1. No category

From www.bloodjournal.org by guest on February 11, 2015. For personal use only.
Effect of Low-Dose Interleukin-2 on Disease Relapse After T-Cell-Depleted
Allogeneic Bone Marrow Transplantation
By Robert J. Soiffer, Christine Murray, Rene Gonin, and Jerome Ritz
T-cell depletion of donor bone marrow has been associated
with an increased risk of dwase relapse after allogeneic
bone marrow transplantation (BMT). Recombinant interlwkin-2 (IL-21, which is capable of increasing the antileukemic activity of peripheral blood lymphocytes obtained
from patients who have undergoneBMT, hasbeen proposed
as a potentially useful agent t o reduce the risk of relapse
post-BMT. We have previously shown that IL-2administered
t o patients at very low doses after BMT is both clinically
tolerable and immunologically active. We now report on the
clinical outcome of 29 patients treated with low-dose IL-2
after CD6-depleted allogeneic BMT for hematologic malignancy. IL-2 was administered by continuous infusion for up
to 3 months beginning at a median of 67 days post-BMT.
Eligibility requirementsfor IL-2 therapy included demonstration of stable engraftment and absence of acute grade 2-4
graft-versus-host disease (GVHD). Low-dose IL-2 was well
tolerated by themajority of patients, with only 4 of 29 subjects withdrawn early. Acute GVHD developed in only one
individual. After 12 weeks of treatment, the mean number
of circulating natural killer cells in patients increased IO-fold
without any significant change in T-cellnumber. Of the
25 patients who received a1 month of IL-2. only 6 have
relapsed. Relapserate and direase-free survival (DFS) were
determined in the 25 patients who completed at least 4
weeks of IL-2 treatment and comparedwith historical controls transplanted at our institution forthe same conditions
and treated with an identical ablative regimen and method
of T-cell depletion. Onlycontrol patients who had survived
disease free for 100 days post-BMT were included in this
analysis. Cox’s proportional hazards regression model suggested that, comparedwith control patients without a history of GVHD, patients treated with IL-2 had a lower risk of
disease relapse (hazard ratio 0.34; range, 0.14 t o 0.82) and
superior DFS (hazard ratio 0.39; range 0.18 to 0.87). A randomized controllod trial of 11-2 immunotherapy after T-celldepleted BMT should now be undertaken.
0 1994 by The American Societyof Hematology.
T
mononuclear cells (PBMC) obtained from patients after
BMT can develop the capacity to lyse cryopreserved autologous leukemic blasts upon exposure to IL-2 in
Moreover, in patients with CML who have undergone a Tcell-depleted allogeneic transplant, the cytolytic activity of
IL-2-stimulated PBMC against leukemic targets has been
correlated with the likelihood of subsequent disease relapse.” The cytotoxic activity of these activated PBMC has
been found to reside predominantly in cells with the immunophenotypic characteristics of natural killer (NK) cell^.^^"^
In a preliminary report, we were able to show that in vivo
administration of very low doses of IL-2 for prolonged periods can selectively expand and activate NK cells without
significantly affecting T cells in patients who have undergone T-cell-depleted allogeneic BMT.I6 Because activated
NK cells from patients post-BMT have been found to exhibit
antileukemic activity,17this approach provides a strategy that
may help compensate for the loss of GVL activity associated
with the removal of alloreactive T cells from donor marrow.
We now report on 29 patients with hematologic malignancies who have been treated with low-dose IL-2 after T-celldepleted allogeneic BMT. We find that daily infusion of IL2 at doses of 2 to 4 X lo5U/m2/d can be tolerated for up to
3 months bymost patients. Low-dose L - 2 treatment can
activate PBMC in these patients and can markedly expand
the number of circulating NK cells. Moreover, multivariable
analyses comparing these patients with historical controls
transplanted at our institution for the same diseases suggest
that IL-2 therapy lowers relapse rates and improves DFS,
and thus, may restore a component of the GVL activity lost
with donor marrow T-cell depletion.
-CELL DEPLETION of donor bone marrow (BM) decreases the incidence of acute graft-versus-host disease
(GVHD) in patients undergoing HLA identical allogeneic
BM transplantation (BMT) for hematologic malignancies.’,2
However, T-cell depletion has been associated with an increased incidence of disease relapse and, as a result, has not
significantly improved disease-free survival (DFS) posttrans~ l a n t .There
~ . ~ is considerable evidence that links the development of GVHD to lower relapse rates POS~-BMT,’-~
and the
increase in disease recurrence after T-cell-depleted BMT
appears to be related, at least in part, to the reduced incidence
of GVHD. This phenomenon has been termed the graftversus-leukemia (GVL) effect.’ In animal systems, it has
been possible to induce GVL activity in the absence of
GVHD.9”1However, the cellular and humoral mechanisms
underlying the GVL effect in humans have not been precisely
elucidated, and efforts to separate GVL activity from GVHD
and its associated toxic sequellae have not been successful
in clinical trials.”
One potential approach to the restoration of GVL activity
after T-cell-depleted allogeneic BMT involves the administration of interleukin-2 (IL-2) in vivo to stimulate immune
reactivity. It has previously been shown that peripheral blood
From the Divisions of Hematologic Malignancies and Biostatistics, Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA.
Submitted January 25, 1994; accepted April 6, 1994.
Supported by NlH Grants No. A129530 and CA41619. R.J.S. is
a recipient of the Baruj Benacerraf Clinical Investigator Award.
Address reprint requests to Robert J. Soiffer, MD, Dana-Farber
Cancer Institute, 44 Binney St, Boston, MA 02115.
The publication costs of this 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 1734 solely to
indicate this fact.
0 1994 by The American Society of Hematology.
0006-4971/94/8403-05$3.00/0
964
MATERIALS AND METHODS
Patient eligibility. Twenty-nine patients who underwent CD6depleted allogeneic BMT at the Dana-Farber Cancer Institute (DFCI)
since 1990 were treated with IL-2 in the posttransplant period. Patients were eligible to begin L 2 therapy after at least 6 weeks had
elapsed after marrow infusion. Initially, only patients transplanted
Blood, Vol 84, No 3 (August l ) , 1994: pp 964-971
From www.bloodjournal.org by guest on February 11, 2015. For personal use only.
LOW-DOSE IL-2 AFTERT-CELL-DEPLETED
BMT
with a history of relapsed acute leukemia, myelodysplastic syndrome
(MDS), non-Hodgkin’s lymphoma (NHL), or accelerated phase
chronic myelogenous leukemia (CML) were eligible for this study.
After the relative safety of IL-2 infusion was established, inclusion
criteria were broadened to include patients transplanted for acute
leukemia in first remission and CML in stable phase. Pretreatment
evaluation in all patients included a physical exam, complete blood
count, serum chemistries, liver function tests, urinalysis, thyroid
function tests, chest x ray, electrocardiogram, and pulmonary function tests. Eligibility criteria at the time of study entry included
good performance status (ECOG 0-1) and near normal parameters
of hepatic, liver, and pulmonary function. Absolute neutrophil count
2500 X 106 celldpL with a platelet count 2 40,OOO/pL independent
of transfusion were required. All patients were required to be free
of active infection at the time of protocol entry. All patients with
evidence of active grades 2-4 GVHD were excluded from study, as
were any patients receiving immune suppressive medications for
GVHD or other indications. Written informed consent was obtained
in all cases. The treatment protocol was approved by the Scientific
Review and Human Subject’s Protection Committees at DFCI.
During the period between January 1990, when studies were begun, and February 1993, a total of 81 patients were transplanted for
acute lymphoblastic leukemia (ALL) acute myeloid leukemia
(AML), CML, MDS, or NHL at our institution. During this time,
29 patients were treated with L-2. Of the 52 patients who did not
receive IL-2 post-BMT, 14 were not eligible for treatment because
they had grade 2-4 acute GVHD, 14 refused for logistical reasons,
8 were patients with early-stage acute leukemia or stable-phase CML
who were transplanted when the treatment protocol was limited to
patients with advanced disease, 6 had experienced early-disease relapse (before day +60), 6 had suffered toxic deaths (before day
+60), 2 had poor performance status, and 2 had experienced graft
failure.
BMTprotocol. The transplant ablative regimen consisted of cyclophosphamide (60 mgkg intravenous [IV]) on two consecutive
days followed by 1,400 cGy total body irradiation delivered in seven
equal 200 cGy fractions over 4 days. One patient who had received
prior radiotherapy was instead treated with busulfan (16 mgkg divided over 4 days) followed by cyclophosphamide (60 mgkg IV X
2). Twenty-eight patients who were treated with L - 2 had received
bone marrow from an HLA-identical sibling donor. One patient
received sibling marrow that was serologically mismatched at a
single HLA locus. BM was collected by standard techniques. BM
mononuclear cells were isolated and treated with anti-T12 monoclonal antibody (MoAb) (anti-CD6) and baby rabbit complement
(Pel Freeze, Brown Deer, WI)as previously described.’*No patients
received immunosuppressive therapy of any type for GVHD prophylaxis, including corticosteroids, methotrexate, or cyclosporine. After
discharge from the hospital, patients routinely received multivitamins, folic acid supplementation, oral acyclovir for herpes simplex
virudvaricella zoster virus (HSVNZV) prophylaxis, and either oral
trimethoprim-sulfamethoxazole or aerosolized pentamidine to prevent Pneumocystis carinii pneumonia.
IL-2 therapy. All patients received L - 2 by continuous intravenous infusion through an indwelling central catheter. Drug was delivered by a portable computerized ambulatory pump (Pharmaciaetec
Model 5100 HF; Pharmaciaetec, StPaul, MN). The treatment was
completely performed on an outpatient basis with one clinic visit
each week. The supply of L - 2 was renewed every 7 days by the
outpatient pharmacist. Recombinant L 2 was provided initially by
Hoffmann-LaRoche (Nutley, NJ; 24 patients) and, more recently by
Amgen, Inc (Thousand Oaks,CA; 5 patients). Because some patients
were treated as part of a phase 1 protocol, the starting dose varied
from 2 X 1 6 U/m’/d to 6 X l6 U/mz/d. Planned duration of therapy
was 3 months. Patients were not routinely treated with any prophy-
965
lactic antipyretic or anti-inflammatory agents while on study. If patients had difficulty tolerating the initial dose level, therapy was
temporarily interrupted and then resumed at a reduced dose level
when symptoms abated. Doses were not escalated above the starting
dose during the course of treatment.
Immunophenotypic studies. PBMC for immunologic studies
were obtained weekly. Blood was collected in preservative-free heparin. PBMC were obtained after Ficoll-Hypaque (Pharmacia, Uppsala, Sweden) density gradient sedimentation. PBMC were analyzed
by direct immunofluorescence for reactivity with a series of MoAbs
using standard techniques. Cells were analyzed for reactivity with a
panel of MoAbs, including T3 (CD3). T4 (CD4). T8 (CDS), NKHl
(CD56). and Tac (CD25) (Coulter Immunology, Hialeah, FL). Immunofluorescence reactivity was determined by automated flow cytometry analyzing 104 cells in each sample (ELITE, Coulter Electronics, Hialeah, FL).
Statistical analysis. The effects of IL-2 therapy upon risk of
disease relapse and DFS post-BMT were evaluated in the 25 patients
who completed at least 4 consecutive weeks of treatment. Comparison was made to historical controls transplanted at our institution
for the same diseases and treated with a similar ablative regimen
and method of T-cell depletion. Patients who died or relapsed S
100 days post-BMT were not included in the control population
because these patients would have been unlikely to have been eligible for IL-2 treatment. There were 92 patients who satisfied the
above criteria. These 92 patients were then classified into two groups,
depending on whether they did (n = 23) or did not (n = 69) have
grade 2-4 GVHD post-BMT. Data were analyzed through September
1, 1993. Descriptive statistics are reported as proportions, medians,
and means. Fisher’s exact test’’ was used to compare proportions
and the Kruskal-Wallis nonparametric analysis of variance testz0was
used to compare the median follow-up time among the groups. Time
to relapse (risk) and DFS curves for each subgroup were constructed
by means of the Kaplan-Meier product limit method.” The log-rank
test” of survival analysis was used tocompare the various subgroups
with respect to their time to relapse (risk) and DFS distributions in
the univariate analyses. In a multivariable analysis, Cox’s proportional hazards regression modelz3 wasused to determine possible
predictors of relapse and DFS. The SAS procedure PHREGz4with
backward selection (5% significance level for removing a predictor)
was used throughout. A test for proportional hazards using timedependent covariates was conducted with respect to time to relapse
(risk) and DFS. The proportional hazards assumption was not violated in both instances. The hazards ratios of each predictor (having
adjusted for the effects of the other significant predictors) and 95%
confidence intervals (CIS)for the hazards ratios were also calculated.
All tests were two tailed.
RESULTS
Patient characteristics. There
were
19
men
and
10
41 years
women who received IL-2. =median age was
(range, 21 to 59 years). CML was the most common diagnosis forwhichpatientsunderwent BMT (n = 15), followed
by AML (n = 7),ALL(n = 3),MDS (n = 2),and NHL(n
= 2). Twelve patientsweretransplantedforeitheracute
leukemia in first remission or CML in stable phase and were
considered to have been transplanted at an “early” stage of
their disease. The remaining 17 patients underwent marrow
transplantation for more advanced disease. IL-2 was begun
at a median of 67 days after marrow infusion (interquartile
range, day +48 to day + l a ) . The median white blood cell
count at the initiationof IL-2 was4.2 X 106/L and the median
platelet count was 121 X 109/L.Six patients began treatment
From www.bloodjournal.org by guest on February 11, 2015. For personal use only.
966
SOIFFERETAL
with 1L-2 at a dose of 2 X 10' U/m2/d, 9 patients at 3 X IO'
U/m2/d, 1 1 patients at 4 X 10' U/m2/d, and 3patientsat6
X IO5 U/m2/d.
Toxicity of low-d(~oseIL-2 @er T-cell-depleted BMT.
The toxicities encountered during treatment with low-dose
IL-2 after CD6-depleted allo-BMT are summarized in Table
1. The most common toxicities observed were fatigue, fever,
nausea, and vomiting. Sideeffectsweremore
frequently
notedatthe
higher starting doses.Indeed, none of the3
patients who were initially treated at 6 X IO5 U/m2/d were
able to tolerate more than 2 weeks at this
dose, largely because of constitutional and gastrointestinal symptoms. Overall, 25/29 patients were able to complete at least 4 weeks of
therapy; 17/29 patients completed at least 8 weeks of treatment. Reasonsfor earlydiscontinuationwerecatheter-related sepsis (3),bronchiolitis(2),gastrointestinaltoxicity
(2), disease relapse(2), pancytopenia ( I ) , hypersensitivity
(l), and acute GVHD ( l ) . The source of 1L-2 (HoffmannLaRoche, Amgen) did not appear to influence the frequency
or severity of side effects. The single
patient who developed
acute GVHD had grade 3 disease, presenting with an erythematous rash and profuse watery diarrhea at 90 days postBMT. The diagnosis wasconfirmed histologically with biopsies of the skin andsigmoid colon. GVHDdeveloped 7
weeks after beginning IL-2 and persisted long after discontinuation. This patientlater
developed fatalAspergillus
pneumonitis while receiving corticosteroids. Although
several other patients did develop rashes while receiving IL-2,
they were transient and usuallyresolved
spontaneously.
None of these patients had histologic evidence of GVHD
when their skin was biopsied.
Immunologic effects. Immunophenotypic analysis of
PBMC obtained frompatients receiving IL-2 showeda selective increase in the number of circulating CD56+ NK cells
without, in general, a change in CD3+ T-cell number (Fig
1). This contrasts sharply with the pattern of immunologic
reconstitutionobservedin
our non-IL-2-treated CD6-depleted allogeneic transplant recipients in whom no increase
Table 1. Toxicity of Low-Dose Recombinant IL-2 After T-cellDepleted BMT
Toxicity
Fatigue
Fever (>38"C)
Nausea/vomiting
Myalgias
Rash
Diarrhea
Edemalweight gain
(21)
Dyspnea (21)
Catheter infection
(17)
Thyroid dysfunction
Thrombocytopenia ( 5 2 0 x 109/L)
Neutropenia (50.5x 109/L)
Acute GVHD
Azotemia (creatinine 2 2 . 0 mg/dL)
Jaundice (bilirubin 22.0 mg/dL)
Hypotension requiring pressors
* Includes all 29 patients treated with IL-2.
N*
W O )
6
6
5
2 (7)
2 (7)
1 (3)
(3)
0 (0)
0 (0)
0 (0)
0
4
8
12
Weeks on IL-2
Fig 1. Effect of low-dose IL-2 on circulating NK cell and T-lymphocyte number. The mean number of NK cells and T lymphocytes in
patients receiving IL-2 after CD6-depleted allogeneic BMT is displayed over time. Values represent data from 25 patients who received greater than 1 month of therapy. Standarderrors were within
10% of the mean value at all data points.
incirculating NK cells is observed beyond6weekspostBMT.''.'' Dual-color immunofluorescence showed that the
majority of NK cells noted in these patientscoexpressed
CD16 (FcyR,,, receptor) and lackedexpression of CD3.
Treatment with IL-2 did not induce expression of CD25, the
T
low-affinity IL-2 receptor chain, on eitherNKcellsor
cells. However,IL-2 therapy didincrease thedensity of
expressionofp75,theintermediate-affinity
IL-2 receptor
chain, on NK cells. Enhanced expression of the p75 IL-2
receptor chain was not induced on T lymphocytes at these
dose levels of IL-2. Cytolytic activity against NK-sensitive
and NK-resistant tumor targets was increased in all patients
receiving [L-2. The timing of IL-2 initiation post-BMT appeared to influence theimmunologic response.Figure2
shows thatpatients in whom IL-2was begun 5 60 days
post-BMT had a more striking increase in both total lymphocytes (Fig 2A) and number of CD56+ NK cells(Fig 2B)
than those in whom IL-2 was begun later. This phenomenon
was independent of IL-2 dose and suggests that there may
be aperiodafter
BMTduring which NKcellsaremore
susceptible to immunologic manipulation.
Patient outcome. Eighteen of the 25 patients who received 2 1 month of IL-2 therapyremainalive,
free of
disease at a median follow-up of 21 months post-BMT. Six
patients have relapsed,4 of whom received truncatedcourses
of IL-2 because of side effects. The observed relapses occurred in 4 patientswith AML transplanted in 2 second
complete remission, 1 patient with stable-phase CML, and
1 patient with MDS. Five of 14 patients (36%) who began
IL-2 more than 2 months after BMT relapsed compared with
only 1/11 (9%) patients in whom IL-2 was initiated within
2 months of BMT. None of the patients receiving low-dose
IL-2 developedcytomegalovirus or any other viral infections
after transplant.
From www.bloodjournal.org by guest on February 11, 2015. For personal use only.
LOW-DOSE IL-2 AFTERT-CELL-DEPLETED
A
967
BMT
Table 2. Comparative Patient Characteristics
T
4O00
Group 1
l+lL-2)
B
30oo
T
T
I
Group 3'
no IL-2)
(no GVHD.
no IL-2)
23
69
25
N
Age (yrs)
(median)
Sex
Male
Female
Disease
CML
AM L
ALL
NHL
MDS
Stage
Early
Advanced
Median follow-up
(rnos)
Week on IL-2
Group 2"
(+GVHD.
41 (21-56) 37
(19-59)
33 .l6
(20-57)
18
7
(74%) 34 (49%)
,038
(26%) 35 (51%)
(72%) 17
(28%) 6
13 (52%)
6 (24%)
2 (8%)
2 (8%)
2 (8%)
P Value
(43%)
30 >.5
17 (25%)
4 (17%) 15 (22%)
2 (9%)
4 (6%)
3 (4%)
1 (4%)
8 (35%)
8 (35%)
37>.5
1 1 (44%)
12 (52%) (54%)
32 (46%)
14 (56%) 11 (48%)
23
(8-43) 42
(10-112)
.003
(12-80)50
Includes only patients who were alive, NED at
BMT.
day +l00 post-
there was a predominance of males among patients who
received IL-2. As expected, the median follow-up of control
patients was longer than that of patients who received IL-2.
The probability of disease relapse for patients treated with
IL-2 is displayed in Fig 3. Among patients without a history
of GVHD, those who received IL-2 had a lower relapse rate
than those who did not receive IL-2 (log-rank, P = .042).
20oo
0
l0
0
Week on IL-2
Fig2. Timing of IL-2 administration and NK cell response.The
effect of IL-2 on absolute lymphocyte count (A) and the number of
CD%+ NK cells (B) is compered in patients in whom treatment was
initiated before (n = 11) and after day +60 post-BMT (n = 14).
We compared the clinical outcome of patients receiving
IL-2 with that of patients transplanted for the same malignancies at our institution since 1984. Control patients were
treated withan identical method of donor-marrow T-cell
depletion and ablative regimen." We separated controls into
two groups of patients who either did or did not have evidence of grade 2-4 GVHD post-BMT, as the presence of
GVHD would have precluded entry onto our IL-2 trial. We
eliminated from the control groups all patients who either
relapsed or died before day +l00 post-BMT because these
patients would have been unlikely to have been eligible for
IL-2 treatment. Clinical variables were well balanced among
IL-2-treated patients (group 1, n = 25) and control patients
with (group 2, n = 23) or without (group 3, n = 69) grade
2-4 GVHD (Table 2). In particular, there were no significant
differences among the groups with respect to patient age,
diagnosis, or stage of disease at the time of BMT, although
l-OI
0.8.
Group 3
0
1
2
3
4
5
6
7
8
9
1
0
Years post-BMT
Fig 3. Risk of relapse in patients treated with IL-2 post-BMT. The
Kaplan-Meier estimate of the risk of disease relapse for patients
treated with greater than 4 weeks of IL-2 post-BMT (group 1, n = 25)
is displayed and compered
with the risk of disease relapsein control
patients with (group2, n = 23) or without (group 3, n = 6 9 ) a history
of grade 2-4 GVHD. Patients who had relapsed or died before day
+l
00 post-BMT were excludedfrom the control groups. Among patients free ofGVHD, relapse rate was superiorfor patients receiving
11-2 (group 1 v group 3, log-rank P = .042).
From www.bloodjournal.org by guest on February 11, 2015. For personal use only.
968
SOIFFER ET AL
Table 3. Multivariate Analysis of Prognostic Factors:
Risk of Relapse
Prognostic Factor
Estimated
Coefficient
P Value
-On
Hazard Ratio
(95% Cl)
IL-2 v no IL-2 (group 1 v
0.34
,016 -1.075
group 3)
GVHD v no GVHD
(group 2 v group 3)
-1.006
Advanced- v early-stage
disease2.59
(1.47-4.59)
,001
0.953
Older v younger
(1.02-3.14)
1.79 .0430.581
(0.14-0.82)
,023
(0.15-0.87)
0.37
Cox's proportional hazards model, using a variety of clinical
factors including age, sex, diagnosis, disease stage, degree
of HLA disparity, presence of GVHD, and treatment with
IL-2 was then used to determine predictors of relapse in the
overall cohort of patients. As illustrated in Tables 3 and 4,
factors such as early disease stage and younger age at the
time of BMT were associated with a decreased relapse risk.
As anticipated, the development of grade 2-4 GVHD was
also associated with a lower risk of disease relapse (group
2 versus group 3, hazards ratio 0.37, 95% C1 0.15 to 0.84).
Like patients with GVHD, patients treated with IL-2 experienced a lower relapse rate compared with control patients
without GVHD who did not receive IL-2 (group 1 versus
group 3, hazards ratio 0.34,95% C1 0.14 to 0.82). The extent
of the decrease in the risk of relapse associated with IL-2
therapy was similar to that observed in patients with GVHD.
Figure 4 illustrates the estimated probability of DFS for
patients receiving IL-2. DFS was superior for patients treated
with IL-2 compared with control patients free of GVHD
(log-rank, P = .MI). When Cox's proportional hazards
model was applied, treatment with IL-2 was found to be an
independent predictor of an improved DFS (group 1 versus
group 3, hazards ratio 0.39, 95% C1 0.18 to 0.87) (Table 3).
In contrast, the presence of GVHD was not associated with
an improvement inDFS because of increased transplantrelated mortality observed in patients with GVHD.26
DISCUSSION
The widespread use of donor-marrow T-cell depletion in
patients undergoing allogeneic BMT has been limited by
reports of a higher frequency of disease relapse posttrans~ 1 a n t . lIt. ~has been postulated that marrow purging removes
components that serve to control the growth of residual malignant cells that have survived the ablative regimen. Considerable experimental evidence supports the proposition that T
Table 4. Multivariate Analysis of Prognostic Factors: DFS
Estimated
Coefficient
Prognostic
Factor
IL-2 v no IL-2 (group 1 v
group 3)
-0.937
Advanced- v early-stage
(1.32-3.63)
,002
0.783
disease2.19
Older v younger
2.01 ,0070.701
PValue
Hazard Ratio
(95% Cl)
,021
(0.18-0.87)
0.39
(1.21-3.35)
o*21
0.04.
0
Group 3
..
1
.
,
2
3
.
4
.
5
.
.
6
7
8
9
1
0
Years post-BMT
Fig 4. Effect of IL-2 administrationon DFS post-BMT. The KaplanMeier estimate of DFSfor patients treated with greater than 4 weoks
of IL-2 post-BMT (group 11 is displayed and compared with the DFS
in control patients who did (group 21 or did not (group 3) have grade
2 through 4 GVHD. Patients who had relapsed or died before day
+l00 post-BMT were excluded from the control groups. Among patients free of GVHD, DFS was superior for patients receiving IL-2
(group 1 v group 3, log-rank P = .041).
lymphocytes are central to the process of eliminating residual
disease.*^*' Allo-reactive T cells may exert their antileukemic
activity directly against tumor targets or, indirectly, through
the elaboration of secondary cytokines such as tumor necrosis factor, IL-1, or y - i n t e r f e r ~ n .The
~ ~ .therapeutic
~~
application ofdonor T lymphocytes has been underrecent investigation. There have been some preliminary reports indicating
that infusion of T lymphocyte-rich buffy coat cells from
HLA-identical allogeneic donors can induce hematologic
and cytogenetic remissions in patients with CML who have
relapsed p~sttransplant.~'.~'
However, buffy-coat infusion at
the time of BMT, designed to induce GVL activity, has
been associated with an increase in fatal GVHD without a
decrease in disease relapse rate." Clinical trials are also
underwayinwhichfixed
numbers of T lymphocytes are
being added back to donor bone marrow after exhaustive Tcell depletion in an attempt to restore GVL activity without
compromising GVHD prophylaxis. Unfortunately, the appropriate number of T cells or subset of T cells required to
accomplish this goal remains unknown.
It is likely that the destruction of residual malignant cells
post-BMT is mediated through several mechanisms, and
some evidence suggests that, like T lymphocytes, NK cells
may play an important role. In experimental models, NK
cell number and activity have been found to correlate with
resistance to metastatic spread of tumors.32In humans, diminished NK cell function has been associated with cancer
progression of both solid and hematologic malignan~ies.~~.'~
After allogeneic marrow transplantation, JL-2-activated NK
cells can mediate major histocompatability complex (MHC)nonrestricted killing of tumor cells obtained from patients
before BMT.I3-l5Therefore, it is possible that exploitation
From www.bloodjournal.org by guest on February 11, 2015. For personal use only.
LOW-DOSE IL-2 AFTER T-CELL-DEPLETED BMT
of the cytotoxic activity ofNK cells may be an effective
way to approach the problem of preventing relapse posttransplant. In addition, because NK cells have been suggested to
be important in the control of viral infection, their presence
and activity may help to reduce some of the infectious complications typically observed after BMT.
The current study confirms that circulating NK cells can
be safely expanded and activated in patients who have undergone T-cell-depleted allogeneic BMT. The absence of Tcell stimulation at these low doses probably accounts for
the low incidence of GVHD observed during this trial. The
contrasting response of T cells and NK cells to low-dose IL2 derives from the distribution of IL-2 receptors on these
cell types. The IL-2 receptor is comprised of several subunits
with differing affinities for the c y t ~ k i n e .Resting
~ ~ . ~ ~T cells
do not express the high-affinity heterodimeric receptor responsible for mediating the proliferative response to IL-2
whereas a fraction of resting NK cells have detectable highaffinity receptor on their surface. Moreover, the p75 chain
of the IL-2 receptor, which binds IL-2 with intermediate
affinity and may play a role in mediating cytolytic function,
is expressed on virtually all resting NK cells, but not on
resting T lymphocytes. Thus, by administering IL-2 in low
doses rather than in the high doses characteristically used in
immunotherapy trials for patients with metastatic renal cell
carcinoma and melan0ma,3'.~~
we can selectively manipulate
NK cell number and function and avoid the induction of
GVHD by activated T cells.
Multivariable analysis of our data indicates that, in patients without evidence of GVHD, treatment with IL-2 is
associated with a reduction in the rate of disease relapse
post-BMT. Because L - 2 has thus far not been administered
in the context of a prospective randomized trial, the multivariate analyses in this study were necessarily performed on
historical controls. These controls were comparable in terms
of clinical characteristics and virtually identical in terms of
clinical treatment. The exclusion of patients who had either
relapsed or died within the first 100 days post-BMT eliminated from the control population patients who were unlikely
to have been placed on IL-2 treatment. The reduction in the
rate of disease relapse associated with IL-2 treatment was
similar to that found in patients who developed GVHD after
BMT at our institution. Notably, treatment with IL-2 was
also associated with improved DFS compared with control
patients who were free of GVHD. Despite the fact that control patients with GVHD relapsed less often than control
patients without it, their overall DFS was not superior because of a higher incidence of fatal complications. Our data
suggests that low-dose IL-2 therapy can provide a way to
reduce the incidence of disease recurrence that is independent of GVHD and free of its toxic consequences. Moreover,
our results provide further evidence in support of an important role for NK cells in the GVL process.
Despite these encouraging results, it still remains unclear
exactly how to maximize the immunologic effects of lowdose IL-2 therapy. We have observed that dose escalation
may be difficult because of the development of intolerable
side effects. As doses of IL-2 are increased, the selective
stimulatory effect on NK cells will ultimately be lost because
969
of saturation of intermediate-affinity receptors on other cell
types. Efforts should be directed at discovering ways to further activate the NK cell compartment already expanded by
low-dose IL-2, perhaps by exposure to other immunomodulatory agents such as IL-12,39IL-6,40or lin~mide.~'
The appropriate duration of treatment with IL-2 is also uncertain.
Previous trials of L - 2 post-BMT have involved high doses
of therapy for relatively short intervals."'" In contrast, lowdose IL-2, because it is well tolerated, can be continued for
an extended period. Prolonged immune stimulation may be
important to the GVL phenomenon as witnessed by the
strong association between de novo chronic GVHD and freedom from posttransplant disease relap~e."~
However, the inconvenience of maintaining central venous catheters and
their susceptibility to infection may make prolonged treatment problematic. Alternative routes of IL-2 administration
(ie, subcutaneous) need to be examined to determine if they
can produce equivalent degrees of selective NK cell expansion and activation post-BMT.
The appropriate timing of IL-2 initiation post-BMT also
must be ascertained. Animal studies have suggested that the
beneficial effect of IL-2 may be lost if it is not administered
close to the time of marrow infusion.&Our preliminary findings suggest a more robust NK cell response with earlier
initiation of treatment. There may be a window of opportunity in which IL-2 therapy is immunologically and clinically
most effective. However, it is not clear that patients can
tolerate IL-2 in the very early posttransplant period. A recent
report on the administration of IL-2 to patients with ALL
after autologous BMT showed that early treatment produced
considerable toxi~ity.4~
Like recipients of T-cell-depleted allogeneic marrow, patients undergoing autologous BMT are at higher risk of relapse than those undergoing unpurged allogeneic BMT."'
Because the pace and character of lymphoid reconstitution
differ between recipients of autologous and T-cell-depleted
allogeneic marrow, it is not certain that the immunologic
responses to IL-2 will be identical. Nevertheless, our previous studies have shown that the NK cell response to lowdose IL-2 is similar in these two groups.16 Although no
conclusions have been drawn regarding the efficacy of IL-2
in the autologous BMT setting, one group has suggested that
it lowers the relapse rate of patients undergoing auto-BMT
for AML.49 IL-2 administration might also be useful after
unpurged allogeneic BMT for patients with advanced malignancies at high risk of relapse. However, it has not as yet
been extensively studied, partially because it is feared that
IL-2 would induce severe GVHD." Because low-dose IL-2
appears to selectively activate NK cells, and not T lymphocytes, this approach wouldbeworth
investigating in this
setting. However, it is possible that allo-reactive T cells,
present after unpurged allo-BMT, but generally absent after
T-cell-depleted BMT, would be stimulated at lower doses
of IL-2 because of the presence of high-affinity IL-2 receptors on their surface. Any clinical trials with IL-2 after unpurged allo-BMT must be undertaken with caution.
It is hoped that the restoration of GVL activity after Tcell-depleted allogeneic marrow transplantation will lead to
an improvement in long-term DFS for patients with hemato-
From www.bloodjournal.org by guest on February 11, 2015. For personal use only.
970
SOIFFER ET AL
logic malignancies. Our results suggest that low-dose IL-2
administration for a prolonged period post-BMT may help
to accomplish this goal.
ACKNOWLEDGMENT
We would like to acknowledge the contributions of Sandra Dunstan, RN,andLynn
Colicchio, RN, whohelped to care for our
patients in outpatient clinic; Steven Chartier and Heather Collins for
processing blood samples for immunophenotyping, Suzan Lazo who
performed the FACS analysis, and Patrice Noonan for her help in
the preparation of this manuscript.
REFERENCES
1. Hale G, Cobbold S, Waldmann H: T-cell depletion with
CAMPATH- 1 in allogeneic bonemarrow transplantation. Transplantation 45:753, 1988
2. Wagner JE, Santos GW, Noga SJ, Rowley SD, Davis J, Vogelsang GB, Farber ER, Zehnbauer BA, Sara1R, Donnenberg AD:
Bone marrow graft engineering by counterflow centrifugal elutriation: Results of a phase 1-11 clinical trial. Blood 75:1370, 1990
3. Goldman J, Gale R, Horowitz M, Biggs JC, Champlin RE,
Gluckman E, Hoffmann RG, Jacobsen SJ, Marmont AM, McGlave
PB, Messner HA, Rimm AA, Rozman C, Speck B, Tura S, Weiner
RS, Bortin MM: Bone marrow transplantation for chronic myelogenous leukemia in chronic phase: Increased risk of relapse associated
with T cell-depletion. Ann Intern Med 108:806, 1988
4. Marmont AM, Horowitz MM, Gale RP, Sobocinski K,Ash
RC, van Bekkum DW, Champlin RE, Dicke KA, Goldman JM,
Good RA, Herzig RH, Hong R, Masaoka T, Rimm AA, Ringden
0, Speck B, Weiner RS, Bortin MM: T-cell depletion of HLAidentical transplants in leukemia. Blood 78:2120, 1991
S. Weiden PL, Floumoy MS, Thomas ED, Prentice R, Fefer A,
Buckner CD, Storb R: Anti-leukemic effect of graft-versus-hostdisease in human recipients of allogeneic-marrow grafts. N Engl J
Med 300:1068, 1979
6. Weisdorf DJ, Nesbit ME, Ramsay NKC, Woods WG, Goldman
AI, Kim TH, Hurd DD, McGlave PB, Kersey JH: Allogeneic bone
marrow transplantation for acute lymphoblastic leukemia in remission: Prolonged survival associated with acute graft-versus-host-disease. J Clin Oncol 5: 1348, 1987
7. Horowitz MM, Gale RP, Sondel PM, Goldman JM, Kersey J,
Kolb H, Rimm AA, Ringden 0, Rozman C, Speck B, Truitt RL,
Zwaan FE, Bortin M M Graft-versus-leukemia reactions after bone
marrow transplantation. Blood 75:555, 1990
8. Antin J: Graft-Versus-Leukemia: No longer an epiphenomenon. Blood 82:2273, 1993
9. Slavin S, Ackerstein A, Naparstek R, Weiss 0, Weiss L: The
graft-versus-leukemia (GVL) phenomenon: Is GVL separable from
GVHD? Bone Marrow Transplant 6:155, 1990
10. Sykes M, Romick M, Sachs D: Interleukin-2 prevents graftversus-host disease while preserving the graft-versus-leukemiaeffect
of allogeneic T cells. Proc Natl Acad Sci USA 875633, 1990
1 1. Johnson B, Drobyski W, Truitt R: Delayed infusion of normal
donor cells after MHC-matched bone marrow transplantation provides an antileukemia reaction without graft-versus-host disease.
Bone Marrow Transplant 11:329, 1993
12. Sullivan KM, Weiden PL, Storb R, Witherspoon RP, Fefer
A, Fisher L, Buckner CD, Anasetti C, Appelbaum FR, Badger C,
Beatty P, Bensinger W, Berenson R, Bigelow C, Cheever MA, Clift
R, Deeg HJ, Doney P, Greenberg P, Hansen JA, Hill R, Loughran
T, Martin P, Neiman P, Peterson FB, Sanders J, Singer J, Stewart
P, Thomas ED: Influence of acute and chronic graft-versus-host
disease on relapse and survival after bone marrow transplantation
from HLA-identical siblings as treatment of acute and chronic leukemia. Blood 73:1720, 1989
13. Higuchi CM, Thompson JA, Cox T, Lindgren CG, Buckner
CD, Fefer A: Lymphokine-activatedkiller function following autologous bone marrow transplantation for refractory hematological malignancies. Cancer Res 495509, 1989
14. Machnnon S, Hows J, Goldman J: Induction of in vitro graftversus-leukemia activity following bone marrow transplantation for
chronic myeloid leukemia. Blood 76:2037, 1990
IS. Hauch M, Gazzola M, Small T, Bordignon C , Barnett L,
Cunningham I, Castro-Malaspina H, O’Reilly R, Keever C : Antileukemia potential of interleukin-2 activated natural killer cells after
bonemarrow transplantation for chronic myelogenous leukemia.
Blood 75:2250, 1990
16. Soiffer RJ, Murray C, Cochran K, Cameron C, WangE,
Schow PW,Daley JF, Ritz J: Clinical and immunologic effects
of prolonged infusion of low-dose recombinant interleukin-2 after
autologous and T-cell-depleted allogeneic bone marrow transplantation. Blood 79517, 1992
17. Hercend T, Takvorian T, Nowill A, Tantravahi R, Moingeon
P, Anderson KC, Murray C, Bohuon C, Ythier A, Ritz J: Characterization of natural killer cells with antileukemia activity following
allogeneic bone marrow transplantation. Blood 67:722, 1986
18. Soiffer RJ, Murray C, Mauch P, Anderson KC, Freedman
AS, Rabinowe SN, Takvorian T, Robertson Mi, Spector N, Gonin
R, Miller KB, Rudders R, Freeman A, Blake K, Coral F, Nadler
LM,Ritz J: Prevention of graft-versus-host disease by selective
depletion of CD6 positive T lymphocytes from donor bone marrow.
J Clin Oncol l0:1191, 1992
19. Mehta C, Patel N: A network algorithm for the exact treatment
of Fisher’s exact test in RxC contingency Tables. J Am Stat Assoc
78:427, 1983
20. Zar JH: Biostatistical Analysis (ed 2). 1984, p 390
21. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457, 1958
22. Armitage P, Berry G: Statistical methods in medical research.
Oxford, UK, Oxford University, 1987, p 429
23. Cox D: Regression models and life-tables (with discussion).
J Roy Stat Soc 34:187, 1972
24. SAS/STAT@Software Technical Report P-229: The PHREG
procedure. Cary, NC, SAS Institute Inc. 1992, p 433
25. Soiffer RJ, Bosserman L,Murray C, Cochran K,Daley J,
Ritz J: Reconstitution of T cell function following CD6-depleted
allogeneic bone marrow transplantation. Blood 75:2076, 1990
26. Soiffer RJ, Gonin R, Murray C, Robertson MJ, Cochran K,
Chartier S, Cameron C, Daley J, Levine H, Nadler LM, Ritz J:
Prediction of graft-versus-host disease by phenotypic analysis of
early immune reconstitution after CD6-depleted allogeneic bone
marrow transplantation. Blood 82:2216, 1993
27. Falkenburg J, Faber L, van den Elshout M, van LuxemburgHeijs S, Hooftmanj-den Otter A, Smith W, Voogt P, Willemze R:
Generation of donor-derived antileukemic cytotoxic T-lymphocyte
responses for treatment of relapsed leukemia after allogeneic HLAidentical bone marrow transplantation. J Immunol 14:305, 1993
28. Symington F W , Storb R: Mechanisms of tissue damage in
human acute graft-versus-host disease. J Immunol Immunopharmacol 9:214, 1992
29. Piguet PF: Tumor necrosis factor and graft-vs-host disease,
Ferrara J, Atkinson K (eds): Graft-Versusin Burakoff SJ, Deeg W ,
Host Disease: Immunology, Pathophysiology and Treatment. New
York, NY, Dekker, 1990, p 225
30. Drobyski W, Keever C, Roth M, Koethe S, Hanson G,
McFadden P, Gottschall J, Ash R, van Tuinen P, Horowitz M, Flomenberg N: Salvage immunotherapy using donor leukocyte infusions as treatment for relapsed chronic myelogenous leukemia after
From www.bloodjournal.org by guest on February 11, 2015. For personal use only.
LOW-DOSE IL-2 AFTERT-CELL-DEPLETED
BMT
allogeneic bone marrow transplantation: Efficacy and toxicity of a
defined T-cell dose. Blood 82:2310, 1993
31. Kolb H, Mittermuller J, C l e m C, Holler E, Ledderose G,
Brehn G, Hemi M, Wilmanns W: Donor leukocyte transfusions for
treatment of recurrent chronic myelogenous leukemia in marrow
transplant patients. Blood 76:2462, 1990
32. Introna M, Mantovani A: Natural killer cells in human solid
tumors. Cancer Metastasis Rev 2:337, 1983
33. Fujimiya Y, Chang W-C, Bakke A, Horwitz D, Pattengale P:
Natural killer (NK) cell immunodeficiency in patients with chronic
myelogenous leukemia. Cancer Immunol Immunother 24213, 1987
34. Choe B, Frost P, Morrison N. Rose N: Natural killer cell
activity of prostatic cancer patients. Cancer Invest 5:285, 1987
35. Smith D: The interleukin 2 receptor. Ann Rev Cell Bioi 5:291,
1989
36. Robertson M, Ritz J: Biology and clinical relevance of human
natural killer cells. Blood 762421, 1990
37. Rosenberg S, Lotze M, Yang J, Aebersold P, Linehan W,
Seipp C, White D: Experience with the use of high dose IL-2 in the
treatment of 652 cancer patients. Ann Surg 210474, 1989
38. Margolin K, Rayner A, Hawkins M, Atkins M, Dutcher J,
Fisher R, Weiss G, Doroshow J, Jaffe H, Roper M, Parkinson D,
Weirnik P, Creekmore S , Boldt D: Interleukin-2 and lymphokineactivated killer cell therapy of solid tumors: Analysis of toxicity and
management guidelines. J Clin Oncol 7:486, 1989
39. Soiffer R, Robertson M, Murray C, Cochran K, Ritz J: Interleukin-12 augments cytolytic activity of peripheral blood lymphocytes from patients with hematologic and solid malignancies. Blood
82:2790, 1993
40. Smyth M, Chtaldo J: Comparison of the effect of IL-2 and
IL-6 on the lytic activity of purified human peripheral blood large
granular lymphocytes. J Immunol 146:1380, 1991
41. Bengtsson M, Simonsson B, Carlsson K, Nilsson B, Smedmyr
B, Termander B, Oberg G, Totterman TH: Stimulation of NK cell,
T cell, and monocyte functions by the novel immunomodulator linomide after autologous bone marrow transplantation. Transplantation 532382, 1992
42. Gottlieb D, Brenner M, Heslop H, Bianchi A, Bello-Fernandenz E, Mehta A, Newland A, Galzka A, Scott E, Hoffbrand A,
97 1
Prentice H:A phase I clinical trial of recombinant interleukin-2
following high dose chemo-radiotherapy for haematological malignancy: Applicability to the elimination of minimal residual disease.
Br J Cancer 60610, 1989
43. Blaise D, Olive D, Stoppa AM, Viens P, Pourrneau C, Lopez
M, Attal M, Jasmin C, Monges G, Mawas C, Mannoni P, Palmer
P, Franks C, Philip T, Maraninchi D Hematologic and immunologic
effects of the systemic administration of recombinant interleukin-2
after autologous bone marrow transplantation. Blood 76: 1092, 1990
44. Higuchi CM, Thompson JA, Peterson FB, Buckner CD, Fefer
A: Toxicity and immunomodulatory effects of interleukin-2 after
autologous bone marrow transplantation for hematologic malignancies. Blood 77:2561, 1991
45. Sullivan KM, Storb R, Buckner CD, Fefer A, Fisher L,
Banaji M, Hansen J,
Weiden PL, Witherspoon RP, Appelbaurn
Martin P, Sanders JE, Singer J, Thomas ED: Graft-versus-host-disease as adoptive immunotherapy in patients with advanced hematologic neoplasms. N Engl J Med 320:828, 1989
46. Charak B, Brynes R, Chogyoji M, Kortes V, Tefft M, Mazumder A: Graft versus leukemia effect after transplantation with
interleukin-2-activated bone marrow. Transplantation 56:31, 1993
47. Weisdorf D, Anderson P, Blazar B,Uckun F, Kersey J, Ramsay N: Interleukin 2 immediately after autologous bone marrow
transplantation for acute lymphoblastic leukemia-a phase I study.
Transplantation 5961, 1993
48. Kersey JH, Weisdorf D, Nesbit ME, LeBien W ,Woods
WG, McGlave PB, Kim T, Vallera DA, Goldman AI, Bostrom B,
Hurd D, Ramsay NKC: Comparison of autologous and allogeneic
bone marrow transplantation for treatment of high risk refractory
acute lymphoblastic leukemia. N Engl J Med 317:461, 1987
49. Benyunes MC, Massumoto C, York A, Higuchi CM, Buckner
CD, Thompson JA, Peterson FB, Fefer A: Interleukin-2 with or
without lymphokine-activated killer cells as consolidative immunotherapy after autologous bone marrow transplantation for acute myelogenous leukemia. Bone Marrow Transplant 12:159, 1993
50. Favrot M, Floret D, Negrier S,Cochat P, Bouffet E, Dacheng
2, Franks C, Bijman T, Brunat-Mentigny M, Philip I, Philip T:
Systemic interleukin-2 therapy in children with progressive neuroblastoma after high dose chemotherapy and bone marrow transplantation. Bone Marrow Transplant 4:499, 1989
FR.
From www.bloodjournal.org by guest on February 11, 2015. For personal use only.
1994 84: 964-971
Effect of low-dose interleukin-2 on disease relapse after T-celldepleted allogeneic bone marrow transplantation
RJ Soiffer, C Murray, R Gonin and J Ritz
Updated information and services can be found at:
http://www.bloodjournal.org/content/84/3/964.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.