In the present era of almost daily “landmark trial” publications, the literature this past week took a slightly more introspective turn. Two separate journals turned the spotlight back on the uneasy relationship between commerce and science – and well written and thoughtful editorials accompany each.
An analysis in the week before last’s BMJ called attention to the increasing trend toward designing trials with composite endpoints – and suggested that this tendency is potentially misleading. In a sample of 114 cardiovascular trials with composite primary end points, mortality was virtually always a part of the composite but typically contributed the fewest events and reflected the smallest treatment effect. When that happens, it creates “an important and plausible risk” that the study’s conclusions will mislead readers to “attribute reductions in mortality to interventions that do not, in fact, reduce death rates,” according to the authors, led by Dr Ignacio Ferreira-González (Universitat Autònoma de Barcelona, Spain). This latest analysis adds an interesting tidbit to the growing debate about how close commercial interests are to the heart of medical research – the accompanying editorial by Nick Freemantle (who took on this issue before with the PROACTIV trial) is worth reading. BMJ Link
The Lancet features an interesting letter in the Correspondence section commenting on the above, in specific regards to the ASSENT trial from last year – and presents both sides of the issue. In this issue there is a nice review on the importance of autopsy in medical education – so if you need a reminder for why it is important to ask, it’s worth reading. Lancet Link
In keeping with the theme – a survey in last weeks NEJM found that most physicians (94%) reported some type of relationship with the pharmaceutical industry, and most of these relationships involved receiving food in the workplace (83%) or receiving drug samples (78%). More than one third of the respondents (35%) received reimbursement for costs associated with professional meetings or continuing medical education, and more than one quarter (28%) received payments for consulting, giving lectures, or enrolling patients in trials. Cardiologists were more than twice as likely as family practitioners to receive payments. Family practitioners met more frequently with industry representatives than did physicians in other specialties, and physicians in solo, two-person, or group practices met more frequently with industry representatives than did physicians practicing in hospitals and clinics. Three separate articles in the Perspective Section present a balanced account of the Prescription Drug User Fee Act (now up for its q5 year renewal) and call attention to the relationship between the FDA and the pharmaceutical industry. NEJM Link.
Investigators in a systematic review for the second issue of 2007 in the Cochrane Library, which was published online, found that the newer long-acting insulin analogs seem only marginally better at preventing nighttime hypoglycemia than traditional NPH insulin, and neither insulin glargine (Lantus) nor insulin detemir (Levemir) were superior to NPH insulin at protecting against diabetes-related morbidity and mortality or improving quality of life. The 8 studies that comprise this review were mediocre – but suggested that the real effect of the newer basal insulins is directly comparable to NPH in terms of long-term metabolic control AND episodes of severe hypoglycemia – with only a slight reduction in episodes of nocturnal hypoglycemia. Better quality data would be more useful for making the decision about which basal insulin to use, but given the BIG difference in cost, this review certainly puts a damper on the enthusiasm for the newer insulins – and will certainly come to bear in formulary decisions. Cochrane Link
And back to the weekly landmark trials…
An analysis from the The Clopidogrel in Unstable Angina to Prevent Recurrent Ischemic Events in Patients Undergoing Percutaneous Coronary Intervention (PCI-CURE) trial database concluded that the adjusted risk of major bleeding within 8 months after the procedure was 2.2-fold greater in patients on aspirin at a dosage of at least 200 mg/day than in those on 100 mg/day or less – and with similar efficacy in terms of death, MI, and stroke. This observation has popped up a few times in other studies and has prompted a larger prospective randomized trial – the OASIS-7/CURRENT. Skeptics make mention of special populations to whom this data may not pertain (eg diabetics and those with a putative aspirin resistance –>for decent editorial on the latter see BMJ 2004;328:477-479) . This data was presented as an abstract and has not been subjected to a peer-reviewed process.
The MERLIN-TIMI 36 investigated the effects of ranolazine in non-ST elevation acute coronary syndromes. Ranolazine demonstrated a non-significant trend toward lowering the COMPOSITE end point of cardiovascular death, MI, or recurrent ischemia (see above). Cardiovascular death and MI were virtually identical between the two groups. Recurrent ischemia was decreased by a statistically significant margin – and this accounted for the small difference seen in the primary end-point. JAMA Link
-Sean Cavanaugh, MD, Associate Editor Clinical Correlations