[ed note: Michael Jorrin, a medical writer who we have dubbed “Doc Gumshoe” (he’s not a doctor), shares his thoughts with us once or twice a month — what follows is his latest missive, the words and opinions below are his alone.]
I didn’t plan it this way, but the timing of this Doc Gumshoe post couldn’t have been much better. As I was gathering material, news items from the American Diabetes Association’s 74th scientific meeting, which ran from the 13th to the 17th of June, kept popping up.
“Popping up” isn’t a bad way to describe what these news releases were doing. They were mostly modest pops, like back yard Fourth of July fireworks. Some of the news media covering this event did throw around the B word, but if any of the agents discussed at the meeting attain Blockbuster status, it will be because of the magnitude of the problem (and the opportunity!) posed by diabetes, and not because any of the drugs appear to have the capacity to work genuine miracles.
So, what’s the magnitude of the problem?
In Doc Gumshoe’s previous piece about diabetes (“The Why and the What of Diabetes Management”) there was a Tiny Addendum that noted that the Centers for Disease Control had upped their estimate of the number of persons in the US from about 26 million in 2011 to about 29 million today. That’s an increase of a million per year. Note that those figures include both Type 1 and Type 2 diabetes, and that somewhere between 25% and 30% of all diabetes is undiagnosed. Previously the ADA had stated that there were about 1.9 million newly diagnosed cases of diabetes in adults per year. Yes, those numbers are estimates, albeit careful estimates based on the best data.
And here’s another estimate from the CDC: 86 million people in the US have what is labelled “prediabetes,” i.e., insulin resistance or impaired glucose tolerance. The CDC projects that without intervention, about 30% of those persons would have Type 2 diabetes within 5 years.
That would up the total diabetic population of our fair land to somewhere in the neighborhood of 55 million individuals.
My bet is that we won’t reach that figure – that’s a worst-case scenario, and a lot of effort is being lavished on diabetes awareness, screening, healthy living, etc. Just because Mayor Michael Bloomberg’s proposed ban on half-gallon sugary drinks (or whatever size) didn’t go through doesn’t mean that Attention Isn’t Being Paid.
All the same, Doc Gumshoe wouldn’t be a bit surprised if we got to 35 – 40 million cases of frank diabetes by the year 2020.
Here’s something else: United Healthcare projects that the costs of diabetes treatment could reach $500 billion by 2020. I don’t know whether that’s meant to strike terror into the hearts of the citizenry, who would have to fork over that moolah, or to provide an incentive to the pharmaceutical industry to come up with drugs that could snaffle a chunk of that loot. However, as I write these words, the ghost of Miss Truesdell is looking over my shoulder and shaking her head. She doesn’t think that $500 billion number makes sense. The 2012 medical costs totaled $176 billion, and the $69 billion in lost productivity brings the total to $245 billion. I am skeptical about adding lost productivity to the costs of diabetes treatment, but even if we include lost productivity, I wonder whether the cost will really double to $500 billion in 8 years. Anyway, however we figure it, the whole pie is immense and Pharma wants its slice.
Big Pharma steps up to the plate
Despite the recommendation by McKinsey & Company a couple of years ago that big pharmaceutical outfits should ease back on R & D and concentrate their efforts on their “core competency,” that being sales and marketing, Big Pharma is lavishing lots of resources on R & D in the diabetes area. Here’s a quick list (by no means exhaustive) of the clinical trial activity of some of those outfits:
No surprise that Novo Nordisk (NVO) tops the list, since they are the long-established leader in diabetes treatment products, specifically a range of insulins, but also the incretin mimetic Victoza (liraglutide). Further down, we’ll be discussing the benefits vs potential risks of the entire class of drugs that relate to incretins and the rapid insulin response, as well as the duel between Victoza and Eli Lilly’s (LLY) dulaglutide.
And here’s a list of the top 10 diabetes drugs in terms of sales:
As you see, seven of the ten are insulin products in one of several forms, and five are Novo Nordisk products. This may come as a surprise, since the great majority of persons with diabetes have Type 2 diabetes (T2DM), which is, at least initially, usually treated with non-insulin agents. Several factors contribute to the global dominance of insulin:
- One, eventually almost all persons with T2DM wind up needing supplementary insulin, as their pancreatic beta cells quit functioning after long years of battling against systemic insulin resistance. (We discussed this at greater length in the previous blog.)
- Two, many persons with T2DM, particularly in less-developed regions of the world, have advanced disease at the time of diagnosis, likely beyond the stage where merely increasing the sensitivity of insulin receptors does much good. Exogenous insulin is the quick answer, and mostly does the trick.
Metformin, which is a usual first-line drug for treating T2DM, is also huge seller globally, but it’s relatively cheap. If the sales of generic metformin plus branded formulations such as Glucophage (Bristol Myers Squibb (BMY)) are added, the total dollar amount only hits somewhere between $1.5 billion and $2.0 billion.
However, another trend worthy of our notice: all three of the top sellers that are not insulin products are drugs that target first-phase insulin release through the incretin response. And many of the drugs discussed at the ADA meeting, particularly the ones that attracted the most attention, also address the first-phase insulin response. Why is this?
A bit more about incretins and the first-phase insulin response
In a non-diabetic person, the pancreatic beta cells are secreting insulin all the time, because every cell in our bodies requires insulin to enable the process of converting glucose into energy. Even when we’re sound asleep, our hearts pump, our lungs suck in air, and our brains have dreams or nightmares, as the case may be. We don’t turn off.
This constant insulin release is called the basal insulin response. It accounts for about half the total. The other half, called the first-phase, or sometimes prandial insulin response, is in direct response to food intake. When nutrients of any kind get into the digestive system, a couple of different peptides are released from the gut, mostly from the small intestine, although some are also released from the colon. These peptides are designated GLP-1 and GIP and are generally termed incretins.
What they do is send a message to the beta cells that it’s time to step up insulin secretion, because a load of glucose is being dumped into the bloodstream. In addition, they send a message to the liver that it’s time to stop breaking down glycogen into glucose and releasing that glucose into the bloodstream – it’s okay for the liver to do that when there’s no food in the digestive system, but we don’t need that extra glucose right after a meal.
Incretin release starts quite soon – about 15 minutes after any nutrients pass into the small intestine – and continues for up to 2 hours. The response of the beta cells to this signal is actually more rapid than their response to the presence of glucose in the blood. This has been demonstrated by studies timing insulin response to intravenous glucose injections compared with insulin response to food in the gut. Food in the gut triggers a quicker response.
However, the half-lives of these two peptides are very short – GLP-1 about 2 minutes, and GIP about 5 to 7 minutes. That’s because incretins are broken down by an enzyme labelled DPP-4. The release of incretins and their breakdown by DPP-4 is clearly a homeostatic process – the organism only requires a robust first-phase insulin release during digestion (while glucose is being sent to the bloodstream) but not after.
But in persons with diabetes, the incretin response is not sufficient to trigger the release of enough insulin to bring the blood glucose to normal levels. This may result in that adverse feedback loop, in which high blood glucose brings about increased insulin release, which in turn leads to resistance at the insulin receptors and continued high blood glucose.
Two classes of drugs attempt to remedy this defect. One class consists of the inhibitors of DPP-4 (the breakdown enzyme), thus prolonging the life of the incretins. DPP-4 inhibitors are called “gliptins.” Merck’s (MRK) Januvia (sitagliptin), second on the list of best-selling diabetes drugs, is one of these.
Strong competition for Januvia
One current challenger is Galvus (vildagliptin). Novartis (NVS) is putting the pedal to the metal on this one. Currently 139 clinical trials with vildagliptin are at various stages. Galvus has not received FDA approval, even though Novartis has been knocking at the door since 2007, the reason being that there were hints that, along with other incretin-based drugs, there might be an association with pancreatic cancer. Note, there was no evidence that vildagliptin itself ever led to pancreatic cancer in a patient, and the FDA and the European Medicines Association (EMA) have reviewed huge amounts of evidence regarding sitagliptin (Januvia) and exenatide (Byetta), an incretin mimetic, and found no evidence of increased incidence of pancreatic cancer. The FDA/EMA statement and a summary of the evidence was published in the New England Journal of Medicine in February of this year (N Engl J Med 2014; 370:794).
My guess is that as the results are announced of the many safety studies with Galvus, FDA approval for this drug will be forthcoming. Galvus has performed very well in clinical studies; for example, reducing mean HbA1C levels in older patients to essentially normal levels, and also minimizing some of the adverse effects (i.e., hypoglycemia) associated with some other antidiabetic drugs. A competitive disadvantage versus Januvia is that Galvus needs to be taken twice daily, whereas Januvia can be taken once daily. Galvus is likely to be a big boost for Novartis.
The duel between Victoza (liraglutide) and Eli Lilly’s dulaglutide
Both of those drugs are incretin mimetics. Incretin mimetics are not synthetic insulins. Instead, they are entirely different molecules that nonetheless closely mimic the activity of incretins, which is to say that they tell the pancreatic beta cells to release insulin, and they tell also slow the process of converting stored fat (glycogen) to glucose. They do this by activating the GLP-1 receptors on the surface of pancreatic beta cells, but only when there are excessive levels of glucose in the blood. When glucose returns to normal levels, insulin secretion also returns to normal or basal levels – that is, the prandial insulin phase peters out.
Incretin mimetics work in tandem with native incretin secretion from the gut. However, an important difference is that the incretin mimetics are not degraded by DPP-4. While native incretins have a very short half-life, mostly less than 10 minutes, incretin mimetics have a much longer half life; therefore it is feasible for these agents to be dosed as drugs.
Lilly began a series of Phase 3 clinical trials with dulaglutide in 2010, comparing their new agent with a wide range of antidiabetic drugs, including insulin glargine and insulin lispro, metformin, the sulfonylurea glimepride (Amaryl), the DPP-4 inhibitor Januvia (sitagliptin), which is the 2nd-biggest diabetes seller, and the incretin mimetic Byetta (exenatide), as well as in patients with kidney disease, hepatic impairment, patients taking digoxin, and women on oral contraceptives. Trials to date have been uniformly successful; for example, dulaglutide achieved greater reductions in HbA1C than either insulin glargine alone or a combination of insulins glargine and lispro, and also bested Januvia and Byetta.
A greater challenge for Lilly has been the head-to-head comparison with Novo-Nordisk’s Victoza (liraglutide), the top-selling incretin mimetic, and the number 5 seller in the diabetes sweepstakes. In the AWARD-6 trial, dulaglutide lowered HbA1C slightly more than Victoza, by 1.42 percentage points from baseline, compared with 1.36 for Victoza. This means that for many patients who present with HbA1C levels in the neighborhood of 8%, dulaglutide alone could be enough to bring them within acceptable limits, even though it is not likely that this new agent will be approved as monotherapy for newly diagnosed patients.
The very small difference in favor of dulaglutide will not be enough to change the prescribing practices of most physicians. However, there is a major difference. Whereas Victoza is given by injection every day, dulaglutide is given, also by injection, once a week. When dulaglutide receives FDA approval in the US, it will be marketed under the name Trulicity.
Victoza did perform better than dulaglutide/Trulicity in one respect. Patients on Victoza in the AWARD-6 trial lost a bit more weight than those on dulaglutide – 3.6 kg versus 2.9 kg. That result has prompted Novo Nordisk to study Victoza as an obesity drug, perhaps to compensate for their anticipated lost sales when dulaglutide hits the market.
As I have said before, I’m reluctant to use the B-word when talking about a drug’s prospects, but if there’s one where that word might be appropriate, it’s dulaglutide/Trulicity. This couldn’t come at a better time for Lilly, which, as some of you have likely noticed, needs something to wipe the egg off its face after recent flubs with Alzheimer’s and cancer drugs. Dulaglutide will very likely bring substantial rewards for Lilly.
… and now for something new and trendy?
Some industry insiders are calling it a “bionic pancreas.” Here’s how it works. First, for comparison, the standard insulin pump, used by millions of persons with diabetes, mostly Type 1, requires the patient to monitor his or her blood glucose levels by the standard finger-stick method, calculate how much insulin he/she needs at that moment, and instruct the pump to deliver that amount of insulin. It was a huge advance over the old way of getting insulin, which consisted of several daily self-administered subcutaneous injections. The insulin pump is more accurate and can deliver insulin more gradually than the insulin injection. The user can tinker with the way the pump delivers insulin, ranging from a bigger dose initially followed by a trickle, perhaps mimicking the prandial / basal insulin responses, to a more even release throughout the 24 hours. Nonetheless, it’s far from perfect, and some insulin pump users continue to experience potentially dangerous episodes of low blood glucose, and also insufficiently controlled elevated blood glucose.
The so-called bionic pancreas tries to eliminate these issues. The patient has two subcutaneous needle-like devices implanted. One delivers insulin, and monitors blood glucose levels. The other delivers glucagon, the hormone that signals the liver to turn stored glycogen into glucose. The implanted needles also monitor the body fluids for insulin and glucagon respectively, so that these two agents are released into the body as needed. The devices are connected to externally-worn insulin and glucagon pumps.
What controls the whole system is that the information from the implanted devices is transmitted to a smartphone, which does the calculations and instructs the pumps as to how much insulin or glucagon to deliver.
A pilot version of this system was tested by Mass General Hospital in 20 adults and 32 young people. In the pilot version, the patients still used the finger-stick method to monitor their levels, but then input that data to their smartphones. The results strongly favored the experimental system over the insulin pump, lowering blood glucose, in the words of the chief investigator, Steven Russell, MD of Mass General, “to levels that have been shown to dramatically reduce the risk of diabetic complications.”
Perhaps more important, it reduced the incidence of hypoglycemic episodes by 37% in the adults and by 50% in the young people. Although in the long run, complications are what destroy the lives of people with diabetes, on a day-to-day basis what patients and health-care providers most have to watch out for are the low blood glucose episodes – so called “insulin reactions” – which can cause fainting, shock, or coma.
If my instincts are correct, this is a development that will have enormous impact on the lives of persons with diabetes, both in the short and long-term. The likely first market for these devices will be younger persons with Type 1 diabetes. But the Type 2 diabetic who can no longer control the disease with non-insulin agents will also certainly benefit and likely jump aboard..
The next time you see a person looking intently at a device, it might not be a device-addict checking for tweets, but a diabetic controlling his/her condition “on the go!”
As for which medical device outfit will jump on this to make and market the “bionic pancreas,” Doc Gumshoe will keep his ear to the ground.
Salsalate lowers blood glucose in Type 2 diabetes
Salsalate is an NSAID (non-steroidal anti-inflammatory drug) in the salicylate family, related to aspirin, which is used to treat osteoarthritis. There were suggestions that it might be effective in treating diabetes as long ago as the 19th century, but these were based on anecdotal evidence and not seriously pursued once insulin deficiency was identified as the underlying pathology in diabetes. However, it has long been thought that inflammation had a role in the pathogenesis of diabetes, and this has now been confirmed in well-conducted clinical studies.
A shorter study, 14 weeks, evaluating the effectiveness of salsalate was published in 2010. This has now been followed by a 48 week study in 286 subjects with T2DM, comparing salsalate with placebo. HbA1C levels in subjects taking salsalate were 37% lower than in those taking placebo.
And patients in the salsalate group needed fewer additional medications to control their diabetes. In addition to its beneficial effects on blood glucose, salsalate demonstrated some benefit in reducing some markers of cardiovascular disease, although the salsalate cohort experienced small increases in LDL cholesterol.
More than propelling salsalate into the front line of agents used to manage diabetes, it provides clinicians with an additional mechanism of action through which to reduce blood glucose levels. The salsalate research was conducted by the Joslin Diabetes Center, which will continue to investigate this drug’s usefulness in the entire spectrum of cardiovascular disease, including diabetes. However, I think we can anticipate research initiatives into managing diabetes through the anti-inflammatory pathway. A likely avenue would be an oral formulation combining an agent that maximizes native insulin utilization with an anti-inflammatory agent with optimum glucose-lowering effectiveness as well as cardiovascular benefits. This would have multiple advantages. It would help with patient compliance, since the fewer pills the patient has to remember to take, the better. It would address two distinct disease mechanisms – insulin deficiency and inflammation. And it would minimize adverse effects, particularly hypoglycemia. Doc Gumshoe predicts that we won’t have to wait long for such a combination to be developed and marketed, and that it will be a bonanza for the pharma outfit that gets there first.
And one more tiny addendum
Yes, stuff keeps popping up. About a week ago (as I was writing this), a paper was published in the Journal of the American Medical Association (JAMA. 2014 Jun 11;311(22):2288-96) comparing two groups of persons with T2DM who were not attaining acceptable HbA1C levels with metformin alone and required intensification of their drug regimen. This was not a controlled clinical trial, but a large observational study. There were two cohorts, one which added insulin to metformin, and a second which added a sulfonylurea. The main outcome measures were cardiovascular events and all-cause death. While there were no statistical differences in cardiovascular events, all-cause death was about 44% higher in the subjects taking insulin than in those taking a sulfonylurea. The possible causes of this large, significant difference need further investigation, and they could range from baseline differences in the subjects (i.e., the patients who got insulin were sicker to start with), to patients’ problems with managing their insulin dosage, which might be improved with the “bionic pancreas” we discussed earlier. At this point, it’s a red flag. We’ll keep watching.
* * * * * * *
Do please keep the comments coming! They are very important to me, since it lets me know if I’m on the right track sending you stuff that you’re interested in. Right now, I’m planning my next piece – I’m thinking about the placebo effect, which has gotten some attention lately in elevated circles. The question is, is it real? And beyond that, can it be put to use? We’ll see!