Pre-treatment with chemotherapy can enhance the antigenicity and immunogenicity of tumours by promoting adaptive immune responses

Abstract

Background: Some cancer patients are immuno-compromised, and it has been long felt that immune-intervention is not compatible with standard chemotherapies. However, increasing evidence suggests that standard chemotherapy drugs may stimulate beneficial changes in both the immune system and tumour.

Methods: We have assessed the expression of human leucocyte antigen class 1 (HLA1) on tumour cells before and after chemotherapy agents (cyclophosphamide, oxaliplatin or gemcitabine). In addition, we show that chemotherapy-stressed tumour cells may release cytokines that enhance the interactions between dendritic cells (DCs) and T cells into growth media.

Results: Here we report that some chemotherapy agents can increase HLA1 expression in tumour cells, even when expression is low. Increases were associated with killing by cytotoxic T cells, which were negated by HLA1-blockade. Furthermore, T-cell function, as indicated by increased proliferation, was enhanced as supernatants derived from tumours treated with chemotherapy augmented DC-maturation and function.

Conclusion: There is evidence that a facet of immune surveillance can be restored by appropriate chemotherapy agents. Also, tumours exposed to some chemotherapy may secrete cytokines that can mature DCs, which ultimately enhances T-cell responses.

Figure 1

Chemotherapy increases human leucocyte antigen class 1 (HLA1) expression in tumour cells. Tumour cell lines were cultured with equi-active ∼IC25 concentrations of each of the drugs for 3 days before the assessment of HLA1 (HLA–ABC) expression by flow cytometry. Representative histograms of expressions in MCF7 are shown (inset), where I = isotype control, 0 = control, o = oxaliplatin, c = cyclophosphamide and g = gemcitabine. Each data column is a representative of the mean and s.d. of at least three separate experiments. In vivo HCT116 data generated from the mean and s.e.m. of five mice.

Figure 2

Human leucocyte antigen class 1 (HLA1) expression is associated with cytotoxic T-cell function. A549, HCT116 and MCF7 cell lines were cultured with equi-active ∼IC25 concentrations of each of the drugs for 3 days. Drugs were then removed and the cells were washed before being admixed with cytotoxic T cells at a ratio of 20 : 1. (A) Cultures were maintained for a further 24 h before the assessment of cell death/viability by the lactate dehydrogenase (LDH) and methylthiazoletetrazolium (MTT) assays. There were significant increases in cytoxicity (LDH: upper panel) and concomitant decreases in cell number (MTT: lower panel) in those cultures where gemcitabine was used. (B) Blockade of HLA1 in HCT116 and MCF7 cells by using antibodies negated the cell death associated with increased HLA1 expression. There was little change to cytotoxicity in cells treated with the isotype control antibody, which was used at 10 μg ml⁻¹. Each data column is a representative of the mean and s.d. of at least three separate experiments.

Figure 3

Supernatants from tumours treated with chemotherapy stimulate dendritic cells (DCs). DC-derived monocytes were cultured with exhausted culture media from A549 and MCF7 treated cells. The geometric mean fluorescence intensities (GMFIs) and percentages (percentage of cells) of expressions of the maturation markers CD80, CD83 and CD86 were assessed after 24 h by flow cytometry. Representative density plots with %cells distribution within each quadrant (A) and histograms highlighting the GMFI within the defined range (B) are shown for DCs exposed to A549-derived media. These media were from cultures treated with cyclophosphamide (CPM), oxaliplatin (OXP) or gemcitabine (GEM). Tumour-free basal medium (BM) and media from untreated tumour cells (control) were also included. (C) There were generally increases in the markers (either in terms of GMFI and/or percentage of cells) upon exposure to supernatants from cyclophosphamide (C), oxaliplatin (O) or gemcitabine (G) compared with untreated tumours (0), which was most clear in the A549 cohort of samples. Each column is a representative of the mean of at least three separate experiments and s.d. has been omitted for clarity.

Figure 4

Supernatant-stimulated dendritic cells (DCs) increase the number of T-cell colonies. DCs matured with supernatant (CM) derived from A549 treated with cyclophosphamide (CPM), oxaliplatin (OXP) or gemcitabine (GEM) were admixed with allogeneic T cells and the number of colonies present enumerated on day 4. There were significantly larger numbers of colonies in wells containing DCs pre-exposed to chemotherapy-treated tumour. Typical pictures of colonies seen in each condition are shown, and each data column. Each data point is a representative of the mean and s.d. of five separate experiments.

Figure 5

Supernatant-stimulated DCs increase the proliferation of T cells in vitro. DCs matured with supernatant (CM) derived from A549 treated with cyclophosphamide (CPM), oxaliplatin (OXP) or gemcitabine (GEM) were admixed with allogeneic T cells loaded with carboxyfluorescein succinimidyl ester (CFSE). (A, B) Proliferation as indicated by a downward shift of CFSE mean fluorescence intensity was increased after culture with the positive control of PHA. Both CD4+ and CD8+ subsets were assessed. (C) There was a trend of increased percentage of proliferating T cells cultured with DCs stimulated with CM, which was significant (P<0.03) in the samples involving GEM. Each data column is a representative of the mean and s.d. of three separate experiments.