|
|||||||||
|
|
|
|
|
|
|
|
|
|
|
Prepublished online as a Blood First Edition Paper on May 13, 2002; DOI 10.1182/blood-2002-02-0419.
TRANSPLANTATION
From the Laboratory of Experimental Transplantation and
Laboratory of Experimental Radiobiology, University of Leuven, Belgium.
A murine model of minor histocompatibility antigen-mismatched bone
marrow transplantation (BMT) was used to study the role of timing of
donor lymphocyte infusion (DLI) in eliciting graft-versus-host (GVH)
and graft-versus-leukemia (GVL) reactivity. We gave DLI at weeks 3 and
12 after BMT and related its ability to induce a GVL effect with (1)
evolution of T cell chimeric status and (2) the extent to which DLI
could elicit lymphohematopoietic GVH (LHGVH) reactivity. All mice
remained free of GVH disease, but only week 3 DLI chimeras exhibited a
significant GVL response when challenged with host-type leukemia cells.
In these week 3 DLI chimeras, host-reactive T cells were found to
proliferate in vivo (5- [and-6]-carboxyfluorescein diacetate,
succinimidyl esther [CFSE]-labeled DLI inocula,
TCR-V The immune recognition and elimination of residual
tumor cells by engrafted donor cells, designated the
graft-versus-leukemia (GVL) effect, constitutes the main curative
potential of allogeneic (allo) hemopoietic stem cell transplantation
(HSCT) in hematologic malignancies.1 Unfortunately, the
GVL effect seems to be closely associated with graft-versus-host
disease (GVHD), still a major cause of posttransplantation
morbidity and mortality. The infusion of donor lymphocytes (donor
lymphocyte infusion [DLI]) following HSCT allows reinduction of
remission in patients with posttransplantation leukemic
relapse.2-5 New HSCT protocols are currently being
developed, in which the exploitation of the GVL response occupies an
important place. The new pretransplantation and posttransplantation
conditioning regimens are tuned to be immunosuppressive rather than
nonmyeloablative, thereby aiming at establishing mutual tolerance
between host and graft, so as to allow allogeneic cells to eradicate
host hematopoietic and malignant cells.6-8 As a part of
these regimens, DLIs are given to patients with relapse or to those
failing to develop full chimerism.9-14 Observations in
patients, treated with DLI, have indicated that the incidence and
severity of GVHD are high when DLI is applied early after HSCT but
significantly lower when DLI is delayed for weeks or months after
transplantation.15,16 Similarly, in animal models of allo
BMT, donor lymphocytes (DLs) can be safely infused, once a sufficient
time interval after transplantation has elapsed.17-23
Moreover, in murine bone marrow (BM) chimeras, DLI did induce distinct
GVL effects.17,21,22 These and other clinical and
experimental data indicate that, whereas GVH and GVL responses may
share effector cells and target antigens, the GVL effect may be
dissociated from GVHD.
The reason for waning susceptibility to GVHD in the posttransplantation
period is not well understood. Shortly after transplantation, the
inflammatory cytokine storm, brought about by the conditioning regimen,
may amplify the immune reactivity of host-reactive donor T
cells.24 Alternatively or in addition to this, the time
interval between transplantation and DLI may allow for development of
regulatory cells that keep host-reactive T cells in
check.19,20,23,25-30 Another poorly defined issue is the
nature of the GVL effector cells. Because GVHD31,32 and
the use of T cell-replete marrow grafts31,33 were shown
to confer protection against disease relapse, donor T cells can be
presumed to be the primary mediators of the GVL response. Furthermore,
in animal models, both CD4+ and CD8+ GVL
effectors have been described,1,34 and in humans, both CD4+ and CD8+ T-cell lines and clones with
antileukemic specificity have been reported.1,35-39
However, in the wake of the T-cell response, other cell populations,
such as natural killer (NK) cells, may be recruited and may participate
in the GVL response. NK and lymphokine-activated killer (LAK) cells
have indeed been implicated as players in the GVL effect in humans and
in mice.40-45 GVL target antigens may be malignancy
specific or overall host specific and the latter antigens may be
restricted to hematopoietic lineages or may be broadly expressed.
Hematopoietic tissue-restricted minor histocompatibility antigens
(miHCags) have been described and donor-derived T-cell clones
from patients undergoing allo HSCT have been isolated, targeting not
only normal hematopoietic cells but also leukemic cells and leukemic
precursors.35,36,46-49 In patients treated with DLI,
successful GVL responses are often associated with conversion to
complete donor chimerism,11,14,15,50 supporting the
concept of a lymphohematopoietic graft-versus-host (LHGVH)
response as part of the GVL mechanism.
Successful application of DLI after allo HSCT requires striking the
delicate balance between eliciting the GVL response and causing GVHD,
the latter of which still constitutes a major treatment-related complication.6-9 Further study into the effector
mechanisms of both immune phenomena may help to design regimens that
predictably achieve GVL effects without GVHD. Here, we used a murine
model of miHCag-mismatched bone marrow transplantation (BMT),
previously shown to be well suited for studying the immunoregulatory
mechanisms of DLI. In this model, GVHD was avoided when DLI was delayed
for 3 weeks after BMT while a distinct GVL response was allowed to develop.21 In the present study, DLI was performed at
different time points after BMT, as from the moment that DLI was
previously shown to be safe.21 The capacity of DLs to
induce a GVL response was studied and was related to the degree of
pre-existing chimerism and to the extent to which DLs were able to
elicit alloreactivity in vivo and in vitro. Frequency analysis of T
cells expressing specific host-reactive T-cell receptor V Animals
BMT
Delayed DLI Three and 12 weeks after BMT, chimeric mice were infused via a tail vein with 50 × 106 donor-type splenocytes.Leukemia challenge BW5147.3 cells (AKR mouse lymphoma; American Type Culture Collection, Rockville, MD) were used to study the GVL effect. The tumor cells were taken from frozen stock and maintained in vitro for a limited number of passages, from which cells were taken for experiments. For each in vivo challenge, the in vivo behavior of tumor cells was controlled for by inoculating untreated host-type mice and monitoring leukemia-free survival. One week after DLI, chimeric mice were injected in a tail vein with 5 × 106 leukemia cells. Mice were weighed and observed for clinical signs of leukemic disease. Moribund animals were killed for postmortem examination.Scoring of T-cell chimerism and frequency of host-reactive donor T cells At different time intervals after BMT and DLI, T-cell chimerism and the frequency of T cells expressing specific TCR-V chains were
studied by flow cytometry, using the FACStar plus (Becton Dickinson,
Erembodegem, Belgium). Venous blood was obtained from the animals by
intracardiac puncture or by cutting the tail. Following red blood cell
lysis using NH4Cl, cells were labeled with fluorescein isothiocyanate (FITC)-, phycoerythrin (PE)- or peridinin chlorophyll protein (PerCP)-conjugated antibodies directed against Thy1.1 (Serotec), Thy1.2 (Serotec), TCR-V6, TCR-V3, TCR-V8.3, or
TCR-V11, CD4, CD8, or CD3 (Pharmingen, Becton Dickinson,
Erembodegem, Belgium; and Caltag Laboratories, Synbio bv, AM Uden, The Netherlands).
Mixed lymphocyte reaction Responder cells (nylon wool-enriched chimeric or control splenocytes) and stimulator cells (host-type splenocytes) were cultured for 5 days in RPMI 1640, supplemented with 10% fetal calf serum (FCS), antibiotics, and 5 × 10 5 2-mercaptoethanol,
at a concentration of 5 × 106 cells/mL and a ratio of
1:1, in a final volume of 200 µL/well, in flat-bottomed 96-well
microculture plates. Prior to culture, stimulator cells were treated
with mitomycin C (Kyowa Hakko Kogyo, Tokyo, Japan), as described
previously.21 DNA synthesis was assayed by adding 1 µCi
(0.037 MBq) (methyl-3H) thymidine (Radio Chemical Centre,
Amersham, Buckinghamshire, United Kingdom) per well during the last 18 hours of culture. Thereafter, the cells were harvested on glass filter
paper, and the counts per minute (cpm) were determined in a liquid
scintillation counter. Results are expressed as stimulation index (SI):
SI = (cpm of stimulated cells  cpm of nonstimulated cells)/cpm
of nonstimulated cells.
In vitro labeling of cells with CFSE for in vivo transfer and tracking of DLI cells Prior to infusion, donor-type splenocytes were labeled with 5- (and-6)-carboxyfluorescein diacetate, succinimidyl esther (CFSE; Molecular Probes, Europe, Leiden, The Netherlands). Splenocytes were suspended at a concentration of 50 × 106 cells/mL in RPMI 1640 and incubated with CFSE at a final concentration of 5 µM for 5 minutes at room temperature. Cells were subsequently washed 3 times with phosphate-buffered saline (PBS), supplemented with FCS 20%, counted, and resuspended for intravenous infusion.Statistics The Mann-Whitney U test and the Kruskal-Wallis multiple comparison Z test were used to estimate the level of statistical significance of difference between groups of data. The log-rank test was used to estimate the level of significance of the difference in survival between groups (P < .05 was considered as evidence for statistical significance).
Clinical GVH and GVL reactivity following DLI at 3 and 12 weeks after BMT We have shown in previous studies that infusion of 50 × 106 donor-type splenocytes at 3 weeks after BMT failed to induce clinical GVHD, while generating a powerful GVL effect.21 To see whether this would still be the case at 12 weeks, we conducted an experiment in which week 3 and week 12 chimeras were challenged with 50 × 106 donor-type splenocytes. Control groups consisted of week 3 and week 12 chimeric animals not given DLI. Survival, weight changes, and clinical signs of GVHD were recorded. Figure 1 shows that, like week 3 chimeras, these week 12 chimeras failed to show weight loss or any other sign of GVHD (hunched back, hair loss, diarrhea; results not shown). Survival was 100% in both groups. The absence of GVHD was confirmed by histology (absence of lymphocytic infiltration in skin, gut, liver; results not shown).
In a similarly designed experiment, week 3 and week 12 DLI chimeras
given either DLI or no treatment were inoculated, 1 week after DLI,
with 5 × 106 BW5147.3 leukemia cells. Weight and signs
of leukemic infiltration were monitored. When moribund, the animals
were killed for postmortem examination. The Kaplan-Meier survival
curves (Figure 2) show that whereas DLI
prolonged disease-free survival time in week 3 chimeras, it failed to
do so in week 12 chimeras. Week 3 DLI chimeras, week 12 DLI chimeras,
and untreated control AKR mice were tested together in 2 experiments,
reconfirming the previously reported GVL effect of week 3 DLI21 and revealing the absence of a significant GVL
response after DLI at 12 weeks. Two additional experiments were
performed, which further confirmed the inability of DLI, when performed
at week 12 after BMT, to elicit a GVL response. That animals died of
leukemic disease was confirmed on postmortem histopathologic
examination.
Donor (host-reactive TCR-V 6-TCR-PE monoclonal antibodies, the
frequency and fluorescence intensity of persisting CFSE+
cells was determined in the T-cell population and in the
TCR-V6-expressing T-cell population. Figure
3, panels A and B, shows that
CFSE+CD3+ cells persisted both in week 3 and
week 12 chimeras. However, the frequency of CFSE+ cells
among total CD3+ cells was markedly higher in week 3 chimeras as compared to week 12 chimeras: 5.7% (SE = 0.7, n = 12)
in week 3 and 2.1% (SE = 0.1, n = 11) in week 12 chimeras (mean of
values obtained in 2 identically designed experiments;
P < .0001 for comparison between groups as tested by the
Mann-Whitney U test). Furthermore, analysis of the
distribution of CD3+ cells according to their fluorescence
intensity revealed that, in week 3 chimeras, a significant proportion
of CFSE+ cells had completed 3 cell divisions; 4 peaks were
identified on the histogram, corresponding to serial 2-fold decreases
of fluorescence intensity. In contrast, in week 12 chimeras, only one
peak could be discerned, the position of which corresponds to the peak
with highest fluorescence intensity in week 3 chimeras (Figure 3C-D).
Peripheral blood lymphocytes (PBLs) of chimeras, given DLI with
unlabeled donor-type spleen cells, were used as negative controls for
analysis of the CFSE fluorescence profiles (n = 5/group). Analysis of
PBL CFSE fluorescence on day 7 after week 12 DLI equally revealed one
population of CFSE+CD3+ cells only, with a CFSE
fluorescence profile identical to that of the population seen on day
14. These data indicate that DLI cells, injected at 3 weeks,
proliferate for at least 14 days, whereas they do not appear to
proliferate during this period when injected into week 12 chimeras.
Donor-antihost reactivity, as it occurs in GVHD, was associated with
expansion of TCR-V6-expressing T cells (see "Expansion and clonal
deletion of T cells expressing specific TCR-V chains in GVHD and
after BMT and DLI"). Although the absolute counts were low, frequency
analysis of TCR-V6+CD3+ cells showed that,
14 days after DLI, the relative number of TCR-V6+ cells
among total CD3+ cells was higher in week 3 than in week 12 chimeras (2.5% ± 0.29% SE, respectively, 1.2% ± 0.07% SE,
n = 9-10/group, one of 2 representative experiments;
P < .001 for comparison between groups as tested by the
Mann-Whitney U test; results not shown). Furthermore,
V6-TCR+CD3+CFSE+ cells in week 3 chimeras exhibited CFSE fluorescence of varying intensity, whereas in
week 12 chimeras, TCR-V6+CFSE+ T cells
exhibit a more uniform fluorescence profile, suggesting that
TCR-V6+ cells constitute a subpopulation of T cells that
proliferate after infusion into the week 3 chimeric host (Figure
3E-F).
Nonclinical antihost reactivity following DLI at 3 and 12 weeks after BMT To test the possibility that DLI does elicit antihost reactivity, which remains limited to the lymphohematopoietic system, experiments were conducted in which we compared week 3 and week 12 chimeras for the effect of DLI on the degree of chimerism. Figure 4 shows proportions of host and donor lymphocytes just before and 7 days after DLI. As can be seen, mixed chimerism had developed at 3 weeks after BMT. In these recipients not given DLI, the level of donor T-cell chimerism increased only very slowly, whereas in those given DLI, it was converted within 1 week into near-complete donor T-cell chimerism (Figure 4A). In week 12 chimeras, near-complete donor T-cell chimerism had already established spontaneously at the time of DLI, and changes, occurring in the subsequent period of 1 week, whether DLI was given or not, were insignificant (Figure 4B). These data indicate that DLI at week 3 induces vigorous GVH reactivity, characterized by a rapid replacement of host-type by donor-type T cells. Because near-complete donor T-cell chimerism is established at week 12, conclusions with regard to LHGVH reactivity cannot be made based on chimerism status.
Expansion and clonal deletion of T cells expressing specific
TCR-V 6, -V7, -V8.1,
and -V9 T cells. C3H donor-type mice do not carry the Mtv-7
provirus, but are infected with the Mtv-6 retrovirus, encoding the
Mls-2 antigen, leading to clonal deletion of TCR-V3 T
cells.52 Clonal deletion of TCR-V11-expressing T cells
and persistence of TCR-V8.3-expressing T cells is common to both strains. We have previously shown that GVHD occurs when DLs are administered on the day of BMT.21 Using an identical
experimental setup, we confirmed that GVHD, characterized by weight
loss, hair loss, and a hunched back is associated with a marked
expansion of splenic TCR-V6+ T-cell-T cells, when
compared to healthy BM chimeras that had received transplants with
depleted BM only (Figure 5). In both GVHD
and control animals, the proportion of TCR-V11+ cells
(clonally deleted in both strains) remained low and the frequency of
TCR-8.3+-expressing T cells (expressed in both mouse
strains) fell within the range of normal untreated host- and donor-type
mice. These data clearly show that in animals developing GVHD, a
selective expansion occurs of host-reactive CD4+ and
CD8+ T cells expressing the TCR-V6 chain.
Based on the aforementioned results, the frequency of
TCR-V
To test whether the long-term increased frequency of
TCR-V
Ex vivo antihost reactivity early after DLI at 3 and 12 weeks after BMT To functionally evaluate the ability of DL to mount antihost reactivity after their infusion into a week 3 or week 12 chimeric host, we tested the capacity of lymphocytes to mount a proliferative response, ex vivo, in a standard mixed lymphocyte reaction (MLR). Considering that, reportedly, lymphocytes recirculate within hours to the spleen,53 we studied the donor-antihost immune reactivity within the first days after DLI. Spleen cells were taken from week 3 and week 12 chimeras, 48 hours after DLI, and stimulated with host-type antigens in a standard MLR. Splenocytes of normal untreated donor-type mice were used as controls for donor-antihost reactivity. As can be seen from Figure 8, spleen cells of both week 3 and week 12 DLI chimeras were capable of mounting a proliferative response in vitro, which was equally strong, although significantly weaker than that of control donor-type splenocytes. Splenocytes from chimeras not given DLI were unable to generate a proliferative response.
The model used for this study consisted of irradiated AKR mice, reconstituted with T cell-depleted C3H BM. As previously shown, such mice develop mixed T-cell chimerism at 3 weeks after transplantation. DLI in the first week induces GVHD, whereas DLI at week 3 fails to induce such a lethal response, but produces a beneficial GVL effect.21 In view of the waning susceptibility to GVHD in the early posttransplantation period, and because GVHD and the GVL effect may share certain target antigens and effector cells, we asked the questions (1) whether further delay of the DLI would similarly be associated with waning of the GVL effect and (2) whether changes in DLI-elicited responses could be related to continued evolution in the chimeric status of the recipient mice. Animals receiving DLI either at week 3 or 12 remained free of clinical GVHD during the entire observation period. This is in keeping with previous reports and indicates that DLI can be safely performed after allo BMT, once a sufficient time interval, the length of which may depend on the model used, has elapsed.17-23 Whereas the previously reported GVL effect of DLI at week 321 was reconfirmed, the data indicated that, when infused at week 12, DLs are not allowed to develop antileukemic activity. The presence or absence of the GVL effect was associated with substantially different immunologic antihost reactivity in vivo. Both in week 3 DLI and week 12 DLI chimeras, infused DLs remained detectable for at least 14 days. A subpopulation of CFSE+ T cells expanded in vivo in week 3 chimeras during the first 14 days after infusion. In week 12 chimeras, however, CFSE+ T cells did not appear to divide on day 14 nor at the earlier stage of day 7 after DLI. From these data, it can be concluded that in week 3 chimeras, T cells proliferate in response to host antigens, whereas in week 12 chimeras, these host-reactive T cells are not stimulated or not allowed to proliferate to the same extent. Following week 3 DLI, host-type lymphocytes were rapidly eliminated with conversion of pre-existing mixed into near-complete chimerism within 7 days, indicating that host-reactive cells generated a marked LHGVH reaction. The frequency of TCR-V In chimeras not given DLI, clonal deletion after allo BMT was shown to
be a long-lasting and specific process; at all time points,
TCR-V In week 3 chimeras given DLI, conversion of mixed into complete donor
chimerism was associated with a progressive expansion of
TCR-V In week 12 DLI chimeras, DLI did not elicit any detectable antihost
reactivity, as evident from the lack of proliferation of infused
CFSE+ T cells and the lack of expansion of
TCR-V Our observations bring forward important aspects of GVL and GVH responses developing after DLI and after allo BMT in general. They suggest that the GVL reactivity of DLI is, at least in part, occurring as part of the T cell-mediated LHGVH response and that the LHGVH reaction with the subsequent conversion to a more advanced donor chimeric state, rather than the full donor chimeric state itself, may be crucial to the GVL response of DLI. The GVL response may therefore wane as time elapses after DLI and remain operative as long as host-reactive T cells from the DLI inoculum proliferate. Importantly, the data suggest that, for donor cells from the DLI inoculum to be able to eradicate leukemia cells, a certain degree of residual host lymphohematopoietic chimerism may be required so that a sufficiently strong LHGVH reaction is elicited. A key role has been ascribed to host-derived APCs in eliciting GVH responses62 and they may therefore also be instrumental in eliciting LHGVH and GVL effects. Therefore, that adequate timing of DLI was so critical for obtaining the GVL effect probably derives from the change in chimeric status taking place after BMT. The direct involvement of residual host-type APCs in generating LHGVH and GVL reactivity should be investigated using a model in which host- and donor-type chimerism can be distinguished both in T and non-T-cell lineages. In patients treated with DLI, a GVL response often coincides with an increase in donor chimerism11,14 and because clinical data do seem to indicate that the success of procedures such as DLI may depend on chimerism status, frequent lineage-specific chimerism analysis has recently been advocated as a guideline to novel transplantation strategies.63 Characterization of lineage-specific chimerism in patients undergoing allo BMT revealed that, in patients with late relapses from acute leukemia or myelodysplastic syndrome, the only cells of host origin were leukemic cells.64 Theoretically, in such complete donor chimeric hosts, the simultaneous infusion of host-type APCs with the DLI inoculum may bring about sufficient GVH reactivity to produce a GVL response. Our findings add to the understanding of the effectiveness of DLI in clinical practice and may be useful for the development of DLI as a preventive strategy in patients undergoing allo BMT.
Submitted February 8, 2002; accepted April 23, 2002.
Prepublished online as Blood First Edition Paper, May 13, 2002; DOI 10.1182/blood-2002-02-0419.
Supported by grants from the National Fund for Scientific Research (FWO) Flanders, from the ASLK Cancer Research Fund and from the Belgian Federation against Cancer. A.D.B. is a fellow of the FWO.
The publication costs of this article were defrayed in part by page charge payment. Therefore, and solely to indicate this fact, this article is hereby marked "advertisement" in accordance with 18 U.S.C. section 1734.
Reprints: Mark Waer, Laboratory of Experimental Transplantation, University of Leuven, Campus Gasthuisberg, O&N 8, Herestraat 49, B-3000 Leuven, Belgium; e-mail: mark.waer{at}med.kuleuven.ac.be.
1. Porter DL, Antin JH. The graft-versus-leukemia effects of allogeneic cell therapy. Annu Rev Med. 1999;50:369-386[CrossRef][Medline] [Order article via Infotrieve].
2.
Slavin S, Naparstek E, Nagler A, et al.
Allogeneic cell therapy with donor peripheral blood cells and recombinant human interleukin-2 to treat leukemia relapse after allogeneic bone marrow transplantation.
Blood.
1996;87:2195-2204 3. Slavin S, Naparstek E, Nagler A, Ackerstein A, Kapelushnik J, Or R. Allogeneic cell therapy for relapsed leukemia after bone marrow transplantation with donor peripheral blood lymphocytes. Exp Hematol. 1995;23:1553-1562[Medline] [Order article via Infotrieve].
4.
Kolb HJ, Mittermuller J, Clemm C, et al.
Donor leukocyte transfusions for treatment of recurrent chronic myelogenous leukemia in marrow transplant patients.
Blood.
1990;76:2462-2465 5. Porter DL, Roth MS, McGarigle C, Ferrara JL, Antin JH. Induction of graft-versus-host disease as immunotherapy for relapsed chronic myeloid leukemi | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
4/group and total n = 43 and
25 for day 21 and day 84, respectively). In right panels, venous blood
was taken at week 4 (day 28) or 13 (day 91) from animals that had
either been given no infusion or DLI at week 3 or week 12, respectively, after BMT. Bars represent mean ± SE of 3 to 5 identically designed experiments (n 
