Clinical and Immunologic Characteristics of Patients with Chronic Graft-Versus-Host Disease Persisting Seven or More Years after Diagnosis

Background

Chronic graft-versus-host disease (cGVHD) is a leading cause of morbidity and mortality after allogeneic hematopoietic stem cell transplant (SCT).  Median duration of systemic immunosuppression (IS) for cGVHD is approximately 2 years; however, 15% of patients still require IS 7 years after SCT.  Previous studies identified factors that were associated with longer duration of IS or cGVHD.  Our aim was to identify clinical and immunologic factors associated with cGVHD that is persistent ≥7 years after cGVHD diagnosis. 

Methods

Patients were drawn from a prospective cross-sectional study of the natural history of cGVHD at the National Institutes of Health (NCT00092235).  A cohort of patients who enrolled on the study ≥7 years from the time of cGVHD diagnosis, and thus had persistent cGVHD (pcGVHD), was compared to those who enrolled <1 year from cGVHD diagnosis.  Univariate analysis was performed, and significant factors were used in multivariable logistic regression.  With a large number of exploratory variables, all p-values<0.005 were considered statistically significant; 0.005≤p<0.05 represented strong trends.

Results

There were 38 patients with pcGVHD and 83 control patients <1 year from cGVHD diagnosis.  The two groups did not differ in age or performance status.  Factors that were significantly associated with pcGVHD (Table 1) included: bone marrow (BM) stem cell source (result divergent from previous studies), myeloablative conditioning, greater lung involvement, lower ferritin, lower IS, positive ENA (autoantibody panel), higher CD19 cells, CD4 cells, and immunoglobulin levels.  Factors with a trend towards significance included female donor, higher Lee cGVHD symptom scale scores, more sclerotic skin involvement and less erythematous skin involvement, and more active cGVHD as determined by therapeutic intent at the time of evaluation.  Interestingly, there were no differences in factors associated with disease severity (NIH global severity, clinician and patient rated severity, number of organs involved and organ scores), other inflammatory markers (ESR, CRP), functional capacity, distance from home to the transplant center, nor other factors identified in previous studies (bilirubin, HLA disparity).  In a multivariable analysis that included 11 variables, BM stem cells, +ENA, higher NIH lung score, higher platelets, and higher IgA levels were significant independent predictors of pcGVHD.

Conclusions

Although cGVHD is often self-limited, late forms requiring long duration of IS exist, and the predictive factors or underlying pathogenesis are unknown.  Our analysis of cGVHD patients who enrolled on a clinical trial ≥7 years after diagnosis showed more lung and sclerotic skin involvement, less inflammatory signs, and higher B cells, immunoglobulins, and autoantibodies.  These new findings suggest that pcGVHD may reflect irreversible damage rather than an active immune process; however, standardly accepted measures of disease severity were not associated with pcGVHD, suggesting that further tools are needed to differentiate accumulated damage from active disease.  Distance from transplant center, which may contribute to access to care, was not associated, although increasing IS was recommended more frequently for pcGVHD patients.  Factors that were previously identified as associated with longer duration of IS were not different or conflicted with our findings, suggesting that further study is needed.  To further elucidate potential immune dysfunction in patients with pcGVHD, our ongoing studies are measuring BAFF, cytokine, and chemokine levels.  Our results contribute to further hypotheses in the pathogenesis and contributing factors in patients who have pcGVHD.

Table 1. Univariate analysis - factors significantly associated with pcGVHD

	≥7 years from diagnosis (N=38)	<1 year from diagnosis (N=83)	p-value
BM stem cells (% patients)	58	11	<.0001
Myeloablative conditioning (% patients)	86	45	<.0001
+ENA (% patients)	18	1	.001
NIH lung score (% with 2-3)	50	22	.002
Intensity of IS (% with mod/high)	47	80	.0005
Prednisone mg/kg/day*	0	0.2	.001
Lee cGVHD symptom scale – breathing score*	5	2	.001
FEV1*	69	93	.001
Platelets*	278	193	<.0001
Albumin*	4	3.5	<.0001
Ferritin*	150	1240	<.0001
CD4 cells*	534	327	.002
CD19 cells*	266	87	.0004
IgG*	756	516	.0005
IgA*	145	32	<.0001

*median