Improved Overall Survival in Patients with Grade III and IV Acute Graft-Versus-Host Disease

Introduction: In the last two decades, there have been multiple reasons for the improved outcomes of patients undergoing allogeneic HSCT including more precise HLA matching, the advent of reduced intensity conditioning regimens, use of novel antimicrobial agents, increased options for immunosuppression and better patient selection.  However, the collective impact of these advances on outcomes of patients who develop grade III or IV acute GVHD remains unclear.

Methods: We performed a retrospective analysis of 428 patients with grade III or IV acute GVHD treated at Massachusetts General Hospital and Dana-Farber Cancer Institute from 1997-2012.We compared overall survival (OS), progression-free survival (PFS) from underlying disease, and GVHD-related mortality (GRM) in 3 cohorts based on the year of transplantation: 1997-2001 (n=111), 2002-2006 (n=112), and 2007-2012 (n=205) using multivariate analysis adjusting for age, diagnosis, disease status at transplant, grade of acute GVHD, performance status, comorbidities, conditioning intensity, HLA match, donor source, and donor-patient gender match.

Results: The most recent cohort of patients with grade III or IV acute GVHD (2007-2012) was significantly older (31% > 60 in 2007-2012, 12% > 60 in 2002-2006, and 4% > 60 in 1997-2001, p < 0.001), and had lower proportion (35%) of grade IV acute GVHD compared to earlier cohorts (52% in 1997-2001, and 38% in 2002-2006, p= 0.01).  The use of reduced intensity conditioning, unrelated donors, and peripheral blood stem cells among patients with grade III-IV acute GVHD increased over time.   In the unadjusted analysis, 12-month OS increased over time (28% in 1997-2001, 35% in 2002-2006, 48% in 2007-2012, P = 0.004) reflecting a decrease in 12-month GRM (54% in 1997-2001, 51% in 2002-2006, 31% in 2007-2012, P = 0.0001) and improvement in 12-month PFS (27% in 1997-2001, 30% in 2002-2006, 42% in 2007-2012, p = 0.02).  After multivariate analysis, the period of transplantation remained a strong predictor for OS, and the hazard ratio (HRs) in 1997-2001 and 2002-2006 compared with 2007-2012 were 1.55 (95% CI, 1.0-2.3, p = 0.03), and 1.54 (95% CI, 1.2-2.1, p = 0.004), respectively.  Similarly, more recent period of transplantation remained a significant predictor for lower GRM [HRs in 1997-2001 and 2002-2006 compared with 2007-2012 were 1.65 (95% CI, 1.0-2.6, p=0.04), and 2.28 (95% CI 1.6-3.3, p < 0.001)] and longer PFS [HRs in 1997-2001 and 2002-2006 compared with 2007-2012 were 1.5 (95% CI, 1.0-2.2, p = 0.04), and 1.49 (95% CI 1.1-2.0, p = 0.007)].

Conclusion: The outcomes of patients who develop grade III or IV acute GVHD after allogeneic HSCT has improved over time with lower GRM and improvement in PFS and OS.  Our results suggest that the multiple innovations in transplantation care have led to improved outcomes in patients with grade III and IV acute GVHD, and outcomes are not as poor as historically reported.