Background. Although the prognosis of acute lymphoblastic leukemia (ALL) has improved considerably in recent years, some cases still exhibit therapy-resistance and late relapse. CD9-CXCL12-CXCR4 pathway has been implicated in hematopoietic and leukemic stem cell homing, promoting cell adhesion and survival in bone marrow stromal niches and mediating cell dissemination to secondary lymphoid organs. We characterized the CD9-CXCL12-CXCR4 expression in specific biological subgroups of childhood ALL, comparing, in some cases, diagnosis vs relapse, in order to investigate the role of these genes, in the process of leukemia relapse and to preliminarily evaluate their impact as prognostic markers, even for those subtypes with well-established good outcome.Materials and Methods. We analyzed bone marrow (BM) samples from 68 children with ALL, 3 cases with chronic myeloid leukemia, as control for t(9;22) positive leukemia, and 4 healthy donors (HDs). Patients were enrolled and treated at our Center from 2000 to 2010. We reverse-transcribed 500 ng of patients’ diagnostic RNA and performed a Real-time PCR using the SYBRTM Green PCR Master Mix (Applied Biosystems®), calculating the median fold-changes (MFCs) among different reactions and comparing with HDs. GUS gene mRNA was used as an internal positive control, showing no significant variation in our experiments. Each PCR experiment was carried out in duplicate. Sequences of oligonucleotides used for experiments, together with RQ-PCR protocol were previously published (Gandemer V. et al Leukemia Research 2010)Results. We analyzed for CD9, CXCL12 and CXCR4 expression 39 children with t(12;21) positive ALL, 12 cases with t(1;19) positive ALL, 8 cases with t(9;22) positive ALL, and 9 cases resulted negative for chromosomal translocation screening. Among the t(12;21) positive ALL we found that CD9 was overexpressed in 13 out of 39 patients (33%), CXCL12 in 14 out of 39 (35%) and CXCR4 in 9 out of 39 (23%), respectively. We noted that in 2 cases who presented a late relapse, one (CT13) showed a normal expression of the analyzed genes, conversely the other case (CT24), who suffered from a isolated extramedullary relapse, showed a high overexpression of CD9. In the t(1;19) subgroup we found only one case with overexpression of both CD9 and CXCR4 genes, respectively. Moreover we noted that comparing diagnosis vs relapse in one case, we observed a dramatic increased expression of CXCL12 (CT70 diagnosis 3,33 FCs vs relapse 38 FCs). In the subgroup of children with t(9;22) positive ALL, we observed an extremely high expression of the CD9-CXCL12-CXCR4 genes. In particular, we found an overexpression of CD9 in 4 out of 8 (50%) cases with this subtype of ALL; interestingly, all these case showed a BM relapse. One of these (CT56) showed an overexpression of all the three pathway’s components. As negative counterpart, we also analyzed three cases with a t(9;22) positive CML, finding a normal expression of all genes. In order to determine this pathway’s expression in children presenting an ALL without known chromosomal aberration, we analyzed other 9 cases. In only one (CT44) we found an overexpression of both CD9 and CXCL12. This patient suffered from an early relapse (<24 months from diagnosis).Conclusions. Our preliminary findings demonstrated that the CD9-CXCL12-CXCR4 pathway is mainly altered in children with t(12;21) and t(9;22) positive ALL. In the latter subgroup we found a linear correlation between overexpression and impending relapse, confirming that this pathway can be potentially used as prognostic marker. Among t(1;19) positive ALL, we found few cases with single gene aberration. More importantly, we found an overexpression of one or two pathway's components in diagnostic samples of those cases, who presented a subsequent relapse. These data need to be confirmed in a larger population and will pave the way to identify new prognostic marker and/or therapeutic target, since CXCR4 has already antagonist drugs in clinical use, showing very recent promising results (Randhawa S et al. British Journal of Hematology 2016). DisclosuresNo relevant conflicts of interest to declare.
Read full abstract