Late Onset Acute Gvhd (late aGVHD) after Allogeneic Hematopoietic Stem Cell Transplantation (HCT)

Introduction: NIH Consensus criteria for diagnosis of chronic GVHD (cGVHD) have been examined in several retrospective studies evaluating cGVHD and its comparison to other GVHD syndromes. There is less data regarding late aGVHD with onset beyond 100 days after HCT. We evaluated a cohort of allogeneic HCT recipients from 2007 to 2012 who developed persistent, recurrent or denovo late aGVHD.

Patients and Methods: 75 patients (median age 50.7 years) were classified as late aGVHD.  Donors were HLA-identical sibling in 41.3%, UCB in 45.3%, URD in 13.3%. 46.7% recipients received PBSC and 8% marrow as the graft source. 64% received a reduced intensity conditioning. Median follow up was 4 years.

Results: The cumulative incidence of any late aGVHD was 14.7% (95% CI: 11.6-17.8) and grade II-IV late aGVHD was 11.7% (95% CI 8.7 - 14.7) at one year (Figure 1). 52% had persistent (defined as active aGVHD that persisted without resolution after 100 days), 39% had recurrent (aGVHD that recurred following resolution after 100 days) and 9% had de-novo late aGVHD. Median time to onset was 155 days for de-novo late aGVHD, and 143 days for recurrent late aGVHD (Table). Persistent late aGVHD was considered to have a time of onset of 100 days. Outcomes were estimated from onset of late aGVHD. (Table). One and 2 year overall survival was 73% (95% CI 63-84) and 59% (95% CI 49-72). Figure 2 shows the cumulative incidences of outcomes after late aGVHD. Progression to cGVHD or overlap syndrome was seen in 45% (95% CI 35-57) by two years. Durable discontinuation of immunosuppression (IS) was achieved in 15%, 41% and 57% of persistent, recurrent and de-novo late aGVHD. In multivariate analysis, PBSC [HR 2.1, 95%CI: 1.2-3.5, p=0.01] and myeloablative conditioning [HR 1.8, 95%CI: 1.0-3.2, p=0.04)] were associated with higher risks of developing late aGVHD.

Conclusions: We present results of clinical presentation and outcomes in patients with late aGVHD. Amongst patients with late aGVHD, durable discontinuation of IS was seen in only 15% of those with persistent late aGVHD, indicating need for intensified therapy in this group. PBSC and myeloablative conditioning were associated with a higher risk of developing late aGVHD. Modulation of these risk factors (such as intensified or longer duration of GVHD prophylaxis in these higher risk groups) may allow reduced risks and improved transplant outcomes.

Table: Comparison of late aGVHD categories with classic aGVHD

	De-novo n=7	Persistent n=39	Recurrent n=29	Classic Acute n=268
Median time to onset, range	155 (101-262)	100	143 (102-342)	36 (7-180)
2 year overall survival after onset (95% CI)	83%(27-97)	45% (28-60)	74% (53-87)	50% (44-57)
1 year NRM after onset	0%	26% (12-40)	14% (4-26)	24% (19-29)
Progression to cGVHD within 2 years of onset	42% (16-75)	51% (36-66)	38% (23-56)	31% (26-37)

Figure 1. Cumulative Incidence of Grade II-IV late aGVHD

Figure 2. Cumulative incidence of outcomes following late aGVHD