- Email: [email protected]
Urinary tract infections in patients with diabetes treated with dapagliflozin
Journal of Diabetes and Its Complications 27 (2013) 473–478
Contents lists available at ScienceDirect
Journal of Diabetes and Its Complications journal homepage: WWW.JDCJOURNAL.COM
Urinary tract infections in patients with diabetes treated with dapagliflozin☆,☆☆ Kristina M. Johnsson a,⁎, Agata Ptaszynska b, Bridget Schmitz b, Jennifer Sugg c, Shamik J. Parikh c, James F. List b a b c
AstraZeneca, Mölndal, Sweden Bristol-Myers Squibb, Princeton, NJ, USA AstraZeneca, Wilmington, DE, USA
a r t i c l e
i n f o
Article history: Received 3 January 2013 Received in revised form 29 April 2013 Accepted 7 May 2013 Available online 10 July 2013 Keywords: Dapagliflozin Glucosuria Sodium glucose cotransporter 2 SGLT2 Urinary tract infection UTI
a b s t r a c t Aims: Urinary tract infection is common in patients with type 2 diabetes. Possible causative factors include glucosuria, which is a result of treatment with sodium glucose cotransporter 2 (SGLT2) inhibitors. Dapagliflozin is an investigative SGLT2 inhibitor with demonstrated glycemic benefits in patients with diabetes. Data from dapagliflozin multi-trial safety data were analyzed to clarify the association between glucosuria and urinary tract infection. Methods: Safety data from 12 randomized, placebo-controlled trials were pooled to evaluate the relationship between glucosuria and urinary tract infection in patients with inadequately controlled diabetes (HbA1c N 6.5%–12%). Patients were treated with dapagliflozin (2.5, 5, or 10 mg) or placebo once daily, either as monotherapy or add-on to metformin, insulin, sulfonylurea, or thiazolidinedione for 12–24 weeks. The incidence of clinical diagnoses and events suggestive of urinary tract infection were quantified. Results: This analysis included 3152 patients who received once-daily dapagliflozin (2.5 mg [n = 814], 5 mg [n = 1145], or 10 mg [n = 1193]) as monotherapy or add-on treatment, and 1393 placebo-treated patients. For dapagliflozin 2.5 mg, 5 mg, 10 mg, and placebo, diagnosed infections were reported in 3.6%, 5.7%, 4.3%, and 3.7%, respectively. Urinary glucose levels, but not the incidence of urinary tract infection, increased progressively with dapagliflozin dosage. Most identified infections were those considered typical for patients with diabetes. Discontinuations due to urinary tract infection were rare: 8 (0.3%) dapagliflozin-treated patients and 1 (0.1%) placebo-treated patient. Most diagnosed infections were mild to moderate and responded to standard antimicrobial treatment. Conclusions: Treatment of type 2 diabetes with once-daily dapagliflozin 5 or 10 mg is accompanied by a slightly increased risk of urinary tract infection. Infections were generally mild to moderate and clinically manageable. This analysis did not demonstrate a definitive dose relationship between glucosuria and urinary tract infection. © 2013 Elsevier Inc. All rights reserved.
1. Introduction Urinary tract infection (UTI) is a common occurrence in patients with type 2 diabetes (Boyko, Fihn, Scholes, Abraham, & Monsey, 2005; Donders, 2002; Geerlings, 2008; Goswami et al., 2000; Joshi, Caputo, Weitekamp, & Karchmer, 1999), although a precise cause–effect relationship has not been determined. Multiple factors may be involved, ☆ Clinical trial registration numbers: NCT00263276, NCT00972244, NCT00528372, NCT00736879, NCT00528879, NCT00855166, NCT00357370, NCT00680745, NCT00683878, NCT00673231, NCT00643851, NCT00859898. Available at http://clinicaltrials.gov. ☆☆ Disclosure Statement: The studies included in this analysis were sponsored by Bristol-Myers Squibb and AstraZeneca. K.M. Johnsson, J. Sugg, and S.J. Parikh are employees and stockholders of AstraZeneca. A. Ptaszynska, B. Schmitz, and J.F. List are employees and stockholders of Bristol-Myers Squibb. ⁎ Corresponding author. AstraZeneca, R&D Mölndal, SE-43183 Mölndal, Sweden. Tel.: +46 31 7065562. E-mail address: [email protected] (K.M. Johnsson). 1056-8727/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jdiacomp.2013.05.004
such as hyperglycemia, glucosuria, and neurogenic bladder (Geerlings, 2008; Hoepelman, Meiland, & Geerlings, 2003; Rackley, 2011; Rayfield et al., 1982; Sauerwein, 2002). Hyperglycemia can impede the body's resistance to microorganisms and lead to increased adherence of bacteria to uroepithelial cells (Geerlings, 2008; Rayfield et al., 1982). There are inadequate data quantifying the relationship between the level of glucosuria, bacterial growth in the urine, and incidence of UTI. In an analysis of the growth of Escherichia coli in human urine, the growth rate of E. coli was enhanced when glucose was added to urine, although very high glucose concentrations resulted in a decreased rate when the urinary pH was not kept constant (Geerlings, Brouwer, Gaastra, Verhoef, & Hoepelman, 1999). In a statistical analysis of a retrospective study of women with diabetes, glucosuria was not identified as a risk factor for UTI (Geerlings et al., 2000a, 2000b). Glucosuria is of particular interest due to its relationship to an investigational class of antidiabetes therapy that targets sodium glucose cotransporter 2 (SGLT2), the protein that mediates reabsorption of
474
K.M. Johnsson et al. / Journal of Diabetes and Its Complications 27 (2013) 473–478
most of the glucose filtered by the kidney. SGLT2 inhibitors act independently of insulin secretion or action to reduce glucose reabsorption, thus resulting in increased urinary glucose excretion with a corresponding reduction in blood glucose levels (Ghosh, Ghosh, Chawla, & Jasdanwala, 2011; Rahmoune et al., 2005; Salvatore et al., 2011). Dapagliflozin is a novel, orally administered, selective, potent SGLT2 inhibitor in development for the treatment of type 2 diabetes (Han et al., 2008; Meng et al., 2008). Dapagliflozin has been studied in patients in placebo-controlled clinical studies as monotherapy, as add-on therapy to other standard antidiabetic treatments, and as firstline combination therapy with metformin (Bailey, Gross, Pieters, Bastien, & List, 2010; Ferrannini, Ramos, Salsali, Tang, & List, 2010; List, Woo, Morales, Tang, & Fiedorek, 2009; Rosenstock, Vico, Wei, Salsali, & List, 2012; Strojek et al., 2011; Wilding et al., 2009; Wilding et al., 2012). In each study, patients who received dapagliflozin achieved significantly greater glycemic control than those who received placebo, and dapagliflozin was generally well tolerated. Data from clinical pharmacology studies demonstrated dose-related increases in glucosuria for doses of 2.5, 5, and 10 mg (Parikh et al., 2011); increases were smaller at doses N 10 mg (List, Woo, Morales, Tang, & Fiedorek, 2009). Because its mechanism of action leads to increased urinary glucose excretion, controlled clinical trials of dapagliflozin offer an opportunity to evaluate the relationship between glucosuria and UTI. 2. Subjects, materials and methods 2.1. Clinical trials and patients Data from 12 randomized, placebo-controlled trials (Appendix 1) were pooled to evaluate the relationship between pharmacologically induced urinary glucose excretion and UTI in patients with type 2 diabetes treated with dapagliflozin. Details regarding the methodologies of these trials are provided in the accompanying article “Vulvovaginitis and balanitis in patients with diabetes treated with dapagliflozin” (Johnsson et al., 2013). Patients with a history or risk factors for UTI were not excluded from participation in the clinical trials. All studies were conducted in accordance with Good Clinical Practice guidelines and the Declaration of Helsinki and were approved by the institutional review boards or independent ethics committee of each participating center. All patients provided written informed consent to participate in their respective clinical trials. 2.2. Safety signal detection and quantification Due to the potential for an increased risk in the incidence of UTI associated with treatment-induced glucosuria, a comprehensive approach was undertaken to broaden the potential for detection of any signal indicating an increased risk. Throughout the dapagliflozin clinical trials program, a rigorous effort was made to capture all signs, symptoms and events suggestive of UTI. In addition to spontaneous reports of symptoms by study participants, investigators proactively questioned patients at each visit about symptoms suggestive of UTI (either experienced currently or since the previous study visit). This proactive questioning was designed to address the possibility that patients might not recognize certain symptoms as being relevant. Supplemental case-report forms were provided to obtain more detailed information for assessment. For the initial analysis, a broad net was generated using a set of 63 prespecified, preferred terms (Appendix 2) from the Medical Dictionary for Regulatory Activities (MedDRA) version 13.0 to capture signs and symptoms (eg, dysuria) suggestive of UTI, as well as specific clinical diagnoses. These preferred terms were referred to as “events suggestive of UTI.” When a prespecified, preferred term was reported,
a case-report questionnaire was completed by the investigator to fully describe the case and the risk factors. In a second, more specific analysis, 49 prespecified, preferred terms for clinical diagnosis were used to quantify diagnosed UTIs (Appendix 2). To increase the specificity of diagnosis, investigators were also asked to obtain a urine culture to confirm diagnosis in patients with suspected UTI. Study protocol required the drug to be withheld in patients with clinical evidence of pyelonephritis or presumed urosepsis until treatment of the infection was complete and clinical recovery had occurred. 3. Results 3.1. Patients The population for this analysis included 4545 patients from the 12 clinical trials. Treatment groups included 3152 patients who received once-daily dapagliflozin (2.5 mg [n = 814], 5 mg [n = 1145], or 10 mg [n = 1193]) and 1393 patients who received placebo. These study groups were generally balanced with respect to baseline demographics and disease characteristics (Table 1). Baseline mean HbA1c was 8.1% to 8.4%. More than 85% of patients had BMI ≥ 25 kg/m 2, and N55% had BMI ≥30 kg/m 2. A varied range of disease duration (mean 5.3–6.7 years) and progression were represented in the patient population. In dapagliflozin-treated patients, mean exposure ranged from 148.2 to 150.5 days. Mean duration of exposure for those treated with placebo was 149.4 days. A dose-dependent increase in glucosuria was documented at 24 weeks in this pooled analysis; mean change from baseline in spot fasting urine glucose excretion for placebo, 2.5 mg, 5 mg, and 10 mg was − 241.0 mg/dL (SE 36.2), + 1480.5 mg/dL (SE 68.6), + 2149.9 (SE 66.7), and + 2592.3 (SE 65.6), respectively. Where long-term data for glucose excretion were available, levels were consistent with the short-term analysis. 3.2. Events suggestive of UTI Of the 63 prespecified, preferred MedDRA terms used to capture signs, symptoms, and events suggestive of UTI, 16 were reported in ≥ 1 patient (Appendix 2 Figure). Greater proportions of patients in the dapagliflozin 5- and 10-mg groups experienced events suggestive of UTI (7.3% and 6.5%, respectively) than did patients in the dapagliflozin 2.5-mg group and the placebo group (4.2% and 4.5%, respectively) (Appendix 3 Figure). Most episodes were mild or moderate in intensity across all treatment groups. These events were more common in women than in men in all treatment groups. Signs, symptoms, and events suggestive of UTI that were reported in ≥1% of women in one or more of the dapagliflozin groups were the diagnosis of UTI, the symptom of dysuria, and the diagnosis of cystitis (Table 2). UTI and cystitis also occurred in ≥ 1% of women in the placebo group. Dysuria and UTI were reported in ≥1% of men in one or more of the dapagliflozin groups but not in the placebo group. Urine cultures were obtained in 39% to 50% of dapagliflozin patients and in 50% of placebo patients with events suggestive of UTI. Approximately two thirds of the cultures across all treatment groups were positive; most of the organisms identified were those commonly seen in patients with type 2 diabetes (eg, E. coli, Klebsiella pneumoniae, and Proteus). To begin to understand the potential risk factors for UTI, incidence rates were determined based on various subgroups, including categories of baseline HbA1c (b8%, ≥ 8% to b9%, and ≥ 9%), age (b 65 and ≥ 65 years), gender, and history of recurrent infection (Fig. 1). For all subgroups, there was variation in the background rates of events suggestive of UTI; however, within each subgroup, there was a proportionate increase in incidence for dapagliflozin vs placebo. There
K.M. Johnsson et al. / Journal of Diabetes and Its Complications 27 (2013) 473–478
475
Table 1 Demographic and baseline characteristics of patients. Characteristic
Placebo (n = 1393)
Dapagliflozin 2.5 mg/d (n = 814)
Age (y), mean (SD) Sex, n (%) Women Men Race, n (%) White Black/African American Asian Other Weight (kg), mean (SD) Body mass index, n (%) ≥25 kg/m2 ≥30 kg/m2 Diabetes duration (y), mean (SD) HbA1c (%), mean (SD) a
5 mg/d (n = 1145)
10 mg/d (n = 1193)
55.3 (10.45)
57.0 (10.06)
55.4 (10.35)
55.1 (10.14)
677 (48.6) 716 (51.4)
400 (49.1) 414 (50.9)
581 (50.7) 564 (49.3)
598 (50.1) 595 (49.9)
653 (80.2) 15 (1.8) 127 (15.6) 19 (2.3) 86.1 (19.55)
907 33 181 24 86.0
(79.2) (2.9) (15.8) (2.1) (19.03)
976 35 152 30 88.0
(81.8) (2.9) (12.7) (2.5) (19.43)
715 (87.8) 472 (58.0) 6.7 (6.94) 8.1 (0.87)
1003 660 5.7 8.4
(87.6) (57.6) (6.64) (1.09)
1082 700 6.2 8.3
(90.7) (58.7) (6.65) (1.05)
1129 38 201 25 87.4
(81.0) (2.7) (14.4) (1.8) (19.46)
1241 (89.1) 794 (57.0) 5.3a (6.37) 8.4 (1.16)
n = 1392.
were no clear trends regarding dosing within any subgroup. More events suggestive of UTI were documented in patients ≥65 years old vs younger patients treated with dapagliflozin; more events were documented in females vs males across all treatment groups. The rate of events suggestive of UTI also were analyzed in patients who completed the short-term treatment period and continued in a long-term treatment period of their respective trials for ≤102 weeks of total treatment (dapagliflozin n = 2160, placebo n = 694). In the long-term analyses, the proportions of patients with events suggestive of UTI were slightly higher vs the short-term treatment period. In the short-term treatment period, proportions were 4.2% to 7.3% in the dapagliflozin groups and 4.5% in the placebo group; in the combined treatment periods, proportions were 9.4% in the dapagliflozin groups and 6.6% in the placebo group. An analysis of time to onset of first event suggestive of UTI indicated that patients treated with dapagliflozin 5 or 10 mg were at greater risk for a first event than those who received dapagliflozin 2.5 mg or placebo, starting at approximately 8 weeks after treatment initiation (Appendix 4 Figure). Across all treatment groups in the combined treatment periods, first events occurred relatively early during the trials; a first event was more likely to occur in the first 24 weeks than after 24 weeks. 3.3. Clinical diagnosis of UTI Of the 49 prespecified, preferred MedDRA terms related to specific clinical diagnoses of UTI, 9 were reported in ≥1 patient (Appendix 2 Figure). A total of 230 patients were diagnosed with UTI: 56 for placebo, 35 for dapagliflozin 2.5 mg, 76 for dapagliflozin 5 mg, and 63
for dapagliflozin 10 mg. The rates of clinically diagnosed cases were higher in the dapagliflozin 5- and 10-mg groups (5.7% and 4.3%, respectively) vs the dapagliflozin 2.5-mg and placebo groups (3.6% and 3.7%, respectively) (Appendix 3 Figure). The incidence of diagnosed infection was analyzed for statistical significance, both for the total pooled population and for subgroups stratified by baseline HbA1c (Table 3). The frequency of UTI was significantly greater vs placebo for patients treated with dapagliflozin 5 mg in the total pooled population; no other comparisons were statistically significant. Overall, clinical diagnosis of UTI was more common in dapagliflozin-treated patients than in the placebo group. The rates were higher for women than men across treatment groups. Treatment for clinically diagnosed UTI was administered by clinical investigator or by patient's physician. Treatment was not specified in the trial protocols and was initiated according to physician choice and local treatment guidelines. Of the 174 dapagliflozin-treated patients documented as having a clinical diagnosis of UTI, treatment was documented in 78%. Most of these cases were managed with one course of standard antimicrobial treatment (Fig. 2). However, 5.7%, 1.3%, and 15.9% of episodes in the dapagliflozin 2.5-, 5-, and 10-mg groups, respectively, and 14.3% of episodes in the placebo group required more than 1 course of treatment. Pyelonephritis was a rare occurrence with similar frequencies in each treatment group: 2 (0.2%) in the dapagliflozin 2.5-mg group, 1 (0.1%) in the dapagliflozin 5-mg group, none in the dapagliflozin 10-mg group, and 1 (0.1%) in the placebo-treated group. The only case that was considered by the investigator to be a serious adverse event was in the placebo group.
Table 2 UTI infections occurring in ≥1% of patients in any treatment group, by gender. No. (%) of patients
Placebo (n = 1393)
Dapagliflozin 2.5 mg/d (n = 814)
5 mg/d (n = 1145)
10 mg/d (n = 1193)
Women Signs, symptoms, and events suggestive of UTI UTI Dysuria Cystitis Clinical diagnoses of UTI UTI Cystitis Men Signs, symptoms, and events suggestive of UTI Dysuria UTI Clinical diagnoses of UTI UTI
677 (49)
400 (49)
581 (51)
598 (50)
35 (5.2) 6 (0.9) 10 (1.5)
21 (5.3) 3 (0.8) 2 (0.5)
48 (8.3) 10 (1.7) 7 (1.2)
39 (6.5) 12 (2.0) 7 (1.2)
35 (5.2) 10 (1.5) 716 (51)
21 (5.3) 2 (0.5) 414 (51)
48 (8.3) 7 (1.2) 564 (49)
39 (6.5) 7 (1.2) 595 (50)
4 (0.6) 3 (0.4)
2 (0.5) 4 (1.0)
8 (1.4) 6 (1.1)
13 (2.2) 4 (0.7)
3 (0.4)
4 (1.0)
6 (1.1)
4 (0.7)
476
K.M. Johnsson et al. / Journal of Diabetes and Its Complications 27 (2013) 473–478
Fig. 1. Percentage of patients with ≥1 event suggestive of UTI by baseline subgroups: (A) HbA1c category, (B) age category, (C) gender, and (D) history of recurrent UTI. Data represent placebo-controlled pool from short-term, double-blind treatment periods and include data after rescue.
Two patients in the dapagliflozin 2.5-mg group experienced urinary tract events identified by the investigator as serious adverse events: 1 with UTI and 1 with malacoplakia vesicae. One episode of urosepsis (not a prespecified, preferred MedDRA term) was reported in the dapagliflozin 10-mg group as a serious adverse event. Dapagliflozin was not discontinued, and the patient completed the study. Few patients discontinued treatment as a result of specific clinical diagnoses of UTI: 8 (0.3%) patients in the dapagliflozin groups and 1 (0.1%) patient in the placebo group. Among dapagliflozintreated patients, 2 patients in the 2.5-mg group discontinued because of pyelonephritis, 2 in the 5-mg group and 3 in the 10-mg group discontinued because of UTI, and 1 in the 2.5-mg group discontinued because of cystitis. In the placebo group, 1 patient discontinued because of pyelonephritis. The proportions of patients with a history of recurring UTIs were relatively small across treatment groups (2.0%–2.8%). During the
study period, greater proportions of these patients with history were diagnosed with UTI (17.1%–21.1%) than those without a history of recurrent infection (3.2%–5.5%). There was no clear trend for recurrence with dapagliflozin vs placebo among patients with history of recurrent infection. During the trials, recurrence of UTI events was reported in 15.3% to 19.6% of dapagliflozin-treated patients vs 7.7% of placebo-treated patients. In the trials up to 2 years, recurrent UTI was reported in 22.9% of dapagliflozin-treated patients vs 13.6% of placebo-treated patients. The frequency of recurrent infection during the trials did not appear to be dose dependent. 4. Discussion A rigorous effort was made to accurately determine the level of risk of UTI associated with dapagliflozin treatment by means of the dual approach of spontaneous reporting and proactive questioning of
Table 3 Patients with diagnosed UTI. Placebo (n = 1393)
Total pooled population No. patients with events (%) 52 (3.7) P value vs placebo Baseline HbA1c category, no. patients with event (%) b8% 19 (3.3) P value vs placebo ≥8% and b9% 17 (4.0) P value vs placebo ≥9% 16 (4.2) P value vs placebo
Dapagliflozin 2.5 mg/d (n = 814)
5 mg/d (n = 1145)
10 mg/d (n = 1193)
29 (3.6) .8373
65 (5.7) .0201
51 (4.3) .4823
15 (3.9) .5822 11 (3.8) .8721 3 (2.2) .4272
24 (5.5) .0791 26 (6.7) .0826 15 (4.7) .7441
17 (3.2) .9751 20 (5.3) .3843 14 (4.9) .6591
Statistical comparisons were conducted for each dapagliflozin group vs placebo using chi-square tests, unless the number of events is ≤5 in either group, in which case, Fisher exact test was used.
K.M. Johnsson et al. / Journal of Diabetes and Its Complications 27 (2013) 473–478
477
Fig. 2. Characterization of episodes of UTI: (A) documentation of non-specific signs, symptoms vs clinical diagnosis; (B) treatment of clinically diagnosed cases.
patients. It was believed that proactive questioning would increase the reporting of events suggestive of UTI and remind patients to be watchful for the symptoms discussed at clinic visits, thereby encouraging them to report any relevant events. Proactive questioning should also remind patients to disclose any treatments administered outside the study. Clinical diagnosis of UTI was made in the dapagliflozin studies by collection of urine cultures, by opinion of experienced clinicians, or by response to treatment administered after a presumed diagnosis. Because it is common in clinical practice to treat symptomatic UTIs without obtaining a culture, relying solely on culture results in these trials would have led to an erroneously low reported incidence of infection. Conversely, if urinalyses had routinely been conducted on all study participants without regard to the presence or absence of symptoms, cases of asymptomatic bacteriuria or pyuria not warranting treatment would have skewed the reported incidence of UTI in the opposite direction. The protocol-specified
approach to the detection and quantification of a signal for increased UTI was reflective of general clinical practice and was considered to be a realistic and optimal means of quantifying increased risk. There is a presumed and considerably debated association between glucosuria and UTI. In our Phase 3 trials, pooled data indicate a small increase in the incidence of UTI in patients treated with dapagliflozin 5 or 10 mg vs those treated with placebo. However, the incidence was relatively low across all treatment groups (b6%), the rate of infection in the 2.5-mg group was similar to that seen with placebo, and there was no consistent dose-related trend across studies. Generally, first events of UTI occurred relatively early in the studies. Clinical diagnosis of UTI occurred more commonly in women than in men. The pathogens reported were types commonly seen in patients with type 2 diabetes. Most events were of mild to moderate intensity and resolved with one course of standard antimicrobial treatment. Discontinuations of dapagliflozin due to clinical diagnosis
Fig. 3. Comparison of dose-related trends in glucosuria and clinical diagnosis of UTI in the dapagliflozin trials.
478
K.M. Johnsson et al. / Journal of Diabetes and Its Complications 27 (2013) 473–478
of UTI were rare, which indicates that study participants found these events to be tolerable. The proportions of dapagliflozin-treated patients and placebo-treated patients diagnosed with pyelonephritis were identical (0.1% for each); the only case leading to discontinuation of treatment was in the placebo group. It has been established that treatment with dapagliflozin produces glucosuria with a reduction in blood glucose levels (Ferrannini, Ramos, Salsali, Tang, & List, 2010; Komoroski et al., 2009; List, Woo, Morales, Tang, & Fiedorek, 2009). In this pooled safety analysis of 12 clinical trials, glucose excretion was shown to be progressively greater with increased doses of dapagliflozin; although as Fig. 3 illustrates, a similar dose-dependent relationship with the incidence of UTI was not demonstrated. While the data suggest that glucosuria is a risk factor for the development of UTI, it is important to recognize potential confounding factors. As an example, patients with a history of recurrent UTIs, including patients treated with placebo, experienced much higher rates of UTI during the trials, indicating a predisposition to infection unrelated to glucosuria. The variation in patient histories with regard to recurrent UTIs limits the ability to generalize findings to the overall patient population. This general pattern also holds true for the relationship between glucosuria and genital infection in dapagliflozin-treated patients (Johnsson et al., 2013); however, the signal for increased risk is clearer for genital infection. This inconsistency may reflect a greater effect of glucosuria in promoting the growth of fungal pathogens typically associated with genital infection vs bacterial pathogens typically associated with UTI. Further study to better elucidate the relationship between glucosuria and these infections is warranted. 5. Conclusions This pooled safety analysis of 12 clinical trials determined that treatment of type 2 diabetes with once-daily dapagliflozin 5 or 10 mg is accompanied by a small increase in risk of UTI, but no evidence of increase in pyelonephritis. These infections were considered typical for patients with type 2 diabetes, they generally did not lead to discontinuation of treatment, and recurrence was uncommon. Infections were generally mild to moderate, clinically manageable, and not associated with complications. A dose relationship with the 5- and 10-mg doses was not evident. Acknowledgments This study was funded by Bristol-Myers Squibb and AstraZeneca. Ann L. Davis, MPH, CMPP, an employee of Bristol-Myers Squibb, assisted in the development of the manuscript. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jdiacomp.2013.05.004. References Bailey, C. J., Gross, J. L., Pieters, A., Bastien, A., & List, J. F. (2010). Effect of dapagliflozin in patients with type 2 diabetes who have inadequate glycaemic control with metformin: a randomised, double-blind, placebo-controlled trial. The Lancet, 375(9733), 2223–2233. Boyko, E. J., Fihn, S. D., Scholes, D., Abraham, L., & Monsey, B. (2005). Risk of urinary tract infection and asymptomatic bacteriuria among diabetic and nondiabetic postmenopausal women. American Journal of Epidemiology, 161(6), 557–564. Donders, G. G. (2002). Lower genital tract infections in diabetic women. Current Infectious Disease Reports, 4(6), 536–539.
Ferrannini, E., Ramos, S. J., Salsali, A., Tang, W., & List, J. F. (2010). Dapagliflozin monotherapy in type 2 diabetic patients with inadequate glycemic control by diet and exercise: a randomized, double-blind, placebo-controlled, phase 3 trial. Diabetes Care, 33(10), 2217–2224. Geerlings, S. E. (2008). Urinary tract infections in patients with diabetes mellitus: epidemiology, pathogenesis and treatment. International Journal of Antimicrobial Agents, 31(Suppl 1), S54–S57. Geerlings, S. E., Brouwer, E. C., Gaastra, W., Verhoef, J., & Hoepelman, A. I. (1999). Effect of glucose and pH on uropathogenic and non-uropathogenic Escherichia coli: studies with urine from diabetic and non-diabetic individuals. Journal of Medical Microbiology, 48(6), 535–539. Geerlings, S. E., Stolk, R. P., Camps, M. J., Netten, P. M., Collet, T. J., & Hoepelman, A. I.; Diabetes Women Asymptomatic Bacteriuria Utrecht Study Group. (2000a). Risk factors for symptomatic urinary tract infection in women with diabetes. Diabetes Care, 23(12), 1737–1741. Geerlings, S. E., Stolk, R. P., Camps, M. J. et al.; Diabetes Women Asymptomatic Bacteriuria Utrecht Study Group(2000b). Asymptomatic bacteriuria may be considered a complication in women with diabetes. Diabetes Care, 23(6), 744–749. Ghosh, R. K., Ghosh, S. M., Chawla, S., & Jasdanwala, S. A. (2012). SGLT2 inhibitors: a new emerging therapeutic class in the treatment of type 2 diabetes mellitus. The Journal of Clinical Pharmacology, 52(4), 457–463. Goswami, R., Dadhwal, V., Tejaswi, S., Datta, K., et al. (2000). Species-specific prevalence of vaginal candidiasis among patients with diabetes mellitus and its relation to their glycaemic status. The Journal of Infection, 41(2), 162–166. Han, S., Hagan, D. L., Taylor, J. R., Xin, L., et al. (2008). Dapagliflozin, a selective SGLT2 inhibitor, improves glucose homeostasis in normal and diabetic rats. Diabetes, 57(6), 1723–1729. Hoepelman, A. I., Meiland, R., & Geerlings, S. E. (2003). Pathogenesis and management of bacterial urinary tract infections in adult patients with diabetes mellitus. International Journal of Antimicrobial Agents, 22(Suppl), 235–243. Johnsson, K. M., Ptaszynska, A., Schmitz, B., Sugg, J., Parikh, S. J., & List, J. F. (2013). Vulvovaginitis and balanitis in patients with diabetes treated with dapagliflozin. Journal of Diabetes and its Complications, 27(5), 479–484. Joshi, N., Caputo, G. M., Weitekamp, M. R., & Karchmer, A. W. (1999). Infections in patients with diabetes mellitus. The New England Journal of Medicine, 341(25), 1906–1912. Komoroski, B., Vachharajani, N., Feng, Y., Li, L., Kornhauser, D., & Pfister, M. (2009). Dapagliflozin, a novel, selective SGLT2 inhibitor, improved glycemic control over 2 weeks in patients with type 2 diabetes mellitus. Clinical Pharmacology and Therapeutics, 85(5), 513–519. List, J. F., Woo, V., Morales, E., Tang, W., & Fiedorek, F. T. (2009). Sodium-glucose cotransport inhibition with dapagliflozin in type 2 diabetes. Diabetes Care, 32(4), 650–657. Meng, W., Ellsworth, B. A., Nirschl, A. A., McCann, P. J., et al. (2008). Discovery of dapagliflozin: a potent, selective renal sodium-dependent glucose cotransporter 2 (SGLT2) inhibitor for the treatment of type 2 diabetes. Journal of Medicinal Chemistry, 51(5), 1145–1149. Parikh, S. J., Johnsson, K. M., Ptaszynska, A., Schmitz, B., et al. (2011). Characterization of urinary tract infections in the setting of pharmacologically-induced glucosuria. Presented at: 71st Scientific Sessions of the American Diabetes Association, June 24–28, 2011. San Diego, CA. Rackley, R. (2011). Neurogenic bladder. Medscape, Available at: http://emedicine. medscape.com/article/453539-overview. Accessed December 11, 2011. Rahmoune, H., Thompson, P. W., Ward, J. M., Smith, C. D., Hong, G., & Brown, J. (2005). Glucose transporters in human renal proximal tubular cells isolated from the urine of patients with non-insulin-dependent diabetes. Diabetes, 54(12), 3427–3434. Rayfield, E. J., Ault, M. J., Keusch, G. T., Brothers, M. J., Nechemias, C., & Smith, H. (1982). Infection and diabetes: the case for glucose control. American Journal of Medicine, 72(3), 439–450. Rosenstock, J., Vico, M., Wei, L., Salsali, A., & List, J. F. (2012). Effects of dapagliflozin, an SGLT2 inhibitor, on HbA(1c), body weight, and hypoglycemia risk in patients with type 2 diabetes inadequately controlled on pioglitazone monotherapy. Diabetes Care, 35(7), 1473–1478. Salvatore, T., Carbonara, O., Cozzolino, D., Torella, R., Nasti, R., Lascar, N., et al. (2011). Kidney in diabetes: from organ damage target to therapeutic target. Current Drug Metabolism, 12(7), 658–666. Sauerwein, D. (2002). Urinary tract infection in patients with neurogenic bladder dysfunction. International Journal of Antimicrobial Agents, 19(6), 592–597. Strojek, K., Yoon, K. H., Hruba, V., Elze, M., Langkilde, A. M., & Parikh, S. (2011). Effect of dapagliflozin in patients with type 2 diabetes who have inadequate glycaemic control with glimepiride: a randomised, 24-week, double-blind, placebocontrolled trial. Diabetes, Obesity & Metabolism, 13(10), 928–938. Wilding, J. P., Norwood, P., T'joen, C., Bastien, A., List, J. F., & Fiedorek, F. T. (2009). A study of dapagliflozin in patients with type 2 diabetes receiving high doses of insulin plus insulin sensitizers: applicability of a novel insulin-independent treatment. Diabetes Care, 32(9), 1656–1662. Wilding, J. P., Woo, V., Soler, N. G., Pahor, A., Sugg, J., Rohwedder, K., & Parikh, S.; Dapagliflozin 006 Study Group. (2012). Long-term efficacy of dapagliflozin in patients with type 2 diabetes mellitus receiving high doses of insulin: a randomized trial. Annals of Internal Medicine, 156(6), 405–415.
Contents lists available at ScienceDirect
Journal of Diabetes and Its Complications journal homepage: WWW.JDCJOURNAL.COM
Urinary tract infections in patients with diabetes treated with dapagliflozin☆,☆☆ Kristina M. Johnsson a,⁎, Agata Ptaszynska b, Bridget Schmitz b, Jennifer Sugg c, Shamik J. Parikh c, James F. List b a b c
AstraZeneca, Mölndal, Sweden Bristol-Myers Squibb, Princeton, NJ, USA AstraZeneca, Wilmington, DE, USA
a r t i c l e
i n f o
Article history: Received 3 January 2013 Received in revised form 29 April 2013 Accepted 7 May 2013 Available online 10 July 2013 Keywords: Dapagliflozin Glucosuria Sodium glucose cotransporter 2 SGLT2 Urinary tract infection UTI
a b s t r a c t Aims: Urinary tract infection is common in patients with type 2 diabetes. Possible causative factors include glucosuria, which is a result of treatment with sodium glucose cotransporter 2 (SGLT2) inhibitors. Dapagliflozin is an investigative SGLT2 inhibitor with demonstrated glycemic benefits in patients with diabetes. Data from dapagliflozin multi-trial safety data were analyzed to clarify the association between glucosuria and urinary tract infection. Methods: Safety data from 12 randomized, placebo-controlled trials were pooled to evaluate the relationship between glucosuria and urinary tract infection in patients with inadequately controlled diabetes (HbA1c N 6.5%–12%). Patients were treated with dapagliflozin (2.5, 5, or 10 mg) or placebo once daily, either as monotherapy or add-on to metformin, insulin, sulfonylurea, or thiazolidinedione for 12–24 weeks. The incidence of clinical diagnoses and events suggestive of urinary tract infection were quantified. Results: This analysis included 3152 patients who received once-daily dapagliflozin (2.5 mg [n = 814], 5 mg [n = 1145], or 10 mg [n = 1193]) as monotherapy or add-on treatment, and 1393 placebo-treated patients. For dapagliflozin 2.5 mg, 5 mg, 10 mg, and placebo, diagnosed infections were reported in 3.6%, 5.7%, 4.3%, and 3.7%, respectively. Urinary glucose levels, but not the incidence of urinary tract infection, increased progressively with dapagliflozin dosage. Most identified infections were those considered typical for patients with diabetes. Discontinuations due to urinary tract infection were rare: 8 (0.3%) dapagliflozin-treated patients and 1 (0.1%) placebo-treated patient. Most diagnosed infections were mild to moderate and responded to standard antimicrobial treatment. Conclusions: Treatment of type 2 diabetes with once-daily dapagliflozin 5 or 10 mg is accompanied by a slightly increased risk of urinary tract infection. Infections were generally mild to moderate and clinically manageable. This analysis did not demonstrate a definitive dose relationship between glucosuria and urinary tract infection. © 2013 Elsevier Inc. All rights reserved.
1. Introduction Urinary tract infection (UTI) is a common occurrence in patients with type 2 diabetes (Boyko, Fihn, Scholes, Abraham, & Monsey, 2005; Donders, 2002; Geerlings, 2008; Goswami et al., 2000; Joshi, Caputo, Weitekamp, & Karchmer, 1999), although a precise cause–effect relationship has not been determined. Multiple factors may be involved, ☆ Clinical trial registration numbers: NCT00263276, NCT00972244, NCT00528372, NCT00736879, NCT00528879, NCT00855166, NCT00357370, NCT00680745, NCT00683878, NCT00673231, NCT00643851, NCT00859898. Available at http://clinicaltrials.gov. ☆☆ Disclosure Statement: The studies included in this analysis were sponsored by Bristol-Myers Squibb and AstraZeneca. K.M. Johnsson, J. Sugg, and S.J. Parikh are employees and stockholders of AstraZeneca. A. Ptaszynska, B. Schmitz, and J.F. List are employees and stockholders of Bristol-Myers Squibb. ⁎ Corresponding author. AstraZeneca, R&D Mölndal, SE-43183 Mölndal, Sweden. Tel.: +46 31 7065562. E-mail address: [email protected] (K.M. Johnsson). 1056-8727/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jdiacomp.2013.05.004
such as hyperglycemia, glucosuria, and neurogenic bladder (Geerlings, 2008; Hoepelman, Meiland, & Geerlings, 2003; Rackley, 2011; Rayfield et al., 1982; Sauerwein, 2002). Hyperglycemia can impede the body's resistance to microorganisms and lead to increased adherence of bacteria to uroepithelial cells (Geerlings, 2008; Rayfield et al., 1982). There are inadequate data quantifying the relationship between the level of glucosuria, bacterial growth in the urine, and incidence of UTI. In an analysis of the growth of Escherichia coli in human urine, the growth rate of E. coli was enhanced when glucose was added to urine, although very high glucose concentrations resulted in a decreased rate when the urinary pH was not kept constant (Geerlings, Brouwer, Gaastra, Verhoef, & Hoepelman, 1999). In a statistical analysis of a retrospective study of women with diabetes, glucosuria was not identified as a risk factor for UTI (Geerlings et al., 2000a, 2000b). Glucosuria is of particular interest due to its relationship to an investigational class of antidiabetes therapy that targets sodium glucose cotransporter 2 (SGLT2), the protein that mediates reabsorption of
474
K.M. Johnsson et al. / Journal of Diabetes and Its Complications 27 (2013) 473–478
most of the glucose filtered by the kidney. SGLT2 inhibitors act independently of insulin secretion or action to reduce glucose reabsorption, thus resulting in increased urinary glucose excretion with a corresponding reduction in blood glucose levels (Ghosh, Ghosh, Chawla, & Jasdanwala, 2011; Rahmoune et al., 2005; Salvatore et al., 2011). Dapagliflozin is a novel, orally administered, selective, potent SGLT2 inhibitor in development for the treatment of type 2 diabetes (Han et al., 2008; Meng et al., 2008). Dapagliflozin has been studied in patients in placebo-controlled clinical studies as monotherapy, as add-on therapy to other standard antidiabetic treatments, and as firstline combination therapy with metformin (Bailey, Gross, Pieters, Bastien, & List, 2010; Ferrannini, Ramos, Salsali, Tang, & List, 2010; List, Woo, Morales, Tang, & Fiedorek, 2009; Rosenstock, Vico, Wei, Salsali, & List, 2012; Strojek et al., 2011; Wilding et al., 2009; Wilding et al., 2012). In each study, patients who received dapagliflozin achieved significantly greater glycemic control than those who received placebo, and dapagliflozin was generally well tolerated. Data from clinical pharmacology studies demonstrated dose-related increases in glucosuria for doses of 2.5, 5, and 10 mg (Parikh et al., 2011); increases were smaller at doses N 10 mg (List, Woo, Morales, Tang, & Fiedorek, 2009). Because its mechanism of action leads to increased urinary glucose excretion, controlled clinical trials of dapagliflozin offer an opportunity to evaluate the relationship between glucosuria and UTI. 2. Subjects, materials and methods 2.1. Clinical trials and patients Data from 12 randomized, placebo-controlled trials (Appendix 1) were pooled to evaluate the relationship between pharmacologically induced urinary glucose excretion and UTI in patients with type 2 diabetes treated with dapagliflozin. Details regarding the methodologies of these trials are provided in the accompanying article “Vulvovaginitis and balanitis in patients with diabetes treated with dapagliflozin” (Johnsson et al., 2013). Patients with a history or risk factors for UTI were not excluded from participation in the clinical trials. All studies were conducted in accordance with Good Clinical Practice guidelines and the Declaration of Helsinki and were approved by the institutional review boards or independent ethics committee of each participating center. All patients provided written informed consent to participate in their respective clinical trials. 2.2. Safety signal detection and quantification Due to the potential for an increased risk in the incidence of UTI associated with treatment-induced glucosuria, a comprehensive approach was undertaken to broaden the potential for detection of any signal indicating an increased risk. Throughout the dapagliflozin clinical trials program, a rigorous effort was made to capture all signs, symptoms and events suggestive of UTI. In addition to spontaneous reports of symptoms by study participants, investigators proactively questioned patients at each visit about symptoms suggestive of UTI (either experienced currently or since the previous study visit). This proactive questioning was designed to address the possibility that patients might not recognize certain symptoms as being relevant. Supplemental case-report forms were provided to obtain more detailed information for assessment. For the initial analysis, a broad net was generated using a set of 63 prespecified, preferred terms (Appendix 2) from the Medical Dictionary for Regulatory Activities (MedDRA) version 13.0 to capture signs and symptoms (eg, dysuria) suggestive of UTI, as well as specific clinical diagnoses. These preferred terms were referred to as “events suggestive of UTI.” When a prespecified, preferred term was reported,
a case-report questionnaire was completed by the investigator to fully describe the case and the risk factors. In a second, more specific analysis, 49 prespecified, preferred terms for clinical diagnosis were used to quantify diagnosed UTIs (Appendix 2). To increase the specificity of diagnosis, investigators were also asked to obtain a urine culture to confirm diagnosis in patients with suspected UTI. Study protocol required the drug to be withheld in patients with clinical evidence of pyelonephritis or presumed urosepsis until treatment of the infection was complete and clinical recovery had occurred. 3. Results 3.1. Patients The population for this analysis included 4545 patients from the 12 clinical trials. Treatment groups included 3152 patients who received once-daily dapagliflozin (2.5 mg [n = 814], 5 mg [n = 1145], or 10 mg [n = 1193]) and 1393 patients who received placebo. These study groups were generally balanced with respect to baseline demographics and disease characteristics (Table 1). Baseline mean HbA1c was 8.1% to 8.4%. More than 85% of patients had BMI ≥ 25 kg/m 2, and N55% had BMI ≥30 kg/m 2. A varied range of disease duration (mean 5.3–6.7 years) and progression were represented in the patient population. In dapagliflozin-treated patients, mean exposure ranged from 148.2 to 150.5 days. Mean duration of exposure for those treated with placebo was 149.4 days. A dose-dependent increase in glucosuria was documented at 24 weeks in this pooled analysis; mean change from baseline in spot fasting urine glucose excretion for placebo, 2.5 mg, 5 mg, and 10 mg was − 241.0 mg/dL (SE 36.2), + 1480.5 mg/dL (SE 68.6), + 2149.9 (SE 66.7), and + 2592.3 (SE 65.6), respectively. Where long-term data for glucose excretion were available, levels were consistent with the short-term analysis. 3.2. Events suggestive of UTI Of the 63 prespecified, preferred MedDRA terms used to capture signs, symptoms, and events suggestive of UTI, 16 were reported in ≥ 1 patient (Appendix 2 Figure). Greater proportions of patients in the dapagliflozin 5- and 10-mg groups experienced events suggestive of UTI (7.3% and 6.5%, respectively) than did patients in the dapagliflozin 2.5-mg group and the placebo group (4.2% and 4.5%, respectively) (Appendix 3 Figure). Most episodes were mild or moderate in intensity across all treatment groups. These events were more common in women than in men in all treatment groups. Signs, symptoms, and events suggestive of UTI that were reported in ≥1% of women in one or more of the dapagliflozin groups were the diagnosis of UTI, the symptom of dysuria, and the diagnosis of cystitis (Table 2). UTI and cystitis also occurred in ≥ 1% of women in the placebo group. Dysuria and UTI were reported in ≥1% of men in one or more of the dapagliflozin groups but not in the placebo group. Urine cultures were obtained in 39% to 50% of dapagliflozin patients and in 50% of placebo patients with events suggestive of UTI. Approximately two thirds of the cultures across all treatment groups were positive; most of the organisms identified were those commonly seen in patients with type 2 diabetes (eg, E. coli, Klebsiella pneumoniae, and Proteus). To begin to understand the potential risk factors for UTI, incidence rates were determined based on various subgroups, including categories of baseline HbA1c (b8%, ≥ 8% to b9%, and ≥ 9%), age (b 65 and ≥ 65 years), gender, and history of recurrent infection (Fig. 1). For all subgroups, there was variation in the background rates of events suggestive of UTI; however, within each subgroup, there was a proportionate increase in incidence for dapagliflozin vs placebo. There
K.M. Johnsson et al. / Journal of Diabetes and Its Complications 27 (2013) 473–478
475
Table 1 Demographic and baseline characteristics of patients. Characteristic
Placebo (n = 1393)
Dapagliflozin 2.5 mg/d (n = 814)
Age (y), mean (SD) Sex, n (%) Women Men Race, n (%) White Black/African American Asian Other Weight (kg), mean (SD) Body mass index, n (%) ≥25 kg/m2 ≥30 kg/m2 Diabetes duration (y), mean (SD) HbA1c (%), mean (SD) a
5 mg/d (n = 1145)
10 mg/d (n = 1193)
55.3 (10.45)
57.0 (10.06)
55.4 (10.35)
55.1 (10.14)
677 (48.6) 716 (51.4)
400 (49.1) 414 (50.9)
581 (50.7) 564 (49.3)
598 (50.1) 595 (49.9)
653 (80.2) 15 (1.8) 127 (15.6) 19 (2.3) 86.1 (19.55)
907 33 181 24 86.0
(79.2) (2.9) (15.8) (2.1) (19.03)
976 35 152 30 88.0
(81.8) (2.9) (12.7) (2.5) (19.43)
715 (87.8) 472 (58.0) 6.7 (6.94) 8.1 (0.87)
1003 660 5.7 8.4
(87.6) (57.6) (6.64) (1.09)
1082 700 6.2 8.3
(90.7) (58.7) (6.65) (1.05)
1129 38 201 25 87.4
(81.0) (2.7) (14.4) (1.8) (19.46)
1241 (89.1) 794 (57.0) 5.3a (6.37) 8.4 (1.16)
n = 1392.
were no clear trends regarding dosing within any subgroup. More events suggestive of UTI were documented in patients ≥65 years old vs younger patients treated with dapagliflozin; more events were documented in females vs males across all treatment groups. The rate of events suggestive of UTI also were analyzed in patients who completed the short-term treatment period and continued in a long-term treatment period of their respective trials for ≤102 weeks of total treatment (dapagliflozin n = 2160, placebo n = 694). In the long-term analyses, the proportions of patients with events suggestive of UTI were slightly higher vs the short-term treatment period. In the short-term treatment period, proportions were 4.2% to 7.3% in the dapagliflozin groups and 4.5% in the placebo group; in the combined treatment periods, proportions were 9.4% in the dapagliflozin groups and 6.6% in the placebo group. An analysis of time to onset of first event suggestive of UTI indicated that patients treated with dapagliflozin 5 or 10 mg were at greater risk for a first event than those who received dapagliflozin 2.5 mg or placebo, starting at approximately 8 weeks after treatment initiation (Appendix 4 Figure). Across all treatment groups in the combined treatment periods, first events occurred relatively early during the trials; a first event was more likely to occur in the first 24 weeks than after 24 weeks. 3.3. Clinical diagnosis of UTI Of the 49 prespecified, preferred MedDRA terms related to specific clinical diagnoses of UTI, 9 were reported in ≥1 patient (Appendix 2 Figure). A total of 230 patients were diagnosed with UTI: 56 for placebo, 35 for dapagliflozin 2.5 mg, 76 for dapagliflozin 5 mg, and 63
for dapagliflozin 10 mg. The rates of clinically diagnosed cases were higher in the dapagliflozin 5- and 10-mg groups (5.7% and 4.3%, respectively) vs the dapagliflozin 2.5-mg and placebo groups (3.6% and 3.7%, respectively) (Appendix 3 Figure). The incidence of diagnosed infection was analyzed for statistical significance, both for the total pooled population and for subgroups stratified by baseline HbA1c (Table 3). The frequency of UTI was significantly greater vs placebo for patients treated with dapagliflozin 5 mg in the total pooled population; no other comparisons were statistically significant. Overall, clinical diagnosis of UTI was more common in dapagliflozin-treated patients than in the placebo group. The rates were higher for women than men across treatment groups. Treatment for clinically diagnosed UTI was administered by clinical investigator or by patient's physician. Treatment was not specified in the trial protocols and was initiated according to physician choice and local treatment guidelines. Of the 174 dapagliflozin-treated patients documented as having a clinical diagnosis of UTI, treatment was documented in 78%. Most of these cases were managed with one course of standard antimicrobial treatment (Fig. 2). However, 5.7%, 1.3%, and 15.9% of episodes in the dapagliflozin 2.5-, 5-, and 10-mg groups, respectively, and 14.3% of episodes in the placebo group required more than 1 course of treatment. Pyelonephritis was a rare occurrence with similar frequencies in each treatment group: 2 (0.2%) in the dapagliflozin 2.5-mg group, 1 (0.1%) in the dapagliflozin 5-mg group, none in the dapagliflozin 10-mg group, and 1 (0.1%) in the placebo-treated group. The only case that was considered by the investigator to be a serious adverse event was in the placebo group.
Table 2 UTI infections occurring in ≥1% of patients in any treatment group, by gender. No. (%) of patients
Placebo (n = 1393)
Dapagliflozin 2.5 mg/d (n = 814)
5 mg/d (n = 1145)
10 mg/d (n = 1193)
Women Signs, symptoms, and events suggestive of UTI UTI Dysuria Cystitis Clinical diagnoses of UTI UTI Cystitis Men Signs, symptoms, and events suggestive of UTI Dysuria UTI Clinical diagnoses of UTI UTI
677 (49)
400 (49)
581 (51)
598 (50)
35 (5.2) 6 (0.9) 10 (1.5)
21 (5.3) 3 (0.8) 2 (0.5)
48 (8.3) 10 (1.7) 7 (1.2)
39 (6.5) 12 (2.0) 7 (1.2)
35 (5.2) 10 (1.5) 716 (51)
21 (5.3) 2 (0.5) 414 (51)
48 (8.3) 7 (1.2) 564 (49)
39 (6.5) 7 (1.2) 595 (50)
4 (0.6) 3 (0.4)
2 (0.5) 4 (1.0)
8 (1.4) 6 (1.1)
13 (2.2) 4 (0.7)
3 (0.4)
4 (1.0)
6 (1.1)
4 (0.7)
476
K.M. Johnsson et al. / Journal of Diabetes and Its Complications 27 (2013) 473–478
Fig. 1. Percentage of patients with ≥1 event suggestive of UTI by baseline subgroups: (A) HbA1c category, (B) age category, (C) gender, and (D) history of recurrent UTI. Data represent placebo-controlled pool from short-term, double-blind treatment periods and include data after rescue.
Two patients in the dapagliflozin 2.5-mg group experienced urinary tract events identified by the investigator as serious adverse events: 1 with UTI and 1 with malacoplakia vesicae. One episode of urosepsis (not a prespecified, preferred MedDRA term) was reported in the dapagliflozin 10-mg group as a serious adverse event. Dapagliflozin was not discontinued, and the patient completed the study. Few patients discontinued treatment as a result of specific clinical diagnoses of UTI: 8 (0.3%) patients in the dapagliflozin groups and 1 (0.1%) patient in the placebo group. Among dapagliflozintreated patients, 2 patients in the 2.5-mg group discontinued because of pyelonephritis, 2 in the 5-mg group and 3 in the 10-mg group discontinued because of UTI, and 1 in the 2.5-mg group discontinued because of cystitis. In the placebo group, 1 patient discontinued because of pyelonephritis. The proportions of patients with a history of recurring UTIs were relatively small across treatment groups (2.0%–2.8%). During the
study period, greater proportions of these patients with history were diagnosed with UTI (17.1%–21.1%) than those without a history of recurrent infection (3.2%–5.5%). There was no clear trend for recurrence with dapagliflozin vs placebo among patients with history of recurrent infection. During the trials, recurrence of UTI events was reported in 15.3% to 19.6% of dapagliflozin-treated patients vs 7.7% of placebo-treated patients. In the trials up to 2 years, recurrent UTI was reported in 22.9% of dapagliflozin-treated patients vs 13.6% of placebo-treated patients. The frequency of recurrent infection during the trials did not appear to be dose dependent. 4. Discussion A rigorous effort was made to accurately determine the level of risk of UTI associated with dapagliflozin treatment by means of the dual approach of spontaneous reporting and proactive questioning of
Table 3 Patients with diagnosed UTI. Placebo (n = 1393)
Total pooled population No. patients with events (%) 52 (3.7) P value vs placebo Baseline HbA1c category, no. patients with event (%) b8% 19 (3.3) P value vs placebo ≥8% and b9% 17 (4.0) P value vs placebo ≥9% 16 (4.2) P value vs placebo
Dapagliflozin 2.5 mg/d (n = 814)
5 mg/d (n = 1145)
10 mg/d (n = 1193)
29 (3.6) .8373
65 (5.7) .0201
51 (4.3) .4823
15 (3.9) .5822 11 (3.8) .8721 3 (2.2) .4272
24 (5.5) .0791 26 (6.7) .0826 15 (4.7) .7441
17 (3.2) .9751 20 (5.3) .3843 14 (4.9) .6591
Statistical comparisons were conducted for each dapagliflozin group vs placebo using chi-square tests, unless the number of events is ≤5 in either group, in which case, Fisher exact test was used.
K.M. Johnsson et al. / Journal of Diabetes and Its Complications 27 (2013) 473–478
477
Fig. 2. Characterization of episodes of UTI: (A) documentation of non-specific signs, symptoms vs clinical diagnosis; (B) treatment of clinically diagnosed cases.
patients. It was believed that proactive questioning would increase the reporting of events suggestive of UTI and remind patients to be watchful for the symptoms discussed at clinic visits, thereby encouraging them to report any relevant events. Proactive questioning should also remind patients to disclose any treatments administered outside the study. Clinical diagnosis of UTI was made in the dapagliflozin studies by collection of urine cultures, by opinion of experienced clinicians, or by response to treatment administered after a presumed diagnosis. Because it is common in clinical practice to treat symptomatic UTIs without obtaining a culture, relying solely on culture results in these trials would have led to an erroneously low reported incidence of infection. Conversely, if urinalyses had routinely been conducted on all study participants without regard to the presence or absence of symptoms, cases of asymptomatic bacteriuria or pyuria not warranting treatment would have skewed the reported incidence of UTI in the opposite direction. The protocol-specified
approach to the detection and quantification of a signal for increased UTI was reflective of general clinical practice and was considered to be a realistic and optimal means of quantifying increased risk. There is a presumed and considerably debated association between glucosuria and UTI. In our Phase 3 trials, pooled data indicate a small increase in the incidence of UTI in patients treated with dapagliflozin 5 or 10 mg vs those treated with placebo. However, the incidence was relatively low across all treatment groups (b6%), the rate of infection in the 2.5-mg group was similar to that seen with placebo, and there was no consistent dose-related trend across studies. Generally, first events of UTI occurred relatively early in the studies. Clinical diagnosis of UTI occurred more commonly in women than in men. The pathogens reported were types commonly seen in patients with type 2 diabetes. Most events were of mild to moderate intensity and resolved with one course of standard antimicrobial treatment. Discontinuations of dapagliflozin due to clinical diagnosis
Fig. 3. Comparison of dose-related trends in glucosuria and clinical diagnosis of UTI in the dapagliflozin trials.
478
K.M. Johnsson et al. / Journal of Diabetes and Its Complications 27 (2013) 473–478
of UTI were rare, which indicates that study participants found these events to be tolerable. The proportions of dapagliflozin-treated patients and placebo-treated patients diagnosed with pyelonephritis were identical (0.1% for each); the only case leading to discontinuation of treatment was in the placebo group. It has been established that treatment with dapagliflozin produces glucosuria with a reduction in blood glucose levels (Ferrannini, Ramos, Salsali, Tang, & List, 2010; Komoroski et al., 2009; List, Woo, Morales, Tang, & Fiedorek, 2009). In this pooled safety analysis of 12 clinical trials, glucose excretion was shown to be progressively greater with increased doses of dapagliflozin; although as Fig. 3 illustrates, a similar dose-dependent relationship with the incidence of UTI was not demonstrated. While the data suggest that glucosuria is a risk factor for the development of UTI, it is important to recognize potential confounding factors. As an example, patients with a history of recurrent UTIs, including patients treated with placebo, experienced much higher rates of UTI during the trials, indicating a predisposition to infection unrelated to glucosuria. The variation in patient histories with regard to recurrent UTIs limits the ability to generalize findings to the overall patient population. This general pattern also holds true for the relationship between glucosuria and genital infection in dapagliflozin-treated patients (Johnsson et al., 2013); however, the signal for increased risk is clearer for genital infection. This inconsistency may reflect a greater effect of glucosuria in promoting the growth of fungal pathogens typically associated with genital infection vs bacterial pathogens typically associated with UTI. Further study to better elucidate the relationship between glucosuria and these infections is warranted. 5. Conclusions This pooled safety analysis of 12 clinical trials determined that treatment of type 2 diabetes with once-daily dapagliflozin 5 or 10 mg is accompanied by a small increase in risk of UTI, but no evidence of increase in pyelonephritis. These infections were considered typical for patients with type 2 diabetes, they generally did not lead to discontinuation of treatment, and recurrence was uncommon. Infections were generally mild to moderate, clinically manageable, and not associated with complications. A dose relationship with the 5- and 10-mg doses was not evident. Acknowledgments This study was funded by Bristol-Myers Squibb and AstraZeneca. Ann L. Davis, MPH, CMPP, an employee of Bristol-Myers Squibb, assisted in the development of the manuscript. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jdiacomp.2013.05.004. References Bailey, C. J., Gross, J. L., Pieters, A., Bastien, A., & List, J. F. (2010). Effect of dapagliflozin in patients with type 2 diabetes who have inadequate glycaemic control with metformin: a randomised, double-blind, placebo-controlled trial. The Lancet, 375(9733), 2223–2233. Boyko, E. J., Fihn, S. D., Scholes, D., Abraham, L., & Monsey, B. (2005). Risk of urinary tract infection and asymptomatic bacteriuria among diabetic and nondiabetic postmenopausal women. American Journal of Epidemiology, 161(6), 557–564. Donders, G. G. (2002). Lower genital tract infections in diabetic women. Current Infectious Disease Reports, 4(6), 536–539.
Ferrannini, E., Ramos, S. J., Salsali, A., Tang, W., & List, J. F. (2010). Dapagliflozin monotherapy in type 2 diabetic patients with inadequate glycemic control by diet and exercise: a randomized, double-blind, placebo-controlled, phase 3 trial. Diabetes Care, 33(10), 2217–2224. Geerlings, S. E. (2008). Urinary tract infections in patients with diabetes mellitus: epidemiology, pathogenesis and treatment. International Journal of Antimicrobial Agents, 31(Suppl 1), S54–S57. Geerlings, S. E., Brouwer, E. C., Gaastra, W., Verhoef, J., & Hoepelman, A. I. (1999). Effect of glucose and pH on uropathogenic and non-uropathogenic Escherichia coli: studies with urine from diabetic and non-diabetic individuals. Journal of Medical Microbiology, 48(6), 535–539. Geerlings, S. E., Stolk, R. P., Camps, M. J., Netten, P. M., Collet, T. J., & Hoepelman, A. I.; Diabetes Women Asymptomatic Bacteriuria Utrecht Study Group. (2000a). Risk factors for symptomatic urinary tract infection in women with diabetes. Diabetes Care, 23(12), 1737–1741. Geerlings, S. E., Stolk, R. P., Camps, M. J. et al.; Diabetes Women Asymptomatic Bacteriuria Utrecht Study Group(2000b). Asymptomatic bacteriuria may be considered a complication in women with diabetes. Diabetes Care, 23(6), 744–749. Ghosh, R. K., Ghosh, S. M., Chawla, S., & Jasdanwala, S. A. (2012). SGLT2 inhibitors: a new emerging therapeutic class in the treatment of type 2 diabetes mellitus. The Journal of Clinical Pharmacology, 52(4), 457–463. Goswami, R., Dadhwal, V., Tejaswi, S., Datta, K., et al. (2000). Species-specific prevalence of vaginal candidiasis among patients with diabetes mellitus and its relation to their glycaemic status. The Journal of Infection, 41(2), 162–166. Han, S., Hagan, D. L., Taylor, J. R., Xin, L., et al. (2008). Dapagliflozin, a selective SGLT2 inhibitor, improves glucose homeostasis in normal and diabetic rats. Diabetes, 57(6), 1723–1729. Hoepelman, A. I., Meiland, R., & Geerlings, S. E. (2003). Pathogenesis and management of bacterial urinary tract infections in adult patients with diabetes mellitus. International Journal of Antimicrobial Agents, 22(Suppl), 235–243. Johnsson, K. M., Ptaszynska, A., Schmitz, B., Sugg, J., Parikh, S. J., & List, J. F. (2013). Vulvovaginitis and balanitis in patients with diabetes treated with dapagliflozin. Journal of Diabetes and its Complications, 27(5), 479–484. Joshi, N., Caputo, G. M., Weitekamp, M. R., & Karchmer, A. W. (1999). Infections in patients with diabetes mellitus. The New England Journal of Medicine, 341(25), 1906–1912. Komoroski, B., Vachharajani, N., Feng, Y., Li, L., Kornhauser, D., & Pfister, M. (2009). Dapagliflozin, a novel, selective SGLT2 inhibitor, improved glycemic control over 2 weeks in patients with type 2 diabetes mellitus. Clinical Pharmacology and Therapeutics, 85(5), 513–519. List, J. F., Woo, V., Morales, E., Tang, W., & Fiedorek, F. T. (2009). Sodium-glucose cotransport inhibition with dapagliflozin in type 2 diabetes. Diabetes Care, 32(4), 650–657. Meng, W., Ellsworth, B. A., Nirschl, A. A., McCann, P. J., et al. (2008). Discovery of dapagliflozin: a potent, selective renal sodium-dependent glucose cotransporter 2 (SGLT2) inhibitor for the treatment of type 2 diabetes. Journal of Medicinal Chemistry, 51(5), 1145–1149. Parikh, S. J., Johnsson, K. M., Ptaszynska, A., Schmitz, B., et al. (2011). Characterization of urinary tract infections in the setting of pharmacologically-induced glucosuria. Presented at: 71st Scientific Sessions of the American Diabetes Association, June 24–28, 2011. San Diego, CA. Rackley, R. (2011). Neurogenic bladder. Medscape, Available at: http://emedicine. medscape.com/article/453539-overview. Accessed December 11, 2011. Rahmoune, H., Thompson, P. W., Ward, J. M., Smith, C. D., Hong, G., & Brown, J. (2005). Glucose transporters in human renal proximal tubular cells isolated from the urine of patients with non-insulin-dependent diabetes. Diabetes, 54(12), 3427–3434. Rayfield, E. J., Ault, M. J., Keusch, G. T., Brothers, M. J., Nechemias, C., & Smith, H. (1982). Infection and diabetes: the case for glucose control. American Journal of Medicine, 72(3), 439–450. Rosenstock, J., Vico, M., Wei, L., Salsali, A., & List, J. F. (2012). Effects of dapagliflozin, an SGLT2 inhibitor, on HbA(1c), body weight, and hypoglycemia risk in patients with type 2 diabetes inadequately controlled on pioglitazone monotherapy. Diabetes Care, 35(7), 1473–1478. Salvatore, T., Carbonara, O., Cozzolino, D., Torella, R., Nasti, R., Lascar, N., et al. (2011). Kidney in diabetes: from organ damage target to therapeutic target. Current Drug Metabolism, 12(7), 658–666. Sauerwein, D. (2002). Urinary tract infection in patients with neurogenic bladder dysfunction. International Journal of Antimicrobial Agents, 19(6), 592–597. Strojek, K., Yoon, K. H., Hruba, V., Elze, M., Langkilde, A. M., & Parikh, S. (2011). Effect of dapagliflozin in patients with type 2 diabetes who have inadequate glycaemic control with glimepiride: a randomised, 24-week, double-blind, placebocontrolled trial. Diabetes, Obesity & Metabolism, 13(10), 928–938. Wilding, J. P., Norwood, P., T'joen, C., Bastien, A., List, J. F., & Fiedorek, F. T. (2009). A study of dapagliflozin in patients with type 2 diabetes receiving high doses of insulin plus insulin sensitizers: applicability of a novel insulin-independent treatment. Diabetes Care, 32(9), 1656–1662. Wilding, J. P., Woo, V., Soler, N. G., Pahor, A., Sugg, J., Rohwedder, K., & Parikh, S.; Dapagliflozin 006 Study Group. (2012). Long-term efficacy of dapagliflozin in patients with type 2 diabetes mellitus receiving high doses of insulin: a randomized trial. Annals of Internal Medicine, 156(6), 405–415.