Part 2:Characterization And Implications Of The Initial Estimated Glomerular Filtration Rate ‘dip’ Upon Sodium-glucose Cotransporter-2 Inhibition With Empagliflozin in The EMPA-REG OUTCOME Trial

Contact: Audrey Hu Whatsapp/hp: 0086 13880143964 Email: audrey.hu@wecistanche.com

Pls click here to Part 1

Based on a cutoff of P < 0.1 for treatment interaction, UACR categories, and angiotensin-converting enzyme inhibitor (ACEi)/angiotensin receptor blocker (ARB) use at baseline were included in the subsequent multivariate analyses (Figure 2a). Other baseline factors, such as age, HbA1c, hematocrit, and blood pressure (BP), were not associated with higher or lower odds for an initial ‘eGFR dip’ with empagliflozin versus placebo (interaction P > 0.1) (Supplementary Figure S3).

In multivariate logistic regression with backward selection (using P < 0.05 for retention of the interaction in the model), only diuretic treatment and KDIGO risk category at baseline were identified as independent predictors of an initial ‘eGFR dip’ with empagliflozin. Figure 2b presents ORs for initial ‘eGFR dip’ with empagliflozin overall (left) and across the 8 subgroups for combinations of diuretic use and KDIGO risk at baseline (right). In participants with low KDIGO risk and not receiving diuretic treatment (n ¼ 1993), empagliflozin was associated with a relatively low OR of 1.6 (95% CI, 1.2–2.1) for an initial ‘eGFR dip’; however, OR increased to 2.7 (95% CI, 1.9–3.7) in low KDIGO risk

category participants receiving diuretic therapy at baseline (n ¼ 1182). The increasingly severe KDIGO risk category was further associated with an increased risk of an initial ‘eGFR dip’ (Figure 2b, right).AE profile across subgroups by predictive baseline factors for an empagliflozin-induced ‘eGFR dip’

Cistanche can help with kidney disease

To assess whether the increased risk of an empagliflozin- induced ‘eGFR dip’ was associated with increased risk of AEs, we investigated overall AEs and kidney AEs in participant subgroups based on predictive baseline factors. These AE analyses were based on reporting by study investigators. Kidney AEs refer to reporting of the narrow Standardised MedDRA Query ARF, which includes AKI. As shown in Figure 3, within both treatments, participants on diuretics at baseline had higher rates of serious (Figure 3a) and kidney (Figure 3b) AEs compared with participants not treated with diuretics. These rates were further increased in higher KDIGO risk categories, especially in high and very high KDIGO risk. Regardless of diuretic treatment and KDIGO risk category, AEs were generally lower or similar with empagliflozin versus placebo (Figure 3 and Supplementary Table S2).

From baseline to week 4, overall and serious AE rates were not increased with empagliflozin in any subgroup (Supplementary Table S3). Kidney AEs were slightly elevated in empagliflozin-treated participants (48 of 4635 [0.1%– 4.9%] with empagliflozin vs. 16 of 2317 [0.5%–3.5%] with placebo across KDIGO risk and diuretic use groups) (Supplementary Table S3). In addition, kidney AEs leading to treatment discontinuation were reported more frequently with empagliflozin, especially in the participants not treated with diuretics at baseline (Supplementary Table S3). However, this affected only 9 patients with kidney events out of a total of 84 AEs leading to discontinuation of empagliflozin treatment, all of which reported the preferred term ‘renal impairment.’

Figure 2| Odds ratios (ORs) for an ‘estimated glomerular filtration rate (eGFR) dip’ with empagliflozin versus placebo: factors included in the multivariate prediction analysis (a) and multivariate analysis yielding 8 subgroups of combinations of diuretic use and Kidney Disease: Improving Global Outcomes (KDIGO) risk (b). (a) Participants treated with at least 1 dose of study drug who had eGFR (Chronic Kidney Disease Epidemiology Collaboration [CKD-EPI]) values available for baseline and week 4. OR (95% confidence interval [CI]) for an ‘eGFR dip’ for empagliflozin versus placebo. Logistic regression with baseline factor, treatment, and interaction of baseline factor with treatment showing baseline factors with P < 0.1 for interaction. (b) Participants treated with at least 1 dose of study drug who had eGFR values available for baseline and week 4. OR (95% CI) for an ‘eGFR dip’ for empagliflozin versus placebo. (Left) Logistic regression including treatment, sex, baseline body mass index category, baseline glycated hemoglobin category, baseline eGFR category, geographical region, and age. (Right) Number of patients per subgroup (diuretic/KDIGO risk category): no/low, 1993; no/moderate, 1063; no/high, 511; no/very high, 215; yes/low, 1182; yes/moderate, 855; yes/high, 490; yes/very high, 295. Following a backward selection procedure, the multivariate logistic regression model included factors for use of diuretics and KDIGO risk category at baseline (P ¼ 0.1542), treatment (P ¼ 0.0006), use of angiotensin-converting enzyme inhibitors/angiotensin II receptor blockers at baseline (P < 0.0001), and treatment-by-use of diuretics and KDIGO risk category at baseline interaction (P ¼ 0.0006) as categorical variables. *Prognosis of chronic kidney disease (CKD) according to 2012 KDIGO guidelines. ACEi, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; UACR, urine albumin-to-creatinine ratio.

Risk for CV and kidney outcomes with empagliflozin across subgroups by predictive baseline factors

For further analyses of CV and kidney outcomes, the 8 subgroups of predictive baseline factors were pooled into 2 groups based on their OR above versus below or equal to the overall OR for an empagliflozin-induced ‘eGFR dip.’ Thereby, participants with no diuretic use and any KDIGO risk and those with diuretic use and low KDIGO

Figure 3| Adverse events (AEs) by a subgroup of baseline predictive factors for an initial ‘estimated glomerular filtration rate dip’: serious (a) and kidney (b) AEs. Analysis based on participants treated with at least 1 dose of study drug using events obtained on treatment þ 7 days. Kidney AE: narrow Standardised Medical Dictionary for Regulatory Activities Query acute renal failure reported by study investigators, which included the preferred term acute kidney injury. KDIGO, Kidney Disease: Improving Global Outcomes.

risk at baseline was pooled into the category ‘OR # 2.7,’ while participants with baseline diuretic use and moderate-to-high KDIGO risk were pooled into the category ‘OR > 2.7.’

CV death, HHF, and incident or worsening nephropathy (Figure 4a), as well as additional CV and kidney outcomes (Supplementary Figure S4) were consistently reduced by empagliflozin versus placebo in these 2 pooled subgroups (all P values for interaction >0.1) (Figure 4a and Supplementary Figure S4). Consistent results were retrieved when all 8 sub-groups were investigated individually (data not shown).

Event rates for all outcomes after week 4 were higher in participants with an increased OR for an ‘eGFR dip’ compared with an OR below or equal to the overall OR (Figure 4a and SupplementaryFigureS4)in both treatment groups. Therefore, we analyzed the association of an ‘eGFRdip’withoutcomesand found a tendency toward a slightly increased risk of CV-related death (HR, 1.24; 95% CI, 0.95– 1.62) and HHF (HR, 1.18; 95% CI, 0.86– 1.63) and a significantly increased risk for incident or worsening nephropathy (HR, 1.22; 95% CI, 1.05– 1.44) in participants with an ‘eGFRdip,’ consistently withempagliflozin and placebo treatment(allPvaluesfortreatment-by-‘eGFRdip’ interaction around 0.5–0.8).

Impact of the ‘eGFR dip’ on the treatment effect of empagliflozin on CV and kidney outcomes: a mediation analysis

We then assessed whether the treatment effect of empagliflozin on these outcomes was affected by the ‘eGFR dip.’Figure 4b presents the HRs for CV death, HHF, and incident or worsening nephropathy following week 4 from the primary analysis, along with HRs after additional adjustment for the ‘eGFR dip,’ along with the resulting percent media- tion of the treatment effect attributable to the ‘eGFR dip.’ The 95% CIs of the treatment effect from the primary analysis and the adjusted analysis overlapped, and the ‘eGFR dip’ resulted in – 14.7%, – 11.7%, and – 10.2% of the empa- gliflozin treatment effect being mediated for CV death, HHF, and incident or worsening nephropathy, respectively.

Effects of cistanche: improve kidney function

DISCUSSION

An initial eGFR decline of approximately –3 to –5 ml/min per 1.73 m2 on treatment has been reported across SGLT2i out- comes trials,2,4,6,8 but its categorization and potential impli- cations for safety and efficacy have not yet been explored. In EMPA-REG OUTCOME, the proportion of participants with an initial ‘eGFR dip’ >10% was doubled with empagliflozin versus placebo, but a more pronounced eGFR decline >30% was rare in both arms. Nevertheless, the wide variability of the initial eGFR changes after treatment initiation was present in both arms, reflecting the biological variability of eGFR previously reported in healthy and diseased cohorts, including diabetic patients.25-27 In the participants randomized to empagliflozin, the mean eGFR over time remained stable after week 12 in all dipping categories, even in participants with more pronounced initial eGFR decline (>30%). Stabilization of eGFR regardless of the degree of acute changes after initiation of another SGLT2i, luseogliflozin was recently reported in Japanese patients with T2D.28

Figure 4| Cardiovascular (CV) and kidney outcomes by baseline predictive factors for an initial ‘estimated glomerular filtration rate (eGFR) dip’ and proportion of the risk reduction in outcomes following week 4 mediated by an ‘eGFR dip.’ (a) Participants treated with $1 dose of study drug who had eGFR (Chronic Kidney Disease Epidemiology Collaboration) values available at baseline and at week 4. Risk reduction overall (all patients) and by baseline diuretic use/Kidney Disease: Improving Global Outcomes (KDIGO) category subgroups categorized according to their risk below or equal to and above the overall effect for an initial ‘eGFR dip’ with empagliflozin versus placebo: odds ratio (OR) #2.7 versus >2.7. OR for an ‘eGFR dip’ in empagliflozin versus placebo in the overall population: 2.7 (95% confidence interval, 2.3–3.0). Overall dipping OR lower (#2.7): participants with no diuretic use at baseline and any KDIGO risk or diuretic use at baseline and low KDIGO risk. Increased dipping OR (>2.7): participants with baseline diuretic use and KDIGO risk moderate to high. Results are based on Cox regression with factors for treatment, age, sex, baseline body mass index (BMI), baseline glycated hemoglobin (HbA1c), region, subgroup, and subgroup-by-treatment interaction. (b) Percentage of empagliflozin treatment effect mediated by an ‘eGFR dip’ resulting from a comparison of the treatment effect from the primary model with the treatment effect from the model also adjusted for ‘eGFR dip’ based on landmark analyses after week 4. A Cox proportional hazards model was used, with adjustment for the treatment group, age, sex, baseline BMI, baseline HbA1c, baseline eGFR, and region in participants treated with $1 dose of study drug who had eGFR values available at both baseline and week 4. HHF, hospitalization for heart failure; HR, hazard ratio.

An initial decline in eGFR after treatment initiation followed by stabilization of kidney function during the chronic maintenance therapy is highly reminiscent of the eGFR responses observed with RAAS inhibitors in several trials.14,16,29-33 In such trials, initial ‘eGFR dipping’ versus ‘eGFR non-dipping,’ mostly within the first 3–6 months, was inversely related to the course of eGFR over the subsequent 3– 4 years. Patients with an initial eGFR increase were generally reported to have steeper eGFR slopes during chronic maintenance therapy compared with those with an initial eGFR decrease irrespective of the use of an ARB,16,32 ACEi, or other BP-lowering agents, such as direct renin inhibitors or beta-blockers.29,33 A meta-analysis of trials in patients with preexisting kidney impairment suggested that a beneficial relationship holds especially for serum creatinine increases of #30% and for patients with a starting creatinine value >1.4 mg/dl,14 but this was not confirmed in all trials.32

In EMPA-REG OUTCOME, the majority of participants received randomized treatment in addition to preexisting RAAS inhibition. With empagliflozin, stabilization of long-term mean eGFR occurred in all dipping categories with differing baseline kidney function, including the subset with a >30% eGFR decline. With RAAS inhibitors, the extent of an initial eGFR decline is thought to indicate treatment response,14,29 but in our analysis, we found a comparable stabilization of long-term mean eGFR in all ‘eGFR dipping’ categories on empagliflozin treatment. Recently reported data on treatment with canagliflozin or dapagliflozin to show that diabetic patients with chronic kidney disease stage 3b–4.34,35 Future studies of empagliflozin in patients with impaired kidney function, such as EMPEROR (NCT03057977, NCT03057951)36,37 and EMPA-KIDNEY (NCT03594110),38 may reveal whether this can also be confirmed with empagliflozin.

herb cistanche

In the current analysis, eGFR increased following empagliflozin treatment cessation in all ‘eGFR dipping’ categories. This long-term treatment.39 Interestingly, in our study, median eGFR values at 1 month after empagliflozin cessation increased even in the ‘eGFR non-dipper’ group to a similar extent as in the ‘eGFR intermediate’ group. Such an improvement in eGFR in participants who did not exhibit an initial ‘eGFR dip’ may raise questions regarding the pathophysiology of the initial ‘eGFR dip’ with empagliflozin. Although the restoration of tubuloglomerular feedback is likely an important kidney contributor to the mechanism of action of SGLT2i,40,41 additional kidney mechanisms, such as tubular protection, reduced hypoxia and inflammation, or long-term effects of natriuresis, also may contribute to the renoprotective effects observed with SGLT2i.42-45 Loss of randomization and apparent differences in baseline characteristics between ‘eGFR dippers’ on empagli-flozin versus those on placebo did not allow for a direct comparison of AEs and outcomes within and across ‘eGFR dipping’ categories. Therefore, we focused on identifying baseline factors that could be predictive of an initial ‘eGFR dip’ after initiation of empagliflozin versus placebo. Participants on diuretic therapy, particularly those on loop diuretics and thiazides, were more likely to experience such an initial ‘eGFR dip,’ which was additionally and independently increased by higher KDIGO risk category. ACEi/ARB treatment at baseline also increased the OR for an ‘eGFR dip’ with empagliflozin versus placebo, but the interaction with empagliflozin treatment was statistically nonsignificant, especially in the multivariate analysis. Thus, ACEi/ARB use at baseline was higher in participants with an initial ‘eGFR dip’ compared with ‘eGFR non-dippers’ on empagliflozin but was not predictive of an ‘eGFR dip’ with empagliflozin treatment. Also, hemodynamic or volume markers, such as baseline BP, hematocrit, and hemoglobin, were not associated with an empagliflozin-induced ‘eGFR dip.’

Empagliflozin did not increase the rate of AEs regardless of kidney risk or diuretic therapy at baseline. Most importantly, reporting of kidney AEs, including AKI, did not raise safety concerns in any of the subgroups. We found numerically elevated proportions of kidney AEs only for the initial treatment phase up to week 4, and few events leading to discontinuation of empagliflozin treatment, especially in participants not on diuretics at baseline. However, all referred to the preferred term ‘renal impairment,’ suggesting that these AEs may have reflected the initial empagliflozin-induced ‘eGFR dip.’ These safety data add to the reassuring findings on RAAS inhibition, with an eGFR decline of up to 20% after treatment initiation considered acceptably safe.46

CV and kidney outcomes were consistently reduced with empagliflozin across the subgroups based on predictive factors for an initial ‘eGFR dip.’ Furthermore, empagliflozin treatment was associated with improved CV mortality, HHF, and kidney outcomes after week 4 in the mediation analysis, which was not substantially weakened by adjustment for an initial ‘eGFR dip,’ as shown by the similar HRs with largely overlapping CIs. These data seem to contrast with previously reported data for RAAS inhibition, in which an initial ‘eGFR dip’ was suggested to be a positive prognostic marker associated with beneficial kidney outcomes and reduced declines in kidney function in patients with and without diabetes and in patients with and without impaired kidney function.14,16,29,31,47-50 However, more recent analyses showed that an ‘eGFR dip’/creatinine increase with RAAS inhibition might not be predictive of CV and kidney outcomes or associated with an increased risk.17-19,27,51,52

Our data are in line with findings from previous analyses on EMPA-REG OUTCOME, showing that outcome event rates were consistently reduced with empagliflozin in sub- groups across various conditions,2,53-59 such as participants using diuretic background medication at baseline,53 or across KDIGO risk categories.60 Our findings are also in line with results from the CV mediation analysis showing that eGFR over time had no or only negligible effects on the empagliflozin treatment effect on CV mortality, which was mainly mediated by changes in markers of plasma volume.23

Effects of cistanche: treating kidney diseases

Limitations

Our study has several limitations. First, this was a post hoc analysis of a CVoutcomes trial that was not prespecified and of an exploratory nature. In addition, correction for multiple testing was not applied. Only single measurements of eGFR were available, and owing to their high biological variability, we might not have been able to correctly classify each patient. Also, this analysis did not include a randomized placebo-corrected comparison of ‘eGFR dippers’ versus ‘eGFR non-dippers,’ because dipping categories were defined post-randomization and had significant differences in baseline characteristics, and the ‘eGFR dip’ itself was already influenced by empagliflozin treatment. For a direct comparison, a dedicated trial that randomized partici- pants after stratification for individual ‘eGFR dipping’ responses would be required. However, by comparing the subgroups based on predictive baseline factors, we did not break treatment randomization and were able to assess an empagliflozin treat- ment effect, and we shed some light on the impact of an ‘eGFR dip’ on outcomes occurring after week 4 in the mediation analysis.

Conclusions

An initial ‘eGFR dip’ >10% affected approximately 1 in 4 of the study participants treated with empagliflozin in EMPA-REG OUTCOME; a more pronounced initial eGFR decline >30% was rare. Participants with more advanced kidney disease and/ or on diuretic therapy at baseline were more likely to experience an initial ‘eGFR dip’ >10%. However, empagliflozin treatment appears to be safe and associated with improved CVand kidney outcomes, irrespective of identified baseline predictive factors. In addition, the initial ‘eGFR dip’ did not have a major impact on the long-term CV and kidney benefits observed with empagliflozin in patients with T2D and CV disease.

DISCLOSURE

BJK has received grant support from the IZKF Wuerzburg (Interdisziplinäres Zentrum für klinische Forschung, project ZZ-13) and honoraria from Boehringer Ingelheim. IR, MM, and AKW are employees of Boehringer Ingelheim, the manufacturer of empagliflozin. MvE was an employee of Boehringer Ingelheim at the time of the study, and is now an employee of Nestlé Health Science, Epalinges, Switzerland. CW reports honoraria from AstraZeneca, Bayer, Boehringer Ingelheim, Eli Lilly, Mitsubishi, and MSD. MW reports serving as a scientific advisor for Bayer, Relypsa, Janssen, Boehringer Ingelheim, AstraZeneca, MSD, Sanofi, Vifor, Akebia, and Boston Scientific and holding the following National Institutes of Health grants: U01 DK116095, R01 HL132372, R01 HL127422, R01 DK066013, U01 DK1061022. GLB reports serving as a member of scientific advisory boards or as a consultant for Merck, Janssen, Novo Nordisk, AstraZeneca, Boehringer Ingelheim, Bayer, Relypsa, and Reata and as a steering committee member for Janssen and Bayer (principal investigator, kidney outcome trial). DZIC has received honoraria from Boehringer Ingelheim-Lilly, Merck, AstraZeneca, Sanofi, Mitsubishi-Tanabe, Abbvie, Janssen, Bayer, Prometic, BMS, and Novo Nordisk and has received operational funding for clinical trials from Boehringer Ingelheim-Lilly, Merck, Janssen, Sanofi, AstraZeneca, and Novo Nordisk. NS has received honoraria from Amgen, AstraZeneca, Boehringer Ingelheim, Eli Lilly, Novo Nordisk, Pfizer, and Sanofi and grant funding from Boehringer Ingelheim. HLH has consultancy agreements with AbbVie, Astellas, AstraZeneca, Boehringer Ingelheim, Janssen, Fresenius, and Merck, and all honoraria goes to his employer. SEI reports honoraria from Intarcia Therapeutics, Daiichi-Sankyo, Lexicon Pharmaceuticals, Janssen, Sanofi, AstraZeneca, Boehringer Ingelheim, and Novo Nordisk. BZ has received grant support from Boehringer Ingelheim, AstraZeneca, and Novo Nordisk and consulting fees from AstraZeneca, Boehringer Ingelheim, Eli Lilly, Janssen, Merck, Novo Nordisk, and Sanofi Aventis.

DATA STATEMENT

The sponsor of the EMPA-REG OUTCOME trial (Boehringer Ingelheim) is committed to responsible sharing of clinical study reports, related clinical documents, and patient-level clinical study data. Researchers are invited to submit inquiries via the following website: https://trials. Boehringer-ingelheim.com.

ACKNOWLEDGMENTS

This study was funded by the Boehringer Ingelheim & Eli Lilly and Company Diabetes Alliance. Assistance with the preparation of the tables and figures, supported financially by Boehringer Ingelheim, was provided by Matthew Smith, Sally Neath, and Paul Hitchcock of Elevate Scientific Solutions. The authors are fully responsible for all content, were involved at all stages of manuscript development, and have approved the final version of the manuscript.

SUPPLEMENTARY MATERIAL

Supplementary File (PDF)

Figure S1. eGFR over time for the subcategory of an initial eGFR decline >30% from baseline in empagliflozin-treated participants. Figure S2. (A) eGFR over time by ‘eGFR dipping’ category in placebo-treated participants. (B) Mean eGFR at last value on treatment and follow-up per ‘eGFR dipping’ categories in placebo-treated participants.

Figure S3. (A) ORs for an ‘eGFR dip’ with empagliflozin versus placebo: analysis of diuretic subgroups. (B) Baseline factors not associated with an ‘eGFR dip’ with empagliflozin versus placebo. Figure S4. Additional cardiovascular and kidney outcomes by baseline predictive factors for an initial ‘eGFR dip.’

Table S1. Baseline characteristics for empagliflozin-treated partici- pants including a subcategory of an initial eGFR decline >30%. Table S2. Adverse events of special interest by a subgroup of predictive baseline factors for an ‘eGFR dip.’

Table S3. Adverse events of special interest by a subgroup of predictive baseline factors for an ‘eGFR dip,’ baseline to week 4.

herb cistanche

REFERENCES

1. Zinman B, Wanner C, Lachin JM, et al. Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes.N Engl J Med.2015;373:2117-2128.

2. Wanner C, Inzucchi SE, Lachin JM, et al. Empagliflozin and progression of kidney disease in type 2 diabetes.N Engl J Med.2016;375:323-334.

3. Neal B,Perkovic V, Mahaffey KW, et al. Canagliflozin and cardiovascular and renal events in type 2 diabetes. N Engl J Med.2017;377:644-657.

4. Wiviott SD, Raz l, Bonaca MP, et al. Dapagliflozin and cardiovascular outcomes in type 2 diabetes. N Engl J Med.2019;380:347-357.

5. Mosenzon O, Wiviott SD, Cahn A, et al. Effects of dapagliflozin on development and progression of kidney disease in patients with type 2 diabetes:an analysis from the DECLARE-TIMI 58 randomised trial.Lancet Diabetes Endocrinol. 2019;7:606-617.

6. Perkovic V, Jardine MJ, Neal B, et al. Canagliflozin and renal outcomes in type 2 diabetes and nephropathy. N Engl J Med.2019;380:2295-2306.

7. Heerspink HJ, Johnsson E, Gause-Nilsson I, et al. Dapagliflozin reduces albuminuria in patients with diabetes and hypertension receiving renin-angiotensin blockers. Diabetes Obes Metab.2016;18:590-597.

8. Heerspink HJ, Desai M, Jardine M, et al. Canagliflozin slows progression of renal function decline independently of glycemic effects. J Am Soc Nephrol. 2017;28:368-375.

9. U.S. Food and Drug Administration. FDA drug safety communication: FDA strengthens kidney warnings for diabetes medicines canagliflozin (Invokana, Invokamet) and dapagliflozin (Farxiga, Xigduo XR). [06-14-2016]. Available at: https://wwwfda.gov/drugs/drug-safety-and-availability/fda-drug-safety-communication-fda-strengthens-kidney-wamings-diabetes-medicines-canagliflozin. Accessed June 9, 2020.

10. Nadkarni GN, Ferrandino R, Chang A,et al. Acute kidney injury in patients on SGLT2 inhibitors: a propensity-matched analysis. Diabetes Care. 2017;40:1479-1485.