The impact of HMG-CoA reductase inhibition on the incidence and severity of graft-versus-host disease in patients with acute leukemia undergoing allogeneic transplantation

To the editor:

Acute graft-versus-host disease (GVHD) is the major cause of morbidity and mortality after allogeneic stem-cell transplantation (SCT).¹  In the December 15, 2007, issue of Blood, Zeiser and colleagues²  reported protective effects of atorvastatin against acute GVHD in a major histocompatibility complex (MHC) mismatched mouse model, while preserving graft-versus-leukemia (GVL) activity and donor engraftment. Through a series of elegant experiments the authors were able to delineate different effects of atorvastatin in donor T cells (Th-2 polarization) and recipient antigen presenting cells (downregulation of costimulatory molecules and MHC-II expression) leading to reduction in acute GVHD. Statins have shown promise in Th-1–mediated autoimmune diseases by promoting a Th-2 bias.^3,4  Hori et al reported activity of statins in patients with refractory chronic GVHD by promoting Th-2 polarization.⁵  The effects of statins on acute GVHD after human allogeneic stem cell transplantation have not been previously reported.

We therefore asked whether statin use at the time of allogeneic SCT would result in reduced acute GVHD while sparing GVL activity. Sixty-seven consecutive patients with acute leukemia underwent T cell–replete allogeneic SCT between June 2002 and October 2006 at our institution. Patients taking statins (defined as any 3-hydroxy-3-methylglutaryl-coenzyme-A [HMG-CoA] reductase inhibitor) at 40 mg/day or more for at least 1 month before and 3 months after allogeneic SCT (n = 10) were compared with those without a history of statin use (n = 57). The median age was 43 years (range, 20-73 yrs). Diagnosis included acute myeloid leukemia (AML; n = 49) and acute lymphoblastic leukemia (ALL; n = 18). The 2 groups had similar baseline characteristics including age, diagnosis, stem cell source, donor type, degree of HLA-compatibility, GVHD prophylaxis, conditioning regimen, and disease risk category (standard-risk defined as patients with acute leukemia receiving a transplant in first complete remission; Table 1). Acute GVHD was scored according to modified Glucksberg criteria.⁶  The rate of grade 2-IV acute GVHD was 10% (n = 1) in the statin group compared with 40% (n = 23) in the no-statin group (P = .08). Fifty-six patients were evaluable for chronic GVHD. No difference in the incidence of chronic GVHD was seen in patients using statins (55%) compared with those in the no-statin group (57%; P = .9). No patient in either group experienced primary or secondary engraftment failure. On subgroup analysis of patients with AML only (n = 49), a significantly reduced incidence of grade 2-IV acute GVHD was seen in the statin group (0%) compared with 43% (n = 18) in the no-statin group (P = .02). Rates of chronic GVHD were 43% and 58% in similar order (P = .68). We further tested to determine whether statin use, while reducing acute GVHD, mitigated the GVL effect in patients with AML. Kaplan-Meier estimates of progression-free survival (PFS) at 3years in AML patients with or without statin use were 54% and 28%, respectively (P = .17; Figure 1). This nonsignificant trend of improved PFS indicates that the GVL is preserved in patients using statins at the time of allografting.

In summary, our findings suggest that the recent observations made by Zeiser and colleagues in a murine model may have clinical relevance and hints at the potential of statins in reducing acute GVHD while preserving donor engraftment and GVL effect, at least in AML. Our results should be interpreted with caution given the limited and retrospective nature of this analysis. The optimal dose and timing of statin use to prevent GVHD is unknown. Whether priming healthy donors with statins before stem cell mobilization would provide added protection against acute GVHD compared with statin use in transplantation patients warrants further investigation.