Improved vascular response to a novel polymeric blend PLLA/PLGA coronary scaffold material: preclinical study

Abstract

Thick struts and polymeric crystalline material are the potential mechanism of failure of the first generation bioresorbable scaffolds (BRS). We evaluated a novel fully amorphous bioresorbable scaffold (BRS) made of poly-L-lactide/poly-L-glycolide (PLLA/PLGA, Apollo, Biostent Consortium, Poland) in a porcine coronary restenosis model. Uncoated BRS (PLLA 120 μm, n = 7; PLLA 150 μm, n = 8; PLLA/PLGA 200 μm, n = 10) were implanted in 12 swine using OCT guidance and 110% overstretch. Follow-up included angiography, OCT, and histopathology at 30 and 90 days. At 30 days, OCT showed no significant differences in stenosis or neointimal hyperplasia between groups as represented by percent area stenosis (PLLA 120: 41% ± 17, PLLA 150: 52 ± 20%, PLLA/PLGA: 55 ± 6%; p = 0.07 ) and neointimal area (PLLA 120: 2.01 ± 0.8 mm², PLLA 150: 2.57 ± 1.1 mm², PLLA/PLGA: 2.02 ±0.8 mm²; p = 0.39). The average number and proportion of struts with no inflammation (score 0) was 7.2 fold higher in the PLLA/PLGA when compared to PLA 120 (p = 0,03). The endothelialization was nearly complete and comparable in all groups. At 90 days, PLLA/PLGA remained patent, covered, and free of restenosis, with positive remodeling and late lumen enlargement only in PLLA/PLGA group (lumen area 30 vs. 90 day: 1.86 vs. 3.40 mm², p = 0.02). The novel PLLA/PLGA BRS demonstrated improved healing, reduced inflammation and positive remodeling, supporting its potential for next-generation BRS.

Introduction

The concept of bioresorbable scaffolds (BRS) represents a pivotal innovation in interventional cardiology, aiming to provide transient vessel support with the promise of leaving no permanent implant behind. Since their inception, first-generation BRS have been lauded for their potential to reduce long-term complications associated with metallic stents, such as chronic inflammation and late thrombosis. Despite the revolutionary technology, clinical data accumulated over the past decade has illuminated substantial challenges, including higher rates of scaffold thrombosis and restenosis compared to their metallic counterparts¹. These failures have been attributed to factors such as scaffold design, material properties, and strut thickness, leading to a re-evaluation of BRS technology². Particularly, strut thickness of 150 micron and more has been opted for one of the most significant limitations of BRS, contributing to inferior vessel healing. However, the material crystallinity also remains a major limitation and cause of excessive inflammation, historically seen in orthopedics^3,4. Therefore, the pursuit of less crystalline and more amorphous material, yet meeting the mechanical criteria has been one of the technological goals. Acknowledging these limitations, the field has witnessed the emergence of novel-generation scaffolds, which offer innovations such as thinner struts and enhanced biocompatible materials, aiming to improve endothelialization and reduce biomechanical stress⁵. Early studies on these novel scaffolds have shown promising results, indicating improvements in deliverability, vessel wall apposition, and reduced inflammatory response. However, the ultimate goal of BRS technology—to leave nothing behind without compromising the efficacy and safety of the procedure when compared to contemporary metallic drug eluting stent—remains a challenging frontier. As we stand on the cusp of a new era in scaffold development, there is a clear and present need to refine our approach to bioresorbable technology. This includes rigorous exploration into novel materials that can provide adequate support with less or no inflammation and more controlled, likely faster resorption rates, and extensive clinical studies to validate their long-term safety and effectiveness. Herein, we evaluate the novel, fully amorphous bioresorbable material – polylactide and polyglycolide blend for construction of coronary scaffold and compare them with currently available crystalline polylactide scaffolds in the porcine coronary in-stent restenosis model.

Methods

Study device description

The studied BRS (Apollo, Biostent Consortium, Poland) is a balloon expandable stent manufactured from a proprietary, fully biodegradable, and amorphous polymer (Fig. 1), made of polylactide (PLA) and polyglycolide (PGA). This development was made by adding 50% weight PLLA (50/50 PLGA (85/15) / PLLA) homopolymer fraction in order to fix the target shape in the post-production. Thanks to the addition of a polyglycolide, the degradation time needed to achieve positive remodeling of the vessel after implantation was shortened and is estimated at one year in-vitro by means of hydrolysis⁶. The PLLA/PLGA stents under this process and proprietary annealing (patent: PL240665B1) increased the radial force up to 20%. The in-vitro data suggest also earlier degradation estimated at 1.5-2 years (Data on file Biostent consortium). The strut thickness is 200 μm. Stent sizes included in this study were 3.0 and 3.5 in diameter and 15 mm length.

Fig. 1.

Optical light microscopy. Upper panel depicts PLLA crystalline stent and strut, lower panel depicts amorphous PLLA/PLGA stent and strut. PLLA – poly-L-lactide, PLGA – poly-L-glycolide.

Control devices description

The control BRS are made from poly(L-lactide) (PLLA) bioabsorbable polymer tubes (Zeus Industrial Products, Inc.) either made from 150 to 120 μm. Stents were cut from tubes according to the structure design developed during the research, using a femtosecond laser, under a gas shield, in the cold, proprietary annealed and crimped over the balloon. BRS production process was the same in control and study device. The design shape is similar and material is identical to the the Absorb stent (150 micron) and Esprit (Abbott, Santa Clara, California, USA.

Study design

A study scheme is presented in Fig. 2. A total of 12 domestic swine (Sus scrofa domesticus) of both sexes were enrolled in the study. Arterial segments, without side branches of all three coronary arteries (RCA, LAD, LCX) were screened for stent implantation. After the quantitative coronary analysis (QCA), 25 segments were eligible for the study. Uncoated BRS were implanted with optical coherence tomography (OCT) guidance and 110% overstretch. These included three BRS types according to polymer and strut thickness PLLA/PLGA 200 mm (n = 7), PLLA 150 mm (n = 6), and PLLA 120 mm (n = 7) were implanted for 30 days, and PLLA/PLGA 200 mm (n = 3), PLLA 150 mm (n = 2) for 90 days. Subsequently, control coronary angiography with OCT was performed and swine were sacrificed. All arterial segments were dissected and harvested for pathological analysis.

Fig. 2.

Study schematic. The flowchart depicts study protocol, including methods used, number of animals, studied groups and follow-up. PLLA – poly-L-lactide, PLGA – poly-L-glycolide, OCT – optical coherence tomography, DFU – days follow-up.

Experimental procedures and in-stent restenosis model

The study was reviewed and approved by an external ethical committee (approval no 3/2018) for compliance with regulations prior to study initiation. All animals received standard of care outlined in the study protocol and in accordance with the Act of Animal Welfare and the “Principles of Care of Laboratory Animals”⁷. The animals were loaded with double anti-platelet therapy 3 days before intervention. After an overnight fast, animals were anesthetized using a combination of Ketamine (10–20 mg/kg, IM) and Xylazine (0.05–0.2 mg/kg, IM), and Atropine (0.02–0.05 mg/kg, SC). After the animal was sedated, it was transported to the preparation room where intravenous lines were placed in the auricular marginal veins, and intravenous fluids (Ringers or balanced crystalloid solution) were administered throughout the procedure. ECG electrodes were applied over the shaved metacarpal and metatarsal zones to monitor ECG throughout the procedure. Anesthesia was maintained using 1–3% Isoflurane with 100% oxygen, inhaled, or Propofol 0.1–0.2 mg/kg/min, IV. A bolus of Fentanyl was administered to increase depth of general anesthesia and provide analgesia. Heart rate, respiration rate, O₂ saturation, and invasive blood pressure were monitored and recorded in the anesthesia form at regular intervals. The animal was placed in dorsal recumbency, and the hair was removed from the access areas. A vascular sheath (6F) was placed in the right or left femoral artery utilizing a Seldinger technique. A heparin bolus (100–150 IU/kg, IV) was administered to achieve an ACT > 250 seconds prior to first treatment delivery. Angiographic images of the vessels were obtained using contrast media. QCA was performed to identify and measure the prospective deployment site. Target stent diameter was set to be ~ 110% of the average baseline target segment diameter. Following final angiography and OCT was performed, and the catheter and sheath were removed. At the designated time points (30 and 90 days), animals were anesthetized and prepared as described for device implant. Pre-sacrifice angiography and OCT were performed. At terminal follow up, animals were euthanized while under general anesthesia by IV injection of euthanasia solution (pentobarbital sodium + pentobarbital – 133.3 mg + 26.7 mg mL).

Optical coherent tomography (OCT)

The OCT images of all tested coronary segments were acquired at 30 and 90 days follow up with a non-occlusive technique using the LUMIEN™ imaging system (St. Jude Medical, St. Paul, MN). The OCT catheter was pulled back at 20 mm/s and OCT images were generated at 100 frames/s. Contrast medium (OMNIPAQUE 350; GE Healthcare) was continuously flushed through a guiding catheter during imaging catheter pullback to ensure blood clearance from the coronary arteries. Cross-sectional OCT images were analyzed at 1 mm intervals. Reference vessel area was calculated at 5 mm distance from proximal and distal edge of stent. Neointimal area (NA) was calculated by subtracting the lumen areas (LA) from the stent areas (SA). Percent area stenosis was calculated with the following formula: (neointimal area / stent area) x 100%, whereas percent diameter stenosis was calculated with: [(average stent diameter – average minimal lumen diameter) / vessel reference diameter] x 100%.

Quantitative coronary angiography (QCA)

QCA analysis was performed utilizing QAngio XA Software (Medis Medical Imaging Systems, Leiden, The Netherlands). In the baseline procedures the following parameters were measured using the guiding catheter as a standard for calibration: minimal lumen diameter (MLD) within the treated segments, reference diameter (RD) measured at the proximal and distal portions of the treated sites, and the balloon and balloon-to-artery ratio. Ind the follow-up points the percentage of artery stenosis (%AS) and diameter stenosis (%DS) were calculated as: (reference lumen area—minimum lumen area)/reference lumen·100% and (reference vessel diameter—MLD)/reference diameter·100%) and the late loss was calculated as (MLD at follow-up - MLD at post-implantation).

Histological analysis

Following vessel harvesting, stented segments were fixed in 10% normal buffered formalin (NBF). For light microscopy all treated vessels were embedded in methyl-methacrylate and then 40–50 micron sections from the proximal, mid and distal portion of each stented segment were obtained. These sections were stained with hematoxylin and eosin (H&E). All sections were evaluated using semi-quantitative scoring criteria. To evaluate the extent inflammatory reaction the following grade was used: 0 - ¼ wires < 10 cells, 1 - up to ¼ wires > 10 cells, 2 - up to ½ wires > 10 cells, 3 - more than ½ wires > 10 cells, 4 − 2 or more granulomatous foci. To evaluate the vessel damage, a respective criteria were utilized: 0 - Preserved intimal elastic membrane; tunica media compressed, not torn, 1 - Ruptured intimal elastic membrane; tunica media compressed, not torn, 2 - Ruptured intimal elastic membrane tunica media interrupted, tunica elastica externa compressed, 3 - All of the above, as well as tunica elastica externa interrupted. Neointimal Maturity Score estimates the proportion of neointimal areas containing smooth muscle cells (SMC): 0 - No SMC in the tunica intima, 1 - Up to 25% SMC in the tunica intima, 2–25-75% SMC in the tunica intima, 3 - >75% in the tunica intima; Endothelial Score was assessed with regards to percentage coverage of artery circumference: 1 - <25%, 2 -25-50%, 3–50-75%, 4 - >75%. Each strut in the section was scored and the mean inflammation and injury score for each section was calculated and reported.

Statistical analysis

Non-parametric data were expressed as number and percentage. Normally distributed parametric data are expressed as average and standard deviation, non-normally distributed as median and interquartile range (IQR). All treatment groups were tested for normality with the Shapiro-Wilk test and were followed by either an ordinary one-way ANOVA for normally distributed data or the Kruskal-Wallis test for non-normally distributed data. Tukey’s post hoc multiple comparisons test was performed for the former and Dunn’s post hoc multiple comparisons test for the latter. The type 1 error rate for multiple comparisons was corrected by specifying the significance at 0.05 in Tukeys test, which corrects for Type 1 Error rate. Dunn’s Test is adjusted with the Bonferroni method.

The significance level was chosen at p < 0.05. Statistical analyses were performed utilizing MedCalc version 22.016 and Statistica v.12.0.

Ethical compliance

The study was approved by the Local Ethics Committee for Animal Research, Decision No. 1140/2015, conducted in compliance with the Animal Welfare Act and the “Guide for the Care and Use of Laboratory Animals,”⁷ and reported according to ARRIVE guidelines.

The animals used in the study were obtained from Jerzy Smolarz, Łukasiewicza 5, 43–241, Łąka, Poland., a certified supplier complying with all relevant ethical and welfare standards.

Results

All animals (12) were successfully implanted with BRS and survived uneventfully up to 30 (n = 9) or 90 (n = 3) days. In one case there was a need for a distal overlap implantation of a bare metal stent (BMS) due to severe spasm and dissection. This vessel was excluded from analysis. Besides, there were no other adverse events reported on the treatment day, neither during the observation period.

QCA results

Average RVD, balloon RD and the overstretch proportion were similar in all groups. Post-stent deployment, all devices were noted to have deployed evenly with no dissections. At 1 and 3 months the MLD was statistically comparable between groups (Fig. 3). However, %DS at 30 days was the lowest in PLLA 120 group (PLLA 120: 31% ± 12, PLLA 150: 34% ± 14, PLLA/PLGA: 52 ± 14; ANOVA p = 0.01, PLLA 120 vs. PLLA/PLGA p = 0.02, mean difference = -21 95% CI -39% -3% ). Conversely, at 90 days a markedly lower %DS was reported in PLLA/PLGA 200 when compared with PLLA 150 (38%±8 vs. 62%±0.2 p = 0.03, mean difference = -24% 95% CI -42% - -6%), Table 1.

Fig. 3.

Representative cross-sectional images of bioresorbable stents – upper panel – pathology, mid panel – peri strut magnification, lower panel– optical coherence tomography. All vessels remained patent. All struts examined were fully covered by neointima in all groups. No malapposition, stent fracture, or late strut discontinuity was observed in any of the groups. PLLA – poly-L-lactide, PLGA – poly-L-glycolide.

Table 1.

Quantitative coronary angiography results at baseline and after implantation as well as at 30 and 90-days follow up.

FU Time [days]	Stent	Unit [mm]	Quantitative coronary angiography (QCA)
			Baseline				Post implantation		Follow up
			MLD	RD	Ballon RD	Overstrech proportion	MLD	RD	MLD	RD	%DS	LL
30	PLGA/PLLA 200	mean	2.14	2.46	2.33	1.08	1.84	2.23	1.08	2.30	51.94	0.76
		SD	0.21	0.21	0.30	0.14	0.38	0.24	0.29	0.45	13.78	0.53
	PLLA 120	mean	2.04	2.40	2.29	1.15	1.80	2.45	1.41	1.96	30.95	0.40
		SD	0.39	0.40	0.13	0.19	0.39	0.23	0.27	0.38	11.83	0.52
	PLLA 150	mean	2.24	2.64	2.35	1.09	1.93	2.32	1.31	1.97	33.78	0.62
		SD	0.52	0.40	0.34	0.23	0.36	0.30	0.38	0.41	14.36	0.63
		p value	0.63	0.57	0.91	0.76	0.83	0.28	0.15	0.23	0.01	0.47
90	PLLA/PLGA 200	mean	2.54	2.45	2.43	0.99	1.72	2.56	1.51	2.29	37.65	0.21
		SD	0.66	0.03	0.06	0.22	0.62	0.28	0.23	0.42	7.69	0.77
	PLLA 150	mean	2.42	2.69	2.31	0.96	2.11	2.59	1.08	2.81	61.56	1.03
		SD	0.21	0.17	0.10	0.04	0.12	0.11	0.23	0.59	0.16	0.11
		p value	0.79	0.29	0.29	0.79	0.39	0.90	0.16	0.42	0.03	0.21

OCT results

Following deployment of the devices, OCT was performed in 11 animals (no baseline OCT analysis was performed in one animal due to the imaging system malfunction; also, the vessel with implanted BMS was excluded from the analysis. A malapposition concerning single struts was present in 1 PLLA/PLGA 200 scaffold. There was no other malapposition noted in any implanted vessels. Also, a scanty thrombus was seen in 3 PLLA 150 and in 2 PLLA/PLGA 200 scaffolds, that did not impair the blood flow and was not present in the follow-up OCT imaging.

At 30 days the stent areas were slightly, but significantly lower in the PLLA/PLGA group (PLLA 120 : 4.97 ± 0.2 mm², PLLA 150: 4.85 ± 0.5 mm², PLLA/PLGA: 3.88 ± 0.94 mm²; PLLA 120 vs. PLLA/PLGA p = 0.01, mean difference = 1.09 95% CI: 0.23–1.94; PLLA150 vs. PLLA/PLGA p = 0.03, mean difference 0.98 95% CI: 0.09–1.87). The lumen area was significantly higher in the PLLA 120 but not in PLLA 150 (PLLA 120: 2.95 ± 0,9 mm², PLLA 150: 2.28± 0.9 mm², PLLA/PLGA: 1.86 ± 0.6 mm², p = 0.03, PLLA 120 vs. PLLA/PLGA p = 0.04). There were no differences between percent area stenosis (PLLA 120: 41% ± 17, PLLA 150: 52 ± 20%, PLLA/PLGA: 55 ± 6%; p = 0.07), percent diameter stenosis (PLLA 120: 24 ± 13%, PLLA 150: 33 ± 16%, PLLA/PLGA: 34 ± 5%; p = 0.07) as well as neointimal area (PLLA 120: 2.01 ± 0.8 mm², PLLA 150: 2.57 ± 1.1 mm², PLLA/PLGA: 2.02 ±0.8 mm²; p = 0.39) (Fig. 4).

Fig. 4.

Optical coherence tomography results at 30 and 90 days follow-up. At 30 days, OCT showed no significant differences in stenosis or neointimal area between groups. PLLA/PLGA exhibited less inflammation and near complete endothelialization. At 90 days, PLLA/PLGA remained patent, covered, and free of restenosis, with positive remodeling. PLLA – poly-L-lactide, PLGA – poly-L-glycolide, DFU – days follow-up.

At 90 days, the reported stent area (7.84 ± 2.1 mm² vs. 5.32 ± 0.1 mm², p = 0.20, mean difference = 2.52, 95% CI: -2.39–7.43) and consequently neointimal area (4.44 ± 0.98 mm² vs. 2.19 ± 0.02 mm², p = 0.05, mean difference = -2.25, 95% CI -2.25–0.07) were significantly greater in PLLA/PLGA 200 group as compared to PLLA 150. However, the %AS was comparable between the PLLA/PLGA 200 and PLLA 150 (55 ± 8% vs. 41 ± 1%; p = 0.09) (Fig. 4). A temporal (30- and 90-day follow-up) stent and lumen area enlargement was reported in OCT exclusively for PLLA/PLGA (SA: 3.91± 1 vs. 7.84 ± 2 mm², p < 0.01, mean difference = 3.96 95% CI 1.87–6.06; LA: 1.86 ± 1 vs. 3.40 ± 1 mm², p = 0.02, mean difference = 1.54 95% CI: 0.27–2.82, Fig. 4).

Histopathology results

Vessel injury

At 30 days the average in-stent injury score was low for PLLA 120 and PLLA/PLGA 200 indicating a good apposition of the struts to the internal elastic lamina (Fig. 5). There were only focal struts that penetrated IEL and were contacting the media, typically within a single quadrant, with some interrupting the media (in PLLA 120 group). A significantly larger vessel injury was observed in PLLA 150 group with most of the struts penetrating IEL and majority of them interrupting the media. The 90-day cohort expressed higher levels of injury, comparable between groups, with some struts contacting the tunica elastica.

Fig. 5.

Qualitative Histopathologic Assessment of stent efficacy at the 30- and 90-day follow-up. PLLA120 (n = 7), PLLA 150 (n = 6) and PLLA/PLGA 200 (n = 7) showed low injury score at 30 days with lesser inflammation observed in PLLA/PLGA scaffolds at 90 days. A – Inflammation Score; B – Vessel Damage; C - Neointimal Maturity score; D - Endothelial Score. All Graphs are represented as median ± IQR. ** - p < 0.01. PLLA – poly-L-lactide, PLGA – poly-L-glycolide, DFU – days follow-up.

Vessel healing

Good vessel wall healing at both time points was characterized by majority (30 days) or almost entirety (90 days) of vessel being covered by fully matured neointima with advanced reendothelialization present at all sites (Fig. 6). No thrombus was observed in any vessel.

Fig. 6.

Qualitative Histopathologic Assessment of stent induced inflammation at the 30-day follow-up of the PLLA120 (n = 7), PLLA 150 (n = 6) and PLLA/PLGA 200 (n = 7). Each stent was assessed at 6 or 7 slices at each strut, and the results were summed up for each score. Values are presented as means and error bars represent the standard deviation. At each score the data was checked for normality, followed by either ANOVA or Kruskal-Wallis analysis with Tukey’s or Wilcoxon’s post hoc tests respectively. * - p < 0.05. PLLA – poly-L-lactide, PLGA – poly-L-glycolide.

Device biocompatibility

Overall, the inflammation scores were low for all vessels (characterized by limited infiltration with lymphocytes, macrophages, and multinucleated giant cells) on both time points except for PLLA 150 cohort. At 30 days the mean inflammation score tended to be lower by 60% in PLLA/PLGA when compared to PLA 150 and comparable to PLLA 120 (1.8 ± 1.2 vs. 3 ± 1.2. vs. 1.7 ± 1.1 ; p = 0.11 Fig. 5). The average number and proportion of struts with no inflammation (score 0) was 7.2 fold higher in the PLLA/PLGA when compared to PLA 120 (p = 0.02, Fig. 6). The endothelialization was advanced and comparable between groups both at 30- and 90-day follow-up. Similarly, neointima was considered mature in all groups both at 30 and 90 days. In the PLLA 150, the substantial inflammation with eosinophil-rich granuloma formation was observed in the stented vessel segments at both time points (Fig. 7).

Fig. 7.

Histopathology evaluation. Staining H&E. S = strut; L = lumen; M = muscular layer; White double arrow = neointimal layer. A. PLLA/PLGA200 30 DFU. The number of lymphocytes at the proximity of strut is lower than 5 cells - inflammation grade 0. Magnification 100-fold. B. PLLA120 30 DFU. Strut is surrounded by neointimal layer containing scanty dispersed lymphocytes – inflammation grade (1) Magnification 100-fold. C.PLLA150 30 DFU. Struts are surrounded by dense dark confluent lymphocytic infiltrations. Histiocytic aggregation with large multinucleated giant macrophages is encircled yellowish – inflammation grade 4. Muscular layer intact. Magnification 50-fold. D. PLLA/PLGA200 90 DFU. Struts are surrounded by fibrous neointima. Scanty clusters of lymphocytes (red encircled) – inflammatory grade 1 to (2) Muscular layer intact. Brownish deposits of hemosiderin are present. Magnification 100-fold. E. PLLA150 90 DFU. Struts are surrounded by dense dark confluent lymphocytic infiltrations. Histiocytic aggregation with two giant multinucleated macrophages located between struts – inflammation grade 4. Muscular layer intact. Magnification 50-fold. PLLA – poly-L-lactide, PLGA – poly-L-glycolide, DFU – days follow-up.

Discussion

First generation BRS were clinically and biologically inferior to metallic counterparts due to multiple factors. Their high strut thickness plays important role in creation of more significant vessel wall injury and flow disturbances which result in platelet and thrombus accumulation as well as increased neointimal formation and restenosis^8,9. Additionally, the stent strut material, highly crystalized PLA caused stent discontinuities, increased inflammation, prolonged and uncontrolled biodegradation¹⁰. Implantation imperfections like vessel under or oversizing, and inadequate expansion also played a significant role in early and late events¹¹.

Therefore, in our study, we evaluated the biological response to a novel, amorphous polymeric blend coronary scaffold in an in-stent coronary restenosis model of a domestic swine as the one that most closely resembles human coronary anatomy, hemodynamics, hemocompatibility vascular response and physiology¹². The novel material was compared head-to-head with current golden standard - the PLA scaffolds made either from 120- or 150-micron struts. All devices were uncoated. Therefore, the outcome is unbiased by the drug release or surface polymer coating. All implantations were feasible and successful. The radial stent was sufficient as assessed by the implantation force and angiographic outcome. At 28 days, the surrogates of restenosis - the neointimal formation (%AS, %DS) were similar between PLLA/PLGA and PLA 150 groups, with a trend toward numerically lower numbers in the PLA 120-micron stents. At 90 days there was a notable lumen enlargement solely in the PLLA/PLGA group in contrary to PLA counterparts and first generation Absorb scaffold¹³. Importantly despite higher strut thickness the average number of struts per stent with no inflammation was highest in the PLLA/PLGA group when compared to PLA 150 and PLA 120 (19.5 vs. 14.1 vs. 2.7, p = 0.03). In contrary, there was a trend toward lowest number of struts with highest inflammation and granulomatous formation in the PLLA/PLGA blend (4.1 vs. 4.2 vs. 14.3 p = 0.07). These outcomes were sustained up to 90 days follow-up, with lower inflammation and vessel injury in the 90 day follow up.

The potential rationale for decreased inflammation in the PLLA/PLGA blend are multiple. Although PLA can be generally considered a safe biomaterial, various reactions can occur depending on the actual composition and production method of the poly-L-lactide (co)polymer. PLA/PGA (co)polymers can be tailored to suit various applications. This, however, implicates also that for each PLA formulation and/or production technique the final resulting material or product will have to be evaluated for its specific safety characteristics. Therefore, it is important to evaluate final stent platform, not material itself, as was made in current study.

The difference in vessel healing responses between crystalline PLA and amorphous PLLA/PLGA blend stents can be attributed to several material properties and their interactions with the biological environment. The degradation rate, surface properties, hydrophilicity, mechanical properties as well an immune response may play a potential role.

1. Degradation rate.

Although not fully evaluated in current study, crystalline PLA generally has a slower degradation rate compared to amorphous PLLA/PLGA blends. This slower degradation can prolong the presence of the stent in the vessel, potentially leading to a prolonged inflammatory response.

Amorphous PLLA/PLGA blends through hydrolysis of their ester bonds degrade more quickly and in a more controlled manner, which can reduce the duration of the inflammatory response and allow for more rapid tissue healing and integration as it degrades faster¹⁴.

• 2.Surface properties:

Crystalline structures tend to be more rigid and less flexible, which can lead to mechanical irritation of the vessel wall thus triggering inflammation. This is confirmed in the clinical the ABSORB II clinical trial, in which first generation crystalline PLA scaffold did not improve vasomotion¹³.

Amorphous materials are generally more flexible and conformable, reducing mechanical irritation and promoting better interaction with the vessel tissue¹⁵.This is important in relation to implantation in coronary arteries, which may facilitate restoration of vasomotion and relaxation as shown in report by Stefan Verheye et al.¹⁶with the Bioadaptor stent.

• 3.Hydrophilicity and drug release.

Crystalline PLA might have less predictable biological response and biodegradation due to its dense and less permeable structure. Amorphous PLLA/PLGA blends often have better hydrophilicity, which can enhance the initial cell attachment and proliferation. It is due to dominated content of the amorphous phase and lower hydrophobicity generated by glycolic acid. Although not evaluated in current study, the drug release profiles from these materials tend to be more predictable and controlled. Crystalline PLA might have less predictable and less efficient drug release profiles due to its dense and less permeable structure^17,18.

• 4.Mechanical properties.

Crystalline polymers are more brittle and may cause micro-injuries to the vessel wall, exacerbating the inflammatory response.

Amorphous polymers generally exhibit better mechanical properties for vascular implants, such as flexibility and toughness, which can reduce the risk of stent-induced injury and subsequent inflammation¹⁹.

• 5.Immune response:

Crystalline regions of PLA can stimulate a higher immune response due to their rigidity and density, leading to more pronounced inflammation.

Amorphous regions are more biocompatible and can be processed by the body more efficiently, resulting in a milder immune response^20,21 which has been proven in our study.

In summary, the above mentions factors could have collectively contributed to the improved vessel healing observed with amorphous PLLA/PLGA blend stents compared to those made from crystalline PLA. The balance between degradation rate, mechanical properties, surface characteristics, and immune response is crucial in determining the overall biocompatibility and effectiveness of stent materials in promoting vessel healing and reducing restenosis.

The lumen enlargement observed in the PLLA/PLGA blend is similar to this observed previously in literature with PLA counterparts. Although mechanism is not fully understood we believe it is similar as PLA is a major component of our polymeric blend. Therefore, the molecular breakdown play a major role in which over weeks to months, the polymer chains break down into smaller molecules such as lactic acid or glycolic acid, which are metabolized naturally by the body. The loss of structural integrity weakens the stent, and allows vessel adaptive remodeling. The controlled degradation phase encourages adaptive growth of the vessel wall, reinforcing the lumen without compromising vessel compliance.

From the regulatory point of view we believe this material will not require any additional requirements when compared to PLA. These standards for all bioresorbable scaffolds were most lately described in the EAPCI task force²² document and we believe they will apply as well to this type of polymer.

To the best of our knowledge this paper reports for the first time improved vessel healing with utilization of the novel polymeric material as represented by lower inflammation and vessel injury score regardless of polymeric strut thickness. Overall, the biological response to a thicker PLLA/PLGA blend was similar to a PLA 120-micron stent. Methodologically this was achieved by utilization of similar stent designs, but different polymeric materials and strut thicknesses. As novel BRS technologies focus mainly on strut thickness, herein we present that polymer material improvement could also be crucial for improved biological and clinical outcomes. Therefore, clinical implications could be important and hint crucial for the development of future resorbable technologies.

Limitations

The limitations of this study include the nature of an experimental model as a human clinical surrogate and utilization of a healthy domestic swine, without underlying atherosclerotic disease. Although determined to evaluate safety and vascular response, the efficacy analysis and full translation into clinical setting is limited. The ninety-day follow-up sample is modest, which goal was to primarily evaluate feasibility and safety. The strut thickness of studied stent is high, yet despite expected higher injury the inflammation was lower. However, the main goal of the study was to evaluate influence of the material, not thickness. Further development of PLLA/PLGA blend scaffold is warranted to achieve significantly thinner struts.

Conclusions

Basing on the current findings, we conclude that improved vessel healing, with lesser inflammation were shown in a novel amorphous PLLA/PLGA BRS material when compared to thinner strut PLA counterparts. In contrary to PLA platforms, positive remodeling and lumen enlargement was shown in the studied, novel PLGA/PLA platform. Therefore, not only strut thickness, but also polymeric material developments can result in improved biological response to a novel BRS technologies, which deserve further clinical evaluation.

Impact on daily practice

Based on this experimental study, the novel amorphous material PLLA/PLGA may offer the improved healing, with lesser inflammation with regard to PLA BRS with lower strut thickness. Further material development may translate into improved biological response in the clinical setting and implementation of novel BRS.

Abbreviations

BRS

Bioresorbable scaffold

PLLA

Poly-L-lactide

PLGA

Poly-L-glycolide

QCA

Quantitative coronary angiography

OCT

Optical coherence tomography

MLD

Minimal lumen diameter

RVD

Vessel reference diameter

%AS

Percentage of area stenosis

%DS

Percentage of diameter stenosis

Author contributions

Wrote the main manuscript text: PP.B, MK, KŁ, MGPrepared the manuscript: KŁ, MM, MK Reviewed the manuscript: PK, KM, JK, JMConceived the study concept: PP.B, JN, RS, MS, JK, JM, KSConducted final evaluation and provided supervision of the work: KM, PE.B.

Funding

The project financed by The National Centre for Research and Development; grant no StrategMed2/269760/1/NCBR/2015).

Data availability

The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request.

Competing interests

The authors declare no competing interests.