Despite the accumulation of genetic mutations likely to generate antigenic neoepitopes, acute lymphoblastic leukaemia (ALL) cells are poor at eliciting anti-tumour immune responses. Similarly to dendritic cells (DCs), B cells have antigen presenting capacity. In B cell ALL (B-ALL), blast expression of co-stimulatory molecules is down regulated whilst endogenous (tumour) antigens are presented through MHC class I and II[1], thus mimicking the DC peripheral tolerance mechanism. This leads to antigen specific anergy in naïve T cells capable of recognising the malignant cells and an anti-tumour immune response is not mounted. A robust DC response, presenting tumour antigens alongside co-stimulation, should be sufficient to overcome this since no intrinsic T cell defect exists in these patients. However, several studies have reported deficiencies in DC number and function in patients with B-ALL. Maecker et al. demonstrated that paediatric patients with B-ALL have significantly reduced blood DC counts at diagnosis, a reduction not attributable to disease induced bone marrow hypoplasia since blood DC numbers do not correlate with granulocyte or monocyte numbers or haemoglobin[2]. Mami et al. identified that CD34+ peripheral blood mononuclear cells from patients with B-ALL at diagnosis are ineffective at generating functional DCs in vitro suggesting that B-ALL cells, or leukaemia derived factors, may contribute to a block in DC development[3]. In this study, we monitored DCs and plasma cytokines in paediatric patients with B-ALL at diagnosis and throughout remission induction chemotherapy, in peripheral blood (PB) and the bone marrow (BM) tumour microenvironment, to identify on treatment changes in DCs and DC regulatory cytokines. 17 paediatric patients diagnosed with precursor B cell ALL were enrolled in this study. Matched PB and BM aspirate samples were collected at time points co-ordinating with treatment protocol (diagnosis, and induction days 8 and/or 15 and 29). Mononuclear cells and plasma samples were obtained by density gradient centrifugation. DCs were identified by flow cytometry (based on expression of HLA-DR(+), lack of B, T, NK and monocyte markers (CD3-, CD19-, CD56-, CD14-)). Plasma cytokines were analysed by Luminex multi-plex immunoassay. Our study confirmed that blood DCs are low in paediatric patients with B-ALL at diagnosis and that this scenario is mirrored in the BM tumour microenvironment. PB and BM DCs remained low through days 8 and 15 of induction remission chemotherapy however, we identified day 29 of induction chemotherapy as a key time point in the re-establishment of BM DCs, when a significant increase was recorded compared to both earlier time points and day 29 PB samples. Concordantly, high levels of two homeostatic chemokines were recorded in PB and BM plasma samples at diagnosis, CCL19 and CCL21. While CCL19 expression remained consistent throughout induction chemotherapy, CCL21 declined rapidly and was not detectable in any PB or BM day 29 plasma sample, coinciding with the rapid reduction of blasts that occurs during induction therapy. The re-emergence of DCs in the bone marrow and other lymphoid organs may represent the first stages in re-establishment of effective anti-tumour immunosurveillance and contribute to the cure of B-ALL. With overall survival rates greater than 90% it is clear that effective immunosurveillance, capable of maintaining a state of heath post-treatment, is re-established in most patients. When and how this occurs in the treatment process is not yet understood. However, CCL21 may play an important role in mediating immune tolerance to B-ALL, since mouse studies have demonstrated its ability to protect tumours from immune rejection[4]. Here, the re-establishment of BM DCs occurred concordantly to the reduction of CCL21 in the BM tumour microenvironment suggesting that over-expression of CCL21 may contribute to DC suppression in paediatric B-ALL patients. [1] Cardoso et al. (1996) Blood 88(1), 41-48; [2] Meaker et al. (2006) Leukemia 20(4), 645-649; [3] Mami et al. (2004) Br. J. Heamatol. 126(1), 77-80; [4] Sheilds et al. (2010) Science 328, 749-752. Disclosures No relevant conflicts of interest to declare.
Read full abstract