1/ sGLT2i educational symposia #KidneyWk #Tweetorial. Trying this out for @AJKDonline rather than a blog post. This is a live tweet thread of the info from this great session.
@AJKDonline 2/ Before we start, if you haven’t already done so, check out the amazing #Tweetorial by @aishaikh on this topic:
@AJKDonline @aishaikh 3/ Initial studies of this mechanism of action was on phlorizin isolated from apple trees, further research identified SGLT carrier proteins.
@AJKDonline @aishaikh 4/ Glucose is a key energy source and cannot traverse lipid cell membrane by diffusion. SGLT2s are 60-80kDa proteins with 580-718 amino acids. The terminal ends are both extracellular.
@AJKDonline @aishaikh 5/ Facilitated diffusion glucose transporters (GLUTs) are slightly different, and have their terminal ends intracellularly, these transporters are not emphasized in this session.
@AJKDonline @aishaikh 6/ SGLTs are symporters of Na and glucose. At least 13 identified in various organs - brain, kidney, intestine, lung, etc. SGLT2s are in S1 and S2 cells in proximal tubules. They are low affinity/high capacity transporters that resorb bulk of filtered glucose.
@AJKDonline @aishaikh 7/ Genetic defects in the SGLT2 gene are associated with high concentrations of glucose in urine. Can cause Fanconi-Bickel syndrome with fasting hypoglycemia, postprandial hyperglycemia, proximal tubule dysfunction, and dwarfism.
@AJKDonline @aishaikh 8/ Dr. Cherney elegantly summarizes the effects of the class of these drugs into two categories:
-Hemodynamic effects
-Nonhemodynamic effects (eg hypoxia, inflammation, etc)
-Hemodynamic effects
-Nonhemodynamic effects (eg hypoxia, inflammation, etc)
@AJKDonline @aishaikh 9/ How do these work? They block reabsorption of sodium and glucose, resultant glucosuric effect. Lowers A1C by about 0.7% and induces natriuresis and decreases body weight by 3 kg. 
@AJKDonline @aishaikh 10/ Hemodynamic hypothesis is based on tubuloglomerular feedback. In states of reduced renal perfusion, there is less Na delivered to macula densa. This can be due to hypotension, sepsis, vol depletion, etc.
@AJKDonline @aishaikh 11/ With reduce Na delivery to macula densa, this leads to decreased adenosine generation. Adenosine is the stimulus for vasoconstriction, so with less of this, there is afferent vasodilation which should improve GFR in low-flow states.
@AJKDonline @aishaikh 12/ Let’s consider hyperglycemia in poorly controlled DM. With high glucose concentrations in proximal tubule, there is increased Na-Glu reabsorption by SGLT2. This also leads to reduced Na delivery to macula densa.
@AJKDonline @aishaikh 13/ Although the low Na delivery in hyperglycemia is NOT due to renal hypoperfusion, the macula densa doesn’t know the difference, and responds with afferent vasodilation, increased intraglomerular pressure, and hyperfiltration.
@AJKDonline @aishaikh 14/ This is where the drug comes in. SGLT2-i block this hyperreabsorptive state, leads to more sodium delivery to macula densa, resulting in afferent vasoconstriction and reduces intraglomerular pressure via its effect on TG feedback.
@AJKDonline @aishaikh 15/ This effect was demonstrated in animal models in vivo, showing reduction in afferent arteriole diameter after empaglifozin, supporting the TG feedback hypothesis.
@AJKDonline @aishaikh 16/ These results were reproduced in humans with T1DM based with hyperfiltration with inulin clearances. GFR decreased from 170 to 130 ml/min with SGLT2-i.
@AJKDonline @aishaikh 17/ This GFR drop occurs RAPIDLY – within 2 hours in animal models. In humans, after a single dose of SGLT2i, eGFR dropped by 5-7 ml/min, highlighting that this is an acute hemodynamic effect.
@AJKDonline @aishaikh 18/ This is not unlike RAAS inhibition, except those work hemodynamically on the efferent arteriole. This makes the trial data important, as SGLT2i was added to RAAS-i.
@AJKDonline @aishaikh 19/ What are the clinical outcomes? In EMPA-REG – acute dip in GFR, followed by stabilization of GFR over time, and preservation compared to placebo.
@AJKDonline @aishaikh 20/ When the drug is stopped, the effect on GFR is completely reversible after a 30-day washout.
@AJKDonline @aishaikh 21/ Shifting to non-hemodynamic effect. SGLT2-I reduce IL-6, an inflammatory marker, which is dependent on glucose. First demonstrated in animals, reproduced in humans with dapa.
@AJKDonline @aishaikh 22/ Too much to go over, but there are lots of different pathways in which SGLT2i may attenuate progression of CKD. May also explain cardioprotection as well.
@AJKDonline @aishaikh 23/ The evidence that this SGLT2i benefit in kidney are independent of glucose lowering. Mostly based on urine albumin:cr ratio data.
@AJKDonline @aishaikh 24/ Reduction of UACR was independent of HbA1C, weight, and SBP. A1C lowering essentially the same compared to other antidiabetic meds like glimiperide, but did not see the same reduction in UACR.
@AJKDonline @aishaikh 25/ Even in T1DM there are benefits, evaluated with sotaglifozin. Had the same GFR dip with reduced UACR by 20-30%, and improved BP. Outlined in image below.
@AJKDonline @aishaikh 26/ Dr. Heerspink from the Netherlands kicks off the next portion focusing on trials. Shout-out to IDNT, but points out that despite renoprotection, there is still a very high residual risk of primary outcome.
@AJKDonline @aishaikh 27/ Again, reemphasizing the reversibility of the hemodynamic effect, akin to discontinuing ACEi or ARB, highlighting the hemodynamic effects.
@AJKDonline @aishaikh 28/ 3 big CV trials - EMPA-REG, CANVAS, DECLARE demonstrated stabilization of eGFR even though they were normal GFR and not looking at kidney outcomes.
@AJKDonline @aishaikh 29/ Important point. The GFR dip was not associated with AKI biomarkers – KIM1, NGAL, LFABP, reflecting that the dip does not reflect organ damage. If anything, the studies above have noted a reduction in AKI if any effect is there.
@AJKDonline @aishaikh 30/ CREDENCE trial population had lower baseline GFR compared to earlier SGLT2-I trials. Some of these patients had eGFR > 30 ml/min at time of randomization.
@AJKDonline @aishaikh 31/ CREDENCE notes similar baseline risk compared to IDNT, patients were on RAASi. Adding SGLT2i further reduced rate of GFR decline by 60%.
@AJKDonline @aishaikh 33/ What about GFR < 30 ml/min in CREDENCE? High risk cohort, relatively small, but similar improvements noted.
@AJKDonline @aishaikh 34/ However, notes that there is no drop in GFR. Possibly because of reduced nephron mass, but more likely reflecting regression to the mean, because a high risk cohort was selected at the start.
@AJKDonline @aishaikh 35/ AKI safety? Data from CREDENCE shows that despite the hemodynamic effects, the risk of AKI were unchanged or lower in the long-term.
@AJKDonline @aishaikh 36/ Because of hemodynamic benefits, there is theoretical use of this in other types of nondiabetic kidney disease. Has even been trialed in small FSGS pilots.
@AJKDonline @aishaikh 37/ DAPA-HF data from late breaking yesterday included diabetics and non-diabetics with CHF. There was 29% RRR in dapa group, but low overall incidence (1.2 vs 1.6%). Secondary renal outcomes by diabetic status showed benefit in both.
@AJKDonline @aishaikh 38/ The DIAMOND trial is looking at dapa in nondiabetic kidney disease. 1 gram + proteinuria, included pts with IgAN, 2 FSGS, HTN, etc. Study closed, but results pending
@AJKDonline @aishaikh 39/ DAPA-CKD and EMPA-KIDNEY also ongoing. Anticipating results by 2021, 4000 patients plus in these data.
@AJKDonline @aishaikh 40/ These trials are important because they are clinically meaningful and translate to routine practice.
@AJKDonline @aishaikh 41/ Closing argument in meta-analysis. Note the slope of GFR decline in the year before. Initiation of SGLT2i resets this and attenuates the slope. Thx for reading to the end!
