Does Expression Of CD200 Provide a Target For LIC In Childhood ALL?

Although the survival of children with acute lymphoblastic leukaemia (ALL) has improved considerably, there are still around 20% of cases that go on to relapse. These patients do not respond well to current therapies and there is a need to develop treatment strategies to directly target the cells that initiate the leukaemia to allow eradication of the disease. Leukaemia initiating cells (LIC) have been shown to be present in several immunophenotypic subpopulations evaluated in NSG assays. Identifying a marker that is expressed by all LIC populations may allow their discrimination from normal haemopoietic stem cells and more specific targeting of these cells. Whole genome expression studies have recently highlighted genes, including CD58, CD97 and CD99 that were over expressed in B-ALL cases when compared to normal CD19⁺/CD10⁺ cells. The addition of these markers to the minimal residual disease (MRD) flow cytometric panel increased the sensitivity of detection by 10-fold to one leukaemia cell in 10⁵ bone marrow (BM) cells. We have previously investigated the functional capacity of ALL cells sorted for CD58, CD97 and CD99 in combination with CD34 in vivo. However, none of these antigens were specific for LIC. CD200 has been described as another marker that improved MRD sensitivity by flow cytometry and consequently we investigated its expression in LIC populations and the functional ability of these cells in vivo. Expression of CD34, CD19 and CD200 was assessed in 12 B-ALL cases (6 pre B-ALL, 6 c-ALL) and 6 cord blood samples (CB) by flow cytometry. Whole genome microarray analysis was also performed on unsorted cells and sorted CD34⁺/CD19⁺, CD34⁺/CD19^-, CD34^-/CD19⁺ and CD34^-/CD19^- subpopulations to compare expression of genes between these LIC and the bulk leukaemia. Cells from 5 of these patients were also sorted on the basis of expression of CD34 and CD200 and the functional ability of the sorted subpopulations was assessed in NSG mice. The microarray analyses demonstrated that CD200 was over expressed in all the subpopulations and was especially high in the CD34⁺/CD19⁺ and the CD34^-/CD19⁺ subpopulations. CD200 was significantly over expressed in the CD34⁺/CD19⁺ subpopulation when compared to its expression in the CD34⁺/CD19^- and CD34^-/CD19^- subpopulations (p=0.009, F=6.3, F_critical=4.5). Flow cytometric analyses confirmed that CD200 was over expressed in unsorted B-ALL cells compared to unsorted CB (54.6%±8.1% vs 0.07±0.09% respectively, p=0.0002) and expression was higher in CD34⁺/CD19⁺ (68.4%±10.8% p=0.0006), CD34⁺/CD19^- (39%±9.2% p=0.01), CD34^-/CD19⁺ (48.4%±8.6% p=0.002) and CD34^-/CD19^- (5%±1.8% p=0.3) subpopulations compared to CD34⁺/CD38^- CB cells (2%±1.1%). Over expression of CD200 on these ALL subpopulations indicates that it may be useful in characterisation of LIC. When cells from 5 of these cases were sorted, the CD34⁺/CD200⁺ subpopulation accounted for the greatest proportion of cells (60.5%±16.1%). Expression of CD19 was also high in this subpopulation (65±11%). CD34⁺/CD200^- cells represented the smallest subpopulation (0.5%±0.1%) and these cells had low expression of CD19 (4±7%). The CD34^-/CD200⁺ and CD34^-/CD200^- subpopulations represented 18%±11.4% and 21%±120.6% of the bulk cells, respectively. CD19 expression was higher in the CD34^-/CD200⁺ subpopulation (30±14%) than in CD34^-/CD200^- cells (7±5%). Results from 4 cases to date indicate that engraftment was achieved with the CD34⁺/CD200⁺ subpopulation in every case (4-92% human leukaemia). In 3 cases, comparable engraftment levels were also achieved using CD34^-/CD200⁺ cells (10.4-86.1%). No engraftment was detected using the CD34⁺/CD200^- or CD34^-/CD200^- subpopulations, with the exception of 1 case where engraftment levels of 18-48% were achieved. These levels were lower than those achieved with CD200⁺ cells from the same patient. Interestingly when BM from mice engrafted with CD200^- cells from this patient were analysed they now expressed CD200 (12-35%), indicating cell differentiation had occurred in vivo. Assessment of self-renewal capability of these LIC populations is ongoing. These data suggest that CD200 may be a useful marker for discriminating LIC cells from normal haemopoietic cells, irrespective of expression of CD34 and CD19 in these cases.