- Email: [email protected]
Endothelial Dysfunction after Hematopoietic Stem Cell Transplantation: Role of the Conditioning Regimen and the Type of Transplantation
Endothelial Dysfunction after Hematopoietic Stem Cell Transplantation: Role of the Conditioning Regimen and the Type of Transplantation Marta Palomo,1 Maribel Diaz-Ricart,1 Carla Carbo,2 Montserrat Rovira,3 Francesc Fernandez-Aviles,3 Carmen Martine,3 Gabriela Ghita,3 Gines Escolar,1 Enric Carreras3 There is endothelial activation and damage in hematopoietic stem cell transplantation (HSCT). The impact of the conditioning and type of HSCT on endothelial dysfunction in the early phases of HSCT has been evaluated. Plasma samples were obtained before and at different times after autologous and allogeneic HSCTwith and without early complications. Changes in soluble markers of endothelial damage (VWF, ADAMTS-13, sVCAM-1, sICAM-1, and sTNFRI) were measured. There were changes in all markers evaluated that followed different patterns in auto and allo settings. For VWF and sTNRI, progressive increases from day Pre to day 14 and to day 21 were observed in the auto and the allo group, respectively. ADAMTS-13 activity correlated inversely with VWF levels. Levels of sVCAM-1 decreased until day 7, and raised significantly to day 14 and to day 21 in the auto and the allo HSCT, respectively. No significant changes were detected for sICAM-1. Our results confirm that there is endothelial damage at the early phases of HSCT, apparently induced by the consecutive effects of the conditioning, the proinflammatory agents used during transplantation, the translocation of endotoxins across the damaged gastrointestinal tract, and the engraftment. However, the comparative analysis between patients with and without complications suggests that none of these markers has diagnostic or prognostic value. Biol Blood Marrow Transplant 16: 985-993 (2010) Ó 2010 American Society for Blood and Marrow Transplantation
KEY WORDS: Endothelial dysfuntion, Inflammation markers, Hematopoietic stem cell transplantation
INTRODUCTION The endothelium has been recognized as a highly active metabolic and endocrine organ, producing a multitude of different molecules, including vasoactive peptide hormones, growth factors, coagulation factors, and adhesion molecules. It is an active biologic interface between blood and tissues that regulates the
From the 1Hemotherapy-Hemostasis Department, Centre de Diagno`stic Biome`dic, Institut d’Investigacions Biome`diques August Pi i Sunyer, Hospital Clınic, Universitat de Barcelona, Barcelona, Spain; 2Immune Disease Institute, Department of Pathology, Harvard Medical School, Boston, Massachusetts; and 3 Stem Cell Transplantation Unit, Hematology Department, Institut d’Investigacions Biome`diques August Pi i Sunyer, Hospital Clınic, Universitat de Barcelona, Barcelona, Spain. Financial disclosure: See Acknowledgments on page 992. Correspondence and reprint requests: Marta Palomo, Pre-doctoral Hemotherapy-Hemostasis Department, Hospital Clinic, Villarroel 170, 08036 Barcelona, Spain (e-mail: palomo_marta@ hotmail.com). Received October 8, 2009; accepted February 4, 2010 Ó 2010 American Society for Blood and Marrow Transplantation 1083-8791/$36.00 doi:10.1016/j.bbmt.2010.02.008
delicate balance between vasoconstriction and vasodilatation, coagulation and fibrinolysis, proliferation and apoptosis, as well as between transient adhesion and diapedesis of blood-borne leukocytes [1,2]. The endothelium is capable of a number of responses to local challenges. Some stimuli can produce a physiologic endothelial activation to maintain the properties of the endothelium and its environment, including changes in the vasomotor tone, development of a pro-coagulant activity, increase of the leukocyte adhesiveness, and the loss of its barrier function, allowing the passage of certain elements and solutes through the endothelial lining [1,2]. However, other stimuli can originate a true endothelial dysfunction, which produces not only abnormal functional changes but also structural changes (nuclear vacuolization, edema, and cytoplasm fragmentation, denudation, and lost of adhesion to the extracellular matrix). Depending on the extension of the damage, the endothelial dysfunction can be localized (in 1 or several organs) or systemic [1-3]. During hematopoietic stem cell transplantation (HSCT), endothelial cells (ECs) can be activated and damaged by the chemoradiotherapy included in the conditioning regimen, the cytokines produced by the 985
986
M. Palomo et al.
injured tissues, bacterial endotoxins translocated through the damaged gastrointestinal tract, some drugs used during the procedure (as granulocytecolony stimulating factor [G-CSF] or calcineurin inhibitors), the complex process of the engrafment, and allogeneic reactions with the donor-derived immune cells [4], among others. There is evidence suggesting that some complications appearing early after HSCT, as veno-occlusive disease of the liver [5,6], capillary leak syndrome [7], thrombotic microangiopathy (TMA) [8,9], engraftment syndrome (ES) [10], diffuse alveolar hemorrhage [11], idiopathic pneumonia syndrome, and graft-versus-host disease (GVHD) [12,13], can have their origin in a localized or systemic endothelial damage. EC activation and damage can be evaluated through the analysis of several soluble markers, including plasma thrombomodulin, von Willebrand factor (VWF), and soluble forms of adhesion molecules [14], the enumeration of circulating ECs [15], and the evaluation of morphologic and functional changes induced on cultured ECs [16,17]. The present study was designed as a extension of our initial in vitro studies on the proinflammatory and prothrombotic effects of the sera of patients receiving HSCT on human umbilical vein ECs in culture [4]. The effect of the different types of HSCT, and the conditioning regimens applied, on the integrity of the endothelium was evaluated in vivo. The potential diagnostic and/or prognostic utility of the soluble markers evaluated as indicators of endothelial related complications in HSCT was also explored. PATIENTS AND METHODS Patients We included 46 selected patients (mean age 48.5 6 14.1 years, ranging from 23-71 years; 20 female, 26 male), suffering from hematologic diseases, and receiving autologous (n 5 22) or allogeneic HSCT (n 5 24, 7 from related and 17 from unrelated donors) in our institution from January 2006 to January 2007. Diagnosis included acute lymphoblastic leukemia (ALL), (n 5 3), acute myelogenous leukemia (AML) (n 5 10), chronic lymphocytic leukemia (CLL) (n 5 2), chronic myelogenous leukemia (CML) (n 5 1), Hodgkin lymphoma (HL) (n 5 6) and non-Hodgkin lymphoma (NHL) (n510), myelodysplastic syndrome (MDS) (n 5 2), and multiple myeloma (MM) (n 5 12). The 4 following conditioning regimens were applied: (1) BEAM: BCNU 300 mg/m2 at day 27; etoposide 400 mg/m2 at days 26, 25, 24, and 23; cytosine arabinoside 200 mg/m2 at days 26, 25, 24, and 23; and melphalan (Mel) 140 mg/m2 at day 22 (n 5 11); (2) MLF: Mel 100 mg/m2 at days 23 and 22 (n 5 11); (3) Cy/TBI: cyclophosphamide 60 mg/kg at days 27 and 26 and total body irradiation 12 Gy in 4 fractions at days 24,
Biol Blood Marrow Transplant 16:985-993, 2010
23, 22, and 21 (n 5 14); and (4) Flu/MLF: reducedintensity conditioning (RIC) with fludarabine 30 mg/ m2 at days 27 to 23, followed by Mel 70 mg/m2 at days 23 and 22 (n 5 10). Cyclosporine (CsA) from day 27 plus short course of methotrexate or mycophenolate mofetil were administered for GVHD prophylaxis in patients receiving Cy/TBI or Flu/MLF conditioning, respectively. Twenty-seven patients had an HSCT without complications defined by the absence of signs of organ dysfunction (bilirubin \2.0 mg/dL, urea \1.5 mg/dL, lactate dehydrogenase (LDH) \300 U/L, glutamatepyruvate transaminase \100 U/L), absence of GVHD, veno-occlusive disease (VOD), TMA, or sepsis, no need for plasma exchange (plasmapheresis, dialysis, hemofiltration), no administration of fresh frozen plasma or steroids, no respiratory failure, and no need of ventilation or vasopressors. Nineteen patients developed transplant-related complications (TRC): engrafment syndrome (ES) in the autologous group (n 5 10), and GVHD (n 5 7) and VOD (n 5 2), both in the allogeneic HSCT. ES was diagnosed in 5 patients from each treatment group in the autologous setting, with the presence of noninfectious fever (.38 C) plus at least 1 of the following events: pulmonary infiltrates with no signs of infection, cardiac failure or pulmonary embolism; hypoxemia; skin rash involving .25% of body surface area; started 24 hours before or after the first appearance of neutrophils in peripheral blood. Acute GVHD (aGVHD) in patients receiving allogeneic HSCT was graded by classical criteria [18]. Hepatic VOD was diagnosed and classified by the presence of painful hepatomegaly, fluid retention, and weight gain and elevated serum bilirubin concentration .2 mg/dL, according to the Baltimore criteria [19]. In addition, plasma samples from 5 additional patients developing TMA (diagnosis based on clinical features characterized by neurologic disturbance, microangiopathic hemolytic anemia, renal insufficiency, and thrombocytopenia) were collected at the time of the thrombotic complication (41 6 18 days). Control samples from healthy donors (n 5 44, mean age 34 6 16.2 years, ranging from 21 to 54 years; 18 female, 26 male) were included in each determination. Informed consent was obtained from all patients and controls for blood utilization. The study was approved by the ethical committee of the Hospital Clınic and was carried out according to the principles of the Declaration of Helsinki. Sample Collection Blood samples were obtained at different time points of the study: Pre (before starting conditioning regimen), day 0 (immediately before the infusion of the hematopoietic stem cells), and days 7, 14, and 21
Soluble Markers of Endothelial Damage in HSCT
Biol Blood Marrow Transplant 16:985-993, 2010
BEAM 2,5
2,5
*
2
2 1,5
1,5
1
1 0,5
0,5
*
0 Pre
Day 0
Day 7
2
*
1,5
*
*
*
2 1,5 1
1 0,5
*
0 Pre
Day 0
Day 7
2
0,5 0
Day 14 Day 21
1
1
0,5
0,5
*
0 Day 0
Day 7
0
Day 14 Day 21
FLU/MLF
2,5 2
*
1,5
*
2,5
*
*
2 1,5 1
1 0,5
*
0,5 0
0 Pre
VWF
2 1,5
Pre
2,5
2,5
*
1,5
Day 14 Day 21
Cy/TBI 2,5
0
MLF
2,5
987
Day 0
Day 7
Day 14 Day 21
ADAMTS-13
Figure 1. Changes in plasmatic levels of vWF and ADAMTS-13 activity. Graphs represent changes in VWF (C) and ADAMTS-13 activity (B) levels in plasma samples from recipients of autologous HSCT with BEAM and MLF conditioning treatments, and of allogeneic HSCT with Cy/TBI and Flu/MLF treatments, as indicated. Levels were determined at the different time points of the study: before starting conditioning regimen (Pre), immediately before the infusion of the hematopoietic stem cells (day 0), and days 7, 14, and 21 after the HSCT. Data are expressed as ratios with respect to the Pre value and correspond to mean 6 SEM (*P \.05; n 5 6, 5, 5, and 8 for the BEAM, MLF, Cy/TBI, and Flu/MLF, respectively).
after the HSCT. Samples were always obtained through a central line to avoid contamination with endothelial cells. Plasma samples were obtained by centrifugation of blood at 800 g for 15 min and immediately stored at 220 C until the day of the analysis.
considered statistically significant when P \ .05. The SPSS statistical package 17.0.0 (SPSS Inc, Chicago, IL) was used for all analyses.
RESULTS Analysis of Soluble Markers of Endothelial Dysfunction The levels of plasma VWF, sVCAM-1, sICAM-1, and sTNFRI were measured by the enzyme-linked immunoabsorvent assay (ELISA) (American Diagnostica, Grifols SA, Barcelona, Spain; Chemicon, Grupo Taper SA, Madrid, Spain; and HycCult biotechnology b.v, Uden, The Netherlands, Diagnostica Stago SA). ADAMTS-13 activity was analyzed in citrated plasma by using a fluorescence resonance energy transfer assay using a truncated synthetic 73-amino acid VWF peptide as a substrate (FRETS-VWF73 assay) [20,21]. In addition, the presence of IgG anti-ADAMTS-13 (Technoclone GmbH, Viena) was also explored in those patients that developed HSCT-associated TMA. All measurements were performed in duplicate and concentrations determined from the standard curve according to the manufacturer’s instructions. Statistics Results are expressed as mean 6 standard error of the mean (SEM) and as ratios with respect to values at day Pre. Statistical analysis was performed with raw data using the Student’s t-test for paired samples and analysis of variance (ANOVA). Results were
Changes in Plasma Levels of VWF Changes in plasma VWF values from patients without TRC followed different kinetics depending on the type of transplant (see ratios on Figure 1). In the autologous setting under treatment with either BEAM or MLF, we found a progressive increase in values of plasma VWF along the follow-up of the patients until day 14, followed by a decrease from day 14 to day 21. In absolute numbers, levels of VWF in patients treated with BEAM were of 97.7% 6 15.7% at day Pre, and increased to 126.1% 6 16.2% at day 0, 159.4% 6 25.8% at day 7, with a peak of 181.1% 6 22.9% at day 14 (P\.05 versus day Pre), and decreased to 164.0% 6 26.3% at day 21. In patients treated with MLF, VWF levels were of 81.0% 6 11.1% at day Pre, increased to 95.1% 6 11.5% at day 0, 136.9% 6 11.1% at day 7, with a peak of 148.8% 6 12.1% at day 14 (P \ .05 versus day Pre), and decreased to 137.0% 6 12.7% at day 21. In the allogeneic group, the increase in VWF plasma levels after both treatments, Cy/TBI and Flu/ MLF, was statistically significant at all time points studied and persistent until day 21, being more notable with Cy/TBI (see Figure 1). Values of plasma VWF in patients treated with Cy/TBI were of 108.7% 6 10.7%
988
M. Palomo et al.
Biol Blood Marrow Transplant 16:985-993, 2010
at day Pre, and increased to 142.6% 6 9.7% at day 0, 165.4% 6 12.2% at day 7, 174.6% 6 7.5% at day 14, and 189.8% 6 15.2% at day 21 (P \ .05 versus day Pre for all values). In patients treated with Flu/MLF, VWF levels were of 95.2% 6 7.0% at day Pre, increased to 133.0% 6 8.4% at day 0, 150.2% 6 9.7% at day 7, 157.8% 6 12.9% at day 14, and 169.2% 6 12.4% at day 21 (P \ .05 versus day Pre for all values). No significant differences were observed when comparing the initial levels of VWF in the 4 groups of patients (day Pre) with levels in a group of controls (83.5% 6 11.3%, mean 6 SEM, n 5 44). Modifications in the ADAMTS-13 Activity The activity of the metalloprotease ADAMTS-13 in plasma from patients without TRC showed a negative association with plasma VWF levels (see Figure 1). Although basal ADAMTS-13 activity values (day Pre) were in the normal range, there was certain variability between the 4 groups under study (82.1% 6 6.1% in BEAM, 125.2% 6 7.7% in MLF, 88.5% 6 7.5% in Cy/TBI, and 108.8% 6 7.1% in Flu/MLF). A decline in the protease activity after the conditioning regimen appeared to be a constant finding. In absolute values, ADAMTS-13 activity in the patients treated with BEAM was of 82.1% 6 6.1% at day Pre, 80.7% 6 5.4% at day 0, 69.7% 6 6.5% at day 7, 59.4% 6 6.1% at day 14 (P \ .05 versus day Pre), and 66.8% 6 8 .8% at day 21. In patients treated with MLF, ADAMST-13 activity was of 125.2% 6 2.7% at day Pre, 101.8% 6 7.1% at day 0, 114.3% 6 sVCAM-1
2,5
Changes in the presence of the soluble endothelial dysfunction markers VCAM-1 and ICAM-1 in plasma samples from patients without TRC from the autologous and allogeneic groups are shown in Figure 2, expressed as ratios. As a general observation, the kinetics for sVCAM-1 changes along the study were similar when comparing BEAM and MLF, in the
1,5
1
*
0,5
*
1
*
0,5 0
Pre
Day 0
Day 7
Day 14
Day 21
Pr e
BEAM
sVCAM-1
2,5
*
2 1,5
*
*
1 0,5 0
sICAM-1
2
1,5
0
Presence of Soluble Adhesion Molecules VCAM-1 and ICAM-1 in Plasma
2,5
*
2
9.1% at day 7, 95.4% 6 6.7% at day 14 (P \ .05 vs. day Pre), and 92.9% 6 9.9% at day 21. In the allogeneic setting, ADAMTS-13 activity in patients under Cy/TBI treatment was of 88.5% 6 7.5% at day Pre, 88.1% 6 8.4% at day 0, 78.4% 6 4.5% at day 7, 61.7% 6 9.1% at day 14 (P \ .05 versus day Pre), and 81.5% 6 14.3% at day 21. In patients treated with Flu/MLF, ADAMTS-13 activity was of 108.8% 6 7.2% at day Pre, 92.0% 6 7.0% at day 0, 82.7% 6 9.1% at day 7, 69.3% 6 9.9% at day 14 (P \ .05 versus day Pre), and 72.1% 6 7.6% at day 21. Moreover, during the present study, we had the opportunity to evaluate 5 cases of TMA. Evaluation of ADAMTS-13 activity and VWF levels was carried out at the time of diagnoses. In these patients, ADAMTS-13 activity was reduced (42.3% 6 5.9%, mean 6 SEM for the n 5 5), and no inhibitor against ADAMTS-13 was detected in any of the samples. The plasma levels of VWF in the 5 patients were significantly higher than control values (180% 6 20% versus 83.5% 6 11.3%, P \ .05).
Day 0
Day 7
Day 14
Day 21
MLF
sICAM-1
2,5 2 1,5 1 0,5
* Pr e
Day 0
Day 7
Day 14
Day 21
Cy/TBI
0
Pre
Day 0
Day 7
Day 14
Day 21
FLU/MLF
Figure 2. Changes in the levels of sVCAM-1 and sICAM-1 in patients without TRC from the autologous and allogeneic groups. Graphs represent changes in soluble VCAM-1 and ICAM-1 levels in plasma from recipients of autologous HSCTwith BEAM (C) and MLF (B) conditioning treatments, and of allogeneic HSCTwith Cy/TBI (:) and Flu/MLF (D) treatments. Levels were determined at the different time points of the study: before starting conditioning regimen (Pre), immediately before the infusion of the hematopoietic stem cells (day 0), and days 7, 14, and 21 after the HSCT. Data are expressed as ratios with respect to the Pre value and correspond to mean 6 SEM (*P \ 0.05; n 5 6, 5, 5, and 8 for the BEAM, MLF, Cy/TBI, and Flu/MLF, respectively).
autologous group, and also when comparing Cy/TBI and Flu/MLF, in the allogeneic group. No significant changes were detected when analyzing plasma levels of sICAM-1. In the autologous group, we observed a slight decline in sVCAM-1 from day Pre to day 0 (from 283.9 6 32 ng/mL at day Pre to 237.7 6 12.4 ng/mL at day 0 in patients treated with BEAM, and from 484.1 6 35.1 ng/mL at day Pre to 393.8 6 34 ng/mL in the MLF group), remaining similar until day 7 (259.8 6 69.5 ng/mL in BEAM, 349.3 6 49.2 ng/ mL in MLF), increasing significantly to a maximum peak at day 14 (439.8 6 70.3 ng/mL in BEAM, 583.3 6 65.8 ng/mL in MLF, P \ .05 for both with respect to each Pre value), and decreasing slightly from day 14 to day 21 (409.6 6 95.4 ng/mL in BEAM, 514.1.3 6 49.6 ng/mL in MLF). Changes of sICAM-1 levels in the same group of patients were not as notable as those observed for sVCAM-1, and showed a moderate tendency to increase until day 14 (see Figure 2). In absolute numbers for the BEAM group, levels of sICAM-1 were of 189.5 6 24.7, 196.6 6 23.6, 208.1 6 7.8, 284.7 6 58.2, and 241.1 6 24.2 ng/mL, at days Pre, 0, 7, 14, and 21, respectively. For the MLF group, levels were of 194.7 6 25.1, 183.3 6 24.5, 197.6 6 48.6, 261.2 6 65.4, and 245.8 6 18.3 ng/mL, at days Pre, 0, 7, 14, and 21, respectively. In the allogeneic group and in association with both treatments, Cy/TBI and Flu/MLF, the plasma sVCAM-1 levels decreased significantly after the conditioning regimen (from 588.9 6 149.8 ng/mL at day Pre to 289.8 6 89.9 ng/mL at day 0, P \ .05, for Cy/TBI, and from 711.5 6 170.6 ng/mL at day Pre to 642.6 6 129.9 ng/mL at day 0, P \ .05), followed diverging tendencies from day 0 to day 7 (314.4 6 60.3 ng/mL in Cy/TBI and 405.6 6 84.9 ng/mL in Flu/MLF), and increased progressively to day 14 (572.5 6 121.3 and 685.6 6 138.1 ng/mL in Cy/TBI and FLU/MLF, respectively) and significantly to day 21 (750.1 6 125.9 and 1001.1 6 172.2 ng/mL in Cy/ TBI and Flu/MLF, respectively, P \ .05 for both with respect to Pre value). Changes in sICAM-1 levels showed the same tendency as that observed for sVCAM-1, although much lower and never reached statistical significance. No significant differences were observed when comparing the initial levels of both sVCAM-1 and sICAM-1 in the 4 groups of patients (day Pre) with levels in a group of controls (434.7 6 31.5 ng/mL and 247.2 6 21.5 ng/mL, respectively, mean 6 SEM, n 5 44). Soluble TNF Receptor I Plasma Levels Profiles of plasmatic soluble TNF receptor I in patients without TRC are represented in Figure 3.
989
Soluble Markers of Endothelial Damage in HSCT
Biol Blood Marrow Transplant 16:985-993, 2010
sTNFRI
3,5 3
*
2,5 2 1,5 1
*
0,5 0
Pr e
Day 0
Day 7
BEAM
Day 14 MLF
3,5
*
3 2,5
*
2
Day 21
*
*
1,5 1 0,5 0
Pr e
Day 0 Cy/TBI
Day 7
Day 14
Day 21
FLU/MLF
Figure 3. Levels of plasmatic soluble TNF receptor I in patients without TRC. Graphs representing changes in soluble TNFRI levels in plasma from recipients of autologous HSCTwith BEAM (C) and MLF (B) conditioning treatments, and of allogeneic HSCT with Cy/TBI (:) and Flu/ MLF (6) treatments. Levels were determined at the different time points of the study: before starting conditioning regimen (Pre), immediately before the infusion of the hematopoietic stem cells (day 0), and days 7, 14, and 21 after the HSCT. Data are expressed as ratios with respect to the Pre value and correspond to mean 6 SEM (*P \.05; n 5 6, 5, 5, and 8 for the BEAM, MLF, Cy/TBI and Flu/MLF, respectively).
In the autologous group, the analysis of TNFRI showed a progressive increase from day Pre (1160.7 6 358.6 pg/mL in BEAM and 889 6 76.8 pg/mL in MLF group) to day 14, at which values reached a statistical significant peak (2299.3 6 805 pg/mL in BEAM and 1420.6 6 135 pg/mL in MLF) (P \ .05 respect day Pre) and decreased to day 21 (1642.6 6 140.5 pg/mL in BEAM and 1273.3 6 169.9 pg/ml in MLF). In the allogeneic setting, we observed a progressive increase of soluble TNFR I from day Pre (858.8 6 126.4 pg/mL in Cy/TBI and 910.4 6 140.4 pg/mL in Flu/MLF) to day 21, being much more notable in Flu/MLF than in Cy/TBI group (2562.3 6 617.1 pg/mL, and 1606 6 411.8 pg/mL, respectively). No significant differences were observed when comparing the initial levels of both TNFRI in the 4 groups of patients (day Pre) with levels in a group of controls (1007.9 6 211.5, n 5 44).
990
M. Palomo et al.
Biol Blood Marrow Transplant 16:985-993, 2010
a 250
*
*
200 150
*
*
100
* *
50 0
Pre
Day 7
Day 0 ES
Day 14
Day 21
VWF levels than in patients without any complication, although no statistical significance was reached. In fact, changes in plasma levels of VWF along the HSCT treatment versus day Pre reached statistical significance at the same time points (days 7, 14, and 21, P \ .05 in all cases) in both groups of patients, with TRC and without complications (Figure 4). Comparison of values for ADAMTS-13 activity, sVCAM-1, sICAM-1, and sTNFRI among patients with and without complications in each group did not show a clear tendency and never reached statistical significance (data not shown).
No ES
b DISCUSSION
250 200
*
150
*
100
*
*
*
*
50 0 Pre
Day 0
Day 7
GvHD
Day 14
Day 21
No GvHD
c 250 200 150 100 50 0
Pre
Day 0
Day 7 VOD
Day 14
Day 21
No VOD
Figure 4. Kinetics of VWF in patients with TRC versus those without complications. Graphs show VWF levels in recipients of autologous HSCT (a) and allogeneic HSCT (b and c) with () and without (o) transplant related complications. ES 5 engraftment syndrome (n 511); GVHD 5 graft-versus-host disease (n 5 7); VOD 5 venooclusive disease (n 5 2). Data are expressed as absolute values in % (mean 6 SEM) (*P \.05; for days 7, 14, and 21 versus day Pre).
Evaluation of the Prognostic Value of the Markers Analyzed When comparing the kinetics for VWF in patients with TRC versus those without complications, some differences could be observed (see Figure 4). In the auto group, those patients that developed ES showed consistently higher levels of VWF than those that did not develop any complication associated with HSCT. When analyzing allogeneic HSCT recipients who developed GVHD or VOD, we also found higher
We have previously demonstrated, by using an in vitro approach, that the exposure of ECs in culture to sera from patients that have received an HSCT causes endothelial activation and damage, and that these effects are not limited to the allogeneic but also to the autologous setting [4]. In the present study, we have extended our investigation to characterize the proinflammatory effect of the different conditioning regimens commonly applied in both modalities of HSCT, and to find potential diagnostic and/or prognostic markers of the early TRC apparently originated by endothelial damage. We have observed changes in plasma levels of VWF, ADAMTS-13 activity, sVCAM-1, and sTNFRI in association with both types of transplant, although following different kinetics. The degree of these changes was related to the intensity of the conditioning treatment. Although all these markers are good indicators of endothelial activation and damage, from the comparative analysis of the results from patients with and without complications, it can be derived that none of these markers seems to have prognostic value. The evaluation of changes in the release of VWF revealed different endothelial activation kinetics during HSCT probably depending on the conditioning regimen, the use of G-CSF or CsA, the translocation of endotoxins during the period of aplasia, the engraftment and, probably, the allo-reactivity. The release of VWF at the moment of the allogeneic transplant (day 0) is persistently higher than in the autologous setting, a result that is in accordance with the obtained by Salat et al. [14]. This fact could be explained by the higher cytotoxic effect of Cy in combination with TBI [22], or by the well-known endothelial toxicity of Flu [23] and MLF. It is interesting to point out that in all our studies the toxic effect of MLF on the endothelium is similar to that exerted by BEAM, which is a regimen traditionally considered more intense than MLF. In addition, in the allo setting a role of the CsA administered from day 27 could not be excluded. It has been clearly demonstrated that CsA produces endothelial
Biol Blood Marrow Transplant 16:985-993, 2010
damage and induces the release of VWF [24,25]. More difficult to explain is the evolution of VWF levels from day 0 to day 114 in autologous HSCT and to 121 in allogeneic HSCT. In the auto setting, the use of GCSF, agent with a well-known proinflammatory effect [26], can contribute to the endothelial damage from day 17 to day 114. Similarly, in the allo setting CsA can contribute to the damage observed along all the study period. However, these 2 factors could not explain the progressive increase in VWF levels from day 0 to 17 in autologous HSCT and up to day 121 in allogeneic HSCT. We believe that this observation can be the consequence of both the translocation of endotoxins through the damaged gastrointestinal tract during the aplastic phase, followed by the endothelial dysfunction that the complex process of engraftment may induce. From day 0, the stem cell infused adhere to bone marrow endothelium interacting with selectins and several surface adhesion molecules as VCAM [27]. After adherence, the transendothelial migration and intraparenchymal homing to hematopoietic niches take place [28]. Finally, after homing and proliferation, the opposite process take place, which results in the release of neutrophils to peripheral blood, a process that usually occurs between days 9 and 12 in autologous HSCT and between days 14 and 20 in allogeneic HSCT. Finally, a possible role of allo-reactivity in this late period of allo-HSCT could not be excluded. Althought other studies had evaluated several soluble markers after day 128 [14,29,30], unfortunately, the design of our study did not allow the evaluation of the endothelial function after day 121. ADAMTS-13 activity results were in good inverse correlation with those obtained for VWF, as previously reported by Mannucci et al. [31]. It is interesting to note that ADAMTS-13 activity decreased significantly when VWF values increased around 50% over the initial value. In addition, ADAMTS-13 activity in the patients included in the study never decreased under 60%, with the exception of those patients developing TMA in which ADAMTS-13 activity was of 42.3% 6 5.9% (value in the normal range). These observations contribute to discard a pivotal role of the metalloproteinase in the development of TMA associated with the HSCT [29]. The release of soluble forms of adhesion receptors sVCAM-1 and sICAM-1 is a reflection of the activation of the endothelium [32]. Our results showed a significant decrease of sVCAM-1 along the conditioning treatment. This decrease was more notable after the conditioning with Cy/TBI, probably because of the high intensity of this treatment. After this initial decline, these receptors increased following the same kinetics observed for VWF depending on the type of HSCT. Conditioning implies the use of aggressive immunosuppressive and cytotoxic treatments that may interfere with the synthesis of proteins. The induced
Soluble Markers of Endothelial Damage in HSCT
991
expression of VCAM-1 and ICAM-1 is largely dependent on synthesis of new mRNA and protein because there are no storage forms of this endothelial adhesion proteins [33]. In the auto setting, levels of sVCAM raised from day 17, at which G-CSF is administered, to day 114, just after the engraftment time. In the allo setting, the progressive increase in sVCAM-1 levels could be mainly because of the engraftment and CsA toxicity. The fact that values of sVCAM-1 were more relevant than those of sICAM-1 could be attributed to a more restricted expression of the former to the endothelium [34], and it suggests a direct endothelial effect of the conditioning process. Apart from this, because the kinetics of these adhesive molecules correlate with the presence of inflammatory cells in the circulation, the role of leukocytes along the HSCT treatment cannot be disregarded. Leukocytes could contribute with the release of either proinflammatory cytokines or even adhesion molecules to the circulation. sVCAM-1 has been shown to increase in response to proinflammatory cytokines such as tumor necrosis factor-alpha (TNF-a), which we have evaluated through the release of sTNFRI. Similar to the other markers evaluated in this study, the kinetics of sTNFRI changes between the 2 types of HSCT. We were unable to detect reliable changes in TNF-a in the plasma of the patients studied (data not shown), as reported by Andersen et al. [35]. In fact, a variable tendency in the changes of plasma levels of TNFa has been described by other authors in many other pathophysiologic conditions [36]. Presence of soluble receptors for TNF-a (sTNFR) has been reported to follow a good correlation with the cytokine. Soluble TNFRs act as inhibitors of TNF-a activity and represent a natural protective mechanism against the cytotoxic effect of this protein by preventing its binding to the surface receptors. In fact, TNF-a circulates bound to sTNFRI in a cryptic inactive form [37]. The expression of sTNFRI seems to be regulated not only by presence of TNF-a but also by other proinflammatory mechanisms. In the present study, sTNFRI increased from Pre to 0 in the allogeneic setting, but also in the autologous. In addition, levels of plasma sTNFRI in the patients studied followed kinetics similar to those described for VWF suggesting the same pathogenetic events previously mentioned. The main difference was that in the allogeneic setting these levels were higher in association with the Flu/MLF treatment, although this observation is of unclear clinical significance. The comparative analysis of data between patients with and without complications did not show a clear tendency or statistical significance for all the markers analyzed, but did for VWF levels. Levels of plasma VWF in patients with complications were consistently higher in all groups studied, although these differences
992
M. Palomo et al.
never reached statistical significance. Therefore, from the present study we can conclude that the markers evaluated have to be ruled out as prognostic markers of endothelial related complications associated to the HSCT. These results may be ascribed either to the nature of the markers or to the low number of patients included, although they may suggest additional mechanisms for the development of these complications in addition to the endothelial damage. In conclusion, there is endothelial activation and damage in association with both autologous and allogeneic HSCT, as demonstrated by the increased plasma levels of the markers evaluated. The kinetics of changes observed in patients without transplantrelated complications indicate that the engrafment induces maximal endothelial activation in both autologous and allogeneic HSCT. The progressive increase in the markers evaluated in the allogeneic setting may be indicating that the initiation of the allogenicity process is the main cause of endothelial damage in this type of HSCT. Although previous reports strongly suggest that endothelial damage may play a major role in the development of TRC, our present results suggest additional mechanisms. Search for diagnostic and prognostic markers, and for prevention strategies should be the subject of future investigation.
ACKNOWLEDGMENTS Financial discolusre: This work has been partially supported by grants FIS PI060260 and FIS PI080156 (Fondo de Investigaciones de la Seguridad Social), German Jose Carreras Leukaemia Foundation (R 07/ 41v), SGR2005-00952 (Generalitat de Catalunya), SAF2006-08003 (Ministerio de Ciencia y Tecnologıa), and RD06/0009/1003 (Red HERACLES, Instituto de Salud Carlos III). We thank Fulgencio Naval on, Jose M. Rodrıguez, Patricia Molina, and Marc Pino for their technical assistance in this study.
REFERENCES 1. Aird WC. Spatial and temporal dynamics of the endothelium. J Thromb Haemost. 2005;3:1392-1406. 2. Aird WC. The endothelium: a primer. UpToDate 17. 1. Available at: www.uptodate.com, 2009. 3. Mohler ER. Endothelial dysfunction. UpToDate 17. 1. Available at: www.uptodate.com, 2009. 4. Palomo M, Diaz-Ricart M, Carbo C, et al. The release of soluble factors contributing to endothelial activation and damage after hematopoietic stem cell transplantation is not limited to the allogeneic setting and involves several pathogenic mechanisms. Biol Blood Marrow Transplant. 2009;15:537-546. 5. DeLeve LD, Ito Y, Bethea NW, McCuskey MK, Wang X, McCuskey RS. Embolization by sinusoidal lining cells obstructs the microcirculation in rat sinusoidal obstruction syndrome. Am J Physiol Gastrointest Liver Physiol. 2003;284:G1045-G1052.
Biol Blood Marrow Transplant 16:985-993, 2010
6. DeLeve LD, Wang X, Kanel GC, et al. Decreased hepatic nitric oxide production contributes to the development of rat sinusoidal obstruction syndrome. Hepatology. 2003;38:900-908. 7. Nurnberger W, Willers R, Burdach S, Gobel U. Risk factors for capillary leakage syndrome after bone marrow transplantation. Ann Hematol. 1997;74:221-224. 8. George JN, Li X, McMinn JR, Terrell DR, Vesely SK, Selby GB. Thrombotic thrombocytopenic purpura-hemolytic uremic syndrome following allogeneic HPC transplantation: a diagnostic dilemma. Transfusion. 2004;44:294-304. 9. George JN. ADAMTS13, thrombotic thrombocytopenic purpura, and hemolytic uremic syndrome. Curr Hematol Rep. 2005; 4:167-169. 10. Spitzer TR. Engraftment syndrome following hematopoietic stem cell transplantation. Bone Marrow Transplant. 2001;27:893-898. 11. Afessa B, Tefferi A, Litzow MR, Krowka MJ, Wylam ME, Peters SG. Diffuse alveolar hemorrhage in hematopoietic stem cell transplant recipients. Am J Respir Crit Care Med. 2002;166: 641-645. 12. Biedermann BC, Sahner S, Gregor M, et al. Endothelial injury mediated by cytotoxic T lymphocytes and loss of microvessels in chronic graft versus host disease. Lancet. 2002;359:2078-2083. 13. Biedermann BC. Vascular endothelium and graft-versus-host disease. Best Pract Res Clin Haematol. 2008;21:129-138. 14. Salat C, Holler E, Kolb HJ, Pihusch R, Reinhardt B, Hiller E. Endothelial cell markers in bone marrow transplant recipients with and without acute graft-versus-host disease. Bone Marrow Transplant. 1997;19:909-914. 15. Woywodt A, Scheer J, Hambach L, et al. Circulating endothelial cells as a marker of endothelial damage in allogeneic hematopoietic stem cell transplantation. Blood. 2004;103:3603-3605. 16. Serradell M, Diaz-Ricart M, Cases A, et al. Uremic medium causes expression, redistribution and shedding of adhesion molecules in cultured endothelial cells. Haematologica. 2002;87: 1053-1061. 17. Serradell M, Diaz-Ricart M, Zurbano MJ, et al. Uremic media affects hemostatic properties of human endothelial cells in culture and increases the production of von Willebrand factor. Nefrologia. 2002;22:33-41. 18. Sullivan KM, Agura E, Anasetti C, et al. Chronic graft-versushost disease and other late complications of bone marrow transplantation. Semin Hematol. 1991;28:250-259. 19. Jones RJ, Lee KS, Beschorner WE, et al. Venoocclusive disease of the liver following bone marrow transplantation. Transplantation. 1987;44:778-783. 20. Kokame K, Nobe Y, Kokubo Y, Okayama A, Miyata T. FRETSVWF73, a first fluorogenic substrate for ADAMTS13 assay. Br J Haematol. 2005;129:93-100. 21. Kremer Hovinga JA, Mottini M, Lammle B. Measurement of ADAMTS-13 activity in plasma by the FRETS-VWF73 assay: comparison with other assay methods. J Thromb Haemost. 2006;4:1146-1148. 22. Eissner G, Kohlhuber F, Grell M, et al. Critical involvement of transmembrane tumor necrosis factor-alpha in endothelial programmed cell death mediated by ionizing radiation and bacterial endotoxin. Blood. 1995;86:4184-4193. 23. Eissner G, Multhoff G, Gerbitz A, et al. Fludarabine induces apoptosis, activation, and allogenicity in human endothelial and epithelial cells: protective effect of defibrotide. Blood. 2002; 100:334-340. 24. Tiemann C, Prohaska W, Ko¨rfer R, Ko¨rner M, Brinkmann T, Kleesiek K. Effect of cyclosporine A on the release of tissue factor pathway inhibitor from endothelial cells in heart transplant patients and cell culture. Eur J Clin Chem Clin. 1997;35:661-667. 25. Woywodt A, Schroeder M, Mengel M, et al. Circulating endothelial cells are a novel marker of cyclosporine-induced endothelial damage. Hypertension. 2003;41:720-723. 26. Fuste B, Mazzara R, Escolar G, Merino A, Ordinas A, DiazRicart M. Granulocyte colony-stimulating factor increases expression of adhesion receptors on endothelial cells through activation of p38 MAPK. Haematologica. 2004;89:578-585.
Biol Blood Marrow Transplant 16:985-993, 2010
27. Frenette PS, Subbarao S, Mazo IB, von Andrian UH, Wagner DD. Endothelial selectins and vascular cell adhesion molecule-1 promote hematopoietic progenitor homing to bone marrow. Proc Natl Acad Sci USA. 1998;95: 14423-14428. 28. Vermeulen M, Le PF, Gagnerault MC, Mary JY, Sainteny F, Lepault F. Role of adhesion molecules in the homing and mobilization of murine hematopoietic stem and progenitor cells. Blood. 1998;92:894-900. 29. Peyvandi F, Siboni SM, Lambertenghi DD, et al. Prospective study on the behaviour of the metalloprotease ADAMTS13 and of von Willebrand factor after bone marrow transplantation. Br J Haematol. 2006;134:187-195. 30. Matsuda Y, Hara J, Osugi Y, et al. Serum levels of soluble adhesion moluecules in stem cell transplantation-related complications. Bone Marrow Transplant. 2001;27:977-982. 31. Mannucci PM, Capoferri C, Canciani MT. Plasma levels of von Willebrand factor regulate ADAMTS-13, its major cleaving protease. Br J Haematol. 2004;126:213-218.
Soluble Markers of Endothelial Damage in HSCT
993
32. Pigott R, Dillon LP, Hemingway IH, Gearing AJ. Soluble forms of E-selectin, ICAM-1 and VCAM-1 are present in the supernatants of cytokine activated cultured endothelial cells. Biochem Biophys Res Commun. 1992;187:584-589. 33. Carlos TM, Harlan JM. Leukocyte-endothelial adhesion molecules. Blood. 1994;84:2068-2101. 34. Elangbam CS, Qualls CW Jr., Dahlgren RR. Cell adhesion molecules-update. Vet Pathol. 1997;34:61-73. 35. Andersen J, Heilmann C, Jacobsen N, Bendtzen K, Muller K. Cytokines and soluble tumour necrosis factor I receptor levels during pretransplant conditioning in allogeneic stem-cell transplantation. Int Immunopharmacol. 2005;5:67-71. 36. Gustot T, Lemmers A, Louis E, et al. Profile of soluble cytokine receptors in Crohn’s disease. Gut. 2005;54:488-495. 37. Van Zee KJ, Kohno T, Fischer E, Rock CS, Moldawer LL, Lowry SF. Tumor necrosis factor soluble receptors circulate during experimental and clinical inflammation and can protect against excessive tumor necrosis factor alpha in vitro and in vivo. Proc Natl Acad Sci USA. 1992;89:4845-4849.
KEY WORDS: Endothelial dysfuntion, Inflammation markers, Hematopoietic stem cell transplantation
INTRODUCTION The endothelium has been recognized as a highly active metabolic and endocrine organ, producing a multitude of different molecules, including vasoactive peptide hormones, growth factors, coagulation factors, and adhesion molecules. It is an active biologic interface between blood and tissues that regulates the
From the 1Hemotherapy-Hemostasis Department, Centre de Diagno`stic Biome`dic, Institut d’Investigacions Biome`diques August Pi i Sunyer, Hospital Clınic, Universitat de Barcelona, Barcelona, Spain; 2Immune Disease Institute, Department of Pathology, Harvard Medical School, Boston, Massachusetts; and 3 Stem Cell Transplantation Unit, Hematology Department, Institut d’Investigacions Biome`diques August Pi i Sunyer, Hospital Clınic, Universitat de Barcelona, Barcelona, Spain. Financial disclosure: See Acknowledgments on page 992. Correspondence and reprint requests: Marta Palomo, Pre-doctoral Hemotherapy-Hemostasis Department, Hospital Clinic, Villarroel 170, 08036 Barcelona, Spain (e-mail: palomo_marta@ hotmail.com). Received October 8, 2009; accepted February 4, 2010 Ó 2010 American Society for Blood and Marrow Transplantation 1083-8791/$36.00 doi:10.1016/j.bbmt.2010.02.008
delicate balance between vasoconstriction and vasodilatation, coagulation and fibrinolysis, proliferation and apoptosis, as well as between transient adhesion and diapedesis of blood-borne leukocytes [1,2]. The endothelium is capable of a number of responses to local challenges. Some stimuli can produce a physiologic endothelial activation to maintain the properties of the endothelium and its environment, including changes in the vasomotor tone, development of a pro-coagulant activity, increase of the leukocyte adhesiveness, and the loss of its barrier function, allowing the passage of certain elements and solutes through the endothelial lining [1,2]. However, other stimuli can originate a true endothelial dysfunction, which produces not only abnormal functional changes but also structural changes (nuclear vacuolization, edema, and cytoplasm fragmentation, denudation, and lost of adhesion to the extracellular matrix). Depending on the extension of the damage, the endothelial dysfunction can be localized (in 1 or several organs) or systemic [1-3]. During hematopoietic stem cell transplantation (HSCT), endothelial cells (ECs) can be activated and damaged by the chemoradiotherapy included in the conditioning regimen, the cytokines produced by the 985
986
M. Palomo et al.
injured tissues, bacterial endotoxins translocated through the damaged gastrointestinal tract, some drugs used during the procedure (as granulocytecolony stimulating factor [G-CSF] or calcineurin inhibitors), the complex process of the engrafment, and allogeneic reactions with the donor-derived immune cells [4], among others. There is evidence suggesting that some complications appearing early after HSCT, as veno-occlusive disease of the liver [5,6], capillary leak syndrome [7], thrombotic microangiopathy (TMA) [8,9], engraftment syndrome (ES) [10], diffuse alveolar hemorrhage [11], idiopathic pneumonia syndrome, and graft-versus-host disease (GVHD) [12,13], can have their origin in a localized or systemic endothelial damage. EC activation and damage can be evaluated through the analysis of several soluble markers, including plasma thrombomodulin, von Willebrand factor (VWF), and soluble forms of adhesion molecules [14], the enumeration of circulating ECs [15], and the evaluation of morphologic and functional changes induced on cultured ECs [16,17]. The present study was designed as a extension of our initial in vitro studies on the proinflammatory and prothrombotic effects of the sera of patients receiving HSCT on human umbilical vein ECs in culture [4]. The effect of the different types of HSCT, and the conditioning regimens applied, on the integrity of the endothelium was evaluated in vivo. The potential diagnostic and/or prognostic utility of the soluble markers evaluated as indicators of endothelial related complications in HSCT was also explored. PATIENTS AND METHODS Patients We included 46 selected patients (mean age 48.5 6 14.1 years, ranging from 23-71 years; 20 female, 26 male), suffering from hematologic diseases, and receiving autologous (n 5 22) or allogeneic HSCT (n 5 24, 7 from related and 17 from unrelated donors) in our institution from January 2006 to January 2007. Diagnosis included acute lymphoblastic leukemia (ALL), (n 5 3), acute myelogenous leukemia (AML) (n 5 10), chronic lymphocytic leukemia (CLL) (n 5 2), chronic myelogenous leukemia (CML) (n 5 1), Hodgkin lymphoma (HL) (n 5 6) and non-Hodgkin lymphoma (NHL) (n510), myelodysplastic syndrome (MDS) (n 5 2), and multiple myeloma (MM) (n 5 12). The 4 following conditioning regimens were applied: (1) BEAM: BCNU 300 mg/m2 at day 27; etoposide 400 mg/m2 at days 26, 25, 24, and 23; cytosine arabinoside 200 mg/m2 at days 26, 25, 24, and 23; and melphalan (Mel) 140 mg/m2 at day 22 (n 5 11); (2) MLF: Mel 100 mg/m2 at days 23 and 22 (n 5 11); (3) Cy/TBI: cyclophosphamide 60 mg/kg at days 27 and 26 and total body irradiation 12 Gy in 4 fractions at days 24,
Biol Blood Marrow Transplant 16:985-993, 2010
23, 22, and 21 (n 5 14); and (4) Flu/MLF: reducedintensity conditioning (RIC) with fludarabine 30 mg/ m2 at days 27 to 23, followed by Mel 70 mg/m2 at days 23 and 22 (n 5 10). Cyclosporine (CsA) from day 27 plus short course of methotrexate or mycophenolate mofetil were administered for GVHD prophylaxis in patients receiving Cy/TBI or Flu/MLF conditioning, respectively. Twenty-seven patients had an HSCT without complications defined by the absence of signs of organ dysfunction (bilirubin \2.0 mg/dL, urea \1.5 mg/dL, lactate dehydrogenase (LDH) \300 U/L, glutamatepyruvate transaminase \100 U/L), absence of GVHD, veno-occlusive disease (VOD), TMA, or sepsis, no need for plasma exchange (plasmapheresis, dialysis, hemofiltration), no administration of fresh frozen plasma or steroids, no respiratory failure, and no need of ventilation or vasopressors. Nineteen patients developed transplant-related complications (TRC): engrafment syndrome (ES) in the autologous group (n 5 10), and GVHD (n 5 7) and VOD (n 5 2), both in the allogeneic HSCT. ES was diagnosed in 5 patients from each treatment group in the autologous setting, with the presence of noninfectious fever (.38 C) plus at least 1 of the following events: pulmonary infiltrates with no signs of infection, cardiac failure or pulmonary embolism; hypoxemia; skin rash involving .25% of body surface area; started 24 hours before or after the first appearance of neutrophils in peripheral blood. Acute GVHD (aGVHD) in patients receiving allogeneic HSCT was graded by classical criteria [18]. Hepatic VOD was diagnosed and classified by the presence of painful hepatomegaly, fluid retention, and weight gain and elevated serum bilirubin concentration .2 mg/dL, according to the Baltimore criteria [19]. In addition, plasma samples from 5 additional patients developing TMA (diagnosis based on clinical features characterized by neurologic disturbance, microangiopathic hemolytic anemia, renal insufficiency, and thrombocytopenia) were collected at the time of the thrombotic complication (41 6 18 days). Control samples from healthy donors (n 5 44, mean age 34 6 16.2 years, ranging from 21 to 54 years; 18 female, 26 male) were included in each determination. Informed consent was obtained from all patients and controls for blood utilization. The study was approved by the ethical committee of the Hospital Clınic and was carried out according to the principles of the Declaration of Helsinki. Sample Collection Blood samples were obtained at different time points of the study: Pre (before starting conditioning regimen), day 0 (immediately before the infusion of the hematopoietic stem cells), and days 7, 14, and 21
Soluble Markers of Endothelial Damage in HSCT
Biol Blood Marrow Transplant 16:985-993, 2010
BEAM 2,5
2,5
*
2
2 1,5
1,5
1
1 0,5
0,5
*
0 Pre
Day 0
Day 7
2
*
1,5
*
*
*
2 1,5 1
1 0,5
*
0 Pre
Day 0
Day 7
2
0,5 0
Day 14 Day 21
1
1
0,5
0,5
*
0 Day 0
Day 7
0
Day 14 Day 21
FLU/MLF
2,5 2
*
1,5
*
2,5
*
*
2 1,5 1
1 0,5
*
0,5 0
0 Pre
VWF
2 1,5
Pre
2,5
2,5
*
1,5
Day 14 Day 21
Cy/TBI 2,5
0
MLF
2,5
987
Day 0
Day 7
Day 14 Day 21
ADAMTS-13
Figure 1. Changes in plasmatic levels of vWF and ADAMTS-13 activity. Graphs represent changes in VWF (C) and ADAMTS-13 activity (B) levels in plasma samples from recipients of autologous HSCT with BEAM and MLF conditioning treatments, and of allogeneic HSCT with Cy/TBI and Flu/MLF treatments, as indicated. Levels were determined at the different time points of the study: before starting conditioning regimen (Pre), immediately before the infusion of the hematopoietic stem cells (day 0), and days 7, 14, and 21 after the HSCT. Data are expressed as ratios with respect to the Pre value and correspond to mean 6 SEM (*P \.05; n 5 6, 5, 5, and 8 for the BEAM, MLF, Cy/TBI, and Flu/MLF, respectively).
after the HSCT. Samples were always obtained through a central line to avoid contamination with endothelial cells. Plasma samples were obtained by centrifugation of blood at 800 g for 15 min and immediately stored at 220 C until the day of the analysis.
considered statistically significant when P \ .05. The SPSS statistical package 17.0.0 (SPSS Inc, Chicago, IL) was used for all analyses.
RESULTS Analysis of Soluble Markers of Endothelial Dysfunction The levels of plasma VWF, sVCAM-1, sICAM-1, and sTNFRI were measured by the enzyme-linked immunoabsorvent assay (ELISA) (American Diagnostica, Grifols SA, Barcelona, Spain; Chemicon, Grupo Taper SA, Madrid, Spain; and HycCult biotechnology b.v, Uden, The Netherlands, Diagnostica Stago SA). ADAMTS-13 activity was analyzed in citrated plasma by using a fluorescence resonance energy transfer assay using a truncated synthetic 73-amino acid VWF peptide as a substrate (FRETS-VWF73 assay) [20,21]. In addition, the presence of IgG anti-ADAMTS-13 (Technoclone GmbH, Viena) was also explored in those patients that developed HSCT-associated TMA. All measurements were performed in duplicate and concentrations determined from the standard curve according to the manufacturer’s instructions. Statistics Results are expressed as mean 6 standard error of the mean (SEM) and as ratios with respect to values at day Pre. Statistical analysis was performed with raw data using the Student’s t-test for paired samples and analysis of variance (ANOVA). Results were
Changes in Plasma Levels of VWF Changes in plasma VWF values from patients without TRC followed different kinetics depending on the type of transplant (see ratios on Figure 1). In the autologous setting under treatment with either BEAM or MLF, we found a progressive increase in values of plasma VWF along the follow-up of the patients until day 14, followed by a decrease from day 14 to day 21. In absolute numbers, levels of VWF in patients treated with BEAM were of 97.7% 6 15.7% at day Pre, and increased to 126.1% 6 16.2% at day 0, 159.4% 6 25.8% at day 7, with a peak of 181.1% 6 22.9% at day 14 (P\.05 versus day Pre), and decreased to 164.0% 6 26.3% at day 21. In patients treated with MLF, VWF levels were of 81.0% 6 11.1% at day Pre, increased to 95.1% 6 11.5% at day 0, 136.9% 6 11.1% at day 7, with a peak of 148.8% 6 12.1% at day 14 (P \ .05 versus day Pre), and decreased to 137.0% 6 12.7% at day 21. In the allogeneic group, the increase in VWF plasma levels after both treatments, Cy/TBI and Flu/ MLF, was statistically significant at all time points studied and persistent until day 21, being more notable with Cy/TBI (see Figure 1). Values of plasma VWF in patients treated with Cy/TBI were of 108.7% 6 10.7%
988
M. Palomo et al.
Biol Blood Marrow Transplant 16:985-993, 2010
at day Pre, and increased to 142.6% 6 9.7% at day 0, 165.4% 6 12.2% at day 7, 174.6% 6 7.5% at day 14, and 189.8% 6 15.2% at day 21 (P \ .05 versus day Pre for all values). In patients treated with Flu/MLF, VWF levels were of 95.2% 6 7.0% at day Pre, increased to 133.0% 6 8.4% at day 0, 150.2% 6 9.7% at day 7, 157.8% 6 12.9% at day 14, and 169.2% 6 12.4% at day 21 (P \ .05 versus day Pre for all values). No significant differences were observed when comparing the initial levels of VWF in the 4 groups of patients (day Pre) with levels in a group of controls (83.5% 6 11.3%, mean 6 SEM, n 5 44). Modifications in the ADAMTS-13 Activity The activity of the metalloprotease ADAMTS-13 in plasma from patients without TRC showed a negative association with plasma VWF levels (see Figure 1). Although basal ADAMTS-13 activity values (day Pre) were in the normal range, there was certain variability between the 4 groups under study (82.1% 6 6.1% in BEAM, 125.2% 6 7.7% in MLF, 88.5% 6 7.5% in Cy/TBI, and 108.8% 6 7.1% in Flu/MLF). A decline in the protease activity after the conditioning regimen appeared to be a constant finding. In absolute values, ADAMTS-13 activity in the patients treated with BEAM was of 82.1% 6 6.1% at day Pre, 80.7% 6 5.4% at day 0, 69.7% 6 6.5% at day 7, 59.4% 6 6.1% at day 14 (P \ .05 versus day Pre), and 66.8% 6 8 .8% at day 21. In patients treated with MLF, ADAMST-13 activity was of 125.2% 6 2.7% at day Pre, 101.8% 6 7.1% at day 0, 114.3% 6 sVCAM-1
2,5
Changes in the presence of the soluble endothelial dysfunction markers VCAM-1 and ICAM-1 in plasma samples from patients without TRC from the autologous and allogeneic groups are shown in Figure 2, expressed as ratios. As a general observation, the kinetics for sVCAM-1 changes along the study were similar when comparing BEAM and MLF, in the
1,5
1
*
0,5
*
1
*
0,5 0
Pre
Day 0
Day 7
Day 14
Day 21
Pr e
BEAM
sVCAM-1
2,5
*
2 1,5
*
*
1 0,5 0
sICAM-1
2
1,5
0
Presence of Soluble Adhesion Molecules VCAM-1 and ICAM-1 in Plasma
2,5
*
2
9.1% at day 7, 95.4% 6 6.7% at day 14 (P \ .05 vs. day Pre), and 92.9% 6 9.9% at day 21. In the allogeneic setting, ADAMTS-13 activity in patients under Cy/TBI treatment was of 88.5% 6 7.5% at day Pre, 88.1% 6 8.4% at day 0, 78.4% 6 4.5% at day 7, 61.7% 6 9.1% at day 14 (P \ .05 versus day Pre), and 81.5% 6 14.3% at day 21. In patients treated with Flu/MLF, ADAMTS-13 activity was of 108.8% 6 7.2% at day Pre, 92.0% 6 7.0% at day 0, 82.7% 6 9.1% at day 7, 69.3% 6 9.9% at day 14 (P \ .05 versus day Pre), and 72.1% 6 7.6% at day 21. Moreover, during the present study, we had the opportunity to evaluate 5 cases of TMA. Evaluation of ADAMTS-13 activity and VWF levels was carried out at the time of diagnoses. In these patients, ADAMTS-13 activity was reduced (42.3% 6 5.9%, mean 6 SEM for the n 5 5), and no inhibitor against ADAMTS-13 was detected in any of the samples. The plasma levels of VWF in the 5 patients were significantly higher than control values (180% 6 20% versus 83.5% 6 11.3%, P \ .05).
Day 0
Day 7
Day 14
Day 21
MLF
sICAM-1
2,5 2 1,5 1 0,5
* Pr e
Day 0
Day 7
Day 14
Day 21
Cy/TBI
0
Pre
Day 0
Day 7
Day 14
Day 21
FLU/MLF
Figure 2. Changes in the levels of sVCAM-1 and sICAM-1 in patients without TRC from the autologous and allogeneic groups. Graphs represent changes in soluble VCAM-1 and ICAM-1 levels in plasma from recipients of autologous HSCTwith BEAM (C) and MLF (B) conditioning treatments, and of allogeneic HSCTwith Cy/TBI (:) and Flu/MLF (D) treatments. Levels were determined at the different time points of the study: before starting conditioning regimen (Pre), immediately before the infusion of the hematopoietic stem cells (day 0), and days 7, 14, and 21 after the HSCT. Data are expressed as ratios with respect to the Pre value and correspond to mean 6 SEM (*P \ 0.05; n 5 6, 5, 5, and 8 for the BEAM, MLF, Cy/TBI, and Flu/MLF, respectively).
autologous group, and also when comparing Cy/TBI and Flu/MLF, in the allogeneic group. No significant changes were detected when analyzing plasma levels of sICAM-1. In the autologous group, we observed a slight decline in sVCAM-1 from day Pre to day 0 (from 283.9 6 32 ng/mL at day Pre to 237.7 6 12.4 ng/mL at day 0 in patients treated with BEAM, and from 484.1 6 35.1 ng/mL at day Pre to 393.8 6 34 ng/mL in the MLF group), remaining similar until day 7 (259.8 6 69.5 ng/mL in BEAM, 349.3 6 49.2 ng/ mL in MLF), increasing significantly to a maximum peak at day 14 (439.8 6 70.3 ng/mL in BEAM, 583.3 6 65.8 ng/mL in MLF, P \ .05 for both with respect to each Pre value), and decreasing slightly from day 14 to day 21 (409.6 6 95.4 ng/mL in BEAM, 514.1.3 6 49.6 ng/mL in MLF). Changes of sICAM-1 levels in the same group of patients were not as notable as those observed for sVCAM-1, and showed a moderate tendency to increase until day 14 (see Figure 2). In absolute numbers for the BEAM group, levels of sICAM-1 were of 189.5 6 24.7, 196.6 6 23.6, 208.1 6 7.8, 284.7 6 58.2, and 241.1 6 24.2 ng/mL, at days Pre, 0, 7, 14, and 21, respectively. For the MLF group, levels were of 194.7 6 25.1, 183.3 6 24.5, 197.6 6 48.6, 261.2 6 65.4, and 245.8 6 18.3 ng/mL, at days Pre, 0, 7, 14, and 21, respectively. In the allogeneic group and in association with both treatments, Cy/TBI and Flu/MLF, the plasma sVCAM-1 levels decreased significantly after the conditioning regimen (from 588.9 6 149.8 ng/mL at day Pre to 289.8 6 89.9 ng/mL at day 0, P \ .05, for Cy/TBI, and from 711.5 6 170.6 ng/mL at day Pre to 642.6 6 129.9 ng/mL at day 0, P \ .05), followed diverging tendencies from day 0 to day 7 (314.4 6 60.3 ng/mL in Cy/TBI and 405.6 6 84.9 ng/mL in Flu/MLF), and increased progressively to day 14 (572.5 6 121.3 and 685.6 6 138.1 ng/mL in Cy/TBI and FLU/MLF, respectively) and significantly to day 21 (750.1 6 125.9 and 1001.1 6 172.2 ng/mL in Cy/ TBI and Flu/MLF, respectively, P \ .05 for both with respect to Pre value). Changes in sICAM-1 levels showed the same tendency as that observed for sVCAM-1, although much lower and never reached statistical significance. No significant differences were observed when comparing the initial levels of both sVCAM-1 and sICAM-1 in the 4 groups of patients (day Pre) with levels in a group of controls (434.7 6 31.5 ng/mL and 247.2 6 21.5 ng/mL, respectively, mean 6 SEM, n 5 44). Soluble TNF Receptor I Plasma Levels Profiles of plasmatic soluble TNF receptor I in patients without TRC are represented in Figure 3.
989
Soluble Markers of Endothelial Damage in HSCT
Biol Blood Marrow Transplant 16:985-993, 2010
sTNFRI
3,5 3
*
2,5 2 1,5 1
*
0,5 0
Pr e
Day 0
Day 7
BEAM
Day 14 MLF
3,5
*
3 2,5
*
2
Day 21
*
*
1,5 1 0,5 0
Pr e
Day 0 Cy/TBI
Day 7
Day 14
Day 21
FLU/MLF
Figure 3. Levels of plasmatic soluble TNF receptor I in patients without TRC. Graphs representing changes in soluble TNFRI levels in plasma from recipients of autologous HSCTwith BEAM (C) and MLF (B) conditioning treatments, and of allogeneic HSCT with Cy/TBI (:) and Flu/ MLF (6) treatments. Levels were determined at the different time points of the study: before starting conditioning regimen (Pre), immediately before the infusion of the hematopoietic stem cells (day 0), and days 7, 14, and 21 after the HSCT. Data are expressed as ratios with respect to the Pre value and correspond to mean 6 SEM (*P \.05; n 5 6, 5, 5, and 8 for the BEAM, MLF, Cy/TBI and Flu/MLF, respectively).
In the autologous group, the analysis of TNFRI showed a progressive increase from day Pre (1160.7 6 358.6 pg/mL in BEAM and 889 6 76.8 pg/mL in MLF group) to day 14, at which values reached a statistical significant peak (2299.3 6 805 pg/mL in BEAM and 1420.6 6 135 pg/mL in MLF) (P \ .05 respect day Pre) and decreased to day 21 (1642.6 6 140.5 pg/mL in BEAM and 1273.3 6 169.9 pg/ml in MLF). In the allogeneic setting, we observed a progressive increase of soluble TNFR I from day Pre (858.8 6 126.4 pg/mL in Cy/TBI and 910.4 6 140.4 pg/mL in Flu/MLF) to day 21, being much more notable in Flu/MLF than in Cy/TBI group (2562.3 6 617.1 pg/mL, and 1606 6 411.8 pg/mL, respectively). No significant differences were observed when comparing the initial levels of both TNFRI in the 4 groups of patients (day Pre) with levels in a group of controls (1007.9 6 211.5, n 5 44).
990
M. Palomo et al.
Biol Blood Marrow Transplant 16:985-993, 2010
a 250
*
*
200 150
*
*
100
* *
50 0
Pre
Day 7
Day 0 ES
Day 14
Day 21
VWF levels than in patients without any complication, although no statistical significance was reached. In fact, changes in plasma levels of VWF along the HSCT treatment versus day Pre reached statistical significance at the same time points (days 7, 14, and 21, P \ .05 in all cases) in both groups of patients, with TRC and without complications (Figure 4). Comparison of values for ADAMTS-13 activity, sVCAM-1, sICAM-1, and sTNFRI among patients with and without complications in each group did not show a clear tendency and never reached statistical significance (data not shown).
No ES
b DISCUSSION
250 200
*
150
*
100
*
*
*
*
50 0 Pre
Day 0
Day 7
GvHD
Day 14
Day 21
No GvHD
c 250 200 150 100 50 0
Pre
Day 0
Day 7 VOD
Day 14
Day 21
No VOD
Figure 4. Kinetics of VWF in patients with TRC versus those without complications. Graphs show VWF levels in recipients of autologous HSCT (a) and allogeneic HSCT (b and c) with () and without (o) transplant related complications. ES 5 engraftment syndrome (n 511); GVHD 5 graft-versus-host disease (n 5 7); VOD 5 venooclusive disease (n 5 2). Data are expressed as absolute values in % (mean 6 SEM) (*P \.05; for days 7, 14, and 21 versus day Pre).
Evaluation of the Prognostic Value of the Markers Analyzed When comparing the kinetics for VWF in patients with TRC versus those without complications, some differences could be observed (see Figure 4). In the auto group, those patients that developed ES showed consistently higher levels of VWF than those that did not develop any complication associated with HSCT. When analyzing allogeneic HSCT recipients who developed GVHD or VOD, we also found higher
We have previously demonstrated, by using an in vitro approach, that the exposure of ECs in culture to sera from patients that have received an HSCT causes endothelial activation and damage, and that these effects are not limited to the allogeneic but also to the autologous setting [4]. In the present study, we have extended our investigation to characterize the proinflammatory effect of the different conditioning regimens commonly applied in both modalities of HSCT, and to find potential diagnostic and/or prognostic markers of the early TRC apparently originated by endothelial damage. We have observed changes in plasma levels of VWF, ADAMTS-13 activity, sVCAM-1, and sTNFRI in association with both types of transplant, although following different kinetics. The degree of these changes was related to the intensity of the conditioning treatment. Although all these markers are good indicators of endothelial activation and damage, from the comparative analysis of the results from patients with and without complications, it can be derived that none of these markers seems to have prognostic value. The evaluation of changes in the release of VWF revealed different endothelial activation kinetics during HSCT probably depending on the conditioning regimen, the use of G-CSF or CsA, the translocation of endotoxins during the period of aplasia, the engraftment and, probably, the allo-reactivity. The release of VWF at the moment of the allogeneic transplant (day 0) is persistently higher than in the autologous setting, a result that is in accordance with the obtained by Salat et al. [14]. This fact could be explained by the higher cytotoxic effect of Cy in combination with TBI [22], or by the well-known endothelial toxicity of Flu [23] and MLF. It is interesting to point out that in all our studies the toxic effect of MLF on the endothelium is similar to that exerted by BEAM, which is a regimen traditionally considered more intense than MLF. In addition, in the allo setting a role of the CsA administered from day 27 could not be excluded. It has been clearly demonstrated that CsA produces endothelial
Biol Blood Marrow Transplant 16:985-993, 2010
damage and induces the release of VWF [24,25]. More difficult to explain is the evolution of VWF levels from day 0 to day 114 in autologous HSCT and to 121 in allogeneic HSCT. In the auto setting, the use of GCSF, agent with a well-known proinflammatory effect [26], can contribute to the endothelial damage from day 17 to day 114. Similarly, in the allo setting CsA can contribute to the damage observed along all the study period. However, these 2 factors could not explain the progressive increase in VWF levels from day 0 to 17 in autologous HSCT and up to day 121 in allogeneic HSCT. We believe that this observation can be the consequence of both the translocation of endotoxins through the damaged gastrointestinal tract during the aplastic phase, followed by the endothelial dysfunction that the complex process of engraftment may induce. From day 0, the stem cell infused adhere to bone marrow endothelium interacting with selectins and several surface adhesion molecules as VCAM [27]. After adherence, the transendothelial migration and intraparenchymal homing to hematopoietic niches take place [28]. Finally, after homing and proliferation, the opposite process take place, which results in the release of neutrophils to peripheral blood, a process that usually occurs between days 9 and 12 in autologous HSCT and between days 14 and 20 in allogeneic HSCT. Finally, a possible role of allo-reactivity in this late period of allo-HSCT could not be excluded. Althought other studies had evaluated several soluble markers after day 128 [14,29,30], unfortunately, the design of our study did not allow the evaluation of the endothelial function after day 121. ADAMTS-13 activity results were in good inverse correlation with those obtained for VWF, as previously reported by Mannucci et al. [31]. It is interesting to note that ADAMTS-13 activity decreased significantly when VWF values increased around 50% over the initial value. In addition, ADAMTS-13 activity in the patients included in the study never decreased under 60%, with the exception of those patients developing TMA in which ADAMTS-13 activity was of 42.3% 6 5.9% (value in the normal range). These observations contribute to discard a pivotal role of the metalloproteinase in the development of TMA associated with the HSCT [29]. The release of soluble forms of adhesion receptors sVCAM-1 and sICAM-1 is a reflection of the activation of the endothelium [32]. Our results showed a significant decrease of sVCAM-1 along the conditioning treatment. This decrease was more notable after the conditioning with Cy/TBI, probably because of the high intensity of this treatment. After this initial decline, these receptors increased following the same kinetics observed for VWF depending on the type of HSCT. Conditioning implies the use of aggressive immunosuppressive and cytotoxic treatments that may interfere with the synthesis of proteins. The induced
Soluble Markers of Endothelial Damage in HSCT
991
expression of VCAM-1 and ICAM-1 is largely dependent on synthesis of new mRNA and protein because there are no storage forms of this endothelial adhesion proteins [33]. In the auto setting, levels of sVCAM raised from day 17, at which G-CSF is administered, to day 114, just after the engraftment time. In the allo setting, the progressive increase in sVCAM-1 levels could be mainly because of the engraftment and CsA toxicity. The fact that values of sVCAM-1 were more relevant than those of sICAM-1 could be attributed to a more restricted expression of the former to the endothelium [34], and it suggests a direct endothelial effect of the conditioning process. Apart from this, because the kinetics of these adhesive molecules correlate with the presence of inflammatory cells in the circulation, the role of leukocytes along the HSCT treatment cannot be disregarded. Leukocytes could contribute with the release of either proinflammatory cytokines or even adhesion molecules to the circulation. sVCAM-1 has been shown to increase in response to proinflammatory cytokines such as tumor necrosis factor-alpha (TNF-a), which we have evaluated through the release of sTNFRI. Similar to the other markers evaluated in this study, the kinetics of sTNFRI changes between the 2 types of HSCT. We were unable to detect reliable changes in TNF-a in the plasma of the patients studied (data not shown), as reported by Andersen et al. [35]. In fact, a variable tendency in the changes of plasma levels of TNFa has been described by other authors in many other pathophysiologic conditions [36]. Presence of soluble receptors for TNF-a (sTNFR) has been reported to follow a good correlation with the cytokine. Soluble TNFRs act as inhibitors of TNF-a activity and represent a natural protective mechanism against the cytotoxic effect of this protein by preventing its binding to the surface receptors. In fact, TNF-a circulates bound to sTNFRI in a cryptic inactive form [37]. The expression of sTNFRI seems to be regulated not only by presence of TNF-a but also by other proinflammatory mechanisms. In the present study, sTNFRI increased from Pre to 0 in the allogeneic setting, but also in the autologous. In addition, levels of plasma sTNFRI in the patients studied followed kinetics similar to those described for VWF suggesting the same pathogenetic events previously mentioned. The main difference was that in the allogeneic setting these levels were higher in association with the Flu/MLF treatment, although this observation is of unclear clinical significance. The comparative analysis of data between patients with and without complications did not show a clear tendency or statistical significance for all the markers analyzed, but did for VWF levels. Levels of plasma VWF in patients with complications were consistently higher in all groups studied, although these differences
992
M. Palomo et al.
never reached statistical significance. Therefore, from the present study we can conclude that the markers evaluated have to be ruled out as prognostic markers of endothelial related complications associated to the HSCT. These results may be ascribed either to the nature of the markers or to the low number of patients included, although they may suggest additional mechanisms for the development of these complications in addition to the endothelial damage. In conclusion, there is endothelial activation and damage in association with both autologous and allogeneic HSCT, as demonstrated by the increased plasma levels of the markers evaluated. The kinetics of changes observed in patients without transplantrelated complications indicate that the engrafment induces maximal endothelial activation in both autologous and allogeneic HSCT. The progressive increase in the markers evaluated in the allogeneic setting may be indicating that the initiation of the allogenicity process is the main cause of endothelial damage in this type of HSCT. Although previous reports strongly suggest that endothelial damage may play a major role in the development of TRC, our present results suggest additional mechanisms. Search for diagnostic and prognostic markers, and for prevention strategies should be the subject of future investigation.
ACKNOWLEDGMENTS Financial discolusre: This work has been partially supported by grants FIS PI060260 and FIS PI080156 (Fondo de Investigaciones de la Seguridad Social), German Jose Carreras Leukaemia Foundation (R 07/ 41v), SGR2005-00952 (Generalitat de Catalunya), SAF2006-08003 (Ministerio de Ciencia y Tecnologıa), and RD06/0009/1003 (Red HERACLES, Instituto de Salud Carlos III). We thank Fulgencio Naval on, Jose M. Rodrıguez, Patricia Molina, and Marc Pino for their technical assistance in this study.
REFERENCES 1. Aird WC. Spatial and temporal dynamics of the endothelium. J Thromb Haemost. 2005;3:1392-1406. 2. Aird WC. The endothelium: a primer. UpToDate 17. 1. Available at: www.uptodate.com, 2009. 3. Mohler ER. Endothelial dysfunction. UpToDate 17. 1. Available at: www.uptodate.com, 2009. 4. Palomo M, Diaz-Ricart M, Carbo C, et al. The release of soluble factors contributing to endothelial activation and damage after hematopoietic stem cell transplantation is not limited to the allogeneic setting and involves several pathogenic mechanisms. Biol Blood Marrow Transplant. 2009;15:537-546. 5. DeLeve LD, Ito Y, Bethea NW, McCuskey MK, Wang X, McCuskey RS. Embolization by sinusoidal lining cells obstructs the microcirculation in rat sinusoidal obstruction syndrome. Am J Physiol Gastrointest Liver Physiol. 2003;284:G1045-G1052.
Biol Blood Marrow Transplant 16:985-993, 2010
6. DeLeve LD, Wang X, Kanel GC, et al. Decreased hepatic nitric oxide production contributes to the development of rat sinusoidal obstruction syndrome. Hepatology. 2003;38:900-908. 7. Nurnberger W, Willers R, Burdach S, Gobel U. Risk factors for capillary leakage syndrome after bone marrow transplantation. Ann Hematol. 1997;74:221-224. 8. George JN, Li X, McMinn JR, Terrell DR, Vesely SK, Selby GB. Thrombotic thrombocytopenic purpura-hemolytic uremic syndrome following allogeneic HPC transplantation: a diagnostic dilemma. Transfusion. 2004;44:294-304. 9. George JN. ADAMTS13, thrombotic thrombocytopenic purpura, and hemolytic uremic syndrome. Curr Hematol Rep. 2005; 4:167-169. 10. Spitzer TR. Engraftment syndrome following hematopoietic stem cell transplantation. Bone Marrow Transplant. 2001;27:893-898. 11. Afessa B, Tefferi A, Litzow MR, Krowka MJ, Wylam ME, Peters SG. Diffuse alveolar hemorrhage in hematopoietic stem cell transplant recipients. Am J Respir Crit Care Med. 2002;166: 641-645. 12. Biedermann BC, Sahner S, Gregor M, et al. Endothelial injury mediated by cytotoxic T lymphocytes and loss of microvessels in chronic graft versus host disease. Lancet. 2002;359:2078-2083. 13. Biedermann BC. Vascular endothelium and graft-versus-host disease. Best Pract Res Clin Haematol. 2008;21:129-138. 14. Salat C, Holler E, Kolb HJ, Pihusch R, Reinhardt B, Hiller E. Endothelial cell markers in bone marrow transplant recipients with and without acute graft-versus-host disease. Bone Marrow Transplant. 1997;19:909-914. 15. Woywodt A, Scheer J, Hambach L, et al. Circulating endothelial cells as a marker of endothelial damage in allogeneic hematopoietic stem cell transplantation. Blood. 2004;103:3603-3605. 16. Serradell M, Diaz-Ricart M, Cases A, et al. Uremic medium causes expression, redistribution and shedding of adhesion molecules in cultured endothelial cells. Haematologica. 2002;87: 1053-1061. 17. Serradell M, Diaz-Ricart M, Zurbano MJ, et al. Uremic media affects hemostatic properties of human endothelial cells in culture and increases the production of von Willebrand factor. Nefrologia. 2002;22:33-41. 18. Sullivan KM, Agura E, Anasetti C, et al. Chronic graft-versushost disease and other late complications of bone marrow transplantation. Semin Hematol. 1991;28:250-259. 19. Jones RJ, Lee KS, Beschorner WE, et al. Venoocclusive disease of the liver following bone marrow transplantation. Transplantation. 1987;44:778-783. 20. Kokame K, Nobe Y, Kokubo Y, Okayama A, Miyata T. FRETSVWF73, a first fluorogenic substrate for ADAMTS13 assay. Br J Haematol. 2005;129:93-100. 21. Kremer Hovinga JA, Mottini M, Lammle B. Measurement of ADAMTS-13 activity in plasma by the FRETS-VWF73 assay: comparison with other assay methods. J Thromb Haemost. 2006;4:1146-1148. 22. Eissner G, Kohlhuber F, Grell M, et al. Critical involvement of transmembrane tumor necrosis factor-alpha in endothelial programmed cell death mediated by ionizing radiation and bacterial endotoxin. Blood. 1995;86:4184-4193. 23. Eissner G, Multhoff G, Gerbitz A, et al. Fludarabine induces apoptosis, activation, and allogenicity in human endothelial and epithelial cells: protective effect of defibrotide. Blood. 2002; 100:334-340. 24. Tiemann C, Prohaska W, Ko¨rfer R, Ko¨rner M, Brinkmann T, Kleesiek K. Effect of cyclosporine A on the release of tissue factor pathway inhibitor from endothelial cells in heart transplant patients and cell culture. Eur J Clin Chem Clin. 1997;35:661-667. 25. Woywodt A, Schroeder M, Mengel M, et al. Circulating endothelial cells are a novel marker of cyclosporine-induced endothelial damage. Hypertension. 2003;41:720-723. 26. Fuste B, Mazzara R, Escolar G, Merino A, Ordinas A, DiazRicart M. Granulocyte colony-stimulating factor increases expression of adhesion receptors on endothelial cells through activation of p38 MAPK. Haematologica. 2004;89:578-585.
Biol Blood Marrow Transplant 16:985-993, 2010
27. Frenette PS, Subbarao S, Mazo IB, von Andrian UH, Wagner DD. Endothelial selectins and vascular cell adhesion molecule-1 promote hematopoietic progenitor homing to bone marrow. Proc Natl Acad Sci USA. 1998;95: 14423-14428. 28. Vermeulen M, Le PF, Gagnerault MC, Mary JY, Sainteny F, Lepault F. Role of adhesion molecules in the homing and mobilization of murine hematopoietic stem and progenitor cells. Blood. 1998;92:894-900. 29. Peyvandi F, Siboni SM, Lambertenghi DD, et al. Prospective study on the behaviour of the metalloprotease ADAMTS13 and of von Willebrand factor after bone marrow transplantation. Br J Haematol. 2006;134:187-195. 30. Matsuda Y, Hara J, Osugi Y, et al. Serum levels of soluble adhesion moluecules in stem cell transplantation-related complications. Bone Marrow Transplant. 2001;27:977-982. 31. Mannucci PM, Capoferri C, Canciani MT. Plasma levels of von Willebrand factor regulate ADAMTS-13, its major cleaving protease. Br J Haematol. 2004;126:213-218.
Soluble Markers of Endothelial Damage in HSCT
993
32. Pigott R, Dillon LP, Hemingway IH, Gearing AJ. Soluble forms of E-selectin, ICAM-1 and VCAM-1 are present in the supernatants of cytokine activated cultured endothelial cells. Biochem Biophys Res Commun. 1992;187:584-589. 33. Carlos TM, Harlan JM. Leukocyte-endothelial adhesion molecules. Blood. 1994;84:2068-2101. 34. Elangbam CS, Qualls CW Jr., Dahlgren RR. Cell adhesion molecules-update. Vet Pathol. 1997;34:61-73. 35. Andersen J, Heilmann C, Jacobsen N, Bendtzen K, Muller K. Cytokines and soluble tumour necrosis factor I receptor levels during pretransplant conditioning in allogeneic stem-cell transplantation. Int Immunopharmacol. 2005;5:67-71. 36. Gustot T, Lemmers A, Louis E, et al. Profile of soluble cytokine receptors in Crohn’s disease. Gut. 2005;54:488-495. 37. Van Zee KJ, Kohno T, Fischer E, Rock CS, Moldawer LL, Lowry SF. Tumor necrosis factor soluble receptors circulate during experimental and clinical inflammation and can protect against excessive tumor necrosis factor alpha in vitro and in vivo. Proc Natl Acad Sci USA. 1992;89:4845-4849.