Evaluation of NIH Consensus Criteria for Classification of Late Acute and Chronic Gvhd in Reduce Intensity Conditioning (RIC) Stem Cell Transplantation (SCT).

Chronic graft-versus-host disease (GVHD) occurs frequently after allogeneic stem cell transplantation (SCT) and has an impact on morbidity and survival. The National Institutes of Heath (NIH) consensus criteria for the diagnosis of GVHD, emphasized clinical manifestations of GVHD rather than the classical time of onset (day 100). Incidence and impact in term of relapse and no-relapse mortality (NRM) of this new classification is not well known after RIC.

We retrospectively reviewed 116 consecutive patients (pts) in Saint Louis' Hospital undergoing an SCT for hematologic malignancy and surviving at least day + 100 after RIC between August 2005 and December 2008. We evaluated non-relapse mortality (NRM) and recurrent malignancy. Cumulative incidence was computed using death as a competing event. Incidence of relapse and NRM was counted from 100 days post-transplant for patients without chronic GVHD or from chronic GVHD onset. Patients with relapse/progression before chronic GVHD onset were considered as not having chronic GVHD in these analyses. The association of occurrence of chronic GVHD with the risk of relapse and non-relapse death was analyzed using time-dependent covariates in cause-specific proportional hazards models.

Among 116 pts ( M/F: 71/45), with a median age of 53 years old (19-68 years) 28 pts (24%) were transplanted for acute leukemia in, 11 pts (9%) for chronic leukemia, 27 pts (23%) for lymphoma, 30 pts (26%) for MPD/MDS and 20 pts (17%) for plasma cell disorder. Sixty-three pts (54%) received HLA-identical sibling transplantation, 53 pts (46%) received transplantation from unrelated donors. Source of stem cells was mobilized peripheral blood stem cell for 108 pts (93%), bone marrow for 4 pts (3%) and 4 cord blood (3%).

After a median follow-up of 18 months (range 5-45 months), a total of 67 pts (58%) developed chronic GVHD according to the Seattle day 100 landmark criteria and when using NIH consensus criteria, 55 pts (47%) developed chronic GVHD, including 43 pts (53%) with classic chronic GVHD and 8 pts (10%) overlap syndrome. Patients reclassified included; 3 pts with late onset acute GvHD, 19 pts had recurrent and 8 had persistent acute GVHD (numbers do not to previous sentence because some of these patients latter developed chronic GvHD). The cumulative incidence of chronic GVHD at 36 months was 64% (95%CI; 53%-73%) when using Seattle criteria compared to 56% (95%CI; 45%-67%) with NIH chronic GVHD criteria.

Two-year Cumulative incidences of relapse and NRM using both classifications are summarized in Table.

In Cox model with GvHD as a time dependent covariate, the NRM was significantly higher in patients with late onset, persistent and recurrent acute GVHD compared to no GVHD (hazard ratio (HR) 31, 47 and 30; p = 0.005, p <0.0001, p <0.0001, respectively), whereas the NRM was statistically increased in case of chronic GVHD using Seattle day 100 criteria (HR: 2.8; P=0.034).

The cumulative incidence of chronic GVHD “decrease” about 10% when using NIH consensus criteria compared to Seattle criteria in our cohort of RIC. Most of the NRM occurred beyond 100 days after SCT was due to the increased risk of NRM in patients with late onset, recurrent or persistent acute GVHD.

No relevant conflicts of interest to declare.