Monday, July 30, 2007

Are Insulin Analogs Superior to Regular Human Insulin?

The Institute for Quality and Efficiency in Health Care (IQWiG) is an independent scientific institute based in Germany that evaluates the quality and efficiency of healthcare to evaluate pharmaceuticals, surgical procedures, diagnostic tests, clinical practice guidelines and aspects of disease management programs, following the principles of evidence-based medicine. The Institute's primary goal is to contribute to improvements in healthcare in Germany. The basic idea is for IQWiG to be kind of an independent review board to act as a public watchdog group for Germany's nationalized public health system. They accomplish much of this by reviewing the medical literature to ensure that pseudo-science, which is often sponsored and/or published by drug companies, is not allowed to influence public health decisions paid for by taxpayers unless the science behind it is sound.

In July 2007, IQWiG published a report that concluded that there is currently no evidence available of the superiority of rapid-acting insulin analogs over synthetic "human" insulin in the treatment of adult patients with type 1 diabetes mellitus. This follows an earlier study which reached the same conclusion in patients with type 2 diabetes. The reasons this conclusion was reached is that many of the studies conducted were either too small to be considered statistically reliable, and perhaps more significantly, none of the studies included in their widespread review was blinded, meaning both the patients and the physicians knew which type of insulin was being injected. They concluded that without blinding, there is a danger that patients, knowing their type of insulin, could behave differently (such as testing more frequently, for example), which would subsequently lead to a bias in the results of the studies. The bottom line is that most studies included in their meta-analysis contained serious enough flaws that they were unable to conclude whether insulin analogs were really "superior" to synthetic human insulin in terms of glycemic control as the insulin manufacturers boldly claim.

What was even more remarkable about this particular report was not the conclusion itself, but the fact that it was not the first major review to reach the same conclusion. In 2004 and 2005, The Cochrane Collaboration, which is an international not-for-profit and independent organization dedicated to making up-to-date, accurate information about the effects of healthcare readily available worldwide based in the United Kingdom, reached a strikingly similar conclusion in several reviews. Cochrane serves a similar role for the U.K.'s public health system. Specifically, Cochrane determined that "most studies, 83%, were of poor methodological quality" and that "81% of the studies were sponsored by the insulin analog manufacturers themselves and sponsors were not declared in the remaining 7 studies", in effect, echoing the conclusions reached by IQWiG. Although the Cochrane's Reviews are only accessible with a costly subscription, the Insulin Dependent Diabetes Trust (IDDT) has summarized and documented those studies on its website, which can be found here. Unfortunately, the United States does not have a comparable organization to IQWiG or the Cochrane Collaboration to police pseudo-science in part, because of its lack of a nationalized healthcare system. The American Diabetes Association routinely publishes pseudo-science in its medical journals, including Diabetes, Clinical Diabetes, Diabetes Care and Diabetes Spectrum. Given the ever rising costs of healthcare coverage, its surprising that private insurers in the U.S. do not collaborate in funding third-party research to address these issues in order to ensure they are spending their shareholders' money wisely, but to date, none have emerged. Activist shareholders (such as institutional investors), if they are unhappy with company performance, could theoretically push for change in this situation, but today, there does not seem to be an imperative to do so. As a result, most of these studies are likely to be conducted outside the United States.

Regardless, the widespread adoption of insulin analogs in clinical practice and the fact that they are covered by most insurance companies strongly suggests that much of the healthcare community believes in the use of insulin analogs. But the downsides include complaints of the expense, as well as a somewhat less acknowledged issue of analogs' structural instability. For example, Humalog (insulin lispro rDNA origin) is not currently approved by the FDA for use in insulin pumps (although it is widely prescribed for this purpose off-label) and one reason is because it can crystalize in the tubing (especially if basal rates are relatively low). More subtle examples are the fact that the same dosage does not always yield the same effect day-after-day, leading to more glycemic variability than with older, more stable insulin varieties. Its worth noting that the very definition of an analog is that while its similar in function to insulin, it is different in both structure and origin. Insulin analogs are not insulin and they exist nowhere in nature, and the long-term effects have never been studied. In fact, the longest study done at the time analogs were first approved by the FDA was 3 years (although some manufacturers claim they have longer studies, they have never released their study results). While insulin analogs have been used in clinical practice for a decade now, the more subtle aspects (such as those noted previously) have never been thoroughly investigated. Also, there is some concern that long-term use of analogs may be cancerous, although again, studies behind such claims are lacking.

Many users remain supportive of insulin analogs, largely because of the convenience factor. But newer varieties are in the works now and could spell the end to insulin analogs down the road. Notably, a form of insulin called Viaject that is structurally identical to regular human insulin with some special additives to improve the speed of absorption (see my post here for more details) is now in Phase III clinical trials. Results from the the company's earlier trials were released at the 2007 ADA Scientific Sessions to some fanfare. Similarly, another partnership with Bristol-Myers Squibb and Flamel Technologies (see here for details) is also working on a controlled-release formulation of regular human insulin called Basulin.

As I boldly predicted at the end of 2006, in the not-too-distant future, we may see a return to "Regular" insulin, but with added benefits to match those now found in today's insulin analogs. I am optimistic that these developments could help resolve many of the concerns now seen with today's insulin analogs.


Anonymous said...

I had no idea that Humalog was not FDA approved for pumping! Gosh - how many pumpers do I know (including myself) doing that very thing.

Anonymous said...

The Institute's primary goal is to contribute to improvements in healthcare in Germany.

What a NOVEL idea. Of course, since most Americans believe that we have the #1 healthcare system in the WORLD, why would we even think that we need a watchdog to contribute to improvements in healthcare???


Scott S said...

Sara, doctors can and do prescribe Humalog "off-label" (meaning for a use other than what the FDA has approved), and many users find they have no problems with Humalog in their pumps, so I wouldn't be too nervous about it, although you should be aware of it!

Melody, we're of a very similar mindset on this issue, but I'd like people who read this to consider that sound scientific reviews done abroad are as good as those done here.

Thanks for your comments, both of you!

Anonymous said...


I hope your enthusiasm for the new insulin proves justified, particularly in the sense of an insulin that would be “cannibalized” and released as the blood sugar level increases.

A very smart medical reporter told me one time that insurance companies use high-profile technology products within a certain patient base to go forward and justify requests for rate increases across-the-board for all “insureds.” The costly items are almost eagerly sought and paid for and apply only to a small percentage of the patient-base. In actuality the company-wide increase in rates provides greater profits for the insurer in the long run.