Home / Resources / Exclusive Interviews / Dr. Stanley Schwartz: Thoughts on SGLT-2 Inhibitors

Dr. Stanley Schwartz: Thoughts on SGLT-2 Inhibitors

May 14, 2014

Our publisher, Steve Freed, had a chance to talk with Dr. Stanley Schwartz, asking questions and getting some in-depth and very informative answers on exactly how SGLT-2 Inhibitors work. 

Steve Freed: Regarding the SGLT2 drugs that are now available, I know that a lot of physicians just refuse to write prescriptions because it’s new, and everyone remembers the problems with troglitazone, so they back off. They want to see it on the market for a couple of years. Have you been prescribing it a lot lately?


Dr. Stanley Schwartz: Yes. They tell me I’m one of the several top writers in the country. It’s because I reviewed the data before and after it came out….

I understand the pathophysiology of diabetes. I specifically think I understand why the kidney is abnormal in diabetes, and how this corrects it, because many people say, “Oh, you’re just treating a side-effect. You are not treating a core defect,” so I think I understand the core defect, and I think I understand how to bring the sugar down. If I understand the pathophysiology, and understand the research that’s been done, and the outcomes of the studies, I call this “evidence-based practice,” I will be an early adopter. The whole issue of evidence-based medicine, says, “If no evidence, keep using the ‘lousy’ therapy of the past, sulfonylureas, with its proven undue risks of hypoglycemia, weight gain, b-cell apoptosis.” I do evidence-based practice where, if there is evidence, I use it, but if there is no evidence, I have to come to my own judgment, and judgment based on pathophysiology, reading the primary data, and then, using an agent, getting experience, then I can go off and teach. That is what I’ve been doing, and I feel very comfortable using canagliflozin….

SF: You mentioned treating the core defect. It certainly does not directly treat the core defects when it comes to diabetes, which are beta cell issues, and insulin resistance. It’s kind of unusual for a drug, basically, to treat everything that’s wrong with diabetes except the core defects. How do you respond to that?

Dr. S: Yes, when I speak at the ADA, I’m going to be proposing a new way to diagnose diabetes, because this concept of many core defects, I think, is wrong. I think the key core defect is the beta cells not working for many, many different reasons. The other thing is that even when you talk about core defects, people forget about a different “core defect” called the brain. The brain, in certain situations, wants to be absolutely assured that it gets enough sugar for it to work on, because it doesn’t work on other substrates. That leads to what is going on in the kidney in early diabetes. It turns out there is a sensor in the glomerulus that’s pointed at the urine, not at the blood. The sensor recognizes that there’s increased filtration of sugar getting into the glomerulus. With the concept that the kidney job is to make sure there’s enough sugar for the brain, as soon as it sees extra sugar in the glomerulus, it up-regulates SGLT2 protein, which engenders increased re-absorption of sugar back into the blood. That sensor is trying to keep enough sugar in the blood stream for the brain to work. Once you have this increase in SGLT2 in the kidney proximal tubules, you are getting more reabsorption which keeps the sugar artificially high. So, by using these SGLT2 inhibitors, you are blocking the effect of the up-regulated SGLT2 molecules, and lowering the blood sugar. When you lower the blood sugar, guess what, you get to the core defect. That is, you are reducing glucose toxicity, and the beta cell function is better, and that’s been proven. So, we are not treating a ‘passive player’ by increasing glycosuria; we are treating an active contributor to hyperglycemia, the kidney, and by doing so, decreasing glucotoxicity, improving the core defect, the abnormal beta-cell.

SF: I know that there are different kinds of patients out there. There are type 1 and type 2; there’s late onset diabetes; there are overweight type 2s and underweight type 2s. We now have something called type 3, and we’ll probably have type 4, 5, and 6 in the not too distant future. What kind of patient do you find the greatest success with in using SGLT2 medications?

Dr. S: Anyone with diabetes; but especially overweight people; because they lose weight with these agents because losing sugar in the urine equals losing calories so you are losing weight. The second is, let’s give the patient the benefit of the doubt, they are trying to eat the right diet, but they occasionally eat some wrong stuff, or maybe too frequently. This class modulates or reduces the rise in sugar in the blood when they are eating the wrong stuff, because it smoothes out those peaks that occur. For the person who is not eating as good as they should, it gives tremendous benefit in this regard. The people who have the highest sugars respond the best because it goes out in the kidneys, but you have to be careful because that means they are at greater risk of the potential volume-loss side effects, as well.

SF: So, you explain to the patients that this is not an excuse to eat anything they want, such as a triple banana split sundae, that you will just pee it away?

Dr. S: Before the drugs came out I used to say, “Well, this is the first one where a company can advertise “this a drug that you can cheat with,” but I don’t tell that to patients. But, functionally, that is what it does.

SF: Every drug has its bad points and side effects, its precautions. We know from the studies what these are for the SGLT2s. Number one, in your experience, have you seen much of the side effects of this drug, and how do you counsel your patients when you write a prescription for the SGLT2s?

Dr. S: Yes, there are known issues, or side effects. They have to do with the extra urination that occurs when you have the osmotic load of glucose in the urine. You are going to lose more fluid. When you have sugar in the urine you have a potential risk for increased urinary tract infections, and for vaginal infections, and for balanitis thrush (yeast infections on the tip of the penis). I’ve seen all of it, no question about it, and yet I perceive that my incidence of these side effects have been reduced by my aggressive, pro-active education when I write that first script.

Let’s deal with each of these separately. From the point of view of volume, a patient’s sugar is very high, 300 mg/dl. They come to me with an HbA1c of 12, I’ll tell them they have to follow my diet for two or three days, because the effect of diet upon blood sugar is very effective very quickly. Let’s bring down the blood sugar at least somewhat before I give you this agent and prevent the potential excessive volume loss when you bring your sugar down. That is part one.

Part two, I tell them to drink enough liquid to keep the urine dilute, pale yellow, or colorless, and clear, like water. I tell them to drink any liquid other than alcohol, or drinks sweetened with sugar. Make sure you drink enough. Now, functionally, they can’t really do it because it requires too huge an intake of fluid, but when they try, they succeed to the point of getting a specific gravity of 1.010 to 1.015, and that’s enough to prevent a volume deficit kind of symptoms, predominantly dizziness. Also, in that realm, is because of the potential volume stuff, if a patient’s blood pressure is on the low side, 100-110/60-70, I will actively cut back something as I give them the SGLT2 inhibitor. If they are on a mild diuretic, HCTZ 25/50, I’ll get rid of it. If they are on several blood pressure pills, I’ll reduce or stop one of them, just to prevent low blood pressure as a result of the volume story.

Also, when teaching them to drink enough liquids, I tell them they can’t evaluate how much they are drinking by what they take in, because everybody has different insensible loss of their fluid, loss of water or fluid through pores, whether it’s sweat, or just as vapor. So I tell them the only way they know they are drinking enough is not by what comes in, but what comes out, because if the kidneys says there is not enough fluid in the body, it will reabsorb the fluid, and concentrate the urine that is going out. When the kidney has enough, it lets the extra out, and that’s when you know your urine is dilute, and you know you are drinking enough. So, that’s the volume story.

The infection story: First, we won’t prescribe it in those occasional folks with frequent UTI’s or yeast infections in the past. There is a modest increase of urinary tract infection, ~1%rise in risk. For yeast infections, in men, it is mainly in uncircumcised men and we tell them to get the tip of the penis dry before they leave the bathroom; rarely get recurrent infections. In women, for both urinary infections as well as yeast infections, we tell them to drink enough fluids. We tell them to use fastidious bathroom habits: wipe front to back, and get dry; urinate after intercourse, before you go to sleep, and again, getting dry, and again, rarely recurrent. These are very practical, upfront discussions that we tell them about and educate two or three times before they leave the office with a script for an SGLT2 inhibitor.

SF: There have been questions, because obviously this works directly on the kidneys. We’ve seen a decrease in microalbumin, a decrease in blood pressure, some of the good things, but can it also damage the kidneys?

Dr. S: We don’t think so in several respects. First, when I first heard about the class, we were worried about that exact same thing, but there’s at least a couple of small studies that showed that there are people who genetically have absence of SGLT2, so from birth they have extra urine sugar, and guess what, it seems they have normal life spans. So the presence of extra sugar in the urine, or sugar, at all in the urine, doesn’t seem to damage the kidneys. Number two, we think it’s actually helping the kidneys for the reasons you mentioned, the reduced microalbumin, the reduced blood pressure, reduced weight,and please don’t forget the reduced sugar. All of those things seem to improve kidney function.

There is a modest reduction in eGFR, three to five ml/min over the first few weeks, but in the studies I’ve seen, over three or four months the eGFR seems to return to a baseline status. That’s why we think the FDA did not tell us to stop them automatically with a mild drop in eGFR. They noted the story of reduced GFR and they said stop it if it’s persistently under 45 for canagliflozin, and under 60 for dapagliflozin, but they don’t tell you over what period of time you should give the kidney time to get better. I usually wait three months, but basically it’s telling you that the FDA understands that there’s a modest drop at the beginning, but comes back to normal over several months in the vast majority of cases. We don’t think there is going to be any late long-term adverse events in regard to kidney function with this class of medication.

SF: When they were doing the investigation on this, the SGLT2, they thought that it would actually prevent the absorption of sugar by 90%. The actual result in humans is more like 30%, so now they are doing studies with the SGLT1 inhibitor and the SGLT2 inhibitor to see if they can increase that. Obviously, our bodies do need some sugar, that’s how our bodies function. What are your thoughts as far as why aren’t we getting the greater numbers with the SGLT2? Do you think it is a good idea to look and see if that may be a possibility of adding SGLT1 and SGLT2, that’s basically how it works in the tubules. Your comments?

Dr. S: First, let me be a little bit more precise. SGLT2 is a protein that is in the proximal portion of the tubules of the kidney. The SGLT1 is in the distal part of the proximal tubule of the kidney, but that’s the one that’s also on the gut surface. The issue with each of the SGLT2 inhibitors that are coming out is how much cross-reactivity there is; how much of that SGLT2 inhibitor drug is also seeing the SLGT1 sites. That is what you are alluding to, that is, there may be some reduction in gut glucose absorption, as well, and just depends on the amount of cross-reactivity. Remember when there is a significant cross-reactivity, translate your blocking SGLT1, then the sugar is going out in the gut, and that can lead to GI side effects, basically as diarrhea. That’s not always a good thing. At this point I’m telling you I don’t know enough, because I don’t know the details of research studies that have been done in regards to how much is a good balance of SGLT2 and SGLT1. I just don’t have the data. I’m sure they’ve done some stuff. I just don’t know those answers, but those are the issues, how much cross-reactivity, how much side effect you’ll get because of diarrhea. I’m waiting to see the answers when these drugs are getting closer to the market. Certainly, the current ones are from more like 99% SGLT2 specific, so we are not seeing significant diarrhea signs.

The other issue you allude to is interesting, an action that made the media for the wrong reasons. That is, we know that it drops the renal threshold. Remember, the normal renal threshold is 180. In diabetes, based on the mechanism I told you, the up-regulation of SLGT2, it goes up to 240 before you get glucose released into the urine; but the drugs reduce that renal threshold down to 60, 70, or 80, or something like that, meaning if it would reduce the renal threshold that degree, why aren’t we seeing more than a one percent drop in glycohemoglobin? Why aren’t we seeing ‘the full effect’ of that drop in renal threshold for glucose?

The intriguing thing that made the news was that the body’s response to the increased glucose excretion in the urine is the increased glucose production by the liver. The media saw that article, and said, “Oh what a terrible drug! It’s increasing glucose production,” but they forgot that, guess what, that despite the increase in glucose production by the liver, you are still getting a one percent drop in HbA1c, which is more than most agents that we have available can do. Be that as it may, despite increased glucose product by the liver, we get a very nice glycohemoglobin drop. Could it be more? Yes, and guess what? We have the data that it can be more, that the hepatic issue is significant, because the incretin class, both DPP IV inhibitors, and the GLP1 receptor agonists reduce hepatic glucose production.

There was an abstract at the ADA by Carol Wysham, and I’ve seen this clinically, even before the abstract, and since, certainly, where you get a complete additive benefit of combining SGLT2 inhibitor with DPP IV. Most of the time when you add a second agent, you don’t get the full effect of the pure combination of monotherapy effectiveness, or drop in glycohemoglobin, but here, you get the full additive drop of the DPP IV. You get the full additive drop in blood sugar/HbA1c, of the GLP1-receptor agonist when you combine it with the SGLT2 inhibitors. That’s wonderful right there.

Moreover, you get the weight loss with the SGLT2 combined with the DPP IV, so you could take two oral drugs, not have a potential GI side effect, and get the equivalent effect of giving a GLP1 receptor agonist.

On the other side, if you have a patient who has no problem with infection, or no problem with GI upset with the GLP1 receptor agonist, you get the additive glycemic, you get additive weight loss of 3.5% on one, or 3.5%, so you could lose over several months 7% of your body weight by combining a GLP1 receptor agonist and the SGLT-2 inhibitor. We don’t see the increased glucose production by the liver as a downside, because the drugs still work, and then if you address the increased hepatic glucose production, you can get additive effects when you combine DPP IV or GLP1 receptor agonist.

SF: If you go back to 1950, when all we had was sulfonylureas, treating diabetes was kind of simple, and now, today, it’s actually very confusing. What is your preference? Where have you seen the greatest results? With the DPP IVs in addition to the SGLT2s, or with the GLP1 and the SGLT2s and metformin? Should that be the base to work from?

Dr. S: There’s nothing bad about metformin other than you must use it in the right patients, have to watch for the lactic acidosis. It remains a standard, but I’ll tell you, when I have somebody that comes into the office the first time, and their sugars are very high and overweight, I have no hesitation in giving them a GLP1 receptor agonist and an SGLT2 inhibitor right from the get-go, and not worry about the metformin, because the hassle with metformin is that you should ideally titrate it up over a couple of weeks to make sure they don’t have diarrhea from it, or other kind of GI side effects. So, the weird thing is, I’m using combinations of these two drugs, GLP1 receptor agonists and SGLT2, and they come back a month later, their sugars are superbly improved. They’ve already lost five to fifteen pounds in one month. It’s just remarkable. That’s one type of patient I’ve had with tremendous response using these agents.

The other patient is a patient who is on insulin. Let’s talk several situations. They are on basal bolus insulin, and their bolus is 3-6 units per meal, I’ll reduce their basal, and stop their bolus as I give them GLP1 and SGLT2, and I can now control patients with just the basal plus these two agents. I’ve gotten rid of 3 shots a day. Remember bolus insulin accounts for 85% of outpatient hyperglycemia, so it markedly reduces the risk of hypoglycemia. It’s superb! Then there is the patient who’s eating the wrong stuff and getting hypoglycemic unawareness — I decrease their basal 50% for those issues; and I start those two agents. I can stop the insulin in the office that day if their total dose when they come into the office is <40 units of insulin/day. I use this drug to minimize the need for prandial insulin. Maybe 95% of people on prandial insulin don’t need it if they are on that type of combination. About twenty percent of patients who come in on insulin, can come off insulin by using these agents with a diet. I had just a remarkable benefit and success in using these agents, especially in combination with GLP1 receptor agonists, as well as with some of the other drugs.

I have a short article case report that will be published in a book published by Dr. Draznin, publishing for the ADA. It will be coming out the beginning of next year. This was a woman who was on 15 units Q6h of U500. That’s 300 units a day. She was eating sweets, and also getting hypoglycemic unawareness, didn’t know it, and the physician didn’t recognize it. By the addition of initially the GLP1 receptor agonist, a few months later, April of last year when the SGLT-2 came out, added that, and by having her follow my diet, (she never saw a nutritionist as told), but she stays away from sweets, and then, a few months later, add metformin and piaglitizone. One year after I met her, she’s off insulin, and her glycohemoglobin was 6.9% and she lost 35 lbs. She is doing tremendous. I don’t tell you that you can get everybody that is on that much insulin off insulin. I can tell you that is an exaggerated case of what I see every day in the office our ability to markedly decrease the insulin, and in fact, get some of them off insulin.

SF: From what you are saying, if you had your druthers, and the typical patient walks in who is overweight with a hemoglobin A1c of 7.9%, you would prefer to put them on SGLT2 and GLP1, rather than the DPP IV?

Dr. S: Absolutely. I respect costs, but for those people who have insurance, the companies have very reasonable monthly discount costs for the SGLT2s; for the GLP1s, it is a reasonable, perhaps $20-30 a month, so for $30 or $35 a month you can keep people off insulin, they can lose weight, and their health is markedly improved.

SF: Every insurance company is different with different tier levels: does the cost prevent you sometimes from writing those prescriptions? If you had to pay retail for those two prescriptions, it has to run somewhere over $500 a month.

Dr. S: True, but with co-pay reductions for those with private insurance, and reductions with ‘dual eligibility’ in those on government plans, the cost can be quite manageable.

Prior authorization is often an issue, but we can often find 4-5 reasons an individual would be at risk for adverse events from hypoglycemia, so we can usually get these meds authorized. Moreover, remember the ‘true’ cost of high blood sugar if you include the increased need for glucose test strips, increased ER visits, admissions for hypoglycemia, and complications of DM over time.

SF: Do you have a special form letter that you use to get approval?

Dr. S: Ninety-nine percent of it is done by phone. I do have something. Remember you published this in Diabetes in Control many years ago. I have that little piece of paper that says, “This patient has hypoglycemia on insulin,” and “This patient has hypoglycemia on sulfonylurea.” I check the appropriate box. If not, there is the other box that says, “This patient is at high risk of hypoglycemia.” I check it in everybody, because there are eight other boxes or so that I could check to justify it to the company. They walk up and down stairs; they are a fall risk; they have heart disease; they have CVA/TIA dizziness, already kind of symptoms; they watch little kids; they have a dangerous profession; they use electric tools like blow dryers, hair curlers, and hair straighteners, so I can check four or five or six boxes for everybody, and the company now respects how dangerous hypoglycemia can be for their costs. They have the ADA, two articles, two abstracts in June, [for those] over 65 there are more ER visits and admissions for hypoglycemia than there are for hyperglycemia. Just this month there is an article on insulin that says the same thing, that insulin ER visits due to hypoglycemia are more common for treatment than hyperglycemia, so we have that form check list that we can submit for any of these patients that I’m giving them more expensive drugs from the get go.

SF: Right now we have the two SGLT2s, canagliflozin and dapagliflozin, and Lilly is close to coming out with theirs, empagliflozin. Do you have any preference? Do you see one benefitting more than the other?

Dr. S: I do not have enough data on empagliflozin in my head, so I’m not going to comment on that. Right now there are advantages and disadvantages to each of the current agents. I would suspect that one major issue is that you can use canagliflozin in people whose eGFR is down to 45, and it’s still effective, yet there seems to be less efficacy in eGFRs 45-60 with dapagliflozin, so you are not supposed to use it in dapagliflozin patients with eGFR of 60. That is a big one.

The dapagliflozin has a warning about bladder cancer. Personally, I do not think that is a risk, and that is why the FDA, ultimately, approved it; but guess what, that’s going to disturb its use by patients and physicians, even if they accept my statement that I do not think there is a significant bladder cancer risk. The canagliflozin does not have that warning. Canagliflozin has a warning about potassium, and the dapagliflozin doesn’t. The warning on the potassium, I think, is appropriate, but over-stated. The majority of people, rarely will get a jump in potassium. In people who are older with an eGFR of 45-60, potassium might go up, so you have to be aware of it. In fact, I have my patients come back in two to four weeks to make sure potassium is good, renal function stays fine, that they are drinking enough to stay asymptomatic.

There is a drug interaction with canagliflozin that is not present in the label of dapagliflozin, that is, canagliflozin that it can raise the digoxin levels. The other one was Rifampin. So, there are ‘ifs,’ ‘ands,’ and ‘buts’ on either side. I think people will get experience with both and then make their own decisions.

SF: My last question, and I’m going to put you on the spot, but I don’t want to get you into trouble. I know you are presenting at the upcoming ADA this June. I understand there may be some very interesting results being reported, and maybe some changes. Without getting into trouble or divulging anything that the ADA would get upset with, can you give us any clues as to what we might be hearing at the upcoming ADA.

Dr. S: In terms of what I am saying, or what is going to be presented at the ADA, in general?

SF: Well, just your knowledge of what’s going to happen at the ADA, in general.

Dr. S: I don’t know any ‘late-breaking news. I hope the results of the ten year Kaiser Permanente pioglitazone/bladder cancer study will be presented there, but I have no absolute knowledge that they have the data together. I know the study finished the end of last year. I just hope they have enough data they can provide an abstract at the ADA. I do not know if other stuff is coming out.

I can tell you briefly what I’ve been asked to speak about, and hope I get an interesting response. I’ve been asked to differentiate type 1, type 2, and Latent Autoimmune Diabetes of the Adult (LADA), and how physicians should go about making the right diagnosis. My point is there is very great overlap in those diagnoses, especially because most physicians are not checking antibodies in their patients with diabetes. I will be asking for a potential reclassification of our current diabetes. The last time this was looked at was in early 2003 or so. What I mentioned before, is I think the core defect in DM is the beta cell. I think that if you understand the influences on reduced beta cell function, the glucose toxicity, lipotoxicity, inflammation, that you can then engender a better understanding of how to diagnose people. I think you should measure antibodies. I think we are very close to being able to say that everyone should get their genome analyzed because we are learning so much about the genes that we will be able to infer that different genes will respond better or worse to different therapeutic modalities. If you have this new classification, and understand the drugs, that will have a different sense of which drugs to use for which patients. The challenging statement I made at the end is because incretins reduce glucose toxicity, they can affect resistance of the GLP1 receptor agonist, because of the weight loss, and because they are sure to improve inflammation, then incretins can be used in every diabetic, off label and type 1s, because of its reduction in glucagon, and reducing inflammation, so it might slow the deterioration of beta cells; maybe engender improved beta cell function over time, even in the type 1s. The type 2s have all those benefits. In LADA, its benefit on inflammation has been suggested by a couple of articles. I hope I’ll start some controversy by presenting this new beta cell focus for specification with implications for therapy. I actually have not heard of any new big stuff coming out except one.

SF: That is great. A couple of things that you mentioned brings me to my last question. There is always “the last question,” which never ends, because you keep bringing up such good points that it brings up other questions, which is great. What are your thoughts of using an SGLT2, and/or one of the incretins, either separate or a combination, as a weight loss program?

Dr. S: SGLT2, I don’t see a major role in the metabolic syndrome, or pre-diabetic patient, just because they are not going to be losing that much sugar in the urine. It may have a preventive role, which is a different question, but that is a separate issue. For weight loss and the metabolic syndrome/pre-diabetic, I do not think SGLT2 will make a huge difference.

On the other hand, we know that Novo and Amylin have published data on their GLP1 receptor agonist for weight reduction in metabolic syndrome. Both of them have data in that regard, and it works. On the order of about 75% of people will lose significant amounts of weight, in metabolic syndrome you use these agents. I know for sure that Novo has put in to the FDA for an indication for weight loss, or for obesity. I have no hesitation in using those for weight loss in obesity.

SF: I want to thank you for your time. I’m looking forward to meeting up with you at ADA in San Francisco, and talking about your presentation.

Dr. S: Thank-you, I’ll look forward to it.

Copyright © 2014 HIPER, LLC

See more SGLT-2 Resources