doi:10.1182/blood-2009-02-204156Prepublished online April 29, 2009;2009 114: 709-718

BacigalupoGuido Gini, Nicola Mordini, Adriana Balduzzi, Pietro Leoni, Armando Gabrielli and Andrea

Attilio Olivieri, Franco Locatelli, Marco Zecca, Adele Sanna, Michele Cimminiello, Roberto Raimondi, featuresImatinib for refractory chronic graft-versus-host disease with fibrotic

http://bloodjournal.hematologylibrary.org/content/114/3/709.full.htmlUpdated information and services can be found at:

(3143 articles)Clinical Trials and Observations � (1131 articles)Free Research Articles �

(1625 articles)Transplantation �Articles on similar topics can be found in the following Blood collections

http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://bloodjournal.hematologylibrary.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

Copyright 2011 by The American Society of Hematology; all rights reserved.Washington DC 20036.by the American Society of Hematology, 2021 L St, NW, Suite 900, Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly

For personal use only. by guest on May 10, 2011. bloodjournal.hematologylibrary.orgFrom

TRANSPLANTATION

Imatinib for refractory chronic graft-versus-host disease with fibrotic features

Attilio Olivieri,1 *Franco Locatelli,2 Marco Zecca,2 Adele Sanna,3 Michele Cimminiello,1 Roberto Raimondi,4 Guido Gini,5

Nicola Mordini,6 Adriana Balduzzi,7 Pietro Leoni,5 Armando Gabrielli,8 and *Andrea Bacigalupo9

1Department of Hematology, San Carlo Hospital, Potenza; 2Pediatric Hematology/Oncology, Fondazione Instituto Di Ricerca e Cura a Caratiere Scientifica,

Policlinico San Matteo, University of Pavia, Pavia; 3Unit of Bone Marrow Transplantation, Ospedale Regionale per le Microcitemie, Cagliari; 4Department of

Hematology, St Bortolo Hospital, Vicenza; 5Department of Hematology, Universita Politecnica delle Marche, Ancona; 6Department of Hematology, Santa Croce e

Carle Hospital, Cuneo; 7Department of Pediatric Hematology, San Gerardo Hospital, Monza; 8Department of Internal Medicine, Universita Politecnica delle

Marche, Ancona; and 9Hematology Unit and Stem Cell Transplant Unit San Martino Hospital, Genova, Italy

We previously reported that patients with

fibrotic, chronic graft-versus-host dis-

ease (cGVHD) have antibodies activating

the platelet-derived growth factor recep-

tor pathway. Because this pathway can

be inhibited by imatinib, we performed a

pilot study including 19 patients with re-

fractory cGVHD, given imatinib at a start-

ing dose of 100 mg per day. All patients

had active cGVHD with measurable in-

volvement of skin or other districts and

had previously failed at least 2 treatment

lines. Patient median age was 29 years

(range, 10-62 years), and median dura-

tion of cGvHD was 37 months (range,

4-107 months). The organs involved were

skin (n � 17), lung (n � 11), and bowel

(n � 5); 15 patients had sicca syndrome.

Imatinib-related, grade 3 to 4 toxicity in-

cluded fluid retention, infections, and ane-

mia. Imatinib was discontinued in 8 pa-

tients: in 3 because of toxicity and in

5 because of lack of response (n � 3) or

relapse of malignancy (n � 2). Overall re-

sponse rate at 6 months was 79%, with

7 complete remissions (CRs) and 8 partial

remissions (PRs). With a median follow-up

of 17 months, 16 patients are alive, 14 still in

CR or PR. The 18-month probability of over-

all survival is 84%. This study suggests that

imatinib is a promising treatment for pa-

tients with refractory fibrotic cGVHD. (Blood.

2009;114:709-718)

Introduction

Over the past decade, significant changes have been made in

hematopoietic stem cell transplantation, including wider use of

this procedure in older patients, more transplantations from

unrelated and HLA-disparate related donors, and widespread

use of peripheral blood as stem cell source. All these factors

have contributed to increase the prevalence of chronic graft-

versus-host disease (cGVHD).1

The exact pathogenesis of cGVHD is still incompletely de-

fined,2 and the role of alloreactivity versus autoreactivity remains

an area of debate. Alloreactivity of donor lymphocytes toward

unshared recipient minor histocompatibility antigens is advocated

to interpret cGVHD as a late phase of acute GVHD; the importance

of autoreactivity, however, is suggested by clinical manifestations

of cGVHD that mimic those of autoimmune diseases. In particular,

the typical sicca syndrome in patients with cGVHD resembles that

of patients with Sjogren syndrome, enteropathy can mimic Crohn

disease, and skin manifestations recall systemic sclerosis (SSc).3

Treatment of cGVHD is based on immune-suppressive agents,

usually steroids, associated with a calcineurin inhibitor,4,5 but other

therapies have been tested, including mofetil mycophenolate,

extracorporeal photopheresis (ECP), and anti-CD20 monoclonal

antibody (rituximab [RTX]).6-9 Response is often unsatisfactory,

and prolonged treatment with these agents contributes to the state

of profound immune deficiency that characterizes patients with

cGVHD and is responsible for the increased risk of developing

severe, sometimes even life-threatening or fatal, infectious compli-

cations. Indeed, cGVHD has been identified as the leading cause of

late nonrelapse mortality in survivors of hematopoietic stem cell

transplantation.10

We have recently shown that patients with cGVHD showing

fibrotic/sclerotic manifestations have agonistic antibodies activat-

ing the platelet-derived growth factor receptor (PDGF) receptor.11

The presence and biologic activity of these autoantibodies argue for

a common pathogenic trait in both cGVDH and SSc,11,12 in that

up-regulation of the PDGF receptor intracellular pathway leads to

increased reactive oxygen species, with consequent exaggerated

collagen synthesis, which, in turn, contributes to the pathologic

lesions observed in both cGVHD and SSc.

Much data strongly suggest that, among the profibrotic cyto-

kines, besides PDGF,13 also transforming growth factor � (TGF-�)

can play a relevant role in the pathogenesis of SSc. Indeed, both

PDGF and TGF-� pathways seem to be up-regulated in the skin of

patients with SSc, and, in a murine model of cGVHD, anti–TGF-�

antibodies prevented the development of skin fibrosis.14 Moreover,

blockade of either TGF-� or PDGF signaling has been shown to

reduce the development of fibrosis in various experimental mod-

els.15,16 Recent in vitro data showed that imatinib strongly inhibits

both PDGF and TGF-� intracellular signaling, which is responsible

for the expression of extracellular matrix genes.17 In this experimen-

tal model, the highest concentration of imatinib was 1.0 �g/mL, a

Submitted February 10, 2009; accepted April 13, 2009. Prepublished online as

Blood First Edition paper, April 29, 2009; DOI 10.1182/blood-2009-02-204156.

*F.L. and A.B. contributed equally to this study.

An Inside Blood analysis of this article appears at the front of this issue.

The online version of this article contains a data supplement.

Presented in part (preliminary data in 15 patients) in poster form at 2008 Annual

Meeting of the European Group for Bone Marrow Transplantation (EBMT),

Florence, Italy, March 2008.

The publication costs of this article were defrayed in part by page charge

payment. Therefore, and solely to indicate this fact, this article is hereby

marked ‘‘advertisement’’ in accordance with 18 USC section 1734.

© 2009 by The American Society of Hematology

709BLOOD, 16 JULY 2009 � VOLUME 114, NUMBER 3

For personal use only. by guest on May 10, 2011. bloodjournal.hematologylibrary.orgFrom

value within the mean trough concentration observed after adminis-

tration to patients of drug doses between 200 and 400 mg per day.

Currently, imatinib is widely used in patients with Philadelphia-

positive chronic myelogenous leukemia (CML) and acute lympho-

blastic leukemia. In view of the specific capacity of imatinib to

inhibit both TGF-� and PDGF signaling pathways, the long-

standing clinical experience with this drug, and its good tolerability

in patients with CML and acute lymphoblastic leukemia,18 we

speculated that imatinib could be a possible option for treating

patients with cGVHD, refractory to conventional treatment, when

sclerotic/fibrotic clinical features are present.

Methods

Patient characteristics

Nineteen patients, affected by refractory cGVHD with fibrotic/sclerotic

clinical features, were enrolled in the study; median age was 29 years

(range, 10-62 years). The main characteristics of these patients are detailed

in Table 1. The major involved organs were skin (n � 17), lung (n � 11),

and bowel (n � 5); sicca syndrome was diagnosed in 15 patients. On the

whole, 11 had generalized cutaneous scleroderma features, and 13 patients

had visceral involvement (excluding sicca syndrome). The diagnosis of

sicca syndrome was based on the clinical criteria elaborated by the

American-European Consensus Group, without including serologic mark-

ers, which, by contrast, are considered in patients with the classical Sjogren

syndrome19; in all 15 patients with sicca syndrome a pathologic Schirmer

test was present (see supplemental file for details, available on the Blood

website; see the Supplemental Materials link at the top of the online article).

Among the 11 patients with generalized skin fibrosis, 5 also had visceral

involvement. In the 8 patients who did not show generalized skin sclerosis,

6 had localized skin fibrotic involvement associated with visceral involve-

ment, 1 patient had only diffuse fasciitis (histologically proven), and the

remaining patient had severe lung fibrosis in the absence of any skin

involvement.

The lines of therapy that each patient had received before being treated with

imatinib are reported in Table 1, and details concerning the organ involvement

and combinations for each single patient are reported in Table 2.

End points

The primary end point of this study was to assess the safety [in terms of

incidence of severe adverse events (SAEs) and hematologic tolerance] of

escalating doses of imatinib in patients with sclerotic/fibrotic refractory

cGVHD. Secondary end points were response rate and overall survival

(OS) at 3 and 6 months after beginning imatinib therapy. The study was

designed as a phase 1 to 2 trial, initially planned to enroll 15 patients, and

approved by the Ethic Committee of the Coordinating Center of Potenza

(Eudract no. 2007-001508-19). Imatinib mesylate (Gleevec; 100-mg tab-

lets; Novartis, Switzerland) was administered off label.

In all patients, written informed consent to treatment was obtained after

having exhaustively discussed the potential benefits and risks of imatinib

therapy, in accordance with the Declaration of Helsinki. Written consent to

collect, handle, and store personal data were also obtained from all enrolled

patients. Incapacitated persons were not involved in the study; for patients

younger than 18 years who were treated in the 2 pediatric centers, a written,

informed consent from both parents was obtained. In view of the rapid

accrual of the first 15 patients and the encouraging response rate, the

institutional review board decided, for ethical reasons, to enroll 4 more

patients, while waiting for the next phase 2 multicenter Gruppo Italiano

Trapianto Midollo Osseo study to start. This study was approved by the

Ethic Committee of the Coordinating Center of Potenza.

Inclusion and exclusion criteria

Patients were eligible to be included in the study if they met the following

criteria. (1) They had a diagnosis of extensive cGVHD with fibrotic

scleroderma-like features (this definition included skin generalized fibrosis,

lung fibrosis, gut fibrosis, or any other extensive fibrotic process affecting

normal physiologic functions). The fibrotic process had to be documented

histologically; in detail, in the patients with sclerotic cutaneous involve-

ment (including fasciitis) a skin biopsy had been performed; also in

5 patients who showed gastrointestinal (GI) involvement (liver and gut), the

diagnosis of cGVHD was documented histologically. Only in one patient

with extensive and severe lung fibrosis (UPN 1346), the diagnosis of

cGVHD was not supported by histology and was based both on classic

clinical features, computed tomography scan of the chest and lung

functional test evaluation. A measure of organ function alterations had also

to be available, ie, percentage of body surface area (BSA), total skin

thickness score (TSC), and Zubrod performance status for skin involve-

ment; forced expiratory volume, forced vital capacity, and carbon monoxide

lung diffusion capacity, together with high-resolution chest compute

tomography scan for lung involvement. (2) They had failed at least 2 lines

of immunosuppressive therapy, including steroids, either as first- or

second-line therapy (all 19 patients had received prednisone orally at

dosages ranging from 0.5 to 1 mg/kg per day for at least 3 months after

an induction phase of 7-10 days of prednisone at 2 mg/kg per day).

(3) They had active disease with at least one of the following

manifestations: generalized skin sclerosis, symptomatic bronchiolitis

Table 1. Patients’ characteristics, n � 19

Value

Sex

Female/male 9/10

Median age, y (range) 29 (10-62)

Diagnosis

ALL/AML 5/6

CML/MM 2/2

Thalassemia/breast cancer 1/1

Fanconi/drepanocytosis 1/1

Type of transplantation, n

HLA-id sibling/MUD/CB 15/3/1

Stem cell source, n

BM /PBSC/CB 11/7/1

cGVHD first-line therapy, n

Prednisone 14

MMF 2

CsA-MMF 1

CsA-prednisone 1

Azathioprine-prednisone 1

cGVHD second and successive lines of therapy

Rituximab � ECP 1

Rituximab, mycophenolate 1

Rituximab � ECP � mycophenolate 1

ECP � rituximab � azathioprine 1

Rituximab � ECP � prednisone � CsA 2

Rituximab � ECP � prednisone � CsA � high-dose CTX 1

Rituximab � ECP � prednisone � methotrexate � CTX 1

Miscellaneous, including ECP (no rituximab) 11

Median cGVHD duration, mo (range) 37 (4-107)

Main cGVHD targets

Skin (generalized scleroderma) 11

Skin (localized scleroderma) 6

Eosinophilic fasciitis 1

Lung (BOP/fibrosis) 11

Sicca syndrome, yes/no 15/4

Other visceral manifestations (liver, bowel) 8

ALL indicates acute lymphoblastic leukemia; AML, acute myelogenous leukemia;

CML, chronic myelogenous leukemia; MM, multiple myeloma; HLA-id sibling,

transplant from identical sibling; MUD, matched unrelated donor; CB, cord blood; BM,

bone marrow; PBSC, peripheral blood stem cell; MMF, mofetil mycophenolate; CsA,

cyclosporine; cGVHD, chronic graft-versus-host disease; ECP, extracorporeal-

photochemotherapy; CTX, cytoxan; and BOP, bronchiolitis obliterans pneumonia.

710 OLIVIERI et al BLOOD, 16 JULY 2009 � VOLUME 114, NUMBER 3

For personal use only. by guest on May 10, 2011. bloodjournal.hematologylibrary.orgFrom

Table 2. Details of the main organ involvement in the 19 patients with cGVHD

UPN

Skin fibrotic involvement

Zubrod PS

Lung involvement

Gut involvementBowel

involvement

GI score

Sicca

syndrome Biopsy

Schirmer

test‡

Basal respiratory functional tests

Basal

LFSBSA involved, %*

Total skin

score† FEV, % FVC, % DLCO, %

Liver

(bil > 2/ALP � 3 NR) Bowel

1 100 49 1 No Skin 6

1064 40 20 2 70 65 60 2 Diarrhea§ 2 No Skin 13

99991 100 44 2 Diarrhea§ 2 Yes Skin 2

99992 100 48 2 76 70 66 2 Diarrhea§ 2 Yes Skin 3

1467 Fasciitis: 90 51 3 Yes Skin 5

1602 25 18 1 78 80 70 2 Yes Skin 15

434 50 31 2 64 73 64 3 Yes Skin 13

1346 Lung NE 4 32 35 48 6 No NE 15

122 100 46 2 44 50 60 5 Yes Skin 4

143 25 21 1 N/474 Diarrhea§ 3 Yes Gut 15

1027 75� 43 2 Yes Skin 7

310AR 100¶ 46 2 68 70 68 3 N/479 Yes Gut 2

1992MP 100 47 2 65 72 70 3 2,3/N Yes Gut 3

401PP 100 51 3 60 70 74 3 N/386 Diarrhea§ 3 Yes Gut 5

8881047 80 42 2 Yes Skin 5

1023015 75 41 1 Yes Skin 5

1025779 30 24 2 74 78 72 2 Yes Skin 15

1024411 90 43 3 Yes Skin 5

9999450 30 23 3 70 68 68 1 N/567 No Gut 10

Total 17/19*: 11 skin 75%-100%,

6 skin � 50%

Range, 18-51 Range, 1-4 Range, 1-6 5/19 Range, 2-3 15/19 18/19

*BSA, body surface area (generalized scleroderma was 75%-100% BSA); PS, performance status; NE, not evaluable; bil, bilirubin total (normal range, 0-1.5 mg/dL); ALP, alkaline phosphatase (normal range, 38-126 UI/L); FEV, forced

expiratory volume; FVC, forced vital capacity; DLCO, carbon monoxide lung diffusion capacity; LFS-SCORE, Lung Function Score according to National Institutes of Health (NIH) consensus conference21; GI-SCORE, Gastrointestinal Score

according to NIH consensus conference21; and N, normal.

†According to the modified Rodnan score system.22

‡Millimeters of wetting of the paper after 5 minutes; the value is the mean of both eyes.

§More than 1000 mL per day.

�Ulcer involvement: 25% skin.

¶Erythema: 80% skin.

IMA

TIN

IBF

OR

RE

FR

AC

TO

RY

cG

VH

D7

11

BLO

OD

,16

JU

LY

2009�V

OLU

ME

114,N

UM

BE

R3

F

or personal use only. by guest on M

ay 10, 2011. bloodjournal.hem

atologylibrary.orgF

rom

obliterans pneumonia (in the absence of intercurrent infectious compli-

cations), extensive lung fibrosis, pathologically demonstrated visceral

fibrotic involvement of the gut.

Exclusion criteria were stable disease controlled by standard treatment

(including a daily maintenance dose of prednisone � 0.4 mg/kg per day),

RTX administration in the preceding 6 months, pregnancy, and secondary

malignancy.

Treatment schedule

Imatinib administration was planned for a minimum of 6 months, starting at

the initial dose of 100 mg per day. After 1 month, in absence of SAEs,

toxicity, or intolerance, the dose could be increased to 200 mg per day until

the end of the third month; afterward, if no response had occurred, in the

absence of SAEs, toxicity, or intolerance, patients were allowed to receive

400 mg per day. Total treatment duration depended on the clinical situation

of the patient and his or her response to therapy with imatinib. Concomitant

immunosuppressive treatment was allowed, according to the clinician’s

decision, including steroid maintenance. The concomitant administration of

other myelotoxic drugs was not permitted. If some patients had been

previously included in a salvage program with ECP, the continuation of this

treatment was allowed, whereas RTX administration was not. Supportive

care/other medications were given, according to institutional guidelines or

local practice or both. Patients were strictly monitored during imatinib

treatment to evaluate the treatment safety and unexpected SAEs, according

to the guidelines currently available.20

Response criteria

Response (evaluated at 3 and 6 months after beginning imatinib) was

defined as complete response (CR) or partial response (PR), according to

the criteria proposed by Couriel et al8 integrated by some chronic

GVHD–specific core measures for organ-specific manifestations, as sug-

gested by Pavletic et al.21 CR was defined as resolution of all manifestations

of skin cGVHD (in terms of sclerosis, erythematous rash, or ulcers) and

resolution of all manifestations related to chronic GVHD in the specific

organs involved, except for some irreversible changes.

PR was established according to the response evaluated in the main

involved organs as follows: for generalized skin involvement, PR was

defined as at least 50% improvement of BSA involved (in terms of sclerosis,

erythematous rash, or ulcers). A significant improvement in Zubrod

performance status by 1 or more (confirming a functional improvement)

was also required (Zubrod performance status was attributed by a senior

physician expert in cGVHD management). To exactly measure response

(both in terms of BSA involved and skin thickness) in patients with

cutaneous sclerotic lesions, a TSC, calculated according to the modified

Rodnan skin score (mRSS) system,22 was used. This score consists of an

evaluation of the patient’s skin thickness rated by clinical palpation with the

use of a scale from 0 to 3 (0 � normal skin; 1 � mild thickening;

2 � moderate thickening; 3 � severe thickening with inability to pinch the

skin into a fold) for each of 17 anatomic surface areas of the body: face,

anterior chest, abdomen, (right and left separately) fingers, forearms, upper

arms, thighs, lower legs, dorsum of hands, and feet. These individual values

are added, and the sum is defined as the TSC. Because most patients had

only sclerotic cutaneous involvement (except 2 who also had ulcers or

erythematous rash), the skin response was measured in terms of reduction

of the BSA involved and improvement of the mRSS. PR for ocular cGVHD

was defined as subjective improvement, with at least 50% reduction in the

frequency of artificial tear administration or as improvement in Schirmer

test in one or both eyes of at least 3 mm. For the respiratory tract, PR was

defined as sustained, measurable improvement in pulmonary function tests

(carbon monoxide lung diffusion capacity, forced expiratory volume, or

both) or the ability to reduce corticosteroids by at least 50% or both without

deterioration of pulmonary function. To better standardize the functional

respiratory improvement a lung functional score was calculated before and

after treatment, as suggested by the National Institute of Health Consensus

Conference for cGVHD20; similarly in patients with GI involvement the

responses were also graded according to severity scales from 0 to 3 (details

for response evaluation in the main organs involved are reported in a

supplemental file).

No response (NR) was defined as no change in cGVHD manifestations

or any minor response (MR) not fulfilling the above-mentioned criteria for

PR; patients who experienced early death as a result of cGVHD before the

time chosen for the assessment of response, as well as those requiring an

increase in steroid dosage, were considered as having had NR. Progressive

disease (PD) was defined as any worsening of cGVHD clinical manifesta-

tions while on treatment; patients with CR or PR in one organ and

simultaneous NR or PD in another were considered to have had a mixed

response (MXR). NR, MR, PD, and mixed response were considered to be

treatment failures (TFs).

Statistical considerations

With the use of the Simon minimax design23 15 patients have been planned

to be recruited in the first initial design, based on the following stopping

rules: more than one death potentially correlated to the treatment and less

than 7 responses (considered as CR and PR). Data were analyzed with SPSS

package (Version 13.0 for Windows; SPSS Inc) and NCSS 2007 (NCSS,

PASS, and GESS, Kaysville, UT). OS was measured from the start of

imatinib administration until death because of any reason and calculated

with the Kaplan-Meier method. The probability of imatinib TF was

expressed as cumulative incidence (CI), to adjust the analysis for competing

risks. CI was calculated from the start of imatinib treatment to the last

follow-up, to whichever event was considered TF (either intolerance to the

drug or lack of response), or to a competing event (relapse, development of

second neoplasia, or death because of causes independent of cGVHD).24

Results

Toxicity

The main hematologic and extrahematologic toxicities are reported

in Table 3. Four patients developed relevant extrahematologic side

effects, in 3 cases requiring imatinib discontinuation. Of them,

1 patient (UPN 310AR) developed John Cunningham virus encepha-

litis, and 2 experienced important fluid retention: 1 patient (UPN

Table 3. Toxicity of treatment in the 19 evaluable patients

No. of patients

Hematologic toxicity (grade 3-4 WHO)

Neutropenia/thrombocytopenia/anemia 0/0/1

Hematologic toxicity (grade 1-2 WHO)

Neutropenia/thrombocytopenia/anemia 1/1/1

Extrahematologic toxicity (grade 3-4 WHO)

Pneumonia 1

CNS viral infection (JCV) 1

Pleural effusion 1

Hydric retention/edema 1

Extrahematologic toxicity (grade 1-2 WHO)

Gastroenteric/liver 0/0

Kidney/neurologic 0/0

Cardiovascular 0

Hydric retention/edema 2

Nausea or vomiting 0

Stop imatinib (causes)

Yes/no 8/11

No response of cGVHD 3

Relapse of underlying disease requiring

new treatment 2

Hydric retention* 2

CNS viral infection (JCV) 1

WHO indicates World Health Organization; CNS, central nervous system; JCV,

John Cunningham virus; and cGVHD, chronic graft-versus-host disease.

*Dead.

712 OLIVIERI et al BLOOD, 16 JULY 2009 � VOLUME 114, NUMBER 3

For personal use only. by guest on May 10, 2011. bloodjournal.hematologylibrary.orgFrom

1602) developed pleural effusion and another (UPN 99992)

developed diffuse subcutaneous edema, which both disappeared

after drug discontinuation. In these 3 patients who stopped the

treatment, the dosage of imatinib at the time of drug discontinua-

tion was 100 mg per day. The fourth patient (UPN 1348, who had a

previous history of several episodes of respiratory infection before

imatinib) developed pneumonia, which successfully responded to

antibiotic therapy, without interrupting treatment with imatinib.

One more patient (UPN 1064) with CML developed anemia,

requiring treatment with recombinant human erythropoietin.

With a median observation time of 17 months from start of

imatinib (range, 8-22 months), 11 patients are still on imatinib

treatment, and in 8 patients treatment was discontinued. Reasons

for discontinuation were intolerance (the 3 patients reported

earlier), lack of response (n � 3), and relapse of the original

disorder (n � 2). The first relapsed patient (UPN 1467) was a

woman with multiple myeloma (MM), who had an extramedullary

relapse 6 months after inception of imatinib, requiring combined

therapy with surgical intervention, followed by radiotherapy and

donor lymphocyte infusions; up to now, the patient is alive and

well, without signs of MM and stable improvement of cGVHD

even after imatinib discontinuation. The second patient (UPN 1)

was a woman with acute myelogenous leukemia (AML), who

achieved PR after 6 months of imatinib treatment, but developed

extramedullary (skin) relapse after 8 months of therapy; she

received salvage therapy without response and died of leukemia.

A second tumor occurred in a patient (UPN 1064) with CML,

who developed an abdominal mass; a surgical biopsy showed

pathologic findings of diffuse large cell lymphoma (DLCL). The

dramatic unfavorable evolution of this secondary tumor did not

allow any further treatment, and the patient died 1 month later. We

cannot exclude a relation between the imatinib treatment and the

development of a secondary neoplasia.

Finally, we observed a further fatality in a patient (UPN 99992),

who died because of interstitial pneumonia, possibly attributable to

the profound immune depression, 207 days after imatinib discon-

tinuation; in this case treatment had been stopped because of

toxicity (diffuse subcutaneous edema), and the patient was

unresponsive.

Response

After 3 months of treatment, we observed, among the 19 evalu-

able patients, 1 CR (5%) and 11 PR (58%), with an overall

response rate of 63%, with 2 more patients having MR (see also

Table 4). Afterward, imatinib dosage has been modified as

follows: 1 patient reduced the dose to 50 mg per day because of

fluid retention; in 2 patients dosage was increased to 200 mg per

day, whereas in the remaining patients imatinib was maintained

at 100 mg per day. After 6 months of treatment, 5 patients

converted from PR to CR, and 1 more patient converted from

NR to CR; the patient in CR after 3 months of imatinib

maintained the CR, the overall CR rate being 37% (7 of 19). The

remaining 6 patients in PR maintained this response at 6 months,

and 2 more patients converted from NR or MR to PR, leading to

an overall PR rate of 42% (8 of 19). The overall response rate

(CR � PR) at 6 months was 79%. After 6 months of treatment,

among the 8 patients who had previously been treated with

rituximab, we observed 2 CRs and 6 PRs; among the 11 patients

who had not received rituximab before imatinib, the response

rate was 64%, with 5 CRs and 2 PRs. In 10 patients, imatinib

treatment allowed that steroids were stopped or tapered off or

their dosage significantly reduced (see Table 4). It is remarkable

that 2 patients (UPNs 310AR and 1992MP) with severe lung

involvement and who depended on oxygen were able to

discontinue oxygen administration. At last follow-up, among the

responding patients, 11 are still on imatinib therapy. In the long

term, 2 more patients discontinued imatinib, because of subjec-

tive intolerance (represented in both cases by mild symptoms

such as myalgia and fluid retention) and 1 patient for MM

relapse (all these 3 patients maintained a stable response); one

had died of leukemia relapse. The details of response after

3 and 6 months, considering each organ involved, are reported

in Tables 5 and 6.

Table 4. Response and outcome in the 19 evaluable patients

UPN

Response after

3 mo*

Daily imatinib

dosage, mg

Response after

6 mo

Daily imatinib

dosage, mg Status

Continues

imatinib

Duration of

imatinib

treatment, mo

Concomitant steroid

therapy

1 NR 100 PR 100 Dead No 8 Yes

1064 MxR 100 PR 100 Dead No 6 Stop

99991 PR 100 CR 100 Alive Yes 16 Yes

99992 NR 100 CR 50 Dead No 6 Yes

1467 PR 100 CR 100 Alive Yes 10 Stop

1602 PR 100 PR 100 Alive No 6 Yes

434 PR 100 PR 100 Alive Yes 11 Yes

1346 NR 100 NR 100 Alive No 6 Yes

122 PR 100 PR 200 Alive Yes 15 Decrease 50%

143 PR 100 PR 100 Alive Yes 9 Decrease 70%

1027 PR 100 PR 100 Alive Yes 7 Stop

310AR PR 100 CR 100 Alive Yes 11 Stop

1992MP CR 100 CR 100 Alive Yes 12 Yes

401PP MR 100 MR 100 Alive No 8 Stop

8881047 PR 100 CR 100 Alive Yes 13 Stop

1023015 PR 100 CR 100 Alive Yes 13 Stop

1025779 NR 100 NR 100 Alive No 6 Increase 50%

1024411 NR 100 NR 100 Alive No 6 Increase 50%

9999450 PR 100 PR 200 Alive Yes 9 Decrease 80%

Total 19 pt CR � Pr 12 CR � Pr 15 16/19 11/8 6-16(9) 7 stop, 3 decrease, 2 increase

NR indicates no response; PR, partial remission; MxR, mixed response; CR, complete remission; and MR, minor response.

IMATINIB FOR REFRACTORY cGVHD 713BLOOD, 16 JULY 2009 � VOLUME 114, NUMBER 3

For personal use only. by guest on May 10, 2011. bloodjournal.hematologylibrary.orgFrom

Table 5. Response details at 3 and 6 months: skin, lung, Zubrod PS

UPN

Total skin score

at diagnosis

BSA* involved

at diagnosis,

%

Total skin score

at 3 mo

BSA* involved

at 3 mo, %

Total skin score

a 6 mo

BSA* involved

at 6 mo, %

LFS Zubrod

At diagnosis At 3 mo At 6 mo

At

diagnosis At 3 mo At 6 mo

1 49 100 46 90 21 40 1 1 0

1064 20 40 15 32 5 16 2 2 2 2 2 1

99991 44 100 20 50 0 0 2 1 0

99992 48 100 46 80 0 0 2 2 0 2 2 0

1467 51 90 Fasciitis: 90 25 40 0 0 3 1 0

1602 18 25 9 10 4 4 2 1 0 1 0 0

434 31 50 15 24 7 10 3 2 1 2 1 1

1346 NE NE NE NE NE NE 6 6 6 4 4 3

122 46 100 22 50 10 20 5 4 4 2 1 0

143 21 25 10 10 4 4 1 0 0

1027 43 75 21 30 10 14 2 1 0

310AR 46 100 18 50 0 0 3 2 0 2 0 0

1992MP 47 100 0 0 0 0 3 1 0 2 0 0

401PP 51 100 33 50 16 40 3 2 0 3 2 1

8881047 42 80 14 30 0 0 2 0 0

1023015 41 75 18 36 0 0 1 1 0

1025779 24 30 24 30 20 20 2 2 2 2 2 2

1024411 43 90 40 90 36 90 3 3 2

9999450 23 30 11 12 5 6 1 1 1 3 2 1

Total 19 pt Range,18-51 Mean/range,

73/25-100

Range, 0-46 Mean/range,

40/0-90

Range, 0-36 Mean/range,

15/0-90

Range, 1-6 Range, 1-6 Range, 0-6 Range, 1-4 Range, 0-4 Range, 0-2

CR indicates complete remission; PR, partial remission; MxR, mixed response; NR, no response; LFS-SCORE, Lung Function Score according to NIH consensus conference21; and NE, not evaluable.

*BSA, body surface area (generalized scleroderma: 75%-100%).

714

OLIV

IER

Ietal

BLO

OD

,16

JU

LY

2009�V

OLU

ME

114,N

UM

BE

R3

F

or personal use only. by guest on M

ay 10, 2011. bloodjournal.hem

atologylibrary.orgF

rom

Table 6. Response details at 3 and 6 months: gut, Sicca syndrome

UPN

Bowel involvement, GI score

Liver involvement,

bil > 2/ALP � 3 NR Schirmer test* Sicca syndrome†

At

diagnosis At 3 mo At 6 mo At diagnosis At 3 mo At 6 mo

At

diagnosis At 3 mo At 6 mo

At

diagnosis At 3 mo At 6 mo

1 6 6 10 No No No

1064 13 13 15 No No No

99991 2 1 0 2 7 15 Yes Yes No

99992 2 2 0 3 3 15 Yes Yes No

1467 2 1 0 5 10 15 Yes Yes No

1602 15 15 15 Yes Yes Yes

434 13 13 13 Yes Yes Yes

1346 15 15 15 No No No

122 4 8 12 Yes Yes No

143 N/474 N/228 N/N 15 15 15 Yes Yes Yes

1027 3 2 1 7 11 14 Yes Yes Yes

310AR N/479 N/90 N/84 2 6 15 Yes Yes No

1992MP 2,3/N 1,8/N 1,6/N 3 15 15 Yes No No

401PP N/386 N/240 N/180 5 8 11 Yes Yes Yes

8881047 3 1 0 5 9 15 Yes Yes No

1023015 5 9 15 Yes Yes No

1025779 15 15 15 Yes Yes Yes

1024411 5 5 5 Yes Yes Yes

9999450 N/567 N/567 N/567 10 15 15 No No No

Mean 2 1 0 5 10 15

Range 2-3 1-2 0-1 2-15 5-15 5-15

Total 19 pt Mean/range,

2/2-3

Mean/range,

1/1-2

Mean/range,

0/0-1

Mean/range,

5/2-15

Mean/range,

10/5-15

Mean/range,

15/5-15

15/19 14/19 7/19

CR indicates complete remission; PR, partial remission; MxR, mixed response; MR, minor response; and NR, no response.

*Millimeters of wetting of the paper after 5 minutes; the value is the mean of both eyes.

IMA

TIN

IBF

OR

RE

FR

AC

TO

RY

cG

VH

D7

15

BLO

OD

,16

JU

LY

2009�V

OLU

ME

114,N

UM

BE

R3

F

or personal use only. by guest on M

ay 10, 2011. bloodjournal.hem

atologylibrary.orgF

rom

Outcome

With a median follow-up of 17 months (range, 8-22 months),

16 patients are alive, 14 still maintaining benefit from imatinib

treatment (6 in CR and 8 in PR); 1 more patient maintains MR with

a stable disease. The 18-month Kaplan-Meier estimate of OS

calculated from the inception of imatinib is 85% (95% confidence

interval, 67%-100%; see also Figure 1), whereas the CI of TF was

35% (95% confidence interval, 17%-63%; see also Figure 2). Three

patients died: one with AML (UPN 1) had leukemia relapse while

completely responding to imatinib and died because of disease

progression; one (UPN 1064) died because of secondary neoplasia

(DLCL); a third patient (UPN 99 992), who had a previous history

of lung infections consequent to cGVHD and its treatment, died

because of pneumonia. At time of death, this patient was receiving

steroids plus cyclosporine. No further unexpected SAEs have been

observed during the late follow-up of patients still on Imatinib

treatment.

Discussion

Treatment options for patients developing cGVHD are limited and

often unsatisfactory; steroids can be effective, but many patients

have a suboptimal response and are unable to discontinue steroids

even after months or years of therapy. The addition of a calcineurin

inhibitor may be of benefit, allowing to reduce the dosage of

steroids used, with a consequent lower risk of steroid-related

toxicity, but it proved to be unable to reduce transplantation-related

mortality among patients with chronic GVHD.4 Similarly, the

addition of other agents to the combination of steroids plus

calcineurin inhibitors has still not definitively proved to signifi-

cantly ameliorate the long-term outcome of patients with steroid-

resistant cGVHD.5

In view of these findings, it is not surprising that there is no

standard approach for patients with cGVHD refractory to steroids,

and these patients are entitled to experimental treatment protocols

based on innovative approaches, able to interfere with alloautoreac-

tive response at the basis of chronic GVHD, as well as with the

mechanisms responsible for the development of fibrosis.25,26 Block-

ade of PDGF or TGF-� signaling has been shown to reduce the

development of fibrosis in various experimental models.15-17 How-

ever, agents that selectively inhibit PDGF pathways are not yet

available for clinical application. Similarly, human antibodies

against TGF-� did not show clinical efficacy in patients with SSc.27

In view of in vitro data showing that imatinib mesylate exerts

selective, dual inhibition of the TGF-�/PDGF pathways,17 we

conducted a prospective study evaluating safety and activity of

imatinib mesylate in a series of patients affected by refractory

cGVHD, with fibrotic/sclerotic features. In our cohort, all patients

enrolled in the study had previously failed at least 2 lines of therapy

(including steroids in all 19 patients, RTX, ECP, mofetil mycophe-

nolate, methotrexate, and cyclophosphamide; see Table 1 for

details). Although our patients were heavily pretreated, most of

them responded to imatinib treatment, showing either CR or PR,

and the OS probability at 18 months from imatinib inception was

84%. These results compare favorably with other salvage therapies,

which, in similar settings, obtained very low response rates,6,9,28

except for the series by Jacobsohn et al29 who reported a response

rate of 55% with Pentostatin and for that reported by Couriel et al8

who obtained 61% response rate with ECP. Jacobsohn et al29

prospectively enrolled 53 patients who had failed first-line therapy,

whereas Couriel et al8 retrospectively evaluated 71 patients with

severe cGVHD (8 heavily pretreated, with � 3 lines of immunosup-

pression, including steroids). Both studies showed an encouraging

outcome with an OS of 53% at 1 year with ECP (but only of 19% at

5 years) and a 2-year OS of 70% with Pentostatin. The beneficial

role played by ECP in patients with cGVHD has also been

confirmed in a recently published randomized trial documenting

that it may have a steroid-sparing effect, concomitant with improve-

ment in skin disease.7

A limitation of our study may be represented by the criteria of

inclusion we chose; indeed, in view of the notion that imatinib is a

potent antifibrotic drug, we enrolled only patients with fibrotic/

sclerotic features cGVHD. Therefore, it remains to be tested

whether imatinib mesylate can be effective also in patients with

cGVHD without these clinical features. However, it should be

emphasized that most patients in our series had not only extensive

skin fibrosis, but also many of them had other systemic manifesta-

tions, such as lung involvement, sicca syndrome, or gut involvement.

Considering the degree of pathologic damage present in many

patients with fibrotic features of cGVHD, and the mechanism of

action of this antifibrotic drug presumably associated with a

response slower than that achievable with immunosuppressive

drugs, we decided to have 2 relatively late time points of evaluation

of efficacy, namely 3 and 6 months after imatinib inception, to be

able to observe a late clinical improvement in the organs involved

0.00

0.25

0.50

0.75

1.00

0 5 10 15 20 25

Month from start of Imatinib treatment

Surv

ivalpro

babili

ty(9

5%

CI)

Number of patients at risk: 19 19 19 12 5 0

Overall survival probability

OS = 84% (67-100)

Figure 1. Overall survival probability (OS) of the 19 patients with cGVHD

receiving imatinib. OS was measured from start of imatinib treatment.

Cumulative incidence of treatment failure

Month from start of Imatinib treatment

Cum

ula

tive incid

ence

(95%

CI)

Number of patients at risk: 19 19 7 3 1 0

0.25

0.50

0.75

1.00

5 10 15 20 25

CI = 35% (17-63)

0.000

Figure 2. Cumulative incidence (CI) of treatment failure in the 19 patients with

cGVHD receiving imatinib. CI was calculated from the start of imatinib to whichever

event was considered as treatment failure (intolerance or lack of response).

716 OLIVIERI et al BLOOD, 16 JULY 2009 � VOLUME 114, NUMBER 3

For personal use only. by guest on May 10, 2011. bloodjournal.hematologylibrary.orgFrom

by the fibrotic process. Indeed, most patients converted from NR to

PR or from PR to CR after 3 more months of treatment, suggesting

that in the absence of response an early discontinuation of the

treatment could be premature or even detrimental.

In our protocol we used prospectively defined objective re-

sponse criteria, and the follow-up of our patients exceeds the

short-term outcomes recommended for phase 2 studies by the

National Institutes of Health (NIH) Consensus Conference on

cGVHD.25 We used the response criteria proposed by Couriel et al,8

integrated by some recommendations proposed by the recent NIH

Consensus Conference on cGVHD,21 which permit a more stringent/

precise evaluation of response in specific organs. Unfortunately, as

reported in the NIH Consensus Conference document, no validated

scale exists for assessing sclerotic skin changes of cGVHD, and

more sophisticated skin-specific scores are warranted. A comprehen-

sive scale has been proposed by NIH to evaluate also other

nonfibrotic skin lesions, but it has not yet been validated in

prospective clinical trials; only one recent study retrospectively

used the NIH response criteria.30 Because the secondary end point

of our study was the response rate in patients with fibrotic/sclerotic

features, we quantified the skin fibrotic response by using an

mRSS.22 The validity of this mRSS as a semiquantitative method

has been supported by findings of a series of studies in patients with

systemic sclerosis.31 For the response in organs other than the skin,

we generally observed a response also in those patients showing

involvement of the respiratory tract and of the GI tract. Notably, we

also observed an unexpected improvement of sicca syndrome,

which, so far, has been considered an irreversible sequela. It is

tempting to speculate that imatinib treatment resulted in a reversion

of the fibrosis of the lachrymal gland, permitting a better produc-

tion of tears, as suggested by the improvement observed in the

Schirmer test. However, this peculiar aspect deserves further and

more specific investigations.

The toxicity profile of imatinib in this study was acceptable. The

young age of most of our patients and the significantly lower

dosage of imatinib, compared with the 400 mg per day, usually

administered to patients with CML possibly contributed to this

safety profile. This said, the good response rate observed suggests

that, in this subset of patients, imatinib could be effective even at a

low dose such as 100 mg per day. The toxicities of concern were

represented by the well-known fluid retention associated with

imatinib treatment, whereas the fatalities we observed cannot be

reasonably attributed to imatinib treatment. As mentioned earlier,

one patient developed a secondary DLCL, and 2 patients showed a

relapse of the original malignancy (MM and AML) after imatinib

therapy. Secondary neoplasia and disease relapse/progression are

not infrequent among patients with cGVHD, who require continu-

ous immunosuppressive therapy, potentially impairing the mecha-

nisms of antitumor immune surveillance. However, we cannot

definitively exclude that imatinib has played a role as additive

immunosuppressive agent, as suggested by some reports.32

In conclusion, this preliminary experience indicates that low-

dose imatinib is active and tolerated in patients affected by cGVHD

with sclerotic features and that the responses obtained are sustained

over time, resulting in an encouraging patient outcome, as proved

by the 18-month OS of 84%. Moreover, a number of steroid-

dependent patients were able to either reduce or even discontinue

steroids. Our results support the conduction of further confirmatory

studies with low-dose tyrosin-kinase inhibitors in patients with

refractory cGVHD. In this line, a multicenter prospective Italian

study aimed at confirming these data is ongoing.

Acknowledgments

This work was supported partially by grants from AIRC (Associa-

zione Italiana Ricerca sul Cancro), CNR (Consiglio Nazionale

delle Ricerche), MURST (Ministero dell’Universita e della Ricerca

Scientifica e Tecnologica), European Union (FP6 program AL-

LOSTEM), Fondazione IRCCS Policlinico San Matteo (F.L.), and

Azienda Ospedaliera San Carlo.

Authorship

Contribution: A.O., A.B., and A.G. conceived and designed the

study; F.L., M.Z., A.S., P.L., G.G., R.R., N.M., and A.B. provided

study materials or patients; M.C. and A.O. collected and assembled

data; A.O., F.L., A.B., and M.C. analyzed and interpreted data; and

A.O., F.L., and A.B. wrote the manuscript.

Conflict-of-interest disclosure: The authors declare no compet-

ing financial interests.

Correspondence: Attilio Olivieri, UOC di Ematologia-Centro

Trapianto di Cellule Staminali, Azienda Ospedaliera Regionale San

Carlo, via Potito Petrone 1, Potenza, 85100 Italy; e-mail: attilio.

olivieri@ospedalesancarlo.it.

References

1. Copelan EA. Hematopoietic stem cell transplan-

tation. N Engl J Med. 2006;354:1813-1826.

2. Ferrara JLM, Antin JH. The pathophysiology of

graft-vs.-host disease. In: Blume KG, Forman SJ,

Appelbaum FR, eds. Thomas’ Hematopoietic Cell

Transplantation. 3rd Ed. Malden, MA: Blackwell

Publishing; 2004:353-368.

3. Bell SA, Faust H, Mittermuller J, Kolb HJ, Meurer

M. Specificity of antinuclear antibodies in

scleroderma-like chronic graft -versus-host dis-

ease: clinical correlation and histocompatibility

locus antigen association. Br J Dermatol. 1996;

134:848-854.

4. Koc S, Leisenring W, Flowers ME, et al. Therapy

for chronic graft-versus-host disease: a random-

ized trial comparing cyclosporine plus prednisone

versus prednisone alone. Blood. 2002;100:48-51.

5. Arora M, Wagner JE, Davies SM, et al. Random-

ized clinical trial of thalidomide, cyclosporine, and

prednisone versus cyclosporine and prednisone

as initial therapy for chronic graft-versus-host dis-

ease. Biol Blood Marrow Transplant. 2001;7:265-

273.

6. Kim JG, Sohn SK, Kim DH, et al. Different effi-

cacy of mycophenolate mofetil as salvage treat-

ment for acute and chronic GVHD after alloge-

neic stem cell transplant. Eur J Haematol. 2004;

73:56-61.

7. Flowers MED, Apperley JF, Greinix HT, et al. A

multicenter prospective phase 2 randomized

study of extracorporeal photopheresis for treat-

ment of chronic graft-versus-host disease. Blood.

2008;112:2667-2674.

8. Couriel DR, Hosing C, Saliba R, et al. Extracorpo-

real photochemotherapy for the treatment of

steroid-resistant chronic GVHD. Blood. 2006;107:

3074-3080.

9. Cutler C, Miklos D, Kim HT, et al. Rituximab for

steroid-refractory chronic graft-versus-host dis-

ease. Blood. 2006;108:756-762.

10. Socie G, Stone JV, Wingard JR, et al. Long-term

survival and late deaths after allogeneic bone

marrow transplantation: Late Effects Working

Committee of the International Bone Marrow

Transplant Registry. N Engl J Med. 1999;341:

14-21.

11. Svegliati S, Olivieri A, Campelli N, et al. Stimula-

tory autoantibodies to PDGF receptor in patients

with extensive chronic graft-versus-host disease.

Blood. 2007;110:237-241.

12. Svegliati S, Baroni S, Santillo MR, et al. Stimula-

tory autoantibodies to the PDGF receptor in sys-

temic sclerosis. N Engl J Med. 2006;354:2667-

2676.

13. Bonner JC. Regulation of PDGF and its receptors

in fibrotic diseases. Cytokine Growth Factor Rev.

2004;15:255-273.

14. McCormick LL, Zhang Y, Tootell E, et al. Anti-

TGF-beta treatment prevents skin and lung fibro-

sis in murine sclerodermatous graft-versus-host

disease: a model for human scleroderma. J Im-

munol. 1999;163:5693-5699.

15. Abdollahi A, Li M, Ping G, et al. Inhibition of

platelet-derived growth factor signalling attenuates

pulmonary fibrosis. J Exp Med. 2005;201:925-935.

IMATINIB FOR REFRACTORY cGVHD 717BLOOD, 16 JULY 2009 � VOLUME 114, NUMBER 3

For personal use only. by guest on May 10, 2011. bloodjournal.hematologylibrary.orgFrom

16. Daniels CE, Wikes MC, Edens M, et al. Imatinib

mesylate inhibits the profibrogenic activity of

TGF-� and prevents bleomycin mediated lung

fibrosis, J Clin Invest. 2004;114:1308-1316.

17. Distler JH, Jungel A, Huber LC, et al. Imatinib me-

sylate reduces production of extracellular matrix

and prevents development of experimental der-

mal fibrosis. Arthritis Rheum. 2007;56:311-322.

18. Druker BJ, the IRIS Investigators. Five-year

follow-up of patients receiving imatinib for chronic

myeloid leukemia. N Engl J Med. 2006;355:2408-

2417.

19. Vitali C, Bombardieri S, Jonsson R, et al. Classifi-

cation criteria for Sjogren’s syndrome: a revised

version of the European criteria proposed by the

American-European Consensus Group. Ann

Rheum Dis. 2002;61:554-558.

20. Couriel D, Carpenter PA, Cutler C, et al. Ancillary

therapy and supportive care of chronic graft-

versus-host disease: national institutes of health

consensus development project on criteria for

clinical trials in chronic Graft-versus-host disease,

V: Ancillary Therapy and Supportive Care Work-

ing Group Report. Biol Blood Marrow Transplant.

2006;12:375-396.

21. Pavletic SZ, Martin P, Lee SJ, et al. Response

Criteria Working Group. Measuring therapeutic

response in chronic graft-versus-host disease:

National Institutes of Health Consensus Develop-

ment Project on Criteria for Clinical Trials in

Chronic Graft-versus-Host Disease, IV: Re-

sponse Criteria Working Group report. Biol Blood

Marrow Transplant. 2006;12:252-266.

22. Clements PJ, Lachenbruch PA, Seibold JR, et al.

Skin thickness score in systemic sclerosis: an

assessment of interobserver variability in 3 inde-

pendent studies. J Rheumatol. 1993;20:1892-

1896.

23. Simon R. Optimal two-stage designs for phase II

clinical trials. Control Clin Trials. 1989;10:1-10.

24. Gooley TA, Leisenring W, Crowley J, Storer BE.

Estimation of failure probabilities in the presence

of competing risks: new representations of old

estimators. Stat Med. 1999;18:695-706.

25. Martin PJ, Weisdorf D, Przepiorka D, et al. Na-

tional Institutes of Health consensus develop-

ment project on criteria for clinical trials in chronic

graft-versus-host disease, VI: Design of Clinical

Trials Working Group report. Biol Blood Marrow

Transplant. 2006;12:491-505.

26. Antin JH, Chen AR, Couriel DR, Ho VT, Nash RA,

Weisdorf D. Novel approaches to the therapy of

steroid-resistant acute graft-versus-host disease.

Biol Blood Marrow Transplant. 2004;10:655-668.

27. Denton CP, Merkel PA, Furst DE, et al. Cat-192

Study Group; Scleroderma Clinical Trials Consor-

tium. Recombinant human anti-transforming

growth factor beta1 antibody therapy in systemic

sclerosis: a multicenter, randomized, placebo-

controlled phase I/II trial of CAT-192. Arthritis

Rheum. 2007;56:323-333.

28. Johnston LJ, Brown J, Shizuru JA, et al. Rapamy-

cin (sirolimus) for treatment of chronic graft-ver-

sus-host disease. Biol Blood Marrow Transplant.

2005;11:47-55.

29. Jacobsohn DA, Chen AR, Zahurak M, et al.

Phase II study of pentostatin in patients with corti-

costeroid-refractory chronic graft-versus-host dis-

ease. J Clin Oncol. 2007;25:4255-4261.

30. Cho BS, Min CK, Eom KS, et al. Feasibility of NIH

consensus criteria for chronic graft-versus-host

disease. Leukemia. 2009;23:78-84.

31. Brennan P, Silman A, Black C, et al. Reliability of

skin involvement measures in SSc. Br J Rheuma-

tol. 1992;31:457-460.

32. Seggewiss R, Lore K, Wiestner A, et al. Imatinib

inhibits T-cell receptor–mediated T-cell prolifera-

tion and activation in a dose-dependent manner.

Blood. 2005;105:2473-2479.

718 OLIVIERI et al BLOOD, 16 JULY 2009 � VOLUME 114, NUMBER 3

For personal use only. by guest on May 10, 2011. bloodjournal.hematologylibrary.orgFrom