Brachytherapy for In-Stent Restenosis: Title and subTitle BreakA Distant Second Choice to Drug-Eluting Stent Placement

The greatest recent mechanical advance in percutaneous coronary revascularization (PCR) has been the development of bare-metal stents, which compared with traditional balloon angioplasty substantially reduce angiographic restenosis and the need for repeat target vessel revascularization (TVR). Stents provide a larger arterial lumen diameter immediately postprocedure (acute gain), although their drawback is an increased reparative response of neointimal formation (late loss). Fortunately, the net gain remains greatest with stents compared with other PCR devices. In less complex lesions, the rate of TVR with bare-metal stents is approximately 10% to 15%, although this rate has been reported to be 2- to 3-fold higher in more complex lesions and unique patient subsets.^{1 - 2} In 2003, at a time when the use of bare-metal stents peaked, approximately 1 million coronary stents were placed in patients hospitalized in the United States.³ Even with a conservative estimate, this means at least 100 000 in-stent restenotic lesions occurred, making this an important clinical problem.

During the last decade, numerous modalities have been used to treat in-stent restenosis (ISR); however, each has provided only modest intermediate-term efficacy. These devices—balloon angioplasty, atherectomy (directional, rotational, and laser), and repeat stenting (stent-in-stent)—provided a high rate of immediate technical success and a low rate of ischemic events, but 30% to 60% of patients required another TVR in the subsequent months.^{4 - 7} The only therapy proven to be uniquely effective in treating ISR has been the intracoronary delivery of ionizing radiation as an adjunct to successful repeat PCR. This procedure, referred to as vascular brachytherapy, involves the transient (minutes) placement of a β-emitting or γ-emitting radiation source (seeds, wire, or liquid-filled balloon) within an indwelling coronary catheter. β Radiation (electrons generated from phosphorous 32, strontium/yttrium 90, or rhenium 188) and γ radiation (photons generated from iridium 192) are able to penetrate the arterial wall and break the bonds of single-stranded DNA in actively dividing cells, such as the proliferating smooth muscle cells that cause restenosis. Several randomized, placebo-controlled (dummy radiation source) studies testing β and γ radiation in the setting of ISR-PCR have demonstrated a similar and sharp reduction in repeat revascularization (TVR rates of 15%-30%) with brachytherapy.^{4 - 8}

However, the benefits of radiation therapy have been mitigated by safety concerns, procedural logistics (need for multiple personnel, specialty equipment, radiation barriers), stent-edge restenosis, thrombosis, and more recently evidence of a late TVR catch-up phenomenon.^{9 - 11} An early meta-analysis by Radke et al¹² suggested that balloon angioplasty should be the treatment of choice for ISR, particularly if a sufficient acute procedural result could be achieved, and that brachytherapy should be considered for cases with more refractory types of ISR. Diffuse, proliferative, and totally occlusive forms of ISR represent the spectrum's end of neointimal hyperproliferation and are a strong determinant of worse angiographic and clinical outcome.¹³ Interventional cardiologists have impatiently waited for a safe, effective, and easily deliverable therapy for ISR and, before prospective data were available, many quickly turned to another locally delivered antiproliferative therapy—drug-eluting stents.

Prior small nonrandomized studies^{14 - 16} suggested that drug-eluting stents may be effective for ISR. In this issue of JAMA, Stone et al¹⁷ and Holmes et al¹⁸ report randomized comparisons of vascular brachytherapy vs a paclitaxel-eluting stent and a sirolimus-eluting stent, respectively, as treatment for restenosis within a bare-metal stent. Each trial randomized approximately 400 patients to either radiation therapy or a drug-eluting stent. There were some differences in trial design beyond the type of drug-eluting stent tested. In the TAXUS V ISR trial,¹⁷ randomization was 1:1 and a β-emitting radiation source was used exclusively as the comparator. In the Sirolimus-Eluting Stent with Vascular Brachytherapy for the Treatment of In-Stent Restenosis (SISR) trial,¹⁸ randomization was 1:2 and either a β-emitting or a γ-emitting radiation source was used. Nonetheless, the results of these 2 trials were remarkably consistent and suggest roughly a halving of the risk of restenosis with either drug-eluting stent compared with brachytherapy. The extent of late loss and net gain of lumen diameter among patients receiving a paclitaxel-eluting stent in TAXUS V ISR trial was 0.13 mm and 1.49 mm, respectively, and among patients receiving a sirolimus-eluting stent in the SISR trial were 0.27 mm and 1.00 mm, respectively. Compared with brachytherapy, these results were associated with a 53% reduction in angiographic restenosis and a 61% reduction in clinical restenosis (target lesion revascularization) with a paclitaxel-eluting stent in the TAXUS V ISR trial, and a 33% and 56% respective reduction in these measures with a sirolimus-eluting stent in the SISR trial.

Mechanistically, the superiority of a drug-eluting stent is likely explained by the greater acute gain achieved with stent placement coupled with an extent of late loss that is similar to or lower than that with brachytherapy. In addition, a drug-eluting stent avoids the occasional brachytherapy hazard of an “edge-effect,” in which waning radiation at the source's therapeutic edge may cause more injury than it suppresses. The Figure displays the acute gain and late loss—and graphically by inference the net gain—achieved with currently available treatments for ISR. Examining data from 10 randomized trials^{4 - 7 ,17 - 22} that included 3346 patients, drug-eluting stents produced the greatest net gain compared with the other therapies, and at an average follow-up of 10 months was associated with a much lower rate of repeat revascularization. These data strongly suggest that drug-eluting stent placement should now be the preferred method for treating restenosis within bare-metal stents.

Are these data strong enough to relegate brachytherapy to a second-line choice for ISR (ie, should patients with restenosis routinely be treated with a drug-eluting stent)? The answer is yes with a few exceptions. There may remain a small fraction of patients with ISR who will be better served with brachytherapy. These would include patients with bifurcation restenotic lesions; vessels or lesions with excessive calcification, tortuosity, or angulation; and other scenarios that may make repeat stenting unsuitable or lead to an increased risk of procedure-related ischemic events. Both TAXUS V ISR and SISR trials excluded patients with renal dysfunction (serum creatinine, >2.0 mg/dL [>176.8 μmol/L]), and the optimal ISR therapy for this cohort remains undefined as re-restenosis may be potentially higher in these patients.

Although the rate of restenosis is quite low after drug-eluting stent implantation, it is not zero, and the effective treatment of restenosis within a drug-eluting stent has not been studied. As such, with the increasing displacement of bare-metal stents with drug-eluting stents as the index PCR device of choice, the next major question may shift focus from the optimal strategy for ISR within bare-metal stents to that occurring within drug-eluting stents. It is doubtful that brachytherapy will have a tenable role in the treatment of drug-eluting stent restenosis. Rather, head-to-head trials of drug-eluting stent–in–drug-eluting stent (with the same vs a different antiproliferative agent) will be needed. Given the low expected rate of subsequent TVR, it is plausible that such trials will need to be quite large and take into account other questions, such as the long-term durability of drug-eluting stents for treating ISR, safety of local reapplication of antiproliferative agents, and optimal duration of dual-antiplatelet therapy.