Graft-Vs-Leukemia Effect in Acute Lymphoblastic Leukemia: Mild Acute Graft-Vs-Host Disease Protects Against Relapse and Improves Survival after Allogeneic Transplantation: A CIBMTR Analysis

Abstract

Introduction: Graft-versus-host disease (GVHD) associated with allogeneic donor hematopoietic cell transplantation (alloHCT) provides protection against leukemia relapse, but more severe GVHD may be detrimental and increase mortality.Methods: We examined the net effect of GVHD on overall survival (OS) of 5215 patients with acute lymphoblastic leukemia (ALL) treated with alloHCT in 232 centers between the years 2000 and 2014 and reported to the CIBMTR. We examined three cohorts: a) adults in complete remission (CR1 and 2; n=2593) treated with myeloablative (MAC) or reduced intensity conditioning (RIC), b) pediatric ALL aged 1-18 in CR1/CR2 (n=1619) treated by MAC and c) patients of all ages with advanced ALL (active disease or CR≥3; n=1003) who received MAC. Transplant outcomes were compared by GVHD presence and severity: no GVHD (reference group), grade I-II acute GVHD (aGVHD), grade III-IV aGVHD, plus each with or without chronic GVHD (cGVHD).Results: Patients had B-cell (80%) or T-cell ALL (15%); 26% were Ph+. MAC was used in most patients (93%), followed by HLA-identical sibling (28%), adult HLA-matched unrelated donor (31%) or umbilical cord blood (24%) grafts. 28% received in vivo T cell depletion. Cumulative incidence of relapse and 5-year overall survival (OS) in the three cohorts are shown in table. In multivariate analysis adjusted for age, white blood cell count, donor source, KPS, GVHD prophylaxis and conditioning, aGVHD (both grade I-II and III-IV) and cGVHD yield protection from relapse as compared to no GVHD (HR 0.49-0.69) in both adult and pediatric CR cohorts (Table). Patients with advanced ALL with cGVHD or with grade III-IV aGVHD also had a lower, but still excessive risk of relapse. However, in all 3 cohorts, increased transplant-related-mortality (TRM) accompanied grade III-IV aGVHD (HR 3.91, 3.69, 2.69) and abrogated the protection from relapse resulting in inferior OS (Table). Importantly, adults in CR1/2 with grade I-II aGVHD without cGVHD had a significantly better OS compared to no GVHD (HR 0.83). A similar net survival benefit of grade I-II aGVHD occurred in children in CR1/CR2 (HR 0.76). CGVHD alone or in combination with grade I-II aGVHD augmented protection from relapse (HR varied 0.51-0.67), but had a neutral effect on OS in the two CR cohorts. Patients with advanced ALL had improved OS if they developed cGVHD with or without grade I-II aGVHD.Conclusions: Our data confirm the potent GVL effect of alloHCT in ALL in CR and demonstrate that the net beneficial effect of aGVHD on OS in ALL is confined to patients who experience grade I-II aGVHD and no cGVHD. This favorable net effect extends from pediatric to adult patients who are in remission. Those with advanced disease experience modest survival benefit from cGVHD. Nevertheless, the benefit of relapse protection is abrogated by increased mortality in patients with grade III-IV aGVHD and is weaker in those without any GVHD. Immunity-modulating interventions post-HCT and more effective GVHD therapies may better harness the potential graft-versus leukemia effect in ALL. [Display omitted] DisclosuresFlowers:Pharmacyclics: Consultancy. De Lima:Pfizer: Membership on an entity's Board of Directors or advisory committees; Celgene Corporation: Research Funding. Bachanova:Seattle-Genetics: Consultancy, Membership on an entity's Board of Directors or advisory committees; Novartis Pharmaceuticals Corporation: Membership on an entity's Board of Directors or advisory committees, Research Funding; Oxis: Membership on an entity's Board of Directors or advisory committees, Research Funding; Juno: Membership on an entity's Board of Directors or advisory committees; Zymogen: Consultancy, Membership on an entity's Board of Directors or advisory committees.