|
|
Background Toll-like receptors (TLRs) are essential components of the immune response to fungal pathogens. We examined the role of TLR polymorphisms in conferring a risk of invasive aspergillosis among recipients of allogeneic hematopoietic-cell transplants.
Methods We analyzed 20 single-nucleotide polymorphisms (SNPs) in the toll-like receptor 2 gene (TLR2), the toll-like receptor 3 gene (TLR3), the toll-like receptor 4 gene (TLR4), and the toll-like receptor 9 gene (TLR9) in a cohort of 336 recipients of hematopoietic-cell transplants and their unrelated donors. The risk of invasive aspergillosis was assessed with the use of multivariate Cox regression analysis. The analysis was replicated in a validation study involving 103 case patients and 263 matched controls who received hematopoietic-cell transplants from related and unrelated donors.
Results In the discovery study, two donor TLR4 haplotypes (S3 and S4) increased the risk of invasive aspergillosis (adjusted hazard ratio for S3, 2.20; 95% confidence interval [CI], 1.14 to 4.25; P=0.02; adjusted hazard ratio for S4, 6.16; 95% CI, 1.97 to 19.26; P=0.002). The haplotype S4 was present in carriers of two SNPs in strong linkage disequilibrium (1063 A/G [D299G] and 1363 C/T [T399I]) that influence TLR4 function. In the validation study, donor haplotype S4 also increased the risk of invasive aspergillosis (adjusted odds ratio, 2.49; 95% CI, 1.15 to 5.41; P=0.02); the association was present in unrelated recipients of hematopoietic-cell transplants (odds ratio, 5.00; 95% CI, 1.04 to 24.01; P=0.04) but not in related recipients (odds ratio, 2.29; 95% CI, 0.93 to 5.68; P=0.07). In the discovery study, seropositivity for cytomegalovirus (CMV) in donors or recipients, donor positivity for S4, or both, as compared with negative results for CMV and S4, were associated with an increase in the 3-year probability of invasive aspergillosis (12% vs. 1%, P=0.02) and death that was not related to relapse (35% vs. 22%, P=0.02).
Conclusions This study suggests an association between the donor TLR4 haplotype S4 and the risk of invasive aspergillosis among recipients of hematopoietic-cell transplants from unrelated donors.
Toll-like receptors (TLRs) are transmembrane proteins on the surface of immune cells that detect conserved molecular motifs known as "microbe-associated molecular patterns" from a variety of organisms. They interact with several adapter proteins to activate transcription factors, leading to the production of inflammatory cytokines and the activation of adaptive immunity.3,4 Aspergillus species activate innate immune cells through both toll-like receptor 2 (TLR2) and toll-like receptor 4 (TLR4).5,6,7,8,9,10
Common polymorphisms in TLR genes have been associated with susceptibility to several infections.11 Mutations in TLR adapter molecules (i.e., interleukin-1 receptor–associated kinase 4 [IRAK4], inhibitor of nuclear factor 
B kinase 
[IKK], and inhibitor of nuclear factor B 
[IB]) cause rare inherited immunodeficiencies, further demonstrating the importance of genetic variability in the TLR signaling pathway.11 In the transplant recipient, inhaled molds may interact with epithelial cells from the recipient or phagocytic cells from the donor lineage; both cells express TLRs. We hypothesized that polymorphisms in TLR genes from the donor and the recipient may influence susceptibility to invasive aspergillosis in recipients of hematopoietic-cell transplants. To test this hypothesis, we analyzed the frequencies of single-nucleotide polymorphisms (SNPs) for four candidate TLR genes in a cohort of recipients of hematopoietic-cell transplants and unrelated donors.
Methods
Patients
All patients who received an allogeneic hematopoietic-cell transplant from the Fred Hutchinson Cancer Research Center and the Seattle Cancer Care Alliance between 1989 and 2004 were eligible for this analysis. Specimens were obtained and DNA was extracted by our Human Immunogenetics Program in an ongoing process that was based solely on the availability of specimens, and no preferences were given to any specific clinical variables. For the discovery and validation studies, patients were selected sequentially from the overall pool of specimens available at the time the respective studies were conducted, and no patient was included in both studies.
For the discovery study, we identified 374 patients with available DNA who underwent their first allogeneic hematopoietic-cell transplantation from an unrelated donor between 1995 and 2003. Cases were identified from microbiologic and histopathological reports in clinical charts and classified as "probable" or "proven" invasive aspergillosis, with the use of definitions developed by the Mycoses Study Group of the National Institute of Allergy and Infectious Diseases and the European Organization for Research and Treatment of Cancer.12 Among the 374 patients, 38 were excluded for one or more of the following reasons: both the donor and recipient had a nonwhite or unknown race or ethnic group (32 patients), invasive aspergillosis occurred before the first or after a second transplantation (5), the diagnosis of invasive aspergillosis was classified as "possible" (3), and infection was due to an invasive mold other than aspergillus species (2).
For the validation study, we used a matched case–control design and included recipients of hematopoietic-cell transplants from related donors. A minimal number of 100 cases with at least one matched control was required for odd ratios of approximately 2.3, 2.0, and 1.8, with 80% power and a 5% type I error rate, for allelic frequencies of 10%, 20%, and 30%, respectively (calculated with the use of Quanto 1.2.3 software).13 Among the 223 patients in whom invasive aspergillosis occurred between 1989 and 2004, identified from the overall pool from the Human Immunogenetics Program, 120 patients were excluded for one of the following reasons: the donor had an unknown or a nonwhite race or ethnic group (67 patients), invasive aspergillosis occurred before the first or after a second transplantation (37), or the diagnosis of invasive aspergillosis was classified as "possible" (16). For each of the 103 remaining patients with invasive aspergillosis, 1 to 6 controls were matched according to age (<40 or 
40 years), donor race or ethnic group, donor type, and year of transplantation (±5 years). Among 404 selected controls, 263 had available DNA and were included in the validation study. A subgroup analysis was performed to account for differences in the types of hematopoietic-cell transplants.
Graft-versus-host disease (GVHD) and cytomegalovirus (CMV) disease were diagnosed according to standard definitions.14,15,16 All patients received systemic antibiotic prophylaxis during the period of neutropenia, as well as prophylactic acyclovir, trimethoprim–sulfamethoxazole or dapsone, and fluconazole, until day 75 after hematopoietic-cell transplantation. A small number of patients (35 patients in the discovery study and 13 in the validation study) received itraconazole or voriconazole prophylaxis on the basis of the individual physician's decision or as part of a prospective, randomized study of the efficacy of itraconazole versus fluconazole prophylaxis.17
Patients and donors gave written informed consent for data collection, cell storage, and cryopreservation. Batched genotyping was performed according to protocols approved by the institutional review board of the Fred Hutchinson Cancer Research Center. The study was also approved by the Western Institutional Review Board (Olympia, WA).
Single-Nucleotide Polymorphisms
Previously identified minimal haplotype-tagging SNPs from the Innate Immunity Program in Genomic Applications database (http://innateimmunity.net/) were used for each candidate gene. Five additional SNPs that have previously been reported to be associated with human diseases were also genotyped. A total of 20 SNPs in TLR2, TLR3, TLR4, and TLR9 were identified with the use of a mass-spectrometry genotyping platform (Sequenom).18
Statistical Analysis
Hardy–Weinberg equilibrium and pairwise linkage disequilibrium were calculated with the use of the genwh and pwld programs developed in Stata software, version 9.2 (StataCorp). We considered an R2 value of more than 0.90 to represent strong linkage disequilibrium. Haplotypes were inferred with the use of the expectation-maximization algorithm implemented in the Decipher program (SAGE).19 We grouped all rare haplotypes (frequencies <2%) into one group for each gene. All analyses were performed for white donors, white recipients, or both; race was reported by the donors and recipients.
The probability of invasive aspergillosis according to TLR variants at 6 and 36 months after transplantation was estimated with the use of cumulative incidence curves, with censoring of data at the date of the last follow-up and with death, second transplantation, and morphologic relapse treated as competing risks.20 A period of 36 months was chosen because all cases of invasive aspergillosis in the discovery cohort had occurred during this period, and a period of 6 months was chosen in order to assess early invasive aspergillosis. Comparisons of the risks of invasive aspergillosis and death on the basis of the TLR variants were performed with the use of univariate Cox regression models, unless otherwise specified. Multivariate Cox models were used to evaluate the relative hazards associated with TLR variants. Candidate covariates were selected from the factors associated with aspergillosis or death in the univariate analysis (P
0.15) or in previous studies.1 The covariates were entered one by one in a pairwise model together with the significant TLR variant and kept in the final model if they remained significant (P<0.05) or altered the association with the TLR variant by more than 10%. Conditional logistic regression was used in the analysis of the data from the case–control validation study, with the use of similar methods for selecting variables.
Because of the large number of statistical comparisons made in the discovery study, the P values presented should be interpreted cautiously. To facilitate this interpretation, we report P values that would remain significant with the application of traditional Bonferroni corrections for multiple comparisons. Among the 20 SNPs tested, two pairs of SNPs were found to be in strong linkage disequilibrium: TLR4 1063 A/G and TLR4 1363 C/T (R2=0.96), and TLR9 +1174 G/A and TLR9 1635 A/G (R2=0.96); these findings are consistent with previous observations.21,22 Since the analyses for these SNPs are redundant, 18 independent tests were considered for multiple-testing correction. For haplotypes, we considered four independent tests (i.e., one for each gene). Since the case–control study was performed for validation and was limited to TLR4, no correction for multiple testing was applied in this study.
Results
Discovery Study
The discovery study included 336 patients who underwent allogeneic hematopoietic-cell transplantation after a myeloablative-conditioning regimen (Table 1). The median follow-up time among patients who survived was 84 months (range, 5 to 124). Among the 336 patients, 33 had proven or probable invasive aspergillosis.
|
To assess the risk of invasive aspergillosis according to TLR SNPs and haplotypes in recipients and donors of hematopoietic-cell transplants, we estimated the cumulative incidence of infection 6 months and 36 months after hematopoietic-cell transplantation. Two SNPs in donors of hematopoietic-cell transplants (1063 A/G [D299G] and 1363 C/T [T399I]), which were in strong linkage disequilibrium (R2=0.96), influenced the risk of the development of invasive aspergillosis (see Table S1 in the Supplementary Appendix, available with the full text of this article at www.nejm.org). The association was stronger at 6 months (for TLR4 1363T, the cumulative incidence of invasive aspergillosis was 22% in the presence of the allele vs. 5% in its absence; P=0.002) (Table 2) and remained significant after correction for multiple testing (P=0.03). Another donor SNP in TLR4, –2604G, tended to influence the risk of invasive aspergillosis when the comparison was made at 36 months (cumulative incidence of invasive aspergillosis, 4% in the presence of the allele vs. 13% in its absence; P=0.01), but the association was no longer present when correction for multiple testing was applied (P=0.20).
|
|
The final SNP model included TLR4 –2604G (adjusted hazard ratio, 3.22; 95% confidence interval [CI], 1.02 to 10.16; P=0.05) (Table 3) and 1363T (adjusted hazard ratio, 4.96; 95% CI, 1.52 to 16.24; P=0.008). In the final haplotype model, the hazard ratio for S3 was 2.20 (95% CI, 1.14 to 4.25; P=0.02) and the hazard ratio for S4 was 6.16 (95% CI, 1.97 to 19.26; P=0.002). Both final models also included seropositivity for CMV in the patient, the donor, or both (hazard ratio, 4.90; 95% CI, 1.51 to 15.91; P=0.008) and acute GVHD (hazard ratio, 3.11; 95% CI, 1.22 to 7.90; P=0.02) in the haplotype model, indicating that donor S4 (or TLR4 1363T), CMV seropositivity, and acute GVHD were independent risk factors for invasive aspergillosis. The interaction between positivity for CMV and S4 did not influence the risk of invasive aspergillosis (P=0.23).
|
The characteristics of the patients in the case–control study were similar to those of the patients in the initial study (Table 1), but the validation study included recipients of transplants from both related and unrelated donors, and the median time to invasive aspergillosis after transplantation was shorter (68 days vs. 100 days). The association of donor TLR4 S4 with the risk of invasive aspergillosis was confirmed in the validation study (adjusted odds ratio, 2.49; 95% CI, 1.15 to 5.41; P=0.02) (Table 3). In the subgroup analysis, the association between S4 and invasive aspergillosis was significant among recipients of transplants from unrelated donors (adjusted odds ratio, 5.00; 95% CI, 1.04 to 24.01; P=0.04) (Table 3) but not among recipients of transplants from related donors (adjusted odds ratio, 2.29; 95% CI, 0.93 to 5.68; P=0.07). No association between S3 and invasive aspergillosis was observed in the validation study. However, in this group, haplotype S2 tended to increase the risk of invasive aspergillosis (adjusted odds ratio, 1.86; 95% CI, 1.01 to 3.44; P=0.05); this haplotype was present in TLR4 +12186G carriers, and this allele also increased the risk of invasive aspergillosis in the SNP model (adjusted odds ratio, 2.18; 95% CI, 1.13 to 4.21; P=0.02).
Risk Assessment before Hematopoietic-Cell Transplantation
Since CMV serologic testing and TLR4 haplotype analysis can both be used before hematopoietic-cell transplantation to evaluate the risk of invasive aspergillosis and death, patients were stratified into groups according to CMV serologic status (CMV-negative vs. CMV-positive) and status with respect to donor haplotype S4 (S4-negative vs. S4-positive) (Figure 1). In the discovery study, invasive aspergillosis occurred in 1 of 83 CMV-negative and S4-negative patients (1%), 14 of 136 CMV-positive and S4-negative patients (10%), and 5 of 23 CMV-negative and S4-positive or CMV-positive and S4-positive patients (22%) (Figure 1A). Since the number of patients in some groups was small, all patients who were CMV-positive, S4-positive, or both were grouped together (Figure 1B and 1C). The cumulative incidence of invasive aspergillosis increased from 1% in 83 CMV-negative and S4-negative patients to 12% in 159 patients who were CMV-positive, S4-positive, or both (hazard ratio, 11.95; 95% CI, 1.60 to 89.34; P=0.02, after adjustment for pretransplantation covariates). The cumulative incidence of death not related to relapse increased from 22% among CMV-negative and S4-negative patients to 35% among patients who were CMV-positive, S4-positive, or both (hazard ratio, 1.88; 95% CI, 1.10 to 3.20; P=0.02, after adjustment for pretransplantation covariates) (Table 4). In the validation study, the odds ratio for invasive aspergillosis among patients who were CMV-positive and S4-positive or CMV-negative and S4-positive, as compared with those who were CMV-negative and S4-negative, was 3.10 (95% CI, 1.29 to 7.45; P=0.01), and for patients who were CMV-positive and S4-negative, the odds ratio was 1.54 (95% CI, 0.84 to 2.85; P=0.16), after adjustment for pretransplantation and post-transplantation covariates.
|
These data suggest an association between the TLR4 haplotype S4 in donors of hematopoietic-cell transplants (haplotype frequency, 6%) and invasive aspergillosis in recipients of hematopoietic-cell transplants. This haplotype was present in carriers of two SNPs in strong linkage disequilibrium (1063G and 1363T) that influence TLR4 function.
The TLR genes were initially identified through their homology with the Toll gene in Drosophila melanogaster, which triggers the production of drosomycin and is required for effective protection against fungal infections.23 Mammalian TLR4 was first recognized as the transmembrane receptor for lipopolysaccharide, a key component for the detection of gram-negative bacteria.24 Subsequent studies have shown that TLR4 is also involved in the recognition of fungal ligands, including mannan (derived from Candida albicans)25 and glucuronoxylomannan (derived from Cryptococcus neoformans).26 Several studies have highlighted the role of TLR2 and TLR4 in the innate immune recognition of aspergillus species.5,6,7,8,9,10 Although TLR2-dependent responses have been shown to augment β-1,3 glucan stimulation through dectin-1, the microbial Aspergillus fumigatus TLR2 and TLR4 ligands have not yet been identified.9
TLR4 1063G and 1363T SNPs have been reported to be associated with susceptibility to infections caused by gram-negative bacteria,27,28 C. albicans,29 brucella species,30 respiratory syncytial virus,31 and Plasmodium falciparum.32 Persons who are heterozygous for the 1063G and 1363T SNPs are hyporesponsive to lipopolysaccharide, as measured by the bronchospastic response after inhalation of endotoxin.33 Furthermore, airway epithelial cells isolated from heterozygous persons were reported to have a deficient response to lipopolysaccharide, suggesting that 1063G and 1363T act in a dominant fashion with respect to the wild-type allele.33 However, monocytes and whole blood isolated from heterozygous subjects did not have abnormal responses to lipopolysaccharide,34,35 suggesting that the effects of these polymorphisms may vary among cell types or experimenta l conditions or that polymorphisms other than 1063G and 1363T interfere with the results.
A possible limitation of this study is the fact that genetic associations can be biased by population stratification.36 However, we observed a similar association in two separate studies, suggesting that a confounding effect is unlikely. Another limitation is the possibility that TLR4 SNPs and haplotypes associated with invasive aspergillosis are in linkage disequilibrium with SNPs in another gene, located nearby, that are responsible for the alterations in the phenotype. However, the fact that TLR4 is an essential receptor for aspergillus species and that TLR4 1063 A/G and 1363 C/T SNPs impair TLR4 function suggests that the polymorphisms in TLR4 are probably responsible for the observed phenotypes. Finally, polymorphisms may not be directly associated with invasive aspergillosis, but instead may be associated with other factors, such as acute GVHD or CMV seropositivity, that also influence the end point.37,38 However, this possibility is unlikely, given the multivariate analysis, which identified donor S4, acute GVHD, and CMV seropositivity as three independent risk factors.
Polymorphisms in genes encoding interleukin-10, tumor necrosis factor receptor type 2, TLR1, and TLR6 have previously been shown to influence susceptibility to various clinical forms of aspergillosis.39,40,41,42,43 However, the TLR4 polymorphisms that we found to be significantly associated with invasive aspergillosis were derived from donor cells in hematopoietic-cell transplants, not from recipient cells. Thus, patients who undergo hematopoietic-cell transplantation are chimeras for whom the immune system is derived in part from the recipient (e.g., epithelial cells) and in part from the donor (hematopoietic cells). Polymorphisms in TLR genes expressed on the surface of polymorphonuclear neutrophils10 or resident macrophages, which are from the donor's cell lineage, may impair cellular activation and antifungal responses.
The association of S4 with an increased risk of invasive aspergillosis, observed in the discovery study, was also observed in the validation study, which included recipients of transplants from related donors and from unrelated donors. In the subgroup analysis, the association was present among the patients who received transplants from unrelated donors, as it was in the discovery study, but it was not present among the recipients of transplants from related donors. The reason for these differences is unknown. They may result from differences in the level of immunosuppression between recipients of hematopoietic-cell transplants from related donors and recipients of transplants from unrelated donors, or they may be explained by the relatively small size of the groups after stratification. In any case, these findings should be interpreted in the context of the limitations of subgroup analyses.44 Further studies will be needed to clarify this issue.
In conclusion, this study suggests an association between TLR4 haplotypes in unrelated donors and an increased risk of invasive aspergillosis among recipients of allogeneic hematopoietic-cell transplants. The identification of donors who have an increased risk of severe infection may have implications for prevention strategies in recipients of allogeneic hematopoietic-cell transplants.
Supported by grants from the Swiss Foundation for Medical and Biological Grants (1121), the Swiss National Science Foundation (81LA-65462), and the Leenaards Foundation (to Dr. Bochud); and the U.S. National Institutes of Health (CA18029, CA15704, AI054523, AI051468, AI33484, HL87690, HL088201, and HL69860). Some of the results of this study were obtained with the use of SAGE software, which is supported by a grant from the National Center for Research Resources (RR03655).
Dr. Marr reports receiving consulting fees from Astellas, Basilea, Enzon, Merck, Pfizer, and Schering-Plough, being a member of a data and safety monitoring board for a trial of antifungal prophylaxis for Schering-Plough, and receiving research support from Astellas, Pfizer, Enzon, and Merck; Dr. Boeckh, being a member of a data and safety monitoring board for a human immunodeficiency virus trial for Pfizer, receiving research support from Pfizer and Merck, and receiving consulting fees from Enzon; and Drs. Bochud, Boeckh, Chien, and Marr, having a pending application for a patent on TLR4 SNPs and haplotypes and the risk of aspergillosis. No other potential conflict of interest relevant to this article was reported.
We thank Jeremy Smith, Ravinder K. Sandhu, Craig Silva, Christopher Davis, Peter Choe, and Sanam Hussain for assistance in data collection and handling; Jeffrey Stevens, Gary J. Olsem, Tera R. Matson, Terry Stevens-Ayers, Jennifer Gravley, and Kimberly Harrington for technical assistance; and Véronique Erard for critical review of an earlier version of the manuscript.
Source Information
From the Institute for Systems Biology (P.-Y.B., M.J., S.D.R., S.L., A.A.); the Program in Infectious Diseases (P.-Y.B., K.A.M., W.M.L., A.U., M.B.), the Clinical Research Division (J.A.H.), and Quantitative Genetic Epidemiology, Division of Public Health Sciences (L.P.Z.), Fred Hutchinson Cancer Research Center; and the Department of Medicine, University of Washington (J.W.C., K.A.M., W.M.L., J.A.H., M.B.) — all in Seattle.
Address reprint requests to Dr. Boeckh at the Vaccine and Infectious Disease Institute, Program in Infectious Diseases, Fred Hutchinson Cancer Research Center, 1100 Fairview Ave. N., Seattle, WA 98109, or at mboeckh{at}fhcrc.org.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||
Related Letters:
Toll-like Receptor 4 Polymorphisms and Aspergillosis
Levitz S. M., Shoham S., Cleary J. D., Cervera C., Moreno A., Lozano F., Asakura Y., Komatsu T., Bochud P.-Y., Marr K. A., Boeckh M.
Extract |
Full Text |
PDF
N Engl J Med 2009;
360:634-636, Feb 5, 2009.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
