Survival without Progressive Impairment As a Novel Endpoint in Chronic Graft-Versus-Host Disease

Title: Survival without progressive impairment as a novel endpoint in chronic graft-versus-host disease.

Background: A primary goal in the management of chronic graft-versus-host disease (cGVHD) is to prevent the progression of organ-specific and global functional impairment due to the natural history of cGVHD or its treatment.  We tested a definition of progressive impairment in a multicenter cohort. Methods: Of the measures recommended by the 2005 NIH Consensus Conference, changes in 21 items having face validity for “impairment” were selected from these categories: clinician-reported organ involvement; FEV1; and patient-reported performance status and physical function.  Results: 575 patients were evaluated, and 237 (41%) met criteria for progressive impairment during a median of 39.9 (range 3.8 to 69.2) months of follow-up.  125 (53%) had progressive impairment on the basis of clinically assessed measures, 61 (26%) on the basis of patient-reported performance status or physical function, and 51 (22%) on the basis of more than one category. The top 5 items indicating progressive impairment were: decline in FEV1 by 10% (81/237), decrease in the Human Activity Profile Adjusted Activity Score (HAP AAS) score by ≥0.5 standard deviation (76/237), increase in the global measure of skin sclerotic changes (70/237), increase in the fascia score (57/237), and increase in the global skin score (51/237). Rates of survival without progressive impairment or relapse were 72% at 6 months, 52% at 12 months, and 35% at 24 months. Cumulative incidences of progressive impairment were 20% at 6 months, 33% at 12 months, and 43% at 24 months. Patients with progressive impairment had significant worsening in clinician and patient-assessed cGVHD severity, symptoms and quality of life as compared to those without progressive impairment (Table). Conclusion: Treatments that prevent unacceptable clinical deterioration in patients with chronic GVHD could be identified by using survival without progressive impairment as an endpoint in clinical trials.

	Change from enrollment to visit with progressive impairment (N=237)					Change from enrollment to last visit without impairment (N=130)
Variable	N	Median	Mean	Min	Max	N	Median	Mean	Min	Max	p-value
MD 0-3*	233	0	-0.15	-3	2	128	0	-0.59	-3	1	<0.001
MD 0-10**	232	-1	-0.64	-8	6	128	-2	-1.98	-8	2	<0.001
PT 0-3*	179	0	-0.01	-2	2	59	-1	-0.58	-2	1	<0.001
PT 0-10**	176	0	-0.37	-7	6	63	-2	-1.62	-6	3	<0.001
FACT-BMT	180	-0.92	0	-44.67	54.67	60	13	15	-24.33	45.94	<0.001
Lee symptom summary	190	-0.27	-0.46	-45.22	33.57	63	-6.05	-7.59	-38.81	10.85	<0.001

*       Clinician (MD) or Patient (PT) -rated overall chronic GVHD severity on a 0-3 scale (none, mild, moderate, severe)

**     MD or PT-rated overall chronic GVHD severity on a 0-10 scale

FACT-BMT = functional assessment of cancer therapy, bone marrow transplant