• Aucun résultat trouvé

Helper and cytotoxic lymphocyte responses to chronicmyeloid leukaemia: implications for adoptive immunotherapywith T cells

N/A
N/A
Protected

Academic year: 2021

Partager "Helper and cytotoxic lymphocyte responses to chronicmyeloid leukaemia: implications for adoptive immunotherapywith T cells"

Copied!
8
0
0

Texte intégral

(1)

British Journal of Haematology, 1996,92,587-594

Helper and cytotoxic lymphocyte responses to chronic

myeloid leukaemia: implications for adoptive immunotherapy with T cells

P. LEW ALLE, N. HENSEL, A. GUIMARAES,* D. COURIEL, Y. Z. JIANG, D. MA VROUDIS AND A. J. BARRETT Bone Marrow Transplant Unit, Hematology Branch, National Heart, Lung and Blood Institute, National Institutes of Health, Bethesda, Maryland, U.S.A., and *Servicio de Hematologia, Instituto Portugues de Oncologia 'Prancicso Gentil', Lisboa, Portugal Received 15 August 1995; accepted for publication 14 November 1995

Summary. We studied 19 patients with chronic myeloid leukaemia (CML) to characterize T-cell autologous responses to leukaemia. Third party stimulated alloresponses in mixed lymphocyte reactions were normal in all patients. Using the helper T -lymphocyte precursor frequency (HTLPI) assay we demonstrated a low frequency of T helper cells recognizing autologous leukaemia cells from CML blood (1/850000) and marrow (1/965000). However. similar frequencies to autologous bone marrow and lymphoid cells were also found in normal individuals. In 11 patients studied. HLA-matched siblings had a higher HTLPf to leukaemia than the patient's autologous response (P < 0.004). Alloresponse in mixed lymphocyte reactions (MLR). and autologous HTLPf to leukaemia. were comparable at all stages of disease progression and time from diagnosis. and independent of treatment given. In order to generate autologous cytotoxic

lymphocyte responses to CML. lymphocytes were stimulated with CML cells. Cultures were fed again with CML cells and examined for cytotoxicity after 21 d. Strong lymphokine- activated killer (LAK) cytotoxicity was found against K562 and Daudi cell lines. and to Epstein-Barr virus-transformed allogeneic and autologous lymphoblastoid cell lines. Auto- logous leukaemia cells were lysed to a lesser extent in only 3/13 patients tested. The findings indicate that immune reactivity in CML is normal but suggest that CML cells are relatively resistant to lysis by cytotoxic T cells. The results do not support the existence of a leukaemia-specific T-cell response in CML.

Keywords: chronic myeloid leukaemia, cytotoxic T cells, helper T cells, antileukaemia cytotoxicity, adoptive immu- notherapy.

Indirect evidence suggests that chronic myeloid leukaemia (CML) could be a target for T-cell attack. Allogeneic T cells confer a powerful protective effect against relapse of CML following BMT (reviewed in Butturini & Gale, 1987;

Horowitz et aI, 1990) and donor buffy coat transfusions induce stable remissions in patients with CML recurring after marrow transplantation (Kolb et aI, 1990; van Rhee et aI, 1994). Furthermore, one study indicates that syngeneic T cells may also exert a graft-versus-leukaemia (GVL) effect (Horowitz et aI, 1990). It has been hypothesized that the bcr/

abl fusion protein generated by the t(9;22) translocation in CML could serve as a major histocompatibility complex (MHC)-restricted leukaemia-specific antigen (Barrett & Jiang,

1992). It has also been suggested that autologous T cells recognize myeloid- as opposed to leukaemia-specific antigens on CML cells (Somasundaram et al, 1988a). Proliferation of T cells to autologous CML has been described (Khene et al, 1981; Somasundaram et ai, 1988a; Pawelec et al, 1989).

However, whilst MHC nonrestricted killing of autologous leukaemia has been demonstrated in CML (Somasundaram et al, 1988a, b; Pawelec et ai, 1990), there is no conclusive evidence for MHC-restricted leukaemia-specific autologous or allogeneic T-cell responses to CML (Pawelec et al, 1992;

Barrett et al, 1993) and no evidence to suggest that immune responses play any role in modifying disease progression in CML. The absence of an in vivo autologous antileukaemic response could be due either to a lack of specific recognition of leukaemia by T cells, or to defective immune responses associated with the leukaemic process or with treatment.

To further characterize autologous immune reactivity to CML, we studied proliferative and cytotoxic responses of CML

587

Correspondence: Dr A. J. Barrett, Bone Marrow Transplant Unit.

National Heart. Lung and Blood Institute, Bldg 10. Rm 7C103.

National institutes of Health. 9000 Rockville Pike, Bethesda.

MD 20892-1652, U.S.A.

~ 1996 Blackwell Science 1td

(2)

to autologous leukaemia and allogeneic targets and compared the response of CML patients to that of normal subjects. We demonstrate here that CML patients have normal immune responses but do not specifically recognize their leukaemia.

PATIENTS

Nineteen patients with Philadelphia chromosome (Ph + ) CML. ranging in age between 17 and 55 years. were studied (Table I). They represented a spectrum of disease progression with intervals from diagnosis to study ranging from 1 to 56 months. 13 were in chronic phase. four in accelerated phase and two in blast crisis of CML. Four (including one untreated

patient) were not receiving active treatment at least 1 month prior to study. Treatment prior to. or at the time of study.

included busulphan. interferon-a or hydroxyurea. Following informed consent of patients and their m.A-identical healthy marrow donors. blood and bone marrow samples were obtained during diagnostic or standard bone marrow transplant procedures.

MA TERIALS AND METHODS

Cell separation. Mononuclear cells were obtained by separation on Ficoll-Hypaque using a standard protocol and used immediately. or frozen at a controlled rate in medium containing 20% fetal calf serum (FCS) and 10%

dimethylsulphoxide (DMSO) in RPMI 1640 and stored in

liquid nitrogen. Frozen cells were rapidly thawed according to standard protocol and DNAse I 80 u/ml (Boehringer Mannheim, Mannheim, Germany) was added to bone marrow cell (BMC) samples. Peripheral blood mononuclear cells (PBMC) and BMC were washed three times and resuspended in RPMI-1640 medium with 107 heat-inacti- vated normal human AB serum (NABS) prior to use. PBMC and BMC were depleted of T lymphocytes using a CD2 monoclonal antibody coupled to magnetic beads (Dynal, Inc.). The cells were adjusted to a concentration of no more than 107 cells/ml in 2% fetal calf serum/phosphate-buffered saline with no magnesium or calcium (2% FCS/PBS) and placed in wet ice. Using the manufacturer's guidelines, the number of target cells was estimated and the appropriate volume of Dynabeads M-450 Pan-T (CD2) was added to the cell suspension after washing the beads twice with PHS. The bead-cell suspension was incubated for 30min at 2-4°C on a rotator to ensure bidirectional mixing. Then 5 ml cold 2%

FCS/PBS was added and the beads were collected with a rare earth magnet (Dynal MPC). The supernatant was transferred to a second tube and the beads were washed again with 2%

FCS/PBS. This supernatant was pooled with the first. This CD2 depleted cell suspension was pelleted and resuspended in 10% NABS/RPMI-1640 culture medium.

Cell lines. The Daudi and K562 cell lines were obtained from American Type Culture Collection (Rockville, Md.).

Lymphoblast cell lines (LCL) were generated from patient and donor PBMC by stimulation with phytohaemagglutinin (BACTO PHA-P from Difco Laboratories, Detroit, Mich..

Table I. Patient characteristics and treatment prior to study.

Treatment Interval

diagnosis to study (months) 24 18 5

36 8 7

Patient Previous

Current

(within 1 month of study) 1

2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19

M M F M M M M M M M M M M F M M M M F

CP CP CP CP CP CP CP

CP

CP CP CP CP CP

AP

AP AP AP BC BC

Hu IPN Hu Hu lPN, Hu IPN Bu,Hu Hu Bu,Hu lPN, Hu IPN IPN IPN lPN, Hu IPN lPN, Hu lPN, Hu lPN, Hu

Hu Hu Hu Hu

-

Hu Hu Hu IFN Hu Hu IFN Hu Hu Hu 6

3 56 47

5 12 12 12 12

36 6

18

Cp, CML in chronic phase; AP, accelerated phase; BC, blast crisis. IFN, alpha interferon 9 million to 35 million units weekly;

Hu, hydroxyurea 0.5-4.0 g daily; Bu, busulphan 2-4 mg daily.

@ 1996 Blackwell Science Ltd, British Journal of Haematology 92: 587-594

(3)

Autoreacting HTLP in CML 589 proliferation and; tritiated thymidine uptake in the 9.12 11-2 sensitive murine cell line. The method was adapted from the technique we previously described (Schwarer et aI, 1993).

Stimulator cells were irradiated 5000 cGy, adjusted to a concentration of 106 /ml and plated in 100 ji,l volumes/well in 96-well round-bottomed microtitre plates. Responders were added in 24 replicates of six dilutions at concentrations 4 x 104, 2 X 104, 1 X 104, 0'5 X 104, 0'25 X 104 and 0'125 x 104 cells/well. Stimulators alone were plated in 24 replicates. Third party responders (4 x 104 cells) with test stimulators and third party stimulators (105 cells) with test responders, were plated in 12 replicates each as controls.

Plates were incubated for 64 h at 370C in 5%CO2, frozen at -80°C to stop further proliferation and then thawed.

(Preliminary studies demonstrated that 11-2 assay of frozen/thawed cells produces comparable results to using live but irradiated cells as described in the original method [Schwarer et ai, 1993].) The prepared 11-2-dependent line was added at 2 x 103 cells/well in 25 ji,l volumes. After 20 h incubation at 37°C, 1 ji,Ci 3H-thymidine was added to each well in 25 ji,l volumes and the plates incubated for a further 18 h. Cells were harvested and 3H-thymidine uptake measured.

Calculation ofHTLPf. The background was defined as the mean control value +3 standard deviations of the 3H- thymidine uptake of the 24 control (stimulator only) wells.

Test wells greater than this value were considered positive for IL-2 production. From the fraction of negative wells at each dilution, the frequency of 11-2-producing lymphocytes was calculated using the maximum likelihood method.

Only results with a positive third party control response and a Chi-squared value of 15 or less were considered valid.

Cytotoxicity testing of cultured PBMC. Equal numbers of stimulators (5000 cGy irradiated) and responder PBMC were cultured in 10ml volumes in 25 cm2 tissue culture flasks at a final concentration of 2 x 106 cells/ml in RPMI-1640 supplemented as described above, with 15% human AB serum with or without the addition of 100 ill/ml interferon- a. Flasks were incubated at 37°C in 5% CO2 for 7-10d.

Cultures were then layered onto ficoll-hypaque, centrifuged, the interface harvested and then cultured in 10 ml fresh medium with 15% human AB serum, 5000 cGy irradiated stimulators and 500U/ml (Cetus) IL-2, in six-well (2ml) plates at cell concentrations no higher than 106 /ml, for a further 15 d. Cytotoxicity assays were performed using a microcytotoxicity technique previously described (W,ang et aI, 1993). Target cells (PBMC, LCL, Daudi cells, andK562 cells) were thawed and adjusted to a concentraton of 106 cells/mI. Calcein-AM dye (Molecular Probes, Medford, Ore., U.S.A.) was diluted 1:100 in phosphate-buffered saline and added to target cells at a concentration of 1 ml/106 target cells. The target cells were incubated in a 37°C water bath for 30 min. washed and resuspended in RPMI-1640 sup- plemented with 10% FCS. Viability (> 99%) was determined using trypan blue and the concentration was adjusted to 103 cells/well for cell lines and LCL, and 104 cells/well for PBMC.

Cytotoxicity assays were carried out in 60-well (40 ji,l/well) Terasaki plates (Robbins Scientific. Mountain View, Calif.,

U.S.A.).

2 JLg/ml. followed by infection with Epstein-Barr virus (EBV) derived from a supernatant of B95-8 marmoset cell line (courtesy of J. Maciejewski). EBV-transformed B-cell lines were thereafter maintained for a minumum of 2 months in RPMl supplemented with 10% FCS. These lines were morphologically 100% lymphoblasts expressing CDl9.

CD22 and negative for myeloid or T-cell markers (CD3.

CD56 CDl3. CD33. CD34 negative) (Tosato. 1994). The 9.12 interleukin-2 (IL-2)-sensitive murine cell line was generously provided by Dr Charles Shih (Medical College of Wisconsin). The 9.12 cell line was maintained at a density of 105cells/ml in RPMl-1640 medium supplemented With gentamicin. L-glutamine. 2 lllM sodium pyruvate. 2-mercap- toethanol. 2-ME. (5 x 10-5 M). 10% FCS. and 7'5% Rat T- StimTM With Con A (Collaborative Biomedical Products.

Bedford. Mass.. U.S.A.. 10 JLg/ml concanavlin A, 100 BRMP units/ml 11-2 activity) in 25 mm tissue culture flasks set upright. The cultures were fed three times weekly. New lines were established every 3 months from frozen stock. Before use in the helper T-lymphocyte precursor (HTLP) assay. the line was cultured in 10% FCS/RPMl-1640 medium without T-Stim for 4 h.

Antibodies. Fluoroscein (FITC) or phycoerythrin (PE) labelled monoclonal antibody conjugates of the folloyving specificities were used in our study: CD3/DR (clone UCHT1/

B8.12.2) and CD4/CD8 (clone 13B.8.2/B9.11) obtained from Immunotech, Inc. (Westbrook. Maine. U.S.A.), CD33 (clone P67.6) obtained from Becton Dickinson (San Jose.

Calif.. U.S.A.). lilA-DR (clone HL38) and CDl9 (clone SJ25-C1) obtained from Gen Trak (Plymouth Meeting. Pa..

U.S.A.).

Immunofluorescence staining procedures. 50 JLI of PBMC or BMC (107/ml) were incubated for 30 min in wet ice with 10 JLI of conjugated antibody. The cells were then washed twice with PES and analysed immediately or mixed with 4%

paraformaldehyde and analysed within 24 h. Isotype- specific FITC- or PH-conjugated negative controls were included in all experiments.

Flow cytometry. Flow cytometric assays were carried out using a Becton Dickinson BD F ACScan TM (Becton Dickinson Immunocytometry Systems. San Jose. Calif.. U.S.A.).

Proliferative assays. PBMC from CML patients and normal subjects were tested against CD2-depleted and non-depleted autologous and allogeneic PBMC and BMC. Assays were performed in 96-well round-bottom microtitre plates (Nunc.

Denmark) at a stimulator/responder ratio of 1:1 (2 x 105 cells/well in 200 JLI). Stimulators were irradiated 5000 cGy.

After 5 d incubation at 37°C in 5% CO2. 1 JLCi tritiated fH]thymidine/well was added. After a further 18 h incuba- tion the plates were harvested and 3H uptake was measured using a Wallac Betaplate 1205 scintillation counter (Wallac.

Turku. Finland).

Helper T-lymphocyte precursor frequency (HTLPf) assay.

This assay determines the frequency at limiting dilution ofT- cell progenitors capable of generating an 11-2 producing (T- helper) clo~ in response to a given stimulator. After a 3 d culture any well in which a T-helper cell precursor has responded to the stimulator by clonal expansion contains sufficient 11-2 to be detected by its ability to stimulate

1996

Blackwell Science Ltd. British Journal of Haematology 92: 587-594

(4)

phenotyped with monoclonal antibodies against CD3, CD4, CD8, CD19, CD33. and HLA DR (Table II). Responding cell populations of CML blood and marrow contained relatively fewer T cells than normal controls: The ficoll-hypaque separated fractions contained 69:i: 7% CD3 + cells (normal subjects PBL) and 45:i: 15% (patients PBL). The CD2 + depleted normal or CML marrow used to stimulate lymphocyte responses contained predominantly myeloid cells together with 14-19% B cells. (CD2 depletion of CML marrow resulted in a reduction of CD3 and CD3/DR cells from 24:i: 15% to 5:i: 6% and from 19:i: 8% to 3:i: 2%, respectively.)

Proliferative responses

The 5 d tritiated thymidine uptake of HLA-mismatched.

HLA-matched (donor) and patient autologous PBMC to non- depleted and CD2-depleted (myeloid-cell enriched) normal and leukaemic marrow and blood stimulators was compared (Table ill). There was zero to very low proliferation to autologous T -depleted or non-depleted CML. PBMC and BMC. However. the proliferation of CML responders to HLA mismatched stimulators was comparable to the response of normal third party cells to CML stimulators. Thus CML PBL fractions showed normal alloresponses but did not respond to their autologous CML stimulators. The strong third party response to CML cell fractions demonstrated that the

590 P. Lewalle et al

U.S.A.). Labelled targets were plated in 10ILI volumes.

Effector (E) PBMC were added to targets (T) in 10 ILl volumes at an E:T ratio of 50:1, 25:1 and 5:1, with 12 replicates per E:T ratio. Six wells containing target cells alone and six with medium alone were used to measure maximum (max f) and minimum (min f) fluorescent emission respectively. After plating, plates were centrifuged at 500- 700 rpm for 3 min to enhance cell-cell contact and incubated for 4 h at 37°C in 5% CO2, A working solution of ethidium bromide (EB) 2'5 mgfml was prepared and 15 ILl added to each ml of 10%

bovine haemoglobin quench. 5 ILl of haemoglobin quench- EB was added to each well. The plates were spun at 500 rpm for 3 min bcfore reading fluorescence using a Lambda Scan Fluoroscan (One Lambda, Canoga Pk, Calif.). The fluorescent emission is decreased in proportion to the killing of target cells. The percentage specific lysis was calculated as:

% specific lysis = 1 -[(mean exp emission -mean min f)j (mean max f -mean min f)] x 100

Statistics. Differences in proliferative assay between groups was calculated with a paired t test.

RESULTS

Immunophenotyping

PBMNC from CML patients and normal donor controls were

Table II. Surface phenotype of stimulators and responders used in MLR and HTLPf assay.

Percentage

of cell fraction (f:SD)

CD3 CD4

n

CDSCD19

CD3:3:

Stimulators CML PBMC Normal PBMC

T-depeleted CML marrow

T-depleted normal marrowResponders CML PBMC

Normal PBMC

191810

6

45

(15)69 (7) 5 (6) 7 (2)

31

(9)39 (8) 3 (5) 2 (3)

14

(7)31 (8) 0 (0) 1 (1)

11 (5) 14 (5) 14 (8) 19(8)

25

(14) 26 (8)60

(19) 58 (5)

1918

45 (15) 69 (7)

31

(9) 39 (8)

14 (7)31 (8)

Table ill. Mixed lymphocyte reactivity of CML patients to autologous and allogeneic (pooled third party) stimulators and response of third party to CML PBMC (paired t tests).

Stimulation index (mean)

n Significance

CML PBMC

}NS}p<

0'001

16101611

815 8

1'5 1'8 34,0 1'6 1'6 38,0 28'0 m.A-matched PBMC

NS Third-party PBMC

NS

@ 1996 Blackwell Science Ltd. British Journal of Haematology 92: 587-594

(5)

Autoreacting HTLP in CML

591

100001

'?

0 )(

Q:..J :J:I-

AUTOLOGOUS ALLOGENEIC Fig 2. Comparison of autologous and allogeneic response to PBMC from patients with CML. 11 paired H11.Pf are shown for the patient v patient (autologous) and HLA-identical sibling (allogeneic) using non-T-depleted PBMC as stimulators (P < 0'004, paired t test).

leukaemia cells were nevertheless capable of stimulating alloresponses.

Helper T lymphocyte precursor frequency

HTLPf assays of patient PBL against T -depleted PBMC or BMC stimulators were measured (Fig 1). The limit of detection of HTLP in this assay is a frequency of 1 in 2000000. Autoreacting HTLP were demonstrated in 10/16 PBMC and 6/12 BMC in frequencies ranging from 1/

125000 to 1/930000 (mean 1/842000:1: 700000 and 1/965000:1: 900000 respectively). A similar frequency of autoreacting HTLP also occurred in most normal individuals (range 1/114000 to 1/934000. mean PBMC 1/321000:1:

322000 and BMC 1/978000:1:954000). In 11 pairs of patients and their bone marrow donors. HTLPf of the HLA- identical donor was compared with autologous HTLPf to the leukaemic PBMC (Fig 2). On 9/11 occasions the HLA- identical alloresponder showed a higher HTLPf than the

autologous responder, The difference between the groups was significant (P < 0'004), In both CML patients and nomlal individuals there was no correlation between proliferation (stimulation index) and the HTLPf (r = 0'4), indicating that the two assays measured different parameters,

Relationship of T-cell responses to clinical parameters

To determine whether advancing disease or treatment with interferon-a or chemotherapy affected immune responsive- ness, we compared autologous HTLPf and allogeneic responses with clinical features (Table IV, Fig 3). There was no difference between MLR response or autologous HTLPf according to disease status. time from diagnosis, or Table IV. Comparison of disease and treatment variables with autologous HTLPf to PBMC-.md

BMC.

Autologous

lfI'LPf PBMCfrequency x 10-3

(mean:f:SD) -

Autologous HTLPf BMC frequency x 10-3 (mean:I: SD)

842::1: 7081580::1:

5711080::1:

741 761::1: 735

965 :i: 8991306:i:981 836:i:816

1283 :i: 982

1070:i:

791 830:i:655

1346:i: 828 569:i: 800 j.., ',:F:

Disease stage.

Chronic phase

Accelerated phaseJblast crisis Interval from diagnosis < 1 yearInterval

from diagnosis> 1 yearTreatment.

Interferon No treatment/other

.P > 0.05 in all comparisons (unpaired t tests).

996 Blackwell Science Ltd. British Journal of Haematology 92: 587-594

(6)

592

P. Lewalle et al

A Third party MLR

0 10 20 30 40 SO 60

months from diagnosis

Fig 3. Mixed lymphocyte reactivity to third-party stimulator and autologous HTLPfto PBMC in 18 patients presented according to time from diagnosis. Shaded area = mean :f: standard deviation for MLR and HTLPf results.

treatment with or without interferon. Immune responses and autologous HTLPf in CML therefore appeared to be unaffected by disease progression and treatment given.

Generation of cytotoxic T lymphocytes

Attempts were made to generate cytotoxic T-cell responses to autologous CML cells. PBMC were incubated with equal Table V. Cytotoxicity of lymphocyte cultures stimulated with autologous CML PBMC and 11-2.

Percentage lysis E: T 50: 1

Cell lines Autologous targets Allogeneic targets

PMBC

LCL PBMC LCL K562 Daudi

Patient

3 - 4

0

35

45

45

45j"

j

16 15 25 58 30 36

0 0 75 56 30

0

33

45 53 30 51 25 41 33 42 19 50 42 57

44 48 56 95 6

11

21 31 45 56 55 55 67 63 57 55 54 43 42

40 63 44 68 32 41 52 36 70 70 56 55

61

69

50

46 52 H -

11

17

30 0 0 28

37 17 0 2 0 0 0 0 0

19 - 18 - 5

+

-

57 48

-

20

61 0 0

38

16

6

10

3 4 15

+ = Interferon-a lOOujml.

1996 Blackwell Science Ltd. British Journal of Haematology 92: 587-594

(7)

numbers of irradiated autologous leukaemic PBMC- as stimulators and restimulated on days 7-10 with further irradiated CML cells and 11-2 500U/ml. This re-challenge induced further proliferation in 13 patients studied. Respon- der cells consisted of > 95% CD3 positive cells « 1 % CDl6 + /CD56 + cells) with a preponderance of CD4 cells (mean 66%) over CD8 cells (mean 27%). Mter 21 d of culture, cytotoxicity was measured against autologous leukaemia, autologous LCL, allogeneic PBMC and LCL, K562 cells, and Daudi cells. All lines tested were cytotoxic to K562 and Daudi cell lines (at an E:T ratio of 50:1, mean lysis 53% for Daudi cells and 49% for K562 cells), indicating LAK cell (MHC-unrestricted) cytotoxicity (Table V). There was 19-61% lysis of allogeneic LCL in 9/12 patients and 45-58% lysis of autologous LCL in 3/3 patients tested.

However, lysis of autologous CML cells was lower and less frequent: on only 3/13 occasions was lysis> 20% observed, compared with 6/8 occasions where allogeneic PBMC were lysed> 20%. This suggests relative resistance of CML cells to LAK killing. The addition of IFN-a reproducibly increased lysis against allogeneic LCL targets but gave inconsistent results against autologous and allogeneic PBMC targets and cell lines. In particular, IFN-a did not increase specific killing of the CML target. These findings indicated that repeated stimulation of responder lymphocytes by autologous CML cells in the presence of 11-2, with or without IFN-a, generated non-specific T-cell cytotoxicity.

DISCUSSION

There have been numerous attempts to detect T-cell reactivity to autologous leukaemia using proliferative assays usually by testing lymphocytes obtained in remission stimulated by cryopreserved leukaemia cells collected at diagnosis (Zarling et ai, 1976; Leventhal et aI, 1978; Lee &

Oliver, 1978; Ortaldo et aI, 1977; reviewed in Ashman et aI, 1986). In CML standard therapy only achieves a state of disease control, making it impossible to study a true remission state. Nevertheless, proliferative autologous responses to CML cells have been described (Khene et ai, 1981; Pawelec et ai, 1989, 1994; Somasundaram et aI, 1988a, b).

Interaction of CD4 cells with leukaemia antigens pre- sented via Ill.A Class II and with APC presenting leukaemia antigens is a critical part of the immune response to malignancy (Topalian, 1994; Ostrand & Rosenberg, 1994).

In order to characterize T-cell immune responses to CML, we used proliferation assays to measure third party and autologous responses. Proliferation assays confirmed that T-cell alloresponses in CML were normal and that CML cells were strong allogeneic targets. However, no bulk proliferative responses to CML were detected. To maximize the chance of detecting an autologous T-cell response to leukaemia, we therefore used the HTLPf assay which can detect the presence in the circulation of rare IL-2-producing T cells respoMing to specific stimuli. High frequencies reflect an established immune response which has given rise to a large compartment of memory cells containing HTLP. Low or undetectable frequencies indicate no prior recognition of

Autoreacting HTLP in CML 593 the stimulator (Sharrock et aI, 1990). The ability of the HTLPf assay to detect different degrees of compatibility in mixed lymphocyte nonreactive HLA-identical sibling pairs has recently been used successfully to predict GVHD after marrow transplantation. Between HLA-identical siblings HTLPf of donor to recipient range between 1/10000 and 1/1000000 cells (despite mutual MLR nonreactivity).

Frequencies> 1/100000 are associated with a significantly higher risk of severe GVHD following BMT, suggesting that only high HTLPfhave biological significance (Schwarer et aI, 1993; Theobald et al, 1992). Since HTLPf against allogeneic leukaemia cells are readily demonstrable (Hoffman et al, 1993), we argued that an effective T-helper cell stimulus from autologous leukaemia would produce a clonal expan- sion of leukaemia-specific T cells in vivo measurable as an increased autologous HTLPf. We found that an autologous HTLPf to PBMC or BMC was detectable, but only at frequencies well below 1/100000. Furthermore, a similar low level of autoreactive HTLP was demonstrated in normal individuals. In order to better define the response to normal marrow cells, autologous HTLPf to T -depleted BMC was tested. In both patients and normal subjects the HTLPf to BM was slightly lower than the HTLPf to PBMC. These findings do not support the existence of a leukaemia-directed T-cell response in CML, rather they suggest that the presence of autoreacting T -helper cells is a normal phenomenon. The significance of this low frequency of autoreacting helper T cells is unclear. It may represent a background frequency of tolerized T cells recognizing self antigens.

We went on to examine whether, despite the low HTLPf, it was possible in vitro to expand cytotoxic lymphocytes with specificity for CML. In vitro primed T-cell responses to CML predominantly involved proliferation of CD4, not CD8 cells.

We found normal LAK-cell production but showed that autologous CML cells were relatively resistant to LAK- mediated cytotoxicity. Other studies found defective LAK-cell generation in CML which was restored with interferon treatment and resistance of CML cells to LAK -mediated lysis (Pawelec et aI, 1990; Teichmann et al, 1992). Using culture conditions that have been effective in the generation of leukaemia-specific alloreacting cytotoxic CD8 T cells (Faber et aI, 1992), we were not able to generate T-cell lines exhibiting leukaemia-specific cytotoxicity. These results are consistent with other reports of cytotoxic T-cell lines from CML patients, showing MHC nonrestricted reactivity (Soma- sundaram et al, 1988a). Pawelec et al (1992) raised eight cytotoxic CD4 + clones with indiscriminate cytotoxicity to CML, PHA blasts and some allogeneic targets. Somasun- daram et al {1988a, b) generated CD8 clones cytotoxic to CML and normal marrow cells. They suggested that the response was against myeloid specific rather than leukae- mia-specific antigens.

In conclusion, although we were able to show the existence of autologous helper cells responding to CML in most patients, we found no evidence that this was a leukaemia-specific occurrence, nor that leukaemia-specific cytotoxic T-cell immune responses can be generated from such patients. Although we cannot exclude the possibility that CML patients rapidly lose T-cell antileukaemic function

1996 Blackwell Science Ltd, British Journal of Haematology 92: 587-594

(8)

594 P. Lewalle et al

after initial diagnosis and treatment. the stability of allogeneic and autoreactivity measured in this series makes it less likely that our findings would have been different if more untreated. newly diagnosed. patients had been studied. Our results suggest that autologous T-cell antileukaemia responses play little part in modifying the emergence or evolution of CML. Furthermore. T-cell clones generated in vitro for immunotherapeutic approaches may have limited antileukaemic specificity and efficacy.

ACKNOWLEDGMENT

P. Lewalle was supported by a grant from the Fondation Salus Sanguinis. Belgium.

Ostrand, S. & Rosenberg, S. (1994) Tumor immunotherapy: the tumor cell as antigen presenting cell. Current Opinion in Immunology, 6,722-727.

Pawelec, G., Ehninger, G. & Schmidt, H. (1990) Susceptibility of autologous target cells to lysis by lymphokine activated effectors from interferon treated chronic myelogenous leukemia patients.

Cancer Immunology and Immunotherapy, 32, 167-172.

Pawelec, G., Kalbacher, H. & Bruserud, O. (1992) Tumor-specific antigens revisited: presentation to the immune system of fusion peptides resulting solely from tumor-specific chromosomal translocations. Oncology Research, 4, 315-320.

Pawelec, G., Schmidt, H. & Busch, F. (1989) Antitumor activity in vitro in chronic myelogenous leukemia revealed after treating peripheral cells with cytosine arabinoside. Cancer Immunology and Immunotherapy, 29, 242-246.

Schwarer, A.P., Jiang, Y.Z., Brookes, P.A., Barrett, A.J., Batchelor, J.R., Goldman, J.M. & Lechler, R.I. (1993) Frequency of anti- recipient alloreactive helper T-cell precursors in donor blood and graft-versus-host disease after mA-identical sibling bone marrow transplantation. Lancet, 341, 203-205.

Sharrock, C.E.M., Kaminski, E. & Man, S. (1990) Limiting dilution analysis of human T-cells: a useful clinical tool. Immunology Today, II, 281-286.

Somasunduaram, R., Advani, S.H. & Gengal, G. (1988a) Prolifera- tion of cytotoxic effectors from chronic myeloid leukemia patients and cloning of cytotoxic T cells, Cancer Immunology and Immunotherapy, 27,183-189.

Somasundaram, R., Rao, S.G.A., Advani, S.H. & Gengal, S.G.

(1988b) In vitro generation of effector cells cytotoxic to autologous targets from' chronic myeloid leukemia patients in remission. Cancer Immunology and Immunotherapy, 27, 177-182.

Teichmann, J.V., Ludwig, W.-D. & Thiel, E. (1992) Cytotoxicity of IL-2 induced lymphokine activated killer (LAK) cells against human leukemia and augmentation of killing by interferon and tumor necrosis factor. Leukemia Research, 16,287-298.

Theobald, M., Nierle, T., Bunies, D., Arnold, R. & Heimpel, H. (1992) Host-specific interleukin-2-secreting donor T cell precursors as predictors of acute graft-versus-host disease in bone marrow transplantation between mA-identical siblings. New England Journal of Medicine, 327, 1614-1617.

Topalian, S.L. (1994) MHC II restricted tumor antigens and the role of CD4 + T cells in cancer immunotherapy. Current Opinion in Immunology, 6, 741-745.

Tosato, G. (1994) Generation of Epstein-Barr virus (BEV) immorta- lized B cell lines. Current Protocols in Immunology (ed. by J. E., Coligan, A. M. Kruisbeek, D. H. Margulies, E. M. Shevach and W. Strober), Vol. 1, p. 7.22.1. Library of Congress, Washington.

van Rhee, F., Feng, L., Cullis, J.O., Spencer, A., Cross, N.P.C., Chase, A., Garicichoa, B., Bungay, J., Barrett, A.J. & Goldman, J.M.

(1994) Relapse of chronic myeloid leukemia after allogeneic bone marrow transplantation: the case for giving donor leukocyte transfusions before the onset of hematological relapse. Blood, 83, 3377-3383.

Wang, X.M., Terasaki, P.I., Rankin, G.W., Chia, D., Zhong, H.P. &

Hardy, S. (1993) A new microcellular cytotoxicity test based on calcein AM release. Human Immunology, 37, 264-270.

Zarling, J.M., Raich, P.C., McKeough, M. & Bach, F.H. (1976) Generation of cytotoxic lymphocytes in vitro against autologous human leukaemia cells. Nature, 262, 691.

REFERENCES

Ashman, K.1., O'Keefe, E.D. & Michaels, S.R. (1986) Autologous responses to hwnan leukemia cells. Cancer Immunology and Immunotherapy, 22, 95-99.

Barrett, A.J., Guimariies, A., Cullis, J.O. & Goldman, J.M. (1993) Immunological characterisation of the twnor-specific junction on Philadelphia chromosome positive chronic myeloid leukemia.

Stem Cells, II, (Suppl. 3), 104-108.

Barrett, A.J. & Jiang, Y.Z. (1992) Immune responses to chronic myeloid leukelnia. Bone Marrow Transplantation, 9, 305'~311.

Butturini, A. & Gale, R.P. (1987) The role of T cells in preventing relapse in chronic myelogenous leukelnia. Bone Marrow

Transplantation, 2, 351-354.

Faber, 1.M., van 1uxemberg-Heijs, S.A.P., Willernze, R. & Falkenberg, J.H. (1992) Generation ofleukelnia~reactive cytotoxic T lympho- cyte clones from the HLA-identical bone marrow donor of a patient with leukemia. Journal ofExperimentaI Medicine, 176, 1283-1289.

Hoffman, T., Theobald, M., Bunjes, D., Weiss, M., Heimpel, H. &

Heit, W. (1993) Frequency of bone marrow T cells responding to HLA identical non-leukemia and leukelnic stimulators. Bone Marrow Transplantation, 12, 1-8.

Horowitz, M.M., Gale, R.P., Sondel, P.M., Goldman, J.M., Kersey, J., Kolb, H-J., Rimm, A.A., Ringden, 0., Rozman, C., Speck, B., Fruitt, R.1., Zwaan, F.E. & Bortin, M.M. (1990) Graft-versus-leukaelnia reactions after bone marrow transplantation. Blood, 75, 555-562.

Khene, A.G., Advani, S.H. & Gengal, S.G. (1981) In vitro generation of lymphocytotoxicity to autochthonous leukaelnia cells in chronic myelogenous leukaelnia. British Journal of Cancer, 43,13.

Kolb, H.J., Mitmuller, J., Clemm, Ch., Holler, E., Ledderose, G., Brehm, G., Heim, M. & Wilmans, W. (1990) Donor leukocyte transfusions for treatment of recurrent chronic myelogenous leukelnia in marrow transplant patients. Blood, 76, 2462-2465.

Lee, S.K. & Oliver, R.T.D. (1978) Autologous leukaelnia-specific T cell mediated lymphocytotoxicity in patients with acute myelo- genous leukelnia. Journal of Experimental Medicine, 147, 912-922.

Leventhal, B.G., Mirro, J. & Konior Yarbo, G.S. (1978) Immune reactivity to twnor antigens in leukelnia and lymphoma. Seminars in Hematology, 15, 157-163.

Ortaldo, J.R. & Bonnard, G.D. (1977) Presence of natural killer cell activity in cytotoxic lymphocytes from MLC-CML reactions.

Federation Proceedings, 35, 1325.

@ 1996 Blackwell Science Ltd. British Journal of Haematology 92: 587-594

Références

Documents relatifs

Malignant FL B cells are found admixed to specific stromal and immune cell subsets, in particular CD4 pos T cells displaying phenotypic features of follicular helper T cells (T FH

Evaluation of innate lymphoid cells, mucosal-associated invariant T cells and helper T cells in human autoimmune uveitisP. 52nd Annual Meeting of the French Society for Immunology,

Constructive objectivity was originally described by Brouwer in terms of the mental process dividing the present in two opposite parts, the before and the after,

Con- firming these observations, the inhibitor of iNOS, NMMA, abrogated Th-1 KDC- mediated tumor cell killing and the cytotoxic activity of DC generated from iNOS 2/2 mice

But for finite depth foliations (which have two sided branching), there are many examples where it is impossible to make the pseudo-Anosov flow transverse to the foliation [Mo5] and

Let X be a n-dimensional compact Alexandrov space satisfying the Perel’man conjecture. Let X be a compact n-dimensional Alexan- drov space of curvature bounded from below by

11,49,53 The similarities in cT FH cell dysregulation in pa- tients with HIV infection and PIDs identifies intrinsic (specific gene mutations) and extrinsic/environmental (chronic

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des