Innovation, statistics, and truth

The New England Journal of Medicine had a statement that caught my attention that bore on the topic started on this post and this post about safety in big pharma and food.

…For example, we know that in type 1 diabetes, metabolic control can reduce the risk of microvascular complications. On the other hand, the two largest randomized, placebo-controlled trials in patients with type 2 diabetes, the United Kingdom Prospective Diabetes Study and the University Group Diabetes Program, failed to find a significant reduction in cardiovascular events even with excellent glucose control.2 Moreover, we are facing a troubling paradox: preliminary data that were presented at the meeting and published by Gerrits et al.3 suggest that among the thiazolidinediones — a class of drugs that has been shown to improve metabolic control — rosiglitazone may increase cardiovascular risk whereas pioglitazone may reduce it. Until we have a better grasp of the pathogenesis of atherosclerosis in type 2 diabetes, it will be difficult to design therapies to prevent this complication or even to determine how the currently available agents that act at multiple sites may affect clinical outcomes in very different ways.

Not surprisingly, glycemic control has been the centerpiece of therapeutic interventions in type 2 diabetes for many years. Within the past decade, several new drugs that result in “glycemic durability” — a reduction in the glycated hemoglobin level — have been approved by the FDA. However, change in the glycated hemoglobin level is a relatively poor surrogate for cardiovascular outcomes in type 2 diabetes, accounting for only 5 to 15% of the variation in ischemic risk.2 This finding represents a major dilemma for practitioners, regulatory agencies, and patients who seek the newest and best treatments for this disease.

The controversy surrounding biochemical surrogates versus clinical outcomes was also highlighted at the FDA meeting when the advisory committee reviewed one of the largest randomized trials of rosiglitazone, A Diabetes Outcome Prevention Trial, or ADOPT. In that study, the percentage decrease in glycated hemoglobin was greater with rosiglitazone than with metformin or sulfonylureas, yet the risks of congestive heart failure and cardiovascular ischemia were higher.4 These data suggest that we urgently need to change the regulatory pathway for drugs for the treatment of type 2 diabetes to make clinical outcomes, not surrogates, the primary end points. This is not a radical proposal: 20 years ago, the FDA shifted its primary efficacy end point for osteoporosis drugs from bone mineral density (a reasonable surrogate for the risk of fracture) to fractures themselves. Without a regulatory sea change with regard to diabetes drugs, we are certain to be in the same position 5 years from now that we are in now: we will again find ourselves in possession of a new wonder drug that is designed to treat a devastating chronic disease but that may do more harm than good.

Drugs are approved or removed from the market on the basis of evidence from randomized, controlled trials. In the FDA hearing on rosiglitazone, several meta-analyses (see Table 1) revealed a significant increase in the risk of myocardial ischemic events among patients taking rosiglitazone. However, an interim analysis of the ongoing Rosiglitazone Evaluated for Cardiac Outcomes and Regulation of Glycaemia in Diabetes (RECORD) trial, which was designed specifically to assess cardiovascular risk among patients receiving rosiglitazone, failed to demonstrate a similar risk.5 In addition, two large observational studies, one conducted by Tricare for the Department of Defense and one conducted by WellPoint (the largest health insurer in the United States), noted no appreciable signal of increased cardiovascular risk with either of the available thiazolidinediones (see Table 2). The contrasts among the levels of evidence and the results regarding the safety of rosiglitazone raised new questions about relative and absolute risks but also highlighted the weaknesses of observational studies examining events that are common and whose rates are likely to be increased only slightly by a given drug, even in a large cohort (such as that used by WellPoint, which comprised 160,000 patient records).

This issue led to an interesting sidelight at the meeting. Recently, there have been calls for the FDA to fund and oversee phase 4 postmarketing studies as a means of determining the safety of newly approved drugs. The two observational studies on thiazolidinediones whose results were presented at our meeting had been performed independently of the manufacturers and the FDA. The indeterminacy of their results, due to the inevitable effects of the many confounding variables inherent in such studies, illustrates why this approach alone will neither solve the overriding problems of drug safety nor ultimately help a chronically underfunded federal agency. There is no doubt that it will be costly to undertake true safety and efficacy studies of new drugs using clinical outcomes as primary measures, but in the long run, these efforts will save time, energy, and money.

Such reading is difficult for non-medical, but the thought is clear. The same theme is presented that proper testing is severely deficient and apt to get false negatives for safety as part of the design. It is unlikely that companies would furnish or finish appropriate testing for followup safety concerns that have NO longitudinal studies done. There is a growing share of the market that is tweaking current medications for the sake of patent production and sales but reveal little improvement in efficacy and actual concern of harm.

The gardasil vaccine is innovative and has been approved of course, but the immense pressure to get to market before a rival company predisposes companies to push increasingly more quickly and with greater ‘advertising’ (a governor advocating mandatory vaccination!) that turns the market into a medical experiment of large proportion.

(Next stop is in India where generic antibiotics are made to lower our costs for drugs)