Association between the Polycomb Group (PcG) BMI-1 Gene Expression and Outcome in Chronic Myeloid Leukemia (CML) Patients Receiving Allogeneic Stem Cell Transplantation (allo-SCT).

Despite a consistent molecular abnormality, the BCR-ABL oncogene, and the success of tyrosine kinase inhibitors, CML still exhibits a marked heterogeneity in prognosis. The PcG gene BMI-1 plays an essential role in regulating the proliferative activity of both normal and leukemic stem cells. BMI-1 correlation with outcome was described in different malignancies. We have previously shown that levels of BMI-1 RNA were significantly higher in advanced phase than in chronic phase (CP) CML. In addition, the level of BMI-1 at diagnosis correlated with time to transformation to blast crisis and, in patients treated with hydroxyurea and IFN-a, low BMI-1 expression was associated with an improved overall survival (OS), suggesting that BMI-1 may be a biomarker for the intrinsic heterogeneity of CML (Mohty et al., Blood 2007). Here, we investigated whether BMI-1 and the other previously established prognostic genes (CD7, PR-3 and ELA-2) are implicated in the prognosis of CML in the context of allo-SCT. For this purpose, we studied 84 CP-CML patients who received allo-SCT from HLA-identical related donors. CD7, PR-3, ELA-2 and BMI-1 expression was assessed by Q-RT/PCR in the recipient’s PBMCs collected before allo-SCT. The median expression level for each gene was used to segregate the patients into 2 groups (“low”: gene expression <median, and “high”: gene expression >median). The median follow-up post-allo-SCT was 9.9 (range, 1.7–23.9) years. The median EBMT-Gratwohl score was 3. None of the 4 tested genes showed any significant association with neutrophil or platelet engraftment, or with graft rejection. CD7, PR-3 and ELA-2 expression was not associated with OS. However, in sharp contrast to our previous findings in the non-allo-SCT setting, patients displaying a “high” BMI-1 expression level prior to allo-SCT had significantly better OS than those with “low” expression (P=0.005). When BMI-1 was included in a multivariate survival model and adjusted for the other prognostic variables (EBMT-Gratwohl score, allo-SCT era, and other relevant parameters), a high expression was found to be an independent marker associated with better survival (RR=2.72, 95%CI; 1.1–6.9; P=0.034). Given the impact of BMI-1 expression level on OS, without a significant association with relapse, and since neither BMI-1, nor the other genes showed any significant association with leukemia-free survival, we assessed their impact on transplant-related mortality (TRM). There was a striking and significant association between acute GVHD and BMI-1 expression, not only in overall incidence (“low” BMI-1: grade 0–1 (n=21), grade 2 (n=10), grade 3–4 (n=9); “high” BMI-1: grade 0–1 (n=32), grade 2 (n=9), grade 3–4 (n=1); P=0.005), but also in cumulative incidence at day 100 (48% vs. 24%, P=0.016). In multivariate analysis, a “low” BMI-1 expression level was associated with an increased risk of grade 2–4 acute GVHD (RR=2.85, 95%CI; 1.3–6.4; P=0.011). These results suggest that BMI-1 measured prior to allo-SCT can serve as a biomarker for predicting outcome in CP-CML patients receiving allo-SCT. The presence of a potential donor immune response against BMI-1, which is a genuine tumor-associated antigen, may abrogate any neoplastic proliferative advantage within the leukemia stem cell pool conferred by higher expression of BMI-1. Such measurement allows for tailored therapeutic intervention, including informed recommendation for allo-SCT in patients failing tyrosine-kinase inhibitors.