■ THERAPY FOCUS

SGLT2 inhibitors and risk of genitourinary infections

STEVE CHAPLIN

The SGLT2 inhibitors dapagliflozin, 4500 DPP-4 inhibitors and are 4000 SGLT2 inhibitors now becoming established in the 3500 management of , but 3000 they are associated with an increased 2500 risk of genital and urinary tract infections, compounding the problem 2000 that people with diabetes are already at

Items (1000s) 1500 higher risk than the general population. 1000 This article examines the extent of these 500 risks and how they may affect treatment 0 adherence. 2007/ 2008/ 2009/ 2010/ 2011/ 2012/ 2013/ 2014/ 2015/ 2008 2009 2010 2011 2012 2013 2014 2015 2016 he first sodium-glucose co-transporter 2 (SGLT2) inhibitor, Year Tdapagliflozin, was introduced in the UK in 2012. Since then, dapagliflozin, canagliflozin and empagliflozin have become established in the management of type 2 diabetes and are now Figure 1. Prescribing volume of dipeptidylpeptidase (DPP-4) inhibitors and recommended as options for certain patients for first intensifi- sodium-glucose co-transporter 2 (SGLT2) inhibitors in primary care, England, cation (dual therapy) and second intensification (triple therapy), 2007/8–2015/163 including in combination with .1 They are also options as monotherapy (almost as a last resort) when , a sulfo- nylurea, are unsuitable and if a dipeptidyl peptidase 4 (DPP-4) inhibitor (gliptin) would otherwise be prescribed.2 SGLT2 inhibitors are predominantly prescribed in primary care and the rate of increase in prescribing is matching that of the DPP-4 inhibitors at a similar phase of their life cycle (see Figure 1).3 Uptake is, however, currently relatively low: in 2015, GPs in England wrote 646,000 prescriptions for SGLT2 inhib- itors out of a total of 35 million for “other antidiabetic drugs” (BNF category 6.1.2).4 But that could change in light of evidence that the SGLT2 inhibitors may reduce mortality in patients with diabetes and cardiovascular disease. The recent EMPA-REG OUTCOME trial showed that, over a median follow-up of about three years, empagliflozin 10 or 25mg plus standard care significantly reduced the risk of the primary composite endpoint (cardiovas- cular death, myocardial infarction or nonfatal stroke) compared with placebo (10.5 vs 12.1 per cent; hazard ratio 0.86), as well as cardiovascular death (3.5–3.8 vs 5.9 per cent; hazard ratio 0.59–0.65) and all-cause mortality (5.6–5.8 vs 8.3 per cent; hazard ratio 0.67–0.70).5 Even aspiring to such a target would secure the future of SGLT2 inhibitors in clinical practice. Patients and GPs will therefore need to manage the most fre- quent adverse effects, notably genitourinary infections.

26 ❚ Prescriber December 2016 prescriber.co.uk SGLT2 inhibitors l THERAPY FOCUS ■

SGLT2 inhibition Age (years) Vaginitis Balanitis SGLT2 is the major co-transporter for renal glucose reabsorp- tion. Its inhibition by a gliflozin increases glucose by 18–39 1.91 (1.46–2.48) 7.60 (3.57–16.20) 70–80g per day (and up to about 120g per day for canagliflozin 300mg daily). This is associated with an increase in urinary fre- 40–49 1.96 (1.54–2.50) 5.48 (3.43–8.74) quency due to the osmotic effect of glucose, and volume increases by around 400ml per day (about the same as the 50–59 1.97 (1.57–2.47) 3.73 (2.54–5.49) functional bladder volume in adults). These effects cause a fall 60–69 1.77 (1.44–2.18) 2.06 (1.50–2.82) in blood glucose and, secondary to a reduction in blood volume, a modest decrease in blood pressure. 70+ 1.53 (1.25–1.87) 1.56 (1.11–2.19) Although direct comparisons are lacking, meta-analysis of clinical trials indicates there is little difference in efficacy between Table 1. Relative risk (95% CI) of vaginitis or balanitis according to the SGLT2 inhibitors.6 Overall, the safety of SGLT2 inhibitors in age among people with type 2 diabetes compared with no diabetes clinical trials included in regulatory submissions was similar and (adjusted for index year and prior genital tract infection)12 the incidence of adverse events was comparable with placebo, with 2–6 per cent of patients stopping treatment as a result.7-9 Risk of urinary tract infections SGLT2 inhibitors are associated with a small increase in the Genitourinary infections and diabetes risk of UTI compared with placebo and most episodes are People with diabetes are at increased risk of genitourinary infec- symptomatic.18 Reported figures for the incidence of UTIs in tions. In a UK primary care population, the one-year incidence clinical trials vary but fall within the range 4–9 per cent, with of urinary tract infection (UTI) was about 50 per cent greater a difference of 1.0–1.5 per cent above that reported with pla- among people with diabetes than among those without (46.9 cebo.7-9 More women are affected than men; in phase 3 trials vs 29.9 per 1000 person-years).10 Vaginitis is twice as likely of canagliflozin, 88 per cent of UTI adverse effects were in among women with diabetes than those without (incidence 21.0 women.18 The incidence is much higher among patients with a vs 10.3 per 1000 person-years; adjusted relative risk 1.81) and history of chronic or recurrent UTIs (27 per cent with empagli- balanitis is three times as likely in men with diabetes compared flozin, 24 per cent with placebo7). Data on whether recurrence with those without (incidence 8.4 vs 2.5 per 1000 person-years; is more frequent with SGLT2 inhibitors than with placebo are adjusted relative risk 2.85).11 The incidence of genital infection is contradictory.8,18 Compared with other antidiabetic agents, higher among younger than older adults and the increased risk in SGLT2 inhibitors are associated with a 40 per cent higher people with diabetes is correspondingly greater (see Table 1).12 risk of UTIs.19 Factors that contribute to the higher risk of genitourinary UTIs associated with SGLT2s are described as mainly mild infection in people with diabetes include bacteriuria associ- to moderate in intensity (see Table 2). They reportedly respond ated with glycosuria; increased adherence of Escherichia coli to to standard treatment and are ‘rarely’ a cause of discontinua- uroepithelial cells and increased virulence of Candida albicans tion.7-9 Severe events or events associated with hospital admis- in individuals with suboptimal glycaemic control; and immune sion were reported in 0.1–0.4 per cent of patients in clinical dysfunction associated with hyperglycaemia.12 SGLT2 inhibitors trials; kidney infections and sepsis were ‘rare’. There were no add to the problem: in one phase 2 study, 12 per cent of 198 differences in these outcomes between SGLT2 inhibitors and women with type 2 diabetes were positive for vaginal Candida placebo. spp. at baseline; after 12 weeks’ treatment with canagliflozin, 31 per cent of women who had been culture negative tested Mild Awareness of symptoms that are easily tolerated, positive compared with 14 per cent of women who received causing minimal discomfort and not interfering with placebo or the DPP-4 inhibitor and, of these, one-third everyday activities to one-quarter were symptomatic.13 Moderate Sufficient discomfort is present to cause interference The impact of single episodes of urinary tract infection (UTI) or with normal activity genital infection on quality of life has not been reported. Recurrent UTIs are associated with impaired quality of life, with measures Severe Extreme distress causing significant impairment of 14,15 demonstrating depression, anxiety and reduced social activity. functioning or incapacitation; preventing normal everyday Recurrent vulvovaginal candidiasis impairs quality of life across activities mental and physical health domains to an extent comparable with asthma or COPD and greater than headache/migraine.16,17 Serious Resulted in death or was life-threatening, required inpatient It is generally believed that the increased risk of genito­urinary hospitalisation or prolongation of existing hospitalisation; infections associated with SGLT2 inhibitors is secondary to glyco- resulted in persistent or significant disability/incapacity; or suria, presumably arising from residual urine in the lower urinary was a congenital anomaly/birth defect tract and on the genitalia. This is a class effect and, although direct comparisons are again lacking and data are sometimes dif- Table 2. Definitions of investigator-assessed intensity of adverse events used in ficult to compare,12 the risks appear to be similar for each agent. clinical trials20 prescriber.co.uk Prescriber December 2016 ❚ 27 ■ THERAPY FOCUS l SGLT2 inhibitors

(a) Women 26 all non-canagliflozin 24 22 canagliflozin 100mg daily 20 canagliflozin 300mg daily 18 16 14 12 10 8 6 4

Estimated % subjects with an event 2 0 0 6 12 18 26 52 78 Time (weeks)

(b) Men 15 all non-canagliflozin 14 13 canagliflozin 100mg daily 12 11 canagliflozin 300mg daily 10 9 8 7 6 5 4 3 2

Estimated % subjects with an event 1 0 0 6 12 18 26 52 78 Time (weeks)

Figure 2. Time to first reported genital mycotic infection in (a) women and (b) men during treatment with the SGLT2 inhibitor canagliflozin compared with untreated subjects (note different y-axes)20 Risk of genital infections Genital infections during treatment with SGLT2 inhibitors – mainly mycotic balanitis and vulvovaginitis – are significantly more common than with placebo7-9 and are five times more likely than with other antidiabetic agents.19 The overall inci- dence of genital infections with SGLT2 inhibitor therapy reported in clinical trials was 4–6 per cent compared with about one per cent with placebo but these events were more common among women (with rates of approximately 7–11 per cent) than men. In a pooled population from canagliflozin clinical trials that included older patients and patients with renal impairment or high cardiovascular risk with a mean exposure to canagliflozin of 17 months, the incidence of genital infections was approximately 14–15 per cent with canagliflozin therapy vs 3 per cent with pla- cebo in women and 7–9 per cent vs 1.6 per cent in men.20

28 ❚ Prescriber December 2016 prescriber.co.uk ■ THERAPY FOCUS l SGLT2 inhibitors

It is usually said that most genital infections occur during 3. Health and Social Care Information Centre. Prescribing for diabetes, early treatment but, as experience with canagliflozin shows, first England 2005/06 to 2015/16. August 2016. http://content.digital. episodes continue to be reported for many months (see Figure nhs.uk/catalogue/PUB21158 2).20 Genital infections are described as mild to moderate in most 4. Health and Social Care Information Centre. Prescription cost analysis, cases, according to investigator assessment. The incidence of England 2015. April 2016. https://www.gov.uk/government/statistics/ prescription-cost-analysis-england-2015 severe events or associated treatment discontinuation was 0.7– 5. Zinman B, et al. Empagliflozin, cardiovascular outcomes, and mortal- 1.0 per cent among women and 0.5 per cent among men, though ity in type 2 Diabetes. N Engl J Med 2015;373:2117–28. these figures are based on low sample sizes. The risk of genital 6. Shyangdan DS, Uthman OA, Waugh N. SGLT-2 receptor inhibitors for infection is increased in women with previous episodes and in treating patients with type 2 diabetes mellitus: a systematic review and uncircumcised men.20 About 2 per cent of all women experienced network meta-analysis. BMJ Open 2016;6:e009417. doi:10.1136/ recurrent infection with canagliflozin treatment compared with 0.3 bmjopen-2015-009417. per cent with placebo; of those who experienced an infection while 7. European Medicines Agency. Jardiance (empagliflozin) public assess- taking canagliflozin, 18 per cent had two episodes and 3 per cent ment report. March 2014. had three or more.20 In women, symptoms lasted for six to seven 8. European Medicines Agency. Forxiga (dapagliflozin) public assess- days after treatment for the infection was initiated. The majority of ment report. September 2012. 9. European Medicines Agency. Canagliflozin public assessment report. infections were treated with standard therapy, including self-treat- September 2013. ment, while continuing to take an SGLT2 inhibitor. This involved 10. Hirji I, et al. Incidence of urinary tract infection among patients with topical therapy in about a third, oral therapy in 40 per cent and oral type 2 diabetes in the UK General Practice Research Database (GPRD). 20 and topical therapy in a fifth of patients. J Diabetes Complications 2012;26:513–6. 11. Hirji I, et al. Incidence of genital infection among patients with type Adherence 2 diabetes in the UK General Practice Research Database. J Diabetes Whether these adverse effects reduce adherence to SGLT2 Complications 2012;26:501–5. inhibitors has not been reported, though it is hard to see how 12. Geerlings S, et al. Genital and urinary tract infections in diabetes: it would help. Adverse effects to antidiabetic medication do impact of pharmacologically-induced glucosuria. Diabetes Res Clin Pract reduce adherence, though not to the same extent as simply for- 2014;103:373–81. 13. Nyirjesy P, et al. Evaluation of vulvovaginal symptoms and Candida getting a dose. In one US study, 17 per cent of patients missed colonization in women with type 2 diabetes mellitus treated with cana- doses due to adverse effects compared with 55 per cent who gliflozin, a sodium glucose co-transporter 2 inhibitor. Curr Med Res Opin 21 just forgot. In EMPA-REG OUTCOME, more than a quarter of 2012;28:1173–8. the 7000 participants discontinued medication prematurely but 14. Bermingham SL, Ashe JF. Systematic review of the impact of urinary this was a bigger problem with placebo than empagliflozin (29 tract infections on health-related quality of life. BJU Int 2012;110(11, pt vs 23 per cent).5 Adverse events accounted for discontinuation C):E830–6. doi: 10.1111/j.1464-410X.2012.11337.x. in 13 per cent of people assigned to placebo and 11.5 per cent 15. Renard J, et al. Recurrent lower urinary tract infections have a det- of those treated with empagliflozin. rimental effect on patient quality of life: a prospective, observational study. Infect Dis Ther 2015;4(1):125–35. Conclusion 16. Zhu YX, et al. Health-related quality of life as measured with the Short-Form 36 (SF-36) questionnaire in patients with recurrent vulvovag- The SGLT2 inhibitors are becoming established in the treatment inal candidiasis. Health Qual Life Outcomes 2016;14:65. of type 2 diabetes and, if trial evidence of a reduction in mor- 17. Aballéa S, et al. Subjective health status and health-related quality tality is realised in practice, they will play an increasingly impor- of life among women with recurrent vulvovaginal candidosis (RVVC) in tant role. People with diabetes are at increased risk of UTI and Europe and the USA. Health Qual Life Outcomes 2013;11:169. doi: genital infections and the risk is further increased by treatment 10.1186/1477-7525-11-169. with an SGLT2 inhibitor. Women, younger people and those with 18. Usiskin K, et al. Safety and tolerability of canagliflozin in patients a history of infections are at higher risk. An individual’s history with Type 2 diabetes mellitus: pooled analysis of Phase 3 study results. of genitourinary infection is therefore clearly something that will Postgrad Med 2014;126:16–34. influence the decision to choose an SGLT2 inhibitor for intensifi- 19. Vasilakou D, et al. Sodium-glucose cotransporter 2 inhibitors for cation. Published reports describe most genitourinary events as type 2 diabetes: a systematic review and meta-analysis. Ann Intern Med 2013;159:262–74. mild to moderate in intensity, manageable with standard topical 20. Nyirjesy P, et al. Genital mycotic infections with canagliflozin, a sodium or oral therapy, and rarely causing treatment discontinuation. glucose cotransporter 2 inhibitor, in patients with Type 2 diabetes mellitus: However, UTIs and genital infections significantly impair quality a pooled analysis of clinical studies. Curr Med Res Opin 2014;30:1109–19. of life, at least when they are recurrent, and serious events were 21. Larkin AT, et al. Determinants of adherence to diabetes treatment. reported rarely in clinical trials. J Diabetes 2015;7:864–71.

References Declaration of interests 1. NICE. Type 2 diabetes in adults: management. NG28. December None to declare. 2015. www.nice.org.uk/guidance/ng28 2. NICE. Canagliflozin, dapagliflozin and empagliflozin as monothera- pies for treating type 2 diabetes. TA390. May 2016. www.nice.org.uk/ Steve Chaplin is a pharmacist who specialises in writing on guidance/ta390 therapeutics

30 ❚ Prescriber December 2016 prescriber.co.uk