ROADMAP studie gepresenteerd en FDA safety review

Nieuws - 23 juni 2010

FDA safety review olmesartan


Discussie over de vraag of de FDA safety review van olmesartan van afgelopen week iets is om zorgen over te maken, volgend op de presentatie tijdens het ESH congres deze week van de grootste van twee studies geciteerd in die review, de ROADMAP studie.

Oslo, Norway - Hypertension experts attending the European Society of Hypertension (ESH) European Meeting on Hypertension 2010 this week did not appear to be particularly concerned about the recently announced FDA safety review of the angiotensin receptor blocker (ARB) olmesartan (Benicar, Daiichi Sankyo). Discussion followed the presentation at the meeting of the largest of the two trials cited in that safety review, the Randomized Olmesartan and Diabetes Microalbuminuria Prevention (ROADMAP) study, reported by Dr Hermann Haller (Hannover Medical School, Germany).


Concern centers on an excess risk of cardiac death seen in the olmesartan arm of ROADMAP and another smaller study, Olmesartan Reducing Incidence of End Stage Renal Disease in Diabetic Nephropathy Trial (ORIENT), although the FDA did still conclude that the benefits of the drug "outweigh its potential risks." Most polled by heartwire felt the findings were likely due to chance—given that olmesartan has been on the market in the US since 2002, so any real danger signal would have become apparent by now—or could be explained by the so-called J-curve effect, whereby in certain patients lowering blood pressure to too great an extent can be harmful. But they note that that the lack of any prior large outcomes trials with this drug has done it no favors when it comes to assessing safety.

Presenting ROADMAP, Haller maintained that the number of CV deaths "was few"—15 in the olmesartan arm compared with three in the placebo group—considering that there were almost 5000 participants in the trial and it lasted nearly four years. He added that this was a post hoc finding and that the steering committee "considers these numbers too small to draw any firm conclusions."


Nevertheless, these fatal CV events were more frequent in the olmesartan group, he acknowledged; further analysis of the study reveals that the increase in mortality was seen only in those with preexisting cardiovascular disease, and "most of it occurred in the quartile where the systolic BP was low, below 120 mm Hg," he explained. So this "supports the whole notion that 'the lower, the better' is not true in all patients. It seems that when you have coronary artery disease, BP should not go below 110/70. The J-curve phenomenon is supported by the findings of ROADMAP," he told heartwire. He said the FDA had no choice but to look into this because the outcomes of the study were released "without an analysis why," but "it's clear-cut, it's not a problem of olmesartan, we are sure about that," he noted.

Signal not strong enough to cause real alarm about olmesartan

Olmesartan has been quietly building up market share among ARBs, with worldwide sales of more than $2 billion last year, almost half of which were in the US. But prior to the ROADMAP study, there had been no large outcomes trials with the drug, and this is likely one of the reasons the finding of excess cardiac deaths in the active treatment arm of this trial has prompted some concern. 

"If we had at least one outcomes study showing morbidity and mortality benefit, we would probably ignore this, but since there is none, there remains a distinct question mark," Dr Franz Messerli (St Luke Presbyterian Hospital, New York) told heartwire.

Yet Messerli says this will not deter him from prescribing the drug: "I probably will continue to use it because this signal is not strong enough. It's a good antihypertensive drug, one of the most powerful ARBs that we have at the present time, so yes, I will of course be very careful, but I will continue to use it."

American Society of Hypertension (ASH) president Dr George Bakris (University of Chicago Pritzer School of Medicine, IL), who was a cochair at the session where ROADMAP was presented, agreed that it was concerning that more outcomes studies had not been done with olmesartan: "It's unfortunate that they have not committed to doing more studies."


But he also stands by the drug: "If there were a real effect with olmesartan in the general population, you would have heard about it, because physicians do report things, especially things that are bad, and that hasn't been borne out." He added: "I think the FDA right now is under a lot of pressure from Congress to be ever-vigilant, so I think they are just being extra careful."

The other co-chair of the session, Dr Björn Dahlöf (Sahlgrenska University Hospital, Ostra, Sweden), said he wasn't perturbed by the safety review of olmesartan. "It is most likely a chance finding," he told heartwire.

And ESH secretary Dr Peter Nilsson (Lund University, Sweden) said: "I agree with Prof Haller that the numbers were very low; on the other hand, this should be respected as an important finding, in a way." Nilsson said these findings would not put him off using olmesartan "from a scientific point of view," but from a financial point of view, "losartan has gone off patent, so in every country for every clinician, this is perhaps the [preferred] ARB because of the price."

Which diabetic patients would benefit most from olmesartan?

ROADMAP is the first large outcomes trial with olmesartan; it was conducted in 4447 patients with type 2 diabetes and one or more additional cardiovascular risk factors, but no evidence of microalbuminuria. Participants could have normal blood pressure or well-controlled hypertension. The primary end point was a renal one, time to onset of albuminuria. Patients were randomized to either 40 mg of olmesartan (n=2232) or placebo (n=2215) daily and were all allowed to take additional non-renin-angiotensin system (RAS) antihypertensive medications to reach target BP (<130/80 mm Hg), until the predefined number of adjudicated microalbuminuria events occurred at a median follow-up of 3.2 years. The average baseline BP of participants was 137/80 mm Hg.

Haller explained that previous studies have been conflicting on the issue of whether ARBs can delay the onset of microalbuminuria in normotensive or prehypertensive diabetics.

With this excellent BP control, would we see an effect of olmesartan on our primary end point? We did.

"We achieved very good blood-pressure control," he noted, with almost 80% of patients in the study reaching 130/80 mm Hg or less. "The question was, with this excellent BP control, would we see an effect of olmesartan on our primary end point? We did." There was a significant 23% reduction in the time to onset of microalbuminuria in those taking olmesartan (risk ratio 0.77; p=0.0104). And the majority of this effect of olmesartan was BP-independent, said Haller.

But it's obvious that in the overall patient population, although the difference is significant, it's small, he said, so the question becomes which diabetic patients should be targeted. Subgroup analyses suggest those with high systolic blood pressure (above 138 mm Hg), a low estimated glomerular filtration rate (eGFR) of under 90 mL/min, or those who have the first trace of albumin in the urine, even before microalbuminuria; the latter is especially important, he said. When these three things are taken into account, the number of patients who will profit will go up, Haller noted.

Microalbuminuria is a risk factor for CVD

In terms of the cardiovascular deaths, there were seven cases of sudden death, five fatal MIs, two fatal strokes, and one death related to coronary revascularization in the olmesartan group, compared with a total of three CV deaths—one sudden death and two fatal strokes—in the control group.

"All the excess cardiovascular mortality seen in the olmesartan group was in those with preexisting cardiovascular disease, and we think was due to hypotension. We also have the time course of when this occurred, so [olmesartan] is not a danger for patients, not at all. It's a question of not treating patients with diabetes and CAD to very low blood pressures," he stressed, noting that this fits with the European guidelines, which state that systolic blood pressure in patients with cardiovascular disease should be between 130 and 140 mm Hg.


Session chair Bakris said that "in terms of kidney progression, microalbuminuria is highly variable, and it probably isn't a valid end point for renal outcomes. But it is thought to be a cardiovascular risk marker." It would have been "nice" if the ROADMAP investigators had had longer follow-up and bigger numbers for the cardiovascular outcomes, he said.

Haller replied that that microalbuminuria "is relevant. . . . Although in diabetic nephropathy there are variable outcomes, it is the first indication of microvascular disease in an individual patient." It is "an individual risk factor" and is very strongly associated with future cardiovascular events and relatively easy to measure, so it's number one of all the measurements for organ damage, he noted.


Haller has received honoraria from Daiichi-Sankyo. Messerli has received grant/research support and speaker fees from Daiichi-Sankyo. Bakris disclosed being a consultant or advisor to Abbott, Merck, Gilead, Forest, Novartis, Takeda, CVRx, Boehringer Ingelheim, Servier, and the FDA. He has received grant or research support from GlaxoSmithKline and Forest and has served on the speakers' bureaus for Novartis and Forest. Dahlöf has received payment for consultancy or board membership from MSD, Novartis, Boehringer Ingelheim, Daiichi-Sankyo, and Pfizer and has received speaker fees from all these companies. 

Deel deze pagina met collega's en vrienden: