Multiagent therapies for children with acute lymphoblastic leukaemia (ALL) have resulted in impressive remission rates of around 90%. Unfortunately, relapse rates remain high and most children who relapse will not survive. Innovative antibody and chimeric antigen receptor (CAR) modified T-cell therapies have shown promising results in ALL. However, this is a heterogeneous disease and several groups have shown ALL evolves in a branching fashion. Moreover, some leukaemia stem cell (LSC) populations do not express antigens that are the targets of developing therapies such as CD19, CD20, and CD22. Consequently, these cells may be largely unaffected and could provide a reservoir of resistant cells, from which relapse may develop. The diversity of reported LSC populations may be a reason for treatment failure and improvements in therapy will require targeting of all cells that have the potential to initiate and maintain this disease. In an attempt to identify more appropriate therapeutic targets we conducted an extensive microarray screen of several LSC subpopulations and bulk leukaemia cells from a group of paediatric patients with favourable cytogenetic abnormalities. This screen identified a number of genes that were up-regulated in all LSC subpopulations compared to levels in normal haemopoietic cells. The up-regulated genes encode for cell surface antigens, including the type-1 membrane protein, CD200 and signalling pathways. Expression of CD200 was 12-fold higher in bulk ALL cells compared to normal BM cells (p=0.001). CD200 is a key immunosuppressive molecule that has been implicated as a poor prognostic factor in multiple myeloma and acute myeloid leukaemia. To validate the microarray results, we assessed expression of CD200 in 20 BCP-ALL cases (11 pre B, 9 c-ALL) by flow cytometry and in functional assays. Fifteen of these cases were classified as MRD low risk and had favourable cytogenetic abnormalities and 5 were classified as high risk by MRD and cytogenetic abnormalities. ALL cells were stained with antibodies against CD34, CD19 and CD200 and cells were sorted on the basis of expression/lack of expression of CD34/CD200. The functional ability of the sorted populations was assessed in NSG mice and serial transplantation experiments were performed to assess self-renewal capacity. Overall, the flow cytometric analyses confirmed the microarray data in that expression of CD200 was high in this patient cohort (60±6.6%) compared to normal BM (0.1%±0.04, p ≤ 0.001), regardless of cytogenetic abnormality or MRD risk status. In standard risk cases, BM engraftment was achieved with the CD34+/CD200+ (2-66% human leukaemia) and CD34-/CD200+ (10-26%) subpopulations only, using inocula of 4x104-107 cells. No engraftment was attained using similar numbers of CD200- cells, regardless of expression of CD34 or CD19 in this group. Interestingly, when cells from the 5 patients defined as high risk were inoculated, engraftment was achieved using all four sorted subpopulations. The levels of leukaemia engraftment in these cases ranged from 2-95% using 1x103-5x106 cells and levels were similar across all subpopulations in individual patient samples. Results from serial transplantation studies to date demonstrate the sorted subpopulations were capable of self-renewal. As with the primary transplants, all four subpopulations from high risk samples engrafted while this ability was restricted to CD200+ subpopulations in standard risk cases. FISH analyses have confirmed all the engrafting cells had an abnormal karyotype. These data provide further evidence that surface antigen immunophenotype is not a good predictor of function in paediatric ALL, particularly in high risk cases. However, in standard risk cases it may be possible to eliminate LSC by targeting CD200. Treating human chronic lymphocytic leukaemia cells with anti-CD200 antibodies has been shown to prevent engraftment in NSG mice. Our data suggests that investigation of the effects of anti-CD200 in standard risk childhood ALL are warranted.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution