The “most emailed” New York Times article this past week has been Gina Kolata’s “Looking Past Blood Sugar to Survive With Diabetes.”  Given the high prevalence of diabetes that’s no surprise.  But the article’s thesis that a dogmatic fixation on controlling blood sugar may actually compromise other aspects of diabetes care has no doubt piqued much interest.

There are many angles to this topic and diabetes in general – which affects so many people and intersects with so many issues in medicine.  Here’s a quick summary of the article and a consideration of some of the additional questions it raises.

Summary

As Dr. David Nathan, director of the Diabetes Center at Massachusetts General Hospital said:

“… when you think about it, it’s not the diabetes that kills you, it’s the diabetes causing cardiovascular disease that kills you.”

That quote alone probably sums up the main point of the article best. 

Since insulin controls blood sugar and diabetes– at least in its simplified conception – involves a relative lack of insulin, blood sugar (glucose) control has been the cornerstone of diabetes treatment for over a half-century.  However, the fixation (some would even say dogma) on stabilizing blood sugar can divert attention from the fact that controlling cholesterol levels and blood pressure can be as important – perhaps even more important – than rigorous blood glucose control.  In fact, based on recent guidelines, a cholesterol level of 110 which might otherwise have been cause for comfort for a non-diabetic may, in fact, be considered to be too high for a diabetic patient and thus potentially warrant a statin drug regimen.

That’s not to say, of course, that high blood sugar is innocuous.  As confirmed by numerous studies, chronically elevated blood glucose levels can result in the classic diabetes complications of blindness (diabetic retinopathy), nerve damage (diabetic neuropathy) and kidney damage (diabetic nephropathy).   However, studies (such as the DCCT and UKPDS trials) that have tried to link strict blood glucose control with prevention of the disease’s cardiovascular complications have shown no such link or have been equivocal at best.

Let’s put all this in stark, logical terms:

A. Cardiovascular disease caused by diabetes is what kills you (Dr. David Nathan)

B. Large-scale clinical trials have not shown a significant decrease in cardiovascular disease as a result of strict blood sugar control

If statement “A” is true and statement “B” is true, then does it logically make sense that after dietary measures strict blood sugar control is the primary goal in treating patients with Type 2 diabetes?  That logical inconsistency between what we know as fact and what is actually practiced is the underlying thesis of the article “Looking Past Blood Sugar to Survive With Diabetes.”  It’s a powerful argument and one whose message needs to more widely disseminated.

Now for some additional questions and implications …

Prospective (instead of) Retrospective Drug Safety Review

It has been one of the recurring themes of this blog that (1) drug safety is becoming more important and (2) safety needs to be given even higher priority within the drug approval process especially at pre-market stages. 

Currently, apart from routine pre-clinical toxicity studies and limited Phase I clinical trials, the bulk of safety evaluation is left to what is called post-market surveillance.  With the controversies that have arisen with Vioxx (which was withdrawn from the market) and Avandia (which recently received a “black box” warning), it has become abundantly clear that while post-market surveillance can uncover safety problems it does so rather inefficiently with both time and lives being casualties of the process.  The scientific (e.g. statistical) basis of using post-market surveillance is intrinsically inefficient because it is based on individual, retrospective, voluntary case reports.  However carefully individual physicians may report their adverse events, this is still about as flawed a set of data one can get in order to glean insights from.  It is like the proverbial “finding a needle in a haystack” where the needle is revealed only after thousands, if not millions, of patients are treated by the drug. 

Moreover, there is also the problem of inertia.  In addition to the huge populations that have been treated, billions of dollars have likely been spent on marketing and as time passes these drugs become, ironically, ensconced within the standard of care.  It is not at all easy to upend all that.  While some critics (“pharma bashers” come to mind) may cry foul at the perceived greed of large pharmaceutical companies, a simpler explanation would be that of inertia.  Billions of dollars of fixed investment do not simply disappear just like that.  This is no doubt why, with respect to the Avandia controversy, the FDA has been proceeding with careful and deliberate steps.

So what does this have to do with “Looking Past Blood Sugar to Survive With Diabetes.”  If you look at the statement “cardiovascular disease caused by diabetes is what kills you” then you have a recipe for a more involved, directed and intensive pre-market safety study of potential diabetes drugs (such as Avandia).  The main point is that the current paradigm for pre-market safety evaluation focuses less on potential adverse effects (which need to incorporate a complete knowledge of the disease and associated drug indications) and more on toxicity.

Pre-market safety studies involve two major steps, preclinical toxicity studies and limited phase I clinical trials.

1.      Preclinical toxicity studies generally follow a set of formulaic guidelines that proscribe animal or cell extract models to study general toxicity, immunotoxicity (does the drug cause autoimmune effects?), reproductive toxicity (does the drug cause birth defects?), genotoxicity (does the drug damage DNA?) and carcinogenicity (does the drug cause cancer?).  The operative term here is toxicity and indeed few drugs which are frankly (at least at their clinical dosage levels) reach the market.  The problems with Vioxx and Avandia, however, were not problems of toxicity but rather problems with adverse effects.  With diabetes where we categorically know that heart disease is a major problem, then a more expansive evaluation of the cardiac effects of Avandia prior to phase I clinical trials would have been warranted.  Such studies, however, are not required by current safety protocols and would involve more ingenious use of animal models of heart disease and other more innovative approaches to safety evaluation.  Interestingly, these preclinical toxicity studies are typically done without a specific indication in mind.  That is to say the basic safety protocol is the same whether the drug being tested is a anti-cancer agent or a diabetes medication. This generic approach, while convenient and efficient, is needs to be reformed.

2.      Phase I clinical trials are the next step in pre-market safety evaluation.  Once gross toxicity has been ruled out by the preclinical toxicity studies described above, phase I clinical trials are performed on limited numbers of patients in order to establish acceptable dosage parameters.  On an ethical and practical level, the operative term is limited numbers of patients.  One certainly doesn’t want to expose large numbers of patients to a drug whose toxicity profile is incompletely understood.  Hence careful dosing studies are done with as few patients as possible.  Again, since the focus is on toxicity and its associated dosing limitations, there is no way that these trials can demonstrate any of the more subtle (yet significant) cardiac effects that were seen later on, for example, with Vioxx and Avandia.  This approach, too, needs to be reformed. Indeed, given the special susceptibility of diabetes patients to cardiovascular disease, then a more expansive and detailed pre-market approach to safety would have involved Phase I clinical trials that even with the limited numbers of patients involved had used innovative methods such as surrogate endpoints for cardiac effects to try to elicit these adverse effects in a prospective manner among relatively small numbers of trial subjects.

The point is that if we know that cardiovascular disease kills patients with diabetes, then that sort of information needs to be incorporated into the safety evaluation at the very earliest stages of drug development.  Just as the individual patients highlighted in Gina Kolata’s article who did not benefit from this more expansive understanding of the disease, the population of patients who may have been harmed by Avandia also suffer a collective tragedy from that information not being extensively used early on in the drug development process.

The Rising Importance of Convergent (combination) Medical Technologies

Another theme of this blog has been the increasing importance of convergent medical technologies – which are combinations of drugs and devices, devices and software, etc.  The blog entry “Convergent Medical Technology: Part I - What is it?” defines this new area intersecting among the biopharma, medical device, nanotechnology and IT sectors.  The concept of combination medical products involving combinations of drugs is a subset of this wider trend towards technology convergence and combination. 

Yet another interesting feature of the New York Times article (p.2) was that beyond blood sugar control, the treatment of cardiovascular disease per se is critical to diabetes care, complex drug regimens including statins (cholesterol lowering agents), antihypertensives in conjunction with the usual insulin or oral hypoglycemic (anti-diabetic) drugs are becoming more commonplace.   As the article writes:

Before he left the hospital, Mr. Smith’s doctors told him about his new diabetes regimen: a statin to drive his cholesterol level very low, two drugs to lower his blood pressure, an aspirin, insulin and two drugs to reduce his blood sugar levels. That new list of drugs was what he should have been taking all along.

…

“Right now, without waiting for lots of exciting things that are almost in the pipeline or in the pipeline, starting tomorrow, if everyone did these things — taking a statin, taking a blood pressure medication, and maybe taking an aspirin — you would reduce the heart attack rate by half.”

The article’s last section is entitled “The Burnout” which highlights how quite literally exhaustion and complexity can lead patients to abandon the rigorous regimen required to fully manage their diabetes.  Physicians call this “noncompliance.”  Patient noncompliance – for whatever the reasons involved – can be fatal.

To this end then, a series of new combination drugs – also known as “polypills” that include several of the required medications within one tablet can simplify compliance and minimize errors.  Ultimately because so much of this depends on patient and physician education, such a complex regimen when delivered within a single combination form (or polypill) becomes simple.   Combination drug therapies, taking off-the-shelf drugs in new formulations, may not be glamorous and “cutting-edge” but they can truly play a role in saving lives.

Is Diabetes a Surgical Disease?

A previous blog entry “The Diabetes Divide: Is Diabetes a Surgical or Medical Disease?” was titled specifically to pique interest and perhaps generate a bit of controversy. 

It is common knowledge that diabetes is a medical disease – namely that drug therapy, whether it be via insulin or other medications, is the primary approach to therapy.  Of course, there are surgical aspects to diabetes.  Amputations, for example, are a dreaded complication.  The main point of that article was that a reevaluation of the generalized (e.g. medication-based) approach to systemic blood sugar control may be in order. 

The New York Times article also touches upon this – not the issue of surgery for diabetes per se but the challenges involved in fully restoring the natural physiology of glucose control.  As Gina Kolata writes:

“But no matter how carefully patients try to control their blood sugar, they can never get it perfect — no drugs can substitute for the body’s normal sugar regulation.” 

It is important to consider exactly what is meant by blood sugar control in the current practice of diabetes.  Blood sugar levels are typically sampled from the systemic venous circulation (basically the infamous finger “prick” or, if in a doctor’s office or hospital, a full venous sample).  Dosages of insulin and/or oral hypoglycemics are then titrated to this systemic venous blood glucose level.  As outlined in the blog article, the actual anatomy of insulin regulation and secretion involves a localized distribution and flow of insulin.  Insulin is first secreted within the pancreatic portal circulation which then drains into the hepatic portal (post-intestinal) circulation.  This portal venous insulin exacts its effects first on the liver and then on the systemic circulations.  This implies (and this is significant) that blood insulin levels and blood sugar levels are not evenly distributed throughout the body.  This is what is ultimately meant by Gina Kolata’s comment that “no drugs can substitute for the body’s normal sugar regulation.”  There are other subtleties to this such as the unique circulation within the pancreatic Islets of Langerhans and various neuro-GI networks which also put into question the long-term effectiveness of blood glucose control via systemic, generalized administration of insulin.

Indeed, when a process is unevenly distributed throughout the body – such as a broken leg – the approach to the problem is not a generalized medical approach but rather a localized, surgical approach.  It would not be reasonable, for example, to imagine that a drug would “cure” a broken leg.  Such a problem requires setting the bone (“reducing the fracture” is the technical term) and casting at the site of injury.  Of course, drugs can be used to ameliorate pain or prevent infection (life-saving in their own regard) but they do not fundamentally solve the problem.

To the extent that diabetes involves local pathology and local effects of glucose regulation, this is inspiration for the idea that next-generation therapies for diabetes may ultimately involve more of surgical approaches.  Some may argue that there is no need for that.  They would say that we definitely know that the problem lies with a relative lack of insulin and that we have identified the culprit molecule – insulin - and simply need to substitute for that the relative lack of that molecule.  To that I would reply: why have we not yet cured the disease?

I congratulate Gina Kolata on a great article on an important topic.  A great article often raises more questions than answers and I found that “Looking Past Blood Sugar to Survive With Diabetes?” has indeed done that for me.  Hopefully this blog article will also create more questions for you as well.

Ogan Gurel, MD MPhil
gurel@aesisgroup.com
http://blog.aesisgroup.com/

Avandia Diabetes blood sugar surgery FDA reform drug safety Gina Kolata convergent medical technologies combination medical products high cholesterol hypertension David Nathan Massachusetts General Hospital Aesis Research Group Ogan Gurel