Prednisone and Azathioprine Compared With Prednisone and ...€¦ · prednisone (both survive with 100% performance). In group II, two Patients with aplastic anemia were prepared

546 Blood, Vol 72, No 2 (August), 1988: pp 546-554

Prednisone and Azathioprine Compared With Prednisone and Placebo forTreatment of Chronic Graft-v-Host Disease: Prognostic Influence of

Prolonged Thrombocytopenia After Allogeneic Marrow Transplantation

By K.M. Sullivan, R.P. Witherspoon, R. Storb, P. Weiden, N. Flournoy, S. Dahlberg, H.J. Deeg, J.E. Sanders,

K.C. Doney, F.R. Appelbaum, R. McGuffin, G.B. McDonald, J. Meyers, M.M. Schubert, J. Gauvreau,

H.M. Shulman, G.E. Sale, C. Anasetti, T.P. Loughran, S. Strom, J. Nims, and E.D. Thomas

We conducted a randomized. double-blind comparison of

prednisone and placebo (group I) v prednisone and azathio-

prine (1 .5 mg/kg/day) (group II) as early treatment of

extensive chronic graft-v-host disease (GVHD). Patients

with platelet counts <100.0O0/�sL were placed into ther-

apy with prednisone alone (group III). All three groups

received identical doses of prednisone (1 mg/kg every

other day) and one double-strength trimethoprim-sulfame-

thoxazole (TMP-SMX) tablet twice daily. Between January

1 980 and December 1 983. 1 79 previously untreated

patients were enrolled and 1 64 were evaluable. Patients

randomized to group I (n = 63) and group II (n - 63) were

well matched for prognostic factors; those placed into

group III (n - 38) had more frequent acute GVHD and

progressive onset of chronic GVHD. Median duration of

therapy was 2 years. Complications included diabetes

(5%). aseptic necrosis (5%) and infection. For groups I, II.

and Ill, the respective incidence of infection was dissemi-

nated varicella. 1 1 %. 24%. 34%; bacteremia, 6%. 11%.

C HRONIC graft-v-host disease (GVHD) is a major late

complication in 25% to 45% of survivors of allogeneic

marrow transplantation.’3 Less than 20% of patients with

untreated extensive chronic GVHD survive with Karnofsky

performance scores �7O%.� Therapy with antithymocyte

globulin or corticosteroids given late in the course of disease

results in little benefit.2 In a pilot study of combined immu-

nosuppression with cytotoxic agents and corticosteroids, I 6

of 21 patients survive free of disability.2 However, compari-

son of therapy with azathioprine and prednisone to our

previous experience with prednisone alone was complicated

by the fact that patients given the combination were treated

34%; and interstitial pneumonia. 5%. 1 4%, 1 8%. Recurrent

malignancy was the most frequent cause of death and did

not differ significantly across the groups. Nonrelapse mor-

tality, however. did differ: 21 % in group I. 40% in group II,

and5s%ingrouplll(l vlI.P= .003;l vlll.P=- .001). Forty

patients in group I, 30 in group II, and 10 in group Ill survive

with a minimum follow-up of 3.8 years. Karnofsky perfor-

mance scores for 68 survivors are 90% to 1 00%, scores for

seven survivors are 70% to 89% and scores for five

survivors are <70%. Actuarial survival at 5 years after

transplant is 61 % in group I, 47% in group II, and 26% in

group III (I v II. P = .03; I v III. P = .0001 ). Treatment with

prednisone alone results in fewer infections and better

survival than prednisone and azathioprine in standard-risk

chronic GVHD. Treatment with prednisone alone is less

effective in high-risk patients with thrombocytopenia. and

other strategies are required.

a 1988 by Grune & Stratton. Inc.

in more recent times when earlier diagnosis and therapy,

better supportive care and closer follow-up were used.

Because the benefits and toxicities of long-term cytotoxic

therapy after marrow grafting are unclear, we studied in a

randomized, double-blind trial the use of prednisone and

placebo compared with prednisone and azathioprine in early

treatment of chronic GVHD. Patients with poor marrow

function as a contraindication to azathioprine (platelet

counts < 100,000/zL) were placed into treatment with pred-

nisone alone. We report the results of this controlled clinical

trial, which has a minimum follow-up of 3.8 years after

transplant.

MATERIALS AND METHODS

From the Fred Hutchinson Cancer Research Center and the

University of Washington School ofMedicine, Seattle.

Submitted August 12. 1987; accepted March 29. 1988.

Supported by US Public Health Service Grants No. CA 18221.

CA 18029. CA 09515, CA 15704 from the National Cancer Insti-tute. and HL 36444from the National Institute ofHeart. Lung and

Blood, Department ofHealth and Human Services.

Presented in part at the 29th Annual Meeting of the AmericanSociety ofHematology. Washington. DC. December 5-8. 1987.

Dr Thomas is a recipient of Research Career Award Al 02425from the National Institute of Allergy and Infectious Diseases,

DHHS. Dr. Loughran is a Fellow of the Leukemia Society ofAmerica.

Address reprint requests to Keith M. Sullivan. MD. Fred Hut-

chinson Cancer Research Center, Department ofMedicine. DivisionofOncology, I 124 Columbia St. Seattle, WA 98104.

The publication costs of:his article were defrayed in part by page

charge payment. This article must therefore be hereby marked

“advertisement” in accordance with 18 U.S.C. section 1 734 solely to

indicate this fact.

© 1988 by Grune & Stratton. Inc.

0006-4971/88/7202-0019$3.00/0

From January 1980 to December 1983, 179 patients with exten-

sive (multiorgan) chronic GVHD entered the study. Seventyenrolled patients were randomized to receive prednisone and placebo(group I), and 7 1 were randomized to receive prednisone and

azathioprine (group II). An additional 38 patients with thrombocy-topenia (group III) were placed into treatment with prednisonealone. When an interim analysis showed an increased mortality,4group III was closed and treatment was modified to include alternat-

ing cyclosporine and prednisone. Results of that subsequent studywere presented in another report.5

Fifteen (1 1%) of 141 enrolled patients randomized to groups I and

II had violations of the double-blind treatment protocol and were notevaluable for response. Five patients refused treatment after enroll-ment because they considered that therapy was not required. Fiverefused the study drug and two took open-label azathioprine becausethey or their physicians wished to choose the specific drug regimen.The remaining three patients discontinued treatment before com-pleting the 9-month schedule. The results in these I 5 inevaluable

patients were as follows. In group I, three refused therapy (two died

and one survived with 40% performance); one stopped treatmentafter 3 months and died; one refused study drug, took prednisone and

azathioprine, and died; and two refused study drug and took

For personal use only.on October 22, 2017. by guest www.bloodjournal.orgFrom

prednisone (both survive with 100% performance). In group II, two Patients with aplastic anemia were prepared for transplant withrefused therapy (both died); three refused study drug (one died, one cyclophosphamide, 50 mg/kg on each of four successive days.’

survived with 50% performance, and one survived with 100% perfor- Patients with hematologic malignancies received high-dose cyclo-mance); one complied poorly with treatment and died; one refused phosphamide or other chemotherapy regimens followed by totalstudy drug, took prednisone and azathioprine, and died; and one body irradiation (TBI) given as a single lO-Gy dose or as a

stopped treatment after 6 months and survived with 100% perfor- fractionated I 2- to I 7.5-Gy dose.7’8 All but 14 patients receivedmance. marrow from HLA-identical siblings; those 14 received marrow

Demographic data of 164 evaluable patients are listed in Table 1. from HLA-nonidentical donors.’ Three HLA-identical recipients

Table 1 . Patient Characteristics at Study Entry

Randomized Placed

Prednisone AlonePrednisone + Placebo Prednisone + Azathioprine

Characteristic (Group I) (Group II) (Group III)

No. of evaluable patients 63 63 38

0

35

2

0

PREDNISONE ± AZATHIOPRINE IN CHRONIC GVHD 547

Abbreviations: ALL, acute lymphoblastic leukemia; ANL, acute nonlymphoblastic leukemia; CML, chronic myelogenous leukemia; GVHD, graft-v-host

disease; MTX, methotrexate.

Diagnosis

Aplastic anemia

Refractory ANL/ALL

Relapse ANL/ALL

2nd+ Remission ANL/ALL

1st Remission ANL/ALL

Accelerated o�’ blast crisis CML

Chronic phase CML

Lymphoma/other

No. (%) of patients

9

2/0

6/6

1/10

13/3

4

7

2/0

7

4/0

8/6

1/3

12/3

8

7

2/2

2

0/0

1/3

2/8

1 1/1

4

4

2/0

<10 yr

10-30 yr

>30 yr

Median (range) age in years

Sex (M/F)

HLA-nonidentical donor (%)

Bone marrow T-ceII depletion

Acute GVHD prophylaxis

None

ShortMTX(11 d)

Standard MTX (102 d)

Cyclosporine (180 d)

MTX/cyclosporine ( 1 1 / 1 80 d)

Grade of Acute GVHD

0

II

III

IV

Total grade ll-lV (%)

Type of onset of chronic GVHD (%)

De novo

Quiescent

Progressive

Presentation of chronic GVHD (%)

Subclinical always

Subclinical -‘ clinical

Clinical always

No. (%) Entering treatment

�4 mo after transplant

5- 1 2 mo after transplant

> 1 2 mo after transplant

Median (range) values at entry

Time to treatment (months after transplant)

Karnofsky score (%)

Serum bilirubin (mg/dI)

WBC count (x 1034iL)

Platelet count (x 103/s1)

8(13)

42 (66)

13 (21)

23 (2-43)

44/19

4 (6)

0

5

46

9

3

21

13

19

10

0

29 (46)

21 (33)

31 (49)

11 (17)

6(10)

18 (28)

39 (62)

35 (56)

21 (33)

7(11)

3 (2-38)

70(60-95)

0.9 (0.2-14.0)

5.4 (1.6-19.0)

170(100-442)

5 (8)

44(70)

14 (22)

24 (1-48)

47/16

4 (6)

0

6

49

5

3

24

12

11

16

0

27 (43)

24 (38)

27 (43)

12 (19)

8 (13)

17 (27)

38 (60)

34 (54)

26 (41)

3 (5)

4 (2-33)

70 (30- 100)

0.7 (0.2-39.0)

5.4 (2.2-20.0)

168(100-600)

6 (16)

22 (58)

10(26)

23 (3-47)

28/10

6 (16)

8

6

9

14

24 (63)

8 (21)

17 (45)

13 (34)

5 (13)

9 (24)

24 (63)

33 (87)

5 (13)

0

3 (3-8)

70(30-90)

1.0 (0. 1-14.0)

4.3 (1.5-14.8)

39 (6-80)

For personal use only.on October 22, 2017. by guest www.bloodjournal.orgFrom

548 SULLIVAN ET AL

were given marrow purged of T lymphocytes.’#{176} All others received

unmodified allogeneic marrow. The prophylaxis, grading, and treat-ment ofacute GVHD were reported �

Chronic GVHD had a progressive onset if it followed as a direct

extension of acute GVHD. Quiescent chronic GVHD developed

after resolution of acute GVHD, whereas de novo chronic GVHDwas not preceded by acute GVHD. Diagnosis was established upon

review of clinical, laboratory, and histologic data by previouslypublished criteria.2” Subclinical chronic GVHD was defined ashistologic evidence ofchronic GVHD on both the blind skin and oralbiopsies without signs or symptoms of clinical disease. Clinicalchronic GVHD was defined as both histologic and clinical evidence

of chronic GVHD. At study entry, all patients were in hematologicremission with donor marrow engraftment and had not receivedprior treatment for chronic GVHD. Previous GVHD prophylaxis

was discontinued upon study entry.Protocols and consent forms were approved by the Institutional

Review Board of the Fred Hutchinson Cancer Research Center. All

three groups received prednisone given in oral divided doses forweeks I and 2 and thereafter as a single oral morning dose (Table 2).

Patients in groups I and II received study drug in addition toprednisone. Study drug was assigned by random permutations of a

set of numbers known only to the protocol registrar and was suppliedas unmarked 50-mg scored tablets of either placebo or azathioprine.

Study drug was given as a single evening dose of 1.5 mg/kg/day.Patients in all three treatment groups received prophylactic

TMP-SMX. Adults received one double-strength (160 mg TMP)tablet twice daily, and children received 75 mg/m2 TMP twice daily.

Patients with life-threatening TMP-SMX allergies received prophy-lactic penicillin or cephalexin. Patients with less severe allergic

histories were rechallenged with TMP-SMX. Patients receivedsupportive care with artificial tear replacements, sun-blocking

creams, and oral caloric and protein supplements as required.

After 9 months of treatment, patients were reevaluated in Seattlewith physical examination, assessment of Karnofsky performancescore, laboratory studies (blood urea nitrogen, serum creatinine,

complete blood count, and liver function tests), pulmonary functiontests, Schirmer’s test and biomicroscopy, nutritional status, and

routine skin and oral mucosa biopsies and other biopsies as� The drug code was not broken and response totreatment was determined in a blinded manner. The followingcriteria were used to judge treatment response. Progressive disease

after 2 months of treatment or stable disease with persistingKarnofsky scores <50% after 9 months of treatment was consideredno response. The drug code was broken at study failure. Clinically

inactive chronic GVHD, but biopsies showing continued GVHD

activity with no new organ involvement after 9 months of therapy

was considered partial response. Clinically inactive disease andbiopsies showing no GVHD activity after 9 months of therapy was

Table 2. Alterna te-Day Predn isone Regimen

Prednisone

WeekofTherapy

(mg/kg/�aIly)

DayA DayS

1 1.0 1.0

2 1.0 1.0

3 2.0 0.5

4 2.0 0.25

5 2.0 0.12

6 2.0 0

7 1.5 0

8 1.25 0

9-36 1.0 0

considered complete response. When treatment was stopped after a

complete response but clinical and histologic disease activityreturned, the patient was considered to have a flare of chronic

GVHD. Therapy was reinstituted for another 9 months. For patientswith a complete response, therapy was discontinued. For patients

with a partial response, treatment continued. If chronic GVHDsymptoms were clinically active and disabling after 18 months oftreatment, patients were declared a failure.

Results were analyzed as of June 1, 1987. Survival rates wereestimated by the Kaplan-Meier method, and comparison statistics(two-sided significance levels) were calculated using the log-rankand stratified log-rank tests.’9’�#{176}

RESULTS

Entry data. As shown in Table 1 , patients randomized to

groups I and II were well matched for treatment and

prognostic factors. Twenty-four (38%) patients in group I, 25

(40%) in group II, and 14 (37%) in group III started

treatment while chronic GVHD was still subclinical. Forty-

four (70%) of these 63 patients with subclinical disease at

diagnosis developed clinical chronic GVHD during treat-

ment. Only in 19 (12%) of the 164 evaluable patients did

chronic GVHD remain subclinical throughout therapy.

Patients in group III did not appear to differ from random-

ized patients in diagnosis, age, acute GVHD prophylaxis,

pretreatment Karnofsky scores or bilirubin values, or time to

treatment. By definition, all had platelet counts <100,000/

sL. The leukocyte counts were slightly lower in group III

patients, but only two patients had leukocytes <2,000/zL at

entry. Median pretreatment bone marrow cellularity was

80% of normal in group I, 80% of normal in group II, and

50% of normal in group III.

Response. Nine months after starting therapy, 47 (29%)

of 164 evaluable patients had died or relapsed, 26 (16%) had

failed therapy, and 41 (25%) had a partial response and 50

(30%) had a complete response to treatment (Table 3).

Complete responses occurred in 21 (33%) patients ofgroup I,

23 (37%) ofgroup II, and six (16%) ofgroup III.

Long-term follow-up. Among I I 7 patients surviving in

hematologic remission who completed 9 months of therapy

(Table 3), 50 (93%) of 54 patients in group I, 40 (89%) of 45

patients in group II, and I 5 (83%) of 1 8 patients in group III

returned one or more times to Seattle for reevaluation.

Toxicity and infection. Thirty (18%) of the 164 evalu-

able patients developed one or more of the toxicities listed in

Table 3. Twenty-nine (46%) patients in group I, 39 (62%) in

group II, and 28 (74%) in group III developed one or more

infections listed in Table 3. Disseminated varicella zoster,

bacteremia, and interstitial pneumonia were more frequent

in patients randomized to receive azathioprine as compared

with placebo recipients. Patients in group III had more

frequent infections than did patients in group I who received

identical treatment.

Among 28 episodes of bacteremia, eight were owing to

pneumococcus (five occurring when patients were off TMP-

SMX), six Staphylococcus, six Pseudomonas, five Hemo-

phi/us influenzae, and three other gram-negative organisms.

Twenty episodes of interstitial pneumonia (six Pneumocystis

pneumoniae, five varicella, four idiopathic, three cytomega-

lovirus (CMV), one Legionella, and one herpes simplex)

For personal use only.on October 22, 2017. by guest www.bloodjournal.orgFrom

PREDNISONE ± AZATHIOPRINE IN CHRONIC GVHD 549

Table 3. Results of Treatment

Randomized Placed

Prednisone + Placebo Prednisone + Azathioprine Prednisone Alone

Result (Group I) (Group II) (Group Ill)

No. of evaluable patients 63 63 38

Response (%) after 9-mo therapy

Complete response 21 (33) 23 (37) 6 (16)

Partial response 18 (29) 17 (27) 6 (16)

Failed treatment 1 5 (24) 5 (8) 6 (16)

Died or Relapsed before 9 mo 9 (14) 18 (29) 20 (52)

Median (range) Karnofsky score after 9-mo therapy (%)

Complete responders 100 (90-100) 100 (80-100) 100 (90-100)

Partial responders 95 (80-100) 90 (80-100) 90 (80- 100)

Failed treatment 70 (30-95) 30 (30-90) 70 (50-90)

Toxicity of therapy

Diabetes mellitus 3 5 1

Aseptic necrosis 4 3 1

Neutrophils < 1 ,000/ML 0 4 2

Platelets < 1 00.00O/�L 1 3 -

Gastrointestinal hemorrhage 1 2 2

Severe osteoporosis 1 0 1

Psychosis 1 0 0

Episodes of infection (no. of patients, percentage of patients)

Varicella zoster (total) 18 (18, 29%) 25 (25. 40%) 17 (15, 39%)

Localized 1 1 (1 1. 17%) 10(10, 16%) 2 (2. 5%)

Disseminated 7(7. 11%) 15(15,24%) 15(13,34%)

Bacteremia 4 (4, 6%) 7 (7. 1 1%) 17 (13, 34%)

Interstitial pneumonia 3 (3. 5%) 9 (9. 14%) 8 (7, 18%)

Noninterstitial pneumonia 13 (10, 16%) 14 (10, 16%) 17 (13, 34%)

were observed. All six episodes of Pneumocystis pneumoniae

developed in 31 patients either not given or discontinuing

TMP-SMX. Forty-four episodes of noninterstitial pneumo-

nia developed, 21 owing to unknown organisms and the

remainder owing to Pneumococcus ( 1 1 ), Hemophilus

influenzae (four), fungus (four), gram-negative organisms

(two) and staphylococcus (two).

Drug dosage. Prednisone was given in the first 9 months

in full dose schedule to 141 (86%) patients; the remaining

patients had some dose modifications owing to the toxicities

listed in Table 3. Permanent reduction in prednisone dose

was observed in eight (5%) patients: three to 75% to 99%,

three to 50% to 74%, and two to 25% to 40% protocol dose.

Azathioprine was given in reduced dose in four (6%)

patients: two to 75% to 99% and two to 50% to 74% protocol

dose. Prophylactic TMP-SMX was not given or was discon-

tinued owing to side effects in nine (14%) patients in group I,

nine ( 14%) in group II, and I 3 (34%) in group III. Twelve of

these 3 1 patients received no additional antibiotics. The

remaining 19 received cephalexin (nine), penicillin (eight),

or ampicillin (two).

Clinical manifestations. Sites of disease noted in 145

patients with clinical chronic GVHD were dermal (79%),

hepatic (73%), oral (72%), ophthalmic (47%), intestinal

(16%), myofascial (11%), pulmonary (1 1%), esophageal

(6%), and serosal (2%). Intestinal chronic GVHD was more

frequent in patients in group III. Laboratory abnormalities

included rheumatoid factor reactivity (16%), hypogamma-

globulinemia (IgG < 650 mg/dL, 15%) and autoantibody

production (8%). Scleroderma with joint contractures devel-

oped in 1 1 (8%) of 145 evaluable patients with clinical

chronic GVHD.

Mortality. Twenty-three (37%) patients in group I, 33

(52%) in group II, and 28 (74%) in group III died (Table 4).

Recurrent malignancy was the most common cause of death.

Kaplan-Meier estimates of recurrent malignancy were 26%

in group I, 33% in group II, and 39% in group III (I v II,

P = .95; I v III, P = .23). Infection was the most common

cause of nonrelapse mortality. The remaining deaths were

owing to progressive GVHD without infection, hemorrhage,

or organ failure. One patient with a long pretransplant

history of paranoia refused psychiatric therapy and commit-

ted suicide.

Figure 1 shows the probability of death from nonrelapse

causes. By actuarial estimate, 21% of patients in group I,

40% in group II, and 58% in group III died of nonmalignant

causes (P = .0001). Among randomized patients, nonrelapse

mortality was significantly increased in azathioprine recip-

ients (P = .003). When group I was compared to group III,

the increase in nonrelapse mortality in patients with throm-

bocytopenia was significant (P = .0001).

Prognostic factors. Mortality did not appear related to

gender, acute GVHD prophylaxis, or time to study entry.

Table 5 lists factors in relation to mortality from all causes.

Age >20 years (P = .02), prior grade II through IV acute

GVHD (P = .03), the progressive onset of chronic GVHD

(P = .0001), and failure to respond to 9 months of treatment

(P = .04) were associated with increased mortality in log-

rank analyses stratified by treatment group.

Actuarial mortality in 19 patients in whom chronic

For personal use only.on October 22, 2017. by guest www.bloodjournal.orgFrom

550 SULLIVAN ET AL

Table 4. Causes of Death

Randomized Placed

Prednisone + Placebo Prednisone + Azathioprine Prednisone AloneCause (n-63) (n-63) (n-38)

Bacterial infection 2 1 0 6

Interstitial pneumonia

Varicella zoster 1 1 2

Cytomegalovirus 0 0 2

Idiopathic 0 2 0

Pneumocystis 0 1 0

Other Infection 1 1 0

Fungal infection 1 3 3

Progressive GVHD without infection 1 1 1

Respiratory failure without IP 1 1 1

Heart failure 0 1 1

Varicella zoster without IP 1 1 0

Hemorrhage 0 1 1

Myasthenia gravis 0 0 1

Suicide 1 0 0

Recurrent malignancy 14 10 10

Total 23 33 28

Abbreviations: GVHD, graft-versus-host disease; IP, interstitial (nonbacterial, nonfungal) pneumonia.

100 GVHD remained subclinical throughout therapy was 58% as

80 p<00001 compared with 50% mortality in 46 patients with subclinicalTh,’ornbocylopenio, p/OCed: , clinical chronic GVHD and 52% mortality in 99 patients

0 -1�� P’ednisone(n-38) with clinical chronic GVHD at diagnosis. Patients with/0 40 � Prednisone #Azothioprine(n�63) sustained subclinical chronic GVHD had higher mortality

Randomized: owing to an increased rate of relapse of leukemia. Figure 22: � ‘P,edn,sone#P/oceboj’n:63) shows the probabilityofrelapse in 146 patients with hemato-

0 1 2 3 4 5 6 7 8 9 10 11 logic malignancies in relation to the course of chronicYears after Transplant GVHD. Three-way statistical analysis showed that relapse of

malignancy was significantly increased in patients in whomFig 1 . Kaplan-Meier product limit estimates of the probability chronic GVHD remained subclinical throughout observation

of death from nonrelapse (ie. transplant-related) causes. (P = .0003).

Table 5. Mortality Risk Factors

No. of Dead/No. of Evaluable Patients

Groupl Groupll Grouplll Total(%)

Factor (n-63) (n-63) (n-38) (n- 164)

Patient age

�20yr 8/31 6/17 11/17 25/65 (38)

>20 yr 15/32 27/46 17/21 59/99 (60)

Grade acute GVHD

0 6/21 10/24 4/8 20/53 (38)

I 5/13 7/12 4/6 16/31 (52)

II 6/19 6/11 8/9 20/39 (51)

Ill 6/10 10/16 11/14 27/40(68)

IV 0/0 0/0 1/1 1/1 (100)

Type of chronic GVHD onset

Denovo 6/21 10/24 4/8 20/53 (38)

Quiescent 12/31 14/27 12/17 38/75 (51)

Progressive 5/11 9/12 12/13 26/36 (72)

Presentation of chronic GVHD

Subclinical always 3/6 4/8 4/5 1 1/1 9 (58)

Subclinical -. clinical 7/ 1 8 6/ 1 7 7/9 20/44 (45)

Clinical always 13/39 23/38 17/24 53/101 (52)

Response after 9-mo therapy

Complete response 2/21 7/23 1/6 10/50 (20)

Partialresponse 5/18 4/17 4/6 13/41 (31)

Failed treatment 7/1 5 4/5 3/6 14/26 (54)

Died or relapsed before 9 mo 9 1 8 20 47

For personal use only.on October 22, 2017. by guest www.bloodjournal.orgFrom

600J‘0

40

20

C

100

80

60

40

20

PREDNISONE ± AZATHIOPRINE IN CHRONIC GVHD 551

100

80 p(O.0003

� CIWOyS (n�/9)

(C//fl/CO/ 0/wayS (n�85)

Years after Transplant

Fig 2. Kaplan-Meier product limit estimates of the probabilityof relapse of malignancy in 1 9 patients with subclinical -‘ subclini-cal chronic GVHD. 42 patients with subclinical -‘ clinical chronic

GVHD. and 85 patients with clinical chronic GVHD. Eighteen of the164 evaluable patients (16 with clinical and 2 with subclinical

clinical chronic GVHD) receiving transplants for nonmalignantdisorders were excluded from this analysis.

Status ofsurvivors. Forty (63%) patients in group I, 30

(48%) patients in group II, and 10 (26%) patients in group

III survive (Table 6). Minimum follow-up is 45 months after

transplant. Nine (14%) patients with complete response in

group I, six (10%) in group II, and two (5%) in group III had

flares of chronic GVHD a median of 2 months (range one to

1 1 months) after stopping therapy and were retreated.

Primary sites of flare included skin (ten), liver (four), mouth

(one), eye (one), and muscle (one). Currently, 65 (81%) of

the survivors continue off treatment and are free of chronic

GVHD. The Karnofsky performance scores of the 80 surviv-

ing patients are shown in Table 6.

Figure 3 shows the actuarial survival. Survival estimates

were 61% in group I, 47% in group II, and 26% in group III

(P = .001 ). When randomized patients in groups I and II

were compared, survival was reduced in azathioprine recip-

ients (P = .03). When groups I and III were compared,

survival was impaired in patients with thrombocytopenia

(P= .0001).

DISCUSSION

The well-being of the marrow graft recipient may depend

on events developing long after the patient leaves the trans-

plant center. Collaboration between the primary care physi-

RondomizedPrednisone # Placebo (n�63)

Randomized

Prednisone #Aza/fliopnne (n:63)

rh(ombocy/openio, placedp<o.oo1 Prednisone (n�38)

C0 1 2 3 4 5 6 7 8 9 10 11

Years after Transplant

Fig 3. Kaplan-Meier product limit estimates of the probabilityof survival among 63 evaluable patients in group I (predni-sone + placebo). 63 evaluable patients in group II (predni-

sone + azathioprine), and 38 patients in group Ill (prednisone). Ticmarks represent 40 group I, 30 group II, and 10 group Ill patientssurviving as of June 1 #{149}1 987. Minimum follow-up is 3.8 years after

transplant.

cian and investigators at the center assumes increasing

importance as the success and use of marrow transplantation

increases.2’ This collaboration is vital in conducting long-

term, double-blind clinical trials such as we describe. We

found that with appropriate education and responsiveness,

physician and patient compliance was enthusiastic and only

15 (8%) of 179 patients enrolled in the study were not

evaluable. Moreover, although 80% of our transplant recip-

ients reside outside the Pacific Northwest, I 05 (90%) of I I 7

patients completing 9 months of protocol therapy returned to

Seattle for long-term follow-up evaluation.

With early treatment, runting, scleroderma, and joint

contractures are now uncommon developments. Indeed, most

patients have normal Karnofsky scores, and only five (6%) of

80 survivors function at <70% performance. With improved

disability-free survival, new manifestations of chronic

GVHD, such as obliterative bronchiolitis2225 and myasthenia

gravis,26’27 may be observed. Nevertheless, with minimum

follow-up of 3.8 years and with 65 (81%) of 80 survivors

currently off all treatment, the present study appears suffi-

ciently mature to detect most causes of disability.28

The double-blind, placebo-controlled design allowed

assessment of the role of azathioprine in early treatment of

chronic GVHD. Prior clinical2 and experimental29 studies

Table 6. Status of Survivi ng Patients

5tatus

Randomized Placed

Prednisone Alone(n - 38)

Prednisone + Placebo

(n - 63)Prednisone + Azathioprine

(n - 63)

No. of surviving patients

Median (range) follow-up

Months after transplant

Months after study entry

Treatment characteristics

Median (range) duration of therapy (mo)

No. of patients still on therapy (%)

No. of patients stopping therapy

No. of patients still off therapy

Median (range) duration off therapy (mo)

Current Karnofsky score

100%-90%

89%-70%

<70%

40

68 (48-125)

63 (42-88)

24 (9-72)

8 (20)

35

32

48 (3-78)

34

4

2 (5%)

30

72 (51-95)

64 (44-90)

22 (9-90)

6 (20)

28

24

46 (2 1 -74)

26

3

1 (3%)

10

66 (45-89)

6 1 (42-85)

24 (9-57)

1 (10)

10

9

33 (2-64)

8

0

2 (20%)

For personal use only.on October 22, 2017. by guest www.bloodjournal.orgFrom

552 SULLIVAN ET AL

suggested that azathioprine has activity against GVHD. The

current trial is the first randomized study of chronic GVHD,

and with its design the effect of immunodeficiency owing to

uncontrolled GVHD and immunodeficiency owing to cyto-

toxic therapy could be dissociated and analyzed. Surprising-

ly, with early treatment the control ofGVHD was equivalent

in recipients of prednisone alone and prednisone plus aza-

thioprine. However, viral and bacterial infections were more

frequent with combination treatment. The nonrelapse mor-

tality was significantly increased in standard-risk patients

receiving additional azathioprine (21% in group I v 40% in

group II). Interstitial pneumonia appeared to be increased30

but hepatic abnormalities3’ and significant myelosuppression

were not more common in azathioprine recipients.

The study design also allowed assessment of the contribu-

tion of cytotoxic therapy in preventing recurrent malignancy

after transplant. The relapse rates did not differ significantly

in placebo and azathioprine recipients. In previous studies,

the contribution of chronic GVHD to an antileukemic effect

had been less clear since patients with more severe chronic

GVHD were usually treated with azathioprine.32’33 The pres-

ent study fails to show an antileukemic effect of azathioprine

treatment.

Although some experimental studies suggest that graft-

v-leukemia (GVL) reactivity can be obtained without

augmenting GVHD,34’35 other studies question this dissocia-

tion.36 The current clinical trial does not support the sugges-

tion that GVHD can be separated from the GVL effect.

Patients with chronic GVHD remaining subclinical through-

out treatment had a 55% probability of relapse as compared

with 28% and 25% rates of relapse, respectively, in patients

with clinical or subclinical �- clinical chronic GVHD

(P .0003). This increased recurrence of leukemia nullified

any survival advantage that might be gained in treating

subclinical disease.

Infection remains a distressingly frequent complication of

chronic GVHD. Patients with thrombocytopenia have a poor

response to therapy and an increased number of infections

even without cytotoxic treatment. This dynamic relation of

chronic GVHD to opportunistic infection and recovery of

immune function is a complex, interrelated, and time-

dependent process.37 Numerous reports demonstrate multi-

plc B and T cell defects in patients with chronic GVHD.3”2

Patients with chronic GVHD often develop bacterial and

. interstitial pneumonias and bacteremia.43’” Infection with

encapsulated organisms such as Pneumococcus and Hemo-

phi/us influenzae are especially frequent.45 Use of prophy-

lactic TMP-SMX reduces the incidence of late interstitial

pneumonia in patients with chronic GVHD. In a previous

report of patients with chronic GVHD, we found a 28%

actuarial incidence of late interstitial pneumonia in patients

treated with prednisone without TMP-SMX v I 5% in

untreated patients (ie, no immunosuppression and no antibi-

otics) v 8% in patients treated with prednisone and TMP-

SMX (P = .001).” Because patients with chronic GVHD

demonstrate impaired immunoglobulin regulation and a

pattern of infection resembling that associated with the

primary immunodeficiencies, long-term administration of

intravenous (IV) immune globulin may be another method to

prevent infection.�’ Controlled studies are currently in prog-

ress to test this hypothesis. Finally, prompt recognition and

treatment of varicella zoster can also lower mortality.

Because of the seriousness of varicella in these patients,

prompt treatment with IV acyclovir appears to be fully

warranted.47

Prognostic factors in this study included patient age, acute

GVHD, progressive onset of chronic GVHD, failure to

respond to 9 months of therapy, and thrombocytopenia. The

reason for increased mortality in patients with thrombocyto-

penia is poorly understood” Pretreatment thrombocytope-

nia was associated with impaired marrow function, as cvi-

denced by a median megakaryocyte cellularity of 50% of

normal (compared to 80% median megakaryocyte cellularity

in patients with normal platelet counts). Viral-induced or

cell-mediated49 suppression of megakaryocytopoiesis could

account for persisting thrombocytopenia. Recent in vitro

studies suggest immune-mediated defects in hematopoietic

progenitor cell function in patients with chronic GVHD.5#{176}

The use of cyclosporine in this setting of thrombocytopenia

appears to be of value and is the subject of another report.5

We conclude that in patients with chronic GVHD with

normal marrow function, prednisone led to fewer infections

and better survival than therapy with prednisone and aza-

thioprine. Early treatment abated the disabling natural

history of clinical extensive chronic GVHD. Patients with

subclinical GVHD did not benefit from therapy due to an

increased rate of recurrent leukemia. Persisting thrombocy-

topenia identified a group of patients with poor response to

prednisone and frequent, often fatal, infection. Improved

supportive care and treatment are needed for high-risk

chronic GVHD patients who have poor marrow function or

who fail initial therapy.

ACKNOWLEDGMENT

We are indebted to referring physicians for management of the

patients as well as inpatient and outpatient physicians, nurses, andsupport staff of the Fred Hutchinson Cancer Research Center andSwedish Hospital Medical Center. We appreciate the support ofJack Pelkey and Burroughs Wellcome, Research Triangle Park,

NC, for their support and supply of the study drug. We especiallythank Deborah Gayle and Marianne Hansen for superb assistance inlong-term follow-up.

REFERENCES

1 . Shulman HM, Sullivan KM. Weiden PL, McDonald GB,Striker GE, Sale GE, Hackman R, Tsoi M-S, Storb R, Thomas ED:

Chronic graft-versus-host syndrome in man. A long-term clinico-pathological study of 20 Seattle patients. Am J Med 69:204, 1980

2. Sullivan KM, Shulman HM, Storb R, Weiden PL, Wither-spoon RP, McDonald GB, Schubert MM, Atkinson K, Thomas, ED:

Chronic graft-versus-host disease in 52 patients: Adverse natural

course and successful treatment with combination immunosuppres-sion. Blood 57:267, 1981

3. Sullivan KM. Deeg Hi, Sanders JE, Shulman HM, Wither-spoon RP, Doney K, Appelbaum FR, Schubert MM, Stewart P.Springmeyer 5, McDonald GB, Storb R, Thomas ED: Late compli-

cations after marrow transplantation. Semin Hematol 21:53, 19844. Sullivan KM. Storb R, Flournoy N, Weiden P. Shulman H,

For personal use only.on October 22, 2017. by guest www.bloodjournal.orgFrom

PREDNISONE ± AZATHIOPRINE IN CHRONIC GVHD 553

Deeg Hi, Witherspoon, Thomas ED: Preliminary analysis of a

randomized trial of immunosuppressive therapy of chronic graft-vshost disease (GVHD). Blood 60:173a, 1982 (suppl) (abstr)

5. Sullivan KM, Witherspoon RP, Storb R, Deeg HJ, Dahlberg

5, Sanders iE, Appelbaum FR, Doney KC, Weiden P. Anasetti C,Loughran T, Hill R, Shields A, Yee G, Shulman H, Nims J, Strom

5, Thomas ED: Alternating-day cyclosporine and prednisone for

treatment of high-risk chronic graft-v-host disease. Blood 72:555,

1988

6. Storb R, Thomas ED, Buckner CD, Appelbaum FR, Clift RA,Deeg Hi, Doney K, Hansen JA, Prentice RL, Sanders iE, StewartP, Sullivan KM. Witherspoon RP: Marrow transplantation for

aplastic anemia. Semin Hematol 21:27, 19847. Thomas ED, Buckner CD, Banaji M, Clift RA, Fefer A,

Flournoy N, Goodell BW, Hickman RO, Lerner KG, Neiman PE,Sale GE, Sanders iE, Singer i, Stevens M, Storb R, Weiden PL:

One hundred patients with acute leukemia treated by chemotherapy,total body irradiation, and allogeneic marrow transplantation. Blood

49:511, 1977

8. Thomas ED: Marrow transplantation for malignant diseases(Karnofsky Memorial Lecture). J Clin Oncol 1:517, 1983

9. Beatty PG. Clift RA, Mickelson EM, Nisperos B, Flournoy N,

Martin PJ, Sanders JE, Stewart P, Buckner CD, Storb R, Thomas

ED, Hansen iA: Marrow transplantation from related donors otherthan HLA-identical siblings. N Engl J Med 313:765, 1985

10. Martin Pi, Hansen JA, Buckner CD, Sanders JE, Deeg HJ,

Stewart P, Appelbaum FR, Clift R, Fefer A, Witherspoon RP,Kennedy MS. Sullivan KM, Flournoy N, Storb R, Thomas ED:

Effects of in vitro depletion of T cells in HLA-identical allogeneicmarrow grafts. Blood 66:664, 1985

1 1. Glucksberg H, Storb R, Fefer A, Buckner CD, Neiman PE,Clift RA, Lerner KG, Thomas ED: Clinical manifestations ofgraft-versus-host disease in human recipients of marrow from HLA-

matched sibling donors. Transplantation 18:295, 197412. Thomas ED, Storb R, Clift RA, Fefer A, iohnson FL,

Neiman PE, Lerner KG, Glucksberg H, Buckner CD: Bone-marrowtransplantation. N EngI i Med 292:832, 895, 1975

13. Deeg Hi, Storb R, Thomas ED, Flournoy N, Kennedy MS.Banaji M, Appelbaum FR, Bensinger WI, Buckner CD, Clift RA,Doney K, Fefer A, McGuffin R, Sanders iE, Singer i, Stewart P,

Sullivan KM. Witherspoon RP: Cyclosporine as prophylaxis forgraft-versus-host disease: A randomized study in patients undergo-ing marrow transplantation for acute nonlymphoblastic leukemia.

Bbood6S:1325, 198514. Doney KC, Weiden PL, Storb R, Thomas ED: Treatment of

graft-versus-host disease in human allogeneic marrow graft recip-ients: A randomized trial comparing antithymocyte globulin and

corticosteroids. Am J Hematol I I : 1 , 1981

15. Storb R, Deeg Hi, Whitehead i, Appelbaum F, Beatty P,Bensinger W, Buckner CD, Clift R, Doney K, Farewell V, Hansen i,

Hill R, Lum L, Martin P, McGuffin R, Sanders J, Stewart P.

Sullivan K, Witherspoon R, Yee G, Thomas ED: Methotrexate andcyclosporine compared with cyclosporine alone for prophylaxis of

acute graft versus host disease after marrow transplantation forleukemia. N EngI J Med 314:729, 1986

16. Sullivan KM: Graft-versus-host disease, in Blume KG, Petz

LD (ed): Clinical Bone Marrow Transplantation. New York,Churchill Livingstone, 1983, pp 91-129

17. Schubert MM, Sullivan KM, Morton TH, Izutsu KT, Peter-son DE, Flournoy N, Truelove EL, Sale GE, Buckner CD, Storb R,Thomas ED: Oral manifestations ofchronic graft-host disease. Arch

Intern Med 144:1591, 198418. McDonald GB, Shulman HM, Sullivan KM. Spencer GD:

Intestinal and hepatic complications of human bone marrow trans-plantation (Parts I and II). Gastroenterology 90: pp 460 and 770,I 986

1 9. Kaplan EL, Meier P: Nonparametric estimation from incom-

plete observations. J Am Stat Assoc 53:457, 195820. Mantel, N: Evaluation of survival data and two new rank

order statistics arising in its consideration. Cancer Chemother Rep50:163, 1966

21. Bortin MM, Rimm AA: Increasing utilization of bone mar-row transplantation. Transplantation 42:229, 1986

22. Ralph DD, Springmeyer SC, Sullivan KM. Hackman RC,Storb R, Thomas ED: Rapidly progressive air-flow obstruction inmarrow transplant recipients: Possible association between oblitera-

tive bronchiolitis and chronic graft-versus-host disease. Am RevRespir Dis 129:641, 1984

23. Kurzrock R, Zander A, Kanojia M, Vellekoop L, Spitzer G,iagannath 5, Schell S, Peters L, Dicke K: Obstructive lung disease

after allogeneic bone marrow transplantation. Transplantation37:156, 1984

24. Rosenberg ME, Vercellotti GM, Snover DC, Hurd D,McGlave P: Bronchiolitis obliterans after bone marrow transplanta-

tion. Am J Hematol 18:325, 198525. Clark iG, Schwartz DA, Flournoy N, Sullivan KM. Craw-

ford SW, Thomas ED: Risk factors for airflow obstruction inrecipients of bone marrow transplants. Ann Intern Med 107:648,

1987

26. Smith CIE, Aarli iA, Biberfeld P. Bolme P. Christensson B,Gahrton G, Hammarstrom L, Lefvert A-K, Lonnqvist B, Matell G,

Pirskanen R, Ringden 0, Svanborg E: Myasthenia gravis afterbone-marrow transplantation. Evidence for a donor origin. N EngI i

Med 309:1565, 1983

27. Bolger GB, Sullivan KM, Spence AM, Appelbaum FR,

iohnston R, Sanders JE, Deeg Hi, Witherspoon RP, Doney KC,Nims i, Thomas ED, Storb R: Myasthenia gravis after allogenic

bone marrow transplantation: Relationship to chronic graft-versus-host disease. Neurology 36:1087, 1986

28. Witherspoon RP, Buckner CD, Sullivan KM, Fisher L, Clarki, Storb R, Thomas ED: Causes of late failure in patients trans-

planted for acute leukemia. Bone Marrow Transplant 2:92, 1987(suppl 1)

29. Shand FL: Attenuation of murine graft-versus-host reactivityby azathioprine. Transplantation 30:55, 1980

30. Bedrossian CWM, Sussman i, Conklin RH, Kahan B: Aza-thioprine-associated interstitial pneumonitis. Am i Clin Pathol

82:148, 1984

3 1 . Loiseau D, Degos F, Degott C, Carnot F, Kreis H: Cholesta-

sis after azathioprine administration in renal transplant recipients:Report of 7 cases. Clin Transplant 1:88, 1987

32. Weiden PL, Sullivan KM, Flournoy N, Storb R, Thomas ED,the Seattle Marrow Transplant Team: Antileukemic effect of

chronic graft-versus-host disease. Contribution to improved survivalafter allogeneic marrow transplantation. N EngI i Med 304:1529,1981

33. Sullivan KM. Fefer A, Witherspoon R, Storb R, BucknerCD, Weiden P. Schoch G, Thomas ED: Graft-versus-leukemia in

man: Relationship of acute and chronic graft-versus-host disease torelapse of acute leukemia following allogeneic bone marrow trans-plantation, in Truitt RL, Gale RP, Bortin MM (eds): CellularImmunotherapy of Cancer. Liss, New York, 1987, pp 39 1-399

34. Bortin MM, Truitt RL, Rimm AA, Bach FH: Graft-versus-leukaemia reactivity induced by alloimmunisation without augmen-

tation of graft-versus-host reactivity. Nature 28 1:490, 197935. Moscovitch M, Slavin 5: Anti-tumor effects of allogeneic

bone marrow transplantation in (NZB X NZW) F1 hybrids withspontaneous lymphosarcoma. i Immunol 132:997, 1984

36. Tutschka PH, Berkowitz SD, Tuttle 5, Klein i: Graft-versus-leukemia in the rat-The antileukemic efficacy of syngeneic

and allogeneic graft-versus-host disease. Transplant Proc 19:2668,

1987

For personal use only.on October 22, 2017. by guest www.bloodjournal.orgFrom

554 SULLIVAN ET AL

37. Sullivan KM: Acute and chronic graft-versus-host disease in

man. Int J Cell Cloning 4:42, 1986 (suppl 1)

38. Noel DR, Witherspoon RP, Storb R, Atkinson K, Doney K,Mickelson EM, Ochs HD, Warren RP, Weiden PL, Thomas ED:

Does graft-versus-host disease influence the tempo of immunologicrecovery after allogeneic human marrow transplantation? An obser-

vation on 56 long-term survivors. Blood 51:1087, 1978

39. Tsoi M-S, Storb R, Dobbs 5, Kopecky KJ, Santos E, Weiden

PL, Thomas ED: Nonspecific suppressor cells in patients with

chronic graft-vs-host disease after marrow grafting. J Immunol123:1970, 1979

40. Witherspoon RP, Storb R, Ochs HD, Flournoy N, Kopecky

Ki, Sullivan KM, Deeg Hi, Sosa R, Noel DR. Atkinson K, Thomas

ED: Recovery of antibody production in human allogeneic marrowgraft recipients: Influence of time posttransplantation, the presence

or absence of chronic graft-versus-host disease, and antithymocyteglobulin treatment. Blood 58:360, 1981

41. Saxon A, McIntyre RE, Stevens RH, Gale RP: Lymphocytedysfunction in chronic graft-versus-host disease. Blood 58:746,

1981

42. Lum LG: The kinetics of immune reconstitution after human

marrow transplantation. Blood 69:369, 198743. Atkinson K, Farewell V, Storb R, Tsoi M-S, Sullivan KM,

Witherspoon RP, Fefer A, Clift R, Goodell B, Thomas ED: Analysis

of late infections after human bone marrow transplantation: Role of

genotypic nonidentity between marrow donor and recipient and of

nonspecific suppressor cells in patients with chronic graft-versus-host disease. Blood 60:714, 1982

44. Sullivan KM. Meyers JD, Flournoy N, Storb R, Thomas ED:

Early and late interstitial pneumonia following human bone marrowtransplantation. Int i Cell Cloning 4:107, 1986 (suppl 1)

45. Winston DJ, Schiffman G, Wang DC, Feig SA, Lin C-H,

Marso EL, Ho WG, Young LS, Gale RP: Pneumococcal infectionsafter human bone-marrow transplantation. Ann Intern Med 91:835,

1979

46. Sullivan KM: Immunoglobulin therapy in bone marrow

transplantation. Am J Med 83:34, 1987

47. Locksley RM, Flournoy N, Sullivan KM, Meyers iD: Infec-tion with varicella-zoster virus after marrow transplantation. i

InfectDis 152:1172, 1985

48. First LR, Smith BR, Lipton J, Nathan DG, Parkman R,Rappeport JM: Isolated thrombocytopenia after allogeneic bonemarrow transplantation: Existence of transient and chronic throm-bocytopenic syndromes. Blood 65:368, 1985

49. Gerwirtz AM, Sacchetti MK, Bien R, Barry WE: Cell-mediated suppression of megakaryocytopoiesis in acquired amega-

karyocytic thrombocytopenic purpura. Blood 68:619, 1986

50. Atkinson K, Norrie 5, Chan P. Zehnwirth B, Downs K, Biggs

i: Hematopoietic progenitor cell function after HLA-identical sib-ling bone marrow transplantation: Influence of chronic graft-

versus-host disease. Int J Cell Cloning 4:203, 1986

For personal use only.on October 22, 2017. by guest www.bloodjournal.orgFrom

1988 72: 546-554  

Doney and FR AppelbaumKM Sullivan, RP Witherspoon, R Storb, P Weiden, N Flournoy, S Dahlberg, HJ Deeg, JE Sanders, KC prolonged thrombocytopenia after allogeneic marrow transplantationtreatment of chronic graft-v-host disease: prognostic influence of Prednisone and azathioprine compared with prednisone and placebo for 

http://www.bloodjournal.org/content/72/2/546.full.htmlUpdated information and services can be found at:

Articles on similar topics can be found in the following Blood collections

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

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

http://www.bloodjournal.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.Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American Society of

For personal use only.on October 22, 2017. by guest www.bloodjournal.orgFrom