October 29, 2013
8 min read

Coming to Terms with SYMPLICITY HTN-3: Renal Denervation Fails to Reduce Primary Efficacy Endpoint

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

In what was arguably the first big news story in cardiology in 2014, Medtronic announced in a press release that SYMPLICITY HTN-3, the highly anticipated pivotal trial assessing the company’s Symplicity renal denervation catheter in patients with severe treatment-resistant hypertension, failed to meet the primary efficacy endpoint of change in office systolic BP from baseline to 6 months.

Although found safe by the Data Safety Monitoring Board, as the study met the safety endpoint of major adverse events 1 month following randomization and renal artery stenosis to 6 months, the news that the first sham-controlled randomized trial on the strategy missed the efficacy mark has sent a shockwave through the specialty, because many cardiologists had heralded renal denervation as the next major breakthrough technology.

As we await the presentation of the data, which is scheduled for the upcoming American College of Cardiology Scientific Sessions at the end of March, Cardiology Today’s Intervention spoke with two experts to get their first impressions on these landmark results: one a US-based endovascular specialist who was involved in the SYMPLICITY HTN-3 trial — Krishna J. Rocha-Singh, MD, with Prairie Heart Institute at St. John’s Hospital, Springfield, Ill. — and the other an interventional cardiologist from Europe who was not involved in the trial — Horst E. Sievert, MD, with the CardioVascular Center Frankfurt in Germany.

Horst E. Sievert, MD

Horst E. Sievert, MD 

Horst E. Sievert

As we await more details from SYMPLICITY HTN-3, it’s important to note that renal denervation did not come out of the blue. It has been shown to be effective in a number of animal models, in surgical experience for decades and in clinical practice. Effects on left ventricular hypertrophy and diastolic function have also been observed and can hardly be explained by a placebo effect.We all have seen patients with severe long-lasting resistant hypertension who became normotensive after denervation.

On the other hand, we have also noticed that the effect is often not as dramatic in clinical practice as in clinical trials. The responder rate seemed to be lower and the mean decrease of BP was less impressive. SYMPLICITY HTN-3 was a very rigorous randomized trial that included a sham procedure. So it could well be that almost everything we have seen before was a placebo effect.

But there may be other reasons as well. SYMPLICITY HTN-3 required optimal medical treatment and mandated adjustment of medications within the weeks before the trial. The centers had been under some pressure to enroll patients, so this titration period may have been rather short in many patients. However, it may take weeks to months to experience the full effects of antihypertensive medications. After the titration period, only 2 weeks of stable medication were required, which means that both treatment groups started during a phase when their BP was decreasing (due to the titration of medications). This may have reduced the treatment effect. The investigators should report when and how medications had been changed before the 2 weeks of stable medication period.

There is also the matter of patient selection and the question of whether North Americans are different, or perhaps their lifestyle and medication use are different. In SYMPLICITY HTN-3, patients had been selected “very carefully” — may be too carefully. It could also be that “really resistant” patients are also resistant to renal denervation, and that actually patients who had not been treated in such an optimal way as in SYMPLICITY HTN-3 or who are not fully compliant to medical treatment respond better.

Other reasons that could have led to the failure of renal denervation to reduce the efficacy endpoint include:

  • Too high expectations: It may be that the effect is much smaller than expected, so the study was under-powered.
  • Too short follow-up: We know from other renal denervation trials that there had been many “late-responders” beyond 6 months.
  • Technique: Perhaps the current technology using radiofrequency was not good enough.
  • There may also be, as some people have argued, a learning curve.

In terms of how these results may affect use of renal denervation in Europe, I can only speak to its potential impact in Germany. In the country, health care insurance companies will use the Medtronic press release to deny payment for renal denervation — which is, of course, unacceptable. But these days they use any opportunity. Referring physicians so far have not changed their referral pattern. I have received many phone calls. Most said they hope we will continue with renal denervation.

At this point, I think it would be a mistake to just close the book on renal denervation for resistant hypertension. I have received some emails from colleagues who had always been against renal denervation, and the messages give me the impression they feel happy looking at the press release. This is the wrong attitude to have! We are talking about patients who are suffering a very high CV risk and do not have a treatment option.

It is our obligation to do more research and find out who benefits from renal denervation and who does not, and to look into other technologies. We have good reason to assume that technologies delivering energy more deeply into the tissue may work better. Several other companies have already announced that they will continue with their programs and we should support these attempts. Future trials should be designed very carefully. For example, the “big day” and the “let’s measure once more” effects should be taken into account. Whether this requires more sham-controlled trials or just more objective parameters can be debated.

Disclosure: Sievert reports receiving consulting fees, study honoraria and/or travel expenses from Abbott, Access Closure, AGA, Angiomed, Aptus, Atrium, Avinger, Bard, Boston Scientific, Cardiac Dimensions, CardioKinetix, CardioMEMS, Coherex, Contego, CSI, CVRx, EndoCross, ev3, FlowCardia, Gore, Guided Delivery Systems, InSeal Medical, Lumen Biomedical, HLT, Lifetech, Lutonix, Maya Medical, Medtronic, NDC, Occlutech, Osprey, Ostial, PendraCare, pfm Medical, ReCor, ResMed, Rox Medical, SentreHeart, Specranectics, SquareOne, Trireme, Trivascular, Venus Medical, Veryan and Vessix; stock options from Access Closure, Cardiokinetix, Coherex, Lumen Biomedical and SMT; and grant research support from Cook Medical and St. Jude Medical.

Krishna J. Rocha-Singh, MD

Krishna J. Rocha-Singh, MD 

Krishna J. Rocha-Singh

Currently, all we know is what was in the Medtronic press release, so everything is conjecture at this point. That said, I was personally very surprised by the trial results. I, like the other researchers involved in the study, knew the results from the SYMPLICITY HTN-1 and HTN-2 trials and am now wondering what went wrong.

For now, I can only hypothesize as to what caused the renal denervation treatment arm to fail to meet the primary efficacy endpoint. The first possible reason may be the Symplicity Flex device itself. Did the catheter perform as it did in pigs and were the translational models appropriate? When we perform renal denervation, we operate in a “black box,” as Felix Mahfoud, MD, has noted. We receive no intra-procedural, real-time feedback as to the potential adequacy of the anatomic denervation. There will soon be data correlating the number of ablation sites per renal artery and subsequent office BP effect. In preliminary data that Dr. Mahfoud has presented at conferences, which has not been through peer review, there does not appear to be a correlation between the number of ablation sites above four to six and office BP response. Still, the only predictor of a favorable BP response based on the analyses of SYMPLICITY HTN-1 and HTN-2 is the baseline office BP: the higher the office BP, the more probable the response.

This leads into the second possible reason, which may have been the procedure. Sometimes you get an “error code 50,” and we don’t know what that means other than an inadequate ablation has occurred.

Third, which I suspect may be the biggest potential reason, is the trial design and the subsequent patient selection. When we enrolled in SYMPLICITY HTN-1, we started out conducting a European study that had the prerequisite of stable medical therapy on three or more medications, one being a diuretic. There were some sites where diuretic use was not mandated because the operators didn’t think the patients would be compliant; however, these patients were on stable and compliant medical therapy. So in SYMPLICITY HTN-3, stable medical therapy morphed into maximal tolerable, optimal medical therapy, where there was more diligence in making sure that patients were on maximal medication doses, one being a diuretic, but not just any diuretic, a thiazide diuretic, the use of which was highly encouraged. In general, patients wouldn’t be enrolled if they didn’t have this “optimal” diuretic or reasons for their intolerance provided. So these are very subtle nuances that evolved between the trial designs, which may have subsequently defined a different patient population. In the US pivotal trial, there was a prerequisite that the demographics of the trial reflect the US population. Although I don’t know what percentage of patients were African Americans, we do know that in this population, the responsiveness to specific medical therapies is different. We had no inkling from the European data as to what the response of renal denervation was in this cohort of patients. However, this may not even matter, but it is an additional variable.

Also, if you did get a patient qualified for the SYMPLICITY HTN-3 trial in the United States, they had to have stable BP for at least 2 weeks. The question then becomes was 2 weeks sufficient? In SYMPLICITY HTN-1 and HTN-2, I suspect the vast majority of patients were recruited from European Hypertension Centers of Excellence; these patients may have been “no options” patients, managed by a team of specialists and on a stable and consistent medication regimen, for at least 6 weeks prior to screening. So, this may all fall into the whole issue of a different trial design, and if so, I think this is a good thing. Physicians want to know the population renal denervation works best in and you only get that assessment from a large, robust trial design. Another reason for the failure of SYMPLICITY HTN-3 could have been clinical indication. There may be issues with the heterogeneity of the hypertensive population and the associated problems, the lack of consensus with regards to ideal treatment, concerns regarding patient medication adherence, and the simple fact that patients who get placed into clinical trials typically have a certain degree of favorable BP response; the potential impact of all these variables must be considered in future trial designs.

Finally, the sham-control arm may have performed better than anyone anticipated. There was a lot of debate over the need for a sham-control arm and, frankly, Medtronic was, and has been proven, correct for including it as part of the study design. We also know from SYMPLICITY HTN-3 results that renal denervation is safe, which is an important conclusion, as the device is currently being used in many countries outside the United States to treat other indications including HF and cardiac arrhythmias.

Even in light of the failed primary efficacy endpoint, it is still vitally important that we continue to investigate renal denervation. There should not be collateral damage from this specific device to other devices — whether monopolar, bipolor or ultrasound, or neurotoxins, because we cannot assume that they won’t be effective based on the results of this specific device and trial design. Nor do I believe that the device iteration or the denervation field is dead. This is just the ‘end of the beginning.’ Over the years, there have been a number of failures with first iterations and trials. If industry and physicians had walked away after their first regulatory failure, for example, we wouldn’t have stenting in the superficial femoral artery. In time, I think we will look back upon the SYMPLICITY HTN-3 trial as another example of “failing forward”: the trial may have failed, but as we learn from this, we will move the field forward.

Disclosure: Rocha-Singh is a paid consultant for Cardiosonic, Medtronic and Vessix/Boston Scientific. He was an investigator in the SYMPLICITY HTN-1 and HTN-3 trials.