Recognition of Clonogenic Leukemic Cells, Remission Bone Marrow and
HLA-identical
Donor Bone Marrow by
CD8+
or
CD4+ Minor
Histocompatibility
Antigen-specific Cytotoxic
T
Lymphocytes
Laura M. Faber,* John van der Hoeven,* ElsGoulmy,*Annelies L. Hooftman-den Otter,* Simone A.P.van Luxemburg-Heijs,* Roel Willemze,* and J. H.Fredenk Falkenburg*
*Laboratory of Experimental Hematology, Department of Hematology, and the
tDepartment
ofImmunohematology and Bloodbank, University Medical Center, 2333 AALeiden, The NetherlandsAbstract
Introduction
Weinvestigated whether minor
histocompatibility
(mH) an-tigen-specificcytotoxicTlymphocytes (CTL)-can discrimi-natebetweenleukemichematopoietic progenitor cells(leu-kemic-HPC) from AML or CML patients, the HPC from theirremissionbonemarrow(remission-HPC),and normal HPC fromtheirHLA-identical siblingbone marrowdonor (donor-HPC). Specific lysisbyCD8+ CTL cloneswas
ob-served notonly oftheleukemic-HPCbutalsoof the
donor-HPC in3/4 patient/donor combinations expressingmH an-tigenHA-1, 3/5 combinations expressingmH antigen HA-2, 2/3combinations expressingmHantigenHA-3,and2/2
combinations expressing mH antigen HY-Al. In four pa-tient/donorcombinationstherecognition ofthedonor-HPC
wasclearly lessthanoftheleukemic-HPC,indicating
differ-ential
susceptibility
to lysis by these mH CTL clones. In addition, differential recognition of leukemic-HPC and re-mission-HPC within seven patients was analyzed. In one patient expressing the HA-2antigen onthe leukemic cellstherecognition of the remission-HPC wasclearlylessthan
oftheleukemic-HPC. One CD4+CTLcloneshowedspecific
lysis
of theleukemic-HPC fromanAMLpatientandaCML patientaswellasofnormalremission-HPCanddonor-HPC.These resultsillustrate that ingeneral CD8+andCD4+mH
antigenspecific CTL clones donot differentially recognize
leukemic-HPC and normal-HPC. However, differences in
susceptibility
tolysis of malignant versusnormalcells maycontributeto adifferential GVL effect.(J.
Cln.
Invest. 1995. 96:877-883.) Key words: leukemia * clonogenic(leuke-mic) hematopoietic progenitor cells * differential recogni-tion *graft-versus-host disease*graft-versus-leukemia
reac-tivity
Address correspondencetoL.M.Faber, MD,Department of Hematol-ogy,Building 1, C2-R, University Medical Center,P.O. Box9600, 2300 RCLeiden,TheNetherlands. Phone:71-262267, FAX:71-225555.
Receivedfor publication15November 1993 and acceptedinrevised
form4May1995.
1. Abbreviations usedin this paper: AML, acute myeloid leukemia; BM, bone marrow; BMT, bone marrowtransplantation;CML, chronic
myeloid leukemia; CTL,cytotoxic T lymphocyte;GVHD, graft-versus-hostdisease; GVL, graft-versus-leukemia; HPC, progenitor cells; mH, minorhistocompatibility.
Allogeneic bonemarrowtransplantation(BMT
)'
has been asso-ciated withanimmune-mediated anti-leukemic effect, the graft-versus-leukemia(GVL) effect. Tlymphocytes from the donor marrow graftmayberesponsible for this GVL effect since T celldepletion of the graft is correlated with anincreased risk ofleukemic relapse after BMT (1-6). Sinceacorrelationhas been found between theoccurrenceofGVHD withadecreased risk of leukemicrelapseafter BMT(2,4, 7-12)it is hypothe-sizedthatthedonor-derivedT lymphocytesthatcause GVHD mayalsobe the mediators of the GVL reactivity. More recently, direct clinical evidence foraGVL effect has beendemonstrated by several investigators (13-15). Hematological and cytoge-netic remissions after leukocyte transfusion from the original marrow donor for Philadelphia-positive patients with chronic myeloid leukemia (CML) who relapsed afterallogeneicBMT have been reported. In theetiology of both GVL and GVHD minorhistocompatibility (mH) antigensmayplayanimportant role(16-20). Invitro, wehavedemonstrated that CD8+mH antigen specific cytotoxic T lymphocyte (CTL) clones gener-atedfrom patients with GVHDafterallogeneic BMT,are capa-ble ofantigen-specific lysis of freshly obtained leukemic cells, and ofantigen specific growthinhibitionof theclonogenic leu-kemicprecursorcells(leukemic-HPC) from unrelated patients (21, 22).Furthermore, wehaveshownthat CD8+ mHantigen specific CTL clonescanrecognizemHantigens on hematopoi-eticprogenitor cells(HPC) (23-25). Recently,wehave dem-onstratedthatmHantigen specific CD4+ CTL clones could be generatedfrom apatientwithchronicmyeloid leukemia (CML) with acute GVHD grade HI-IV afterallogeneic BMT (Faber, L.M., S.A.P. vanLuxemburg-Heijs, W.F.J. Veenhof, R. Wil-lemze, and J.H.F. Falkenburg, manuscript submitted for publi-cation). As previously illustrated, some mH-specific CTL clones recognize both hematopoietic and non-hematopoietic cells, whereas other reactivities appearto berestricted tocell types ofhematopoietic origin (26).Inthisstudy we investigated whether mHantigen specific CD8+orCD4+ CTL clones gen-erated frompatientswithGVHDdifferentiallyrecognize leuke-mic-HPC frompatients with AML or CML and normal HPC from their HLAgenotypically identical sibling donor. Further-more,weevaluated whether thesemHspecificCTL clonescan discriminatebetween leukemic-HPCandthe remission progeni-tor bonemarrow cells (remission-HPC), within the same pa-tient withAML. Wedemonstratethatalthoughingeneralthese CD8+and CD4+mHantigenspecificCTL clonesdonotdiffer-entiallyrecognizeleukemic-HPC and normalHPC,differences in susceptibility to lysis ofmalignant versus normal cellscan
beobservedsuggestingthat insomepatients this may resultin aspecific GVL effect.
J.Clin. Invest.
© TheAmericanSocietyfor ClinicalInvestigation,Inc.
Methods
Generation of mH antigen-specific CTL clones. CD8+ mH
antigen-specific CTL clones with specific cytotoxicity for the mH specificity HA-1, HA-2, or HA-3 were generated from patients after allogeneic BMTaspreviously described (16, 27).HA-1 andHA-2-specific
recog-nition isrestrictedby A2, HA-3 recognition is restricted by HLA-Al. As control effector cells CTLcloneswith anti-HLA-A2 or
anti-HLA-A1 specificity (positive controls) or a HLA-B8 restricted mH
antigenspecificity (negative control)wereused. Thefrequency of the
mHantigens recognized by theHA-1,HA-2, HA-3 specificCTLclones studied in arandom populationon the lymphocytes from individuals
expressing the relevant class Irestriction moleculewas69%forHA-1,
95%for HA-2, and 88% for HA-3 (27). TheHLA-Al-restricted HY-specific CTL clone used in this study was generated froma female
patient whorejectedaT-lymphocyte-depletedbonemarrowgraft from
her HLA-phenotypically identical donor (28).
TheCD4+ mH antigen-specific HLA-DR2restricted CTL clones, clone 11 was generated froma patient with CMLwith acute GVHD gradeHI-IV afterallogeneic BMT (Faber, L.M.,S.A.P.van
Luxemburg-Heijs,WJ.F.Veenhof,R.Willemze,and J.H.F.Falkenburg, manuscript submittedfor publication). The second CD4+ CTL clone used, clone
S,wasgeneratedbefore HLA-identicalallogeneicBMTfrom the sibling donordirectedagainsttheleukemic cellsfrom the recipient (29). Clone S,showed HLA-DR2restrictedlysis oftherecipient leukemic cells and DR2 positiveEBVtransformed Bcells (EBV-LCL). PHAstimulated Tlymphocytes from bothrecipient and donor werenotrecognized by this clone.
TheCTL cloneswereculturedin RPMI mediumsupplemented with
15%pooledhumanserumandhumanrecombinantinterleukin-2
(r-IL-2) 300IU/ml and expanded by weekly restimulation with irradiated, specificmHantigenpositiveEBV-LCLplusfreshly isolated peripheral bloodmononuclearcells(MNC)from randomdonors. The CTL clones
were stored in liquid nitrogen. The day before use the clones were
thawed andculturedovernight in RPMI plus 15% pooled humanserum
and rIL-2300IU/ml.
5"Crreleaseassay. Standard51Crreleaseassays were performedas
described previously (29, 30). Aseffector cells, the CTL cloneswere
used;astargetcellsTlymphocytepopulations, leukemic cells, remission bonemarrow from the patient,bone marrow from the donoror EBV-LCL'swereused. Targetcells,labelledwith 51Cr and effector cellswere
incubated at various ratio's. To measure spontaneous release of5'Cr,
targetsuspensionswereanalyzed in the absence of effector cells. Maxi-mumreleasewasdetermined byadding 0.1 ml of thetargetsuspension
to 0.1 mlZaponinesolution. The percentage specific lysis obtained in
a51Cr releaseassay wasdeterminedasfollows: 100%x (experimental release cpm - spontaneous release cpm) / (maximum release cpm
- spontaneous releasecpm).
Targetcells. Afterinformedconsentleukemic cells werecollected
from the bone marrow ofpatients with AML or CML at diagnosis. Patients with >95% morphologically recognizable leukemic cells in theirbonemarrowwereselected.Remission bonemarrow wasobtained
incompleteremissionfrom the patients with AML aftertreatmentwith daunorubicin and arabinosyl-cytosine (Ara-C) (induction course) and subsequently with high dose Ara-C and amsacrin (consolidation course). (Normal) donorbone marrow was obtained from the
HLA-identicalsibling donorof patients whoweretransplanted.The cellswere
centrifuged over Ficoll Isopaque (density 1.077 g/cm3, 1,000 g, 20
min) and the interphase cells were harvested, and cryopreserved in liquid nitrogen. Beforeuse, the cells werethawed, washed twice, and
resuspended in RPMI containing 15% prescreened ABserum.
T lymphocyteswereexpandedby stimulating 107peripheral blood MNC from patient or donor with 0.2% phytohemagglutinin (PHA) (Difco Laboratories, Detroit, MI) in RPMIplus 15%human ABserum
forthreedays. Thecells werethenwashed andfurther cultured inthe
absence ofPHA inthepresenceof human rIL-2 300 IU/ml,foratleast threemoredaysbefore testing.
StableEBV-LCLwereestablished byin vitrotransformation of107
peripheralblood MNC withEBV supernatant.Thecells were washed
and further cultured in RPMIplus 10% FCS (GIBCO BRL, Gaithers-burg, MD).
Cell-mediated inhibition of clonogenic (leukemic)progenitor cell growth (31). Todetermine specific reactivity of the anti-mHCTLclones
with clonogenic leukemicprogenitorcells (leukemic-HPC), remission
bone marrowprogenitor cells(remission-HPC) anddonorbone marrow
progenitorcells(donor-HPC) 3.l04targetcells(leukemic cellsor bone marrowcells)in0.1mlmediumconsisting of Iscove'sModified
Dulbec-co's Medium(IMDM) plus 15% human AB serum wasmixed with irradiated(l5Gy) CTLateffector/target (E/T) ratio1:1. The cell mix-ture wascentrifuged (1,000 g, 15 s) to establishcell-cellcontact be-tween CTL andthe targetcells, and then preincubatedfor 4 h at37TC. After incubation the cellswereresuspended and cultured as asingle
cellsuspensionat aconcentration of3.104targetcells/mlin a30-mm culture dishes in 1 ml IMDMcontaining 10% prescreened humanAB
serum, 2U/mi erythropoietin (Cilag AGInternational,Zug, Switzer-land), 10ng/mi GM-CSF(SandozLtd., Basle, Switzerland), 10 ng/ mlG-CSF(Amgen,CA), 50 ng/ml SCF (Amgen, CA), 25 ng/mi IL-3(Sandoz Ltd., Basle,Switzerland),0.47grams/liter human transferrin (Behringwerke AG,Marburg,Germany) saturated with FeCl3,5. 10-5 mol/l mercaptoethanol, andmethylcellulose(Methocel4000 cps;Fluka, Freiburg,Germany)at afinal concentrationof1.1%(wt/vol).Thecells wereincubatedin afullyhumidified atmosphere of5%CO2and37°C. After 14d ofculture, the numberof leukemic colonies (defined as aggregatesofatleast50undifferentiated cells), CFU-Gcolonies
(de-finedasaggregatesof>50neutrophilicoreosinophilicgranulocytes), CFU-M (defined as aggregates >50 monocytic cells), and BFU-E
colonies(definedascoloniesconsisting of>50hemoglobinized cells)
werescored, using a inverted microscope. The colonygrowth in the presenceof effectorcells isexpressed aspercentage ofgrowth of the
numberof colonies observedinthe untreated(grown in theabsenceof effectorcells) control culture.Tocontrolforantigen-nonspecific inhibi-tion ofcolony growth dueto secretion of humoral inhibitory factors into the culture medium, effectors werealso plated togetherwiththe target cells atthe same E/Tratios directlyin thesemisolidmedium, without 4 h preincubation, to prevent cell-cell contact. The colony
growth in the presenceof effectorcells, without4hpreincubation, is expressedaspercentageofgrowth ofthenumberofcoloniesobserved in the untreated(grownin the absence ofeffectorcells)controlculture. Specificcell-mediatedgrowth inhibitionwasdeterminedasfollows:100 -(colony growthinthepresenceof cell-cellcontact/colony growthin the absence of cell-cellcontact)x 100%.Specific cell-mediated growth inhibition of <20% wasconsiderednegative. Inhibition of 20-50%
wasconsidered minorinhibition,inhibition of>50%positive.
Anincreased growth of CFU-M colonieswasobserved when the CD4+CTLclones,clone 11 andclone S, wereused aseffectorcells, compared to the untreated control cultures. Therefore a conditioned
medium wasmade from both CD4+ clones and addedto all culture
dishes atafinalconcentration of20%when thecorresponding CD4+ CTL clonewasusedaseffectorcell.
Forcytogeneticanalysisof thegrowingcoloniesaclonogenicassay ofthe leukemiccellsfrompatients 8 and11 wasperformed. 106 cells werecultured under thesameconditionsasdescribedandwereharvested
after 7 d of culture. Themethylcellulosewasremovedbywashingwith RPMIand the cells wereresuspendedintheoriginalculturemedium
without methylcellulose and cultured in suspension in 50 ml tissue culture flasks for 18 hours at 37°C, 5% C02, 98% humidity. After colcemid arrest,metaphasespreadswerepreparedaccordingtostandard
techniques. Atleast 30metaphases wereexamined for each chromo-somalanalysis.
Results
Leukemiccells from sevenpatients withAMLofvarious sub-typesaccordingtothe FABclassification and from fourpatients with CML in chronic phase were analyzed. Only patients
Table I. Characteristics of the Leukemic Cells and Colony Growth oftheLeukemicPrecursorCells per3.10'PlatedCells
HLA-Type FAB-class. Karyotype Leuk-HPC
1. A2 AML M4/5 46XYdel(19)(pl3) 26
2. A2 AML M2 46XYtrisomie 10 36
3. A1,B8 AMLM5 46XY 286
4. Al, A2, B8 CML CP* 46XY t(9;22) 118 5. Al,A2,B8 CMLCP* 46XYt(9;22) 223
6. A2 CML CP* 46XXt(9;22) 212
7. Al,B8 AMLM5 46XY 383
8. A2 AML M2 46XXt(8;21) 366
9. A2 AMLMO 46XX 80
10. A2, DR2 AMLM2 46XY 149
11. DR2 CMLCP* 46XX t(9;22) 353
*Chronic phase. Forpatient
ments.
11, 1.10'cellswereplatedin the
experi-pressingHLA-Al,-A2,and/or HLA-DR2ontheirlymphocytes
were selected for this study, corresponding with the HLA-re-striction element from the mH antigen-specific CTL clones used.Table I shows thecharacteristicsof the leukemicsamples
from the I1 patients used.Cytogenetic analysis of the growing colonies frompatients8 and 11 after 7 d of culture showed that
93%of themetaphasescontained thet(8;21) translocationand 100% of the metaphases contained the t(9;22) translocation,
respectively.Furthermore,thehigh percentage (>95%)of the leukemic cells in the samplesusedexcluded thepossibility of asignificantcontribution of colonies derived from residual nor-malprogenitorcells present in the cell suspension.
Table H shows the specific reactivity of the CD8+ and CD4+ mHCTLclones with theT lymphocytesorEBV-LCL
from the patients and their HLA-identical sibling donors. In
general the mHspecific CTL clones didrecognizetheT lym-phocytes from both the patient and HLA-identical donor ex-pressingthe relevant mHantigen. However intwo patient/do-norcombinations differences in lysis of T lymphocytes from
thepatientversusTlymphocytes fromthe donorby twoCTL cloneswasobserved. The HA-I and HA-3 specificCTL clones showedspecific lysis of the T lymphocytes from patient 1 and 4,respectively, but not of the T lymphocytes from their HLA-identical donor, illustrating that the specific mH antigen was notexpressed by the T lymphocytes from the donors. When mHantigen-specific lysis of the Tlymphocytes was observed theleukemic cells from the patients and the bone marrow cells from their HLA-identical sibling donor were further tested in thecell-mediated cytotoxicity assays described.
HLA-A2andHLA-AJ restricted recognitionby themH
spe-cific
CTL clones. The HLA-A2 restricted HA-i and anti-HA-2 CTL clone and the HLA-A1 restricted anti-HA-3 and anti-HY CTL clone showed similarcytotoxicity againstthe leu-kemic cells from the patientsand the normal bone marrow cell fromtheir HLA-identical sibling donorasmeasured in the51Cr release assay (TableHI). WhentheseCTLcloneswere further testedintheclonogeniccytotoxicity assay,similar results were observed from the HA-1, HA-3, and HY-Al specific CTL clones (Table IV). When the leukemic cells from the patient andthe normal bonemarrowcells from the HLA-identical do-nor were recognized in the5"Cr
release assay, mH antigen-specific growth inhibition was observed of both the leukemic-HPC fromthepatient and HPCfrom the HLA-identical donor. Different results were obtained from the HLA-A2 restricted HA-2specific CTL clone. This CTL clone recognized the HA-2 antigen expressed onthe Tlymphocytes and leukemic cells frompatient 1 and2, and theTlymphocytes and normalbone marrowcellsfrom theirHLA-identical donor in the5"Cr
release assay.However, in theclonogenic assay the HA-2specificCTL clone inhibited the leukemic-HPC from patient 1 and 2, but showedlower inhibition of the HPC fromtheir HLA-identical donor.In conclusion the HA-1, HA-2 and HA-3 specific CTL clones in general recognize both normal and leukemic HPC, but may discriminate betweenleukemic-HPC from thepatient and normalHPC from the HLA-identical sibling donor.
To investigate whether these mH antigen specific CTL clones differentially recognize leukemic and normal precursor cells fromindividualpatients with AML (Table I), the
leuke-Table II.Reactivity of mH-specific CTL Clones with Normal Lymphocytes from the Patients and Their HLA Genotypically Identical Donor Patient/HLA-identicaldonor 1 \ 2 3 4 5 6 CTL-clone P D P D P D P D P D P D HA-1 ++ -- ++ ++ __ __ ++ ++ ++ ++ __ __ HA-2 ++ + ++ ++ -- -- ++ ++ ++ ++ ++ ++ HA-3 _++ ++ ++ ++ ++ mH-B8 -- -- -- -- -- -- -- -- -- -- -- --HY-Al -- -- -- -- ++ ++ ++ ++ * * * * a-Al -- -- -- -- ++ ++ ++ ++ ++ ++ -- --a-A2 ++ ++ ++ ++ __ __ ++ ++ ++ ++ ++ ++
Specific lysisof PHA stimulatedTlymphocytesfrom thepatients(P),generated from remission bone marrow cells of theAMLpatients and from the leukemic bonemarrowcellsof the CMLpatients,andfrom their HLAgenotypicallyidenticalsiblingdonor(D) ina
51Cr
release assayatETratio3:1. (-) 0-10%lysis; (+) 10-25%lysis; (++)25-100%lysis; *not tested. The anti-HLA-Al and anti-HLA-A2 CTL cloneswereusedas
TableIII. Specific Lysis of the Leukemic Cells from the Patients and of the Normal Bone Marrow Cells from Their HLA Genotypically Identical Sibling Donor
Patient/HLA-identicaldonor 1 2 3 4 5 6 CTL-clone P D P D P D P D P D P D HA-1 35 -2 50 61 * * 35 39 57 29 * * HA-2 20 26 39 51 * * 21 29 70 35 24 36 HA-3 * * * * 13 30 30 -4 92 51 * * HY-Al * * * * 25 27 35 46 * * * *
Specific lysis oftheleukemic cellsfrom thepatients (P)andofthe normal bone marrow cellsfromtheir HLAgenotypicallyidenticalsiblingdonor (D) wasmeasured using astandard5'Crreleaseassayinduplicate at E/Tratio3:1. Data areexpressed as percentage specific lysis. *Data not
applicable because therestriction elementwasnotpresent.Theanti-HLA-Al and anti-HLA-A2 CTL clones (data not shown) were used as positive controls(specific lysis> 25%),the HLA-B8 restrictedmH-specific CTL cloneas anegative control (specific lysis < 10%).
mic-HPC and the remission-HPC from the same patient was further tested in the clonogenic cytotoxicity assay when mH
antigen-specific lysis of the T lymphocytes was observed. In generalmH antigen-specific growthinhibitionwasobservedof both the leukemic-HPC and remission-HPC (Table V). Differ-entresultswereobtained from theHLA-A2 restricted anti-HA-2 specific CTL clone. This CTL clone recognized the HA-2
antigen expressedonthe Tlymphocytes, the leukemic cells and remission bonemarrowfrompatients 1,2,8,9,and 10 in the5'Cr release assay (data not shown). HA-2specific growthinhibition by this CTL clonewasobserved of both the leukemic-HPC and remission-HPC frompatients 2,8,9, and 10.However, the HA-2 specificCTL clone inhibited theleukemic-HPC, but
repeat-edlynotthe remission-HPC frompatient1(the experimentwas repeated three times atdifferent days), indicating differential
susceptibility tolysis bythis CTL clone(Table V).
Inconclusionfrom the four tested CD8+ mHspecificCTL clonesonly oneCTL clone,the HA-2 specific CTLclone
dis-criminated between leukemic-HPC cells and remission-HPC from onepatient.
DR-2restricted recognitionbythe CD4+ CTL clones. The twoDR-2restricted CTLclones, clone 11 and clone S recog-nized the EBV-LCL from AML patient 10 (Table VI). In a 51Cr release assay these two CTL clones showed no specific
lysis of the leukemic cells and remission bone marrow (remis-sion-BM) from patient 10. The low specific lysis of the leuke-mic cells frompatient 10 could not be explained by a low DR-expression of these leukemic cells, since the DR DR-expression was high as measured by FACS analysis. However when the CTL clone 11 was further tested in the cell-mediated clonogenic assay mH antigen specific growth inhibition was observed of both the leukemic-HPC and the remission-HPC from patient 10. Nodifferential recognition by this CTL clone was observed. Clone Sshowed no oronly minor specific growth inhibition of theleukemic-HPC and remission HPC from patient 10 in the
clonogenic assay. ThetwoDR-2 restricted CTLclones, clone 11 and clone S bothrecognizedtheEBV-LCL frompatient 11 and herHLA-genotypicallyidentical donor (Table VI). Clone 11 showed low but significant specific lysis of the leukemic cellsfrompatient 11 in the
5"Cr
release assay in three separateexperiments (19±4%). This was probably in part due to the
relatively low expression of class II molecules on the CML cells (<50%). When this CTL clone was further tested in thecell-mediatedclonogenic assay mHantigen specific growth
inhibitionwasobservedof both the leukemic-HPC and the do-nor-HPC frompatient 11. CloneS, the CTL clone that recog-nized the EBV-LCL frompatient11 anddonor 11,and showed minorrecognitionof theleukemic cells from patient 11 in the
TableIV. HLA-AJ andHLA-A2RestrictedGrowthInhibitionof Clonogenic LeukemicPrecursorCellsfrom thePatientsandofthe HematopoieticProgenitorCellsfromTheirHLA GenotypicallyIdenticalSiblingDonor
Patient/HLA-identicaldonor 1 2 3 4 5 6 CTh-clone P D P D P D P D P D P D HA-1 80 17 80 83 * * 95 87 76 85 * * HA-2 60 31 88 49 * * 58 54 69 65 50 53 HA-3 * * * * 75 71 99 1 94 99 * * HY-Al * * * * 73 86 99 99 * * * *
HLA-A1 and HLA-A2restrictedmHspecific growthinhibition ofclonogenic leukemicprecursor cellsfrompatientswith AMLorCML(P) and of normalhematopoieticprecursorcells(BFU-E + CFU-GM)from their HLA-identical donor(D). E/Tratio 1:1 is shown. Similar resultswere obtainedatE/Tratio's 0.3:1 and0.1:1,respectively.The anti-HLA-Al and anti-HLA-A2CTLcloneswereusedaspositivecontrols(specificcell
mediatedcolony growth inhibition>50%),theHLA-B8 restricted mHspecificCTLcloneas anegativecontrol(specificcell-mediatedcolony growth inhibition < 20%).
Table V. HLA-AJ andHLA-A2 RestrictedGrowth Inhibition of theClonogenic Leukemic Progenitor Cells and the Remission HematopoieticProgenitor Cellsfrom the Same Patient
Patient 1 2 3 7 8 9 10 CTL-clone L R L R L R L R L R L R L R HA-1 80 57 80 96 * * * * 2 9 1 30 4 1 HA-2 60 1 88 80 * * * * 72 55 78 74 53 89 HA-3 * * * * 75 75 64 58 * * * * * * HY-Al * * * * 73 90 70 98 * * * * * *
Growthinhibition of clonogenic leukemicprecursorcells(L)andhematopoieticprogenitor cells in remission(R)from thesameAMLpatient by theHLA-Al,HLA-A2 and HLA-DR2restrictedmHspecific CTLclones. E/T ratio 1:1 isshown.Similarresults were obtainedatE/T ratio's0.3:1
and0.1:1, respectively. *Data notapplicable becausetherestrictionelement was not present. The anti-HLA-Al and anti-HLA-A2 CTLclones
wereusedaspositivecontrols(specificcell mediatedcolonygrowth inhibition>50%),the HLA-B8 restricted mHspecificCTL cloneas anegative control (specificcell-mediatedcolony growth inhibition< 20%).
5"Cr
release assay (14±5%,n = 3). However, this clone did notshow specific growth inhibition of theleukemic-HPC and donor-HPC from patient 11 in the cell-mediated clonogenic assay.Discussion
Inallogeneic bone marrowtransplantation, clinical observations haveindicated an antileukemic effect of donor-derived T lym-phocytes, the GVL effect (2, 4, 32, 33). An association has been found between the occurrenceof GVHD and GVL reactivity, althoughclinicalobservations and animal studies have indicated that GVL may exist in the absence of GVHD (2, 32-36).
Several studies have suggested that both CD8+ and CD4+ T
lymphocytes are involved in GVHD and GVL reactivity and may recognize polymorphic antigens in the context of HLA-class I andclass II molecules, respectively, thatareprocessed by the recipient cells (37-46). Previously, we have shown that CTL clones directedagainstmHantigens generated from patients with GVHD afterallogeneicBMT, arecapableof
anti-gen-specific lysisinvitro offreshlyobtained leukemic cells, and ofantigen-specificgrowth inhibition of the leukemic progenitor cells from unrelatedpatients (21,22). Theseexperiments sup-portthehypothesisthatalloreactive CTL clones directedagainst
polymorphic antigens presented by recipient cellscancontribute totheanti-leukemiceffect.
Here, we investigated whether leukemic cells fromAML orCML patients and normal (progenitor) bone marrow cells obtainedin remission orfrom their HLA-identical sibling donor canshowdifferentialsusceptibilitytolysis bymHCTL clones. From the 4 CD8+ and 2 CD4+ CTLclones tested threemH specific CTLclones, the HA-1, the HA-2 and the HA-3 specific CTL clone showed differentialrecognitionoftheleukemic-HPC anddonor-HPCin 4of 14testedpatient/donorcombinations. In these combinations theleukemic-HPCwererecognized bythe mH specific CTL clonesandthe' normal HPC fromthe HLA-identical donor were not recognized. Only one mH specific CTL clone,the HA-2 specificCTL clone showed differential recognition of theleukemic-HPC and remission-HPC fromone of the fivepatients. Inthispatient 1,expressingtheHA-i and HA-2antigenonthe leukemic-HPC thesusceptibility to lysis of the remission-HPC by the HA-2 specific CTL clone was repeatedly (n= 3)muchlower thanof theleukemic-HPC.This couldnotbeexplained by theimpairedexpression of theHLA restriction molecule because the anti-HLA-A2 CTL cloneand the HLA-A2 restricted HA-1specific CTL clones did recognize theremission-HPC from patient 1. We recently have demon-strated that low specific lysis of lymphocytic leukemia cells
Table VI.HLA-DR2 RestrictedSpecificLysisofLeukemicCells, RemissionBone MarrowCellsand Normal Bone Marrow Cells from the HIA GenotypicallyIdentical Donor
Patient10 Patient/HLA-identicaldonor11
Lysist5'Crrelease assay* Growthinhibitiont LysisttCrreleaseassay* Growth inhibitiont
EBV Leu Rem-BM Leuk-HPC Rem-HPC EBVpatient Leuk EBV-donor Donor-BM Leuk-HPC Donor-HPC
Clone 1 81 7 10 87 65 72 19 73 9 55 53
Clone S 49 2 9 33 2 34 14 30 2 16 18
*HLA-DR2restricted mHspecific lysisofEBV-LCL, leukemic cells (Leuk), remission bone marrow cells (Rem-BM) and donor bone marrow cells
(Donor-BM)wasmeasuredusingastandard
ttCr
release assay induplicate atE/Tratio3:1. Data are expressed as percentagespecific lysis. *HLA-DR2restricted mHspecific growthinhibition of clonogenic leukemic progenitor cells (Leuk-HPC) from a patient with AML(patient 10)orCML(patient 11)and of remissionhematopoieticprogenitor cells(Rem-HPC)frompatient 10or normalhematopoieticprogenitor cells from the
HLA-identical donor(Donor-HPC)ofpatient IL.ETratio1:1.Both CD4+CTLclonesshowednospecific lysis of control patient,HLA-DR2positive
by the HA-2 CTL clone could be explained by a low surface expression of these adhesion molecules as compared to targets that werewell recognized (21). The recognition by these latter mH CTL clones could exclude the possibility of a too low surfaceexpression of the adhesion molecules CDl la/CD18.
Sincehematopoiesis in patients with AML in remission may beof clonal origin, clinical remission may be the persistence of apreleukemic clone during remission (47-49). To investigate whether a possible preleukemic hematopoiesis of the remission-BM may explain similarities in recognition of remission and leukemic precursor cells, mH specific recognition of the remis-sion-BM wascompared to the recognition of the bone marrow from theHLA-identical sibling donor. From the 3 HLA-identi-cal sibling pairs tested (AML patient 1, 2 and 3) in one pair differential recognition was observed (data not shown). The remission-HPC frompatient 2 were strongly lysed by the HA-2 specific CTL clone, but only minor recognition of the HPC from the HLA-identical sibling donorwasobserved. Since also from 1 patient the leukemic cells were more susceptible to lysis than the normal HPC, these results may suggest that a preleukemic hematopoiesis of the remission-BM may be the causeof therelatively high lysis of the remission-BM as
com-paredtothe donorbonemarrow.On the otherhand differential susceptibilitytolysis by the HA-2 specific clone duetogenetic predisposition may a more likely explanation for this phenom-ena. Since low specific lysis of normal HPC's by the HA-2 CTL clonewasalso observed by Marijtetal in 2of 6 normal healthy bone marrow donors this differential susceptibility to lysis may be explained byagenetic difference in HA-2 antigen
expression (25).
From the twotested CD4+ CTL clones clone 11 showed
specific lysis of the leukemic-HPC froman AML patientand CML patientas wellas of normal remission-HPC and donor-HPCrespectively. Since the specific lysisof these targets was low in the 54Cr release assay these results confirm the high
sensitivity of the clonogenic assay as a tool for the study of
cellularlydefinedantigensonHPC's(31). Clone S, the CD4+ CTLclone thatrecognizesin vitrotransformed EBV-LCL from DR2positiveindividuals andsome(EBVnegative) AML sam-ples did not show specific growth inhibition of the leukemic-HPC and HPC's from both patients in the clonogenic assay. However thisclone did recognizeboth the leukemic cells and EBV-LCLfrom theAMLpatientagainstwhom thisCTL clone wasgenerated (29). SincetheEBV-LCL from the
HLA-geno-typically identical donor of thisAMLpatientwerealso
recog-nized,the antigen recognized by clone S appearednot tobea classic mHantigen and is still unclear.
From these data we conclude that in general mH specific CTLclones (CD8+ and CD4+ CTLclones) donot differen-tially recognize leukemic-HPC and normal HPC. This is in agreementof studiesperformed in animals indicating that mH antigen specific CTLdo notdiscriminate between normal and malignant hematopoietic cells (12). The differential
suscepti-bilitytolysisof leukemic-HPC fromtherecipient andthe donor-HPC in 4patient/donor combinations by the HA-1, HA-2, or HA-3specific CTL clones may explain aGVLreactivity after
allogeneic
BMT. Inourexperiments recognitionofspecific tar-get organs involved in GVHDbythemHCTL clones has not beentested.Recently astudyhasbeenperformedtodetermine theexpressionof humanmHantigensonvarious tissues(26).The minorantigens HA-3, -4,-6,-7, andHY weredetectedon all tissues tested (PHA-blasts, EBV-LCL, purified Tcells, B
cells, monocytes, immature thymocytes, skin-derived cultured
fibroblasts, keratinocytes,melanocytes, culturedepithelialcells ofkidney proximaltubuli andumbilicalcord vein derived endo-thelialcells),but HA-1 and HA-2 could only be demonstrated oncell types ofhematopoietic origin. Therefore, both mH spe-cific CTL clonesrecognizingmHexpressed to ahighextenton
leukemic-HPCascompared to normal HPC, and clones
recog-nizingmHthatareonly expressed on hematopoietic cells, may be relevantfor thesuppression of leukemia relapses after alloge-neic BMT.
In conclusion, our results indicated that in general mH-specific CTL clones do not differentially recognize leukemic HPC and normal HPC, nordo the leukemic HPC exhibit a
differentialsensitivitytolysisorgrowth inhibition that discrimi-nates them from normal hematopoietic cells. However, in a
minorityof thecasesstudied,differentialsusceptibility of leuke-mic cellsovernormalhematopoietic cells can be demonstrated. Thus, insomepatientssuchmH-specific CTL clones may con-tributeto adifferential GvL effect.
Acknowledgments
The authors thank M.E.M.Kraakman forperforming EBV-transforma-tion ofBcells andDr. A.Wessels forperforming cytogenetic analysis
of thegrowing colonies.
This workwas supported bygrantsfromthe DutchCancer Society
(Koningin Wilhemina Fonds)and the J. A. CohenInstitute for
Radiopa-thology and Radiation Protection.J. H.FrederikFalkenburgisaSpecial Fellow of theRoyal Netherlands Academy ofArtsandSciences.
References
1. Butturini, A., and R. P. Gale. 1989. Clinical trials of T-celldepletion:
currentcontroversies,futuredirections.InBone MarrowTransplant:Current
Con-troversies.R. P.Galeand R. E.Champlin,editors. Alan R.Liss, Inc.NewYork. 495-509.
2.Horowitz, M.M., R. P. Gale,P. M.Sondel,J. M. Goldman,J. Kersey, H. J. Kolb, A. A. Rimm, 0. Ringden,C. Rozman, B. Speck, R. L. Truitt,
F. E.Zwaan,and M. M. Bortin. 1990.Graft-versus-leukemiareactionsafterbone marrowtransplantation. Blood. 75:555-562.
3.Goldman,J.M.,R. P.Gale,M. M.Horowitz,J. C.Biggs,R. E.Champlin,
E.Gluckman,R.G.Hoffmann,S. J.Jacobsen,A. M.Marmont,P. B.McGlave,
H.Messner,A.Rimm,C.Rozman,B.Speck,S.Tura,R.Weiner,and M. Bortin.
1988.Bonemarrowtransplantation for chronic myelogenousleukemia inchronic phase.Increased riskforrelapse associatedwith T-celldepletion.Ann.Intern.
Med. 108:806-814.
4. Barrett,A.J., M. M.Horowitz,R. P.Gale,J. C.Biggs,B. M.Camitta,
K. A.Dicke, R. A.Gluckman, R. A. Good,R. H. Herzig,M. B.Lee,A. M.
Marmont,T.Masoaka,N. K.C.Ramsey,A. A.Rimm,B.Speck,F. E.Zwaan,
and M. M.Bortin. 1989. Marrowtransplantationforacutelymphoblastic leuke-mia: factorsaffecting relapseandsurvival.Blood. 74:862-871.
5.Marmont,A.M.,M.M.Horowitz,R. P.Gale,K.Sobocinski,R.C.Ash,
D. W.vanBekkum,R. E.Champlin,K. A. Dicke,J.M.Goldman,B.Speck,
R.S.Weiner,and M. M.Bortin. 1991. T celldepletionofHLA-identical
trans-plantsin leukemia.Blood. 78:2120-2130.
6.Apperley,J.F.,F. R.Mauro,J. M.Goldman,W.Gregory,C. K.Arthur,
J.Hows,W.Arcese,G.Papa,F.Mandelli,D.Wardle,P.Gravett,G.Franklin,
G.Bandini,P.Ricci,S.Tura,A.lacone,G.Torlontano,W.Heit,R.Champlin,
and R. P. Gale. 1988. Bonemarrowtransplantationfor chronicmyeloidleukaemia in first chronicphase: importanceofagraft-versus-leukaemiaeffect. Br. J. Haema-tol.69:239-245.
7. Weiden,P.L.,N.Flournoy,E. D.Thomas,R. Prentice,A.Fefer,C. D.
Buckner,and R. Storb. 1979.Antileukemiceffect ofgraft-versus-hostdisease in humanrecipientsofallogeneic-marrow grafts.N.Engl.J.Med. 300:1068-1073.
8.Weiden,P.L.,K. M.Sullivan,N.Flournoy,R.Storb,and E. D. Thomas. 1981. Antileukemic effect of chronicgraft-versus-hostdisease: contributionto
improved survival after allogeneic marrow transplantation. N. Engl. J. Med. 304:1529-1533.
9.Wagner,J.E.,M.Zahurak,S.Piantadosi,R.B.Geller,G. B.Vogelsang,
J. R. Wingard, R. Saral, C. Griffin, N. Shah. B. A. Zehnbauer,R. Armbinder, W.
Bums,R.Jones,W. Stratford,S.Rowley,A.Yeager,and G. W. Santos. 1992.
Bone marrowtransplantationof chronicmyelogenousleukemiain chronicphase: evaluation of risks and benefits. J. Clin. Oncol. 10:779-789.
10. Frassoni, F., M. Sessarego, A. Bacigalupo, P. Strada,M. Repetto,S. Occhini,R.Defferari, andA. Marmont.1988.Competition between recipientand donor cellsafterbone marrowtransplantation for chronic myeloid leukemia.Br. J.Haematol. 69:471-475.
11.Frassoni,F., D. Scarpati, A. Bacigalupa, V. Vitale, R.Corvo,S.Miceli,
F. Gualandi, D.Occhini, M. Clavio, and M. T. Van Lint. 1989. The effect of total body irradiation dose and chronic graft-versus-host disease on leukemic relapseafter allogeneic bone marrow transplantation. Br. J. Haematol. 73:211-216.
12. Truitt, R. L., A. V. LeFever,C. C. Y. Shih, J. M. Jeske, and T. M. Martin. 1990.Graft-vs-leukemiaeffect. In Graft-vs-Host-Disease: Immunology,
Pathophysiology, and Treatment. H. J.Burakoff,H.J. Deeg, I. Ferrara,and K. Atkinson, editors.Dekker,New York. 177-204.
13. Kolb,H.J., J.Mittermueller, C. Clemm,C. Holler, G. Ledderose, G. Brehm, M. Heim, and W.Wilmanns. 1990. Donorleukocyte transfusions for treatmentofrecurrentchronic myelogenous leukemia in marrowtransplant pa-tients. Blood. 76:2462.
14.Porter, D.L.,M.S.Roth,C.McGarigle,R. N.James,J. L. M.Ferrara,
and J. H. Antin.1994.Induction of Graft-versus-Host disease asimmunotherapy
forrelapsechronicmyeloidleukemia. N.Engl.J.Med.330:100-106. 15.Bir,B. M. A.M.,A.Schattenberg,E.J. B. M.Mensink,A.GeurtsVan
Kessel, T. F.C. M. Smetsers, G. H. J. N. Knops,E. H. P. Linders, and T. DeWitte. 1993. Donorleukocyteinfusions for chronicmyeloidleukemia after
allogeneicbonemarrowtransplantation. J.Clin. Oncol. 11:513.
16. Goulmy, E. 1988. Minorhistocompatibility antigensin man andtheir role in transplantation.Transplant.Rev.2:29-53.
17.Perreault, C.,F.Decary,S.Brochu,M.Gyger,R.Belanger,and D.Roy.
1990. Minorhistocompatibility antigens. Blood.76:1269-1280.
18. Sondel, P. M., J. A.Hank,T.Wendel,B.Flynn,and M.J. Bozdech. 1983. HLA identicalleukemiacells and T cellgrowthfactor activatecytotoxicT cell
recognition ofminor locushistocompatibility antigensinvitro. J. Clin. Invest.
71:1779-1783.
19.Goulmy, E.,A.Termijtelen,B. A.Bradley,and J. J. van Rood. 1977. Y-antigen killing by T cells of women is restricted by HLA. Nature (Lond.). 266:544-545.
20.Goulmy, E.,J. W.Gratama,E.Blokland,F. E.Zwaan,andJ. J.vanRood. 1983. A minortransplantation antigendetected by MHC-restrictedcytotoxicT
lymphocytesduringgraft-versus-hostdisease. Nature(Lond.). 302:159-161.
21. vanderHarst, D.,E. Goulmy, J. H.F.,Falkenburg, Y. M. C. Kooy-Winkelaar, S. A. P. Luxemburg-Heijs, H.M. Goselink, and A. Brand. 1994.
Recognition ofminorhistocompatibility antigens onlymphocytic andmyeloid
leukemic cellsby cytotoxicT-cell clones. Blood. 83:1060-1066.
22.Falkenburg,J.H.,H. M.Goselink,D. vanderHarst, S.A.vanLuxemburg Heijs,Y. M.KooyWinkelaar,L. M.Faber,J. de Kroon, A.Brand,W. E.Fibbe,
R.Willemze,and E.Goulmy. 1991. Growthinhibition ofclonogenicleukemic precursorcellsbyminorhistocompatibilityantigen-specific cytotoxicT
lympho-cytes. J.Exp.Med. 174:27-33.
23. Voogt,P.J.,E.Goulmy,W. E.Fibbe,W. F.Veenhof,A.Brand,and J. H.
Falkenburg. 1988.MinorhistocompatibllityantigenH-Yis expressedonhuman
hematopoieticprogenitor cells.J.Clin.Invest.82:906-912.
24.Voogt,P.J.,E.Goulmy,W. F.Veenhof,M.Hamilton,W. E.Fibbe,J. J. vanRood,and J. H.Falkenburg. 1988. Cellularlydefined minorhistocompatibility antigensaredifferentially expressedonhumanhematopoietic progenitorcells. J.
Exp.Med 168:2337-2347.
25.Marijt,W. A.F.,W. F. J.Veenhof,E.Goulmy,R.Willemze,J.J.van
Rood,andJ. H. F.Falkenburg. 1993. Minor histocompatibility antigen
HA-1,-2,-4and HYspecificcytotoxicTcell clones inhibit humanhematopoieticprogenitor
cellgrowth byamechanism that isdependentondirect cell-cell contact.Blood.
82:3778-3785.
26.deBueger, M., A.Bakker,J. J. vanRood,F. van derWoude,and E.
Goulmy. 1992. Tissue distribution of human minorHistocompatibility antigens.
J. Immunol. 149:1788-1794.
27. vanEls,C.A.,J.D'Amaro,J.Pool,E.Blokland, A.Bakker,P.J. van
Elsen,J.J. vanRood,and E.Goulmy. 1992. Immunogenetics of humanminor histocompatibility antigens:theirpolymorphismand immunodominance.
Immuno-genetics.35:161-165.
28.Voogt,P.J.,W. E.Fibbe,W. A.Marijt,E.Goulmy,W. F.Veenhof,M.
Hamilton, A.Brand, F. E. Zwaan, R. Willemze,J. J. vanRood, and J. H. F.
Falkenburg. 1990.Rejectionofbone-marrowgraft byrecipient-derived cytotoxic Tlymphocytes againstminorhistocompatibility antigens.Lancet. 335:131-134. 29.Faber,L.M.,S. A. P. VanLuxemburg-Heijs, R.Willemze,and J. H. F.
Falkenburg. 1992.Generation of leukemia-reactive cytotoxic T lymphocyte clones from the HLA-identical bone marrow donor of a patient withleukemia. J. Exp. Med. 176:1283-1289.
30. Goulmy, E. 1982. HLA-A-B restriction ofcytotoxic T cells.In HLA-Typing: Methodology and Clinical Aspects. S. Ferrone and B. G.Solheim, editors. CRC Press, Inc. New York. 105-122.
31. Voogt, P. J., W.E. Fibbe, W. F. J. Veenhof, A. Brand, E.Goulmy, J. J. van Rood, and J. H. F.Falkenburg. 1987. Cell-mediated lysis of humanhematopoietic
progenitor cells.Leukemia. 1:427-431.
32.Fefer, A., K. M. Sullivan, P. Weiden, C. D. Buckner, G. Schoch, R.Storb, and E. D. Thomas. 1987.Graft versus leukemia effect in man: therelapserate ofleukemia is lowerafter allogeneic than after syngeneictransplantation. Prog.
Clin. Biol. Res. 244:401-408.
33.Jones, R. J., R. F. Ambinder, S. Piantadosi, and G. W. Santos. 1991. Evidence of agraft-versus-lymphoma effectassociated withallogeneicbone mar-rowtransplantation.Blood. 77:649-653.
34.Schreiber,K.L.,andJ. Forman. 1990. Effect ofgraft-versus-hostdisease onanti-tumorimmunity.J. Immunol.144:2018-2026.
35.Aizawa, S.,and T. Sado. 1991. Graft-versus-leukemiaeffect in
MHC-compatibleandincompatibleallogeneicbone marrowtransplantationof
radiation-induced,leukemia-bearingmice.Transplantation. 52:885-889.
36. Glass,B., L.Uharek,W.Gassmann,B.Focks,H.Bolouri,H.Loeffler,
and W. Ruchholtz.1992.Graft-versus-leukemia activity afterbone marrow
trans-plantationdoes notrequiregraft-versus-hostdisease. Ann. Hematol.64:255-259.
37.Kaminski,E., J.Hows,S.Man,P.Brookes,S.Mackinnon,T.Hughes, 0. Avakian,J. M.Goldman,and J. R.Batchelor. 1989.Prediction ofgraftversus host disease byfrequency analysisofcytytoxicTcells after unrelated donor bone marrowtransplantation.Transplantation. 48:608-613.
38.Irschick,E. U., F.Hladik,D.Niederwieser,W.Nussbaumer,E.Hooler,
E.Kaminski, and C. Huber. 1992. Studies on the mechanism of tolerance of
graft-versus-host disease inallogeneicbone marrowrecipientsatthe levelofcytotoxic
T-cell precursorfrequencies.Blood. 79:1622-1628.
39.Champlin, R.,W.Ho,J. Gajewski,S.Feig,M.Burnison,G.Holey,K.
Greenberg,I.Lee,I.Schmid,and J.Giorgi. 1990.Selectivedepletion ofCD8+ Tlymphocytesforpreventionofgraft-versus-hostdisease afterallogeneicbone marrowtransplantation.Blood.76:418-423.
40.Deacock, S., A. Schwarer,R. Batchelor, J. Goldman, and R. Lechler. 1992. Arapid limiting dilutionassayformeasuring frequenciesofalloreactive,
interleukin-2-producing T cells in humans. J. Immunol. Methods. 147:83-92. 41.Sosman,J.A., K. R.Oettel,S. D.Smith,J. A.Hank,P.Fisch,and P. M. Sondel. 1990. Specific recognition of human leukemic cells byallogeneic T cells: II.Evidence for HLA-D restricted determinants on leukemic cells that are crossreactive with determinants present on unrelatednonleukemiccells. Blood.
75:2005-2016.
42.Truitt,R.L. and A. A.Atasoylu.1991. Contribution of CD4+ and CD8+ Tcells tograft-versus-hostdisease andgraft-versus-leukemiareactivity after
trans-plantationof MHC-compatiblebone marrow. Bone marrow transplant. 8:51-58. 43.Okunewick,J.P.,D.L.Kociban,L. L.Machen, and M. J. Buffo. 1991. The role of CD4 and CD8 T cells in the graft-versus-leukemia response in Rauschermurineleukemia. Bone marrow transplant. 8:445-452.
44.Theobald,M., T.Nierle,D. Bunjes, R. Arnold, and H.Heimpel. 1992.
Host-specificinterleukin-2-secretingdonor T-cell precursors aspredictorsof acute
graft-versus-hostdiseasein bone marrow transplantation between HLA-identical
siblings.N.Engl.J.Med. 327:1613-1619.
45.Nierle,T., D.Bunjes,R. Arnold,H.Heimpel,and M. Theobald. 1993.
Quantitativeassessmentofposttransplant host-specific Interleukin-2-secreting
T-helpercell precursors inpatientswith and without acutegraft-versus-hostdisease after allogeneic HLA-identical sibling bone marrow transplantation. Blood. 81:841-848.
46.Theobald, M., and D. Bunjes. 1993.Pretransplantdetection of human minorhistocompatibility antigen-specificnaive and memory
interleukin-2-secre-tingTcellswithin class Imajor histocompatibility complex (MHC)-restricted
CD8+ and class II MHC-restricted CD4+ T-cell subsets.Blood. 82:298-306. 47.Busque,L., and D. G. Gilliland. 1993. Clonal evolution in acute myeloid leukemia. Blood. 82:337-342.
48.Fialkow,P.J.,J. W.Singer,W. H.Raskind,J. W.Adamson,R. J.Jacobson, I. D.Bernstein,L. W.Dow,V.Najfeld,and R. Veith. 1987. Clonal development, stem-celldifferentiation, and clinicalremissions in acute nonlymphocytic leuke-mia. N.Engl.J.Med.317:468-473.
