Immunogenetic Cross-Talk in Patients Transplanted for AML: CMV Reactivation Is Not a Strong Stimulus for Immune Response Against Leukemia

Early cytomegalovirus (CMV) re-activation after allogeneic hematopoetic stem cell transplantation (allo-HCT) for AML has been correlated with enhanced graft versus leukemia (GvL) effect, thus implying an immunogenetic cross- talk between donor-derived cytotoxic response against both CMV and leukemia. Data on this are conflicting and correlation could not be  confirmed for special group populations. In this study we explore whether CMV re-activation favors the outcome in a group of 49 AML patients consecutively allo-transplanted between 2008-2012. Forty-nine patients (29 male, 20 female), aged 14-63 years (median 42) underwent allo-HCT with myelo- ablative (42/49) or reduced-intensity conditioning (7/49), receiving graft from an HLA- identical sibling (25/49), matched volunteer (23/49) or haplo-identical (1) donor.  Most patients were transplanted in complete remmision (CR), 22/49 in CR1 and 11/49 in CR2 and 16/49 patients had refractory disease. Monitoring of CMV viral load was performed weekly for 3 months with real-time PCR, or more if patients were treated for GvHD. Upon clinical suspicion, other samples were tested.  Overall, 19/49 patients (39%) presented with CMV reactivation {viremia (>=1000 copies/ml) in 16/49, gastritis in 2/49, cystitis in 1) at a median of 46 days post-tranplant (13-402). Leukemia relapse occured in 7/19 patients with CMV at a median of 189 days (2-512), versus 7/30 CMV-free patients (37% vs 23%, p=ns). Disease free survival (DFS) was 591 days for CMV patients (95% CI: 417-939) and 609  for CMV-free (95% CI: 474-871, p=ns). Overal survival (OS) was 746 (95% CI:472-963) and 625 (95% CI:504-899) days, respectively, p=ns. Only acute and chronic GvHD were correlated with longer DFS and OS. Sixteen out of 19 CMV (84%) patients presented with increasing/full donor chimerism 30 days after reactivation, due to expansion of donor cytotoxic lymphocytes; however, this did not protect from imminent relapse. No particular clinical correlations could be found even for 14/19 patients with early (<100 days) CMV reactivation (RR: 28% vs 28% for all other patients, DFS: 714 vs 503 days, OS: 746 vs 625 days, p=ns for all comparisons). In conclusion, CMV reactivation in AML transplanted patients is not related to graft allo-reactivity. Late relapses in this cohort of AML patients, transplanted often in remission, are not influenced by immunogenetic dynamics at the first semester, when CMV occurs. Regardless CMV re-activation, OS was comparable in all patients, providing evidence on a succesfull pre-emptive treatment of CMV. Larger prospective studies are strongly warranted in order to explore possible immunogenetic interactions.