Intracranial atherosclerotic stenosis (ICAS) is one of the causes of ischemic stroke. Recurrent stroke is known to occur in approximately 17% of patients with ICAS in 1 year, even with the best medical management [2,4]. Based on technical and clinical experience in the field of coronary artery diseases, endovascular treatment with stents has been attempted for patients with ICAS. Stenting in ICAS patients raises two major concerns : peri-procedural complications and high restenosis rates [2,4,8]. Regarding peri-procedural complications, recent studies have shown better outcomes than previous studies due to advancements in medical management, endovascular techniques and devices [2,4,8]. Regarding the issue of high restenosis rates after cerebral arterial stenting, a proper solution has not yet been identified. Drug-eluting stents (DESs) and drug-eluting balloons (DEBs) in the coronary system, which release drugs that inhibit intimal hyperplasia, have also been used for ICAS with good preliminary results in reducing restenosis [1,2,6,7,9,10,12,13,16]. However, applying these drugs to cerebral arteries may provoke neurotoxicity [2,5,12].

To our knowledge, no report has yet been published on this issue. The authors performed this study to evaluate neurotoxicity due to the local effects of high-dose paclitaxel and rapamycin using a rat model with transient blood-brain barrier (BBB) opening.

The procedures for handling and caring for animals adhered to the guidelines that are in compliance with current international laws and policies (National Institute of Health Guide for the Care and Use of Laboratory Animals, Publication No. 85-23, 1985, revised 1996), and they were approved by the Institutional Animal Care and Use Committee of Kangwon National University (KIACUC-12-0014).

Most of rat brains were stained with EB, with a various range of grades from 1 to 3 (Supplementary Table 1). There were no histological abnormalities in the sham group (Supplementary Table 2, Supplementary Figs. 1 and 2). No neurotoxicity was observed (Bederson’s grade 0 in all cases) in any rats regardless of drug type, concentration or time period (Fig. 4 and Supplementary Figs. 3-14). Moderate degrees of inflammatory cell aggregation and neuronal damage were identified in 1 of 15 sections of just one rat brain (No. 4, day 14 after injection of 600 μg human paclitaxel dose; Supplementary Figs. 15 and 16). No rat showed neurological abnormalities.

The restenosis rate after treating symptomatic ICAS with bare-metal stents is known to be as high as approximately 30% [8,12]. Moreover, DESs and DEBs for ICAS caused lower rates of restenosis, ranging from 0% to 5% and from 0% to 13%, respectively [2,12]. Using DESs and DEBs in cerebral arteries could provoke neurotoxicity. Fortunately, chronic exposure of the brain to low doses of the drugs released from DEBs and DESs appears safe, based on the lack of reports about neurotoxicity due to systemic effects of the drugs in interventional cardiology over 15 years. And, not a few preliminary preclinical and clinical studies showed no local and systemic complications, yet [1,6,7,9,10,12,13,16]. In the early period after DES and DEB placement, high amounts of the drugs are released from the devices within the cerebral arteries. These high amounts could have harmful local effects on the brain. For example, 75–80% of the drugs are released into the bloodstream after DEB inflation, and only 10–15% are attached to the target vessel wall [18]. This study is the first animal experiment examining neurotoxicity due to the local effects of intra-arterial injection of high-dose paclitaxel and rapamycin. Our experiment showed no neurotoxicity, even though the drugs more easily penetrated into the brain tissues due to mannitol-induced BBB opening and higher doses, up to 4-times higher than the general dose on devices, were administered.

The major adverse event related to paclitaxel is peripheral neuropathy [17]. It has been reported that peripheral neuropathy of any grade can occur with 100 to 300 mg/m² paclitaxel. The cumulative dose that causes severe neuropathy is known to be approximately 1000 mg/m² [17]. For example, considering the surface dose (3 μg/mm²) and specifications of SeQuent^® Please NEO (B. Braun Medical, Melsungen AG, Berlin, Germany), a novel paclitaxel-coated balloon ranging from 2.0×10 mm to 4.0×40 mm, the total surface dose is approximately 380 μg–3 mg. As the total surface dose ranges between 1/180–1/60 of the dose that causes peripheral neuropathy, paclitaxelrelated neurotoxicity hardly seems to occur, even if a full dose is released from the highest specification of the balloon. Moreover, paclitaxel cannot normally cross the BBB, and only two cases of grand mal seizure, leading to suspicion of central nervous system toxicity, have been reported [15]. Each ovarian cancer patient experienced seizures after intravenous injection of more than 250 mg/m² paclitaxel in early clinical trials. However, the direct causality of paclitaxel is not clear because one patient had a brain metastasis and sub-therapeutic blood levels of phenytoin, and paclitaxel was not detected in the cerebrospinal fluid of the other patient.

Rapamycin is an immunosuppressant used for solid organ transplantation and a major drug coated on DESs. Rapamycin- related complications mainly include myelosuppression and hyperlipidemia. Although rapamycin easily crosses the BBB, neurotoxicity has not been reported except in five cases of posterior reversible encephalopathy, which was suspected to be caused by rapamycin-related neurotoxicity simply based on recovery after rapamycin discontinuation [5,11]. Therefore, even though our experiment did not show high-dose rapamycininduced neurotoxicity in a rat model, using rapamycin-coated devices needs to be avoided until the adverse effects in humans are clearly shown.

There are a few limitations in this study. First, only two prototype of drugs for the DESs and DEBs were used. However, these data would be applicable because newer drugs are derivatives of rapamycin and paclitaxel. Second, it is an animal experiment. Third, the degree of BBB opening in rat models was a little inhomogeneous, however, there was no tendency of the degree of BBB opening according to some factors. The authors did not attempt additional experiments, sacrificing more animals. Further experiments in vivo or in vitro under the control of the degree of BBB opening can be necessary. Fourth, rats’ neurological states were evaluated simply with a triage test which cannot detect minor changes.

There were no histological or neurological evidences of neurotoxicity by intra-arterial injection of paclitaxel and rapamycin in a rat model with transient BBB opening. This finding is expected to be helpful in consideration of the safety of drugs to reduce in-stent and post-procedural restenosis.