diabetes: GLP1 agonists vs SGLT2 antagonists

i have written many blogs on the differences between the diabetes meds GLP1s and SGLT2s, but i think it might be useful to have a summary review that weaves in new info.

GLP1 receptor agonists:

-- these restore a normal physiologic response to food, the "incretin effect", that is aberrant in diabetics

    --relevant human incretins are glucagon-like peptide-1 (GLP1) and glucose-dependent insulinotropic polypeptide (GIP)

    --in diabetics GLP1 is reduced by 30-40%

    --deficiency of GLP1 exacerbates insulin deficiency, glucagon excess, and also reduces CNS control of satiety/gastic emptying

--advantages of GLP1s  include:

    --they can lead to very dramatic reductions in A1c levels [i have had several patients with A1c's in the 10-14% range who are largely non-adherence to diabetic diets and are being treated with insulin and metformin, but adding GLP1 agonists have had their A1cs plummet to the 6-7% range!!]

    --they are cardioprotective: in high-risk patients, there was a 13% decrease in the composite of death from cardiovascular causes, nonfatal MI and nonfatal stroke, with significant decreases in death from cardiovascular causes and death from all causes (see http://gmodestmedblogs.blogspot.com/2016/06/liraglutide-decreases-cardiovascular.html ), and they have an FDA indication for reducing major adverse cardiovascular events, heart attack, stroke and cardiovascular death

    --they are renoprotective: 22% decrease in adverse renal outcomes, a composite of new-onset persistent macroalbuminuria, persistent doubling of serum creatinine and an eGFR <45, the need for continuous renal replacement therapy with no reversible cause of renal disease, or death from renal disease (see http://gmodestmedblogs.blogspot.com/2017/09/liraglutide-also-decreases-adverse.html )

    --they are associated with weight loss, 2.3kg in the above liraglutide study. there is a gradient of GI adverse effects (mostly nausea or anorexia) related to the actual GLP1 potency, with dulaglutide (esp at the 0.75mg dose) having the least potency, and semaglutide the most (and semaglutide is also the most effective on A1c, if tolerated)

--one disadvantage is that GLP1s have been available only as injections, most of them are once-a-week. but semaglutide was

approved as a pill which may even be more potent than injectable liraglutide; and, it is now available  (see http://gmodestmedblogs.blogspot.com/2019/06/diabetes-oral-semaglutide-better-than.html )

    --the average wholesale price (AWP) of these meds is all pretty high: semaglutide pills come as 3mg (for 1^st month only for treatment initiation, does not lower blood sugar), 7mg, and 14mg, one month supply $926.92 for each strength. Advised by manufacturer NOT to cut pills in ½.  Injectable dulaglutide is $956.76. Injectable semaglutide is $973.26 (though po empagliflozin, an SGLT2 is a mere $626 a month)

SGLT2 antagonists:

--these meds are NOT physiologic; they cause massive glycosuria.

--they have significantly less effect on A1c levels

-- one concern is some subterfuge in the studies; eg the initial study for the first SGLT2 empagliflozin (see http://gmodestmedblogs.blogspot.com/2015/12/empagliflozin-good-and-bad.html):

    -- Composite of death from CV causes, nonfatal MI or nonfatal stroke decreased 14%, from 12.1% to 10.5%

    -- BUT, buried in the supplementary materials (and not in the article!!) was the fact that there was no benefit in MI or stroke, but lower death from “other CV causes”: this includes “fatal cases that were not assessable due to a lack of information and were presumed to be cardiovascular deaths”.

--that being said, it is clear that all of the SGLT2s do decrease heart failure hospitalizations (they are, after all, loop-like diuretics).  But is this a good thing???

    --we know from older studies (eg SOLVD) that even asymptomatic patients having heart failure with reduced ejection fraction (HFrEF) have a mortality advantage if put on an ACE inhibitor

    --we also know that the only meds in patients with symptomatic HFrEF that have a mortality benefit are ACE inhibitors/ARBs, some b-blockers, and mineralocorticoid antagonists (eg spironolactone). There are no studies showing mortality benefit from loop diuretics (not studied that i know of. but likely do work, since these are the mainstays for urgent patients with severe heart failure who would likely have a high mortality otherwise....). But, these studies of SGLT2s do not find a significant decreased mortality from heart failure (only decreased hospitalization). so, maybe it is better for diabetics to actually have heart failure develop on an SGLT2 and then be put on a med that has known mortality benefit??????

--another recent study found a decrease in gout with SGLT2s. this was attributed to their known effect of decreasing uric acid levels, on the order of 0.5 mg/dL. But, we can achieve these not-so-large uric acid decreases by lifestyle changes (and these lifestyle changes have myriad benefits beyond uric acid levels...). my experience with a few patients is that stopping sodas (and their attendant hit of high-fructose corn syrup) lowers the uric acid level more than the SGLT2s. and there are lists of foods/drinks to avoid that can lead to significant decreases in uric acid levels, with likely similar benefit to SGLT2s in reducing new cases of gout (by the way, there is increasing argument that we should be checking uric acid levels more, eg in all hypertensive patients (this is a contested point, but the major European guidelines do call for checking uric acid levels routinely in all hypertensive patients: see http://gmodestmedblogs.blogspot.com/2020/01/uric-acid-threshold-for-increased-cad.html , which also highlights the potential cardiac benefits of lowering uric acid levels)

--in terms of renal effects: a canagliflozin study found a 30+% decrease in renal-specific outcomes of the combo of ESRD, doubling of creatinine, and death from renal causes, as well as of ESRD by itself  (see http://gmodestmedblogs.blogspot.com/2019/06/diabetes-guideline-update-pushing-sglt.html )

--adverse reactions are many, though several of them are infrequent, but include: lower extremity amputations, increased fracture risk (diabetics already have an increased risk of osteoporosis), diabetic ketoacidosis (even at very low blood sugar levels, eg  <200 mg/dL), severe mycotic infections, increased UTIs (and severe, ICU-bound cases of urosepsis), Fournier's gangrene. there are an array of FDA alerts about these problems: see http://gmodestmedblogs.blogspot.com/2019/06/diabetes-guideline-update-pushing-sglt.html

Commentary:

--it should be noted that all of the above studies were drug-company sponsored. 

--there are always concerns about what the "control" group received:

    --in the studies, the control groups were given additional meds so that the A1c's sort-of matched that of the intervention group (GLP1 or SGLT2), which brings up:

        --the other meds given to the "control" group may well increase their likelihood for cardiac adverse events (typically patients were given TZDs, without commenting if this was rosiglitazone!!!, a documented cardiotoxin; or they were also given more insulin or sulfonylureas, which also may well increase adverse cardiac outcomes presumably through hyperinsulinemia, which is associated with many potentially serious adverse consequences including being prothrombotic by increasing plasminogen activator inhibitor-1, increasing platelet adhesiveness, and being directly atherosclerotic -- eg see https://www.ncbi.nlm.nih.gov/pubmed/3277013 ). also, the Agency for Healthcare Research and Quality (AHRQ) noted that sulfonylureas vs metformin were associated with a 50-70% higher relative risk of cardiovascular mortality (see http://gmodestmedblogs.blogspot.com/2016/04/diabetes-update-ahrq-and-pioglitazone.html) [so, one potential issue is: do GLP1s or SGLT2s actually decrease cardiac events, or is it mostly because the control groups got more of the standard diabetic meds which increase these events??]

        --and, even with this, there was still a gap of about 0.5 percentage points of improved A1c control with the GLP1/SGLT2 over the control group, which could affect the renal outcomes (well-documented decrease in nephropathy with improved A1c's)

--there are significant concerns about adverse effects of these new meds. one recent study found that about 50% of people on either GLP1/SGLT2 meds stopped them because of intolerance. My caveat here is that i have been a very strong advocate for GLP1s for a long time, with my preferred therapy being metformin first and adding a GLP1 as a second agent when needed. I have seen remarkable successes with this med combo, which has the added value that neither of them incur significant hypoglycemia (unless coupled with insulin or sulfonylurea). And, i do very strongly push GLP1s with my patients, touting their beneficial cardiac effects (and the fact that about 80% of diabetics die from cardiovascular disease!!). My reasonably reliable guess is that only about 10% of my patients have intolerance to GLP1s leading to their discontinuation (esp nausea, and occ with such loss of appetite that they lose too much weight), though I do occasionally have to dial back to a less potent dose, such as dulaglutide 0.75mg; however, i can often subsequently increase that dose after a few months.  An enthusiastic support for the med may lead to fewer episodes of med intolerance???? (I will do a blog soon on placebo/nocebo effects)

--there are studies finding added effectiveness of combining the SGLT2s with GLP1s to achieve better A1c control

--so, my general feeling (which is probably pretty evident above):

    --GLP1s restore a specific deficit of a physiologic hormone that is deficient in diabetics; they are a targeted and specific very effective therapy; they seem to confer real cardiovascular protection; they can be well-tolerated; and their function fits in with a reasonable model of diabetes (not enough insulin being released to cover a meal)

    --SGLT2s create a nonphysiologic effect on the kidneys by dramatic increases in glycosuria (leading to several of the adverse effects, though SGLT2 do have the benefit of excreting more uric acid), have a less significant effect on A1c, and their cardiac benefits are largely preventing heart failure hospitalizations (vs the seemingly more significant decreases in mortality: see http://gmodestmedblogs.blogspot.com/2019/04/diabetes-assn-update-dapagliflozin-and.html).  the FDA gave an indication for SGLT2s for preventing heart failure; but their indication for the GLP1 liraglutide was for reducing the risk for myocardial infarction, stroke, and cardiovascular death….

So,

 the price of these meds is pretty exorbitant, not so surprisingly. But these are both great meds to add to our treatment of diabetes. Given the high risk of cardiovascular disease in diabetics, I really support a strategy of starting with metformin, then adding a GLP1. And, if not tolerated or other problem, try SGLT2.  Then, if GLP1 tolerated but does not work sufficiently, I have been adding an SGLT2 to GLP1 along with metformin, with good effect. Then, if that does not work, back to the regulars: insulin, sulfonylureas, pioglitazone…. 

I would also add my concern about the DPP-4 inhibitors. their function is to poison a widely expressed enzyme involved in many different effects throughout the body, which includes leading to an increased GLP1 levels (hence their benefit in diabetics):

    -- but they do have myriad effects (shotgun approach to the body: see http://www.glucagon.com/pdfs/MulvihillER-E2014-1035.pdf).

    -- these DPP4s have only small effects on A1c levels, without ancillary benefits in terms of all-cause mortality, cardiovascular mortality, MI, stroke.

    -- there are an array of potential adverse effects, such as heart failure, IBD, severe joint pain

    -- hence, seems likely to me that since DPP4s increase the GLP1 levels but lack the cardiac benefits of the real GLP1 agonists, this is likely due to the DPP4s having adverse effects on other systems in the body that counter these positive GLP1 effects. ie, best to stick with the GLP-1s themselves!!! (eg see http://gmodestmedblogs.blogspot.com/2018/03/dpp-4-inhibitors-increasing-ibd.html ). so, from my perspective, not much utility for DPP4s...

geoff​

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

to get access to all of the blogs:

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list