Food and Drug Administration: Cardiovascular and Renal Drugs Advisory Committee, 98th Meeting, January 6th-7th, 2003
Cozaar (losartan potassium), an angiotensin AT1 receptor antagonist presently approved for the management of hypertension and nephropathy in Type II diabetics, was proposed by Merck for a new supplemental indication to reduce cardiovascular morbidity and mortality. The application was based on new data arising from the Losartan Intervention For Endpoint Reduction in Hypertension (LIFE) Study, a large, multinational, double-blind, parallel, randomized, active controlled study of 9193 patients with a mean follow-up of 4.8 years. Patients aged between 55 and 80 years with ECG-confirmed left ventricular hypertrophy and blood pressures between 160-200 mm Hg systolic and 95-115 mm Hg diastolic who were off antihypertensive medications were eligible. After a withdrawal period from all antihypertensive medications and a 2-week placebo run-in, patients were randomized to either 50 mg of losartan or 50 mg of atenolol (the active comparator). Subsequently, a titration scheme aimed at achieving similar blood pressure reductions in each group with a target <140/90 mm Hg was undertaken. This involved the introduction of hydrochlorothiazide, uptitration of losartan or atenolol, or the addition of other antihypertensive agents (excluding ACE inhibitors and β-blockers).
At the conclusion of the trial, sitting trough blood pressure measurements were comparable: ie, systolic blood pressure fell by 30.2 mm Hg in the losartan-treated patients and 29.1 mm Hg in the atenolol group, and diastolic pressure fell 16.6 and 16.8 mm Hg, respectively. Pulse pressure was reduced by 13.6 mm Hg with losartan and by 12.4 mm Hg with atenolol (P<0.001).
According to the prespecified analysis, which took into account baseline LVH and the Framingham risk score as covariants, the antihypertensive strategy anchored with losartan produced a 13% relative risk reduction over that achieved by the antihypertensive strategy anchored with atenolol in the primary composite end point, which consisted of cardiovascular death, stroke, and myocardial infarction (10.2-8.9%, P=0.021). Importantly, however, the analysis of the components of the primary composite revealed that the treatment effect was heterogeneous and losartan’s benefit primarily related to a 25% reduction in the risk of stroke (5.7-3.01%, P=0.001).
Amongst the multiple sub groups examined in LIFE were 533 black primarily US patients. A nominally significant qualitative interaction in the outcomes in blacks versus non-blacks was observed with Losartan; the Committee viewed this difference as more than a chance finding given the magnitude of the interaction coupled with prior knowledge of biological differences in black/non-black responses of the renin angiotensin system.
Separately, the committee also noted that the high concomitant use of diuretic therapy in both groups made it difficult to ascertain the individual contributions of atenolol and diuretic to the efficacy of the comparator arm.
After extensive discussion, they accepted the convincing evidence on the reduction of the stroke end point as a meaningful clinical consequence of losartan’s effect beyond that reasonably attributable to the small difference in blood pressure reduction achieved between the two treatment groups. Although they unanimously recommended approval of a claim for establishment of the efficacy of losartan as part of an antihypertensive strategy that usually included a diuretic, they felt strongly that labeling should provide a caveat regarding the qualitatively different effect of losartan-based regimens in black patients and that this claim could not extend beyond stroke reduction to a claim for the reduction of other cardiovascular events.
Carvedilol, a nonselective β-blocker with vasodilating properties currently approved for the treatment of essential hypertension and heart failure was brought forward by Glaxo SmithKline for consideration of a revision in its current labeling as it relates to use in survivors of acute myocardial infarction. The sponsor’s request arose primarily from an evaluation of the Carvedilol Post Infarct Survival Control and LV Dysfunction (CAPRICORN) of 1959 subjects with left ventricular ejection fraction of <40% evaluated within 21 days of an acute myocardial infarction while receiving contemporary, concomitant post-infarct management, including ACE Inhibitors.
The patients were randomized to placebo or carvedilol at 6.25 mg bid; carvedilol was then uptitrated as tolerated to 25 mg bid over several weeks and patients were monitored for a mean of 15 months. Although initially designed with a primary end point of mortality when it commenced in June of 1997, the trial was amended in July of 1999 to incorporate a new coprimary end point of all cause mortality or cardiovascular hospitalization. Conservative analytical adjustments were made requiring a P value of 0.045 as adequate evidence that the drug reduced the incidence of the new end point and 0.005 for demonstration of affect on the original primary end point of all cause mortality. This adjustment to the end point was in response to a recommendation by the Data and Safety Monitoring Board, which took into account slow trial enrollment and a lower-than-anticipated overall mortality as well as the emergence of new data attesting to the benefit of β-blockers in patients with congestive heart failure.
Although the amended trial failed to achieve either its original mortality end point (hazard ratio 0.77, 95% CI, 0.60-0.98, P=0.031) or the new coprimary end point of death or cardiovascular hospitalization (hazard ratio 0.92, 95% CI, 0.80-1.07, P= 0.297), the committee voted unanimously to approve carvedilol for the reduction of mortality in this patient population. Important concerns were raised about the study not meeting its prespecified statistical assumptions. The unanimity of the committee was, however, based in part on the relative consistency of the effect amongst prespecified sub groups and the wealth of other data on the efficacy of β-blockers in reducing mortality amongst post-infarction populations as well as supporting data from the CarvedilOl ProspEctive RaNdomIzed CUmulative Survival (COPERNICUS) Trial, which attested to the efficacy of carvedilol in patients with advanced heart failure, the majority of whom had experienced a prior myocardial infarction. Because of uncertainties associated with recurrent infarction after the index event and issues around the definition of reinfarction, the Committee also unanimously voted to deny the sponsor’s claim for the use of carvedilol for reduction of recurrent myocardial infarction.