Critical limb ischemia: thrombogenic evaluation of two autologous cell therapy products... by Vaclav Prochazka

ORIGINAL RESEARCH Critical limb ischemia: thrombogenic evaluation of two autologous cell therapy products and biologic profile in treated patients Claire Tournois,1,2 Bernard Pignon,3* Marie-Antoinette Sevestre,4* Zoubir Djerada,2 Jean-Claude Capiod,5 Ga€e l Poitevin,2 Anne-Marie Delloup,1 and Philippe Nguyen1,2

BACKGROUND: Cell therapy has been proposed as a salvage limb procedure in critical limb ischemia (CLI). Autologous cell therapy products (CTP) are obtained from patients with advanced peripheral arterial disease to be injected at the site of ischemia. Thrombogenicity of CTPs has not yet been assessed. The objectives were: 1) to assess thrombotic risk in candidates for cell therapy, 2) to evaluate two different CTPs in terms of thrombogenic potential, and 3) to evaluate clinical thrombotic events. STUDY DESIGN AND METHODS: In this ancillary study of a Phase I and II clinical trial, bone marrow (BM)CTPs (n 5 20) and CTPs obtained by cytapheresis (peripheral blood [PB]-CTPs; n 5 20) were compared. Inflammatory and coagulation markers were measured at baseline and 24 hours after CTP implantation. CTP cell content and tissue factor (TF) expression (mRNA and protein) were analyzed. Thrombin generation assessed CTP-related thrombogenicity. RESULTS: All patients presented cardiovascular risk factors. At baseline, the patients’ biologic profile was characterized by high levels of fibrinogen, C-reactive protein (CRP), D-dimer, interleukin (IL)-6, and plasmatic TF, whereas IL-10 was low. Although different in terms of cell composition, both BM- and PB-CTPs support low thrombin generation. Twenty-four hours after implantation, biologic markers remained stable in the PBCTP group, except for IL-6. In the BM-CTP group, a significant increase of IL-6 but also of CRP and D-dimer was observed. Clinically, one single patient developed deep vein thrombosis 24 hours after the implantation of autologous PB-CTP. CONCLUSION: CTPs supported low thrombin generation and were well tolerated after calf implantation.

eripheral arterial disease (PAD) is a common circulatory disease. The annual incidence is approximately 500 to 1000 new cases per million in industrialized countries.1,2 The prevalence is

ABBREVIATIONS: AbTF 5 blocking anti-TF; BM 5 bone marrow; cb-CD341 5 cord blood CD341 cells; cb-ECFCs 5 cord blood endothelial colony-forming cells; CK 5 creatine kinase; CLI 5 critical limb ischemia; CRP 5 C-reactive protein; CTP(s) 5 cell therapy product(s); DVT 5 deep vein thrombosis; Fg 5 fibrinogen; HUVEC(s) 5 human umbilical vein endothelial cell(s); IL-6R 5 interleukin-6 receptor; IL10R 5 interleukin-10 receptor; IM 5 intramuscular; LPS 5 lipopolysaccharide; PAD 5 peripheral arterial disease; PB 5 peripheral blood; PFP 5 platelet-free plasma; TcPO2 5 transcutaneous partial pressure of oxygen; TF 5 tissue factor (Factor III); plTF 5 tissue factor plasma level; rHuTF

5 recombinant human tissue factor; TGA 5 thrombin

generation assay. matologie, CHU Robert Debre ; 2EAFrom the 1Laboratoire d’He , Universite de Reims Champagne3801, SFR CAP-Sante de The rapie Cellulaire, CHU, Reims, Ardenne; and 3Unite edecine Vasculaire, CHU; and France; and the 4Service De M 5 matologie, CHU, Amiens, France. Laboratoire d’He This study was funded by a clinical research hospital program grant (French Ministry of Health, PHRC 2003). *BP and MAS contributed equally to this study. Address reprint requests to: Philippe Nguyen, MD, PhD, , Laboratoire d’H EA-3801, SFR CAP-Sante ematologie, Centre ^ pital Robert-Debr Hospitalier Universitaire Reims, Ho e, 51092 Reims Cedex, France; e-mail: pnguyen@chu-reims.fr. Received for publication February 19, 2015; revision received April 27, 2015; and accepted May 4, 2015. doi:10.1111/trf.13203 C 2015 AABB V

TRANSFUSION 2015;00;00–00 Volume 00, Month 2015 TRANSFUSION 1

TOURNOIS ET AL.

expected to increase with the increasing rate of risk factors such as diabetes and ageing population. Severe forms of PAD lead to critical limb ischemia (CLI). In 25% of patients with CLI, limb amputation cannot be avoided and less than half of the patients are alive without any major amputation after 6 months.1 Since 2002, several studies have shown that autologous transplantation of either bone marrow (BM) or peripheral blood (PB) cell therapy products (CTPs) could be effective in inducing angiogenesis.3,4 In a review of 45 clinical trials, including a total of 1272 patients, the overall adverse event rate was low (4.4 and 3.7%, respectively, in BM- and PBCTP groups).5 The most common adverse events were pain (20.8%), bleeding (17%), and anemia (15.1%). The thrombogenicity of such procedures is not considered in spite of the fact that: 1) CTPs contain high amounts of potentially thrombogenic cells; 2) CTPs are obtained from patients potentially at risk for thrombosis;1,6-8 and 3) autologous CTPs are currently administered by a series of intramuscular (IM) injections, a potentially thrombogenic route of administration. Furthermore, BM aspiration requires general anesthesia and the discontinuation of any ongoing antithrombotic treatment. In this specific surgical condition, the need for an antithrombotic prophylaxis during multiple IM injections of CTPs into the gastrocnemius of the ischemic leg has not been documented. We conducted a prospective Phase I and II clinical trial in three French academic hospitals. The goal of this trial was to evaluate whether IM implantation of autologous BM- or PB-CTP is safe and associated with clinical improvement in CLI patients. The originality of the trial was that PB-CTPs were collected without any granulocyte–colony-stimulating factor mobilization. Indeed, the use of hematopoietic growth factor may induce side effects such as bone pain, arthralgia, and thrombosis.5,9,10 This trial gave us the opportunity to perform a series of ancillary studies: in the first study,11 we analyzed the histologic modifications induced by the injection of CTPs, and in the second study,12 we characterized the cell composition of CTPs. This study aimed at evaluating and characterizing the thrombogenicity of two CTPs. The objective was: 1) to evaluate the baseline proinflammatory and hypercoagulable status of CLI patients; 2) to evaluate the prothrombotic potential risk of nonmobilized PB-CTPs in comparison with BM-CTPs; and 3) to investigate the blood variations of inflammatory and coagulation markers, 24 hours after IM implantation in comparison with clinical outcome. We focused on tissue factor (TF) expression as 1) it is the primary initiator of the extrinsic coagulation pathway, 2) it plays a key role in the cross-talk between inflammation and coagulation,13 and 3) it is involved in angiogenesis.14 We investigated the balance between interleukin (IL)-6 and IL-10, which could have opposite effects in terms of TF expression and thrombogenicity.15 2

TRANSFUSION Volume 00, Month 2015

MATERIALS AND METHODS Patients and BM- and PB-cell products preparation The study protocol was approved by the ethics committee of Champagne Ardennes. The design of the clinical trial can be found at www.ClinicalTrials.gov (Trial Number NCT00533104). Briefly, 40 patients presenting with unilateral CLI but not suitable candidates for nonsurgical or surgical revascularization were included. Transcutaneous partial pressure of oxygen (TcPO2) was measured using a transcutaneous monitoring system (TINA TCM4, Radiometer, Copenhagen, Denmark) under standardized conditions. CTPs were obtained by BM aspiration (BM-CTPs n 5 20) or cytapheresis (PB-CTPs n 5 20). Antiplatelet (PLT) drugs (acetylsalicylic acid or clopidogrel) were stopped 7 days before cell therapy. Vitamin K antagonists or heparin treatments were interrupted, respectively, 3 days and 12 hours before cell harvesting. For the preparation of BM-cell products, 500 mL of BM was collected under general anesthesia through multiple punctures of the posterior iliac crest. BM- and PM-CTP preparation has been previously described.12 BM- and PB-CTPs were suspended in autologous patients’ plasma and were implanted within 3 hours after preparation, by 30 multiple IM injections into the gastrocnemius of the ischemic leg (1 mL per injection). Anticoagulants or/and anti-PLT therapy was resumed 6 hours after IM injection. Follow-up visits were performed on Postprocedure Days 1, 2, 3, 14, and 28 and at Months 3 and 6.

Patient’s blood variables Quantitative determination of IL-6 and TF plasma level (plTF) in EDTA plasma was performed by specific human antibodies using an enzyme-linked immunosorbent assay (ELISA) kit (Quantikine, R&D Systems, Inc., Minneapolis, MN), according to the manufacturer’s instructions. IL-10 EDTA plasma levels were measured by Luminex multianalyte profiling (xMAP) technology using a MAP kit human cytokine panel I (Milliplex, Millipore Corporation, Billerica, MA). C-reactive protein (CRP) serum levels, fibrinogen (Fg), and D-dimer plasma levels were measured using standardized methods (respectively, COBAS, Roche Diagnostics, Meylan, France; STA-R, Diagnostica Stago SAS, Asnie`res sur Seine, France; Vidas D-dimer Exclusion II, bioM erieux, Marcy l’Etoile, France). Creatine kinase (CK) activity and myoglobin plasma levels were measured using standardized methods (COBAS, Roche Diagnostics Meylan).

CTP variables Cell counts BM- or PB-cell product characterization methods have been previously described.12

CELL THERAPY IN CRITICAL LIMB ISCHEMIA

TF and cytokine mRNA content by real-time quantitative reverse transcriptase–polymerase chain reaction The mRNA analysis was realized from 11 BM and 11 PB randomly selected cell products. The following cells were used as controls: endotoxin-stimulated and nonstimulated elutriated human monocytes (n 5 5) according to Poitevin and colleagues,16 cord blood endothelial colony-forming cells (cb-ECFCs, n 5 9), cord blood CD341 cells (cbCD341, n 5 7), and human umbilical vein endothelial cells (HUVECs, n 5 7) as previously described by Cuccuini and coworkers.17 The following primers and Taqman probes (Applied Biosystems, Courtaboeuf, France) were used to analyze and normalize full-length TF, IL-6, IL-6 receptor (IL-6R), IL-10, and IL-10 receptor (IL-10R) alpha mRNA: TF (Hs00175225_m1), IL-6 (Hs00174131_m1), IL-6R (Hs00169842_m1), IL-10 (Hs99999035_m1), IL-10R (Hs00155485_m1), b2-microglobulin (Hs99999907_m1), and GAPDH (Hs99999905_m1). b2-Microglobulin and GAPDH were used as housekeeping genes. Conventional PCR was performed under standard conditions and analyzed on a sequence detector (ABI Prism 7700, Applied Biosystems). The mRNA levels of full-length TF, IL-6, IL-6R, IL10, and IL-10R were normalized to housekeeping genes using the 2–DCt method (=2–[Ct(target) – Ct(housekeeping gene)]).

Statistical analysis Statistical analyses were performed using computer software (SAS, Version 8.0, SAS Institute, Cary, NC). Differences between groups for continuous variables were evaluated using nonparametric Mann-Whitney test. Differences between groups for continuous variables were evaluated using nonparametric test Kruskal-Wallis test. If the F ratio was significant, Dunn’s multiple comparison (post hoc) test was applied to assess significance. Paired comparisons for continuous variables used nonparametric Wilcoxon matched pairs test. Differences for categorical variables were assessed using chi-square or two-tailed Fisher’s exact tests when the expected number in any cell was less than five. Quantitative variables were expressed as median (range) or box-and-whiskers plots (Medcalc Version 7.3 software, https://www.medcalc.org/). The correlation between biologic and/or clinical variables used the nonparametric Spearman test. The significant differences level was defined as *p < 0.05, †p < 0.01, and ‡p < 0.001.

RESULTS Patient characteristics

Detection of TF by antigen ELISA and Western blot BM- and PB-CTPs were lysed according to the cell lysis procedure (cell lysis buffer 1) of the Quantikine ELISA kit (R&D Systems, Inc.). The expression of TF protein was performed as previously described by Cuccuini and colleagues.17

Fluorogenic measurement of thrombin generation assay Immediately after cell separation, PB- or BM-CTPs were stored frozen in liquid nitrogen until analysis. After being thawed, cells were washed twice and resuspended in phosphate-buffered saline 31. Thrombin generation assay (TGA) was performed as previously described17 in PLTfree plasma (PFP), in the presence of synthetic anionic phospholipids and aprotinin. The assay was performed using randomly selected 1 3 104 PB- (n 5 6) or BM-CTPs (n 5 6) with a viability of more than 90%, as determined by trypan blue exclusion or 1 and 5 pmol/L recombinant human TF (rHuTF; Dade Innovin, Siemens, Saint Denis, France). At first, TGA was measured in normal PFP in the presence of cells alone or cells with the irrelevant antibody or cells with a specific antibody blocking the coagulant activity of TF (AbTF; No. 4509, American Diagnostica, Neuville-sur-Oise, France). Secondly, TGA was measured in a Factor (F)VII-deficient plasma (Diagnostica Stago SAS). Lag time (min), time to peak (min), and thrombin peak (nmol/L) mean values were calculated from triplicates using computer software (Thrombinoscope, Synapse BV, Maastricht, the Netherlands).

Forty patients were included in the clinical trial: 20 treated with BM-CTPs and 20 treated with PB-CTPs. They presented severe symptoms and had significant comorbidities (Table 1). TcPO2 levels were 3.9 times lower in smokers compared with the nonsmokers (9.5 [1-74] mmHg in the smokers vs. 37.0 [1-57] mmHg in the nonsmokers; p 5 0.045). Biologic variables on Day 0 are reported in Table 1. CLI patients presented a major inflammatory syndrome with high levels of IL-6, CRP, and Fg (72% of the patients, 64% and 50% higher than normal values, respectively). These inflammation markers are statistically correlated (CRP vs. Fg, r 5 0.72, p 5 0.0002; CRP vs. IL-6, r 5 0.69, p 5 0.0001; and IL-6 vs. Fg, r 5 0.48, p 5 0.0065). At baseline, IL-10 was weakly detectable in plasma (Table 1). Patients with resting pain presented a higher inflammatory syndrome with high CRP (4.6-fold higher, p 5 0.007) and Fg levels (1.37-fold higher, p 5 0.018) in comparison with patients who did not complain of pain. Fg plasma levels were inversely correlated with TcPO2 (r 5 20.37, p 5 0.034). At baseline, D-dimer plasma levels were in the upper range (81% of patients had D-dimer concentrations above the 0.5 mg/mL cutoff, which corresponds to the optimal negative predictive value to rule out venous thrombosis). D-dimer concentrations were not correlated with any other studied clinical and biologic variables. plTF was in the upper range (74% of the patients higher than normal values) and was positively correlated with age (r 5 0.48, p 5 0.0034). At baseline, there was no difference Volume 00, Month 2015 TRANSFUSION 3

TOURNOIS ET AL.

TABLE 1. Patient characteristics* Group Characteristic Age (years) Sex (male/female) Body mass index (kg/m2) TcPO2 (mmHg) Rest pain Active smokers Arterial hypertension Hypercholesterolemia Diabetes mellitus Hemoglobin (g/L) WBCs (3109/L) PLTs (3109/L) IL-6 (pg/mL), <3.2† CRP (mg/L), <5† Fg (g/L), 1.8-3.5† D-dimer (mg/mL), <0.5† plTF (pg/mL), 39 6 7† IL-10 (pg/mL), 2.2 [0.3-102]†

BM (n 5 20)

PB (n 5 20)

71 (45-84) 18/2 25 (20-32) 16 (1-43) 17 (85) 16 (80) 14 (70) 14 (70) 6 (30) 127 (91-149) 7.8 (5.2-13.9) 258 (157-391) 5 (1-45) 6.6 (0.7-60) 3.9 (2.7-4.9) 1.1 (0.4-4.8) 54 (31-130) 1.3 (0.3-29)

65 (37-84) 15/5 27 (19-42) 18 (1-74) 18 (90) 12 (60) 13 (65) 8 (40) 4 (20) 124 (90-151) 7.8 (4.5-16.4) 276 (134-497) 8 (1-139) 10.6 (1.0-120) 3.8 (2.3-6.0) 1.2 (0.2-4.2) 56 (32-104) 0.3 (0.3-89)

* Data are presented as counts (%). Quantitative variables are expressed as median (range). Proinflammatory (IL-6, CRP, and Fg), antiinflammatory (IL-10), and prothrombotic (plTF, D-dimer) biologic markers measured in 40 CLI patients, on Day 0 (baseline) before cell therapy. † Normal values as indicated by the manufacturer.

TABLE 2. Cell counts of BM- and PB-CTPs* Biologic variables

BM-CTPs (n 5 20)

PB-CTPs (n 5 20)

p value†

Hct (%) PLTs (3109/L) Total nucleated cells (3109/L) Total MNCs (3109/L) Lymphocytes (3109/L) Monocytes (3109/L) Erythroblasts (3109/L) Other cells‡ (3109/L) Mature granulocytes (3109/L) CD341 stem cells (3109/L)

4.8 (2.4-12.6) 712 (461-1214) 38 (18-71) 34 (11-52) 18.0 (6.2-44.7) 5.7 (2.1-12.4) 3.4 (0.4-13.6) 2.2 (0-5.6) 6.4 (1.9-33.9) 0.98 (0.20-3.31)

8.7 (2.8-21) 1541 (814-2652) 110 (48-187) 104 (45-186) 65.6 (23.1-159.3) 37.0 (11.3-80.9) None None 8.0 (1.9-31.6) 0.11 (0.04-0.25)

0.0005 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001

NS <0.0001

* BM- and PB-CTPs were implanted by 30 multiple IM injections into the gastrocnemius of the ischemic leg (1 mL per injection). Quantitative variables are expressed as median (range). † Nonparametric Mann-Whitney test. ‡ Other cells: blasts, immature granulocytes, and plasma cells.

between the two groups, whatever the studied clinical or biologic variables. We investigated the potential risk in terms of thrombogenicity of BM- and PB-cell autologous products. To approach this issue, we 1) determined the cell composition of CTPs; 2) evaluated the capacity of these CTPs to express full-length TF, IL-6, and IL-10 and their receptors’ mRNA and TF protein; and 3) studied the ability of CTPs to support TGA.

Cell counts of BM- and PB-cell products The cell composition of BM- and PB-CTPs is presented in Table 2. This composition was quantitatively and qualitatively heterogeneous whatever the type of CTP. As 4

TRANSFUSION Volume 00, Month 2015

expected, the proportions of both mature and immature cell subsets were different: mature cells (including lymphocytes, monocytes, and PLTs) were in higher concentration in PB-CTPs, whereas CD341 stem cells were significantly higher in BM-CTPs.

Expression of TF, IL-6, and IL-10 and their receptors’ mRNA in PB- and BM-cell products TF mRNA was detected in both CTPs (Fig. 1A). For comparison, we measured TF mRNAs in mature or immature hematopoietic cells: unstimulated and lipopolysaccharide (LPS)-stimulated monocytes; cb-CD341 cells; and in endothelial cells, cb-ECFCs and HUVECs. TF mRNA was barely detectable in PB-CTPs, unstimulated monocytes,

CELL THERAPY IN CRITICAL LIMB ISCHEMIA

and cb-ECFCs. TF mRNA was present in LPS-stimulated monocytes, cb-CD341 cells, and HUVECs. TF mRNA was 7.2-fold lower in PB-CTPs than in BM-CTPs. As cytokines are key regulators of TF expression, we searched for an expression of IL-6 and IL-10 mRNA and their receptors in CTPs (Figs. 2 and 3). The choice of these receptors was justified by the fact that we observed an elevation of IL-6 in patients (see above). On the other hand, IL-10 is known to down regulate TF. IL-6 mRNA was 6.8fold lower in PB-CTPs than in BM-CTPs (Fig. 2A). No difference was observed between PB- and BM-CTPs in IL-10 mRNA expression (Fig. 3A). IL-6R (Fig. 2B) and IL-10R (Fig. 3B) mRNA were strongly expressed in both CTPs.

Detection of TF by Western blot and by ELISA method Western blot analysis did not reveal any expression of the full-length TF protein in either CTP (data not shown). TF protein was barely detectable in both BM- and PB-CTPs lysates. Using the ELISA method, TF protein was inferior to the sensitivity threshold (<0.69 pg/mL).

Fluorogenic measurement of TGA To investigate whether TF expressed by both CTPs was functional, we determined the time course of TGA supported by viable and frozen-thawed CTPs. For this, we used a calibrated fluorogenic assay. Figure 1B shows representative TGA curves supported by a BM-CTP or PBCTP. The kinetics of TGA were very late and the amount of formed thrombin was low. No TGA occurred in FVIIdeficient plasma (data not shown). The different variables of TGA, that is, lag time, time to peak, and thrombin peak, are presented in Table 3. No difference was observed between the two CTPs. Of note, we did not observe any inhibitory effect of the AbTF, whatever the TGA variable studied. Taken together, these results show that TF expression is present at a low level in CTPs and that these products induce low TGA.

Fig. 1. Expression of full-length TF mRNA in PB-CTPs ( , n 5 11) and BM-CTPs (ä&#x160;?, n 5 11) (A). Results are expressed in comparison to LPS-stimulated and nonstimulated elutriated human monocytes (n 5 5), cb-CD341 cells (n 5 7), cb-ECFCs (n 5 9), and HUVECs (n 5 7). Fluorogenic measurement of TGA (B). The assay was performed using a representative ( ) PBor ( ) BM-CTP or rHuTF at two concentrations (5 [~] and 1 [ä&#x2030;Ť] pmol/L). TF plasma levels measured at baseline (H0) and 24 hours (H24) after PB-CTP ( , n 5 20) or BM-CTP (ä&#x160;?, n 5 20) implantation (C). Biologic variables are expressed as box-andwhiskers plots. aNormal values as indicated by the manufacturer. 1Aberrant distribution values as indicated by box-andwhiskers plots (Medcalc version 7.3 software).

Variations of blood variables after CTP implantation and safety outcome To evaluate the effect of IM implantation of BM- or PBCTPs, a follow-up was performed before cell harvesting (H0) and 24 hours after implantation (H24). We observed a significant increase of IL-6 (2.9-fold higher; Fig. 2C), CRP (5.5-fold higher), CK activity (1.7-fold higher), and myoglobin (1.5-fold higher; Table S1, available as supporting information in the online version of this paper) in patients who were implanted with BM-CTPs. Myoglobin and CK variations were correlated (p < 0.0001, r 5 0.84) and both were correlated with CRP variations (p 5 0.02, r 5 0.47 between CRP and myoglobin variations; p 5 0.01, r 5 0.44 between CRP and CK variations) in patients of the BM group. The increase of IL-6 was Volume 00, Month 2015 TRANSFUSION 5

TOURNOIS ET AL.

Fig. 2. Expression of IL-6 (A) and IL-6R (B) mRNA in PB-CTPs ( , n 5 11) and BM-CTPs (䊏, n 5 11). Results are expressed in comparison to nonstimulated elutriated human monocytes (n 5 5), cb-CD341 cells (n 5 7), cb-ECFCs (n 5 9), and HUVECs (n 5 7). IL-6 plasma levels measured at baseline (H0) and 24 hours (H24) after PB-CTP ( , n 5 20) or BM-CTP (䊏, n 5 20) implantation (C). aNormal values as indicated by the manufacturer.

moderate (1.4-fold higher) at H24 when PB-CTPs (Fig. 2C) were implanted (but recovered baseline level on Day 3, data not shown). In contrast, the increase of CRP was not significant at H24 in PB-CTP–treated patients (Table S1). Fg was measured without any statistical difference between H0 and H24, whatever the implanted cell product (Table S1). In contrast, we observed a significant increase of anti-inflammatory IL-10 (3.2-fold higher) in patients who were implanted with BM-CTPs (Fig. 3C; but recovered baseline level on Day 3; data not shown). In terms of biomarkers of thrombogenicity, plTF was not different between H0 and H24, whatever the implanted cell product (Fig. 1C). In contrast, D-dimer significantly increased (2.9-fold higher) at H24 only in patients who 6

TRANSFUSION Volume 00, Month 2015

received BM-CTPs (Table S1). For each biomarker of inflammation and thrombogenicity, we searched if the variation observed between H0 and H24 could be explained by the number of implanted cells (total number of cells and number of each cell type as described in Table 2). In this exhaustive analysis, no correlation was found, whatever the cell composition (quantitatively and qualitatively). For safety reasons, hemograms were repeated at H0, H24 (Table S1), H48, and H72 and 1 week (data not shown) after the implantation. During this follow-up, no significant modification of white blood cells (WBCs) and PLT counts was observed. In contrast, a significant decrease of hematocrit (Hct) was observed at H24, in patients of the BM group.

CELL THERAPY IN CRITICAL LIMB ISCHEMIA

Fig. 3. Expression of IL-10 (A) and IL-10R (B) mRNA in PB-CTPs ( , n 5 11) and BM-CTPs (䊏, n 5 11). Results are expressed in comparison to nonstimulated elutriated human monocytes (n 5 5), cb-CD341 cells (n 5 7), cb-ECFCs (n 5 9), and HUVECs (n 5 7). IL-10 plasma levels measured at baseline (H0) and 24 hours (H24) after PB-CTP ( , n 5 20) or BM-CTP (䊏, n 5 20) implantation (C). aNormal values as indicated by the manufacturer.

TABLE 3. TGA* rHuTF (pmol/L)

BM-CTPs

PB-CTPs

TGA variables

Cells

Cells 1 AbTF

Cells

Cells 1 AbTF

p value†

p value‡

Lag time (min) Time to peak (min) Thrombin peak (nmol/L)

2.4 (2.0-3.2) 4.8 (4.3-6.0) 222 (201-242)

5.5 (4.3-7.6) 10.4 (9.1-13.2) 93 (84-116)

28 (18-34) 32 (26-42) 52 (44-58)

31 (18-34) 36 (26-42) 54 (51-58)

31 (18-35) 40 (20-63) 57 (23-59)

33 (16-55) 40 (20-63) 39 (23-56)

NS NS NS

* TGA was measured in normal PFP in the presence of cells with the irrelevant antibody (IgG1) or cells with AbTF. Quantitative variables were expressed as median (range). † Nonparametric Mann-Whitney test (BM-CTPs cells vs. PB-CTPs cells) or (BM-CTPs cells 1 AbTF vs. PB-CTPs cells 1 AbTF). ‡ Nonparametric Wilcoxon matched pairs test: BM-CTPs (cells vs. cells 1 AbTF) or PB-CTPs (cells vs. cells 1 AbTF).

Safety clinical outcome One patient out of 40 developed a deep vein thrombosis (DVT), which represents 2.5% (range, 0.06%-13.16%; binomial distribution). This DVT was located in a mus-

cular vein and was diagnosed 24 hours after the injection of PB-CTP. The affected patient was a nonsmoker, nonobese, 60-year-old man, who presented with diabetes mellitus, high blood pressure, and dyslipidemia. Volume 00, Month 2015 TRANSFUSION 7

TOURNOIS ET AL.

Before implantation, he had a mild anemia and a major inflammatory syndrome. On Day 0, D-dimer plasma level was below 0.5 mg/mL (Table S2, available as supporting information in the online version of this paper). For this patient, CTP was obtained by cytapheresis and was not different in its cell content from other PB-CTPs except for Hct, which was the highest in our series (Hct, 21%). On Day 1, D-dimers increased 7.1-fold and subsequently decreased on Day 2 to normalize on Day 7. Fg and CRP followed comparable kinetics (Table S2). The DVT outcome was favorable with conventional anticoagulant therapy. The patient who developed a DVT was alive without any major amputation 6 months after PB-CTP injection. Six months after cell therapy, 12 among the 20 patients treated with BM-CTPs were alive without any major amputation, six had been amputated at 47 (9-92) days, and two had died (at 82 and 156 days). Among the 20 patients treated with PB-CTPs, six had been amputated at 46 (7-169) days, and one had died at 145 days.

DISCUSSION The objective of this study was to assess the thrombogenicity of two CTPs. CLI patients present prothrombotic risk factors. In the context of autologous implantation (cells suspended in the patient’s plasma), patients are both donors and recipients for CTP. They currently present with inflammatory status at the time of implantation. The thrombogenicity of CTP has not been reported yet. To the best of our knowledge, it is the first report that directly focuses on TF expression and TGA supported by CTPs. We had the opportunity to compare two different types of CTPs obtained after either BM aspiration or cytapheresis. At baseline, CLI patients included in our study presented current cardiovascular risk factors and a proinflammatory status characterized by an elevation of IL-6, CRP, and Fg. IL-6 and CRP were reported as elevated in CLI patients compared to an age-matched control group.7,18 Similarly, Fg not only predicts the severity of PAD, but can also be used as a marker for the future development of PAD.19 To assess the balance between pro- and anti-inflammatory cytokines, we measured IL-10, a cytokine considered as beneficial in preventing atherosclerosis. In our cohort of patients, IL-10 levels were in the lower range. This is in agreement with reported data comparing CLI patients with matched controls.18 We next measured TF in the patient’s plasma samples. TF has been reported to be elevated in early stages of PAD.20-22 This study is the first to measure TF plasma in patients with CLI. Seventy-four percent of patients presented 1.5-fold higher values than normal range. plTF was positively correlated with age but not with D-dimer. Furthermore, the technique we used (ELISA) does not allow 8

TRANSFUSION Volume 00, Month 2015

to determine the origin of the measured TF (soluble form of TF and/or full-length TF antigen associated with different cell-derived microparticles).23,24 D-dimers were found to be elevated in 81% of patients at baseline. No correlation was found between D-dimer levels and any other studied clinical and biologic variables, suggesting that this biomarker is independent. An increased turnover of fibrin associated with high levels of both tissue plasminogen activator and D-dimer was reported in PAD patients.7,18 However, a study concluded that D-dimers were not associated with PAD severity.7 Furthermore, the age-related elevation of D-dimers must be taken into account and most of our patients were more than 70 years old. We next assessed CTP thrombogenicity analyzing cell composition: the cell content of each CTP was very different, PB-CTP containing higher amount of PLTs, red blood cells, mononuclear cells, and neutrophils, all cells being potentially thrombogenic. The method we used for cell preparation (as centrifugation) does not allow us to rule out the presence of microparticles.25 BM- and PB-CTPs were suspended in autologous patients’ plasma and were implanted within 3 hours after preparation into the ischemic leg. These patients’ plasma samples contained high levels of IL-6 and CRP, which are well known as TF inducers. Indeed, the levels of these molecules were above the threshold for TF induction by expressing cells (IL-6, 10-100 pg/mL;26 CRP, 5-100 mg/L).27 We showed that IL-6 mRNAs receptors were strongly expressed in CTPs, regardless their origin. We did not measure CD32, the CRP receptor involved in TF expression, but it is likely to be expressed by monocytes present in CTPs.28 Furthermore, lymphocytes were present in CTPs: this subtype of WBCs was described as mandatory to stimulate CRP and IL-6– induced TF expression by monocytes.29 However, the amount of TF mRNA was low in either CTP. Moreover, TF protein was barely detectable in both products. The fact that TF expression was low suggests that monocytes are not stimulated. The low thrombogenicity was confirmed by the fact that both CTPs supported very limited amount of TGA. Under such conditions, thrombin generation was partially inhibited by AbTF but null in FVII-deficient plasma. The intrinsic system was unlikely involved in this thrombin generation as aprotinin was added in the plasma. One cannot exclude a protease activity of cell preparations, which may explain residual thrombin generation. Under such conditions, the fact that TF was barely detectable in CTPs was rather unexpected. A possible explanation for this low expression of TF in CTPs could be attributed to the counterbalancing effect of IL-10.15 Indeed, IL-10 mRNAs receptors were strongly expressed in both CTPs. We do not keep this explanation since the level of IL-10 in autologous plasma was low and under the threshold (10 ng/mL)30 for TF down regulation, possibly suggesting a production of other TF down regulators such as IL-4 and transforming growth factor-b by the cells

CELL THERAPY IN CRITICAL LIMB ISCHEMIA

present in the CTPs.31 The contact of cells with autologous plasma may affect their properties in the CTPs. In the last part of this study, we evaluated the evolution of biomarkers of inflammation, thrombogenicity and muscle injury (namely, IL-6, IL-10, CRP, Fg, plTF, D-dimer, myoglobin, and CK activity) 24 hours after CTP implantation. We observed a moderate increase of CK activity and myoglobin in patients who were implanted with BM-CTPs. In any case, CK activity and myoglobin were not correlated with D-dimers. A significant increase of IL-6, CRP, and Ddimer after BM-CTP implantation was observed whereas CRP and D-dimer remain stable after PB-CTP implantation. An IL-63 and CRP5,32,33 increase have been previously reported, respectively, 9 to 12 and 24 hours after BM harvest and BM-CTP implantation. In our cohort of patients, the increase of IL-6 was moderate (1.4-fold higher) 24 hours after PB-CTP implantation and recovered to baseline level on Day 3 (compatible with minimal tissue trauma).34 IL-10 blood levels were increased 1 day after BM-CTP implantation. Such an IL-10 increase has been previously reported.18 Muscle implantation itself is not likely to be responsible for such an elevation since the same procedure was used with either type of CTP. The difference between CTPs may be related to the fact that BM aspiration is much more invasive (related to general anesthesia and surgical procedures) in terms of bone and muscle injury than cytapheresis.35,36 The low thrombogenic potential of cell therapy was confirmed by the clinical outcome of patients. plTF did not increase at H24. Most patients were currently under antithrombotic treatments, which were discontinued before the procedure. Thrombosis has been previously reported as a possible adverse event of BM-CTP therapy. In agreement with our study, one trial mentioned one single DVT out of 16 patients who received BM-CTP.33 Another study reported one thrombosis out of 40 treated patients; however, the procedure was different, since the cell infusion was done intraarterially and a low pressure balloon occlusion was used.37 In this study, one single patient developed a muscular calf vein thrombosis 24 hours after the reinjection of PB-CTP. Interestingly, this patient had normal baseline D-dimer. D-dimers along with Fg and CRP peaked at H24 and gradually normalized within 1 week. The DVT outcome was favorable under conventional anticoagulant therapy. Furthermore, histologic examination of legs of the first two patients who had had their leg amputated, treated with BM-CTPs, showed intrafascicular edema and inflammation, but no thrombosis.11 We are currently conducting a randomized trial comparing BM-CTP versus placebo (Trial Number NCT00904501). In conclusion, we demonstrated that CTPs have a low (if any) thrombotic potential. TGA is appropriate as it is sensitive, integrative, and easy to handle. In general, thrombogenicity seems to be underestimated, but we believe that it remains a major safety issue. Cell therapy in cardiovascular diseases is currently proposed as an auto-

logous procedure, in patients at high risk for thrombosis. It may be useful to evaluate thrombogenicity of CTPs in further studies, especially if an intravascular route is planned to infuse cell therapy.37 The need for antithrombotic prophylaxis should also be systematically examined. ACKNOWLEDGMENTS , The authors appreciate the contribution of C. Droull e, C. Mace M.-C. Mulpas, A.H. Moret (PRBI plateforme), S. Gobert, and M. Drame.

CONFLICT OF INTEREST The authors have disclosed no conflicts of interest. Authors had full access to the data.

REFERENCES 1. Norgren L, Hiatt WR, Dormandy JA, et al. Inter-society consensus for the management of peripheral arterial disease (TASC II). Eur J Vasc Endovasc Surg 2007;33 Suppl 1:S1-75. 2. Hirsch AT, Duval S. The global pandemic of peripheral artery disease. Lancet 2013;382:1312-4. 3. Tateishi-Yuyama E, Matsubara H, Murohara T, et al. Therapeutic angiogenesis for patients with limb ischaemia by autologous transplantation of bone-marrow cells: a pilot study and a randomised controlled trial. Lancet 2002;360: 427-35. 4. Wang ZX, Li D, Cao JX, et al. Efficacy of autologous bone marrow mononuclear cell therapy in patients with peripheral arterial disease. J Atheroscler Thromb 2014;21:1183-96. 5. Benoit E, Oâ&#x20AC;&#x2122;Donnell TF, Patel AN. Safety and efficacy of autologous cell therapy in critical limb ischemia: a systematic review. Cell Transplant 2013;22:545-62. 6. Sartori M, Favaretto E, Legnani C, et al. Thrombophilic risk factors and peripheral arterial disease severity. Thromb Haemost 2010;104:71-7. 7. Cassar K, Bachoo P, Ford I, et al. Markers of coagulation activation, endothelial stimulation and inflammation in patients with peripheral arterial disease. Eur J Vasc Endovasc Surg 2005;29:171-6. 8. Bjorck M, Lepkowska Eriksson M, Bylock A, et al. Plasminogen activator inhibitor-1 levels and activity decrease after intervention in patients with critical limb ischaemia. Eur J Vasc Endovasc Surg 2013;46:214-22. 9. Kawachi Y, Watanabe A, Uchida T, et al. Acute arterial thrombosis due to platelet aggregation in a patient receiving granulocyte colony-stimulating factor. Br J Haematol 1996;94:413-6. 10. Jonsson TB, Larzon T, Arfvidsson B, et al. Adverse events during treatment of critical limb ischemia with autologous peripheral blood mononuclear cell implant. Int Angiol 2012; 31:77-84. 11. Pignon B, Sevestre MA, Chatelain D, et al. Histological changes after implantation of autologous bone marrow Volume 00, Month 2015 TRANSFUSION 9

TOURNOIS ET AL.

mononuclear cells for chronic critical limb ischemia. Bone Marrow Transplant 2007;39:647-8. 12. Capiod JC, Tournois C, Vitry F, et al. Characterization and comparison of bone marrow and peripheral blood mononuclear cells used for cellular therapy in critical leg ischaemia: towards a new cellular product. Vox Sang 2009;96:256-65. 13. Mackman N. The many faces of tissue factor. J Thromb Haemost 2009;7 Suppl 1:136-9. 14. Carmeliet P, Mackman N, Moons L, et al. Role of tissue factor in embryonic blood vessel development. Nature 1996;383: 73-5. 15. Poitevin S, Ben Hadj Kalifa Kechiche S, Mac e C, et al. IL-10 inhibits apoptosis and microvesiculation of human monocytes. J Thromb Haemost 2009;7:1241-3. 16. Poitevin S, Garnotel R, Antonicelli F, et al. Type I collagen induces tissue factor expression and matrix metalloproteinase 9 production in human primary monocytes through a redox-sensitive pathway. J Thromb Haemost 2008;6:1586-94. 17. Cuccuini W, Poitevin S, Poitevin G, et al. Tissue factor upregulation in proinflammatory conditions confers thrombin generation capacity to endothelial colony-forming cells without influencing non-coagulant properties in vitro. J Thromb Haemost 2010;8:2042-52. â&#x201A;Źrsson H, Holmberg A, Siegbahn A, et al. Activation of 18. Pa coagulation and fibrinolytic systems in patients with CLI is not normalized after surgical revascularisation. Eur J Vasc Endovasc Surg 2004;27:186-92. 19. Paraskevas KI, Baker DM, Vrentzos GE, et al. The role of fibrinogen and fibrinolysis in peripheral arterial disease. Thromb Res 2008;122:1-12. 20. Blann AD, Amiral J, McCollum CN, et al. Differences in free and total tissue factor pathway inhibitor, and tissue factor in peripheral artery disease compared to healthy controls. Atherosclerosis 2000;152:29-34.

27. Nakagomi A, Freedman SB, Geczy CL. Interferon-gamma and lipopolysaccharide potentiate monocyte tissue factor induction by C-reactive protein: relationship with age, sex, and hormone replacement treatment. Circulation 2000;101: 1785-91. 28. Paffen E, Vos HL, Bertina RM. C-reactive protein does not directly induce tissue factor in human monocytes. Arterioscler Thromb Vasc Biol 2004;24:975-81. 29. Song CJ, Nakagomi A, Chandar S, et al. C-reactive protein contributes to the hypercoagulable state in coronary artery disease. J Thromb Haemost 2006;4:98-106. 30. Noble KE, Harkness D, Yong KL. Interleukin 10 regulates cellular responses in monocyte/endothelial cell co-cultures. Br J Haematol 2000;108:497-504. 31. Paysant J, Soria C, Cornillet-Lefebvre P, et al. Long-term incubation with IL-4 and IL-10 oppositely modifies procoagulant activity of monocytes and modulates the surface expression of tissue factor and tissue factor pathway inhibitor. Br J Haematol 2005;131:356-65. 32. Tachi Y, Fukui D, Wada Y, et al. Changes in angiogenesisrelated factors in serum following autologous bone marrow cell implantation for severe limb ischemia. Expert Opin Biol Ther 2008;8:705-12. 33. Malyar NM, Radtke S, Malyar K, et al. Autologous bone marrow mononuclear cell therapy improves symptoms in patients with end-stage peripheral arterial disease and reduces inflammation-associated parameters. Cytotherapy 2014;16:1270. 34. Jawa RS, Anillo S, Huntoon K, et al. Interleukin-6 in surgery, trauma, and critical care part II: clinical implications. J Intensive Care Med 2011;26:73-87. 35. Pulsipher MA, Chitphakdithai P, Logan BR, et al. Acute toxicities of unrelated bone marrow versus peripheral blood stem cell donation: results of a prospective trial from

21. Makin AJ, Chung NA, Silverman SH, et al. Vascular endothe-

the National Marrow Donor Program. Blood 2013;121:

lial growth factor and tissue factor in patients with established peripheral artery disease: a link between angiogenesis

197-206. 36. Lamplugh SM, Johnson P, Turner WL, et al. Changes in

and thrombogenesis? Clin Sci (Lond) 2003;104:397-404. 22. Gosk-Bierska I, Wysokin Ski W, Karnicki K, et al. Tissue factor, tissue pathway factor inhibitor and risk factors of atherosclerosis in patients with chronic limbs ischemia: preliminary study. Int Angiol 2008;27:296-301. 23. Szotowski B, Antoniak S, Rauch U. Alternatively spliced tissue factor: a previously unknown piece in the puzzle of hemostasis. Trends Cardiovasc Med 2006;16:177-82. 24. Owens AP 3rd, Mackman N. Microparticles in hemostasis and thrombosis. Circ Res 2011;108:1284-97. 25. Lacroix R, Dubois C, Leroyer AS, et al. Revisited role of microparticles in arterial and venous thrombosis. J Thromb Haemost 2013;11 Suppl 1:24-35. 26. Neumann FJ, Ott I, Marx N, et al. Effect of human recombinant interleukin-6 and interleukin-8 on monocyte procoagulant activity. Arterioscler Thromb Vasc Biol 1997;17: 3399-405.

TRANSFUSION Volume 00, Month 2015

serum creatine kinase isoenzyme activities after surgical procedures and cardioversion. Ann Clin Biochem 1979;16:315-9. 37. Walter DH, Krankenberg H, Balzer JO, et al. Intraarterial administration of bone marrow mononuclear cells in patients with critical limb ischemia: a randomized-start, placebo-controlled pilot trial (PROVASA). Circ Cardiovasc Interv 2011;4:26-37.

SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article at the publisherâ&#x20AC;&#x2122;s website: Table S1. Evolution of biological parameters after cell therapy. Table S2. Biological seven days (D) follow-up of the patient who developed DVT, 24 hours (H24) after PBCTP implantation.