Cytokines release inhibition from activated monocytes, and reduction of in-stent neointimal growth in humans

Atherosclerosis 211 (2010) 242–248

Contents lists available at ScienceDirect

Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis

Cytokines release inhibition from activated monocytes, and reduction of in-stent neointimal growth in humans Gabriele Pesarini a , Angela Amoruso b , Valeria Ferrero a , Claudio Bardelli b , Luigia Grazia Fresu b , Laura Perobelli c , Paolo Scappini c , Giuseppe De Luca b , Sandra Brunelleschi b , Corrado Vassanelli a , Flavio Ribichini a,∗ a

Department of Biomedical Sciences and Surgery, University of Verona, Division of Cardiology, Verona, Italy Department of Medical Sciences, University of Piemonte Orientale “A. Avogadro”, Novara, Italy c Laboratory of Clinical Biochemistry and Haematology of the Ospedale Civile Maggiore, Verona, Italy b

a r t i c l e

i n f o

Article history: Received 14 June 2009 Received in revised form 19 January 2010 Accepted 1 February 2010 Available online 10 February 2010 Keywords: Cytokines Monocytes Inflammation Prednisone Restenosis

a b s t r a c t Objective: Atherosclerosis and restenosis are largely ruled by inflammation. The aim of this study was to test the effects of a short-course, high-dose oral prednisone on the release of interleukin-6 (IL-6) and tumour necrosis factor (TNF)-␣ from circulating monocytes and on the neointimal growth that follows bare metal stent (BMS) implantation. In a sub-group of patients activated NF-␬B was also evaluated. Methods: Out of 40 patients with coronary artery disease treated with BMS implantation, 20 were randomly assigned to receive oral prednisone during 40 days according to a standardized protocol. In non-stimulated and stimulated (LPS and PMA) monocytes we evaluated the release of IL-6 and TNF␣, and NF-␬B p50 subunit translocation at baseline, at 10 and 30 days. Late luminal loss (LLL) 9 months after angioplasty was calculated by quantitative coronary angiography. Results: Plasma concentrations of prednisone correlated inversely with IL-6 and TNF-␣ release (R2 = 0.45, p = 0.04 and R2 = 0.69, p = 0.005, respectively) and NF-␬B activation from monocytes (R2 = 0.58, p = 0.01). The reduction of TNF-␣ release and NF-␬B activation were significantly related (R2 = 0.56, p = 0.01). Prednisone patients showed a significantly larger reduction of cytokine release and NF-␬B activation compared to non-treated patients, at 10 days and 30 days. LLL was lower in the prednisone group (0.44 ± 0.35 mm versus 0.80 ± 0.53 mm, p = 0.02) and correlated with reduction of TNF-␣ (R2 = 0.41, p = 0.01). Conclusions: High doses of oral prednisone reduce NF-␬B pathway activation and pro-inflammatory cytokine release in circulating activated monocytes of patients treated with coronary stenting. TNF-␣ release reduction correlates with decreased LLL. © 2010 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Atherosclerosis is a systemic disease that can affect the coronary district as well as other vascular compartments, whose initiation, progression and complications are largely ruled by inflammatory mechanisms [1]. In fact, circulating levels of inflammatory mediators, mainly released by monocyte/macrophages, have been shown to play a determinant role in atherosclerosis [2,3]. Therefore, focusing on inflammatory pathways as therapeutic targets to prevent the development and progression of atherosclerosis is a soundly

∗ Corresponding author at: Catheterization Laboratory of the University of Verona, Ospedale Civile Maggiore, Piazzale A. Stefani 1, 37126, Verona, Italy. Tel.: +39 045 812 2039; fax: +39 045 802 7307. E-mail addresses: fl[email protected], fl[email protected] (F. Ribichini). 0021-9150/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2010.02.004

approach [4]. In the specific setting of coronary artery disease (CAD) the efficacy of systemic immunosuppressive or anti-inflammatory drugs to down-regulate pro-inflammatory mechanisms and reduce restenosis and ischemia has been demonstrated, respectively, in animal models [5] and in humans [6–8]. In particular, the use of corticosteroids bears the capability of inhibiting inflammation, mainly acting on the activation of the NF-␬B pathway [9,10], consequently modulating the vascular response to the injury that follows percutaneous coronary interventions (PCI) with minor and predictable side effects [8]. Steroids easily diffuse through cell membranes and bind to high-affinity cytoplasmatic glucocorticoid receptors. The receptor-steroid complex migrates to the nucleus, where it binds to multiple sites and regulatory elements of DNA, modulating genetic transcription and inhibiting the effects of several factors, like the activated form of NF-␬B (p65/p50 heterodimer). These actions cause transcriptional regulation, thereby altering the expression of genes involved in the immune and inflammatory response, in

G. Pesarini et al. / Atherosclerosis 211 (2010) 242–248

particular cytokine release [9,10]. Glucocorticoids affect also the number, distribution, and function of all types of leukocytes, reducing the accumulation of monocytes and macrophages, neutrophils and eosinophils at the sites of inflammation [8]. However, little is known about the effective drug concentration and the capability of in vivo inhibiting pro-inflammatory cytokines, and how this action may translate into the phenotypic response of reducing neointimal growth after coronary stenting. This study was therefore aimed to: (a) assess the plasmatic concentration ranges of prednisone given orally twice daily; (b) to assess a possible correlation between the plasmatic concentration of the drug, the NF-␬B pathway inhibition and the release of cytokines by activated monocytes of patients treated with PCI and implantation of bare metal stents (BMS); and (c) to assess a possible correlation between the degree of cytokine release inhibition and neointimal proliferation after coronary stenting by quantitative coronary angiography (QCA) at 9–12 months.

2. Methods 2.1. Population and study protocol Patients analysed in this study belong to a largest independent, multi-centre randomized study entitled “Cortisone plus BMS or DES versus BMS alone to eliminate restenosis (CEREA-DES)”, whose protocol is published elsewhere [11]. In brief, the study is aimed at comparing the clinical outcome of non-diabetic patients with CAD after one year of a randomly assigned treatment to either stenting with BMS, stenting with drug-eluting stents, or stenting with BMS followed by oral therapy with prednisone for 40 days. Main exclusion criteria were: age older than 80 years, diabetes mellitus, recent Q-wave myocardial infarction (<2 weeks), renal insufficiency (creatinine >2.5 mg/dl), active peptic ulcer, neoplastic and inflammatory diseases and contraindications to anti-platelet or corticosteroid therapy. Among patients enrolled at our centre, 20 cases assigned to receive oral prednisone treatment after PCI with BMS implantation, and 20 assigned to BMS alone were randomly included in this sub-study.

2.2. Medication Before PCI all patients were pre-treated with a loading dose of either ticlopidine 500 mg/day for at least 48 h before or clopidogrel 300 mg, ideally 6 h before PCI, and conventional doses of aspirin (325–500 mg in patients with acute coronary syndromes; 100–160 mg in stable patients). After successful stent implantation all patients received standard medications, including aspirin 100–160 mg, ticlopidine 250 mg twice daily or clopidogrel 75 mg/day for 1 month. The double anti-platelet treatment was continued for 1 month only in patients with silent or stable ischemia, and for a minimum of 6 months, but ideally for one year, among those treated because of acute coronary syndromes. Statins were given to all patients at a minimum dose of 20 mg/day and up to 80 mg/day, if tolerated. Prednisone was administered following the treatment scheme used in the “Immunosuppressive therapy for the prevention of restenosis after coronary artery stent implantation” (IMPRESS) studies with slight modifications [8]: 1 mg/kg for the first 15 days; 0.5 mg/kg from days 16 to 30; 0.25 mg/kg from days 31 to 40. Oral thiazides and anti-acids/PPI were advised to reduce liquid retention and blood pressure and prevent calcium depletion [11]. Treatment with prednisone was started ideally the same day of PCI, early after the procedure and in all cases within 48 h.

243

2.3. Follow-up All patients were controlled clinically at 10 and 30 days after the procedure, underwent a stress test within 6 months of the PCI and had a subsequent out-patient clinic control. Between 9 and 12 months all patients underwent control angiography by radial approach and quantitative coronary analysis (QCA) was performed by an independent core laboratory. Angiograms were analyzed at baseline, after stenting and at follow-up. Proximal and distal vessel diameters, interpolated reference diameter (IRD), minimum lumen diameter (MLD), percentage of diameter stenosis (DS%) and late lumen loss (LLL) were obtained using the CAAS program version 5 (Pie Medical Imaging, the Netherlands). At 12 months after the index procedure all patients were controlled in the out-patient clinics by assigned independent clinicians to verify the incidence of any major adverse cardiac event comprised in the primary endpoint of the study [11]. 2.4. Biochemical determinations, monocyte isolation and evaluation of inflammatory mediators Hs-CRP was measured at baseline, after PCI and at 30 days using a standard high sensitivity method. Among patients receiving the corticosteroid treatment, blood samples were also obtained to measure the plasma concentrations of prednisone and cortisol at the same time points. Prednisone and Cortisol determinations were also performed in 5 healthy volunteers as a control. Samples were collected in the morning with patients fasting since the evening before and resting for about 30 min before venipuncture. Blood was immediately sent to the laboratory for centrifugation and storage until examination. Prednisone and cortisol were measured by an HPLC method according to Gai et al. [14] with slight modifications, separation being performed by Allure Biphenyl 250 mm × 4.6 mm i.d. column (Restek). The mobile phase was acetonitrile 40% (v/v) and 6-alpha metilprednisolone was used as internal standard. Human monocytes were isolated from heparinized venous blood (20 ml) by standard techniques of dextran sedimentation, Hystopaque (density = 1.077 g/cm3 ) gradient centrifugation (400 × g, 30 min, room temperature) and recovered by thin suction at the interface, as described [12]. Cells were re-suspended in RPMI 1640 medium, supplemented with 5% heat-inactivated foetal bovine serum (FBS), 2 mM glutamine, 50 ␮g/ml streptomycin, 5 U/ml penicillin and 2.5 ␮g/ml amphotericin B. Purified monocyte populations were obtained by adhesion (90 min, 37 ◦ C, 5% CO2 ); cell viability (trypan blue dye exclusion) was usually >98%. Expression of surface markers (CD14, CD68, MHCII) was analysed by flow cytometry, as described [12], yielding >85% pure monocyte populations. The release of interleukin-6 (IL-6) and tumour necrosis factor-alpha (TNF-␣), as well as NF-␬B activation, from circulating monocytes were measured at baseline, at day 10 (under the maximum dosage of oral prednisone at immunosuppressive dose: 1 mg/kg/die) and at day 30 when prednisone had been tapered to anti-inflammatory dose (0.5 mg/kg/die), before switching to the retrieval dose (0.25 mg/kg). To measure cytokine release, monocytes (1 × 106 cells) were challenged with phorbol 12-myristate 13-acetate (PMA; 10−7 M) or Salmonella-derived lipopolysaccharide (LPS; 10 ng/ml) for 24 h; supernatants were collected and stored at −80 ◦ C until use. IL-6 and TNF-␣ in the samples were evaluated by ELISA (Pelikine CompactTM human ELISA kit; sensitivity was 0.5 and 1.4 pg/ml, respectively), following the manufacturer’s instructions (CLB/Sanquin, Netherlands); results were expressed in pg/ml. NF-␬B activation was evaluated in 10 patients per group by determining the nuclear content of p50 subunit, using Trans AMTM NF-␬B p50 Chemi Transcription Factor Assay ELISA kits (Active Motif Europe, Belgium), according to the manufacturer’s instructions. Briefly, monocytes (5 × 106 cells) were stimulated by PMA

244

G. Pesarini et al. / Atherosclerosis 211 (2010) 242–248

10−6 M or LPS 10 ng/ml for 1 h [13] and nuclear extracts were prepared using “Nuclear Extract kit” (Active Motif Europe, Belgium), according to the manufacturer’s instructions. Evaluation of nuclear protein content was performed by BCATM protein assay kit (Pierce, USA), 1 ␮g of the sample being used to evaluate the presence of p50 subunit. This kit assay specifically detected bound NF-␬B p50 subunit in human extracts; activity of p50 was measured by a Rosys Anthos Lucy 1 luminometer and results are expressed as RLU (Relative Luminescence Unit).

mm, was used to assess in-segment neointimal proliferation from the index procedure to the angiographic follow-up. A p value of less than 5% was considered as statistically significant. SPSS version 8 and Microsoft Excel 2003 were used for computer assisted analysis and graphics.

2.5. Statistical analysis

Twenty patients in the BMS group and 20 in the BMS + prednisone group were included; clinical presentation and baseline biochemical characteristics did not differ in the two groups (Table 1). No patient died or had myocardial infarction during follow-up. No differences between the two groups were present at baseline for hs-CRP levels (Table 1). CRP values did not change significantly over time in BMS group, while there was a clear reduction in patients treated with prednisone after 1 month (p = 0.02). Leukocyte counts did not differ between the two groups; as expected, prednisone treatment was associated with a slightly higher leukocyte count (Table 1). Fifteen out of 20 prednisone-treated patients underwent serum quantification of prednisone and endogenous cortisol at 10 days (21.21 ± 10.78 ng/ml and 32.06 ± 36.97 nmol/l, respectively) and 30 days (18.9 ± 16.74 ng/ml and 25.74 ± 12.83 nmol/l); these results are comparable to previously published data with similar drug

Continuous data are expressed as mean ± 1 standard deviation. Population data were evaluated using variance analysis or a non-parametric test (Kruskal–Wallis) where appropriated. The differences between the biochemical parameters in the two groups were assayed by the T-test or a non-parametric test when the distribution was not normal. Linear regression was used to evaluate the relationship between plasmatic prednisone concentrations and the reduction of cytokine release and NF-␬B activation from baseline to day 10, when the steroid was administered at the highest dosage for immunosuppressive action. To evaluate a possible link between the reduction of IL-6 and TNF-␣ release and neointimal proliferation a linear regression was calculated. Cytokine release reduction was calculated as the difference between its baseline and 30-day values and expressed in pg/ml, while late lumen loss, expressed in

3. Results 3.1. Characteristics of the study population

Table 1 Baseline characteristics of the study population. BMS (n = 20)

BMS + PREDNISONE (n = 20)

p

Risk factors Age (years) Males (n, %) Family history of CAD (n, %) Hypertension (n, %) Smoke (n, %) Hypercholesterolemia (n, %) Previous CAD (n, %)

64.5 ± 6.8 15 (75%) 9 (45%) 14 (70%) 12 (60%) 12 (60%) 8 (40%)

63.6 ± 10.8 16 (80%) 10 (50%) 16 (80%) 16 (80%) 15 (75%) 10 (50%)

NS NS NS NS NS NS NS

Clinical presentation Silent ischemia (n, %) Stable angina (n, %) Unstable angina (n, %) NSTEMI (n, %) Previous MI (n, %) LVEF (%)

2 (10%) 4 (20%) 10 (50%) 4 (20%) 2 (10%) 50.3 ± 18.9

3 (15%) 5 (25%) 9 (45%) 3 (15%) 3 (15%) 47.6 ± 22.8

NS NS NS NS NS NS

Therapy before PCI Aspirin (n, %) Thienopyridines (n, %) ACE-inhibitors (n, %) Sartans (n, %) Calcium channel blockers (n, %) Statins (n, %) Beta-blockers (n, %)

20 (100%) 20 (100%) 18 (90%) 2 (10%) 5 (25%) 17 (85%) 12 (60%)

20 (100%) 20 (100%) 16 (80%) 4 (20%) 6 (30%) 18 (90%) 13 (65%)

NS NS NS NS NS NS NS

Biochemical Data Leukocytes basal (cells × 103 ) Leukocytes 30 dd (cellsx × 103 ) Haemoglobin (g/dl) HS-CRP basal (mg/l) HS-CRP 30 dd (mg/l) Fibrinogen (mg/dl) Gamma-GT (U/l) GPT (U/l) Creatinine (mg/dl) Total Cholesterol (mg/dl) LDL Cholesterol (mg/dl) HDL Cholesterol (mg/dl) Positive troponin-I (n, %)

7.6 ± 1.14 9.15 ± 4.03 13.7 ± 1.4 3.22 ± 4.35 3.88 ± 3.57 439.6 ± 80.6 45.9 ± 47.9 20.3 ± 12.4 0.90 ± 0.24 184.7 ± 43.41 126.32 ± 32.61 48.65 ± 13.89 5 (25%)

8.87 ± 3.24 10.51 ± 6.40 14.3 ± 1.1 3.44 ± 3.52 0.87 ± 19.95 433.5 ± 89.6 50.4 ± 40.3 22.5 ± 14.1 0.75 ± 0.41 180.11 ± 57.31 125.46 ± 35.14 45.86 ± 14.31 4 (20%)

NS NS NS NS 0.02 NS NS NS NS NS NS NS NS

Values are given as mean ± SD. Significance values were estimated by means of chi-square test for categorical variables and T-statistics or Kruskal–Wallis test for continuous data (see methods for more details). Abbreviations: Gamma-GT = gamma glutamil transpeptidase; GPT = glutamate pyruvic transaminase; Hs-CRP = high sensitivity C-reactive protein.

G. Pesarini et al. / Atherosclerosis 211 (2010) 242–248

dosages and laboratory techniques [15]. As expected, in patients with lower values of prednisone, endogenous cortisol remained higher, and vice versa. We also determined the plasmatic cortisol value in 5 healthy subjects (277.23 ± 101.66 nmol/l), which, as expected, resulted unsuppressed. Furthermore, in these volunteers plasmatic prednisone was not detectable. 3.2. Evaluation of inflammatory mediators Release of IL-6 and TNF-␣, as well as NF-␬B activation, from unstimulated or stimulated (LPS or PMA) monocytes at baseline, 10 and 30 days are shown for BMS group or BMS + prednisone group (Fig. 1). Prior to PCI intervention, cytokine release did not differ between the two groups, but declined significantly at 10 and 30 days after stent implantation, as a consequence of the attenuated inflammatory response in all patients (Fig. 1a–f). Interestingly, for both IL-6 (Fig. 1a–c) and TNF-␣ (Fig. 1d–f) release, the reduction was significantly higher in patients treated with prednisone. NF-␬B activation was assessed by evaluating the nuclear translocation of p50 subunit. No significant difference in baseline levels of p50 subunit was observed between the two groups (Fig. 1g–i). The amount of nuclear p50 subunit tended to decrease progressively from baseline to day 30 in both groups, but more markedly in prednisone-treated patients (p < 0.001; Fig. 1g–i). 3.3. Quantitative coronary analysis Quantitative coronary analysis was performed at baseline, after optimised stent implantation, and at follow-up. Mean time to angiographic follow-up was 10 ± 1.8 months. All quantitative parameters were treated according to the standard and widely

245

Table 2 Quantitative coronary analysis data. QCA

BMS (n = 20)

BMS + PREDNISONE (n = 20)

Basal values Lesion length (mm) IRD (mm) MLD (mm) DS% (%)

16.57 3.46 0.62 80.85

± ± ± ±

8.15 0.50 0.66 17.41

14.44 2.56 0.68 75.27

± ± ± ±

7.75 0.51 0.35 9.85

NS NS NS NS

Post-PCI values Stent length (mm) MLD (mm) DS% (%) Acute gain (mm)

18.42 2.97 11.42 2.37

± ± ± ±

6.18 0.44 9.76 0.59

17.40 2.37 10.83 1.69

± ± ± ±

6.78 0.38 7.76 0.42

NS NS NS NS

Follow-up values MLD (mm) DS% (%) Late loss (mm)

2.11 ± 0.72 40.98 ± 12.99 0.80 ± 0.53

2.09 ± 0.69 19.74 ± 23.29 0.44 ± 0.35

p

NS 0.03 0.02

Between 9 and 12 months all patients underwent QCA. Prednisone-treated patients showed less neointimal growth, as indicated by the lower Late Lumen Loss values. Value are expressed as mean (mm) ± SD and Kruskal–Wallis test was used to evaluate differences between groups. Abbreviations: IRD = interpolated reference diameter; MLD = minimum lumen diameter; DS% = diameter stenosis percentage; PCI = percutaneous coronary intervention.

accepted definitions for QCA and given in Table 2. No baseline or post-procedural differences were present between the prednisonetreated patients and control patients, but at 9 months elective angiography, the BMS+ prednisone group showed a significantly lower % diameter stenosis (40.98 ± 12.99% versus 19.74 ± 23.29%, p = 0.03) and LLL (0.80 ± 0.53 mm versus 0.44 ± 0.35 mm, p = 0.02).

Fig. 1. Evaluation of inflammatory mediators in monocytes from BMS patients and BMS + prednisone patients. Release of IL-6 (a–c) and TNF-␣ (d–f), and nuclear translocation of NF-␬B p50 subunit (g–i) were evaluated over time (at baseline, at 10 and 30 days after stent implantation) in non-stimulated (a, d and g); LPS-stimulated (b, e and h); or PMA-stimulated (c, f and i) monocytes in the two treatment groups. Cytokine release was evaluated in 20 patients per group. Nuclear translocation of p50 subunit was evaluated in 10 patients per group. Data are plotted as means + S.D. The p value refers to the difference at 30 days between the two groups.

246

G. Pesarini et al. / Atherosclerosis 211 (2010) 242–248

Fig. 2. Linear regression plots for inflammatory mediators. Plasmatic concentrations of prednisone at 10 days were correlated with the reduction of IL-6 (a), TNF-␣ (b) and p50 subunit (c) in LPS-stimulated monocytes from 10 BMS + prednisone patients. (d) Shows the relationship between the reduction of TNF-␣ release and of NF-␬B activation. Every circle represents a single evaluated patient. Determination coefficients are reported: see text for details.

3.4. Linear regression analyses As depicted in Fig. 2a–c, plasmatic concentrations of prednisone at 10 days correlated significantly with the reduction in IL-6 (R2 = 0.45, p = 0.04; TNF-␣ R2 = 0.69, p = 0.005 and NF-␬B activation/p50 subunit (R2 = 0.58, p = 0.01) in LPS-stimulated monocytes. Moreover, the reduction of TNF-␣ release was significantly related to the reduction of NF-␬B activation (R2 = 0.56, p = 0.01, Fig. 2d). Interestingly, the reduction of TNF-␣ release at 30 days from LPS-stimulated monocytes (that is greater in prednisone-treated patients) significantly correlated with a LLL at follow-up angiography in all patients (R2 = 0.41, p = 0.01; Fig. 3), while no correlation was observed between LLL and the reduction of IL-6 release (data not shown).

4. Discussion Oral administration of prednisone twice daily, according to a previously validated protocol aimed at reducing restenosis after PCI and BMS implantation, yields plasmatic concentrations similar to those previously reported in pharmacokinetic studies [15]. Our experiment demonstrates that this prednisone dosage is effective in reducing serum concentrations of hs-CRP, NF-␬B activation and cytokine release from activated circulating monocytes.

Fig. 3. Correlation between TNF-␣ reduction and late lumen loss at angiographic follow-up. The decrease in TNF-␣ release from baseline to day 30 was expressed in pg/ml; late lumen loss was expressed in mm. Data refer to LPS-stimulated samples from all patients.

G. Pesarini et al. / Atherosclerosis 211 (2010) 242–248

CRP, IL-6 and TNF-␣ are known markers and mediators of the inflammatory response to vessel injury induced by PCI and stent implantation. Indeed, previous investigations have highlighted the relation between the over-expression of these two cytokines and of CRP with restenosis after angioplasty [16–19]. Several pathways of immuno-mediated inflammation participate in atherosclerosis and post-PCI restenosis that include humoral (i.e. antibodies, immunocomplexes, complement, cytokines, prostaglandins) and cellular systems, e.g., monocyte/macrophages, lymphocytes, neutrophils, platelets, fibroblasts, endothelial cells and smooth muscle cells (SMC). Monocyte/macrophages play a prominent role in atheromatous plaque development, and are also recruited after vascular injury. Monocytes migrate into the damaged area either as a direct response or through the release of platelet-derived factors. Activated monocytes release high amounts of pro-inflammatory cytokines, that cause vasoconstriction and non-specific recruitment, proliferation and activation of other cells in the vascular wall [1–3]. Inflammation is therefore a key target for a non-specific anti-restenosis treatment, and glucocorticoids are the most effective and well known anti-inflammatory agents. In fact, steroids exert beneficial effects on platelet function, inflammatory cells migration and activation, SMC proliferation and collagen synthesis, thus interfering with several steps of the cascade leading to neointima formation and subsequent LLL [8]. Our investigation confirms that, in BMS-treated patients, oral administration of prednisone twice daily inhibits IL-6 and TNF␣ release from activated monocytes more rapidly after stent implantation. In particular, TNF-␣ has been shown to decrease dramatically in response to corticosteroids, especially in PMAstimulated monocytes. The different degree of inhibition between LPS- and PMA-stimulated monocytes could reflect the heterogeneous intracellular signalling that, in the case of PMA, involves several pathways, while for LPS is mainly focused on NF-␬B activation [13]. The NF-␬B pathway plays a fundamental role in atherosclerosis progression, its activation leading to enhanced transcription and release of pro-inflammatory cytokines such as IL-6 and TNF␣ [3,13]. In this experiment NF-␬B p50 subunit has been measured because it is more abundant than the p65 subunit in human monocyte/macrophages, and because its known involvement in atherosclerosis [21]. The more marked decrease in p50 subunit observed at 30 days in patients treated with prednisone (as compared to BMS patients) reflects the multiple anti-inflammatory effects of the steroids, largely attributed to NF-␬B inhibition [9,10,13]. In this work a significant positive correlation between NF-␬B p50 activation and TNF-␣ release has been evidenced. Interestingly, the reduction of this cytokine also correlated with a significant decrease of the LLL at follow-up. The correlation between inhibition of TNF-␣ release and neointimal proliferation within the stent (as quantitatively assessed in terms of LLL), supports the rationale of focusing on inflammatory pathways as an effective target to prevent restenosis after PCI. Of note, reduction of TNF-␣ release (but not IL-6) in both BMS and BMS+ prednisone patients correlated with LLL reduction. In fact, TNF-␣ is the most potent proliferative mediator released by activated monocytes, and therefore, the most likely to respond to the prednisone treatment. The link between the mechanisms of action of the drug, and the phenotypic change that may translate into clinical benefit is apparent even in this relatively small study. Indeed, the oral prednisone treatment yielded significantly lower indexes of neointimal proliferation and restenosis at follow-up angiography, and the magnitude of the benefit of the treatment compared to controls is very similar to that previously observed in prospective clinical studies [6,7].

247

Considering the ubiquitous nature of atherosclerosis, a major theoretical advantage of a systemic treatment such as the one tested in our study, is its potential for being effective independently of the technical limitations of catheter-based interventions and the locally delivered drug treatment offered by current drugeluting stent technology. Although the use of prednisone in patients with coronary artery disease remains an “off-label” application, the results of our investigation add further evidence in favor of its efficacy. The sustained benefits of this treatment have been recently demonstrated even at long term [20]. Whether systemic therapy with prednisone serves as an adjunctive strategy to refrain atherosclerosis progression in phases of activity of the disease, such as acute coronary syndromes or after percutaneous revascularization procedures, remains a very important matter of further investigation, which deserves to be addressed in large population studies. The main practical limitation of this strategy is the lack of data in diabetic patients, due to the perceived unfavourable effects that high-dose steroids may induce in the diabetic metabolic milieu. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.atherosclerosis.2010.02.004. References [1] Welt FG, Rogers C. Inflammation and restenosis in the stent era. Arterioscler Thromb Vasc Biol 2002;22:1769–76. [2] Fukuda D, Shimada K, Tanaka A, et al. Circulating monocytes and in-stent neointima after coronary stent implantation. J Am Coll Cardiol 2004;43:18–23. [3] Liuzzo G, Santamaria M, Biasucci LM, et al. Persistent activation of nuclear factor kappa-B signaling pathway in patients with unstable angina and elevated levels of C-reactive protein. Evidence for a direct proinflammatory effect of azide and lipopolysaccharide-free C-reactive protein on human monocytes via nuclear factor kappa-B activation. J Am Coll Cardiol 2007;49:185–94. [4] Libby P. Act local, act global, inflammation and the multiplicity of vulnerable coronary plaques. J Am Coll Cardiol 2005;45:1600–2. [5] Ribichini F, Joner M, Ferrero V, et al. Effects of oral prednisone after stenting in a rabbit model of established atherosclerosis. J Am Coll Cardiol 2007;50:176–85. [6] Versaci F, Gaspardone A, Tomai F, et al. Immunosuppressive therapy for the prevention of restenosis after coronary artery stent implantation (IMPRESS Study). J Am Coll Cardiol 2002;40:1935–42. [7] Ribichini F, Tomai F, Ferrero V, et al. Immunosuppressive oral prednisone after percutaneous interventions in patients with multi-vessel coronary artery disease. The IMPRESS-2/MVD study. Eurointervention 2005;2:173–80. [8] Ferrero V, Ribichini F, Pesarini G, et al. Therapeutic potential of glucocorticoids in the prevention of restenosis after coronary angioplasty. Drugs 2007;67:1243–55. [9] Almawi WY, Melemedjian OK. Molecular mechanisms of glucocorticoids antiproliferative effects: antagonism of transcription factor activity by glucocorticoids receptor. J Leukoc Biol 2002;71:9–15. [10] Auphan N, DiDonato JA, Rosette C, Helmberg A, Karin M. Immunosuppression by glucocorticoids: inhibition of NF-Kappa B activity through induction of I Kappa B synthesis. Science 1995;270:286–90. [11] Ribichini F, Tomai F, De Luca G, et al. A multicenter, randomized study to test immunosuppressive therapy with oral prednisone for the prevention of restenosis after percutaneous coronary interventions: cortisone plus BMS or DES versus BMS alone to eliminate restenosis (CEREA-DES)—study design and rationale. J Cardiovasc Med 2009;10:192–9. [12] Amoruso A, Bardelli C, Gunella G, Ribichini F, Brunelleschi S. A novel activity for substance P: stimulation of peroxisome proliferator-activated receptor␥ in human monocytes and macrophages. Br J Pharmacol 2008;154:144– 52. [13] Lavagno L, Gunella G, Bardelli C, et al. Anti-inflammatory drugs and tumor necrosis factor-␣ production from monocytes: role of transcription factor NF-kB and implication for rheumatoid arthritis therapy. Eur J Pharmacol 2004;501:199–208. [14] Gai MN, Pinilla E, Paulos C, Chavez J, Puelles V, Arancibia A. Determination of prednisolone and prednisone in plasma, whole blood, urine and bound to plasma proteins by high-preformance liquid chromatography. J Chromatogr Sci 2005;43:201–6. [15] Ferry JJ, Horvath AM, Bekersky I, Heath EC, Ryan CF, Colburn WA. Relative and absolute bioavailability of prednisone and prednisolone after separate oral and intravenous doses. J Clin Pharmacol 1988;28:81–7. [16] Liuzzo G, Buffon A, Biasucci LM, et al. Enhanced inflammatory response to coronary angioplasty in patients with severe unstable angina. Circulation 1998;98:2370–6.

248

G. Pesarini et al. / Atherosclerosis 211 (2010) 242–248

[17] Hojo Y, Ikeda U, Katsuki T, et al. Interleukin 6 expression in coronary circulation after coronary angioplasty as a risk factor for restenosis. Heart 2000;84: 83–7. [18] Szkodzinski J, Blazelonis A, Wilczek K, et al. The role of interleukin-6 and transforming growth factor-beta1 in predicting restenosis within stented infarct-related artery. Int J Immunopathol Pharmacol 2009;22:493–500. [19] Murayama H, Takahashi M, Takamoto, et al. Deficiency of tumour necrosis factor-alpha and interferon-gamma in bone marrow cells synergistically

inhibits neointimal formation following vascular injury. Cardiovasc Res 2008;80:175–80. [20] Ferrero V, Tomai F, Versaci F, et al. Long-term results of immunosuppressive oral prednisone after coronary angioplasty in non-diabetic patients with elevated C-reactive protein levels. Eurointervention 2009;5:250–4. [21] Brunelleschi S, Bardelli C, Amoruso A, et al. Minor polar compounds in extravirgin olive oil extract (MPC-OOE) inhibits NF-␬B translocation in human monocyte/macrophages. Pharmacol Res 2007;56:542–9.