Skip to main content
NCBI home page
Cancer Immunology, Immunotherapy : CII logoLink to Cancer Immunology, Immunotherapy : CII
. 2004 Jan 23;53(6):510–518. doi: 10.1007/s00262-003-0485-5

Dendritic cells are dysfunctional in patients with operable breast cancer

Sukchai Satthaporn 1, Adrian Robins 2, Wichai Vassanasiri 3, Mohamed El-Sheemy 3,4, Jibril A Jibril 3, David Clark 3, David Valerio 3, Oleg Eremin 1,3,5,✉
PMCID: PMC11032822  PMID: 14740176

Abstract

Background: Dendritic cells (DCs) play a crucial role in presenting antigens to T lymphocytes and inducing cytotoxic T cells. DCs have been studied in patients with breast cancer to define the factors leading to failure of an effective systemic and locoregional anticancer host response. Methods: Purified DCs were obtained from peripheral blood (PB) and lymph nodes (LNs) of women with operable breast cancer, using immunomagnetic bead selection. The stimulatory capacity of DCs in the allogeneic mixed leukocyte reaction (MLR) and autologous T cell proliferation test (purified protein derivative (PPD) as stimulator), the expression of surface markers on DCs and the production of cytokines in vitro by DCs from patients with operable breast cancer and from healthy donors (controls) were studied. Results: 70–75% purified DCs were isolated from PB and LNs. PBDCs and LNDCs from patients with operable breast cancer demonstrated a reduced capacity to stimulate in an MLR, compared with PBDCs from normal donors (p<0.01). Autologous T cell proliferation in patients had a decreased ability to respond to PPD, when compared with controls (p<0.01). However, T cells from patients responded as well as control T lymphocytes in the presence of control DCs. PBDCs and LNDCs from patients expressed low levels of HLA-DR and CD86, and induced decreased interleukin-12 (IL-12) secretion in vitro, compared with DCs from normal donors (p<0.01). Conclusion: These data suggest a defective DC function in patients with operable breast cancer. Switched-off DCs in patients with early breast cancer and decreased IL-12 production may be important factors for progressive tumour growth.

Keywords: Dendritic cells, Breast cancer, Lymph nodes

Introduction

Breast cancer is the most common cancer in women in the developed world. In the UK, which has one of the highest incidence and mortality rates for breast cancer, 1 in 12 women will develop this disease in their lifetime [14]. In humans, most solid tumours appear to successfully evade host defenses and, in particular, fail to elicit a clinically significant and therapeutically relevant anticancer response. Tumour growth is a complex process resulting in uncontrolled growth of cells and tissues, and breast cancer is no exception. Induction of effective tumour immunity can be viewed as a three-step process that includes appropriate presentation of tumour-associated antigens (TAAgs), selection and activation of TAAg-specific T cells as well as non-Ag-specific effectors, and homing of TAAg-specific T cells to the tumour site and effective elimination of tumour cells [39, 40, 46]. Malignant growths may escape immune surveillance due to changes in, and modulation of, these various processes, which contributes to their uncontrolled growth, eventually leading to death of the host. However, a key element in the induction of the anticancer immune response, namely Ag presentation to T cells in a tumour-bearing host (human and animal), has been poorly documented and inadequately studied.

Dendritic cells (DCs), originally identified by Steinman and Cohn (1973), represent the pacemakers of the immune response [47]. Dendritic cells are highly specialized and very effective Ag-presenting cells (APCs) whose primary function is to capture, process and present Ags to naïve T cells in conjunction with major histocompatibility complex (MHC) and costimulatory molecules [4]. Immature DCs capture Ag at peripheral sites and then migrate to T cell areas in secondary lymphoid tissues such as the spleen and lymph nodes, where the activated DCs attract T cells by releasing chemokines and maintain the viability of recirculating T lymphocytes [43]. Although dysfunction of DCs in malignancy is believed to occur, the underlying mechanisms are poorly defined and appear to be heterogeneous in different cancers [19, 22, 38, 44, 49]. To date, there have been few studies published in women with breast cancer. The investigation of DC function in women with breast cancer is important for a better understanding of the key components involved and their modulation in progressive tumour growth in such women. This should provide a better biological framework on which to develop more effective and long-term beneficial therapies. The aims of this study, therefore, were to critically examine DCs obtained from peripheral blood (PB) and regional tumour-draining lymph nodes (LNs) of patients with operable breast cancer. The allogeneic mixed leukocyte reaction (MLR) and the purified protein derivative (PPD) stimulation assay can be regarded as in vitro surrogate measures for Ag-pulsed DCs in vivo in women with breast cancer. To clarify further the possible mechanisms underlying functional defects of putative anticancer defences in breast carcinoma, DC phenotype surface markers (HLA-DR, CD40, CD83 and CD86) and the capacity of DCs to induce and produce cytokines (interleukin-10 (IL-10) and IL-12) in vitro were evaluated.

Materials and methods

Patients

Twenty-eight women with breast carcinoma (T1–3, N0–1 and M0) presenting to the Lincoln Breast Unit gave signed informed consent and were enrolled in the study, as were ten healthy normal donors (controls). Each patient was asked to donate 40 ml of blood prior to undergoing surgery to the breast (mastectomy or wide local excision) and axilla (axillary sample ≥4 nodes removed). Five of the 28 patients had large (>3 cm) or locally advanced (T3, N1) breast cancers and had received neoadjuvant chemotherapy (adriamycin (60 mg/m2) and cyclophosphamide (600 mg/m2), three times weekly for six courses) prior to undergoing surgery. Following removal of the breast tumour and lymph nodes, one or two of the LNs were transected and placed in tissue culture medium (see below). The remaining halves of LNs and the other LNs were sent for histological assessments. Tissues were obtained by prior agreement with the pathologists and using an establish protocol. Approval for the study was given by the North Lincolnshire Ethics Committee. This was a collaborative project between the United Lincolnshire NHS Trust and the University of Nottingham. The various analyses and assays were carried out in the research laboratories of the departments of surgery and immunology, Queen’s Medical Centre, University of Nottingham.

Tissue culture medium (TCM)

Tissue culture medium (TCM) consisting of RPMI 1640 medium, supplemented with 10% heat-inactivated foetal calf serum (FCS), containing 1-mM l-glutamine, penicillin (100 U/ml) and streptomycin (100 μg/ml) (Sigma, UK), was used to prepare mononuclear cells and for various in vitro assays.

Mononuclear cell suspensions

Blood was collected just prior to surgery by venepuncture into syringes containing preservative-free heparin (Sigma, UK.). Peripheral blood mononuclear cells (PBMCs) were isolated by Histopaque-1077 (Sigma, UK) density gradient centrifugation (400 g for 30 min). The resultant mononuclear cell band was removed and washed twice with TCM (250 g for 10 min).

Regional LNs were obtained from patients undergoing surgery for breast cancer (as above). The LNs were cleaned of fat and distended by repeated injection with TCM (using a 2.5-ml syringe and 25-G needle). A single cell suspension was obtained by disaggregating the LNs and passing the suspensions through a stainless steel filter. Residual LN stroma was suspended in TCM containing collagenase (1 mg/ml; type II; Worthington Biochemical, Freehold, NJ) and DNase (0.02 mg/ml; grade II bovine pancreatic DNase; Boehringer, Germany) and incubated at room temperature. To disrupt DC–T cell complexes, ethylenediaminetetraacetic acid (EDTA) (Sigma, UK; 1 ml of 0.1 M, pH 7.2) was added to the cellular digest, and further incubation with agitation was carried out for 5 min. Undigested stromal fragments were removed by passing the suspension through a sterile stainless steel filter. Viable mononuclear cells were prepared by isolation over Histopaque-1077 (Sigma, UK) as described above for blood. Following this procedure, the cells were suspended in 10 ml of TCM containing EDTA (1 ml of 0.1 M, pH 7.2).

T lymphocyte isolation by rosetting with AET-treated sheep erythrocytes

PBMCs were prepared by centrifugation of PB over a Histopaque-1077 gradient, as described above. The PBMCs were then incubated for 30 min with aminoethylisothiouranium bromide (AET)–treated sheep red blood cells. Cells (T lymphocytes) that adhered to the red cells (R+) by forming rosettes and those that did not (R−) were separated on a Histopaque-1077 gradient (400 g for 30 min). T lymphocytes (R+) were recovered from the erythrocyte pellet by disaggregation and lysis of red blood cells with hypotonic saline solution (0.899% NH4Cl) for 5 min and then washing twice in TCM (250 g for 10 min) to remove red cell ghosts. The T lymphocytes (PB: patients and controls; LNs: patients) were used in the in vitro DC assays.

Immunomagnetic cell sorting (negative selection): isolation of dendritic cells

Mononuclear cells (1.5×107 cells)were incubated with a lineage cocktail containing fluorescein isothiocyanate (FITC)–conjugated antibodies (Abs: CD3, CD14, CD16, CD19, CD20 and CD56) from Becton Dickinson Systems (Mountain View, CA) for 30 min in a refrigerator at 6–12°C. The cells were washed twice with wash buffer (0.5% phosphate buffered saline (PBS) containing 0.5% FCS and 0.1 M of EDTA). The cell pellet was incubated with anti-FITC Microbeads (Miltenyi Biotec, UK) for 30 min in a refrigerator at 6–12°C followed by magnetic cell separation. A negative depletion column type LD(Miltenyi Biotec, UK) was used and placed in the magnetic field (Midi MACs; Miltenyi Biotec, UK). The negative (nonreacting) cells passed through the column and were collected as part of the negative fraction.

Allogeneic T cell proliferation assay

The ability of DCs from PB and regional LNs to stimulate allogeneic T cells was assessed in this assay. To minimize the effect of HLA differences between individuals in the allogeneic T cell proliferation, T cells used in all the experiments were collected from a single healthy volunteer, with no previous history of malignant disease and not taking medication during the experimental period. In this assay, DCs were the stimulator cells and T lymphocytes the responder cells. The latter were added at 5×104 cells/well. In each assay, all tests were carried out using three replicates. Appropriate controls were set up in each 96-well plate (Costar, Cambridge, UK). The stimulator DCs were irradiated with 2,000 rads in a cell irradiator (Cs-137). Varying concentrations of irradiated stimulator DCs were then dispensed into appropriate wells. DC to T cell ratios (1:20, 1:40 and 1:80) were set up and the plates were cultured for 6 days at 37°C in a humidified 5% CO2 incubator. Eighteen hours prior to harvesting the cells, 1.0 μCi (37 KBq) of [3H]-thymidine was added to each well. Thymidine uptake in the samples was then counted for a defined period of time (standard deviation <2%) in a liquid scintillation counter (Packard, UK). Mean counts per minute (CPM) were calculated. Control wells, containing T cells or antigen-presenting cells (APCs) alone, incorporated <500 CPM of [3H]-thymidine.

Autologous T cell proliferation assay (PPD stimulation test)

Autologous T cell responder cells, at 5×104 cells/well, were dispensed. Each experiment had at least three replicate samples. Dendritic cells were irradiated with 2,000 rads in a cell irradiator and various concentrations were dispensed into appropriate wells resulting in different DC to T cell ratios (1:20, 1:40 and 1:80). Dendritic cells were cultured with autologous T cells in the presence of 5 μg/ml of PPD for 6 days in 96-well plates. Eighteen hours prior to harvesting the cells, 1 μCi of [3H]-thymidine was added to each well. [3H]-Thymidine uptake was documented, as described above.

Flow cytometric analysis of dendritic cells

The expression of leukocyte phenotypic markers was assessed using fluorescein isothiocyanate (FITC)–conjugated murine monoclonal antibodies (mAbs)–lineage cocktail mAbs (CD3, CD14, CD16, CD19, CD20 and CD56) (Becton Dickinson Systems, UK). HLA-DR and CD40 surface markers were determined using phycoerythrin (PE)–conjugated anti-HLA-DR and anti-CD40 and the APC-conjugated anti-HLA-DR (Pharmingen, UK). The CD83 DC maturation marker amd CD86 costimulatory molecule were assessed using Cy-Chrome–conjugated anti-CD83 and the ECD-conjugated anti-CD86 (Beckman Coulter, UK). Cell preparations were examined using multicolour (3 laser, 6 colour) flow cytometry (EPIC ALTRA; Beckman, Coulter, UK). Cell suspensions (1×106 cells) were incubated with optimal concentrations of fluorescent-conjugated mAbs for 45 min on ice (0–4°C). The cells were then washed in PBS containing 0.1% bovine serum albumin (BSA) and fixed in 3% formaldehyde to exclude dead cells.

Cytokine assays

Supernatants from multiwell plate cultures of DCs and T cells were removed and assayed for cytokine production (IL-10 and IL-12), using commercially available quantitative enzyme-linked immunosorbent assay (ELISA) kits (R&D, UK). The optical density (OD) was directly proportional to the concentration of the respective cytokines in the sample. The amount of cytokine in each sample was estimated by calibrating the OD values of the standards supplied with the kits, using an ELISA reader.

Statistical analysis

Statistical analysis was performed using the Mann-Whitney U-test. Spearman coefficient was applied to examine the correlation. Differences were considered significant when p≤0.05.

Results

Patient characteristics

The patients age, clinical tumour size (T1–3), tumour type, tumour grade, axillary node status (pathological) and hormone receptor status (oestrogen receptor [ER] and progesterone receptor [PR]) are documented in Table 1.

Table 1.

Clinical and pathological characteristics of patients studied. IDC invasive ductal cancer, NST no special type, DCIS ductal carcinoma in situ, ILC invasive lobular cancer, ER eostrogen receptor, PR progesterone receptor, NA not available

Patient number Age Tumour size Tumour pathology Tumour grade Nodal statusa Hormonal receptor
ER PR
1 44 T1 IDC, NST 2 0/4 ++ –
2 53 T2 IDC, NST 3 0/6 – –
3 52 T1 IDC, NST 2 1/5 + +
4 49 T2 IDC, NST 2 0/6 +++ +++
5 84 T2 ILC 2 0/8 +++ +++
6b 52 T3 IDC, NST 2 3/15 – –
7 61 T2 IDC 3 0/13 + +
8 79 T1 IDC, NST, DCIS 2 1/11 +++ +++
9 65 T2 IDC, DCIS 2 0/4 – –
10 64 T1 IDC, NST, DCIS 3 0/4 – –
11 56 T1 IDC, NST 2 0/5 +++ +
12b 69 T3 IDC, NST 2 0/14 + +
13 42 T1 IDC, NST, DCIS 3 0/8 – –
14 63 T1 IDC, NST 1 0/6 +++ +
15 56 T2 IDC, NST 2 0/11 + +
16b 60 T3 ILC, NST 2 1/5 NA NA
17 51 T2 ILC, IDC 2 0/6 + +
18 62 T1 Tubular 1 0/5 +++ ++
19 41 T1 IDC, NST 2 0/6 +++ –
20 62 T1 IDC, NST 2 1/7 ++ ++
21 74 T2 ILC 1 0/8 ++ ++
22b 57 T3 IDC, NST 2 3/10 – –
23 59 T1 IDC, NST, DCIS 2 0/11 ++ ++
24 68 T2 IDC, DCIS 2 0/7 – –
25 74 T1 IDC, NST, DCIS 3 0/8 +++ ++
26 67 T1 IDC, NST 2 0/5 ++ +
27b 59 T3 IDC, NST 2 0/11 ++ ++
28 61 T2 IDC, NST 2 0/5 – –
Open in a new tab

aNodal status: tumour-involved nodes / nodes removed

bPost neoadjuvant chemotherapy

Isolation of dendritic cells

Dendritic cells from the low-density fraction of PBMCs were prepared by using immunomagnetic beads and negative selection. Using this technique, DCs were prepared from PB (10–15 ml; 107 PBMCs) without using cytokines or time-consuming in vitro culture. From blood, approximately 0.5% of PBMCs were isolated and approximately 70% to 75% were DCs (Fig. 1). These cells fulfilled the functional requirements of DCs as assesed by stimulation of allogeneic T cells (see Fig. 2) and autologous T lymphocyte proliferation in response to PPD (5 μg/ml) (data not shown), using healthy donor PBDCs and lymphocytes.

Fig. 1.

Fig. 1

Open in a new tab

Isolation of DCs by immunomagnetic bead selection method (negative selection). Representative results from FACScan analysis of PBDCs isolated by immunomagnetic bead negative selection method from peripheral blood. DCs are Lin− and HLA-DR+

Fig. 2.

Fig. 2

Open in a new tab

Blastogenesis in MLR: allogeneic T lymphocyte (normal donor) proliferation, in cultures containing various ratios of DCs from the peripheral blood of healthy donors. The stimulatory capacity of DCs was estimated from the incorporation of [3H]-thymidine (shown as CPM) during the last 18 h of 6 days of in vitro culture containing T cells (5×104 cells/well) from a normal volunteer and irradiated PBDCs (n=10) from healthy donors. Results shown are means ± SE

In the LNs (obtained from the axilla of patients with breast cancer), cell suspensions were prepared by mechanical and enzymatic digestion. About 1x107 to 2x107 LNMCs were isolated from the portion of LN obtained in the operating theatre. Lymph node dendritic cells were then isolated by the immunomagnetic bead selection technique described for the preparation of PBDCs. With this method, approximately 0.80% to 1.85% of LNMCs were isolated and contained approximately 75% to 80% DCs (data not shown).

Defective MLR response: allogeneic (normal donor) T lymphocyte proliferation with DCs (PB and LNs) from patients with breast cancer

As the T cells from a normal volunteer were used in all the experiments, the level of blastogenesis indicated the stimulatory capacity of PBDCs. Figure 3 shows the incorporation of [3H]-thymidine by allogeneically stimulated proliferating T lymphocytes from a normal volunteer and DCs (PB and LNs) from patients with breast cancer and healthy donors (controls). The levels of blastogenesis (PB and LNs) were significantly lower in patients with breast cancer, compared with the normal controls (PBs) (p<0.01). Cultures containing only T cells or DCs showed a background proliferation of <500 CPM. This data suggests a defective or suboptimal level of DC function in women with breast cancer, both systemically in blood and in the regional LNs draining the breast cancer (Fig. 3).

Fig. 3.

Fig. 3

Open in a new tab

Decreased blastogenesis of MLR. The stimulatory capacity of DCs (PB and LNs) was estimated from the incorporation of [3H]-thymidine (shown as CPM) during the last 18 h of 6 days of in vitro culture containing T cells (5×104 cells/well) from a normal volunteer and irradiated DCs (2.5×103 cells/well) from patients with breast cancer (PBDCs and LNDCs) and healthy donors (control PBDCs). Values shown are means ± SE. Asterisks indicate significant (p<0.01) reduction of PBDC and LNDC activity, compared with healthy control PBDC function

The lowest responses were seen in women who had undergone primary chemotherapy for large / locally advanced cancers, prior to undergoing surgery. There was no correlation of MLRs with tumour size (T1–3), nodal status (N0–2), or hormonal receptor status (ER/PR±) (data not shown).

Normal function of T cells in patients with breast cancer

To characterize the mechanisms of the observed in vitro stimulatory effects, purified T cells from patients with breast cancer (PB and LNs) and DCs from the normal controls were incubated and the ability of DCs to stimulate allogeneic T cells was evaluated, as shown in Fig. 4. No differences were found in the isolated T cells (PB and LNs) of these patients, to undergo stimulation with control DCs (allogeneic). The level of CPM documented was high and comparable with the control blood responses obtained (only data for PB shown).

Fig. 4.

Fig. 4

Open in a new tab

PBDCs from control individuals effectively stimulate PB T lymphocytes from patients with breast cancer (MLR). The stimulatory capacity of PBDCs was estimated from the incorporation of [3H]-thymidine (shown as CPM) during the last 18 h of 6 days of in vitro culture containing T cells (five different patients with breast cancer [5×104 cells/well]) and irradiated PBDCs (control [2.5×103 cells/well]). Control bloods were obtained from six healthy donors

Defective response of autologous T lymphocyte proliferation in patients with breast cancer (PPD stimulation assay)

Antigen-driven autologous T cell proliferation in patients with breast cancer was studied. Because this required the presence of APCs, both DCs and T cells were isolated from the PB and LNs of patients with breast cancer and mixed at different ratios (1:20, 1:40 and 1:80); the PPD stimulator was used at 5 μg/ml. Patients with breast cancer had a significantly decreased ability of their autologous T cells to respond to PPD, when compared with the stimulation elicited by autologous T cells and DCs from normal donors (p<0.01), as shown in Fig. 5. Thus, defective autologous T cell responses were seen in both the blood and regional axillary lymph nodes draining the cancer. There was no correlation of PPD stimulation assay with tumour size (T1–3), nodal status (N0–2), or hormonal receptor status (ER/PR±) (data not shown).

Fig. 5.

Fig. 5

Open in a new tab

Decreased blastogenesis of autologous T lymphocyte proliferation. Cultures containing DCs (PB and LNs) and T lymphocytes (PB and LNs) from patients with breast cancer and healthy donors (control PB) and PPD (5 μg/ml) as the stimulator were set up. The stimulatory capacity of DCs (PB and LNs) was estimated from the incorporation of [3H]-thymidine (shown as CPM) during the last 18 h of 6 days of in vitro culture containing autologous T cells (PB and LNs) (5×104 cells/well) and irradiated DCs (2.5×103 cells/well). Asterisks indicate significant (p<0.01) reduction of PBDC and LNDC activity, compared with healthy control PBDC function

Decreased IL-12 p40 production in cultures containing DCs from patients with breast cancer

DCs from PB and LNs of patients with breast cancer were cultured in vitro with T cells from a single volunteer (MLRs). The supernatants were collected after 5 days of culture and the level of cytokines (IL-10 and IL-12 p40) present were determined by ELISA. The levels of IL-12 p40 produced in allogeneic MLRs containing DCs from patients with breast cancer were significantly decreased, compared with the levels documented in healthy donors (controls) (p<0.01). The levels of IL-10 in the allogeneic MLR supernatants were less than the minimum detectable dose. (Determined by adding two standard deviations to the mean optical density [OD]; see Table 2.)

Table 2.

Production of IL-10 and IL-12 p40 in allogeneic MLR. The amounts of cytokines in the samples were estimated by calibrating the OD of the standards supplied with the kits, using an ELISA reader. The levels of IL-12 p40 in allogeneic MLRs containing DCs from patients with breast cancer were significantly decreased compared with the level obtained from normal controls (p<0.01)

IL-10 (pg/ml) IL-12 p40 (pg/ml)
Control PBDCs None detected 72.95±4.47
Breast cancer PBDCs 0.59 18.1±6.73*
Breast cancer LNDCs 0.61 21.9±9.20*
Open in a new tab

*p<0.05; significant decrease compared with normal control PBDCs

Decreased expression of HLA-DR, CD40 and CD86 on DC surface from patients with breast cancer

The representative staining patterns of HLA-DR, CD40 and CD86 are shown in Fig. 6. The percentage of expression of HLA-DR and CD86 in PB and LNs from patients with breast cancer was significantly lower compared with normal controls (p<0.01). There were significant alterations in the expression of CD40 on DCs of PB from patients with breast cancer, compared with normal controls (p<0.05) (Fig. 6). These data suggest that the DCs isolated from the women with breast cancer are switched off and have poor Ag-presenting capabilities.

Fig. 6.

Fig. 6

Open in a new tab

DCs were stained with multicolour-labeled mAbs. Lin− cells were selectively gated and analysed. Asterisks indicate decreased expression (p<0.01) of HLA-DR, CD40 and CD86 on DC surface (PB and LNs) from patients with breast cancer. Percentage of cells expressing of HLA-DR, CD40 and CD86

Discussion

Although there is considerable interest in the biological role of DCs, studying these cells in vitro has proven difficult because of the small numbers present in blood and tissues and the complexity of the techniques used for their isolation. In addition, there are currently no specific or unique surface markers for this population of cells. The method used in our study is a simple and relatively rapid technique for isolating DCs. The low-density fraction of mononuclear cells was used as the starting population, thus omitting time-consuming in vitro culture and the need for cytokine stimulation [granulocyte-monocyte colony-stimulating factor (GM-CSF), interleukin-4 (IL-4) or tumour necrosis factor-α (TNF-α)]. Also, prolonged in vitro culture activation of progenitor cells may not represent the true in vivo DC status. Culture of mononuclear cells, in particular with FCS at 37°C, activates DCs and changes their size as well as their density, thereby altering purification over density gradients [7, 29, 50, 51, 52]. The gradients used for DC isolation, however, may induce phenotypic and functional change through their osmotic effects [27, 36].

Because of their central role in the induction of immune activity, dysregulation of DC function in malignant disease could result in failure of activation of putative anticancer host defenses and lead to progressive tumour cell growth.

In the few studies published to date, variable alteration in DC function has been reported in both tumour-bearing mice and humans. In rodents, Zou et al. (1992) have reported normal and even increased function of APCs in tumour-bearing mice [54]. Gabrilovich et al. (1996), on the other hand, described diminished DC function and reduced MHC class II Ag presentation [19].

In humans, diminished DC function (assessed predominantly by MLR) has been documented in patients with several different tumour types (head and neck, colorectal and renal cell cancer, hepacellular carcinoma and malignant melanoma). These changes have been seen mainly in blood [9, 21, 22, 38], in cells infiltrating cancers [11, 41, 49] and more recently in LNs [1]. Only two studies, to date, have investigated patients with breast cancer [1, 21]. Our study documents DC function (both MLR and autologous responses) in the blood of women with operable and clinically localized breast cancer, as well as a small subset of women with locally advanced breast cancer following neoadjuvant chemotherapy. Data about DC function in tumour-draining LNs are poorly documented. One study has investigated head and neck cancer and the other study has documented the heterogeneity of phenotypic characteristics in axillary lymph nodes in women with breast cancer [1, 48]. The study reported here has carefully evaluated and documented both the phenotypic profile and Ag-presenting function of DCs in axillary lymph nodes draining breast cancer.

Our data not only confirm more fully the defective DC function documented in the blood of patients with breast cancer, but also document substantial inhibition of DC activity in both blood and tumour-draining LNs in patients with operable and clinically localized breast cancer. We have found no correlation of DC function with tumour volume, as assessed by tumour size and nodal spread; an association described in one previous study but with a larger tumour load [21]. The low levels documented in women who had received primary chemotherapy for large / locally advanced breast cancer are either a reflection of occult tumour disease or inhibition of DC function by chemotherapy [10, 15, 34, 53]. Both postulates are feasible as decreased DC function has been documented in vitro with various chemotherapeutic agents, including doxorubicin, a drug used as neoadjuvant therapy in our patients [10]. Although possible, it is unlikely in our study that cyclophosphamide is toxic to DCs, as it has been shown not to affect DC function, at least in animal studies [32, 35].

The low level of blastogenesis in the allogeneic MLR containing DCs from patients with breast cancer (Fig. 3) is not attributable to either a functional defect of T cells or the influence of serum and other inhibitory factors in patients with breast cancer. The T cells used in the allogeneic MLR were collected from a single normal volunteer, and autologous serum was never used in any of the experiments. Concurrently, patients with breast cancer had a decreased ability of their autologous T cells to respond to PPD, when compared with the normal controls (Fig. 5). These data are consistent with defective function of either T cells or DCs. To characterize the mechanism(s) of the observed effects, purified T cells from breast cancer patients and DCs from a normal control were cocultured in MLRs. The responses of the T cells from patients with breast cancer were comparable with those of healthy donors (Fig. 4). These data suggest that the defects in the functional activity of DCs could be a major factor in the inability of women with progressively growing breast cancers to elicit an appropriate and beneficial anticancer immune response. This defect is both systemic and in the secondary lymphoid structures, the latter being a crucial compartment for initiating the generation of anticancer cytotoxic T cells.

Interleukin 12 production is important for the induction and development of cellular immune responses. Interleukin 12 induces the maturation of DCs and causes an increased expression of costimulatory molecules (CD80, CD86) and CD83 (maturation-specific marker) on the surface of DCs [12, 24]. Interaction of Ag-specific T cells with DCs expressing Ag, linked to MHC class II peptides (MHC class II–TCR) and CD40 (DCs) binding to CD40L (T cells), leads to secretion of IL-12 [6, 28, 42, 45]. The latter leads to rapid development of Th1-mediated cellular immune responses by inducing the differentiation of naïve CD4+ T cells into Th1 cells [25, 33]. Some studies have shown that the development of Th1 cells is required to generate protective anticancer, cell-mediated immune responses [18, 23]. In our study, DCs from blood and LNs of patients with breast cancer produced low levels of IL-12 in in vitro MLRs (Table 2), compared with DCs from normal controls. The low level of DC function documented in our patients with breast cancer, therefore, may be due to this decreased level of production of IL-12. To date, reduced PBDC production of IL-12 has been documented in patients with hepatocellular carcinoma and melanoma; in the latter, with advanced disease [18, 38]. In our study, low level of IL-12 production was seen in women with a small tumour volume and in both circulating PBDCs and those in the regional tumour-draining LNs. These findings, to the best of our knowledge, have not previously been documented.

Although the DC-enriched and T cell–enriched populations in our allogeneic MLRs may contain contaminating B cells and macrophages, the IL-12 p40 is believed to be secreted by DCs only. De Smedt et al. (1997) did not detect mRNA for IL-12 p40 in purified preparations of macrophages or B cells in in vitro culture [16]. Induction of IL-12 by DCs has been shown to play an important role in generating effective antitumour immunity [23, 37].

Interleukin-10 production in the MLRs in our study (Table 2) could not be evaluated because the values were less than the minimum amount detectable by the assay. We have shown recently that the IL-10 levels in serum from patients with operable breast cancer is significantly higher than that found in normal age- and sex-matched controls [2]. The increased level of IL-10 was due to production and secretion by monocyte-macrophages. Interleukin-10 has been shown to have a significant inhibitory effect on several aspects of DC function; for example, the expression of costimulatory molecules and the ability to synthesize IL-12 [8, 30]. Importantly, IL-10–treated DCs can be tolerogenic [5, 13, 17]. Thus, the high level of IL-10 documented in the circulation in our patients may be an important mechanism for switching off DC activity, both in blood and LNs.

DCs become activated and mature following Ag capture, resulting in expression of costimulatory molecules, production and secretion of cytokines and induction of the proliferation of Ag-specific T lymphocytes in secondary foci of lymphoid tissues (e.g. LNs) [4]. Immature or unstimulated DCs, on the other hand, express low levels of MHC class II Ags, costimulatory molecules (CD40, CD80 and CD86) and have a low capacity for stimulation in the allogeneic MLR [4, 7, 36]. In particular, up-regulation of the expression of CD80 and CD86 molecules on DCs, results in enhanced interaction, with CD28 on T lymphocytes and their resultant activation and proliferation [12, 24]. The DC–T-cell interaction is thought to be a two-way interaction. The generation of T cell receptor (TCR) ligands and the immunostimulatory property on DCs develop in sequence, and appear to occur during the migration from the periphery to the T cell areas of LNs. In the present study, we found that the DCs from patients with breast cancer, both in the circulation and in regional tumour-draining LNs, had an inactive and immature phenotype. The costimulatory CD86 molecules, responsible for accessory signals in T cell activation, as well as the MHC class II Ags, were significantly lower than those expressed on PBDCs from normal healthy controls. This is in agreement with data published from several human cancers [1, 21, 38]. However, to the best of our knowledge, this phenotypic profile of DCs has not been previously well characterized in LNs draining breast cancer. Takahashi et al. (1998), in a small study, described heterogeneity of phenotype (CD1a, CD83 and CD86) distribution in axillary LNs using flow cytometry [48]. Preliminary findings in our ongoing studies have shown low levels of expression of CD83 in PB and LN DCs (less than 1%, data not shown), compared with PBDCs in healthy humans [3, 26]. Thus, DCs in patients with progressively growing but clinically localized breast cancer are both immature and switched off. The putative antitumour effect of DCs is dependent on the level of activation and maturation of DCs [31, 38]. Also, lack of costimulatory molecules in the presence of TCR occupancy does lead to T cell tolerance [11]. These concurrent changes in the LNs of patients with breast cancer are likely to result in failure of induction of a tumour-specific T cell cytotoxic response.

In summary, we have documented the phenotype and functions of DCs in patients with operable breast cancer and compared these with those of normal healthy controls. These data show that DCs in the circulation and in tumour-draining regional LNs in patients with operable breast cancer are immature, poorly functioning and probably switched off. Our data not only confirms unequivocally the few studies published to date, but also provides new and important findings in women with clinically localized breast cancer. Our results provide an important insight into the possible mechanisms responsible for the apparent failure of putative locoregional and systemic anticancer host defenses in patients with breast cancer, which may be relevant for other malignancies in humans.

References

Articles from Cancer Immunology, Immunotherapy : CII are provided here courtesy of Springer

RESOURCES