520597
research-article2014
AOPXXX10.1177/1060028013520597Annals of PharmacotherapyPawaskar et al
Research Report
Observational Study of Kidney Function and Albuminuria in Patients With Type 2 Diabetes Treated With Exenatide BID Versus Insulin Glargine
Annals of Pharmacotherapy 2014, Vol. 48(5) 571–576 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1060028013520597 aop.sagepub.com
Manjiri Pawaskar, PhD1, Katherine R. Tuttle, MD2, Qian Li, PhD3, Jennie H. Best, PhD4, and Pamela W. Anderson, MD5
Abstract Background: Kidney disease is common among patients with type 2 diabetes (T2DM). Treatment of diabetes seeks to minimize negative kidney impact. Objective: To assess the changes in kidney function and incidence of albuminuria in patients with T2DM treated for 1 year with exenatide twice daily (BID) or insulin glargine. Methods: Retrospective analyses were performed using an electronic medical record database. Patients inititating treatment with either exenatide BID or insulin glargine between November 2006, and April 2009 comprised the study cohort. The 2 groups were 1:1 propensity score matched on baseline variables yielding 2683 pairs. Measures of kidney function included estimated glomerular filtration rate (eGFR,) and urinary albumin/creatinine ratio (UACR). Results: Mean baseline eGFR was identical between the groups (79 ± 23 mL/min/1.73 m2). At 1-year follow-up, there was no significant difference in mean eGFR between the groups (78 mL/min/1.73 m2 for exenatide BID vs 80 mL/min/1.73 m2 for insulin glargine; P = .39). Despite matching of multiple characteristics, mean baseline UACR was lower in the exenatide BID group (mean = 34 ± 71 mg/g) compared with the insulin glargine group (183 ± 509 mg/g; P = .03). At follow-up, exenatide BID patients had a mean increase in UACR of 104 mg/g, insulin glargine patients had a decrease of 47 mg/g, but the difference was not significant (P = .19). Conclusion: There were no significant differences in change in kidney function or albuminuria at 1 year in patients treated with exenatide BID compared with insulin glargine as administered in routine practice. Keywords albuminuria, diabetes, eGFR, exenatide BID, insulin glargine
Introduction Diabetes mellitus is associated with a number of significant complications, including kidney disease. Diabetes was the leading cause of new kidney failure in the United States, accounting for 44% of all new cases in 2008.1 Given the prevalence of kidney disease and its important association with diabetes, therapeutic treatment of diabetes should seek to minimize negative impact on the kidneys. Both exenatide twice daily (BID) and insulin glargine are subcutaneously injected and are commonly used to treat patients with type 2 diabetes. Exenatide BID is a glucagonlike peptide-1 receptor agonist that has been shown to reduce glycated hemoglobin (A1C) and weight in adults with type 2 diabetes. Exenatide BID is cleared primarily via the kidneys and has not been shown to be directly nephrotoxic in preclinical or clinical studies.2-4 The US prescribing information for exenatide and insulin each contain information on adjusting dose or advising against use of the respective products in patients with reduced kidney function.2,5
Previous studies have reported data regarding kidney function in patients treated with exenatide BID. One study of a single dose of exenatide administered to individuals with varying levels of kidney function (normal to end-stage renal disease [ESRD]) found exenatide BID to be well-tolerated, except by the ESRD participants who experienced significantly decreased clearance of exenatide, and nausea, and vomiting.3 Increased vomiting among patients with ESRD was assumed to be a result of drug accumulation, not 1
Shire Pharmaceuticals, Wayne, PA, USA Providence Sacred Heart Medical Center and University of Washington School of Medicine, Spokane, WA, USA 3 Evidera, Lexington, MA, USA 4 Bristol-Myers Squibb/AstraZeneca, San Diego, CA, USA 5 Eli Lilly and Company, Indianapolis, IN, USA 2
Corresponding Author: Pamela W. Anderson, MD, Eli Lilly and Company, Lilly Corporate Center, Indianapolis, IN 46285, USA. Email: [email protected]
Downloaded from aop.sagepub.com at East Tennessee State University on June 16, 2015
572
Annals of Pharmacotherapy 48(5)
toxicity, suggesting that a decrease in dosage would be necessary to increase tolerance. In addition, because exenatide BID may induce nausea and vomiting with transient hypovolemia, treatment may worsen renal function.2 A retrospective study utilizing data culled from a medical and pharmacy claims database found no significant differences in claims reporting acute renal failure between persons with type 2 diabetes treated with exenatide BID and those on other therapy regimens.6 A small number of postmarketing case studies have reported a temporal relationship between the initiation of exenatide therapy and the development or worsening of renal impairment.6,7 However, there have been postmarketing reports of altered renal function, including increased serum creatinine, renal impairment, and worsened chronic renal failure and acute renal failure, sometimes requiring hemodialysis or kidney transplantation, which are described in the warning section of the exenatide label. Some of these events occurred in patients receiving one or more pharmacological agents known to affect renal function or hydration status, such as angiotensin-converting enzyme inhibitors, nonsteroidal anti-inflammatory drugs, or diuretics.2 Data regarding the safety of insulin glargine in patients with compromised kidney function are sparse except for well-documented increased risk of hypoglycemia.5 The current study sought to examine kidney function and albuminuria at 1 year in patients with type 2 diabetes treated with exenatide BID or insulin glargine in clinical practice settings in the United States, because the potential renal effect of using exenatide BID was unknown. Insulin glargine was used as a comparator.
Materials and Methods Retrospective analyses were performed using data from the General Electric (GE) General Practice electronic medical record database developed and sold by GE Clinical Data Systems. The GE database contains patients from more than 10 000 general practitioners and detailed health care encounters for more than 20 million patients representing 47 states in the United States. Because data are collected nationwide from 10 000 clinicians and 20 million patients, it is difficult to make any assessment about standardization of diabetes care or routine tests performed for these patients nationwide. Physician visits provide information on lab data. Physician prescription orders are captured; however, whether the prescription was filled or refilled or drug samples were provided is not documented. The amount of inpatient data is quite limited and would only be available if documented by the physician in the outpatient medical record. However, the GE database has been extensively used for research purposes.7-13 In accordance with the Health Insurance Portability and Accountability Act of 1996 regulations regarding repository databases, no identifying patient information was provided,
and there was no way to identify patients from identification as utilizing a general physician or a specialist. Hence institutional review board approval was not required for using these data for research purposes. Data that were received were kept confidential per the sponsor’s policy. Patients with type 2 diabetes who initiated exenatide BID (n = 4494) or insulin glargine (n = 5424) therapy between November 1, 2006, and April 30, 2009, were included in the study. To focus on patients newly initiated with the index medication, patients were excluded if they had a prescription order for glucagon-like peptide-1 receptor agonist or insulin in any form in the preindex period. Patients were also excluded if they had been prescribed pramlintide in the preindex period because these are not typical exenatide patients. Patients were also excluded if they had a diagnosis of type 1 diabetes, gestational diabetes, or were pregnant. Patients needed to be continuously eligible in the 12-month preindex and 12-month postindex periods. Continuous eligibility was defined as being active in the GE database and having activity recorded 365 days before and 365 days after the index date. Hence there was no issue of drop-out rates. Patients who died during the study period were not included in the analyses. The diagnosis of type 2 diabetes mellitus required that at least 1 of the following criteria be satisfied: (1) diagnosis of type 2 diabetes mellitus based on an International Classification of Diseases, Ninth Revision (ICD-9) code of 250.X0 or 250.X2; (2) one or more prescription orders for an oral glucose-lowering drug; (3) 2 consecutive fasting blood glucose levels of ≥126 mg/dL; or (4) an A1C ≥7.0%. Detailed inclusion and exclusion criteria have been published previously.6 The patient observation period was 12 months. Estimated glomerular filtration rate (eGFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation: eGFR = 141 × min(Scr/κ,1)α × max(Scr/κ,1)-1.209 × 0.993Age × 1.018 (if female) × 1.159 (if black), where Scr is serum creatinine (mg/dL), κ is 0.7 for women and 0.9 for men, α is −0.329 for women and −0.411 for men, min indicates the minimum of Scr/κ or 1, and max indicates the maximum of Scr/κ or 1. The National Kidney Foundation Kidney Disease Outcomes Quality Initiative stages of CKD were used to group patients by stage as determined by eGFR.14 Urinary albumin/creatinine ratio (UACR) was measured and calculated by patients’ local clinical laboratories. Patients were categorized by the following UACR values ≤30, 30 300 mg/g. The 2 study cohorts (exenatide BID and insulin glargine) were 1:1 propensity score matched on baseline demographics, clinical characteristics, and resource utilization variables. The study assessed the balance between cohorts after matching by paired tests and standardized differences, yielding 2683 matched pairs. Because patients were not randomly assigned to different treatment groups, we used the propensity score matching method to balance the
Downloaded from aop.sagepub.com at East Tennessee State University on June 16, 2015
573
Pawaskar et al Table 1. Baseline Demographic Characteristics of Patients Before and After Matching. Pre–Propensity Score Matching Age in years, mean (SD) Gender, n (%) Male Female Race/Ethnicity, n (%) Unknown White Black Hispanic Other Specialty of current primary care provider, n (%) Unknown General practitioner Endocrinologist Other
Exenatide BID, n = 4494
Insulin Glargine, n = 5424
Post–Propensity Score Matching P Valuea
Exenatide BID, n = 2683
Insulin Glargine, n = 2683
56
(11)
61
(13)
research-article2014
AOPXXX10.1177/1060028013520597Annals of PharmacotherapyPawaskar et al
Research Report
Observational Study of Kidney Function and Albuminuria in Patients With Type 2 Diabetes Treated With Exenatide BID Versus Insulin Glargine
Annals of Pharmacotherapy 2014, Vol. 48(5) 571–576 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1060028013520597 aop.sagepub.com
Manjiri Pawaskar, PhD1, Katherine R. Tuttle, MD2, Qian Li, PhD3, Jennie H. Best, PhD4, and Pamela W. Anderson, MD5
Abstract Background: Kidney disease is common among patients with type 2 diabetes (T2DM). Treatment of diabetes seeks to minimize negative kidney impact. Objective: To assess the changes in kidney function and incidence of albuminuria in patients with T2DM treated for 1 year with exenatide twice daily (BID) or insulin glargine. Methods: Retrospective analyses were performed using an electronic medical record database. Patients inititating treatment with either exenatide BID or insulin glargine between November 2006, and April 2009 comprised the study cohort. The 2 groups were 1:1 propensity score matched on baseline variables yielding 2683 pairs. Measures of kidney function included estimated glomerular filtration rate (eGFR,) and urinary albumin/creatinine ratio (UACR). Results: Mean baseline eGFR was identical between the groups (79 ± 23 mL/min/1.73 m2). At 1-year follow-up, there was no significant difference in mean eGFR between the groups (78 mL/min/1.73 m2 for exenatide BID vs 80 mL/min/1.73 m2 for insulin glargine; P = .39). Despite matching of multiple characteristics, mean baseline UACR was lower in the exenatide BID group (mean = 34 ± 71 mg/g) compared with the insulin glargine group (183 ± 509 mg/g; P = .03). At follow-up, exenatide BID patients had a mean increase in UACR of 104 mg/g, insulin glargine patients had a decrease of 47 mg/g, but the difference was not significant (P = .19). Conclusion: There were no significant differences in change in kidney function or albuminuria at 1 year in patients treated with exenatide BID compared with insulin glargine as administered in routine practice. Keywords albuminuria, diabetes, eGFR, exenatide BID, insulin glargine
Introduction Diabetes mellitus is associated with a number of significant complications, including kidney disease. Diabetes was the leading cause of new kidney failure in the United States, accounting for 44% of all new cases in 2008.1 Given the prevalence of kidney disease and its important association with diabetes, therapeutic treatment of diabetes should seek to minimize negative impact on the kidneys. Both exenatide twice daily (BID) and insulin glargine are subcutaneously injected and are commonly used to treat patients with type 2 diabetes. Exenatide BID is a glucagonlike peptide-1 receptor agonist that has been shown to reduce glycated hemoglobin (A1C) and weight in adults with type 2 diabetes. Exenatide BID is cleared primarily via the kidneys and has not been shown to be directly nephrotoxic in preclinical or clinical studies.2-4 The US prescribing information for exenatide and insulin each contain information on adjusting dose or advising against use of the respective products in patients with reduced kidney function.2,5
Previous studies have reported data regarding kidney function in patients treated with exenatide BID. One study of a single dose of exenatide administered to individuals with varying levels of kidney function (normal to end-stage renal disease [ESRD]) found exenatide BID to be well-tolerated, except by the ESRD participants who experienced significantly decreased clearance of exenatide, and nausea, and vomiting.3 Increased vomiting among patients with ESRD was assumed to be a result of drug accumulation, not 1
Shire Pharmaceuticals, Wayne, PA, USA Providence Sacred Heart Medical Center and University of Washington School of Medicine, Spokane, WA, USA 3 Evidera, Lexington, MA, USA 4 Bristol-Myers Squibb/AstraZeneca, San Diego, CA, USA 5 Eli Lilly and Company, Indianapolis, IN, USA 2
Corresponding Author: Pamela W. Anderson, MD, Eli Lilly and Company, Lilly Corporate Center, Indianapolis, IN 46285, USA. Email: [email protected]
Downloaded from aop.sagepub.com at East Tennessee State University on June 16, 2015
572
Annals of Pharmacotherapy 48(5)
toxicity, suggesting that a decrease in dosage would be necessary to increase tolerance. In addition, because exenatide BID may induce nausea and vomiting with transient hypovolemia, treatment may worsen renal function.2 A retrospective study utilizing data culled from a medical and pharmacy claims database found no significant differences in claims reporting acute renal failure between persons with type 2 diabetes treated with exenatide BID and those on other therapy regimens.6 A small number of postmarketing case studies have reported a temporal relationship between the initiation of exenatide therapy and the development or worsening of renal impairment.6,7 However, there have been postmarketing reports of altered renal function, including increased serum creatinine, renal impairment, and worsened chronic renal failure and acute renal failure, sometimes requiring hemodialysis or kidney transplantation, which are described in the warning section of the exenatide label. Some of these events occurred in patients receiving one or more pharmacological agents known to affect renal function or hydration status, such as angiotensin-converting enzyme inhibitors, nonsteroidal anti-inflammatory drugs, or diuretics.2 Data regarding the safety of insulin glargine in patients with compromised kidney function are sparse except for well-documented increased risk of hypoglycemia.5 The current study sought to examine kidney function and albuminuria at 1 year in patients with type 2 diabetes treated with exenatide BID or insulin glargine in clinical practice settings in the United States, because the potential renal effect of using exenatide BID was unknown. Insulin glargine was used as a comparator.
Materials and Methods Retrospective analyses were performed using data from the General Electric (GE) General Practice electronic medical record database developed and sold by GE Clinical Data Systems. The GE database contains patients from more than 10 000 general practitioners and detailed health care encounters for more than 20 million patients representing 47 states in the United States. Because data are collected nationwide from 10 000 clinicians and 20 million patients, it is difficult to make any assessment about standardization of diabetes care or routine tests performed for these patients nationwide. Physician visits provide information on lab data. Physician prescription orders are captured; however, whether the prescription was filled or refilled or drug samples were provided is not documented. The amount of inpatient data is quite limited and would only be available if documented by the physician in the outpatient medical record. However, the GE database has been extensively used for research purposes.7-13 In accordance with the Health Insurance Portability and Accountability Act of 1996 regulations regarding repository databases, no identifying patient information was provided,
and there was no way to identify patients from identification as utilizing a general physician or a specialist. Hence institutional review board approval was not required for using these data for research purposes. Data that were received were kept confidential per the sponsor’s policy. Patients with type 2 diabetes who initiated exenatide BID (n = 4494) or insulin glargine (n = 5424) therapy between November 1, 2006, and April 30, 2009, were included in the study. To focus on patients newly initiated with the index medication, patients were excluded if they had a prescription order for glucagon-like peptide-1 receptor agonist or insulin in any form in the preindex period. Patients were also excluded if they had been prescribed pramlintide in the preindex period because these are not typical exenatide patients. Patients were also excluded if they had a diagnosis of type 1 diabetes, gestational diabetes, or were pregnant. Patients needed to be continuously eligible in the 12-month preindex and 12-month postindex periods. Continuous eligibility was defined as being active in the GE database and having activity recorded 365 days before and 365 days after the index date. Hence there was no issue of drop-out rates. Patients who died during the study period were not included in the analyses. The diagnosis of type 2 diabetes mellitus required that at least 1 of the following criteria be satisfied: (1) diagnosis of type 2 diabetes mellitus based on an International Classification of Diseases, Ninth Revision (ICD-9) code of 250.X0 or 250.X2; (2) one or more prescription orders for an oral glucose-lowering drug; (3) 2 consecutive fasting blood glucose levels of ≥126 mg/dL; or (4) an A1C ≥7.0%. Detailed inclusion and exclusion criteria have been published previously.6 The patient observation period was 12 months. Estimated glomerular filtration rate (eGFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation: eGFR = 141 × min(Scr/κ,1)α × max(Scr/κ,1)-1.209 × 0.993Age × 1.018 (if female) × 1.159 (if black), where Scr is serum creatinine (mg/dL), κ is 0.7 for women and 0.9 for men, α is −0.329 for women and −0.411 for men, min indicates the minimum of Scr/κ or 1, and max indicates the maximum of Scr/κ or 1. The National Kidney Foundation Kidney Disease Outcomes Quality Initiative stages of CKD were used to group patients by stage as determined by eGFR.14 Urinary albumin/creatinine ratio (UACR) was measured and calculated by patients’ local clinical laboratories. Patients were categorized by the following UACR values ≤30, 30 300 mg/g. The 2 study cohorts (exenatide BID and insulin glargine) were 1:1 propensity score matched on baseline demographics, clinical characteristics, and resource utilization variables. The study assessed the balance between cohorts after matching by paired tests and standardized differences, yielding 2683 matched pairs. Because patients were not randomly assigned to different treatment groups, we used the propensity score matching method to balance the
Downloaded from aop.sagepub.com at East Tennessee State University on June 16, 2015
573
Pawaskar et al Table 1. Baseline Demographic Characteristics of Patients Before and After Matching. Pre–Propensity Score Matching Age in years, mean (SD) Gender, n (%) Male Female Race/Ethnicity, n (%) Unknown White Black Hispanic Other Specialty of current primary care provider, n (%) Unknown General practitioner Endocrinologist Other
Exenatide BID, n = 4494
Insulin Glargine, n = 5424
Post–Propensity Score Matching P Valuea
Exenatide BID, n = 2683
Insulin Glargine, n = 2683
56
(11)
61
(13)
