Introduction

In some patients with coronary artery disease, drug-coated balloons (DCB) represent an attractive therapeutic alternative to drug-eluting stents (DES), currently the undisputed gold standard for percutaneous coronary interventions.^1,2 At present, DCB provide the best means to pursue a leave-nothing-behind strategy following studies demonstrating that bioresorbable scaffolds are inferior, regarding safety and efficacy, to new-generation DES. In Europe, current clinical practice guidelines for coronary revascularization recommend the use of DCB (Class 1, Level of Evidence: A) for patients with in-stent restenosis (ISR).¹ Evidence supporting this recommendation stems from multiple randomized clinical trials (RCT) demonstrating that, in this unique anatomical setting, DCB are superior to classic treatment modalities and equivalent to new-generation DES.³ Notably, in patients with ISR, although DES provide better long-term angiographic results, these only translate into a marginal superiority over DCB regarding target lesion revascularization.³ This clinical benefit is considered insufficient by many to justify systematic implantation of an additional, permanent, metal layer on the vessel wall.²

The value of DCB in selected patients with de novo lesions has been investigated in observational studies, RCT, and meta-analyses,⁴ but the available evidence is clearly less robust than is available in the setting of ISR. A consensus document on DCB highlights the role of this technology in special clinical and anatomical scenarios (patients at high bleeding risk or with acute myocardial infarction and lesions in small vessels or involving bifurcations), outside the niche of ISR.² In small vessels, the relatively large BASKET-SMALL-2 (Basel Stent Kosten Effektivitäts Trial Drug Eluting Balloons vs. Drug Eluting Stents in Small Vessel Interventions) RCT demonstrated that DCB are noninferior to DES for the primary combined clinical endpoint at 1 and 3 years,⁵ although some patients in the control group were treated with first-generation DES. Also in small vessels, the PICCOLETO-2 (Drug Eluting Balloon Efficacy for Small Coronary Vessel Disease Treatment) RCT demonstrated that DCB have a lower angiographic late lumen loss (LLL), as compared with new-generation DES,⁶ though minimal lumen diameter and percent diameter stenosis at follow-up, valuable endpoints in this setting, were similar in both treatment groups. In selected patients with acute myocardial infarction, 2 small RCT supported the use of a DCB-alone strategy.^7,8 In the PEPCAD-NSTEMI (Bare Metal Stent Versus Drug Coated Balloon With Provisional Stenting in Non-ST-Elevation Myocardial Infarction) RCT, DCB were noninferior to stents for the primary composite clinical endpoint at 1 year.⁷ Similarly, the REVELATION (Revascularization With Paclitaxel-Coated Balloon Angioplasty Versus Drug-Eluting Stenting in Acute Myocardial Infarction) RCT suggested that DCB and new-generation DES provided a similar physiological outcome (fractional flow reserve) in the culprit vessel at 9-month follow-up.⁸ For lesions involving bifurcations, the small PEPCAD-BIF (The Paclitaxel-Eluting Percutaneous Coronary Angioplasty [PTCA]-Balloon Catheter for the Treatment of Coronary Bifurcations) RCT showed a lower restenosis rate with DCB compared with conventional balloon angioplasty (BA).⁹ Last, but not least, in high-bleeding risk patients, the DEBUT (Drug-Eluting Balloon in Stable and Unstable Angina) RCT demonstrated that DCB reduced the occurrence of adverse clinical events compared with bare-metal stents.¹⁰ Due to the limitations of the totallity of the evidence for this indication, the use of DCB in de novo lesions is not yet recommended by clinical practice guidelines in any specific clinical or anatomical niche,¹ suggesting that more robust evidence remains an unmet need.^1,2

Some additional important issues shape the field of DCB in the coronary space. First, most RCT on DCB are small and focus on surrogate angiographic endpoints, but large “pivotal” RCT, powered for major hard clinical endpoints, are lacking.^1,2 This may explain why this technology has not yet been approved for clinical use in some countries, including the United States. Second, most of the available evidence supporting the efficacy of DCB in the coronary territory was obtained with DCB eluting paclitaxel and, in particular, with a specific DCB technology.^1,2 Therefore, a class effect cannot be assumed for these devices: not all DCB are created equal.^1,2 Paclitaxel has been classically selected for DCB because its lipophilic properties facilitate the transit and retention at the vessel wall.² Paclitaxel is a cytotoxic drug with a narrow therapeutic range that irreversibly blocks cell division, preventing depolymerization of intracellular microtubules. The safety concerns and controversy regarding the use of paclitaxel-DCB in peripheral vessels have never affected the coronary space where, recently, a large and exhaustive meta-analysis completely dissipated doubts on any potential adverse signal.¹¹ Nevertheless, in new-generation DES, limus-type drugs have completely displaced taxane derivatives, as a result of superior safety and efficacy.¹ Limus-type drugs are potent cytostatic drugs with a wider therapeutic window and reversible effects at the vessel wall. However, limus drugs do not exhibit high lipophilicity and face major difficulties in achieving adequate tissue penetration and retention when eluted from a DCB.²

Fortunately, recent technological advances have been able to overcome these pharmacodynamics challenges, and currently limus-DCB are available for clinical use in many countries. Preliminary clinical experience with the use of limus-DCB has been encouraging.^2,12,13 The largest experience encompasses the use of sirolimus-DCB for real world patients, with either ISR or de novo lesions, with satisfactory mid-term clinical results.^2,12,13 However, RCT comparing limus-DES with reliable comparators are scarce and involve a very limited number of patients (Table 1).^14-16 Whether limus-DCB will fulfil the expectation of being superior to paclitaxel-DCB remains unsettled.

Author	Year	Setting	Devices	Patients, n	LLL, mm	TLR, %
Ali et al14	2019	DES-ISR	Sirolimus-DCB	25	0.17 ± 0.55	12
		Paclitaxel-DCB	25	0.21 ± 0.54	16
Ahmad et al15	2022	De novo	Sirolimus-DCB	35	0.10 ± 0.32	0
		Paclitaxel-DCB	35	0.01 ± 0.33	0
Xu et al16	2022	De novo SVD	Biolimus-DCB	106	0.16 ± 0.29a	5.7
		BA	103	0.30 ± 0.35	10.9

BA = conventional plain balloon angioplasty; DCB = drug-coated balloon; DES = drug-eluting stent; DES-ISR = drug-eluting stent in-stent restenosis; LLL = late lumen loss; RCT = randomized clinical trials; SVD = small vessel disease; TLR = target lesion revascularization.

a P < 0.05.

In this issue of JACC: Cardiovascular Interventions, Xu et al¹⁶ present results of the BIO-RISE CHINA (A Randomized Trial of a Biolimus-Coated Balloon Versus POBA in Small Vessel Coronary Artery Disease) RCT, where a novel biolimus-DCB was compared with BA in patients with de novo lesions in small vessels. The study was conducted in 10 Chinese centers, enrolled 212 patients, with 177 patients (86%) entering in the per-protocol population, and used a superiority design. Lesions were to be <25 mm in length and in vessels with a reference diameter of 2.00 to 2.75 mm on visual estimation. Patients were only randomized after a satisfactory angiographic result was obtained following predilation of the target lesion. The primary endpoint, in-segment LLL at 9 month, was met (biolimus-DCB 0.16 ± 0.29 vs BA 0.30 ± 0.35 mm; P = 0.001). In addition, late lumen enlargement (ie, negative LLL) occurred in one-third of patients in the biolimus-DCB arm vs one-tenth of those in the BA arm (P = 0.007). From a clinical standpoint, target lesion failure (6.7% vs 13.9%) and a patient-oriented combined outcome measure (14.3% vs 21.8%) were similar, although numerically favored the biolimus-DCB group.

Considering the potential implications of this RCT and addressing some methodological nuances are important.¹⁶ First, using conventional BA as the comparator deviates from standard of care for these patients. According to clinical practice guidelines, new-generation DES are the treatment of choice across the spectrum of lesion and patients subtypes including lesions in small vessels.^1,17 Methodologically, however, removing the confounding effects of stent implantation, allows better definition of the treatment effect of biolimus in isolation. Nevertheless, using a classical paclitaxel-DCB as the comparator would have provided unique mechanistic and efficacy insights. Notwithstanding these considerations, BA are still frequently used in clinical practice to treat small vessels, and in this regard, the results of this RCT are of certain clinical relevance.¹⁷

Second, patients with suboptimal angiographic results after lesion preparation were excluded, and this explains a cross-over to stents after randomization of only 3%. The number of patients assessed for eligibility, but eventually not randomized, was also very low 12 of 224 (5.3%). Therefore, it remains unclear whether additional patients were excluded due to inadequate results after predilation requiring DES implantation. This issue should be kept in mind when considering the general clinical applicability of this strategy.

Third, this RCT is a first-in-man study of this novel limus-DCB, using biolimus A9 at a dosage of 3 μg/mm² and polyethylene oxide as excipient. Thus, additional manufacturing details would have been of interest. In this regard provision of preclinical pharmacokinetic and histopathological results, both at the vessel wal and downstream myocardium would have been of major scientific value.

Fourth, angiographic data are paramount considering that LLL was the primary endpoint. The superiority of biolimus-DCB compared with BA regarding LLL was clearly demonstrated. Indeed, this superiority was observed despite a better-than-expected performance of BA in the control arm: LLL after BA was lower than expected and previously reported. This may be explained by the inclusion of very small vessels with rather short lesions (reference vessel diameter 2 mm, lesion length 11 mm, by quantitative angiography) as compared with other studies in small vessels.^1,2,17 By contrast, although the LLL of the biolimus-DCB was halved as compared with BA, it remains significantly higher than in previous studies using paclitaxel-DCB.² Similarly, late lumen enlargement after biolimus-DCB only occurred in one-third of lesions, whereas this phenomenon appears to be more frequently found after paclitaxel-DCB.² Of note, the term positive vascular remodeling should probably be avoided when intracoronary imaging results are not available. One could speculate that a lower rate of late lumen enlargement might suggest a lower cytotoxicity of biolimus compared with paclitaxel. Conversely, the relatively high LLL might also suggest a lower efficacy of biolimus-DCB compared with paclitaxel-DCB. However, historical comparisons are always flawed, and this issue can only be resolved by adequately powered head-to-head comparative studies.^3,4 One such trial is ongoing in the setting of ISR (Biolimus A9™ [BA9™] Drug Coated Balloon [DCB] Study [REFORM]; NCT04079192). Finally, as the methodology and precise system used for the quantitative angiographic analysis (performed at a core lab run by the study’s contract research organization), was not clarified, its potential influence on these differences remains difficult to elucidate.

Last but not least, the clinical implications of surrogate angiographic parameters are particularly elusive on small vessels, where the presence of asymptomatic restenosis may disconnect anatomical results from clinically driven target lesion revascularization.¹⁷ Larger studies are required to confirm the favorable clinical results obtained with this novel biolimus-DCB.

Previous Studies with Sirolimus DCB

Information on the clinical value of limus-DCB remains limited.^12-16,18 In the family of limus-DCB, sirolimus has been the drug most frequently selected and several technologies have been developed that accord coating stability of the drug on the balloon surface and then adequate transfer, concentration, and retention at the vessel wall.^12-16,18 These include the use of nanocarriers, nanoencapsulated sirolimus liquid formulations, and crystalline sirolimus on a butylated hydroxytoluene excipient.^12-16,18 Small single-arm studies with systematic angiographic follow-up have demonstrated the antirestenotic efficacy of limus-DCB. Moreover, several prospective registries of sirolimus-DCB in patients with ISR or de novo lesions have reported satisfactory clinical results with low rates of target lesion revascularization.^12,13 However, data from controlled RCT are scarce (Table 1).^14-16 Ali et al,¹⁴ in a head-to-head RCT including 50 patients with DES-ISR, demonstrated a similar in-segment LLL at 6-month (0.17 ± 0.55 vs 0.21 ± 0.54 mm) and similar rate of 1-year clinical events, with a novel crystalline sirolimus-DCB versus the standard paclitaxel-DCB. In another small RCT in 70 patients with de novo lesions (mean reference vessel diameter 2.76 mm) Ahmad et al¹⁵ showed a similar LLL with crystalline sirolimus-DCB and paclitaxel-DCB (0.10 ± 0.32 vs 0.01 ± 0.33; noninferiority P < 0.05) but, interestingly, late lumen enlargement was more frequently found in the paclitaxel-DCB arm (60% vs 32%; P = 0.019).

In summary, Xu et al¹⁶ should be commended for designing and conducting this RCT providing evidence supporting the efficacy of this biolimus-DCB for de novo lesions in small vessels. Although these results are certainly promising, they remain preliminary, considering that the small sample size was unpowered to detect differences in clinical endpoints. Further studies are required to confirm the relative value of limus-DCB compared with new-generation DES in patients with ISR and de novo lesions. Similarly, additional RCT are warranted to characterize the efficacy of limus-DCB compared with standard paclitaxel-DCB in different clinical and anatomic scenarios.