The Diagnostic Value of Circulating Biomarkers and Role of Drug-Coated Balloons for In-Stent Restenosis in Patients with Peripheral Arterial Disease

Peripheral arterial disease (PAD) is an increasingly pathological condition that commonly affects the femoropopliteal arteries. The current fashionable treatment is percutaneous transluminal angioplasty (PTA), often with stenting. However, the in-stent restenosis (ISR) rate after the stenting of the femoropopliteal (FP) district remains high. Many techniques have been proposed for the treatment of femoropopliteal ISR, such as intravascular brachytherapy, laser atherectomy, second stenting and drug-coated balloons angioplasty (DCB). DCB showed a significantly lower rate of restenosis and target lesions revascularization (TLR) compared to conventional PTA. However, further studies and multi-center RCTs with dedicated long-term follow-up are needed to verify the true efficiency of this approach. Nowadays, the correlation between PAD and inflammation biomarkers is well known. Multiple studies have shown that proinflammatory markers (such as C-reactive proteins) and the high plasma levels of microRNA could predict the outcomes after stent placement. In particular, circulating microRNA-320a, microRNA-3937, microRNA-642a-3p and microRNA-572 appear to hold promise in diagnosing ISR in patients with PAD, but also as predictors of stent patency. This narrative review intends to summarize the current knowledge on the value of circulating biomarkers as predictors of ISR and to foster the scientific debate on the advantages of using DCB in the treatment of ISR in the FP district.


Introduction
Peripheral arterial disease (PAD) is becoming an increasingly common pathology, with an increase in the world average age and an estimated incidence of 200 million cases per year [1,2]. PAD remains one of the most frequent manifestations of atherosclerosis and cardiovascular pathologies, together with coronary artery and cerebrovascular disease. PAD includes the stenosis of non-coronary and non-cerebral arteries and shares classic risk factors for cardiovascular disease [3]. Classical disease can involve the aorto-iliac, femoro-popliteal district and the below-the-knee arteries. Generally, the below-the-knee arteries are more frequently involved in diabetic patients. The most commonly affected district is the femoropopliteal (FP), which can cause claudication or ischemia of the lower limbs, ischemic ulcers or non-revascularizable conditions, leading to critical limb ischemia (CLI) and a risk of amputation. In the past, open surgery was generally indicated for native arterial disease, mainly bypassed with autologous grafts or with prosthetic material, but with the advent of new technologies, the endovascular approach has become predominant. Nowadays, percutaneous transluminal angioplasty (PTA) represents the first line approach of treatment for obstructive disease of the superficial femoral and popliteal artery. Often, the procedural treatment of this district requires stent implantation after PTA for residual flap dissection or very calcified residual stenosis. However, in-stent restenosis (ISR) after stenting of the FP district remains a daunting problem, resulting in a less than 50% patency rate after 3 years [4]. ISR is believed to be due to neointimal hyperplasia caused by post-PTA endothelial damage [5,6], but similarly to coronary stent disease, other factors are involved such as adherence to antiplatelet therapy [7,8] and specific stent factors [9]. The incidence of ISR also varies according to the type of stent used: 19-35% at 1 year with a grooved tube nitinol stent, and 14-17% with a twisted-wire nitinol stent [4]. The anatomical position can also determine ISR (flexion, stretching, etc.). In the past, there was also stent fracturing, which is particularly common with older self-expanding stents [10]. Many techniques have been proposed for the resolution of ISR such as intravascular brachytherapy, laser atherectomy, second stenting and drug-coated balloon (DCB) [11].
Balloon dilatation and stent implantation is associated with vascular injury, followed by repair processes that include endothelialization and neointimal formation, providing for the activation of the inflammatory response and the release of blood inflammatory markers [9][10][11]. The analysis of these biomarkers could prove useful to physicians to predict an eventual predisposition to ISR. DCB has been shown to yield excellent results in the setting of PAD. Angiographic and clinical data highlight the superiority of this technique, and the European clinical practice guidelines endorsed its use in the treatment of ISR (class I, Evidence A) [11][12][13]. The DCB is a traditional drugcoated balloon, mainly with Paclitaxel, which significantly improves the short-and medium-term patency of angioplasty by minimizing neo-intimal hyperplasia [14][15][16][17]. Paclitaxel, with a nominal dose between 2 and 3.5 µg/mm 2 , has lipophilic and hydrophobic properties and, together with a hydrophilic agent, allows the drug to be released to the surface of the artery (Figure 1). There are more than 10 DCBs available in Europe, only three of which are FDA approved. It has been shown that the treatment of ISR with DCB produces better results than traditional angioplasty, while a superiority over drug-eluting stents has not yet been demonstrated [1,18,19]. DCBs are mainly used on the FP axis, particularly on the FP-ISR and rarely on the common iliac or femoral arteries. Another important topic that will be covered in this review are biomarkers. In the literature, microRNA is emerging with great strength as predictors or diagnostic elements for numerous pathologies. ISR, microRNA-320a, microRNA-3937, microRNA-642a-3p and microRNA-572 are proving to be excellent predictors both for evaluating the follow-up and for evaluating the longevity of the patency of the placed stent. This narrative review intends to summarize the value of circulating biomarkers as predictors of ISR, analyze the current knowledge on DCB on ISR and aims to identify the advantages of using DCB in the treatment of ISR of the FP district.

Materials and Methods
We checked the PubMed and Scopus databases from inception to March 2021. The following key word have been used, with relative MeSH terms: (("drug eluting" OR "drug coated" OR "coated" OR "eluting" OR "paclitaxel") AND ("balloon" OR "device" OR "devices" OR "endovascular" OR "stent") AND ("peripheral" OR "femoral" OR "iliac" OR "popliteal" OR "tibial") NOT "coronary") AND ("review" OR "metanalysis" OR "meta-analysis") AND ("in-stent restenosis"), or ((("drug eluting" OR "drug coated" OR "coated" OR "eluting" OR "paclitaxel") AND ("balloon" OR "device" OR "devices" OR "endovascular" OR "stent") AND ("peripheral" OR "femoral" OR "iliac" OR "popliteal" OR "tibial") NOT "coronary") AND ("in-stent restenosis")) NOT ("review" OR "metanalysis" OR "meta-analysis")). All the articles taken into consideration have been revised; even the references have been checked in order not to lose valuable information, and above all, to verify the relevance with the theme of this review. To be considered, the articles had to have studied ISR in PAD. We initially included randomized studies and observational non-randomized studies. Secondary research articles have been also investigated. We also included, in addition to systematic reviews, prospective studies that examined and compared IRS resolution techniques. Outcomes of interest for selected articles have been included in a shared dataset by three independent authors. All disagreements were resolved by consensus or after consultation with the senior author. Considering the narrative nature of this review, no statistical analysis has been performed.

Results
After preliminary evaluation, duplicate removal and manuscript screening, a total of 33 papers have been included in this review. The study design of the included studies is summarized in Table 1, while Table 2 shows the baseline characteristics of the included patients. In all the studies included in this review, the treatment with DCB demonstrated a higher rate of patency and minor rate of restenosis during the follow-up. Furthermore, a significant clinical improvement was highlighted in the majority of the studies, and this result acquires more value if we consider that these data have not been evaluated in all works. Table 3 outlines the results for each study and each treatment arm.

2020
Multicenter RCT DCB vs. ba ISR ≥70% or in-stent occlusion 3 to 27 cm long within the stent and adjacent segments of the SFA and/or popliteal artery occurring >3 months after stent implantation; (2) Rutherford category 2 to 5 ischemia; (3) at least 1 patent runoff vessel; and (4) patient willingness and ability to continue study participation after the initial study procedure.
Patients were excluded if they were not able to give informed consent, had a creatinine level >2.5 mg/dL, were unable to take antiplatelet drugs, or had a planned surgical or endovascular procedure within 15 days of the index procedure.  (4). Patient understands the nature of the procedure and provides written informed consent, prior to enrolment in the study (5). Patient has a projected life-expectancy of at least 24 months (6). Noninvasive lower extremity arterial studies (resting or exercise) demonstrate ankle-brachial index ≤0.8 (7). Patient is eligible for treatment with the Viabahn ® Endoprosthesis (W.L. Gore) (8). Male, infertile female or female of child bearing potential practicing an acceptable method of birth control with a negative pregnancy test within 7 days prior to study procedure Angiographic inclusion criteria (1). Restenotic or reoccluded lesion located in a stent which was previously implanted (>30 days) in the superficial femoral artery, suitable for endovascular therapy (2). Total target lesion length between 4 and 27 cm (comprising in-stent restenosis and adjacent stenotic disease) (3). Minimum of 1.0 cm of healthy vessel (non-stenotic) both proximal and distal to the treatment area (4). Popliteal artery is patent at the intercondylar fossa of the femur to P 3 (5). Target vessel diameter visually estimated to be >4 mm and <7.6 mm at the proximal and distal treatment segments within the SFA (6). Guidewire and delivery system successfully traversed lesion (7). There is angiographic evidence of at least one-vessel-runoff to the foot, that does not require intervention (<50% stenotic) (1). Untreated flow-limiting aortoiliac stenotic disease (2). Presence of a chronic total occlusion, i.e., a complete occlusion of the failed bare stent that cannot be re-opened with thrombolysis or does not allow easy passage of the guidewire by the physician (3). any previous surgery in the target vessel (4). severe ipsilateral common/deep femoral disease requiring surgical reintervention (5). Perioperative unsuccessful ipsilateral percutaneous vascular procedure to treat inflow disease just prior to enrollment (6). Femoral or popliteal aneurysm located at the target vessel (7). Non-atherosclerotic disease resulting in occlusion (e.g., embolism, Buerger's disease, vasculitis) (8). No patent tibial arteries (>50% stenosis) (9). Prior ipsilateral femoral artery bypass (10). Severe medical comorbidities (untreated CAD/CHF, severe COPD, metastatic malignancy, dementia, etc.) or other medical condition that would preclude compliance with the study protocol or 2-year life expectancy (11). serum creatinine >2.5 mg/dl within 45 days prior to study procedure unless the subject is currently on dialysis (12). Major distal amputation (above the transmetatarsal) in the study or non-study limb (13). Septicemia or bacteremia (14). Any previously known coagulation disorder, including hypercoagulability (15). Contraindication to anticoagulation or antiplatelet therapy (16). Known allergies to stent or stent graft components (nickel-titanium or ePTFE) (17). Known allergy to contrast media that cannot be adequately premedicated prior to the study procedure (18). Patient with known hypersensitivity to heparin, including those patients who have had a previous incidence of heparin-induced thrombocytopenia (HIT) type II (19). Currently participating in another clinical research trial, unless approved by W.L. Gore & Associates in advance of study enrolmen t (20). Angiographic evidence of intra-arterial thrombus or atheroembolism from inflow treatment (21). Any planned surgical intervention/procedure within 30 days of the study procedure (22). Target

Diagnostic Implications
PTA of peripheral arteries is associated with vascular wall injury followed by repair processes, including endothelialization and neointimal formation [51]. This process is more expressed after stent implantation. The damage to the vascular wall caused by angioplasty and/or stenting causes inflammatory activation. This mechanism leads to the activation of the proliferative process, which consists of the proliferation, migration and differentiation of vascular smooth muscle cells (VSMC). Therefore, this inflammatory process leads to the release of various blood markers.
Beyond the classic inflammatory indices, microRNAs are assuming a considerable importance. Many scientific papers examine mRNAs as markers of disease or as markers of disease progression (Figure 2). In the case of ISR, we have noted and assumed from the literature that these markers can evaluate the longevity of a stent based on the degree of expression at the plasma level.

Inflammatory Markers
Numerous studies have reported the probable correlation between inflammatory markers and the restenosis process. In particular, some studies have previously reported that high levels of C-Reactive Protein (CRP) could predict the ISR after the BMS implantation [52,53]. Furthermore, Zhu et al., [54], conducted a meta-analysis of six prospective observational trials, which confirmed that the higher level of hs-CRP is associated with a significantly increased risk of ISR (OR 1.16, 95% CI 1.01-1.30; p < 0.05) in patients who underwent percutaneous transluminal coronary angioplasty. Finally, Jakubiak et al. [55], reported in a review that processes such as inflammation, neointimal hyperplasia and neoatherosclerosis, allergy, resistance to antimitotic drugs used for coating stents and balloons, genetic factors, and technical and mechanical factors, could be implicated in restenosis complications. The authors concluded that every effort should be made to develop knowledge about the pathogenesis of ISR after endovascular treatment of PAD, leading to the availability of more and more perfect therapeutic tools in clinical practice. Some studies reported that the activation of the complement system could play a crucial role in the pathogenesis of restenosis [56,57]. Speidl et al. [56] reported that a higher C5a plasma level is associated with an increased risk of restenosis in patients with PAD who underwent peripheral PTA. In this study C5a concentration was measured at baseline and eight hours after the procedure. Median C5a levels increased significantly from 39.7 ng/mL (IQR 27.8 to 55.0) at baseline to 53.8 ng/mL (IQR 35.6 to 85.1, p < 0.001) 8 h post intervention. During follow-up period, 53% of patients developed restenosis, and elevated levels of C5a at baseline were significantly associated with an increased risk for restenosis (p = 0.0092). Furthermore, the authors specified that this effect was independent of nonspecific inflammation as reflected by the plasma levels of CRP in their patients.
From these experiences, it comes to light that the inflammatory mechanisms play a major role in the development of restenosis after PTA and stent implantation. Therefore, the value of inflammatory biomarkers should be more investigated to improve patency rates.

microRNA
Recently, studies have reported that elevated levels of microRNA in patients after stenting could be predictive of ISR. Yuan et al. [58] reported that circulating microRNA-320a and microRNA-572 have promising value in diagnosing ISR in patients with PAD. The authors compared 78 patients with ISR, 68 non-ISR patients and 62 healthy volunteers. The microarray analysis showed significant changes in microRNAs, which were up-regulated or down-regulated in ISR groups compared with non-ISR and healthy volunteers. In fact, the analysis revealed that the expression of plasma microRNA-320a, microRNA-3937, microRNA-642a-3p and microRNA-572 were significantly higher in ISR patients than in the control groups. On the other hand, microRNA-4669 and microRNA-3138 showed significantly lower expression in the ISR group than that in the control groups. In addition, from the entire sample set, testing with quantitative reverse transcriptase-polymerase chain reaction (qRT-PCR) and receiver operating characteristic (ROC) analysis, the patients with ISR showed significantly higher expression levels of microRNA-320a and microRNA-572, suggesting their high potential diagnostic value for ISR detection.
Furthermore, Stojkovic et al. [59] recently reported a study on 62 consecutive PAD patients after infrainguinal PTA with stent implantation. The authors investigated the predictive value of 11 microRNAs for the composite endpoint of restenosis and atherothrombotic events (primary endpoint) and target lesion revascularization (TLR, secondary endpoint), demonstrating that the circulating microRNA-195 could predict restenosis, atherothrombotic events and TLR after PTA with stent implantation in FP district.

Discussion
The new treatments for PAD have considerably simplified the post-operative course compared to the open surgery approach, but a significant increase in ISR has been observed in recent years and is forecasted in the next years considering the number of procedures performed in each center [2,4,5,26,60,61]. Despite reducing the impact for the patient, PTA is giving very poor results in terms of long-term patency and TLR, and DCB is a new technique which is performing adequately in ISR.
As anticipated in the introduction, this manuscript has the intention of providing a narrative review on the current knowledge on the value of circulating biomarkers as predictors of ISR and to foster the scientific debate on the advantages of using DCB in the treatment of ISR in the FP district. In fact, the use of the stent in the FP district will become an increasingly pursued practice in the field of vascular surgery, making it increasingly necessary to study this knowledge in depth.
There is a focus on more complex and combined techniques for the treatment of ISRs which are producing more encouraging results such as DES, again with major limitations such as a stent length which excludes treating long stenosis. DCB shows superior results compared to traditional re-PTA with a stent and is becoming the technique of choice for ISRs [12][13][14][15]62]. Combined techniques such as laser atherectomy and DCB are also being developed and appear to be candidates to become the reference technique, but adequate studies are warranted [1,11,18,19]. In all the studies included in this review, the treatment with DCB demonstrated a higher rate of patency and minor rate of restenosis during the follow-up. Furthermore, Tepe et al. [24] in the COPACABANA trial reported that at the 12-month follow-up, TLR was performed in 18 (49%) of 37 patients in the uncoated group and in 6 (14%) of 43 patients in the single-dose DCB group (p = 0.001). At~2 years after treatment, a remarkable number (14/27, 52%) of TLRs were recorded in the single-dose DCB group. The authors concluded that treatment with DCBs resulted in significantly less 6-month restenosis rate and fewer TLRs up to 24 months than the treatment with uncoated balloons.
In the recent metanalysis performed by Xi et al. [12], despite not considering the most recently published articles, 18 studies (9 RCTs and 9 OSs) have been included with a followup extended to 3 years. The data analysis showed how the treatment of ISR with DCB and DES was comparable, concluding that the treatment with DCB has a proven efficacy and certainly is not inferior to DES. The current guidelines confirm that DCB is a valid treatment for ISR, which obviates various problems brought about by the placement of traditional stents such as the reduction, in terms of time, of the double antiplatelet therapy, required in the placement of stent-in-stent. Another finding was that the restenosis has drastically decreased after the advent of DES and DCB. However, data have emerged that show that PAD patients treated with DCB (paclitaxel) have an increased risk of death, and it is hypothesized that paclitaxel toxicity is the cause of this increased risk. The mechanism of this increase remains unknown, and new studies are needed to investigate this issue. By comparing the DCBs used for the treatment of PAD and the DCBs used for coronary heart disease, it is hypothesized that the problem is the size and therefore the quantity of paclitaxel released. At the coronary level the DCB is very small, and consequently, the quantity of drug released is significantly lower than the DCB used for peripheral arteries (about 10 times greater). Another hypothesis, in addition to the intrinsic quantity of the drug present on the DCB, is the washing of the drug due to the blood circulation, obviously greater in the peripheral arteries than in the coronary circulation. Less circulating paclitaxel may be the reason for the diverging data between coronary and peripheral DCB. An alternative to PCB or Sirolimus-coated balloon (SCB) is being studied, which seems to give encouraging results. The mechanism of SCB is different from paclitaxel: the first falls into the class of cytostatics, the second into the class of cytotoxics, reversibly binding to the FKBP12 receptor forming a complex with rapamycin blocking the cell cycle in the G1 and S phase. Furthermore, Sirolimus has anti-inflammatory properties compared to paclitaxel, which appears to be the target treatment of patients with ISR. This new technology was approved by the European community and obtained the CE mark in 2016, and numerous centers are experimenting with this new method with Sirolimus. Sirolimus might be the alternative to paclitaxel in peripheral and coronary treatment, provided that the mortality data will be confirmed with scientific evidence.
In addition to all the data collected on the efficacy of the specific drug, the anatomy of the stenosis should also be carefully considered. Feng et al. [14] specifically evaluated the length of the stenosis by dividing the patients into two main categories. CTO (chronic total occlusion) >10 cm, in the reported analysis. Those treated with DCB had a 1-year patency free from other treatments, significantly lower compared to chronic total occlusion, with a length less than 10 cm. The measure of stenosis thus became a 1-year predictor of patency. In addition to the length of the stenosis, the degree of calcification was taken into consideration, considered a limiting factor for treatment with DCB13. It was noted that calcification in the arterial vessel reduces the diffusion capacity of the cytostatic/cytotoxic drug to the arterial wall. Another problem for which we are turning to combined procedures, as described above, is the degree of stenosis or calcification that does not allow for the passage of devices. For this reason, many stenoses or ISRs must be pretreated with traditional balloons to fragment the stenosing plaque, bringing the vessel back to an adequate caliber and allowing the absorption of drugs in situ [12][13][14][15].

Conclusions
ISR has been increasingly acknowledged as a daunting complication after percutaneous treatment of PAD. Poor long-term patency and the risk of stent failure in the case of inadequate pharmacological management expose the patient to a risk of limb injury. Multiple studies have shown that proinflammatory markers and high plasma levels of micro-RNA can influence the outcomes after peripheral stenting. This aspect could hold promise for early recognition of ISR, paving the way for future therapeutic tools. The DCB device has been recently introduced to perform target vessel and TLR. DCB have been shown to have promising results and will rapidly be the first line indication for ISR. In particular, DCB showed a significantly lower rate of restenosis and TLR in all the analyzed reports compared to conventional PTA. However, studies with dedicated long-term followup are warranted to verify the true efficiency of a technique. DCB gives excellent results in the short-term outcomes, and therefore it might be reliably speculated that it will become the best practice in complex and complicated PAD treatment.

Conflicts of Interest:
The authors declare no conflict of interest.