The Health Outcome
Graft versus Host Disease (GVHD) is a condition that can result from hematopoietic cell transplant (HCT). It affects a large percentage of individuals who receive HCT; the chronic form occurs in 50% of HLA-matched sibling transplants, and in 64% of patients whose donors were HLA-unmatched. Frequency of the acute form varies widely according to other risk factors such as age and underlying illness: from 19% to 66% in HLA-matched transplant patients, and 70% to 90% of HLA-unmatched transplants patients (1).
GVHD results when mature T lymphocytes are introduced to a recipient through HCT. The T-cells respond to the recipient’s tissues as they would to a foreign protein in the body of the donor. These T-cells can persist for a long period, and possibly proliferate in response to antigen recognition, either those foreign to the recipient or derived from the recipient’s own tissues.
The effects of donor-T-cell alloresponse can take both acute and chronic forms. The acute form occurs within 30 days of transplant. The symptoms are characteristic of systemic cytokine overload. The chronic form usually follows the acute form, though it can also occur de novo. It resembles an autoimmune syndrome and is generally expressed in epidermal tissues.
Lin et al. propose that single-nucleotide polymorphisms in various cytokines may affect the outcome of GVHD by reducing or enhancing the inflammatory response of the recipient. Only
IL10 was found to be significantly associated with disease outcomes, though
IL1B,
IL1RA,
IL6 and
TNFA were also evaluated and found not to be associated. IL-10 is produced by T-cells and macrophages and is thought to influence Th0 cells to differentiate into Th2 cells. These cells favor a humoral immune response, which is antibody-mediated, as opposed to the cytotoxic T-cell-mediated cellular immune response favored by Th1 cells. Because Th1 cells favor an inflammatory response, IL-10 is considered an anti-inflammatory cytokine.
The Finding
Two cohorts of patients undergoing HCT were used in this study; in both cohorts, all patients received donations from HLA-matched siblings. The first cohort included 570 HLA-A2-positive individuals, while the second cohort included 423 mostly HLA-A2-negative individuals. Further analyses were carried out on the combined cohort of 993 individuals.
GVHD was defined and evaluated by standard criteria. Proportional hazards regression models were used with adjustment for known risk factors to evaluate associations between single polymorphisms and disease outcomes. The disease outcomes evaluated included acute GVHD, chronic GVHD and death in remission. In analysis of chronic GVHD, age, sex mismatch, use of total body irradiation, and disease group (reason for HCT) were controlled for. In acute GVHD analysis, the above were adjusted for, in addition to year of transplant, as there was a change in diagnostic ability in 1991. In both of these analyses, death in remission was considered a competing event and these data were censored. In analysis of death in remission, age, time from diagnosis, and disease group were adjusted for. In this analysis and analysis of chronic GVHD, relapse was considered a competing event, and these data were censored. These analyses were carried out in the first cohort, and polymorphisms with hazard ratios for any disease outcome greater than 2.0 or less than .5 were repeated in the second cohort. Trend analyses were also conducted and those with a p-value less than .05 were repeated.
Only the IL10 -592 and -1082 polymorphisms met these criteria. These are both upstream from IL10 exon 1, and are assumed to be in the promoter region. These are two of three proximal single-nucleotide polymorphisms in IL10; an additional two have been identified in the distal region.
In the second cohort, frequency of the IL10-592 allele proved too small to evaluate statistically, though the association was repeated. The cohorts were combined for further analyses. Two additional promoter polymorphisms were analyzed. In the combined cohort of 993 individuals, the genotype groups did not differ significantly by any other risk factors.
In the combined cohort, the IL10 -592 allele was associated with acute GVHD but not with chronic GVHD. Homozygosity for the common allele, with cytosine at the -592 position, was taken as the referent group. The adenine substitution was associated with a protective effect for grade III or IV acute GVHD with a hazard ratio of 0.7 for A/C heterozygotes and 0.4 for A/A homozygotes. The p-value for the test of trend across genotypes was 0.001. Three other IL10 polymorphisms were analyzed for associations with acute GVHD, including -1082, -2763, and -3575, none of which were associated significantly with the outcome. The A/A genotype was associated with a low risk of death in remission (HR, 0.6; CI 0.3 - 1.0; p=.05) compared with the C/C genotype, but the significance of this finding is questionable. Risk of death in remission was not different in the A/C group than in the C/C group.
The IL10 -592 allele is in linkage disequilibrium with other promoter polymorphisms. The authors assumed complete linkage disequilibrium between a -819 allele and the -592 allele. Three proximal haplotypes (A-T-A, A-C-C, and G-C-C at positions -1082, -819, and -592) and the six resulting genotypes were analyzed for association with acute GVHD. The A-T-A/A-T-A group was taken as a reference. The two A-T-A heterozygous groups had hazard ratios higher than null but not significantly. The genotypes without the -592 allele all had hazard ratios significantly higher than null, but in no case was the association strong.
The authors present frequency data within the combined cohort, though they draw no conclusions based on this information. It is difficult to evaluate whether the proportions of underlying diseases in the study groups are comparable to those found in the general population. While there is no evidence, at present, that these polymorphisms may be associated with the underlying disease, the possibility cannot be excluded.
Public Health Implications
The effects observed in this study were small. It is unlikely that a genetic test for IL10 promoter polymorphisms alone will ultimately prove useful for assessing the risk of developing GVHD in a potential transplant recipient. However, this study contributes to the body of literature suggesting an important role for IL-10 in GVHD outcomes. Studies of endogenous IL-10 in HCT patients corroborate these findings (2).
One major weakness of the hypothesis presented here is the lack of evidence that a high-IL-10 phenotype is associated with the -592A genotype. Studies of this association have had conflicting results, with some suggesting that the common C allele is associated with higher IL-10 production, others finding an association between the A allele and higher IL-10 production (3-4). However, these studies were carried out in populations of patients with juvenile rheumatoid arthritis, and systemic lupus erythematosis, respectively. These inflammatory conditions affect cytokine production in such a way as to make these associations inapplicable to other populations, either of well people or those requiring HCT. Though treatment for the diseases underlying the need for HCT transplant are immunosuppressive, the resumption of IL-10 production could be evaluated over time, post-transplant, to assess whether endogenous IL-10 production is associated with promoter polymorphisms as it might affect the population at risk for GVHD.
Although the potential for application at present is limited, studies such as this are vital to understanding the role of IL-10 and other cytokines in immune response. There is great potential for therapeutic use of many cytokines, including IL-10; but thorough evaluation of the effects of endogenous production is a necessary precursor. Given the biologic mechanism proposed here, it will be possible to test the role of these
IL-10 polymorphisms in other diseases, including allergy, autoimmune syndromes, and maternal-fetal GVHD. However, clinical applicability of these findings is limited by a lack of alternative treatment options for those with the highest risk for GVHD, which represent a large majority of the population
.
References
- http://www.emedicine.com/med/topic926.htm
- Holler E, et al. Prognostic significance of increased IL-10 production in patients prior to allogeneic bone marrow transplantation. Bone Marrow Transplant 2000, 25(3):237-41.
- Crawley E, et al. Polymorphic haplotypes of the interleukin-10 5' flanking region determine variable interleuking-10 transcription and are associated with particular phenotypes of juvenile rheumatoid arthritis. Arthritis and Rheumatism 1999 42:1101-1108.
- Gibson A, et al. Novel single nucleotide polymorphisms in the distal IL10 promoter affect IL10 production and enhance the risk of systemic lupus erythematosis. J Immunol 2001, 166:3915-3922.
