Drug-eluting Stents Present Headaches for Regulators

Drug-eluting stents have proved remarkably effective at addressing the shortcomings of first-generation bare metal stents, but they are not without problems of their own. Peter Charlish reports.

Peter Charlish is a contributing editor to The Regulatory Affairs Journals.

Drug-eluting stents (DESs) in 2001 emerged as the magic bullet for preventing once-narrowed arteries from becoming re-clogged. While there are dissenters in the field of cardiology, the devices have produced extraordinary results in company-sponsored clinical trials and experienced a phenomenally rapid adoption by cardiologists worldwide. It now appears that the devices might be unsafe, as reports have fuelled fears that they may be linked with thrombosis. These concerns have led to scrutiny by US regulators and a drop in DES use as physicians revert to the technology's older bare metal stent (BMS) counterparts.

At the beginning of the 21st century, coronary heart disease (CHD) remains the world's biggest cause of death. According to the World Health Organization (WHO), 3.8 million men and 3.4 million women worldwide die each year from CHD, and since 1990 more people have died as a result of CHD than of any other cause¹.

Although genetic factors play a part, the majority of people dying from CHD have one or more risk factors that are associated with lifestyle. In fact, CHD is a relatively recent phenomenon compared with related conditions such as stroke, reflecting changes in personal mores. But while this situation has been recognised and is being addressed in developed countries, where mortality from CHD is falling, the same is not the case elsewhere and the WHO predicts that 82% of the future increase in CHD mortality will occur in developing countries.

Aside from drug therapy, treatment of CHD has largely been directed at re-establishing adequate perfusion of the affected areas of the heart muscle. Since the late 1960s, this has meant surgical revascularisation via coronary artery bypass grafting, where the obstructed coronary arteries are bypassed either with a section of vein removed from the patient's leg (saphenous vein graft) or with a graft obtained from the internal mammary artery. Sometimes, endarterectomy, the removal of an atheromatous obstruction within the lumen of the artery, is performed at the same time.

Coronary bypass is a highly invasive procedure, and efforts to find a less invasive solution to CHD resulted in the development of balloon angioplasty, or percutaneous transluminal coronary angioplasty (PTCA). The procedure, first performed in 1977, involves passing a catheter with a deflated balloon at its tip into the affected artery, and then rapidly inflating the balloon. This compresses and fractures the atheromatous plaque leading to the restoration of blood flow. The catheter is introduced into a peripheral artery, typically the femoral artery but sometimes the radial artery, and advanced to the blocked vessel. This procedure is obviously much less invasive than open-chest surgery and is associated with much more rapid recovery and a lower complication rate. It is also considerably less expensive than bypass surgery.

Over the years, PTCA has proved a highly effective intervention in CHD patients, including in patients who are poor candidates for surgery, typically resulting in a reduction of symptoms such as angina and dyspnoea, a decreased risk of heart attack, and a decreased need for additional medication such as anti-anginals. However, the procedure is not without risk. The main short-term complication is abrupt vessel closure, which occurs when the treated vessel becomes blocked. It typically occurs within the first 24 hours after the procedure. Although relatively uncommon (approximately 3%-5% of cases), it can lead to complications such as death, myocardial infarction, and emergency coronary artery bypass.

A second, longer-term problem with PTCA is that around 30% of coronary arteries treated gradually close up again, a process referred to as restenosis. This occurs mainly in the first six months and is caused partly by the walls of the vessel recoiling against the dilatation, and partly by smooth muscle cell proliferation and fibrointimal hyperplasia, a process referred to as neointimal proliferation. Neointimal proliferation is a response to injury caused during the dilatation, and can occur as a complication of all forms of arterial reconstruction.

A number of approaches to this problem have been investigated, including the use of alternatives to balloon angioplasty such as rotational atherectomy (using a rotating cutting device to remove plaque), and endarterectomy using an excimer laser. Most of these have now been abandoned, partly because they are difficult to use, especially on more distal stenosis, and because they are expensive. One approach that had been successfully used in other areas of surgery where it was necessary to keep the lumen of a tubular structure open (for example the oesophagus or a ureter) had been the use of a stent, a wire metal mesh tube that can be collapsed to a small diameter for insertion and then expanded to lock into place and act as a scaffold. In the case of PTCA, it was necessary to create a device small enough to be inserted into a coronary artery mounted on a balloon catheter, whereupon inflation of the balloon would simultaneously push back the plaque and expand the stent.

The first intracoronary stent was implanted in 1986 in Europe, and it was quickly confirmed that the device could lower the incidence of restenosis following PTCA. In the US, development of a coronary stent was undertaken by Drs Julio Palmaz and Richard Schatz of the University of Texas Medical School in San Antonio, and the first stainless steel Palmaz-Schatz stent, manufactured by Johnson & Johnson, was approved by the FDA in 1994. Subsequent clinical trials confirmed the superiority of stenting over angioplasty alone in terms of both clinical and angiographic outcomes, virtually eliminating problems associated with abrupt vessel closure, and over the next few years a number of other so-called bare metal stents appeared on the market, each claiming advantages over the original models such as greater flexibility and easier delivery.

Complications can occur with bare metal stents, and one such problem to be identified early on was thrombosis within the stent, which was associated with fatal myocardial infarction, so patients were routinely anticoagulated with warfarin. But this in turn tended to cause bleeding from arterial puncture sites and even systemic haemorrhage, a problem that has been addressed by the substitution of antiplatelet drugs such as dipyridamole for warfarin.

Stents generally are associated with a greater increase in lumen diameter than angioplasty alone, largely due to the fact that they reduce the elastic recoil of the dilated vessel. Neointimal proliferation still occurs after stenting (in fact, stents appear to increase it), but this is usually more than compensated for by the reduction in elastic recoil. However, in some cases, it is severe enough to lead to reocclusion of the vessel, which requires a repeat procedure.

A number of ways were investigated to try to inhibit tissue overgrowth following stent placement, including alternative balloon designs, radioactive stents and catheters (brachytherapy), biodegradable stents, adjunctive pharmacotherapy, and stents coated with a cytostatic drug (drug-eluting stents) to inhibit cell division. While each of these approaches met with a certain amount of success it was drug-eluting stents that, despite their relatively high cost, turned out to be the most effective solution, and by the end of 2004 they were used in nearly 80% of all percutaneous coronary interventions (PCI) carried out in the US².

The first DES to be approved in the US was the Cordis Cypher sirolimus-eluting coronary stent, which was approved in April 2003 following expedited review³. The approval was for use in coronary arteries with diameters of 2.5 mm to 3.5 mm and in lengths of 8, 13, 18, 23, 28 and 33 mm. Cordis, a Johnson & Johnson company, hailed the device as a “remarkable and innovative treatment”, adding that “clinical evidence and experience with more than 50,000 patients treated to date in nearly 60 countries suggests the Cypher stent represents the beginning of a new era in interventional cardiology - an era in which the combination of drugs and devices substantially improves patient outcomes.” The device was already available in Europe, Canada, the Middle East and elsewhere.

However, the following July, Cordis issued a “Dear Colleague” letter to doctors in the US stating that it had received adverse event reports of stent thrombosis at the time of implantation or within a few days of implantation, with some centres reporting multiple events⁴. The company suggested that factors impacting the rate of thrombosis might include failure to achieve adequate stent apposition (due to under-deployment) or suboptimal use of antiplatelet medication. It also said that it was aware that some interventionalists had been over-expanding smaller stents for use in larger vessels, and urged doctors to comply strictly with the instructions for use.

In August 2003 the French and Canadian authorities issued safety notices concerning the Cypher stent, followed in September by the UK regulatory agency. In October 2003 the US Food and Drug Administration (FDA) announced that it had received more than 290 reports of thrombosis occurring between one and 30 days after implantation of the Cypher device⁵. “In more than 60 of these reports, the use of the device was associated with the death of the patient; in the remainder, the device was associated with injury requiring medical or surgical intervention.” The FDA said it had also received more than 50 reports, including some deaths, that Cordis considered to be possible hypersensitivity reactions, including pain, rash, respiratory alterations, hives, itching, fever and blood pressure changes. The agency urged doctors to follow the instructions for use of the stent and to be vigilant for any patient symptoms that could be attributed to hypersensitivity.

The FDA issued clarified guidance in November 2003, saying that it considered the Cypher DES to be safe and effective when used according to its labelling, particularly those sections dealing with the selection of patients and the appropriate use of medication⁶. Significantly, it said that the available evidence suggested that the rate of thrombosis was within the expected rate for any stent.

Watching and waiting in the wings was Boston Scientific, which announced in March 2004 that it had received approval from the FDA to market its Taxus Express2 paclitaxel-eluting coronary stent system⁷. Just as Cordis had been almost a year earlier on the launch of Cypher, Boston was upbeat about Taxus. “Broad, consistent clinical data and extensive real-world experience have clearly demonstrated that polymer-based delivery of paclitaxel is a safe and effective therapy that dramatically reduces restenosis. These attributes are complemented by the outstanding deliverability of the Taxus stent system,” said Boston president and chief executive officer Jim Tobin. In an oblique reference to the initial supply difficulties that Cordis hinted had led to cardiologists over-expanding smaller Cypher stents for use in larger vessels, Boston said that it had “ample inventory in all sizes” ready for immediate launch.

Before long, reports of problems with the Taxus stent began to emerge. A story in the New York Times claimed that the FDA was actively seeking reports of possible problems after some cardiologists had reported that the stent had failed to detach from the angioplasty balloon after inflation and others reported that the balloon had failed to deflate at all⁸. For its part, the FDA was reported as saying that, although it was aware of a number of incidents, it was too soon to say whether there was a genuine problem with the device. However, in July 2004 Boston voluntarily recalled 200 Taxus stents “due to characteristics in the delivery catheters that have the potential to impede balloon deflation during a coronary angioplasty procedure⁹”. Two separate manufacturing lots were involved: the problem related to a narrowing in the area where the catheter and balloon are laser welded, which resulted in the potential for impeded deflation and removal of the balloon after stent placement, the company said.

Two weeks later Boston announced it was recalling a further 85,000 Taxus Express2 DESs and 11,000 Express2 bare metal stents because of “characteristics in the delivery catheters that have the potential to impede balloon deflation during a coronary angioplasty procedure¹⁰”.

At this point it is worth mentioning that, although only two DESs have been approved for sale in the US, a number have been approved in Europe and elsewhere including: the Cordis Cypher stent and the Boston Taxus stent; Abbott's Xience V everolimus-eluting stent, which Boston also markets under the name Promus; Medtronic's Endeavor DES, which uses the Abbott rapamycin analogue, ABT-578; Sorin Biomedica's Janus tacrolimus-eluting stent; and Conor Medsystems' CoStar paclitaxel-eluting stent system (which has recently been discontinued because the product failed to meet its primary endpoint in its pivotal study¹¹). The design of the CoStar is interesting in that it is based on a bioresorbable polymer coating, so that after a certain period of time when all the drug has been delivered it effectively becomes a bare metal stent. What is also interesting about the CoStar is that it releases its drug from hundreds of tiny reservoirs. This reservoir platform will be used in a future stent being designed by Johnson & Johnson, which acquired Conor this year. Others in development include Abbott's ZoMaxx zotarolimus-eluting stent.

Alarm bells rang again in October 2004, when a research letter appeared in The Lancet describing four cases of stent thrombosis that occurred late (from 11 months to more than a year) after either a paclitaxel- or a sirolimus-eluting stent had been implanted¹². All four cases, which were confirmed angiographically and resulted in a heart attack, occurred soon after antiplatelet therapy was interrupted. The report led to calls for large-scale registries and postmarketing surveillance studies to quantify the risk of late thrombosis, studies on how to identify patients at particular risk to late thrombosis and, most significantly, studies to determine the optimum duration of antiplatelet therapy. Cordis commented that the number of cases reported was very small and that its Cypher DES had an excellent safety record in clinical trials.

But more bad news emerged at the Annual Scientific Session of the American College of Cardiology in Atlanta, Georgia, in March 2006, when Prof Mattias Pfisterer and colleagues at the Division of Cardiology at the University of Basel in Switzerland reported on the findings of the BASKET-LATE (BAsel Stent Kosten Effektivitäts Trial - LAte Thrombotic Events)¹³. In the study, 826 patients undergoing PTCA were randomised to receive either a DES or a BMS: baseline clinical and lesion characteristics were broadly similar between the two groups. Antiplatelet (clopidrogrel) therapy was withdrawn six months after the procedure. During seven to 18 months' follow-up, the rates of nonfatal myocardial infarction and cardiac death were significantly higher in the DES group, while the overall rate of major adverse cardiac events was slightly, although not significantly, higher in the DES group.

Further data from the BASKET trial were presented at the World Congress of Cardiology/European Society of Cardiology meeting in Barcelona, Spain in September 2006, as were two other studies raising concerns about the safety of drug-eluting stents. In one presentation, Dr Eduardo Camenzind of the University of Geneva described a meta-analysis of all the major, randomised trials of the Cypher and Taxus stents. Analysis of approximately three years' follow-up data showed that death and heart attack, as well as stent thrombosis, were higher in patients who received the first-generation DESs than in patients treated with BMSs. The Cypher stent was associated with a statistically significant 38% increase in risk of heart attack and death, while the Taxus stent was associated with a 16% increase in risk, although that figure was not statistically significant. Dr Camenzind and colleagues' findings were subsequently published in Circulation¹⁴.

The other study, a meta-analysis of all randomised, controlled Cypher and Taxus trials, was presented by Dr Alain Nordmann of the Basel Institute for Clinical Epidemiology. The study concluded that DESs do not reduce total mortality when compared with BMSs, and called for long-term follow-up assessment of cause-specific deaths in patients receiving DESs in order to determine their long-term safety. It also highlighted the risk of DES technology and non-cardiac deaths. Dr Nordmann found that while there was no statistically significant increase in cardiac mortality for either DES compared with the BMSs, Cypher was associated - though also not statistically significantly - with non-cardiac deaths. Separate analyses revealed these deaths to be cancer, stroke and lung disease. Dr Nordmann's analysis was published in the European Heart Journal¹⁵.

The publication of the BASKET-LATE data and the Camenzind and Nordmann analyses prompted the FDA to issue a public update on its position on DESs. In a statement issued on 14 September 2006 the agency said that, while the new evidence raised important questions, it still did not possess sufficient data to be able fully to characterise the mechanism, risks and incidence of DES thrombosis¹⁶.

(There has been speculation outside the FDA that while DESs have been shown to be much better than BMSs at preventing the tissue growth that causes restenosis, this reduced tissue growth means that the stent is exposed to blood for longer, increasing the risk of thrombosis.)

The FDA said that a more formal evaluation of the data in those studies was necessary and that any conclusions would depend on a thorough peer review. The agency also said it was continuing to evaluate information related to the duration of treatment with clopidrogrel as it recognised that the optimum duration of treatment, particularly in patients with more complex lesions than those who had taken part in the original clinical trials in support of Cypher and Taxus approval, had not been defined. A public meeting of the Circulatory System Devices Advisory Panel would be held by the end of the year to explore the matter further.

The advisory panel meeting took place on 7 and 8 December 2006. The first day was used to discuss the use of DESs in accordance with the approved label indications, while the broader use of DESs in “real world” clinical situations and the implications of using DESs outside the approved indications were discussed on the second day¹⁷. The first day included presentations by Cordis and Boston Scientific. The panel's conclusions about the use of DESs in accordance with their labelling were:

• DESs are associated with a clinically important numerical excess of late stent thromboses;
• DESs are not associated with an increased risk of death or myocardial infarction compared with BMSs, although this finding could be due to insufficient sample size, or it could be that there was an increase in death and myocardial infarction secondary to DES thrombosis, but that it was offset by a decrease in these events caused by in-stent stenosis;
• compared with BMSs, DESs are not associated with an increased rate of all-cause mortality;
• safety concerns apply equally to both Cypher and Taxus stents; and
• concerns about the safety of DESs do not outweigh their benefits compared with BMSs.

The conclusions of the second day's discussions were:

• there is an increased risk of adverse events when DESs are used in more complex patients;
• at least 12 months of dual antiplatelet therapy should be recommended for off-label use of DESs;
• there is insufficient data to reach an opinion on whether concerns related to off-label use apply equally to both the Cypher and the Taxus stents;
• DES labelling should state that results from off-label use should not necessarily be expected to be the same as results from the clinical trials used to support marketing approval; and
• there should be larger and longer premarket clinical trials and longer follow-up for postmarketing studies, especially in relation to stent thrombosis events, and more attention should be given to monitoring patient compliance with antiplatelet therapy. (Whether the suggestion for longer premarket trials would be sanctioned is still unclear, and it raises the risk of delays for US entrants that are still at an early stage of their programme.)

In February 2007 the New England Journal of Medicine published a series of articles that were representative of the presentations and discussions that took place at the FDA advisory committee meeting. Observers agreed, however, that there was little new to be gleaned from the articles, and that they left many questions unanswered about using the devices off-label (which reportedly accounts for over 60% of DESs used in the US), the role of antiplatelet therapy, and the long-term safety of DESs.

The latest thinking is that, while percutaneous coronary intervention is known to be useful in acute heart attack and high-risk unstable angina, its use in stable coronary disease patients (who are less likely to suffer a heart attack or die) is debatable - stable disease has been said to account for around 50% of PCI procedures worldwide.

In fact, a major study (Clinical Outcomes Utilizing Percutaneous Coronary Revascularization and Aggressive Guideline-Driven Drug Evaulation - Courage) released in March 2007 found that PCI in stable patients was no better in reducing heart attack and death than cheaper drug therapy¹⁸. This study has been very controversial (it is argued that stents are used to relieve symptoms rather than reduce death). But still, the findings are expected to lead to fewer stents being used. Mainly bare metal stents were used in the study, but the findings were said to be applicable to DESs.

The ongoing uncertainty about DESs has led Henry Waxman (Democrat - California), chair of the US House of Representatives' Committee on Oversight and Government Reform, to write to Boston Scientific and Johnson & Johnson requesting a broad range of information about the safety and off-label use of DESs¹⁹. The letters referred to the December 2006 FDA advisory panel meeting. The requested information has now been provided, and the committee is currently studying it²⁰. Whether the committee's deliberations result in any action remains to be seen.

References

1. World Health Organization, Atlas of Heart Disease and Stroke, 2004

2. Maisel WH, 2007, Unanswered Questions - Drug-Eluting Stents and the Risk of Late Thrombosis, New England Journal of Medicine, 356(10), 981-984

3. Cordis press release, 24 April 2003, www.cordis.com/active/crdus/en_US/html/cordis/downloads/press/2003CYPHERUSApproval423700pm.pdf

4. Cordis “Dear Colleague” letter, Subacute Thrombosis in Association with CYPHERTM Coronary Stents - What is Known?, 7 July 2003, www.fda.gov/bbs/topics/news/cordis_ltr.pdf

5. FDA talk paper, 29 October 2003, www.fda.gov/bbs/topics/ANSWERS/2003/ANS01257.html

6. FDA talk paper, 25 November 2003, www.fda.gov/bbs/topics/ANSWERS/2003/ANS01267.html

7. Boston Scientific press release, 4 March 2004, http://bostonscientific.mediaroom.com/index.php?s=press_releases&item=331

8. F.D.A. Seeks Reports of Stent Problems, The New York Times, 23 April 2004

9. Boston Scientific press release, 2 July 2004, http://bostonscientific.mediaroom.com/index.php?s=press_releases&item=439

10. Boston Scientific press release, 16 July 2004, http://bostonscientific.mediaroom.com/index.php?s=press_releases&item=460

11. Johnson & Johnson press release, 7 May 2007, www.investor.jnj.com/releasedetail.cfm?ReleaseID=241182&year=2007&print=on

12. McFadden EP, Stabile E, Regar E, Cheneau E, Ong ATL, Kinnaird T, Suddath WO, Weissman NJ, Torguson R, Kent KM, Pichard AD, Satler LF, Waksman R and Serruys PW, 2004, Late thrombosis in drug-eluting coronary stents after discontinuation of antiplatelet therapy, The Lancet; 364, 1519-1521

13. Pfisterer ME, Kaiser CA, Bader F, Brunner-La Rocca HP, Bonetti PO and Buser PT, Late clinical events related to late stent thrombosis after stopping clopidogrel: prospective randomized comparison between drug-eluting versus bare-metal stenting. Program and abstracts from the American College of Cardiology 55th Annual Scientific Session, 11-14 March 2006, Atlanta, Georgia, Abstract 422-11

14. Camenzind E, Steg PG and Wijns W, Stent thrombosis late after implantation of first-generation drug-eluting stents: a cause for concern, 2007, Circulation, 115(11), 1433-1439

15. Nordmann AJ, Briel M and Bucher HC, 2006, Mortality in randomized clinical trials comparing drug-eluting stents vs bare metal stents in coronary artery disease: a meta-analysis, European Heart Journal, 27(23), 2784-2814

16. FDA statement on Coronary Drug-Eluting Stents, 14 September 2006, www.fda.gov/cdrh/news/091406.html

17. Summary from the Circulatory System Devices Panel Meeting - 7 - 8 December, 2006, www.fda.gov/cdrh/panel/summary/circ-120706.html

18. Society for Cardiovascular Angiography and Interventions press release, 27 March 2007, www.scai.org/pr.aspx?PAGE_ID=5059

19. Committee on Oversight and Government Reform press release, Chairman Waxman Requests Research and Marketing Information from Pharmaceutical and Medical Device Companies, 5 March 2007, http://oversight.house.gov/story.asp?ID=1199

20. Personal communication, Committee on Oversight and Government Reform, 8 May 2007