The Development of Magnesium-Based Resorbable and Iron-Based Biocorrodible Metal Sca ﬀ old Technology and Biomedical Applications in Coronary Artery Disease Patients

: In the treatment of atherosclerotic disease patients, the adoption of second-generation drug-eluting stents (DES) in percutaneous coronary intervention reduced the occurrence of in-stent restenosis (ISR) and acute stent thrombosis (ST) when compared to bare metal stents and 1st generation DES. However, the permanent encaging of the vessel wall by any of the metallic stents perpetuates the inﬂammation process and prevents vasomotion in the treated segment. Aiming to overcome this issue, the bioresorbable sca ﬀ old (BRS) concept was developed by providing transient vascular radial support to the target segment during the necessary time to heal and disappearing after a period of time. Close to 20 years since BRS technology was ﬁrst reported, the interventional cardiology ﬁeld saw the rise and fall of several BRS devices. Although iron-based BRS is an emerging technology, currently, magnesium-alloy resorbable sca ﬀ olds devices are supported with the most robust data. This manuscript aims to review the concept of magnesium-based BRS devices, as well as their bioresorption mechanisms and the status of this technology, and the clinical outcomes of patients treated with magnesium BRS and to review the available evidence on iron-based BRS technology.


Introduction
In the treatment of atherosclerotic disease patients, the adoption of second-generation drug-eluting stents (DES) in percutaneous coronary intervention reduced the occurrence of in-stent restenosis (ISR) and acute stent thrombosis (ST) when compared to bare metal stents and 1st generation DES [1][2][3][4][5].
The historical limitations of the DES technology, particularly due to the perpetuation of the inflammation process by the permanent encaging of the vessel wall, led to the development of the Bioresorbable Scaffolds (BRS) concept, namely providing vascular radial support (i.e., scaffolding) to the treated segment during the necessary time to heal the vessel wall and disappearing after a period of time [6]. However, close to 20 years since Tamai et al. first reported the BRS technology concept, the field of interventional cardiology saw the rise and fall of several BRS devices [7]. Although iron-based scaffolds are an emerging technology, currently, magnesium-alloy resorbable scaffold devices are supported with the most robust data [8][9][10]. This manuscript aims to review the magnesium-based BRS Magnesium, an alkaline earth divalent cation (Mg 2+ ), is predominantly an intracellular cation, the 2nd most common after potassium and the 4th in the extracellular matrix. Around 60-65% of which was presented as ionized or free [15,16]. The concentration of Mg 2+ in human physiology is determined by intestinal absorption (1/3 of the overall daily western dietary income), bone exchange (i.e., largest body reservoir), and renal depuration taking between 25 to 81 hours [17,18].
Ionic magnesium is essential for some enzymatic reactions, protein synthesis, metabolism, and ion channel regulation. In the later, it inhibits the calcium-induced release channels and stabilizes the cellular membrane by increasing the excitability threshold, preventing eclampsia, arrhythmias, and seizures for example [17][18][19]. Also, the divalent cation characteristics of Mg 2+ shifts the membrane's ionic charge distribution making it prone to negative electromagnetic properties, consequently, lowering in-vivo thrombogenicity compared to the positively charged ones [20][21][22]. The thrombogenic protective effect of Mg 2+ was observed in a series of porcine shunt model studies where magnesium-based BRS had less thrombus formations attached to the struts than durable metallic alloys DES [23][24][25]. Finally, pre-clinical studies also suggest that ionic magnesium may present protective effect against vasoconstriction induced by endothelin-1 [26,27].

Magnesium Alloy Biocompatibility and Resorption
A metal stent biocompatibility study reported a significant dielectric spectroscopy interaction between the metallic stents and cell membrane of erythrocytes, albeit this phenomenon was not related to unfavorable clinical outcomes [28]. For this, magnesium alloy devices were developed aiming to increase the biocompatibility between device versus blood and vessel wall. Magnesium has a high biocompatibility but low tensile force and accelerated corrosion rate (ACR) with an increased hydrogen (H 2 ) evolution rate [9,20], which are H 2 pockets formations next to metal surface and the alkalization (OH − ) of the surrounding solution mostly due to the ACR. Biologically, ACR delays the intimal healing process inducing local necrosis and separating the tissue layers [9,29,30]. The ACR issue solution came from magnesium alloys extensively used in aerospace since the addition of low levels of rare-earth metal elements decelerates the ACR process, thus minimizing the H 2 evolution rate issue. After many attempts, two alloys remained as valuable options for BRS application, the AE21 (SHL-Alubin Ltd., Kiryat Biolik, Israel) and WE43 (Magnesium Elektron Ltd., Swinton, Manchester, UK) [29]. Magnesium AE21 alloy is a composite of 97.0% of Mg, 2.0% of Aluminum (Al), and 1.0% of rare-earth metals [Cerium (Ce), Praseodymium (Pr), and Neodymium (Nd)] [8], and the WE43 alloy has 93.6% of Mg, 3.7-4.3% of Yttrium (Y), 2.4-4.4% of rare-earth metals (predominantly Neodymium, Nd), and 0.4% of Zirconium (Zr) [9,30]. From the biologic point of view, these alloys considerably improved the predictability of the degradation process providing longer platform integrity [29] and optimizing the radial strength-to-weight ratio of metallic Magnesium BRS platforms, which are now comparable to aluminum and steel alloys with a lighter platform [29][30][31].
The complete resorption time of magnesium-based platforms vary depending on magnesium alloy, platform characteristics (strut thickness and width), and the presence or not of polymeric coating. In porcine models implanted with AE21 alloy, uncoated platform device struts were observed at 56 days and an expected complete resorption time ≤89.4 days [8]. In fact, an objective resorption observation data came from a WE43 alloy uncoated platform Lekton Magic (Biotronik AG, Bülach, Switzerland) presenting a complete in-vivo resorption in <2 months [9,32,33]. The addition of polymeric coating in the platform like Magmaris (DREAMS 2G-Biotronik AG, Bülach, Switzerland) significantly delayed the in-vivo resorption time up to 12 months when 95% of platform struts were fully resorbed as a consequence providing effective 3 months of scaffolding vessel support time [34].
The magnesium alloy resorption main chemical process is hydrolysis, where Mg 2+ and two H 2 O molecules are converted to magnesium hydroxide [Mg 2+ (OH − ) 2 ] and hydrogen gas (H 2 ) followed by a progressive replacement of magnesium by calcium (Ca 2+ ) and phosphate (PO 4 3-) into a soft amorphous hydroxyapatite imprint [Ca 5 2+ (PO 4 3-) 3 OH − ]. Based on the in vitro and in-vivo observations, these processes occur simultaneously but in a multi-phased fashion. First, the initial corrosion phase starts with the hydrolytic process being homogeneously distributed over platform surface. In the second phase, the resorption process is now heterogeneous, deeply advancing to the inner parts of the struts through microfractures in the surface of the platform. Finally, in the third and last phase, the calcium and phosphate are now homogenously distributed over the imprint remains of the platform until the complete substitution of the residual Magnesium for soft amorphous hydroxyapatite ( Figure 1) [21,32].

First Experience of a Magnesium Alloy Device
In 2003, Heublein B et al. was the first group to report the concept of a magnesium-based AE21 alloy scaffold (non-commercial). Their uncoated scaffold platforms (mass of 4 mg) were prototyped on extruded 2.0 mm diameter AE21 alloy laser cut tubes with a strut thickness of 150-200 µm (uneven thickness) and 10 mm length [8]. The characteristics of each device is presented in Table 1.
In a pre-clinical swine model study, 11 subjects were included with 9 stents in the left anterior descending artery (LAD), 9 in the left circumflex artery (LCx), and 2 in the right coronary artery (RCA) totalling 20 stents with one implanted per vessel. All treated segments ranged from 2.5 to 3.5 mm in proximal reference vessel diameter (RVD) [8]. Each animal was assessed by angiography followed by necropsy which were performed in 3 pigs (5 devices) at 10 days, 4 pigs (7 devices) at 35 days, and 3 pigs (6 devices) at 56 days. Intravascular ultrasound (IVUS) was performed in the 35-and 56-day group of animals. Overall, the imaging and pathology assessments showed patent treated segments and non-thrombosed devices and non-infarcted myocardium except for 1 of the animals (with 2 devices), which died due to scaffold thrombosis <10 days. Moreover, IVUS assessment did not show any significant intimal thickening up to 56 days after the device implantation [8].
processes occur simultaneously but in a multi-phased fashion. First, the initial corrosion phase starts with the hydrolytic process being homogeneously distributed over platform surface. In the second phase, the resorption process is now heterogeneous, deeply advancing to the inner parts of the struts through microfractures in the surface of the platform. Finally, in the third and last phase, the calcium and phosphate are now homogenously distributed over the imprint remains of the platform until the complete substitution of the residual Magnesium for soft amorphous hydroxyapatite ( Figure 1) [21,32].

Lekton Magic Coronary Stent
Lekton Magic Coronary Stent (Biotronik AG, Bülach, Switzerland) was a single tube WE43 alloy laser sculpted platform with 4 crowns and 4 unbowed cross-links in the longitudinal scaffold axis mounted over a dedicated balloon and no radiopaque markers. The device showed a 5% elastic recoil, collapse pressure of 0.8 atm, and strut thickness/width of 165/80 µm (Table 1) [9].
In the pre-clinical porcine study (n = 33), the animals implanted with Lekton Magic showed a higher post-procedure minimum lumen diameter (MLD) of 1.49 mm than the BMS group (1.34 mm) and a luminal increase from 1.49 to 1.69 mm (p < 0.001) at 12 weeks [9].
The Lekton Magic Stent was studied in critically ischemic peripheral artery disease (PAD) patients (n = 20) and 23 lesions were included in this exploratory study. After 1-month follow-up, normal flow assessed by doppler was observed in 18 patients and arterial occlusions in 2 subjects leading to surgical revascularization in one of the cases [9].

Absorbable Metal Stent First Generation (AMS-1)
The Absorbable Metal Scaffold from the first generation (AMS-1) was a single tube laser sculpted WE43 alloy platform with 4 crowns and 4 inter-crown links, and no radiopaque markers. The device's crossing profile was 1.4 mm, 0.8 atm of collapse pressure, 6-7% of elastic recoil and longitudinal shortening of <5% [33,35]. AMS-1 was a pre-mounted semi-compliant balloon crimped device with strut thickness/width of 165/80 µm, and the available device sizes were 3.0 and 3.5 mm in diameter and 10 or 15 mm in length (Table 1) [36].
AMS-1 safety and feasibility were studied in the PROGRESS-AMS trial. A single arm, prospective, multicenter study enrolled 63 patients with stable coronary artery disease (CAD). In the quantitative angiographic assessment at 4 months, the in-scaffold late lumen loss (LLL) was 1.08 ± 0.49 mm with a proximal margin luminal reduction from 2.60 ± 0.47 mm to 2.23 ± 0.65 mm (p < 0.001). In the IVUS assessment, the overall decrease of external elastic membrane-volume (EEM) of 42% with neointimal formation of 45% and 18% of in-scaffold area decrease. At 12 months, there was 27% of ischemic-driven target lesion revascularization with no death, myocardial infarction (MI), or scaffold thrombosis (ST) events [36].
A patient-level pooled multivariate analysis assessing the TLF rate of Magmaris (Biosolve-II and -III trials) versus Orsiro (Bioflow II trial) showed no significant difference in the unadjusted nor in the adjusted data [41,42]. Still, there is no randomized study comparing Magmaris and a 2nd Generation Metallic DES.

Next Generation of Sirolimus-Eluting Resorbable Metal Scaffold (DREAMS 3rd Generation)
The next generation of Magmaris RMS was announced, the DREAMS 3G. The major differences in this device come from the platform keeping the poly-L-Lactide Acid (PLLA) coating eluted with sirolimus. The DREAMS 3G platform is currently being tested in a variety of strut thickness 99 μm, 117 μm, and 147 μm with the same strut width of 150 μm. The device sizes range from 2.5 to 4.0 mm in diameter and from 13 to 30 mm in length with the same resorption period of 12 months. The major difference between the previous generation is an increase in the scaffolding time to ≥3 months ( Table   1). The DREAMS 3G first-in-man and the randomized controlled trial against 2nd generation DES are planned for 2020 [43]. Magmaris vasomotion testing at 6 and 12 months in Biosolve-II trial was positive in 80% (20/25) and 79% (11/14) of the patients [38]. TLF rate at 6 and 24 months was 3% (4/123) and 6% (7/116) of the patients [38,40], and of these 7 events, 2 were cardiac deaths, 1 MI, and 4 clinically-driven target lesion revascularizations (TLR) [40]. There were no definite or probable ST cases at 24 months [38][39][40].
A patient-level pooled multivariate analysis assessing the TLF rate of Magmaris (Biosolve-II and -III trials) versus Orsiro (Bioflow II trial) showed no significant difference in the unadjusted nor in the adjusted data [41,42]. Still, there is no randomized study comparing Magmaris and a 2nd Generation Metallic DES.

Next Generation of Sirolimus-Eluting Resorbable Metal Scaffold (DREAMS 3rd Generation)
The next generation of Magmaris RMS was announced, the DREAMS 3G. The major differences in this device come from the platform keeping the poly-L-Lactide Acid (PLLA) coating eluted with sirolimus. The DREAMS 3G platform is currently being tested in a variety of strut thickness 99 µm, 117 µm, and 147 µm with the same strut width of 150 µm. The device sizes range from 2.5 to 4.0 mm in diameter and from 13 to 30 mm in length with the same resorption period of 12 months. The major difference between the previous generation is an increase in the scaffolding time to ≥3 months ( Table 1). The DREAMS 3G first-in-man and the randomized controlled trial against 2nd generation DES are planned for 2020 [43].

Iron Alloy Platform Degradation
In the benchmark corrosion test with saline immersion, a pure iron non-polymer covered platform (99.8% of iron) with 26 mm in length and strut thickness of 91 µm releases 41 mg/month of iron into the bloodstream, closely corresponding to the total oral intake of iron for 1 month [44]. Overtime, in the animal model analysis (rabbits), there were no observed signs of iron toxicity up to 18 months [44].
Pure iron devices are highly biocompatible and structurally sturdy, however, they present a long corrosion period and slow clearance from the vessel up to 18 months from implantation [45]. However, iron-alloy processed with nitriding technology, a thermochemical heat treatment with nitrogenation of the iron piece, adds more reliability to corrosion rate when associated with PLLA coating as seen in porcine models at 13 months [45,46].

Biocorrodible Pure Iron Devices
The first study reported with the iron-based BRS came from NOR-I (Devon Medical, Hamburg, Germany). Based on >99.8% of pure iron (Goodfellow, Cambridge, UK), this platform was tested in the pre-clinical environment in the descending Aorta of rabbit models [44]. Overall, 16 animals were implanted with the device and evaluated by angiography and followed by necropsy from 6, 12, or 18 months. Results showed the absence of thrombus, no significant neointima proliferation, and no toxicity (locally or systemically) [44].
Waksman et al. reported a porcine model evaluation of the Pure Iron Stent (Biotronik, Erlangen, Germany) compared to metallic stents at 28 days [47]. Overall, in the 9 pigs implanted with pure iron stents, no intra-luminal thrombus or statistically significant differences in vessel, lumen, and stent areas in the histological analysis were observed [47].

Sirolimus-Eluting Iron Bioresorbable Coronary Scaffold (IBS)
The major difference of this device compared to the pure iron-devices is the nitriding technology and the Zinc buffer covering the whole strut before the sirolimus-eluted PLLA coating, conferring a more reliable corrosion rate to the device and maintaining structural integrity up to 4-6 months and complete corrosion of the struts at 13 months in the pre-clinical rabbit models [45,48]. IBS (Lifetech Scientific, Shenzhen, Guangdong, China) presented a strut thickness of 70 µm, density of 115 µg/cm 2 , diameters of 2.75-3.0 mm and 15 mm in length. The IBS crossing profile was 1.04 mm and maximum diameter increase of 0.75 mm for 3.0 mm device. In total, 48 swines that received IBS were assessed by QCA, OCT, and histology, and their results were compared to the same number of subjects receiving Xience (Abbott Vascular, Santa Clara, CA, USA) at 7 days to 6 months follow-up. IBS results showed no significant difference in area of stenosis compared to Xience up to 6 months, while scaffold or stent thrombosis was observed during the study [45].

Conclusions
Almost 20 years since Tamai et al., the BRS technology concept has evolved, and current devices are aimed to overcome second-generation DES historical issues by promoting a transient scaffolding of the vessel wall, thus recovering vasomotion in the previously scaffolded segment and allowing reintervention of the target segment either by PCI or CABG if needed. This promise was partially achieved by sirolimus-eluting resorbable magnesium scaffold (Magmaris) with a relatively low TLF rate and no ST events reported in Biosolve-II and -III trials clinical outcomes report at 24 months, which was not observed in the polymeric BRS devices. In addition, a multivariate analysis comparing Magmaris and Orsiro clinical outcomes suggested similar TLF rates at 12 months, and there are no randomized studies comparing Magmaris clinical outcomes 2nd Generation DES. Lastly, in the newer generation RMS, DREAMS 3G is currently under development and its performance remains to be seen. While the iron-alloy BRS is emerging as a promising new technology, the in-man performance of these devices remains to be seen.