 Table. Study inclusion and exclusion criteria Table 4. Definitions of the grades of overall strength of evidence Table 5. Characteristics of included studies: Newer diabetes medications within-class comparisons Table 6. Characteristics of included studies: Newer diabetes medications between-class comparisons.. 24 Table 7. Randomized controlled trials comparing canagliflozin and sitagliptin: Results for hemoglobin A1c and weight Table 8. Characteristics of included studies: Newer diabetes medications compared with metformin... 0 Table 9. Characteristics of included studies: Fixed-dose combination products or dual therapy... 8 Table 10. Adverse events of intra-class comparisons Table 11. Meta-analysis comparing adverse events: Exenatide XR to sitagliptin 100 mg Table 12. Adverse events of between-class comparisons: Sitagliptin compared with exenatide XR or liraglutide Table 1. Adverse events of between-class comparisons: Sitagliptin compared with canagliflozin Table 14. Meta-analysis comparing adverse events: Linagliptin 5 mg compared with metformin 1000 mg twice daily Table 15. Meta-analysis comparing adverse events: Sitagliptin compared with metformin... 5 Table 16. Adverse events of sitagliptin compared with metformin... 5 Table 17. Changes in lipid parameters (mean change from baseline, mg/dl): Sitagliptin compared with metformin Table 18. Meta-analysis comparing adverse events: Saxagliptin compared with metformin Table 19. Meta-analysis comparing adverse events: Dapagliflozin 5 mg compared with metformin XR Table 20. Meta-analysis comparing adverse events: Dapagliflozin 10 mg compared with metformin XR. 56 Table 21. Adverse events of Kazano or dual therapy with alogliptin and metformin compared with metformin monotherapy Table 22. Adverse events of Jentadueto or dual therapy with linagliptin and metformin compared with linagliptin or metformin monotherapy Table 2. Adverse events of Janumet or dual therapy with sitagliptin and metformin compared with sitagliptin or metformin monotherapy Table 24. Changes in lipid parameters (mean change from baseline, mg/dl): sitagliptin plus metformin compared with metformin monotherapy Table 25. Summary of evidence by Key Question FIGURES Figure 1. Results of literature search APPENDIXES Appendix A. Boxed warnings for included drugs... 7 Appendix B. Glossary Appendix C. Search strategies Appendix D. Excluded studies and studies of poor quality Appendix E. Meta-analyses Appendix F. Strength of evidence EVIDENCE TABLES Evidence tables are published in a separate document.
8 Acknowledgments We thank Leah Williams, our publications editor, for putting this report into its present form for you to read. We also thank Ellen Shanahan for assistance with project coordination, Claire Baker for assistance with full-text article identification and retrieval, and Christiane Voisin for assistance with the literature search. Suggested citation for this report Feltner C, Wines R, Simon J, Rowe C, Posey R, Jonas D. Drug class review: Newer diabetes medications and combinations. Prepared by the RTI-UNC Evidence-based Practice Center for the Drug Effectiveness Review Project. Oregon Health & Science University. Portland, OR Funding The funding source, the Drug Effectiveness Review Project, is supported by 12 State Medicaid programs. These organizations selected the topic and had input into the Key Questions for this review. The content and conclusions of the review are entirely determined by the Evidence-based Practice Center researchers. The authors of this report have no financial interest in any company that makes or distributes the products reviewed in this report.
9 INTRODUCTION Diabetes mellitus (diabetes) is a chronic disease associated with significant morbidity and healthcare costs. The prevalence of diabetes has increased substantially over the past 2 decades; in 2010, approximately 21 million adults in the United States had diabetes, based on self-report or hemoglobin A1c (HbA1c) levels 6.5%. 1 Among people diagnosed with diabetes, 90% to 95% have type 2 diabetes, while 5% to 10% have type 1 diabetes. 2 Type 1 diabetes is characterized by autoimmune destruction of beta cells of the pancreas resulting in absolute insulin deficiency. Type 2 diabetes encompasses a heterogeneous group of disorders characterized by slow progressive loss of beta cell function and mass, leading to variable degrees of insulin resistance, impaired insulin secretion, and increased hepatic glucose production. Higher glucagon levels relative to insulin also play a significant role in the pathogenesis and management of type 2 diabetes. The 2014 American Diabetes Association treatment guidelines recommend an HbA1c goal of <7% for most nonpregnant adults in order to prevent adverse microvascular and macrovascular outcomes. The guidelines acknowledge that less stringent (HbA1c <8%) or more stringent (A1c < 6.5%) goals may be appropriate for certain populations. Insulin is the standard treatment for type 1 diabetes. Pharmacologic options for type 2 diabetes include sulfonylureas, biguanides, thiazolidinediones, meglitinides, alpha-glucosidase inhibitors, dipeptidyl peptidase-4 (DPP-4) inhibitors, glucagon-like peptide-1 (GLP-1) analogs, sodium-glucose cotransporter 2 (SGLT2) inhibitors, combination products, and insulin. Because of the progressive nature of diabetes, practitioners and patients often experience challenges in reaching and sustaining American Diabetes Association treatment goals. Patients with type 2 diabetes often need to take more than one type of diabetes medication. In 2005 and 2006, 5% of all patients with diabetes were taking 2 classes of antidiabetes medications, and 14% were taking or more classes. 4 Newer Diabetes Medications Within recent years, several new antihyperglycemic agents have been approved (Table 1). These agents offer mechanisms of glycemic control beyond that of traditional oral agents and insulin by targeting alternate gluco-regulatory receptors and hormones such as amylin, GLP-1, glucosedependent insulinotropic peptide (GIP), DPP-4, and sodium-glucose cotransporter 2 (SGLT2). For the purposes of this report, we consider the following to be newer diabetes medications : amylin agonists, DPP-4 inhibitors, GLP-1 analogs, and SGLT2 inhibitors. Amylin is a neuroendocrine hormone co-secreted with insulin from beta cells in response to elevated blood glucose concentrations, and it complements the actions of insulin. GLP-1 and GIP are secreted by L- and K-type cells in the intestinal tract in response to a combination of endocrine and neural signals initiated by the entry of food into the gut. Secretion of GLP-1 and GIP enhance insulin release. Both endogenous GLP-1 and GIP are rapidly degraded by the proteolytic enzyme DPP- 4. SGLT2 is located in the proximal renal tubules and is the main site of filtered glucose reabsorption from the tubular lumen. Inhibition of SGLT2 results in increased urinary excretion of glucose and reduced plasma glucose concentrations.
10 Dual Therapy and Fixed-dose Combination Products For this report, we ve included 5 fixed-dose combination products (FDCPs) approved for the treatment of type 2 diabetes; all combine a DPP-4 inhibitor with metformin (Table 1). In addition to the 5 FDCPs, we ve included studies of the individual components of those FDCPs when used together but in separate pills we refer to this as dual therapy throughout the review. We only evaluate dual therapy when there is a US Food and Drug Administration-approved fixed dose combination product. Table 1. Characteristics of included drugs Drug Class Trade name Administration Labeled indications FDA-approved dosing Type 1: µg with Pramlintide Symlin Type 1 diabetes, meals Type 2 diabetes; Amylin agonist Injectable Type 2: µg with Adjunct with insulin meals Linagliptin DPP-4 Inhibitor Alogliptin DPP-4 Inhibitor Sitagliptin DPP-4 Inhibitor Saxagliptin DPP-4 Inhibitor Exenatide GLP-1 Analog Exenatide XR GLP-1 Analog Liraglutide GLP-1 Analog Canagliflozin SGLT2 Inhibitor Dapagliflozin SGLT2 Inhibitor Alogliptin + Metformin Linagliptin + Metformin Tradjenta Oral tablet Nesina Oral tablet Januvia Oral tablet Onglyza Oral tablet Byetta Injection Bydureon Injection Victoza Injection Invokana Oral tablet Farxiga Oral tablet Kazano Oral tablet Jentadueto Oral tablet Type 2 diabetes; Monotherapy or combination with many antihyperglycemics Type 2 diabetes; Monotherapy or combination with many antihyperglycemics Type 2 diabetes; Monotherapy or combination with many antihyperglycemics Type 2 diabetes; Monotherapy or combination with many antihyperglycemics Type 2 diabetes; Not recommended with insulin, except for insulin glargine Type 2 diabetes; Not recommended with insulin Type 2 diabetes; Not recommended with shortacting insulin (has not been studied) Type 2 diabetes; Monotherapy or combination with many antihyperglycemics Type 2 diabetes; Adjunct to diet and exercise Type 2 diabetes; Adjunct to diet and exercise for patients not adequately controlled on alogliptin or metformin monotherapy Type 2 diabetes; Adjunct to diet and exercise in patients when treatment with both linagliptin and metformin is appropriate 5 mg once daily 25 mg once daily (6.25 or 12.5 mg if renal dysfunction) 100 mg once daily (25 or 50 mg if renal dysfunction) mg once daily (2.5 mg if renal dysfunction) 5 or 10 µg twice daily prior to meals 2 mg once weekly 0.6, 1.2, or 1.8 mg once daily 100 or 00 mg once daily prior to first meal (100 mg if renal dysfunction) 5 or 10 mg once daily in the morning 12.5/ /1000 mg one to two times daily with meals 2.5/500 mg, 2.5/850 mg, or 2.5/1000 mg one to two times daily with meals Sitagliptin + Janumet Type 2 diabetes; 50/500 mg or 50 mg/1000
11 Drug Class Trade name Administration Labeled indications FDA-approved dosing Metformin Oral tablet Adjunct to diet and exercise in adult patients when treatment with both sitagliptin and metformin is appropriate mg once or twice daily with meals Sitagliptin XR + Metformin Saxagliptin + Metformin Janumet XR Oral tablet Kombiglyze XR Oral tablet Type 2 diabetes; Adjunct to diet and exercise in adult patients when treatment with both sitagliptin and metformin is appropriate Type 2 diabetes; Adjunct to diet and exercise when treatment with both saxagliptin and metformin is appropriate 50/500 mg, 50/1000 mg, or 100/1000 mg one to two tablets once daily with meals; Maximum of 100/2000 mg daily 2.5/1000 mg, 5/500, or 5/1000, one to two tablets once daily with evening meal; Maximum of 5/2000 mg daily Abbreviations: DPP-4, Dipeptidyl peptidase-4; GLP-1, glucagon-like peptide-1; SGLT2, Sodium-glucose cotransporter 2 Purpose and Limitations of Systematic Reviews Systematic reviews, also called evidence reviews, are the foundation of evidence-based practice. They focus on the strength and limits of evidence from studies about the effectiveness of a clinical intervention. Systematic reviews begin with careful formulation of research questions. The goal is to select questions that are important to patients and clinicians, then to examine how well the scientific literature answers those questions. Terms commonly used in systematic reviews, such as statistical terms, are provided in Appendix B and are defined as they apply to reports produced by the Drug Effectiveness Review Project. Scope and Key Questions The purpose of this review is to assist healthcare providers, researchers, and policy makers in making clinical decisions, creating formularies, and developing policies regarding newer medications for the treatment of diabetes based on the most current available literature. We compare the efficacy, effectiveness, and tolerability of newer diabetes medications and combinations, and also look for subgroups that may differ in these areas. In the previous update report, 5 placebo-controlled comparisons were included as part of the evidence base. A new, streamlined approach was used for this update, which focuses on head-to-head studies. We compare efficacy and tolerability both within classes of newer diabetes medications and between the classes of newer diabetes medications; we also compare newer diabetes medications with metformin. For this update, we do not include comparisons of newer diabetes medications with pioglitazone or sulfonylureas. We include trials enrolling populations of diabetes patients who are already on treatment with insulin or other oral medications for diabetes as long as they are randomized to an eligible drug and comparator (Table 2). We developed preliminary key questions to identify the populations, interventions, outcomes of interest, and eligibility criteria for studies. A draft of these questions and inclusion and exclusion criteria were posted on the Drug Effectiveness Review Project (DERP) website for public comment. The draft was reviewed and revised by representatives of the organizations participating in the DERP. These organizations approved the following key questions to guide the review for this report:
12 1. What is the comparative efficacy and effectiveness of newer diabetes medications and drug combinations (administered as fixed dose combination products or dual therapy) for children and adults with diabetes mellitus? 2. What is the comparative tolerability and frequency of adverse events for newer diabetes medications and drug combinations (administered as fixed dose combination products or dual therapy) for children and adults with diabetes mellitus?. Are there subgroups of patients based on demographics (age, racial groups, gender), comorbidities (drug-disease interactions, obesity), or other medications (drug-drug interactions) for which newer diabetes medications and drug combinations (administered as combination products or dual therapy) differ in efficacy/effectiveness or frequency of adverse events? METHODS Literature Search To identify articles relevant to each key question, we searched MEDLINE, Embase, the Cochrane Library, and the International Pharmaceutical Abstracts. To identify new trials of drugs in the last report, we searched from December 1, 2008 (1 year before the final searches were run for the previous report 5 ) through November 5, 201. To identify new trials of linagliptin, we searched from September 1, 2011 (1 year before the final searches were run for the Linagliptin Single Drug Addendum 6 ) through November 5, 201. To identify trials of newly approved drugs, we did not limit the search date. To identify trials of dapagliflozin, we conducted additional searches through February 7, 2014, as dapagliflozin was approved after the original searches were completed. We used the generic and brand names of included drugs, and study designs as search terms, limiting the electronic searches to human and English language. We combined the results of all the searches and removed duplicate references. The full search strategies are presented in Appendix C. We attempted to identify additional studies through hand searches of reference lists of included studies and reviews. In addition, we searched the FDA s Center for Drug Evaluation and Research (CDER) for medical and statistical reviews. Finally, we requested dossiers of published and unpublished information from the relevant pharmaceutical companies for this review. All received dossiers were screened for studies or data not found through other searches. All citations were imported into an electronic database (Endnote X.0.2, Thomson Reuters). Study Selection Selection of included studies was based on the inclusion criteria in Table. Studies meeting these criteria and comparing at least one of the drugs of interest with an eligible comparator were included. Eligible drugs and comparators are listed in Table 2. Two reviewers independently assessed titles and abstracts of citations identified through literature searches for eligibility using the criteria below. Full-text articles of potentially relevant citations were retrieved and were
13 again assessed for eligibility by both reviewers. Disagreements were resolved by consensus. Results published only in abstract form were not included because inadequate details were available for quality assessment. Table 2. Eligible drugs and comparators Drug class or drug a Eligible comparators Amylin Agonists DPP-4 Inhibitors, GLP-1 Analog, SGLT2 Inhibitors, Fixed dose combination products, Dual therapy with the component Pramlintide vs. medications of fixed dose combination products, Biguanides (metformin) DPP-4 Inhibitors Each other, Amylin Agonists, GLP-1 Analog, SGLT2 Inhibitors, Sitagliptin, Saxagliptin, Linagliptin, Fixed dose combination products, Dual therapy with the component or Alogliptin vs. medications of fixed dose combination products, Biguanides (metformin) GLP-1 Analog Exenatide, Exenatide XR, or Liraglutide vs. SGLT2 Inhibitors Canagliflozin or Dapagliflozin vs. Fixed-dose Combination Products b Janumet (Metformin + Sitagliptin), Janumet XR (Metformin + Sitagliptin XR), Kombiglyze XR (Metformin + Saxagliptin), Kazano (Metformin + Alogliptin), or Jentadueto (Metformin + Linagliptin) vs. Dual Therapy Metformin + Sitagliptin, Metformin + Sitagliptin XR, Metformin + Saxagliptin, Metformin + Alogliptin, or Metformin + Linagliptin vs. Each other, DPP-4 Inhibitors, Amylin agonists, SGLT2 Inhibitors, Fixed dose combination products, Dual therapy with the component medications of fixed dose combination products, Biguanides (metformin) Each other, Amylin agonists, DPP-4 Inhibitors, GLP-1 Analogs, Fixed dose combination products, Dual therapy with the component medications of included fixed dose combination products, Biguanides (metformin) Monotherapy with one of the component medications of the product or head to head studies comparing 2 fixed dose combination products Monotherapy with one of the component medications Abbreviations: DPP-4, Dipeptidyl peptidase-4; GLP-1, glucagon-like peptide-1; SGLT2, Sodium-glucose cotransporter 2 a Evidence of these interventions (left column) used as monotherapy or as add-on therapy will be included when available. b In Drug Effectiveness Review Project reports, we traditionally refer to the drug products by their generic names wherever possible. For this report, however, we are using the trade names for the FDCPs in an effort to make reading easier.
14 Table. Study inclusion and exclusion criteria Included populations Adults with type 2 diabetes for all included medications Adults and children with type 1 diabetes for Pramlintide (Symlin ) only Excluded populations Individuals with gestational diabetes, pre-diabetes (impaired fasting glucose or impaired glucose tolerance), metabolic syndrome without diabetes, or polycystic ovary syndrome Included intermediate outcomes Hemoglobin A1c (HbA1c) Changes in weight Included health outcomes All-cause mortality Microvascular disease: chronic kidney disease, including renal dialysis, renal transplantation, end-stage renal disease; renal failure with proteinuria, retinopathy including proliferative retinopathy and blindness; peripheral neuropathy Macrovascular disease: cardiovascular events, cardiovascular morbidity (e.g. myocardial infarction and peripheral arterial disease), cardiovascular mortality, stroke/transient ischemic attack, coronary heart disease, cardiovascular procedures, extremity amputation Included harms/adverse events outcomes Overall adverse events Withdrawals due to adverse events Major adverse events (for example diabetic ketoacidosis, non-ketotic hyperosmolar coma) Specific adverse events (for example cancers/neoplasms, infections, hypoglycemia, liver toxicity, liver function abnormalities, gastrointestinal effects, congestive heart failure, adverse changes in lipid concentrations, pancreatitis, weight gain, fractures) Included study designs For intermediate outcomes: randomized controlled trials a For health outcomes: randomized controlled trials For harms: randomized controlled trials, controlled clinical trials, population-based comparative cohort studies focused on adverse events, case-control studies, and reports from voluntary adverse event reporting systems. Duration: 12 weeks of follow-up Sample size: No limit for randomized controlled trials, observational studies with at least 100 subjects a We included extension trials (e.g., longer term follow-up) of randomized controlled trials when patients were maintained on the treatment they were initially randomized to. We considered groups that were switched from one treatment to another during the follow-up period to be an ineligible comparison. Data Abstraction The following data were abstracted from included trials: study design; population characteristics, including sex, age, and ethnicity; eligibility and exclusion criteria; interventions; comparisons; numbers randomized or treated, and the numbers analyzed; and results for each outcome. We recorded intention-to-treat results when reported. If true intention-to-treat results were not reported, but loss to follow-up was very small, we recorded these results and noted that they were modified intention-to-treat results. In cases where only per protocol results were reported, we recorded these results and noted that they were per protocol results. We considered whether results were intention-to-treat, modified intention-to-treat, or per protocol when assessing the internal validity of studies (as described below). Data abstraction was performed by one reviewer and independently checked by a second reviewer, and differences were resolved by consensus. When studies reported duration in number of months, we converted this to number of weeks by multiplying months by 4. and rounding up or down. Number of weeks is presented in the tables of study characteristics throughout the report. When recording data on lipids, we
15 converted mmol/l to mg/dl. To convert total cholesterol and HDL and LDL cholesterol, we used the following formula: divide mmol/l by to get mg/dl. To convert triglycerides, we used the following formula: divide mmol/l by to get mg/dl. When included trials had non-eligible comparators (e.g., placebo comparisons or non-fda approved dosages), we extracted that data into evidence tables (published in a separate document) but did not synthesize or report that data in the results. Validity Assessment Two reviewers independently assessed each study, and differences were resolved by consensus. We assessed the internal validity (quality) of trials based on the predefined criteria (see These criteria are based on those developed by the US Preventive Services Task Force and the National Health Service Centre for Reviews and Dissemination (United Kingdom). 7,8 We rated the internal validity of each trial based on the methods used for randomization, allocation concealment, and blinding; the similarity of compared groups at baseline; maintenance of comparable groups; adequate reporting of dropouts and attrition; loss to follow-up; and the use of intention-to-treat analysis. Trials that had a fatal flaw were rated poor quality; trials that met all criteria were rated good quality; the remainder were rated fair quality. As the fair-quality category is broad, studies with this rating vary in their strengths and weaknesses: The results of some fair-quality studies are likely to be valid, while others are only possibly valid. A poor-quality trial is not valid; the results are at least as likely to reflect flaws in the study design as a true difference between the compared drugs. A fatal flaw may be reflected by one aspect introducing a high risk of bias or by failure to meet combinations of items of the quality assessment checklist. We did not include poor quality studies in our qualitative or quantitative analyses; however, data from poor quality trials is reported in evidence tables (Evidence Tables are published in a separate document). A particular randomized trial might receive 2 different ratings (e.g., one rating efficacy/effectiveness outcomes and one for harms). In these cases, data from one study may be included for one KQ but not included in another KQ. We only rated the quality of trials; although observational studies were eligible for outcomes related to adverse events, we did not identify any that met our inclusion criteria. Grading the Strength of Evidence We graded strength of evidence based on the guidance established for the Evidence-based Practice Center Program of the Agency for Healthcare Research and Quality. 9 Developed to grade the overall strength of a body of evidence, this approach incorporates 4 key domains: risk of bias (includes study design and aggregate quality), consistency, directness, and precision of the evidence. It also considers other optional domains that may be relevant for some scenarios, such as a dose-response association, or plausible confounding that would decrease the observed effect, strength of association (magnitude of effect), publication bias, and selective analysis reporting (e.g. presenting a point estimate without providing a measure of dispersion). We considered all evidence from intermediate outcomes (e.g. HbA1c and weight) to be indirect (not directly reporting health outcomes). Table 4 describes the grades of evidence that can be assigned. Grades reflect the strength of the body of evidence to answer key questions on the comparative effectiveness, efficacy, and harms of the drugs included in this review. Grades do not refer to the general efficacy or effectiveness of pharmaceuticals. Two reviewers assessed
16 each domain for each outcome and differences were resolved by consensus. For this streamlined update report, we did not modify SOE grades for comparisons assessed in prior DERP reports when we found no new trials relevant to a particular comparison. In these cases, SOE grades may reflect evidence that was previously reviewed but not part of the scope of this review (e.g., doseresponse association). We reference the prior DERP for these SOE grades in a footnote to the SOE appendix. We graded the strength of evidence for the outcomes deemed to be of greatest importance to decision makers and those most commonly reported in the literature. For example, these included HbA1c and weight changes, among others. Because of time and resource constraints, we did not grade the strength of evidence for every possible outcome reported everywhere in the included literature. Table 4. Definitions of the grades of overall strength of evidence Grade Definition 9 We are very confident that the estimate of effect lies close to the true effect for High this outcome. The body of evidence has few or no deficiencies. We believe that the findings are stable, i.e., another study would not change the conclusions We are moderately confident that the estimate of effect lies close to the true Moderate effect for this outcome. The body of evidence has some deficiencies. We believe that the findings are likely to be stable, but some doubt remains. We have limited confidence that the estimate of effect lies close to the true effect for this outcome. The body of evidence has major or numerous deficiencies (or both). We believe that additional evidence is needed before concluding either that the findings are stable or that the estimate of effect is close to the true effect. We have no evidence, we are unable to estimate an effect, or we have no Insufficient confidence in the estimate of effect for this outcome. No evidence is available or the body of evidence has unacceptable deficiencies, precluding reaching a conclusion. Data Synthesis We constructed evidence tables showing the study characteristics, quality ratings, and results for all included studies (published in a separate document). We conducted meta-analyses of outcomes reported by at least 2 studies that were homogeneous enough to justify combining their results. In order to determine whether metaanalysis could be meaningfully performed, we considered the quality of the studies and the heterogeneity among studies in design, patient population, interventions, and outcomes. When meta-analysis could not be performed, the data were summarized qualitatively. Random-effects models were used to estimate pooled effects. 10 Forest plots graphically summarize results of individual studies and of the pooled analysis. 11 For continuous outcomes (e.g., HbA1c and weight) measured with the same scale, we report the weighted mean difference (WMD) between intervention and control subjects. For binary outcomes, we report the relative risk (RR) and 95% confidence interval (CI). The Chi-squared statistic and the I 2 statistic (the proportion of variation in study estimates due to heterogeneity) were calculated to assess heterogeneity in effects between studies. 12,1 An I 2 from 0 to 40% might not be important, 0% to 60% may represent moderate heterogeneity, 50% to 90% may represent substantial heterogeneity, and 75% represents considerable heterogeneity. 14 The importance of the observed value of I 2 depends on the magnitude and
17 direction of effects and on the strength of evidence for heterogeneity (e.g. P value from the chisquared test, or a confidence interval for I 2 ). Whenever including a meta-analysis with considerable statistical heterogeneity in this report, we provide an explanation for doing so, considering the magnitude and direction of effects. 14 Potential sources of heterogeneity were examined by analysis of subgroups of study design, study quality, patient population, and variation in interventions. Quantitative analyses were conducted using Stata version When meta-analysis could not be performed, the data were summarized qualitatively. For continuous outcomes (changes in HbA1c and weight), we report both the mean change from baseline for both groups and the between-group difference when these were reported. When only mean change in hemoglobin A1c from baseline and standard error or standard deviation were reported for 2 comparisons, we calculated the difference in means (linagliptin-placebo) and 95% confidence interval if not already reported by the included trials. When describing conclusions and key findings in this report, we sometimes refer to no difference between 2 treatments. We use this wording to indicate that the available evidence did not support a statistically or clinically significant difference between the 2 treatments. For HbA1c outcomes, we note whether differences between groups were statistically significant. When between-group differences in HbA1c were less than 0.%, we did not consider this a clinically meaningful difference; this cut-off value is consistent with what was considered clinically significant in prior DERP reports. 5 For weight outcomes, we did not set a cut off at which we considered a change in weight to be clinically meaningful. We note the magnitude of effect and describe the SOE for the observed effect as discussed above. Public Comment This report was posted to the Drug Effectiveness Review Project website for public comment. We received comments from 5 persons representing pharmaceutical companies. RESULTS Overview Literature searches through November 4, 201 for the current report identified 749 unduplicated citations. We received dossiers from 5 pharmaceutical manufacturers: AstraZeneca, Boehringer- Ingelheim, Janssen, Novo Nordisk, and Merck. Ninety-six additional references were identified through hand searches of systematic reviews and other sources, and 52 additional articles were identified from the dossiers, 17 from the exenatide (Byetta ) dossier (AstraZeneca), from the exenatide XR (Bydureon ) dossier (AstraZeneca), 8 from the liraglutide (Victoza ) dossier (Novo Nordisk), 2 from the canagliflozin (Invokana ) dossier (Janssen), 1 from the Jentadueto dossier (Boehringer-Ingelheim), 1 from the Kombiglyze TM dossier (AstraZeneca), 2 from the pramlintide (Symlin ) dossier (AstraZeneca), 4 from the saxagliptin (Onglyza ) dossier (AstraZeneca), 7 from the sitagliptin (Januvia ) dossier (Merck), and 7 from the dapagliflozin (Farxiga TM ) dossier (AstraZeneca ). We also retrieved 4 excluded references from the reference database of the Newer Diabetes Medications, TZDs, and Combinations DERP report 5 in order to review these publications using new inclusion criteria. From all of these sources, we had a total of 1024 references. In addition to these, we carried forward 12 of the included studies from 5
18 previous DERP reports: Newer Drugs for the Treatment of Diabetes Mellitus, 15 Fixed Dose Combination Drug Products for the Treatment of Type 2 Diabetes and Hyperlipidemia, 16 and the Drug Class Review on Thiazolidinediones, 17 Newer Diabetes Medications, TZDs, and Combinations, 5 Newer Diabetes Medications, TZDs, and Combinations Single Drug Addendum: Linagliptin. 6 By applying the eligibility and exclusion criteria to titles and abstracts of all identified citations, we obtained full-text copies of 266 citations. After re-applying the criteria for inclusion, we ultimately included 21 new publications from our recent literature searches and other sources, plus the 9 includes from previous reports. See Appendix D for a list of excluded studies and reasons for exclusion at the full text stage. Figure 1 shows the flow of study selection. Among the 21 includes from our recent searches, all 21 were randomized controlled trials. Among these, we rated 2 as good quality, 18 as fair quality, and 1 as poor quality. Poor-quality studies are listed in Appendix D. Figure 1. Results of literature search 749 unduplicated references identified from recent searches 275 additional references identified through other sources (127 included references from previous reports, 148 from dossiers and hand searches) 1024 titles and abstracts screened for report 758 references excluded at abstract level 266 full text articles assessed for eligibility in current report (12 references carried forward from previous reports + 14 full-text includes from recent searches) 0 articles (26 studies) included in the review 8 (5) from previous reports, 22 (21) from recent searches 26 full-text articles excluded 17 ineligible publication type ineligible drug or duplicate publication already included in earlier diabetes medication reports 184 ineligible comparison 11 no eligible outcomes reported 8 ineligible study design 6 ineligible population 7 full text not available The 22 articles from recent searches included: 21 trials (22 articles), and 0 observational studies (2 good, 18 fair, 1 poor)
19 Key Question 1. What is the comparative efficacy and effectiveness of newer diabetes medications and drug combinations (administered as fixed dose combination products or dual therapy) for children and adults with diabetes mellitus? I. Intra-class Comparisons (within a class) Key Findings We found 5 fair-quality trials comparing 2 medications within the same class: one compared saxagliptin with sitagliptin and 4 compared 2 different GLP-1 analogs. No trial assessed health outcomes as a primary outcome or evaluated outcomes beyond 6 months; we found insufficient evidence to determine the comparative efficacy for improving health outcomes for all intra-class comparisons. One trial (N=801) found no difference between sitagliptin and saxagliptin for reducing HbA1c or weight over 18 weeks. Evidence was insufficient to determine the comparative efficacy of sitagliptin and saxagliptin for reducing HbA1c or weight (1 trial; unknown consistency). We pooled data from trials (N=1225) comparing exenatide XR with exenatide administered twice daily over 24-0 weeks. Exenatide XR was more efficacious in reducing mean HbA1c than exenatide twice daily: WMD -0.46%; 95% CI, to -0.2 (moderate strength of evidence). Three trials found no difference between exenatide XR and exenatide administered twice daily for weight changes over 24-0 weeks; 1 trial (N=252) two trials found no statistically significant difference between groups, the third trial (N=678) found a small reduction in weight (-0.kg; p<0.001) favoring exenatide twice daily (low strength of evidence for no difference). One trial (N=464) compared liraglutide 1.8 mg once daily with exenatide 10 µg twice daily over 26 weeks. Liraglutide was more efficacious than exenatide in reducing mean HbA1c: -0.%; 95% CI, to -0.18; P< (low strength of evidence). There was no difference in weight loss between liraglutide and exenatide twice daily: 0.8 kg; 95% CI, 0.99 to 0.2, P=0.225 (low strength of evidence). Characteristics of Included Studies Six trials compared 2 drugs within the same class. One trial compared saxagliptin with sitagliptin, 18 one compared liraglutide with exenatide XR, 19 and trials compared exenatide XR (given once weekly) with exenatide given twice daily We rated one trial comparing liraglutide with exenatide XR as poor quality, primarily due to inadequate handling of missing data; study characteristics and results of that trial are provided in evidence tables (published in a separate document). 2 Table 5 presents study characteristics for the 5 trials that we rated as fair quality and included in our analyses. We found no trials assessing intra-class comparisons of amylin agonists or SGLT2 inhibitors.