Abstract
Cytotoxic T-Lymphocyte Antigen-4 (CTLA-4) plays an essential role in T cell homeostasis by restraining immune responses. AG and GG genotypes of donor CTLA-4 SNP rs4553808 in patients after unrelated donor (URD) hematopoietic stem cell transplants (HSCT) have been shown to be an independent predictor of inferior relapse free survival (RFS) and overall survival (OS) compared to the AA genotype in single center studies. We tested the hypothesis that SNP rs4553808 is associated with RFS, OS, non-relapse mortality (NRM) and the cumulative incidence of acute graft-versus-host disease (aGVHD) and chronic graft-versus-host disease (cGVHD) in adults with acute myeloid leukemia (AML) and advanced myelodysplastic syndrome (MDS) undergoing a first 8/8 or 7/8 HLA matched URD HSCT. Multivariable analysis adjusting for relevant donor and recipient characteristics showed no significant association between SNP rs4553808 and OS, RFS, NRM, and incidence of aGVHD and cGVHD. An exploratory analysis of other CTLA-4 SNPs as well as studying the interaction with ATG also demonstrated no significant associations. Our results indicate that CTLA-4 SNPs are not associated with HSCT outcomes.
Keywords: CTLA-4, SNPs, hematopoietic stem cell transplantation
INTRODUCTION
Hematopoietic stem cell transplant (HSCT) is curative therapy for many malignant hematologic diseases. However, the success of HSCT is often limited by disease relapse and graft-versus-host disease (GVHD). There are currently no established biologic predictors of transplant outcomes. Understanding how to achieve the optimal balance between a potent graft-versus-tumor (GVT) and minimizing the risk of graft-versus-host disease (GVHD) remains elusive. Both GVT and GVHD are dependent on maintaining T cell homeostasis, which requires regulation of an intricate system of immunologic checks and balances. One of the regulatory molecules that serve as an important immunologic checkpoint is Cytotoxic T-Lymphocyte Antigen-4 (CTLA-4), which primarily functions in inhibition of T-cell activation. CTLA-4 is a member of the immunoglobulin superfamily, and it, along with other regulatory molecules such as programmed death-1 (PD-1), play an important role in regulation of peripheral tolerance.1 Upon recognition of an antigen by the T-cell receptor (TCR), CTLA-4 competes with CD28 to bind with CD80/86. This signal in turn leads to inhibition of downstream T cell activation and subsequently downregulates the immune response.2 In clinical practice, this mechanism of action has been harnessed as immunotherapy exemplified by the drug ipilimumab, a CTLA-4 blocking monoclonal antibody that augments the immune response and improves overall survival (OS) in patients with metastatic melanoma.3
The impact of single nucleotide polymorphisms (SNPs) of CTLA-4 has previously been demonstrated in solid tumors Specifically, the G allele of rs4553808 has been associated with a positive response to ipilimumab in patients with metastatic melanoma.4 In HSCT studies, CTLA-4 SNPs have been associated with differences in RFS, OS, and GVHD but with discordant results from various investigators.5–9 Previously we have shown that donor SNP rs4553808 is an independent pre-transplant predictor of outcomes in unrelated donor HSCT, with patients receiving transplants from donors with AG or GG genotypes having inferior relapse free survival (RFS) and OS compared to the AA genotype.10 We hypothesized that these results could be validated in a larger, more homogenous cohort of patients and if validated could be basis for pre-transplantation donor CTLA-4 genotyping as a risk-stratification tool that would aid in prediction of relapse risk. Patients at high risk of relapse could be specifically targeted for peri-transplant immunomodulation or pre-emptive therapies to reduce the risk of relapse.
Additionally, the interaction of CTLA-4 SNPs with use of anti-thymocyte globulin (ATG) was also studied. In patients undergoing a matched unrelated donor (URD) HSCT, ATG is often used as part of GVHD prophylaxis. The efficacy of ATG in prophylaxis of GVHD is dependent on its ability to deplete T lymphocytes of the graft.11 We hypothesized that genetic variation in CTLA-4 may interact with the utilization of the immunomodulatory drug ATG as part of the conditioning regimen in URD transplants. If an interaction was observed, then that this could additionally facilitate personalization of GVHD prophylaxis.
STUDY DESIGN
This study was completed in collaboration with the Center for International Blood and Marrow Transplant Research (CIBMTR) and National Marrow Donor Program (NMDP). The CIBMTR is an association of more than 500 transplantation centers worldwide that collects statistical data and maintains longitudinal follow-up on consecutive allogeneic HSCTs. Data are submitted to the Statistical Center at the Medical College of Wisconsin (MCW), and quality is ensured with computerized checks, physician review of data, and onsite audits of centers. Observational data collection by CIBMTR was approved by the Institutional Review Board of the NMDP.
The primary objectives were to study the impact of genetic variation in donor CTLA-4 SNP rs4553808 on RFS (defined as survival without relapse or death), OS (defined as time to death), and non-relapse mortality [(NRM), defined as any death without evidence of disease relapse with relapse treated as a competing risk). All data were censored at date of last follow-up. Secondary endpoints included incidence of grades 2 to 4 acute GVHD (aGVHD), and presence of chronic GVHD (cGVHD). The use of ATG was considered in the analysis. An exploratory analysis of the other 9 tagSNPs (rs231775, rs231779, rs11571315, rs231777, rs3087243, rs16840252, rs231725, rs10197010, rs11571316) of CTLA-4 was also performed to investigate potential additional associations.
Patients over the age of 18 with acute myeloid leukemia (AML) and advanced myelodysplastic syndrome (MDS) in first or second complete remission (CR1, CR2) who underwent a first 8/8 or 7/8 HLA-matched unrelated donor allogeneic HSCT after any conditioning regimen between 2002–2007 were included in this study. HLA typing was verified using DNA-based techniques, as previously described.12 Any GVHD prophylaxis was permitted with the exception of ex vivo T-cell depletion or use of in vivo/in vitro alemtuzumab. Based on CIBMTR data, there were 1463 cases eligible for this study. A sample size calculation indicated 780 patients were required to achieve 80% power to detect a significance level (alpha) of 0.005 (increased stringency due to multiple testing) using a two-sided log-rank test. To study the interaction with ATG, patients who received ATG were oversampled with 50% of patients studied having received ATG.
SNP analysis was performed on samples provided by the NMDP Research Repository with whole genome amplification using the REPLI-g UltraFast Mini-Kit (QIAGEN, Chatsworth, CA) and verified by RNASE-P analysis on the Taqman 7200. Genotyping of amplified DNA was performed with a custom GoldenGate (Illumina, San Diego, CA) combined with VeraCode technology on the BeadXpress according to the manufacturer’s protocol (Illumina).
Each SNP was analyzed in multivariable models adjusting for clinically significant variables using the Cox’s proportional hazards model. Due to multiple testing, P-values of <0.005 were considered significant. Reported p-values were not adjusted for multiple comparisons. Probability of RFS and OS were calculated using the Kaplan-Meier estimator from time of transplant. The cumulative incidence of NRM was calculated with relapse considered a competing risk. The interaction with CTLA-4 SNPs and use of ATG was also tested with 50% of the patient population having received ATG as part of their conditioning regimen.
RESULTS
Table 1 lists patient-, transplant-, and donor-related variables for patients included in the study. The final analysis included 780 patients with a median follow-up of 63 months for survivors. The median age of recipients and donors was age 50 (range 18–74) 34 (range 18–61), respectively. Ninety percent (N=706) of recipients and 82% (N=639) of donors were of Caucasian descent. Eighty percent of patients had been diagnosed with AML (N=624) and 20% with advanced MDS (N=156). Conditioning was 61% myeloablative, 28% reduced intensity and 11% non-myeloablative, with 74% receiving peripheral blood grafts. Seventy-five percent were HLA-matched and 25% single locus mismatched. Tacrolimus with methotrexate or mycophenolate mofetil GVHD prophylaxis was given to 62%, cyclosporine-based prophylaxis to 15% while 23% received other regimens. By design, 50% of patients received ATG. Genotyping showed that 552 (71%) were AA, 208 (27%) were AG, and 19 (2%) were GG (AG, and particularly GG, were associated with worse outcomes in the prior study).
Table 1.
Patient-, transplant-, and donor- related clinical characteristics (N=780)
| Characteristic | N (%) |
|---|---|
| Recipient Age (years) | |
| Median | 50 |
| Range | 18–74 |
| Sex | |
| Male | 403 (52) |
| Female | 377 (48) |
| Recipient Race/Ethnicity | |
| Caucasian | 706 (90) |
| African American | 22 (3) |
| Hispanic | 29 (4) |
| Other/Multiple/Decline | 23 (3) |
| Disease | |
| AML | 624 (80) |
| MDS | 156 (20) |
| Conditioning | |
| Ablative | 474 (61) |
| Reduced intensity | 222 (28) |
| Non-myeloablative | 84 (11) |
| HLA match | |
| 8/8 Matched | 585 (75) |
| 7/8 Mismatch | 195 (25) |
| GVHD prophylaxis | |
| Tacrolimus + MTX or MMF | 480 (62) |
| Cyclosporine + MTX | 119 (15) |
| Other | 181 (23) |
| Stem cell source | |
| Peripheral blood | 576 (74) |
| Bone Marrow | 204 (26) |
| HLA Match | |
| 8/8 Matched | 513 (66) |
| 7/8 (1 MM at HLA-A, -B, -C, or -DRB1) | 267 (34) |
| Donor/Recipient Sex Match | |
| Male/Male | 276 (36) |
| Male/Female | 220 (29) |
| Female/Male | 117 (15) |
| Female/Female | 78 (21) |
| Donor Age | |
| Median | 34 |
| Range | 18–61 |
| <20 years | 9 (1) |
| 20–29 years | 242 (33) |
| 30–39 years | 269 (36) |
| 40–49 years | 174 (23) |
| 50 and older | 54 (7) |
| Donor Race/Ethnicity | |
| Caucasian | 639 (82) |
| African American | 25 (3) |
| Hispanic | 23 (4) |
| Other/Multiple/Decline | 93 (12) |
acute myeloid leukemia (AML); myelodysplastic syndrome (MDS); human leukocyte antigen (HLA); methotrexate (MTX); mycoohenolate mofetil (MMF); mismatch (MM)
Multivariable analysis adjusting for relevant donor and recipient characteristics showed no significant association between SNP rs4553808 and OS, RFS, NRM, and the cumulative incidence of acute, chronic GVHD, and relapse (Figure 1). Multivariable analysis of the other 9 tagSNPs of the CTLA-4 gene showed no association with primary transplant outcomes (all P values >0.005). No interaction was found between CTLA-4 SNPs and use of ATG on outcomes.
Figure 1. Clinical outcomes stratified by donor CTLA-4 SNP rs4553808 genotype.
(A) adjusted cumulative incidence of grades 2–4 aGVHD, (B) adjusted probability of overall survival, (C) adjusted cumulative incidence of non-relapse mortality, (D) adjusted probability of relapse free survival, (E) adjusted cumulative incidence of relapse.
DISCUSSION
Understanding the intricate balance of T cell homeostasis in the post-transplant period remains an important area of investigation. CTLA-4, an important immunologic checkpoint, and its genetic variations have been of interest in the HSCT realm because of its possible important role in immune modulation, especially as it relates to GVHD and the potency of the GVT effect. CTLA-4 SNPs have previously been shown to be associated with RFS, OS, and incidence of GVHD 5–9 as well as a possible surrogate marker for response to donor lymphocyte infusion after HSCT.13 This raised the possibility that a personalized approach to HSCT donor selection and conditioning regimens based on genetic variation could be realized. However, previous studies that have investigated CTLA-4 SNPs have produced discordant results, likely due to small sample sizes and heterogeneous cohorts that in turn make interpretation of the true interaction of CTLA-4 SNPS with transplant outcomes difficult. The negative result from this present study highlights this limitation and underscores the importance of validation of preliminary genomic association studies in larger, more homogenous cohorts.14
The strength of our study is indeed the homogeneity of our patient population with regards to the disease and transplant type as well as its large sample size, particularly compared to previous similar studies. To our knowledge, this study is the largest and most homogenous transplant cohort investigating genetic variation in CTLA-4, although is should be noted that there were differences in other variables such as age, degree of HLA match, conditioning regimens, and post-transplant GVHD prophylaxis. Additionally, this large sample size allowed us to perform an exploratory analysis of other CTLA-4 SNPs as well as study the interaction with ATG. Our methods included increased stringency for statistical significance due to multiple testing, lending further credence to the interpretation of our results and final conclusion.
Limitations of this study include that only donor genetic variation of CTLA-4 was studied. While theoretically, donor genetics likely plays a more important role in post-transplant T cell homeostasis, there certainly could be an interaction with recipient genetic variations that were not accounted for. Secondly, because our study included only patients with AML/MDS in the unrelated donor setting, the application of our conclusion is restricted to such patients. It is possible that other disease and transplant types may produce different results. Haplotypes of the CTLA-4 gene were not studied as in some previous studies, and thus our conclusions are limited only to the single SNP level with regards to genetic variation. Finally, it is also possible that if interaction is present, the impact of a single polymorphism may simply be too small to influence such expansive outcomes as NRM, RFS, and OS.
In conclusion, the observation from our previous pilot study that SNP rs3554808 was an independent predictor of transplant outcomes was not confirmed in this larger study. Moreover, no other CTLA-4 SNPs in our exploratory analysis were identified to be independent predictors of outcomes. Data from this study support the conclusion that donor CTLA-4 SNPs are not associated with unrelated donor HSCT outcomes in patients with AML and MDS. Though the results of this study are negative, it should be considered definitive in such patient populations. Future studies could include investigation of genetic variation of CTLA-4 in other disease and transplant types as well as other immune checkpoints such as PD-1, especially as clinical therapeutics that target PD-1 evolve.
Supplementary Material
Supplemental Figure 1. Diagram of the CTLA-4 gene with identification of previously studied SNPs, currently studied SNPs, and exons.
Acknowledgments
The authors acknowledge Cara Sutcliffe, MS Center for Human Genetics Research and Heidi Chen, PhD, Department of Biostatistics for their assistance with this study.
Footnotes
Authorship: All authors listed participated in the design of the study and the writing of this manuscript.
Declaration of commercial interest: This study was supported by Sanofi-Oncology.
Financial Disclosure: This study was supported in part by research funding from Sanofi-Oncology to S.S. and M. J.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Salama AK, Hodi FS. Cytotoxic T-lymphocyte-associated antigen-4. Clinical Cancer Research. 2011;17(14):4622–8. doi: 10.1158/1078-0432.CCR-10-2232. [DOI] [PubMed] [Google Scholar]
- 2.Teft WA, Kirchhof MG, Madrenas J. A molecular perspective of CTLA-4 function. Annual Review of Immunology. 2006;24:65–97. doi: 10.1146/annurev.immunol.24.021605.090535. [DOI] [PubMed] [Google Scholar]
- 3.Hodi FS, O’Day SJ, McDermott DF, Weber RW, Sosman JA, Haanen JB, et al. Improved survival with ipilimumab in patients with metastatic melanoma. NEJM. 2010;363(8):711–23. doi: 10.1056/NEJMoa1003466. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Breunis WB, Tarazona-Santos E, Chen R, Kiley M, Rosenberg SA, Chanock SJ. Influence of cytotoxic T lymphocyte-associated antigen 4 (CTLA4) common polymorphisms on outcome in treatment of melanoma patients with CTLA-4 blockade. J Immunother. 2008;31(6):586–90. doi: 10.1097/CJI.0b013e31817fd8f3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Sellami MH, Bani M, Torjemane L, Kaabi H, Ladeb S, Ben Othmane T, et al. Effect of donor CTLA-4 alleles and haplotypes on graft-versus-host disease occurrence in Tunisian patients receiving a human leukocyte antigen-identical sibling hematopoietic stem cell transplant. Human Immunology. 2011;72(2):139–43. doi: 10.1016/j.humimm.2010.11.008. [DOI] [PubMed] [Google Scholar]
- 6.Bosch-Vizcaya A, Perez-Garcia A, Brunet S, Solano C, Buno I, Guillem V, et al. Donor CTLA-4 Genotype Influences Clinical Outcome after T Cell-Depleted Allogeneic Hematopoietic Stem Cell Transplantation from HLA-Identical Sibling Donors. Biol Blood Marrow Transplant. 2011;18(1):100–5. doi: 10.1016/j.bbmt.2011.05.021. [DOI] [PubMed] [Google Scholar]
- 7.Vannucchi AM, Guidi S, Guglielmelli P, Glinz S, Lombardini L, Busca A, et al. Significance of CTLA-4 and CD14 genetic polymorphisms in clinical outcome after allogeneic stem cell transplantation. Bone Marrow Transplant. 2007;40(10):1001–2. doi: 10.1038/sj.bmt.1705850. [DOI] [PubMed] [Google Scholar]
- 8.Perez-Garcia A, De la Camara R, Roman-Gomez J, Jimenez-Velasco A, Encuentra M, Nieto JB, et al. CTLA-4 polymorphisms and clinical outcome after allogeneic stem cell transplantation from HLA-identical sibling donors. Blood. 2007;110(1):461–7. doi: 10.1182/blood-2007-01-069781. [DOI] [PubMed] [Google Scholar]
- 9.Azarian M, Busson M, Lepage V, Charron D, Toubert A, Loiseau P, et al. Donor CTLA-4 +49 A/G*GG genotype is associated with chronic GVHD after HLA-identical haematopoietic stem-cell transplantations. Blood. 2007;110(13):4623–4. doi: 10.1182/blood-2007-08-106385. [DOI] [PubMed] [Google Scholar]
- 10.Jagasia M, Clark WB, Brown-Gentry KD, Crawford DC, Fan KH, Chen H, et al. Genetic variation in donor CTLA-4 regulatory region is a strong predictor of outcome after allogeneic hematopoietic cell transplantation for hematologic malignancies. Biol Blood Marrow Transplant. 2012;18(7):1069–75. doi: 10.1016/j.bbmt.2011.12.518. [DOI] [PubMed] [Google Scholar]
- 11.Mohty M. Mechanisms of action of antithymocyte globulin: T-cell depletion and beyond. Leukemia. 2007;21(7):1387–94. doi: 10.1038/sj.leu.2404683. [DOI] [PubMed] [Google Scholar]
- 12.Spellman S, Setterholm M, Maiers M, Noreen H, Oudshoorn M, Fernandez-Vina M, et al. Advances in the selection of HLA-compatible donors: refinements in HLA typing and matching over the first 20 years of the National Marrow Donor Program Registry. Biology Blood Marrow Transplant. 2008;14(9 Suppl):37–44. doi: 10.1016/j.bbmt.2008.05.001. [DOI] [PubMed] [Google Scholar]
- 13.Metaxas Y, Bertz H, Spyridonidis A, Spyroupoulou-Vlachou M, Porzelius C, Finke J. CT60 single-nucleotide polymorphism as a surrogate marker for donor lymphocyte infusion outcome after allogeneic cell transplantation for acute leukemia. Bone Marrow Transplant. 2012;47(3):411–5. doi: 10.1038/bmt.2011.101. [DOI] [PubMed] [Google Scholar]
- 14.Ting C, Alterovitz G, Merlob A, Abdi R. Genomic studies of GVHD-lessons learned thus far. Bone Marrow Transplant. 2013;48(1):4–9. doi: 10.1038/bmt.2012.9. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Figure 1. Diagram of the CTLA-4 gene with identification of previously studied SNPs, currently studied SNPs, and exons.