SEARCH

SEARCH BY CITATION

Keywords:

  • co-stimulatory molecules;
  • myeloid leukaemia;
  • CD80;
  • CD86;
  • first remission

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Surface CD80, CD86, HLA class I, HLA class II, CD11a and CD54 expression on monocytes, T cells and activated B cells
  6. Phenotyping of AML blast samples for CD80 and CD86 expression
  7. Expression of HLA class I, HLA class II, CD11a, CD54 and CD58 on AML blasts
  8. Patient details including cytogenetic analysis
  9. Correlation between expression of co-stimulatory or adhesion molecules analysed and duration of first remission
  10. Discussion
  11. Acknowledgments
  12. References

Summary. Many solid tumours have been shown to lack expression of either of the immune co-stimulatory molecules CD80 (B7·1) or CD86 (B7·2), which is thought to be one of the ways in which tumours may escape immune recognition. We have examined the surface expression of CD80, CD86, human leucocyte antigen (HLA) class I and II, CD11a, CD54, and CD58 on the blast cells from patients with acute myeloid leukaemia (AML) at presentation. CD80 was only rarely expressed on AML blasts and, in those leukaemic cells expressing CD80, the level of expression was low. In contrast, expression of CD86 was detected on the AML blasts in more than half of the samples tested and, in some cases, the level of expression was equivalent to that of mature monocytes and activated B lymphocytes. The percentage of leukaemic blasts expressing CD86 was higher in the M4 and M5 French–American–British (FAB) types, and expression of CD11a and HLA class II was higher in the M4 FAB type. There was no difference in expression of CD80, CD54, CD58, or HLA Class I between different FAB subgroups. There was no significant difference in duration of first remission with expression of CD80, CD86, CD11a, CD54 or HLA class II. However, when expression of CD80 and CD86 were considered together, a significantly longer duration of remission was found. We suggest that these molecules may play a role in immunosurveillance, resulting in prolonged remission in some patients treated for AML.

Immunotherapy is a potentially powerful tool in the fight against cancer. Immunotherapy protocols seek to harness the patient immune system to destroy tumour cells, with precision, while sparing non-malignant tissues. In addition, immunotherapy may engender a state of immunological memory that could provide a constant surveillance against recurrence of disease.

There are a number of reasons why leukaemia is an appropriate candidate for immunotherapy. There have been reports of spontaneous remissions in leukaemia (Dore et al, 1976), and graft-versus-leukaemia (GVL) activity is an established phenomenon as evidenced by the decreased risk of relapse after allogeneic transplantation, compared with autologous transplantation, demonstrating that acute myeloid leukaemia (AML) cells can be recognized by the immune system (Weiden et al, 1979). T-cell depletion of allogeneic marrow is associated with an increased risk of relapse, indicating both an immune mechanism of cure and a role for T lymphocytes in mediating GVL (Marmont et al, 1991). Donor leucocyte infusions are an established treatment for relapsed AML after allogeneic bone marrow transplantation (Kolb et al, 1990; Popat et al, 1995) although with less effect than is seen in chronic myeloid leukaemia, probably due, in part at least, to different disease kinetics (Barrett, 1997).

Although AML-specific tumour antigens have yet to be demonstrated, there are many potential targets present on AML blasts which could function in this capacity. These include peptides derived from the proteins derived from leukaemia-specific fusion genes resulting from the various relevant translocations (Dermine et al, 1995; Bosch et al, 1996), the products of mutated oncogenes (Peace et al, 1994), abnormal tumour suppressor genes (Noguchi et al, 1994), or overexpression of normal myeloid proteins such as myeloperoxidase or proteinase-3 (Brouwer et al, 1994; Dengler et al, 1995).

The major cellular mediators of antitumour immunity are thought to be T cells, which require two signals for their optimal activation and proliferation. Signal 1 is provided by antigen presented in a human leucocyte antigen (HLA) groove binding to a reciprocal T-cell receptor and signal 2 is provided by a co-stimulatory molecule interacting with its ligand on the T cell (Mueller et al, 1989). CD80 and CD86 are important and extensively studied co-stimulatory molecules expressed on ‘professional antigen presenting cells’ (APC) (Freeman et al, 1989; Azuma et al, 1993). APC express HLA class I and II molecules (Kern et al, 1986), and a number of other molecules with adhesion and co-stimulatory functions that are important in the interaction between T cells and APC (Van Seventer et al, 1990; Bachmann et al, 1997). These include lymphocyte function-associated antigen 1 (LFA-1; CD11a/18), intercellular adhesion molecule 1 (ICAM-1; CD54) and LFA-3 (CD58) (Damle et al, 1992; Hodge et al, 1999). Many of these molecules are also important for the interaction of natural killer (NK) cells with their targets (Lowdell et al, 2001). The failure of tumours to express appropriate co-stimulatory molecules has been postulated as one mechanism whereby they might escape immune recognition (Chen et al, 1992).

It has been shown that human AML blasts have variable expression of CD86 and frequently a lack of CD80 expression (Hirano et al, 1996). Subsequently, several murine models of AML have been used to show that an autologous immune response can be generated against the AML after CD80 or CD86 expression has been induced by gene transfer (Dunussi-Joannopoulos et al, 1996; Matulonis et al, 1996; Boyer et al, 1997; Dunussi-Joannopoulos et al, 1997), and that protection is conferred to subsequent exposure to the wild-type tumour (Matulonis et al, 1995). However, a previous study found that CD86 expression by human AML appeared to confer a poor prognosis (Maeda et al, 1998). This was not replicated in a recent publication, although expression of the co-stimulatory molecule CD40 and adhesion molecule CD11a correlated with a poor prognosis (Brouwer et al, 2001).

In this study, we undertook the phenotypic analysis of 61 AML samples for the presence of CD80, CD86, HLA class I, HLA class II, CD54, CD58 and CD11a. These were chosen to examine whether AML blasts have the capacity to express potential tumour antigens on HLA molecules, and what levels of the adhesion and co-stimulatory molecules required for NK- and T-cell activation were present. We then compared the level of expression of these molecules with French–American–British (FAB) classification types, cytogenetic risk groups and the duration of first remission (CR1). We chose duration of first remission as a surrogate marker of survival as those patients who relapse have a very poor outcome (Thalhammer et al, 1996; Lee et al, 2000) and many patients with relapsed AML will undergo allogeneic stem cell transplantation which deletes the autologous immune surveillance.

AML patient characteristics. After informed consent, bone marrow or peripheral blood samples were obtained from patients presenting with AML between 1988 and 2001. AML classification was performed according to standard FAB criteria (Bennett et al, 1976), and cytogenetic analysis was undertaken. The proportion of samples derived from bone marrow and blood were 0·43 and 0·57 respectively. The median age of the patients at diagnosis was 47 years (range 14–89 years). The cohort consisted of 27 females and 34 males. The FAB classification for these patients is described in Table I. Fifty-seven patients were treated with standard [mainly Medical Research Council (MRC) trial] AML chemotherapy, four with oral chemotherapy or supportive care alone. Patients not given potentially curative treatment were excluded from the survival analysis. Seven patients underwent an autologous stem cell transplant and 10 patients underwent a sibling HLA-identical allogeneic stem cell transplant. Because of the known improved leukaemia-free survival after allogeneic stem cell transplant, these patients were censured at the date of transplant for the purpose of leukaemia-free survival.

Table I.  Clinical features of the AML patient group studied.
  1. Of the patients that were intensively treated, seven failed to enter CR1. Karyotypic risk groups were assigned according to MRC AML 10 trial (Wheatley et al, 1999). AutoSCT, autologous haematopoietic stem cell transplantation; AlloSCT, allogeneic haematopoietic stem cell transplantation; WBC, white blood cell count; MDS, myelodysplastic syndrome; CML, chronic myeloid leukaemia.

Number of patients61
Male/female34/27
Age in years, mean (range)47 (14–89)
Mean WBC at diagnosis × 109/l (range)64·45 (0·8–300)
Chemotherapy40
Chemotherapy and AutoSCT 7
Chemotherapy and AlloSCT10
No intensive treatment 4
FAB morphology
 AML M0 3
 AML M1 4
 AML M211
 AML M3 5
 AML M424
 AML M5 8
 AML M6 2
 Arising on MDS or CML 4
Karyotypic risk group
 Good10
 Standard39
 Poor10
 Cytogenetic analysis failed 2

Cytogenetic risk group. Cytogenetic risk group was assigned according to the MRC AML 10 study criteria which identified three risk groups that were predictive for survival following intensive chemotherapy (Wheatley et al, 1999). The groups were defined as good [t(8;21), t(15;17) or inv(16) or FAB type M3 (irrespective of the presence of additional abnormalities)], standard (neither good nor poor), poor [monosomy of chromosomes 5 or 7, del(5q), abn(3q) or a complex karyotype (> 4 four abnormalities with no good-risk features].

Immunophenotyping.  Cells were analysed either directly after collection or after rapid thawing from storage in liquid nitrogen. The samples were washed twice in Hank's balanced salt solution (HBSS) in 12 × 75 mm round-bottomed tubes (Falcon) and pelleted by centrifugation at 400 g. After the final wash, the cells were incubated for 15 min with one of the following phycoerythrin (PE)- or fluorescein isothiocyanate (FITC)-conjugated monoclonal antibodies: 10 µl anti-CD80PE (clone L307·4; Becton Dickinson, Cowley, Oxfordshire, UK), 10 µl anti-CD86-FITC (clone FUN-1; Pharmingen, Cowley, Oxfordshire, UK), 10 µl anti-HLA class I-FITC (clone G46-2·6; Pharmingen), 10 µl anti-HLA-Class II-PE (clone L234; Becton Dickinson), 10 µl anti-CD11a-FITC (clone G25·2; Becton Dickinson), 10 µl anti-CD54-PE (clone LB-2; Becton Dickinson), 10 µl anti-CD58-PE (clone L306·4; Pharmingen) or 10 µl anti-CD40-PE (clone 5C3; Pharmingen). After incubation, the cells were washed once in HBSS and then resuspended in FACSFlow (Becton Dickinson) prior to flow cytometric analysis (FACScan; Becton Dickinson). Ten thousand viable cells, as determined by light scatter, were analysed from each sample using three-colour flow cytometry using a FACScan cytometer and cellquest software (Becton Dickinson). Cells lines and a few samples that were phenotyped, both fresh and after cryopreservation, showed no major differences in the level of antigen expression as found in previous studies (Campos et al, 1988; Maeda et al, 1998).

Expression of CD80, CD86, MHC class I and II, CD11a, and CD54 was also determined on resting T lymphocytes, resting monocytes and activated B lymphocytes from the peripheral blood of six normal donors. T cells were analysed by three-colour labelling with 10 µl anti-CD3-peridinin chlorophyll (PerCP) (clone SK7; Becton Dickinson) and appropriate antibodies as described above. FL3 photomultiplier (PMT) voltage was set after standardization of FL1 and FL2 to the bead preparation. The intensity of FL3 output was not formally standardized against a bead preparation as this parameter was used solely for identification of specific cell subsets rather than precise measurements of fluorescence intensity. The monocyte population was determined by gating and the expression on T lymphocytes by analysis of the CD3-positive population. B lymphocytes were analysed after 24 h incubation with pokeweed mitogen (Sigma, Poole, Dorset, UK) at a concentration of 5 µg/ml. The cells were washed and then incubated with 10 µl anti-CD19-FITC (clone 4G7; Becton Dickinson) and either 10 µl anti-CD80-PE or 10 µl anti-CD86-PE (clone IT2·2; Pharmingen) or 10 µl anti-CD54-PE (clone LB-2; Becton Dickinson) or 10 µl anti-HLA-Class II-PE, all as above, for 15 min at room temperature. To analyse CD11a expression on B cells, the lymphocytes activated with pokeweed mitogen were incubated with 10 µl anti-CD19-PE and 10 µl anti-CD11a-FITC, as above, for 15 min at room temperature. The cells were then analysed by flow cytometry.

Analysis of flow cytometric data.  Flow histograms were constructed for each leukaemic sample for CD80 and CD86 expression. The percentage of positive cells was calculated from each sample as compared with the negative control cells, and the median channel fluorescence was calculated from the positive cell population.

Percentage expression of HLA class I, HLA class II, CD11a and CD54 was also determined from the samples by comparison with isotype-labelled cells. Median channel fluorescence (MCF) was calculated in the positive cell population and compared with the MCF in the isotype cell population.

Statistical analysis.  For statistical analysis of data as indicated in the text and construction of correlation and survival curves, the graphpad prism computer programme was used (GraphPad Software, San Diego, USA).

Surface CD80, CD86, HLA class I, HLA class II, CD11a and CD54 expression on monocytes, T cells and activated B cells

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Surface CD80, CD86, HLA class I, HLA class II, CD11a and CD54 expression on monocytes, T cells and activated B cells
  6. Phenotyping of AML blast samples for CD80 and CD86 expression
  7. Expression of HLA class I, HLA class II, CD11a, CD54 and CD58 on AML blasts
  8. Patient details including cytogenetic analysis
  9. Correlation between expression of co-stimulatory or adhesion molecules analysed and duration of first remission
  10. Discussion
  11. Acknowledgments
  12. References

Six normal donors were used as a source of mononuclear cells to determine CD80, CD86, HLA class I, HLA class II, CD11a (LFA-1) and CD54 (ICAM-1) expression on T cells, monocytes and activated B cells. For the latter two cell types, these data give an indication of the expression of the molecules on antigen presenting cells. These data are presented in Table II.

Table II.  Expression of CD80, CD86, HLA class I, HLA class II, CD11a and CD54 on resting T cells, monocytes and activated B cells.
Antigen Cell type
MonocytesB cellsT cells
  1. Data for median channel fluorescence refer to mean (standard deviation).

CD80% positive cells 83 61  0
Median channel fluorescence421 (15·9)457 (11·3)165 (5·3)
CD86% positive cells 97 81  2
Median channel fluorescence553 (8·3)603 (19·7)175 (7·4)
HLA class I% positive cells100100100
Median channel fluorescence925 (33·6)939 (70·0)857 (56·8)
HLA class II% positive cells 99·5 97·5 15
Median channel fluorescence727 (135·8)825 (142·3)389 (133·6)
CD11a% positive cells 95 94·5 99·5
Median channel fluorescence701 (55·5)498 (89·0)628 (73·9)
CD54% positive cells 96·5 96 81·5
Median channel fluorescence580 (76·5)501 (100·6)354 (85·2)

Phenotyping of AML blast samples for CD80 and CD86 expression

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Surface CD80, CD86, HLA class I, HLA class II, CD11a and CD54 expression on monocytes, T cells and activated B cells
  6. Phenotyping of AML blast samples for CD80 and CD86 expression
  7. Expression of HLA class I, HLA class II, CD11a, CD54 and CD58 on AML blasts
  8. Patient details including cytogenetic analysis
  9. Correlation between expression of co-stimulatory or adhesion molecules analysed and duration of first remission
  10. Discussion
  11. Acknowledgments
  12. References

For fluorescence-activated cell sorting (FACS) analysis, the leukaemic blast population was identified with a gate on the forward-scatter/sideways-scatter (FSC/SSC) dot plot, which sometimes included T cells. The mean percentage of CD3+ cells was 2·1% (SD 2·13), indicating that the samples were not heavily contaminated with CD3-positive T cells. We, therefore, considered ≥ 5% expression of an antigen on AML blasts as positive. On testing with the unpaired t-test, there was no significant difference in expression of any of the antigens between peripheral blood- and bone-marrow-derived cells.

All AML samples were phenotyped for expression of CD80 and CD86. These data are presented in Table III. Five samples were positive for CD80 expression, with 5–27·5% of cells expressing the molecule. In the five positive samples, the mean MCF of the positive cells within the samples was 334·9 (SD 68·2) which, while significantly greater than that of resting T cells (MCF 165·0; P = 0·006 Student's t-test), was significantly lower than that of monocytes (MCF 421; P = 0·007) or activated B cells (MCF 457; P = 0·005).

Table III.  Expression of CD80, CD86, HLA class I, HLA class II, CD11a, CD54 and CD58 on up to 61 samples from patients with acute myeloid leukaemia.
AntigenPercentage positive cellsMedian channel fluorescence
Mean ± SDRangeMean ± SDRange
  1. Data are presented as both percentage of positive blasts and median channel fluorescence.

CD802·1 ± 6·00–27·5334·9 ± 68·1199–404
CD8626·7 ± 28·70–94·9394·1 ± 126·710·4–635
HLA class I98·9 ± 3·682·5–100785·7 ± 251·1611–1963
HLA class II86·0 ± 19·915–100692·0 ± 234·4201–995
CD11a68·7 ± 28·812–100454·0 ± 158·867–682·5
CD5475·6 ± 22·029–99420·0 ± 154·194·5–561·5
CD5879·3 ± 27·122·5–100352·6 ± 212·954·27–906·6

In contrast, 35 samples contained CD86-positive cells, ranging from 5·0 to 90·3%. The mean MCF of the positive cells was 394 (SD 126·7), which was significantly lower than that of monocytes or activated B cells (P < 0·001), but higher than that seen in the T cells (P < 0·001). However, in the samples with the highest expression, the MCF was approaching that seen on monocytes and activated B lymphocytes. The results for percentages of blasts expressing CD80 and CD86, according to FAB type, are presented in Fig 1.

image

Figure 1. Percentage of AML blasts expressing CD80 or CD86, according to FAB subtype. Post MDS, AML arising from myelodysplasia; CML BC, myeloid blast crisis of chronic myeloid leukaemia; M0–M6, FAB subtype of primary AML.

Download figure to PowerPoint

Expression of HLA class I, HLA class II, CD11a, CD54 and CD58 on AML blasts

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Surface CD80, CD86, HLA class I, HLA class II, CD11a and CD54 expression on monocytes, T cells and activated B cells
  6. Phenotyping of AML blast samples for CD80 and CD86 expression
  7. Expression of HLA class I, HLA class II, CD11a, CD54 and CD58 on AML blasts
  8. Patient details including cytogenetic analysis
  9. Correlation between expression of co-stimulatory or adhesion molecules analysed and duration of first remission
  10. Discussion
  11. Acknowledgments
  12. References

Expression of HLA class I and HLA class II is presented in Table III, with 82·5–100% of cells in all samples tested expressing HLA class I (mean 96·6%; SD 7·9) and the majority of samples having 100% of cells positive. The level of expression of HLA class I is comparable to that seen on normal T cells, B cells and monocytes as presented in Table II.

In contrast to HLA class I, the percentage of cells expressing HLA class II in each sample ranged from 15% to 100% (mean 73·23%; SD 29·2). However, despite there being fewer cells in each sample expressing HLA class II, the level of expression was again similar to that seen on monocytes.

The percentage of cells expressing CD11a ranged from 12% to 100% (mean 68·7%; SD 28·8), the percentage of cells expressing CD54 in each sample ranged from 29% to 99% (mean 75·6%; SD 22·0) and the percentage of cells expressing CD58 in each sample ranged from 22·5% to 100% (mean 79·3%; SD 27·1). The results for percentages of blasts expressing HLA class I, HLA class II, CD54, CD58 and CD11a, according to FAB type, are presented in Fig 2.

image

Figure 2. Percentage of AML blasts expressing HLA class I, HLA class II, CD54, CD58, CD11a or CD40, according to FAB subtype. Post MDS, AML arising from myelodysplasia; CML BC, myeloid blast crisis of chronic myeloid leukaemia; M0–M6, FAB subtype of primary AML.

Download figure to PowerPoint

Patient details including cytogenetic analysis

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Surface CD80, CD86, HLA class I, HLA class II, CD11a and CD54 expression on monocytes, T cells and activated B cells
  6. Phenotyping of AML blast samples for CD80 and CD86 expression
  7. Expression of HLA class I, HLA class II, CD11a, CD54 and CD58 on AML blasts
  8. Patient details including cytogenetic analysis
  9. Correlation between expression of co-stimulatory or adhesion molecules analysed and duration of first remission
  10. Discussion
  11. Acknowledgments
  12. References

Of the 61 patients analysed, four received either blood product support or oral palliative chemotherapy and were excluded from the survival analysis. Of the remaining 57 patients, 10 underwent allogeneic stem cell transplant and, because of the known GVL effect, were censured at the time of transplant with regard to duration of first remission. Seven patients underwent autologous stem cell transplant; these patients remained in the analysis in the post-transplant period. Seven patients were refractory to induction chemotherapy and, as immune mechanisms are unlikely to play a role in this setting, these patients were also excluded from the analysis.

Of the 61 patients, cytogenetic analysis was successful in 54 patients. Nine patients had good-risk cytogenetic abnormalities, 36 patients had standard-risk cytogenetics and nine patients had poor-risk cytogenetic abnormalities. To look for an association between antigen expression and cytogenetic risk group, the patients were divided into their cytogenetic risk groups according to whether the blasts expressed CD80, CD86, CD54, CD58 and CD11a. These results are expressed in Fig 3. No significant difference was seen between cytogenetic risk assignment and degree of expression of the molecules analysed.

image

Figure 3. Expression of CD80, CD86, CD40, CD54, CD58 or CD11a according to cytogenetic risk group assignment. Post MDS, AML arising from myelodysplasia; CML BC, myeloid blast crisis of chronic myeloid leukaemia; M0–M6, FAB subtype of primary AML.

Download figure to PowerPoint

Correlation between expression of co-stimulatory or adhesion molecules analysed and duration of first remission

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Surface CD80, CD86, HLA class I, HLA class II, CD11a and CD54 expression on monocytes, T cells and activated B cells
  6. Phenotyping of AML blast samples for CD80 and CD86 expression
  7. Expression of HLA class I, HLA class II, CD11a, CD54 and CD58 on AML blasts
  8. Patient details including cytogenetic analysis
  9. Correlation between expression of co-stimulatory or adhesion molecules analysed and duration of first remission
  10. Discussion
  11. Acknowledgments
  12. References

Relapse was defined as > 5% blasts found in the bone marrow of similar morphology and phenotype to the presentation blasts. The duration of first remission was calculated based on the interval from the date of complete remission until relapse, except in those patients undergoing allogeneic stem cell transplant who were censured at the date of the transplant. We analysed the duration of first remission at 5%, 10%, 20% and 50% for the molecules analysed, but only at 5% and 10% for CD80 as a result of the low numbers in this subgroup. In addition, we also assessed duration of first remission at 5%, 10% and 20% expression of CD80 or CD86. We found no significant difference in duration of first remission with expression of CD80, CD86, CD54, CD58, CD11a or HLA class II alone at the levels analysed. However, when we analysed leukaemia-free survival with expression of either CD80 and/or CD86 at 5% expression, a trend towards prolongation of first remission was seen (P = 0·056), and at 10% expression a significant prolongation of first remission was seen (P = 0·038). These results are presented in Fig 4.

image

Figure 4. Leukaemia-free duration curves for those patients whose AML blasts showed greater than 10% of blasts to express the B7 molecules CD80 or CD86, either singly or together.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Surface CD80, CD86, HLA class I, HLA class II, CD11a and CD54 expression on monocytes, T cells and activated B cells
  6. Phenotyping of AML blast samples for CD80 and CD86 expression
  7. Expression of HLA class I, HLA class II, CD11a, CD54 and CD58 on AML blasts
  8. Patient details including cytogenetic analysis
  9. Correlation between expression of co-stimulatory or adhesion molecules analysed and duration of first remission
  10. Discussion
  11. Acknowledgments
  12. References

This study found that CD80 was rarely expressed but that CD86 was expressed by over 50% of AML samples. Duration of first remission was significantly prolonged in those patients whose AML blasts expressed > 10% of either CD80 or CD86 or the combination.

We also found that all of the AML blast samples expressed both MHC class I and class II, and all the samples analysed expressed CD54, the percentage of cells positive in each sample ranging from 40% to 99%, with the protein expressed at a high density. All AML samples expressed the adhesion molecule CD11a, with between 10% and 100% of cells in each sample expressing the molecule, and in greater than 50% of samples at a similar intensity to that seen on monocytes. Therefore, most of the relevant surface molecules, with the exception of the co-stimulatory molecule CD80 and sometimes CD86, required for both CD4+ and CD8+ T-cell activation, were present on AML blasts.

There are two requirements for the successful delivery of a co-stimulatory signal, the density of expression of the co-stimulatory molecule on the cell surface and the number of those cells within a population. In the case of pre-B ALL, the expression of CD86 required to deliver a co-stimulatory signal and prevent the induction of anergy was found to be low (Cardoso et al, 1996). The numbers of positive cells required within a population is better understood. A recent study (Wendtner et al, 1997), showed that only very small numbers of CD80- or CD86-positive cells in a population are sufficient to stimulate T cells. Using a fixed number of T cells (5 × 104) as few as 300 CD80- or CD86-transduced cells were able to induce a T-cell proliferative response (Wendtner et al, 1997).

Our results confirm those of Hirano et al (1996), showing that CD80 is infrequently expressed on myeloid blasts. We found that CD86 was expressed in 35 of the 61 samples of AML tested, comparable to other reports (Hirano et al, 1996; Maeda et al, 1998). In those samples where CD80 or CD86 was expressed, it was at a intermediate level between that seen on T cells and monocytes. Therefore, we would anticipate that in some of these samples an efficient co-stimulatory signal could be delivered. In vitro assays using AML blasts demonstrated that 5% CD86 expression was sufficient to co-stimulate an allogeneic T-cell response which could be inhibited by cytotoxic T-lymphocyte–associated antigen-4 (CTLA4)-Fc (data not shown).

These data show that AML blasts can act as in vitro stimulators of allogeneic T cells. It has been previously shown that co-stimulation via CD80 on human AML blasts is able to stimulate both an allogeneic T-cell response, and an autologous T-cell response from samples obtained in remission (Notter et al, 2001). It is, therefore, possible that AML blasts expressing CD80 or CD86 might be able to provoke an in vivo autologous CTL response. In addition, NK cells, which are reported to have antileukaemic activity, have been demonstrated to exhibit antitumour cytotoxicity which is triggered through CD80 and CD86 interaction with CD28 (Galea-Lauri et al, 1999; Wilson et al, 1999; Lowdell et al, 2001). The correlation between CD80/CD86 and duration of CR1 after chemotherapy may, therefore, correspond to different patterns of immune recognition of the AML blasts. It is interesting that CD34+ AML blasts have been shown to express CD86 (Re et al, 2002), as this may indicate that immune recognition of AML blasts could occur at the level of a relatively primitive progenitor. This is of relevance in possible leukaemia vaccination strategies which rely on relatively inefficient viral transduction of AML blasts to induce expression of CD80 or CD86, although with adenoviral vectors we have been able to routinely exceed a 10% level of transduction from our cultured AML blasts (unpublished observations).

There are other variables that may be responsible for these findings. Many factors affect duration of CR1, including age, sex, white cell count at presentation, drug efflux pump activity and other, as yet undiscovered, factors relating to underlying cytogenetic abnormalities. There was a similar age and sex distribution in the two groups, and white cell count at presentation was higher in the group of patients with 10% or more of the blasts expressing CD86.

It has been demonstrated using tumorogenic cell lines that expression of CD80 enables silent epitopes on the tumour cells to be recognized, facilitating the generation of a specific antitumour immune response (Johnston et al, 1996). Indeed, co-stimulation through CD28 has been shown to decrease the number of T-cell receptors on the T cell which need to bind with the appropriate antigen (Bachmann et al, 1997). It is possible that with the higher proportion of CD80/CD86-positive AML blasts, an antileukaemic immune response may have been induced against silent epitopes expressed on them.

There have been two previous studies demonstrating similar in vitro data to our own, but in one, CD86 expression appeared to confer a poor prognosis (Maeda et al, 1998) and, in the second, no effect on clinical outcome was observed (Costello et al, 1998). This would be in contrast to the murine models of AML where both CD80 and CD86 expression enabled immune-mediated rejection and conferred protection to subsequent exposure, although the effect with CD86 was less marked (Matulonis et al, 1996).

We (data not shown) and others (Matulonis et al, 1996) have shown that those AML blast samples which had less than 10% of cells expressing CD86 were still able to deliver a co-stimulatory signal capable of inducing allogeneic T-cell proliferation and interleukin 2 production of similar levels to those samples where greater than 60% of AML cells expressed CD86 (data not shown). The duration of CR1 when 10% CD80 and/or CD86 expression is present is significantly prolonged. These data suggest that strategies for the creation of leukaemia vaccines by inducing the expression of co-stimulatory molecules on AML blasts are worthwhile and that not all of the AML cells in a population need to be modified for effective induction of an immune-mediated antileukaemia effect.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Surface CD80, CD86, HLA class I, HLA class II, CD11a and CD54 expression on monocytes, T cells and activated B cells
  6. Phenotyping of AML blast samples for CD80 and CD86 expression
  7. Expression of HLA class I, HLA class II, CD11a, CD54 and CD58 on AML blasts
  8. Patient details including cytogenetic analysis
  9. Correlation between expression of co-stimulatory or adhesion molecules analysed and duration of first remission
  10. Discussion
  11. Acknowledgments
  12. References
  • Azuma, M., Ito, D., Yagita, H., Okumura, K., Phillips, J.H., Lanier, L.L. & Somoza, C. (1993) B70 antigen is a second ligand for CTLA-4 and CD28. Nature, 366, 7679.
  • Bachmann, M.F., Mckall-Faienza, K., Schmits, R., Bouchard, D., Beach, J., Speiser, D.E., Mak, T.W. & Ohashi, P.S. (1997) Distinct roles for LFA-1 and CD28 during activation of naive T cells: adhesion versus costimulation. Immunity, 7, 549557.
  • Barrett, A.J. (1997) Mechanisms of the graft-versus-leukemia reaction. Stem Cells, 15, 248258.
  • Bennett, J.M., Catovsky, D., Daniel, M.T., Flandrin, G., Galton, D.A., Gralnick, H.R. & Sultan, C. (1976) Proposals for the classification of the acute leukaemias. French–American–British (FAB) co-operative group. British Journal of Haematology, 33, 451458.
  • Bosch, G.J., Joosten, A.M., Kessler, J.H., Melief, C.J. & Leeksma, O.C. (1996) Recognition of BCR-ABL positive leukemic blasts by human CD4+ T cells elicited by primary in vitro immunization with a BCR-ABL breakpoint peptide. Blood, 88, 35223527.
  • Boyer, M.W., Vallera, D.A., Taylor, P.A., Gray, G.S., Katsanis, E., Gorden, K., Orchard, P.J. & Blazar, B.R. (1997) The role of B7 costimulation by murine acute myeloid leukemia in the generation and function of a CD8+ T-cell line with potent in vivo graft-versus-leukemia properties. Blood, 89, 34773485.
  • Brouwer, E., Stegeman, C.A., Huitema, M.G., Limburg, P.C. & Kallenberg, C.G. (1994) T cell reactivity to proteinase 3 and myeloperoxidase in patients with Wegener's granulomatosis (WG). Clinical and Experimental Immunology, 98, 448453.
  • Brouwer, R.E., Hoefnagel, J., Van Der, B., Jedema, I., Zwinderman, K.H., Starrenburg, I.C., Kluin-Nelemans, H.C., Barge, R.M., Willemze, R. & Falkenburg, J.H. (2001) Expression of co-stimulatory and adhesion molecules and chemokine or apoptosis receptors on acute myeloid leukaemia: high CD40 and CD11a expression correlates with poor prognosis. British Journal of Haematology, 115, 298308.
  • Campos, L., Guyotat, D., Larese, A., Archimbaud, E., Mazet, L., Ehrsam, A. & Fiere, D. (1988) Expression of immunological markers on leukemic cells before and after cryopreservation and thawing. Cryobiology, 25, 1822.
  • Cardoso, A.A., Schultze, J.L., Boussiotis, V.A., Freeman, G.J., Seamon, M.J., Laszlo, S., Billet, A., Sallan, S.E., Gribben, J.G. & Nadler, L.M. (1996) Pre-B acute lymphoblastic leukemia cells may induce T-cell anergy to alloantigen. Blood, 88, 4148.
  • Chen, L., Ashe, S., Brady, W.A., Hellstrom, I., Hellstrom, K.E., Ledbetter, J.A., Mcgowan, P. & Linsley, P.S. (1992) Costimulation of antitumor immunity by the B7 counterreceptor for the T lymphocyte molecules CD28 and CTLA-4. Cell, 71, 10931102.
  • Costello, R.T., Mallet, F., Sainty, D., Maraninchi, D., Gastaut, J.A. & Olive, D. (1998) Regulation of CD80/B7–1 and CD86/B7–2 molecule expression in human primary acute myeloid leukemia and their role in allogenic immune recognition. European Journal of Immunology, 28, 90103.
  • Damle, N.K., Klussman, K., Linsley, P.S. & Aruffo, A. (1992) Differential costimulatory effects of adhesion molecules B7, ICAM-1, LFA-3, and VCAM-1 on resting and antigen-primed CD4+ T lymphocytes. Journal of Immunology, 148, 19851992.
  • Dengler, R., Munstermann, U., Al-Batran, S., Hausner, I., Faderl, S., Nerl, C. & Emmerich, B. (1995) Immunocytochemical and flow cytometric detection of proteinase 3 (myeloblastin) in normal and leukaemic myeloid cells. British Journal of Haematology, 89, 250257.
  • Dermine, S., Molldrem, J., Parker, K., Jiang, Y., Marvroudis, D., Hensel, N., Couriel, D., Mahoney, M., Coligan, J. & Barrett, J. (1995) Human CD8+ T lymphocytes recognise the fusion region of BCR/ABL hybrid protein present in chronic myelogenous leukaemia. Blood, 86, 158a.
  • Dore, J.F., Bertoglio, J. & Guibout, C. (1976) Antigens associated with human leukemia. Bulletin Du Cancer, 63, 353370.
  • Dunussi-Joannopoulos, K., Weinstein, H.J., Nickerson, P.W., Strom, T.B., Burakoff, S.J., Croop, J.M. & Arceci, R.J. (1996) Irradiated B7–1 transduced primary acute myelogenous leukemia (AML) cells can be used as therapeutic vaccines in murine AML. Blood, 87, 29382946.
  • Dunussi-Joannopoulos, K., Krenger, W., Weinstein, H.J., Ferrara, J.L. & Croop, J.M. (1997) CD8+ T cells activated during the course of murine acute myelogenous leukemia elicit therapeutic responses to late B7 vaccines after cytoreductive treatment. Blood, 89, 29152924.
  • Freeman, G.J., Freedman, A.S., Segil, J.M., Lee, G., Whitman, J.F. & Nadler, L.M. (1989) B7, a new member of the Ig superfamily with unique expression on activated and neoplastic B cells. Journal of Immunology, 143, 27142722.
  • Galea-Lauri, J., Darling, D., Gan, S.U., Krivochtchapov, L., Kuiper, M., Gaken, J., Souberbielle, B. & Farzaneh, F. (1999) Expression of a variant of CD28 on a subpopulation of human NK cells: implications for B7-mediated stimulation of NK cells. Journal of Immunology, 163, 6270.
  • Hirano, N., Takahashi, T., Ohtake, S., Hirashima, K., Emi, N., Saito, K., Hirano, M., Shinohara, K., Takeuchi, M., Taketazu, F., Tsunoda, S., Ogura, M., Omine, M., Saito, T., Yazaki, Y., Ueda, R. & Hirai, H. (1996) Expression of costimulatory molecules in human leukemias. Leukemia, 10, 11681176.
  • Hodge, J.W., Sabzevari, H., Yafal, A.G., Gritz, L., Lorenz, M.G. & Schlom, J. (1999) A triad of costimulatory molecules synergize to amplify T-cell activation. Cancer Research, 59, 58005807.
  • Johnston, J.V., Malacko, A.R., Mizuno, M.T., Mcgowan, P., Hellstrom, I., Hellstrom, K.E., Marquardt, H. & Chen, L. (1996) B7-CD28 costimulation unveils the hierarchy of tumor epitopes recognized by major histocompatibility complex class I-restricted CD8+ cytolytic T lymphocytes. Journal of Experimental Medicine, 183, 791800.
  • Kern, D.E., Klarnet, J.P., Jensen, M.C. & Greenberg, P.D. (1986) Requirement for recognition of class II molecules and processed tumor antigen for optimal generation of syngeneic tumor-specific class I-restricted CTL. Journal of Immunology, 136, 43034310.
  • Kolb, H.J., Mittermuller, J., Clemm, C., Holler, E., Ledderose, G., Brehm, G., Heim, M. & Wilmanns, W. (1990) Donor leukocyte transfusions for treatment of recurrent chronic myelogenous leukemia in marrow transplant patients. Blood, 76, 24622465.
  • Lee, S., Tallman, M.S., Oken, M.M., Cassileth, P.A., Bennett, J.M., Wiernik, P.H. & Rowe, J.M. (2000) Duration of second complete remission compared with first complete remission in patients with acute myeloid leukemia. Eastern Cooperative Oncology Group. Leukemia, 14, 13451348.
  • Lowdell, M.W., Lamb, L., Hoyle, C., Velardi, A. & Prentice, H.G. (2001) Non-MHC-restricted cytotoxic cells: their roles in the control and treatment of leukaemias. British Journal of Haematology, 114, 1124.
  • Maeda, A., Yamamoto, K., Yamashita, K., Asagoe, K., Nohgawa, M., Kita, K., Iwasaki, H., Ueda, T., Takahashi, A. & Sasada, M. (1998) The expression of co-stimulatory molecules and their relationship to the prognosis of human acute myeloid leukaemia: poor prognosis of B7-2-positive leukaemia. British Journal of Haematology, 102, 12571262.
  • Marmont, A.M., Horowitz, M.M., Gale, R.P., Sobocinski, K., Ash, R.C., Van Bekkum, D.W., Champlin, R.E., Dicke, K.A., Goldman, J.M. & Good, R.A. (1991) T-cell depletion of HLA-identical transplants in leukemia. Blood, 78, 21202130.
  • Matulonis, U.A., Dosiou, C., Lamont, C., Freeman, G.J., Mauch, P., Nadler, L.M. & Griffin, J.D. (1995) Role of B7–1 in mediating an immune response to myeloid leukemia cells. Blood, 85, 25072515.
  • Matulonis, U., Dosiou, C., Freeman, C., Lamont, C., Mauch, P., Nadler, L.M. & Griffin, L.D. (1996) B7–1 is superior to B7–2 costimulation in the induction and maintenance of T cell-mediated antileukaemia immunity. Further evidence that B7–1 and B7–2 are functionally distinct. Journal of Immunology, 156, 11261131.
  • Mueller, D.L., Jenkins, M.K. & Schwartz, R.H. (1989) Clonal expansion versus functional clonal inactivation: a costimulatory signalling pathway determines the outcome of T cell antigen receptor occupancy. Annual Review of Immunology, 7, 445480.
  • Noguchi, Y., Chen, Y.T. & Old, L.J. (1994) A mouse mutant p53 product recognized by CD4+ and CD8+ T cells. Proceedings of the National Academy of Sciences of the United States of America, 91, 31713175.
  • Notter, M., Willinger, T., Erben, U. & Thiel, E. (2001) Targeting of a B7–1 (CD80) immunoglobulin G fusion protein to acute myeloid leukemia blasts increases their costimulatory activity for autologous remission T cells. Blood, 97, 31383145.
  • Peace, D.J., Smith, J.W., Chen, W., You, S.G., Cosand, W.L., Blake, J. & Cheever, M.A. (1994) Lysis of ras oncogene-transformed cells by specific cytotoxic T lymphocytes elicited by primary in vitro immunization with mutated ras peptide. Journal of Experimental Medicine, 179, 473479.
  • Popat, U., Sekhar, M., Prentice, H.G., Wycherley, J., Lawler, M. & Jones, F. (1995) Donor leucocyte infusions for relapsed acute leukaemia post BMT. British Journal of Haematology, 89, 5a.
  • Re, F., Arpinati, M., Testoni, N., Ricci, P., Terragna, C., Preda, P., Ruggeri, D., Senese, B., Chirumbolo, G., Martelli, V., Urbini, B., Baccarani, M., Tura, S. & Rondelli, D. (2002) Expression of CD86 in acute myelogenous leukemia is a marker of dendritic/monocytic lineage. Experimental Hematology, 30, 126134.
  • Thalhammer, F., Geissler, K., Jager, U., Kyrle, P.A., Pabinger, I., Mitterbauer, M., Gisslinger, H., Knobl, P., Laczika, K., Schneider, B., Haas, O.A. & Lechner, K. (1996) Duration of second complete remission in patients with acute myeloid leukemia treated with chemotherapy: a retrospective single-center study. Annals of Hematology, 72, 216222.
  • Van Seventer, G.A., Shimizu, Y., Horgan, K.J. & Shaw, S. (1990) The LFA-1 ligand ICAM-1 provides an important costimulatory signal for T cell receptor-mediated activation of resting T cells. Journal of Immunology, 144, 45794586.
  • Weiden, P.L., Flournoy, N., Thomas, E.D., Prentice, R., Fefer, A., Buckner, C.D. & Storb, R. (1979) Antileukemic effect of graft-versus-host disease in human recipients of allogeneic-marrow grafts. New England Journal of Medicine, 300, 10681073.
  • Wendtner, C.M., Nolte, A., Mangold, E., Buhmann, R., Maass, G., Chiorini, J.A., Winnacker, E.L., Emmerich, B., Kotin, R.M. & Hallek, M. (1997) Gene transfer of the costimulatory molecules B7–1 and B7–2 into human multiple myeloma cells by recombinant adeno-associated virus enhances the cytolytic T cell response. Gene Therapy, 4, 726735.
  • Wheatley, K., Burnett, A.K., Goldstone, A.H., Gray, R.G., Hann, I.M., Harrison, C.J., Rees, J.K., Stevens, R.F. & Walker, H. (1999) A simple, robust, validated and highly predictive index for the determination of risk-directed therapy in acute myeloid leukaemia derived from the MRC AML 10 trial. United Kingdom Medical Research Council's Adult and Childhood Leukaemia Working Parties. British Journal of Haematology, 107, 6979.
  • Wilson, J.L., Charo, J., Martin-Fontecha, A., Dellabona, P., Casorati, G., Chambers, B.J., Kiessling, R., Bejarano, M.T. & Ljunggren, H.G. (1999) NK cell triggering by the human costimulatory molecules CD80 and CD86. Journal of Immunology, 163, 42074212.