Risk factors for acute graft-versus-host disease after ... - Nature

Bone Marrow Transplantation (2007) 40, 875–880 & 2007 Nature Publishing Group All rights reserved 0268-3369/07 $30.00

www.nature.com/bmt

ORIGINAL ARTICLE

Risk factors for acute graft-versus-host disease after allogeneic hematopoietic stem cell transplantation: retrospective analysis of 73 patients who received cyclosporin A N Izumi1, T Furukawa2, N Sato3, K Okazuka1, N Tsukada1, T Abe1, T Yano1, T Kurasaki1, M Masuko2, K Toba1, M Takahashi4 and Y Aizawa1 1

Division of Hematology, Niigata University Graduate School of Medical and Dental Sciences, Niigata, Japan; 2Division of Bone Marrow Transplantation, Niigata University Medical and Dental Hospital, Niigata, Japan; 3Nagaoka Red Cross Hospital, Nagaoka, Japan and 4Division of Hematology and Oncology, Graduate School of Health Sciences, Niigata University, Niigata, Japan

Cyclosporin A (CsA) has been used most widely as an immunosuppressive agent for preventing graft-versus-host disease (GVHD). To explore the risk factors including CsA blood levels for grades II–IV acute GVHD, we retrospectively analyzed the data of patients who underwent allogeneic hematopoietic stem cell transplantation in our hospital between March 1989 and July 2001. Seventythree patients (47 males and 26 females) received CsA and short-term methotrexate for GVHD prophylaxis. CsA 1.5 mg/kg was administered as a 3-h infusion twice daily from day 1 until the patient recovered from the toxic gastrointestinal complication. Methotrexate was given at a dose of 15 mg/m2 on day 1 and 10 mg/m2 on days 3, 6 and 11. Grades II–IV acute GVHD occurred in 18 patients (24.7%). Multivariate Cox regression analysis revealed that higher C5 (the whole-blood CsA concentration at 5 h after the start of infusion) before the onset of acute GVHD reduced the onset of grades II–IV acute GVHD with a hazard ratio of 0.994 (95% confidence interval 0.989–0.999) for every increase of 1 ng/ml. Our data indicate that inadequate exposures of CsA can be a vital risk for developing acute GVHD. From our results, we consider that precise monitoring of CsA concentrations and adjustment of CsA dose using the concentration may be effective to prevent the onset of severe acute GVHD. To confirm this finding, further prospective study will be needed. Bone Marrow Transplantation (2007) 40, 875–880; doi:10.1038/sj.bmt.1705834; published online 27 August 2007 Keywords: acute graft-versus-host disease; allogeneic hematopoietic stem cell transplantation; cyclosporin A; risk factor; retrospective clinical study; concentration

Correspondence: Dr T Furukawa, Division of Bone Marrow Transplantation, Niigata University Medical and Dental Hospital, Asahimachi-dori 1-754, Niigata 951-8520, Japan. E-mail: [email protected] Received 19 June 2006; revised 23 July 2007; accepted 24 July 2007; published online 27 August 2007

Introduction Despite advances in the understanding of the disease mechanism, graft-versus-host disease (GVHD) remains a major concern in allogeneic hematopoietic stem cell transplantation (HSCT). The median incidence of clinically important (grades II–IV) acute GVHD is about 40%, and the median incidence of chronic GVHD is between 30 and 45%.1 Acute GVHD is one of the leading causes of early treatment-related mortality, although it is associated with graft-versus-leukemia effect. Thus the management of acute GVHD is a key issue to achieve successful transplantation. The standard pharmacologic regimen for the prevention of acute GVHD is a combination of methotrexate (MTX) and cyclosporin A (CsA). MTX is given at a dose of 15 mg/ m2 on day 1 and 10 mg/m2 on days 3, 6 and 11 (‘short-term’ MTX). CsA is administered intravenously 3 mg/kg/day from day 1 to day 30 followed by oral treatment until day 180.1 Previous studies showed that low trough CsA concentration and reduction to less than 80% of the scheduled CsA dose were associated with the onset of grades II–IV acute GVHD.2–5 However, these studies did not clarify fully whether low CsA concentrations, after the start of administration, are associated with the onset. Recent study reported that the incidence of grades II–IV acute GVHD was lower in the patients treated with twicedaily infusion of CsA than in those treated with the continuous infusion.6 Hence, CsA concentrations after starting administration may play an important role to prevent the disease. In this study, we retrospectively analyzed the data of patients who underwent allogeneic HSCT. In these patients, CsA was administered by 3-h intravenous infusion twice daily for acute GVHD prophylaxis, and whole-blood concentrations were measured consecutively immediately before each infusion (trough level, C0) and at 5 h after the start of each infusion (C5). The objective of this study was to explore the risk factors for grades II–IV acute GVHD including these blood CsA concentrations.

Analysis of risk factors for acute GVHD N Izumi et al

876

Materials and methods Patients We retrospectively analyzed the data of patients who underwent allogeneic HSCT at Niigata University Medical and Dental Hospital between March 1989 and July 2001 for various hematological disorders such as leukemia, aplastic anemia and myelodysplastic syndrome. The data of all patients who received CsA with 3-h infusion twice daily and short-term MTX for GVHD prophylaxis were included in this analysis (n ¼ 73; 47 males and 26 females). The median age of the patients was 29 years (range 16–55). Twenty-three patients received conditioning with total body irradiation (TBI), either 12 Gy (n ¼ 16) or 13.2 Gy (n ¼ 7) given over 3 days in six fractions, followed by etoposide (1 g/m2 2 times) and high-dose cytosine arabinoside (3 g/m2 6 times).7 Twelve patients received 12 Gy TBI and cyclophosphamide (CY, 120 mg/kg). Four patients received 12 Gy TBI, followed by etoposide and CY. Six patients with aplastic anemia received 7.5 Gy total lymphoid irradiation and CY (4 50 mg/kg). All these 45 patients were included as a TBI regimen. Three patients received busulfan (16 mg/kg) and CY (120 mg/kg). Twentyfive patients received busulfan, CY and 3 Gy TBI. These 28 patients were included as a busulfan regimen. Bone marrow cells were reinfused 24–48 h after the last cytotoxic drug dose. No manipulation of the graft was performed. Diagnosis and grading of acute GVHD were based on the criteria of Consensus Conference held in 1994.8 In this analysis, acute GVHD was defined as GVHD that occurred within the first 100 days of allogeneic HSCT by convention. Graft-versus-host disease prophylaxis CsA was administered with the starting dose of 1.5 mg/kg as a 3-h infusion twice daily (every 12 h) from day 1 until the patient recovered from the toxic gastrointestinal complication induced by a conditioning regimen. As a general rule, the dose of CsA was maintained during intravenous administration. Dose increments of intravenous administration of CsA were observed in 17 patients whose trough blood level (C0) was below 100 ng/ml. On the contrary, dose reductions were performed in 24 patients based on an increase in serum creatinine (20 patients) and high trough concentration of above 300 ng/ml (4 patients). However, these dose modifications were not systematically performed according to the CsA blood concentration. Route of CsA administration was converted to oral intake at a ratio of 1:2 or 1:3 when patients were able to intake oral food. The median day of conversion from intravenous to oral CsA was day 48 (range, 22–272). MTX was given at a dose of 15 mg/m2 on day 1 and 10 mg/ m2 on days 3, 6 and 11. MTX was not administered on day 11 if severe mucositis was observed. Patients did not receive other immunosuppressive therapies including steroids, tacrolimus or antithymocyte globulin unless grades II–IV acute GVHD was observed. Drug concentration measurements We measured whole-blood CsA concentrations immediately before each infusion (trough, C0) and at 5 h after the Bone Marrow Transplantation

start of each infusion (C5). And after conversion from intravenous administration to oral intake, C5 was not available and only C0 was measured. CsA concentrations were measured by radioimmunoassay based on monoclonal antibody (RIA)9 until March 1997 (n ¼ 47), thereafter by fluorescence polarization immunoassay based on monoclonal antibody (M-FPIA)10 using TDX system according to the manufacturer’s instructions (Abbott Japan Co Ltd, Tokyo, Japan) until March 1999 (n ¼ 12). From April 1999, CsA concentrations were measured by M-FPIA using AxSYM analyzer according to manufacturer’s instructions (Abbott Japan Co Ltd; n ¼ 14). In this analysis, concentrations measured by RIA and M-FPIA using TDX analyzer were converted into those measured by M-FPIA using AxSYM analyzer according to the correlation equations between each value of different methods. The correlation for all samples analyzed in our institution by both RIA and M-FPIA using TDX was described by the linear regression equation, M-FPIA (TDX) ¼ 1.0515 RIA þ 21.664 (r ¼ 0.988, n ¼ 137). The correlation for samples analyzed by both M-FPIA using TDX and AxSYM in our institution was also described by the linear regression equation, M-FPIA (AxSYM) ¼ 0.89 M-FPIA (TDX)— 12.70 (r ¼ 0.993, n ¼ 100).

Statistical analysis First, we compared the baseline characteristics of the patients with grades II–IV acute GVHD (GVHD group) with those of the patients with no or grade I disease (control group) by w2-test. The means of C0 and C5 for each 10 days after transplantation (days 1–10, 11–20, 21–30, 31–40, 41–50 and 51–60) were calculated for each patient. These periods were selected because after day 60, only two patients developed grades II–IV acute GVHD. C5 was not available for patients who converted the route of CsA administration from intravenous infusion into oral intake. After the patient was diagnosed as grade II–IV acute GVHD, C0 and C5 were eliminated from the analysis. Then we explored the differences of C0 and C5 between the two groups by repeated measurements analysis of variance. In this analysis, group (GVHD or control group), time, patient and group-by-time interaction were included as the factors. To estimate risk factors for grades II–IV acute GVHD, multivariate Cox regression analysis was conducted. The baseline characteristics, which appeared predictive based on the results of w2-test, were included as the factors. Means of C0 and C5 were also included in the model as the time-dependent covariates. This method of analysis was described elsewhere.2 In brief, for every patient who was diagnosed as grades II–IV acute GVHD, a comparison group or risk set was formed that consisted of all other patients without GVHD at the time of the diagnosis. For instance, if a particular patient developed grades II–IV acute GVHD on day 35 after transplantation, this patient’s associated risk set would consist of all patients who did not have this grade of acute GVHD on day 35 after their own transplantation. Then, for each case, the values of the covariates would be compared with those of all patients in the risk set. For simplicity, diagnoses were grouped


877

Results Actual daily dose of cyclosporin A CsA was administered with the starting dose of 1.5 mg/kg as a 3-h infusion twice daily (every 12 h) from day 1 until the patient was converted to oral administration. As a general rule, the dose of CsA was maintained during intravenous administration. The dose modification of CsA was performed to maintain the adequate trough blood level and to get rid of nephrotoxicity in some patients as described above. However, the median of actual daily dose of intravenously administered CsA during first 40 days after transplantation was maintained at 3 mg/kg in our patients (Figure 1).

Daily dose of CsA (mg /kg)

6

5

4

3

2

1 day 0-10

day 11-20 day 21-30

day 31-40

Figure 1 Actual daily dose of cyclosporin A (CsA). The box and whisker plot shows 10, 25, 50, 75 and 90 percentile values. Outliers are indicated by dots. Incidence of Grade II to IV acute GVHD

according to each 10-day period in which they occurred. Accordingly, if the time-dependent covariate was the mean CsA concentration during the 10-day period (days 21–30) preceding day 35, the mean concentration for every patient in the risk set would be evaluated for that 10-day period (days 21–30) and the values then used to compare the index patient with other patients in the risk set. We assumed the effects of C0 and C5 in the following two ways: (1) the onset of acute GVHD was mainly affected by C0 or C5 just before the event; (2) the onset of acute GVHD was averagely affected by C0 or C5 over the treatment period. Competing models were compared by the value of Akaike’s Information Criterion (AIC). We did not use a stepwise procedure because we considered that the sample size was too small to obtain maximum likelihood estimates appropriately. All analyses were performed by using SASs version 8.2 (Cary, NC, USA). Institutional Review Board Approval was obtained at Niigata University Graduate School of Medical and Dental Sciences to conduct this review of existing data and all individual patient data have been de-identified for the statistical analysis.

1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 10

20

30

40

50 60 70 80 Days after transplant

90

100

Figure 2 Cumulative incidence of grades II–IV acute graft-versus-host disease (GVHD). Grades II–IV acute GVHD developed in 18 patients (24.7%).

Risk factors for grades II–IV acute graft-versus-host disease Grades II–IV acute GVHD developed in 18 patients (24.7%; Figure 2). Median time for the onset of GVHD in patients who eventually developed grades II–IV acute GVHD was 34 days (range 10–71 days). Table 1 summarizes the baseline characteristics of the GVHD and control groups. In this table, we defined highrisk diseases as acute leukemia or lymphoma without complete remission, and chronic myelogenous leukemia in accelerated phase according to the previous report.3 Proportion of the high-risk diseases was higher in the GVHD group than in the control group (P ¼ 0.027). Other baseline characteristics were not significantly different between the two groups. Both groups received MTX with a similar dosage (41.0 (s.d. 7.3) mg/m2 in the GVHD group and 41.6 (s.d. 6.6) mg/m2 in the control group).

all retrospectively available data of the patients who received short-term MTX for GVHD prophylaxis. Although the means of C0 values in patients without GVHD were higher than that in the patients with GVHD until the period of days 31–40 (Figure 3), C0 values were not significantly different throughout the treatment period between the two groups (P ¼ 0.708, repeated measurements analysis of variance). While, C5 values were consistently lower in the GVHD group than in the control group (P ¼ 0.023). The differences of the mean C5 levels between the two groups were 47, 48, 68 and 55 ng/ml respectively during each period (Figure 4). We could not explore the differences of C5 between the two groups after the period of days 30–41 in this analysis, because no patient who was administered CsA intravenously developed grades II–IV acute GVHD after day 50. The mean C5 before the onset of grades II–IV acute GVHD was 303 (s.d. 120) ng/ml.

Effect of cyclosporin A blood level on the incidence of grades II–IV acute graft-versus-host disease Figures 3 and 4 summarize the means (71 s.d.) of C0 and C5 during the treatment period, respectively. We used the

Multivariate analysis Three multivariate Cox models were developed and compared by means of Akaike’s Information Criterion Bone Marrow Transplantation


878 Table 1 Baseline characteristics of 73 patients who underwent allogeneic HSCT Patients with grades II–IV acute GVHD (n ¼ 18)

25 (16–55)

31 (18–49)

0.347

43 (78) 12 (22)

13 (72) 5 (28)

0.604

Patient sex, n (%) Male Female

33 (60) 22 (40)

Diagnosis, n (%) AA ALL AML CML MDS NHL

4 16 19 13 3 0

High risk diseases,a n (%) Without With

(7) (29) (35) (24) (5) (0)

14 (78) 4 (22)

2 2 5 6 2 1

300 250 200 150 100 50 0

1-10

41-50 11-20 21-30 31-40 Days after transpantation

51-60

0.172

Figure 3 The means71 s.d. of trough cyclosporin A blood concentra-

(11) (11) (28) (33) (11) (6)

0.270

47 (85) 8 (15)

11 (61) 7 (39)

0.027

Donor, n (%) Sibling Related other than sibling Unrelated

42 (76) 3 (5) 10 (18)

9 (50) 2 (11) 7 (39)

0.107

Donor sex, n (%) Male Female

27 (49) 28 (51)

10 (56) 8 (44)

0.634

6 (33)

0.848

tions (C0) during the treatment interval in patients who did or did not have grades II–IV acute graft-versus-host disease (GVHD). Closed circles are the means in patients with grades II–IV acute GVHD. Open square with a shadow are the means in patients with no or grade I acute GVHD. The C0 values after the patient was diagnosed as grades II–IV acute GVHD were not included in this analysis.

600

Donor-patient sex match, n (%) Male patient with female 17 (31) donor Other combinations 38 (69)

500 400 300 200 100 0 1-10

11-20

21-30

31-40

Days after transplantation

12 (67)

Figure 4 The means71 s.d. of cyclosporin A blood concentrations at 5 h

HLA identical, n (%) Identical GVH antigen mismatch

52 (95) 3 (5)

16 (89) 2 (11)

0.410

Preparative conditioning Busulfan TBI

22 (40) 33 (60)

6 (33) 12 (67)

0.614

Abbreviations: AA ¼ aplastic anemia; ALL ¼ acute lymphoblastic leukemia; AML ¼ acute myeloid leukemia; CML ¼ chronic myelogenous leukemia; GVH ¼ graft-versus-host; GVHD ¼ graft-versus-host disease; HLA ¼ human leukocyte antigen; HSCT ¼ hematopoietic stem cell transplantation; MDS ¼ myelodysplastic syndrome; NHL ¼ non-Hodgkin’s lymphoma; TBI ¼ total body irradiation. a Acute leukemia or lymphoma without complete remission ¼ and chronic myelogenous leukemia in the accelerated phase. *w2-test.

(AIC). According to the AIC, the model based on the assumption that C5 just before the event mainly affects the onset of acute GVHD was the best (Table 2, model 1; AIC ¼ 140.3). The result demonstrated that higher C5 reduced the onset of grades II–IV acute GVHD with a hazard ratio of 0.994 (95% confidence interval 0.989–0.999) for every increase of 1 ng/ml. The model based on the Bone Marrow Transplantation

350 Cyclosporin A (ng /ml)

Median (range) age, years Patient age, n (%) o40 X40

Pvalue*

Patients with no or grade I acute GVHD (n ¼ 55)

Cyclosporin A (ng/ml)

Characteristic

400

after the start of infusion (C5) during the treatment interval in patients who did or did not have grades II–IV acute graft-versus-host disease (GVHD). Closed circles are the means in patients with grades II–IV acute GVHD. Open square with a shadow are the means in patients with no or grade I acute GVHD. The C5 values after the patient was diagnosed as grades II–IV acute GVHD were not included in this analysis.

assumption that the onset of acute GVHD was averagely affected by C5 revealed that both C5 and the high-risk disease were the significant risks, although this model was far less fitted for predicting acute GVHD (Table 2, model 3; AIC ¼ 370.4) than the model 1. The model that included the C0 and the high-risk disease could not demonstrate a statistical significant impact of both factors on the development of grades II–IV acute GVHD (Table 2, model 2). We also included previously reported risk factors such as patient age or donor sex in the model, but these factors were not predictive of the onset.

Discussion This retrospective study demonstrated that low C5 was a significant risk factor for grades II–IV acute GVHD with a


879 Table 2 Model 1 2 3

Multivariate analysis of risk factors for grades II–IV acute GVHD Factor

Hazard ratio

95% confidence interval

P-value

AIC

a

0.994 1.824 0.998 2.587 0.996 1.942

0.989–0.999 0.665–5.007 0.993–1.003 0.966–6.929 0.992–0.999 1.060–3.559

0.019 0.243 0.398 0.059 0.008 0.032

140.3

C5 just before the onset of acute GVHD High risk diseasesc C0b just before the onset of acute GVHD High-risk diseasesc Average C5a before the onset of acute GVHD High-risk diseasesc

148.5 370.4

Abbreviations: GVHD ¼ graft-versus-host disease; AIC ¼ Akaike’s Information Criterion. a Blood concentration of cyclosporin A at 5 h after the start of infusion. b Blood concentration of cyclosporin A before the start of infusion (trough concentration). c Acute myelogenous leukemia without complete response and chronic myelogenous leukemia in accelerated phase.

relative risk of 0.994 for every increase of 1 ng/ml (i.e., 0.994x for every increase of x ng/ml). In our study, the means of C5 during the period of intravenous CsA administration were largely different between the GVHD group and the control group. The means of C5 of the control group were consistently higher than that of the acute GVHD group in the value of 50–60 ng/ml until day 40. And the mean C5 before the onset of grades II–IV acute GVHD was 303 ng/ml. The relative risk will be substantial when such a large decrease in the CsA concentration at the post infusion sampling point is observed. Previous studies showed that low trough CsA concentrations were associated with grades II–IV acute GVHD in HSCT. Yee et al.2 reported that low serum trough concentrations can be a cause of the disease with a relative risk of 0.7 for every increase of 100 ng/ml. Martin et al.5 used the data of children who were transplanted from a matched-sibling or an unrelated donor and reported that trough blood concentration o85 ng/ml was a risk factor. We used the data of all patients who received CsA and short-term MTX for GVHD prophylaxis, and the means of C0 in patients without GVHD revealed to be higher than that in the patients with acute GVHD until day 40 (as shown in Figure 3). However, we could not detect a significant association between the onset of acute GVHD and the means of C0 in our patients. We also included trough CsA concentrations and previously reported risk factors in the multivariate regression model, but we could not find that either of these factors was associated with the onset of grades II–IV acute GVHD. Importantly, our analysis did not imply that trough CsA concentration had no effect on the onset of acute GVHD, but rather, this study does not have enough statistical power to detect this factor’s effect on GVHD onset. We speculated that the dispersion of the means of C0 and a small sample size might prevent reaching statistical significance in our retrospective analysis. The immunosuppressive effects of CsA (the inhibition of calcineurine in lymphocytes) have been demonstrated to be correlated with blood concentration, and the peak of action was at approximately 2 h after Neoral (an improved microemulsified formulation) administration, when the peak absorption occurred in solid organ transplantation.11 And considerable refinement for Neoral dosage adjustment, using the area under the concentration (AUC)–time curve in the first 4 h after administration (AUC0–4) value,

and more recently the 2-h post dose (C2) value has been developed.12,13 A number of clinical trials have employed C2 monitoring in solid organ transplantation and demonstrated superior outcomes in terms of rejection rates and acute toxicities compared to trough-level (C0) monitoring.13,14 In the HSCT setting, intravenous administration of CsA is usually used for acute GVHD prophylaxis. It remains to be determined whether the area under the concentration or peak level, rather than the trough level provides a more accurate measure of immunosuppression in such intravenous administration.15 Our study identified the influence of C5 on grades II–IV acute GVHD quantitatively and suggested the importance of monitoring CsA concentrations after the start of infusion. Although we measured C5 (which was not the peak concentration of 3-h infusion) for technical reasons in this analysis, whether the 5-h after the start of infusion sampling point (C5) is suitable for the measurement of concentration to adjust CsA dosage remains to be clarified. We need to evaluate CsA concentration at various time points to identify the single sample that provides the most accurate surrogate marker for immunosuppression and the CsA dose adjustment. In clinical practice, many institutes monitor the trough concentrations, but fewer institutes monitor the concentrations after the start of infusion. However, concentrations after administration differ greatly because the dosing schedules vary among the institutions.16 For example, some institutes administer 3 mg/kg/day CsA intravenously over 24 h (continuous infusion), and the other institutes administer the same dose over 8 or 12 h. Hence, monitoring only trough concentrations may not be sufficient to optimize CsA therapy. Recent study reported that twice-daily infusion of CsA significantly reduces the incidence of grades II–IV acute GVHD when compared with the continuous infusion.6 In their study, the doses of CsA were adjusted to maintain the trough concentrations to the target levels (150–300 ng/ml in the twice-daily group and 250–400 ng/ml in the continuous infusion group). As a result, trough concentrations were lower in the twice-daily group, and this fact indicates a limitation of trough concentration monitoring. Although authors of the study did not draw any definite conclusion because the actual daily dose was lower in the continuous infusion group, their study suggests the possibility that achieving peak concentrations may be effective to prevent the onset of the disease. From these results, we consider Bone Marrow Transplantation


880

that monitoring CsA concentrations at the time points after the start of infusion may be necessary to achieve the immunosuppressive activity adequately. Our study used the data of patients who received 1.5 mg/ kg/day CsA over 3 h twice a day, and the results of this study should be applied to the same setting only. To assess the effects of CsA concentrations in the wider concentration range, further analysis will be needed including the data of patients administered 3 mg/kg/day CsA over 8, 12 and 24 h. Despite these limitations, this study shows that low CsA concentration at 5 h after the start of infusion is a risk factor for moderate-to-severe acute GVHD. To confirm whether monitoring the CsA concentrations after the start of infusion and dose adjustment using the concentration effectively prevents the onset of acute GVHD, a prospective observational study is now under study.

8

Acknowledgements

9

We thank Miyako Baba and Yoko Tanahashi, Clinical Laboratory Division, Niigata University Medical and Dental Hospital for their excellent technical assistance and helpful discussion.

References 1 Devergie A. Graft vs. host disease. In: Apperley J, Carreras E, Gluckman E, Gratwohl A, Masszi T (eds). Haemopoietic Stem Cell Transplantation. European School of Hematology: Paris, 2004, pp 163–176. 2 Yee GC, Self SG, McGuire TR, Carlin J, Sanders JE, Deeg HJ. Serum cyclosporine concentration and risk of acute graftversus-host disease after allogeneic marrow transplantation. N Engl J Med 1988; 319: 65–70. 3 Nash RA, Pepe MS, Storb R, Longton G, Pettinger M, Anasetti C et al. Acute graft-versus-host disease: analysis of risk factors after allogeneic marrow transplantation and prophylaxis with cyclosporine and methotrexate. Blood 1992; 80: 1838–1845. 4 Martin P, Bleyzac N, Souillet G, Galambrun C, Bertrand Y, Maire PH et al. Relationship between CsA trough blood concentration and severity of acute graft-versus-host disease after paediatric stem cell transplantation from matched-sibling

Bone Marrow Transplantation

5

6

7

10

11 12 13

14

15

16

or unrelated donors. Bone Marrow Transplant 2003; 32: 777–784. Martin P, Bleyzac N, Souillet G, Galambrun C, Bertrand Y, Maire PH et al. Clinical and pharmacological risk factors for acute graft-versus-host disease after paediatric bone marrow transplantation from matched-sibling or unrelated donors. Bone Marrow Transplant 2003; 32: 881–887. Ogawa N, Kanda Y, Matsubara M, Asano Y, Nakagawa M, Sakata-Yanagimoto M et al. Increased incidence of acute graft-versus-host disease with the continuous infusion of cyclosporine A compared to twice-daily infusion. Bone Marrow Transplant 2004; 33: 549–552. Sato N, Furukawa T, Kuroha T, Hashimoto S, Masuko M, Takahashi H et al. High-dose cytosine arabinoside and etoposide with total body irradiation as a preparatory regimen for allogeneic hematopoietic stem-cell transplantation in patients with acute lymphoblastic leukemia. Bone Marrow Transplant 2004; 34: 299–303. Przepiorka D, Weisdorf D, Martin P, Klingemann HG, Beatty P, Hows J et al. 1994 Consensus Conference on acute GVHD grading. Bone Marrow Transplant 1995; 15: 825–828. Schran HF, Rosano TG, Hassell AE, Pell MA. Determination of cyclosporine concentrations with monoclonal antibodies. Clin Chem 1987; 33: 2225–2229. Lindholm A, Napoli K, Rutzky L, Kahan BD. Specific monoclonal radioimmunoassay and fluorescence polarization immunoassay for trough concentration and area-underthe-curve monitoring of cyclosporine in renal transplantation. Ther Drug Monit 1992; 14: 292–300. Halloran P. Calcineurin inhibition-relationship to cyclosporine blood concentration. Focus Med 1998; 13: 15–18. Kahan BD. Therapeutic drug monitoring of cyclosporine: 20 years of progress. Transplant Proc 2004; 36: 378S–391S. Nashan B, Cole E, Levy G, Thervet E. Clinical validation studies of Neoral C(2) monitoring: a review. Transplantation 2002; 73: S3–S11. Levy G, Thervet E, Lake J, Uchida K, Consensus on Neoral C(2): Expert Review in Transplantation (CONCERT) Group. Patient management by Neoral C(2) monitoring: an international consensus statement. Transplantation 2002; 73: S12–S18. Duncan N, Craddock C. Optimizing the use of cyclosporin in allogeneic stem cell transplantation. Bone Marrow Transplant 2006; 38: 169–174. Ruutu T, Niederwieser D, Gratwohl A, Apperley JF. A survey of the prophylaxis and treatment of acute GVHD in Europe: a report of the European Group for Blood and Marrow, Transplantation (EBMT). Chronic leukaemia working party of the EBMT. Bone Marrow Transplant 1997; 19: 759–764.