Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Inherited Variants in Regulatory T Cell Genes and Outcome of Ovarian Cancer

  • Ellen L. Goode ,

    Affiliation Department of Health Sciences Research, Mayo Clinic, Rochester, Minnesota, United States of America

  • Melissa DeRycke,

    Affiliation Department of Health Sciences Research, Mayo Clinic, Rochester, Minnesota, United States of America

  • Kimberly R. Kalli,

    Affiliation Department of Medical Oncology, Mayo Clinic, Rochester, Minnesota, United States of America

  • Ann L. Oberg,

    Affiliation Department of Health Sciences Research, Mayo Clinic, Rochester, Minnesota, United States of America

  • Julie M. Cunningham,

    Affiliation Department of Experimental Pathology, Mayo Clinic, Rochester, Minnesota, United States of America

  • Matthew J. Maurer,

    Affiliation Department of Health Sciences Research, Mayo Clinic, Rochester, Minnesota, United States of America

  • Brooke L. Fridley,

    Affiliation Department of Health Sciences Research, Mayo Clinic, Rochester, Minnesota, United States of America

  • Sebastian M. Armasu,

    Affiliation Department of Health Sciences Research, Mayo Clinic, Rochester, Minnesota, United States of America

  • Daniel J. Serie,

    Affiliation Department of Health Sciences Research, Mayo Clinic, Rochester, Minnesota, United States of America

  • Priya Ramar,

    Affiliation Department of Health Sciences Research, Mayo Clinic, Rochester, Minnesota, United States of America

  • Krista Goergen,

    Affiliation Department of Health Sciences Research, Mayo Clinic, Rochester, Minnesota, United States of America

  • Robert A. Vierkant,

    Affiliation Department of Health Sciences Research, Mayo Clinic, Rochester, Minnesota, United States of America

  • David N. Rider,

    Affiliation Department of Health Sciences Research, Mayo Clinic, Rochester, Minnesota, United States of America

  • Hugues Sicotte,

    Affiliation Department of Health Sciences Research, Mayo Clinic, Rochester, Minnesota, United States of America

  • Chen Wang,

    Affiliation Department of Health Sciences Research, Mayo Clinic, Rochester, Minnesota, United States of America

  • Boris Winterhoff,

    Affiliation Department of Gynecologic Oncology, Mayo Clinic, Rochester, Minnesota, United States of America

  • Catherine M. Phelan,

    Affiliation H. Lee Moffitt Cancer Center & Research Institute, Tampa, Florida, United States of America

  • Joellen M. Schildkraut,

    Affiliation Duke Comprehensive Cancer Center, Duke University, Durham, North Carolina, United States of America

  • Rachel P. Weber,

    Affiliation Department of Community and Family Medicine, Duke University Medical Center, Durham, North Carolina, United States of America

  • Ed Iversen,

    Affiliation Department of Statistical Science, Duke University, Durham, North Carolina, United States of America

  • Andrew Berchuck,

    Affiliation Duke Comprehensive Cancer Center, Duke University, Durham, North Carolina, United States of America

  • Rebecca Sutphen,

    Affiliation University of South Florida College of Medicine, Tampa, Florida, United States of America

  • Michael J. Birrer,

    Affiliation Brigham and Women’s Hospital, Boston, Massachusetts, United States of America

  • Shalaka Hampras,

    Affiliation Department of Cancer Prevention and Control, Roswell Park Cancer Institute, Buffalo, New York, United States of America

  • Leah Preus,

    Affiliation Department of Cancer Prevention and Control, Roswell Park Cancer Institute, Buffalo, New York, United States of America

  • Simon A. Gayther,

    Affiliation University of Southern California, Los Angeles, California, United States of America

  • Susan J. Ramus,

    Affiliation University of Southern California, Los Angeles, California, United States of America

  • Nicolas Wentzensen,

    Affiliation National Cancer Institute, Bethesda, Maryland, United States of America

  • Hannah P. Yang,

    Affiliation National Cancer Institute, Bethesda, Maryland, United States of America

  • Montserrat Garcia-Closas,

    Affiliation Institute of Cancer Research, Sutton, United Kingdom

  • Honglin Song,

    Affiliation University of Cambridge, Cambridge, United Kingdom

  • Jonathan Tyrer,

    Affiliation University of Cambridge, Cambridge, United Kingdom

  • Paul P. D. Pharoah,

    Affiliation University of Cambridge, Cambridge, United Kingdom

  • Gottfried Konecny,

    Affiliation Department of Gynecologic Oncology, University of California, Los Angeles, California, United States of America

  • Thomas A. Sellers,

    Affiliation H. Lee Moffitt Cancer Center & Research Institute, Tampa, Florida, United States of America

  • Roberta B. Ness,

    Affiliation School of Public Health, University of Texas, Houston, Texas, United States of America

  • Lara E. Sucheston,

    Affiliation Department of Cancer Prevention and Control, Roswell Park Cancer Institute, Buffalo, New York, United States of America

  • Kunle Odunsi,

    Affiliation Department of Gynecologic Oncology, Roswell Park Cancer Institute, Buffalo, New York, United States of America

  • Lynn C. Hartmann,

    Affiliation Department of Medical Oncology, Mayo Clinic, Rochester, Minnesota, United States of America

  • Kirsten B. Moysich,

    Affiliation Department of Cancer Prevention and Control, Roswell Park Cancer Institute, Buffalo, New York, United States of America

  •  [ ... ],
  • Keith L. Knutson

    Affiliation Department of Immunology, Mayo Clinic, Rochester, Minnesota, United States of America

  • [ view all ]
  • [ view less ]

Inherited Variants in Regulatory T Cell Genes and Outcome of Ovarian Cancer

  • Ellen L. Goode, 
  • Melissa DeRycke, 
  • Kimberly R. Kalli, 
  • Ann L. Oberg, 
  • Julie M. Cunningham, 
  • Matthew J. Maurer, 
  • Brooke L. Fridley, 
  • Sebastian M. Armasu, 
  • Daniel J. Serie, 
  • Priya Ramar


Although ovarian cancer is the most lethal of gynecologic malignancies, wide variation in outcome following conventional therapy continues to exist. The presence of tumor-infiltrating regulatory T cells (Tregs) has a role in outcome of this disease, and a growing body of data supports the existence of inherited prognostic factors. However, the role of inherited variants in genes encoding Treg-related immune molecules has not been fully explored. We analyzed expression quantitative trait loci (eQTL) and sequence-based tagging single nucleotide polymorphisms (tagSNPs) for 54 genes associated with Tregs in 3,662 invasive ovarian cancer cases. With adjustment for known prognostic factors, suggestive results were observed among rarer histological subtypes; poorer survival was associated with minor alleles at SNPs in RGS1 (clear cell, rs10921202, p = 2.7×10−5), LRRC32 and TNFRSF18/TNFRSF4 (mucinous, rs3781699, p = 4.5×10−4, and rs3753348, p = 9.0×10−4, respectively), and CD80 (endometrioid, rs13071247, p = 8.0×10−4). Fo0r the latter, correlative data support a CD80 rs13071247 genotype association with CD80 tumor RNA expression (p = 0.006). An additional eQTL SNP in CD80 was associated with shorter survival (rs7804190, p = 8.1×10−4) among all cases combined. As the products of these genes are known to affect induction, trafficking, or immunosuppressive function of Tregs, these results suggest the need for follow-up phenotypic studies.


Ovarian cancer is the fifth leading cause of cancer death among women in the United States [1]. Five-year overall survival is approximately 45%, and, even with modern surgical and chemotherapeutic strategies, most cases with advanced disease relapse and succumb to the disease [2], [3]. Rare germline BRCA1 or BRCA2 mutations confer improved survival [4]. Common inherited variants could also influence outcome; genome-wide association studies (GWAS) are underway, but have yet to find survival-associated loci [5]. Consideration of novel biological pathways using in-depth analysis of variation in candidate genes holds promise for the identification of prognostic genetic factors.

Several studies demonstrate the importance of the immune system in ovarian cancer outcome. For example, in one report, cases with evidence of CD3 T cell tumor infiltration (approximately one-half of the cases studied) showed improved progression-free and overall survival [6]. Subsequent studies have refined our understanding of tumor-infiltrating T cells, including one showing that CD8+ T lymphocytes are the primary sub-population of T cells associated with better survival [7]. Along with the finding that tumor antigen-specific T cell responses can be detected in cases, these results suggest that anti-tumor immunity is elicited against ovarian cancers and impacts the clinical course of the disease [8]. Despite this generation of an immune response, however, anti-tumor immunity is counterbalanced by an immune suppressive microenvironment [9]. Of immune suppressive mechanisms, CD4+ regulatory T cells (Tregs) are a primary means of immune evasion in ovarian cancers; these are CD4+ T lineage cells whose primary function is immune regulation [10]. In collaboration with others, we first suggested a role of tumor-infiltrating Tregs in ovarian cancer pathogenesis, reporting higher levels of CD4+ Tregs, measured with immunofluorescence, among cases with poorer survival [11]. Subsequent work supports the importance of CD4+ Tregs in ovarian cancer pathogenesis and outcome [7], [12]. For example, the presence of CD4+ Tregs appears to influence the anti-tumor activity of tumor-infiltrating cytotoxic CD8+ T cells [7]. CD4+ Tregs block both adaptive and innate immune effectors by cell contact mechanisms as well as by soluble mediators [13], [14]. Soluble mediators of suppression commonly associated with CD4+ Tregs include IL-10 and TGF-β, both of which block T cell proliferation and cell-mediated immunity [15], [16], [17]. Other cell surface molecules implicated in suppressing the immune response include B7-H1 (CD274), GITR (TNFRSF18), LAG-3, CTLA-4, and surface-bound TGF-β [18], [19], [20].

Because of the importance of CD4+ Tregs and a role for inherited factors in outcome, we assessed whether common inherited variation related to CD4+ Treg-related genes was associated with ovarian cancer outcome following standard of care therapy. Specifically, we assessed 54 genes key to the induction, trafficking, or immunosuppressive functions of CD4+ Tregs. Utilizing data from a novel candidate gene study combined with existing data, we studied polymorphisms that have been shown to affect the expression of or to tag inherited variation in these genes. Variants associated with outcome in multiple study populations could shed light on immunosuppressive mechanisms in ovarian cancer.

Materials and Methods

Ethics Statement

Protocols were approved by the appropriate institutional review boards (Mayo Clinic Institutional Review Board, Roswell Park Cancer Institute Institutional Review Boards, Duke University Institutional Review Boards, National Cancer Institute Institutional Review Boards, Moffitt Cancer Center Institutional Review Boards) or ethics panel (Royal Marsden Hospital ethics panel, University of Cambridge ethics panel, University College London ethics panel). All participants gave written informed consent.

Candidate Gene SNP Array

Fifty-four genes of key relevance to the biology of Tregs (ACVR2B, AGTR1, CCL11, CCL17, CCL19, CCL2, CCL20, CCL22, CCL3, CCL4, CCL5, CCR4, CCR6, CCR7, CCR8, CD274, CD46, CD80, CD86, CXCL10, CXCL13, CXCR5, DUSP4, EGR2, GPR83, IDO1, IKZF2, IKZF4, IL10RB, IL15RA, IL2RB, IL6ST, IL9, INHBA, INHBB, IRF4, ITGAE, KLF10, LAG3, LRRC32, MDFIC, NRP1, PDCD1, PLAG1, PRNP, RGS1, RGS16, SH3BGRL2, SLC22A2, SMAD3, SOCS2, TNFRSF18, TNFRSF4, and TNFRSF9) were chosen for study (Table S1). The relevance of these genes was established from a PubMed [21] database search which revealed published information that either directly showed or suggested a role for the respective gene products in the induction, immune suppressive function, or trafficking of Tregs. Eighty-five SNPs associated (p<10−6) with lymphocyte mRNA expression of one or more candidate genes (eQTL SNPs) were included [22]. In addition, within five kb of each gene, SNPs tagging other SNPs with minor allele frequency (MAF) ≥0.05 at r2≥0.9 were identified from 60 European-American participants in the Low-Coverage Pilot of the 1000 Genomes Project (for 53 genes) [23] or the SeattleSNPs Variation Discovery Resource (for CCL2) [24], whichever was most informative. Selected tagSNPs (N = 1,451) were optimized to include SNPs with appropriate design scores for an Illumina Goldengate BeadArray Assay, predicted functionality, and to include more than one tagSNP for large groups of correlated SNPs [25]. Additional SNP information is in Table S2.

Candidate Gene Study Participants and Genotyping

Cases genotyped on the Treg custom SNP array included women with pathologically-confirmed invasive primary epithelial ovarian, peritoneal, or fallopian tube cancer enrolled at the Mayo Clinic and Roswell Park Cancer Institute (RPCI). Mayo Clinic cases (N = 905) were ascertained between December 1999 and November 2010 into the Mayo Clinic Ovarian Cancer Case-Control Study (MAY) or the Mayo Clinic Case-Only Ovarian Cancer Study (MAC) and included women aged 20 years or above enrolled through Mayo Clinic’s Divisions of Gynecologic Surgery and Medical Oncology. Sixty eight percent of these cases were enrolled within a week of diagnosis (median time from diagnosis to recruitment was zero days). RPCI cases (N = 167) were residents of Western Pennsylvania, Eastern Ohio, or Western New York, aged 25 years or above and ascertained between January 2004 and May 2009 within six months of diagnosis through Roswell Park Cancer Institute’s Divisions of Gynecological Surgery and Oncology. These cases were enrolled with a median time from diagnosis to recruitment of 80 days. At both sites, DNA was extracted from 10 to 15 mL fresh peripheral blood (using Gentra AutoPure LS Purgene salting out methodology at Mayo Clinic and FlexiGene DNA Kit methodology at RPCI), stored at −80°C, and bar-coded to ensure accurate processing. A total of 1,072 participants were genotyped using a custom Illumina Goldengate BeadArray Assay along with 24 duplicates and 24 HapMap CEU replicates (8 trios). Concordance among study duplicates was 99.998%, no SNPs had unresolved replicate or Mendelian errors, and the mean genotype call rate was 99.88%. Samples with genotyping failure (N = 22) or call rate <97% (N = 11) were excluded as well as samples found to be incorrectly plated (N = 1) or from cases deemed ineligible due to non-epithelial disease (N = 5), borderline tumor behavior (N = 34), or enrollment more than one year prior to diagnosis (N = 5). SNPs with genotyping failure (N = 162), call rate <95% (N = 19), or HWE <0.0001 and poor clustering (N = 15) were excluded. Thus, analyses were based on 994 cases and 1,340 SNPs.

Genome-Wide Association Study

We also analyzed data from the Follow-up of Ovarian Cancer Genetic Association and Interaction Studies (FOCI) collaboration which is part of the National Cancer Institute GAME-ON Post-GWAS initiative and is described elsewhere [26], [27]. In brief, participants were 2,167 self-reported white invasive ovarian cancer cases enrolled in the North Carolina Ovarian Cancer Study (NCO), the NCI Ovarian Case-Control Study in Poland (POL), the Royal Marsden Cancer Study (RMH), the UK Studies of Epidemiology and Risk Factors in Cancer Heredity Ovarian Cancer Study (SEA), the Tampa Bay Ovarian Cancer Study (TBO), and the UK Familial Ovarian Cancer Registry and the UK Ovarian Cancer Population Study (UKR+UKO). Genotyping used Illumina 317 k or 610-Quad Infinium Arrays with imputation to HapMap v 26 using dosage values obtained from the MACH software package [28]. Data on 820 SNPs were available on all cases.

Statistical Analysis

We used Cox proportional hazards regression accounting for left truncation to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for association with overall survival. Survival time was defined as time from diagnosis of ovarian cancer until death from any cause or last follow-up. For each SNP, HRs with 95% CIs were estimated per-allele (i.e., 0, 1, or 2 copies of minor allele), analogous to the Armitage test for trend for binary endpoints. Log-additive Cox proportional hazards regression models were adjusted for study site (MAY+MAC, RPCI, POL, NCO, RMH, SEA, TBO, UKO+UKR), age at diagnosis (<50 years, 50–69 years, >70 years), tumor stage (I or II, III or IV, unknown), tumor grade (low, high, unknown), and race (white, non-white, unknown), modeling direct genotype calls for MAY+MAC and RPCI and imputed allele dosage values where appropriate for FOCI participants. Heterogeneity of HRs across study site was formally examined using study-by-SNP interaction terms and performing likelihood ratio tests; no correction for multiple testing was performed.

Tumor Expression Quantitative Trait Locus (eQTL) Analysis

For 54 genotyped Mayo Clinic cases (33 serous, nine clear cell, eight endometrioid, four mucinous), expression analyses were also performed. Tumor RNA was isolated from fresh frozen samples, using the Qiagen RNEasy protocol and quantitated using a Nanodrop Spectrophotomer (Agilent Technologies, Santa Clara, CA). Total RNA (750 ng) of high quality (RNA integrated number >8.0) was labeled with cyanine 5-CTP or cyanine 3-CTP, using the Low RNA Input Fluorescent Linear Amplification Kit (Agilent Technologies), purified on RNeasy Mini columns (Qiagen), and hybridized to Agilent whole human genome 4×44 K expression arrays (using a mixed reference containing 106 tumor samples). Slides were scanned using the Agilent 2565BA Scanner, and data were normalized using Agilent’s error model and exported by the Agilent Feature Extraction Software (version 7.5.1). Data in the form of the log ratios of signals from individual tumors to signals from the reference mix were used for analysis. For genes with SNPs that were associated with survival at p<0.001, association between genotype and expression probes was assessed using Wilcoxon rank-sum tests.


The nearly 1,000 invasive ovarian cancer cases genotyped in a Treg custom SNP array and approximately 2,600 cases in the FOCI collaboration demonstrated the expected distributions of mortality, age, and clinical features (Table 1). 1,529 deaths were observed during a median follow-up of 5.4 years. In combined analyses of all cases and case groups defined by histology, eight SNPs yielded p<0.001 including six independent at r2<0.95 (Table 2). Results were similar when restricted to 2,518 cases with complete data on stage, grade, and histology. At each SNP, minor alleles were associated with poorer ovarian cancer survival. The most statistically significant association was between survival following clear cell ovarian cancer (N = 217) and minor alleles at an uncommon intronic SNP in the regulator of G-protein signaling 1 (RGS1), suggesting an almost three-fold increased risk of death (p = 2.7×10−5).

Table 1. Distributions of Ovarian Cancer Clinical Characteristics by Study.

Table 2. Regulatory T Cell SNPs Associated with Overall Survival (p<0.001).

Among 272 mucinous ovarian cancers, minor alleles at four SNPs in leucine rich repeat containing 32 (LRRC32) were associated with greater than two-fold poorer survival. Three of these SNPs (rs3781699, rs3197153, rs3781701) were highly correlated (r2≥0.97), thus only one of these is presented in Table 2 (rs3781669 p = 4.5×10−4). A fourth SNP, LRRC32 SNP rs7944357, was only modestly correlated with this three-SNP cluster (r2≤0.26), yet was also associated with survival among cases with mucinous ovarian cancer (p = 8.3×10−4). One other SNP rs3753348 showed association with a more than three-fold risk with survival in mucinous ovarian cancer (Table 2); this SNP resides in the 5 kb region between tumor necrosis factor receptor superfamily members 4 and 18 (TNFRSF4 and TNFRSF18) on chromosome 1.

Among 570 endometrioid ovarian cancer cases, minor alleles at an intronic CD80 tagSNP rs13071247 were associated with a 73% increased risk of mortality (p = 8.0×10−4, Table 2). In addition, analysis of all cases regardless of histology suggested that an eQTL SNP for CD80 (rs7804190) was associated with 14% shorter survival time (p = 8.0×10−4). The rs7804190 eQTL, which resides intronic to MAD1 mitotic arrest deficient-like 1 (yeast) (MAD1L1), is of interest because genotype has been found to correlate with expression of CD80 (p = 6.0×10−7), as well as polymerase (DNA directed), beta (POLB), proline-serine-threonine phosphatase interacting protein 2 (PSTPIP2), and KIAA1128 (p<10−6) in lymphoblastoid cell lines [22]. Of note, no SNPs were associated with survival following serous ovarian cancer (p<0.001), the most common and most lethal histologic subtype. Tests for interaction revealed no site-specific heterogeneity of association for any of the SNPs in Table 2 (p>0.05 for each).

To explore potential mechanisms of the observed SNPs associated with ovarian cancer survival, we examined tumor RNA expression data on 54 Mayo Clinic cases and correlated expression with genotype at the most suggestive survival-associated SNPs described above. At CD80 rs13071247, an association was observed such that heterozygotes and minor allele homozygotes (AC and CC genotypes) had slightly increased tumor expression (median fold change  = 1.04 on the raw scale, 0.06 on the log2 scale) than cases with AA genotype (p = 0.006; Figure 1). No other associations between genotype and tumor RNA expression were observed at p<0.05.

Figure 1. CD80 rs13071247 Genotype and Tumor Expression.

Among N = 54 MAY+MAC invasive ovarian cancer cases, Agilent whole human genome 4×44 K expression array probe A_24_P155632; CD80 log2 ratios of tumor versus reference RNA expression (y-axis) versus genotype (x-axis); dashes indicate median; each symbol represents a unique patient; data points are jittered so all data are visible.


Previous studies support an important role for Tregs in ovarian cancer which appear to foster an immune suppressive microenvironment. Here, we analyzed ovarian cancer survival in relation to inherited Treg genotypes using a combination of customized genotyping and integration of existing genotypes from a collaborating GWAS. While no SNPs yielded suggestive results among the more common serous cases, poorer survival was associated with minor alleles at SNPs in RGS1 (clear cell), LRRC32 and TNFRSF18/TNFRSF4 (mucinous), and CD80 (endometrioid). For the latter, additional data support a CD80 genotype association with CD80 tumor RNA expression. Combined analysis of multiple independent datasets provides greater statistical power than separate discovery and replication analyses [29]; thus, we used this study design and examined the possibility of heterogeneity of results across studies. As no statistically significant heterogeneity across studies was observed, our inference is based on this combined, most powerful approach. It is worth noting that we examined a relatively large number of SNPs across 54 Treg-associated genes and across different histologic subtypes, and therefore we acknowledge that multiple testing issues may exist; some of our highlighted results could indeed be false positive associations. The fact that the genes examined were chosen based on an a priori role in ovarian cancer and that we concentrated only on SNP associations with p-values less than 0.001 lessens, but does not entirely eliminate, this possibility. Nonetheless, this work highlights particular Treg-related genes of interest.

CD80 has been extensively studied for its role in immune responses, yet no focused analysis of SNPs and ovarian cancer outcome has been reported, to our knowledge. It acts as a ligand for both CD28 and CTLA4, leading to proliferation and anergy in naïve T cells, respectively [30], [31], [32]. CTLA4 is constitutively expressed in Tregs, where it is important in suppressing immune responses through a variety of proposed mechanisms, including activation of the indoleamine-2,3-dioxygenase pathway in dendritic cells (DCs) and inhibition of interactions between activated T cells and DCs [33], [34], [35]. We observed that an intronic tagging CD80 SNP was associated with poorer survival of endometrioid cases and with increased tumor CD80 expression, and, among all cases, we observed that an eQTL SNP on another chromosome which associated with CD80 lymphocyte expression was also associated with poorer survival. Altogether, these data suggest that CD80 expression may be in part driven by inherited factors and may lead to increased immune suppression and poorer outcome.

Of interest to clear cell ovarian cancer, the gene product of RGS1, RGS1 or BL34, is a member of the RGS protein family whose members are involved in regulation of G-protein signaling. Specifically, they are GTPase-activating proteins that limit the duration of G-protein signaling [36], [37]. Lymphocyte migration to chemotactic signals are mediated by G-protein signals and previous studies have found that Tregs do not respond as well to chemokine signaling as naïve T cells [38]). Furthermore, RGS1 is more highly expressed in Tregs and expression is inversely correlated with migration [38]. Interestingly, RGS1 gene expression is increased in activated Tregs, and this is mediated by binding of the Treg transcription factor FOXP3 to the RGS1 gene [39]. Associations of RGS1 SNPs have also been observed in numerous T cell-mediated autoimmune diseases, including type 1 diabetes, celiac disease, and multiple sclerosis [40], [41], [42], [43]. Based on these prior studies, it is tempting to speculate that genetically determined RGS1 levels may regulate Treg infiltration into clear cell ovarian cancers and thus contribute to outcome.

A SNP of possible relevance to mucinous disease is in LRRC32 which encodes GARP, a transmembrane protein expressed specifically on naturally occurring activated Tregs, but not resting Tregs [44], [45]. GARP has been shown to be a receptor for inactive, latency-associated peptide (LAP) bound TGF-β [46], [47]. GARP does not induce activation of latent TGF-β; however, it may function in infectious tolerance, converting FOXP3 cells to suppressive FOXP3+ cells [46], [48]. Additionally, GARP is part of a positive feedback loop with FOXP3 in Tregs, which are known to maintain a suppressive tumor microenvironment and prevent an effective immunological response [45]. Our finding of poorer survival among mucinous cases with minor alleles at an LRRC32 SNP suggests that these cases may have increased immune suppression.

We report an association between risk of death due to mucinous ovarian cancer and a SNP in a gene cluster containing TNFRSF18 and TNFRSF4. TNFRSF18 encodes GITR, a co-stimulatory molecule present constitutively on Treg cells and upregulated on naïve T cells after stimulation. Reports are conflicting as to the role of GITR, as it has been reported both to increase [49], [50] and to abolish the suppressive function of Tregs [51], [52]. While the specific mechanism of GITR action remains controversial, it is clear that it modulates Tregs. TNFRSF4 encodes OX40 (CD134), and signaling through OX40 reduces the ability of Tregs to act as suppressor cells by decreasing expression of FOXP3 [53], [54]. Reduced FOXP3 results in decreased miR155 and a subsequent increase in SOCS1 (suppressor of cytokine signaling 1). SOCS1 is a component of a negative feedback loop for IL-2 signaling cascade; increased expression of SOCS1 results in a need for increased IL-2 levels for survival of Tregs [55]. Thus, OX40 signaling in Tregs can modulate suppressor functions both by decreasing their effector function and by requiring higher amounts of IL-2 for continued survival and activation. The intergenic SNP associated with overall survival in cases with mucinous ovarian cancer may act through either a GITR or OX40 mechanism. Dissecting the role of the germline variants may help to identify mechanisms that could explain why the immune system is not able to mount an effective response to ovarian cancers.

In conclusion, our analysis of 3,662 invasive ovarian cancer cases suggests that inherited variants related to Tregs are associated with ovarian cancer outcome in a subtype-specific manner, even after adjustment for known prognostic features. Our findings underscore the importance of subtype-specific analyses in clinical and epidemiological studies of ovarian cancer, given the established disease heterogeneity, with each histologic subtype expressing different patterns of genetic, epidemiologic and clinical characteristics (reviewed by Karst and Drapkin [56]). Future work should include examination of additional study populations, immunological studies, and correlation of inherited variants with other tumor features, such as levels of Treg infiltration.


We are grateful to the family and friends of Kathryn Sladek Smith for their generous support of Ovarian Cancer Association Consortium through their donations to the Ovarian Cancer Research Fund.

Author Contributions

Conceived and designed the experiments: ELG KRK ALO LCH KBM KLK. Performed the experiments: JMC BW GK. Analyzed the data: MJM BLF SMA DJS PR KG RAV DNR H Sicotte CW. Contributed reagents/materials/analysis tools: CMP JMS RPW EI AB RS MJB SH LP SAG SJR NW HPY MG H Sicotte H Song JT PPDP TAS RBN LES KO. Wrote the paper: ELG MD KLK.


  1. 1. Siegel R, Naishadham D, Jemal A (2012) Cancer statistics, 2012. CA Cancer J Clin 62: 10–29.
  2. 2. Hoskins WJ, Bundy BN, Thigpen JT, Omura GA (1992) The influence of cytoreductive surgery on recurrence-free interval and survival in small-volume stage III epithelial ovarian cancer: a Gynecologic Oncology Group study. Gynecol Oncol 47: 159–166.
  3. 3. McGuire V, Jesser CA, Whittemore AS (2002) Survival among U.S. women with invasive epithelial ovarian cancer. Gynecol Oncol 84: 399–403.
  4. 4. Bolton KL, Chenevix-Trench G, Goh C, Sadetzki S, Ramus SJ, et al. (2012) Association between BRCA1 and BRCA2 mutations and survival in women with invasive epithelial ovarian cancer. JAMA 307: 382–389.
  5. 5. Bolton KL, Tyrer J, Song H, Ramus SJ, Notaridou M, et al. (2010) Common variants at 19p13 are associated with susceptibility to ovarian cancer. Nat Genet 42: 880–884.
  6. 6. Zhang L, Conejo-Garcia JR, Katsaros D, Gimotty PA, Massobrio M, et al. (2003) Intratumoral T cells, recurrence, and survival in epithelial ovarian cancer. N Engl J Med 348: 203–213.
  7. 7. Sato E, Olson SH, Ahn J, Bundy B, Nishikawa H, et al. (2005) Intraepithelial CD8+ tumor-infiltrating lymphocytes and a high CD8+/regulatory T cell ratio are associated with favorable prognosis in ovarian cancer. Proc Natl Acad Sci U S A 102: 18538–18543.
  8. 8. Knutson KL, Krco C, Goodman K, Kelemen LE, Low PS, et al. (2006) T cell immunity to the folate receptor alpha is prevalent in women with breast or ovarian cancer. J Clin Oncol 24: 4254–4261.
  9. 9. Preston CC, Goode EL, Hartmann LC, Kalli KR, Knutson KL (2011) Immunity and immune suppression in human ovarian cancer. Immunotherapy 3: 539–556.
  10. 10. Knutson KL, Disis ML, Salazar LG (2007) CD4 regulatory T cells in human cancer pathogenesis. Cancer Immunol Immunother 56: 271–285.
  11. 11. Curiel TJ, Coukos G, Zou L, Alvarez X, Cheng P, et al. (2004) Specific recruitment of regulatory T cells in ovarian carcinoma fosters immune privilege and predicts reduced survival. Nat Med 10: 942–949.
  12. 12. Wolf D, Wolf AM, Rumpold H, Fiegl H, Zeimet AG, et al. (2005) The expression of the regulatory T cell-specific forkhead box transcription factor FoxP3 is associated with poor prognosis in ovarian cancer. Clin Cancer Res 11: 8326–8331.
  13. 13. Maggi E, Cosmi L, Liotta F, Romagnani P, Romagnani S, et al. (2005) Thymic regulatory T cells. Autoimmun Rev 4: 579–586.
  14. 14. Zou W (2006) Regulatory T cells, tumour immunity and immunotherapy. Nat Rev Immunol 6: 295–307.
  15. 15. Roncarolo MG, Bacchetta R, Bordignon C, Narula S, Levings MK (2001) Type 1 T regulatory cells. Immunol Rev 182: 68–79.
  16. 16. Wahl SM, Swisher J, McCartney-Francis N, Chen W (2004) TGF-beta: the perpetrator of immune suppression by regulatory T cells and suicidal T cells. J Leukoc Biol 76: 15–24.
  17. 17. Moore KW, de Waal Malefyt R, Coffman RL, O'Garra A (2001) Interleukin-10 and the interleukin-10 receptor. Annu Rev Immunol 19: 683–765.
  18. 18. Knutson KL (2006) Strong-arming immune regulation: Suppressing regulatory T cell function to treat cancers. Future Oncol 2: 379–389.
  19. 19. Fahlen L, Read S, Gorelik L, Hurst SD, Coffman RL, et al. (2005) T cells that cannot respond to TGF-beta escape control by CD4(+)CD25(+) regulatory T cells. J Exp Med 201: 737–746.
  20. 20. Yang ZZ, Novak AJ, Stenson MJ, Witzig TE, Ansell SM (2006) Intratumoral CD4+CD25+ regulatory T-cell-mediated suppression of infiltrating CD4+ T-cells in B-cell non-Hodgkin lymphoma. Blood 107: 3639–3646.
  21. 21. website. Available: Accessed 2010 Feb 2.
  22. 22. Gamazon ER, Zhang W, Konkashbaev A, Duan S, Kistner EO, et al. (2010) SCAN: SNP and copy number annotation. Bioinformatics 26: 259–262.
  23. 23. The 1000 Genomes Project Consortium (2010) A map of human genome variation from population-scale sequencing. Nature 467: 1061–1073.
  24. 24. Program for Genomic Applications: Seattle SNPs. Available: Accessed 2010 Jun 4.
  25. 25. Sicotte H, Rider D, Poland G, Dhiman N, Kocher J-P (2011) SNPPicker: High quality tag SNP selection across multiple populations. BMC Bioinformatics 12: 129.
  26. 26. White KL, Schildkraut JM, Palmieri RT, Iversen ES Jr, Berchuck A, et al. (2012) Ovarian cancer risk associated with inherited inflammation-related variants. Cancer Res 72: 1064–1069.
  27. 27. Fridley BL, Jenkins GD, Tsai YY, Song H, Bolton KL, et al. (2012) Gene set analysis of survival following ovarian cancer implicates macrolide binding and intracellular signaling genes. Cancer Epidemiol Biomarkers Prev 21: 529–536.
  28. 28. University of Michigan: Center for Statistical Genetics. Available: Accessed 2012 Jan 6.
  29. 29. Skol AD, Scott LJ, Abecasis GR, Boehnke M (2006) Joint analysis is more efficient than replication-based analysis for two-stage genome-wide association studies. Nat Genet 38: 209–213.
  30. 30. Perez VL, Van Parijs L, Biuckians A, Zheng XX, Strom TB, et al. (1997) Induction of peripheral T cell tolerance in vivo requires CTLA-4 engagement. Immunity 6: 411–417.
  31. 31. Greenwald RJ, Boussiotis VA, Lorsbach RB, Abbas AK, Sharpe AH (2001) CTLA-4 regulates induction of anergy in vivo. Immunity 14: 145–155.
  32. 32. Chen L (2004) Co-inhibitory molecules of the B7-CD28 family in the control of T-cell immunity. Nat Rev Immunol 4: 336–347.
  33. 33. Grohmann U, Orabona C, Fallarino F, Vacca C, Calcinaro F, et al. (2002) CTLA-4-Ig regulates tryptophan catabolism in vivo. Nat Immunol 3: 1097–1101.
  34. 34. Tang Q, Adams JY, Tooley AJ, Bi M, Fife BT, et al. (2006) Visualizing regulatory T cell control of autoimmune responses in nonobese diabetic mice. Nat Immunol 7: 83–92.
  35. 35. Tadokoro CE, Shakhar G, Shen S, Ding Y, Lino AC, et al. (2006) Regulatory T cells inhibit stable contacts between CD4+ T cells and dendritic cells in vivo. J Exp Med 203: 505–511.
  36. 36. Ross EM, Wilkie TM (2000) GTPase-activating proteins for heterotrimeric G proteins: regulators of G protein signaling (RGS) and RGS-like proteins. Annu Rev Biochem 69: 795–827.
  37. 37. Hollinger S, Hepler JR (2002) Cellular regulation of RGS proteins: modulators and integrators of G protein signaling. Pharmacol Rev 54: 527–559.
  38. 38. Agenes F, Bosco N, Mascarell L, Fritah S, Ceredig R (2005) Differential expression of regulator of G-protein signalling transcripts and in vivo migration of CD4+ naive and regulatory T cells. Immunology 115: 179–188.
  39. 39. Birzele F, Fauti T, Stahl H, Lenter MC, Simon E, et al. (2011) Next-generation insights into regulatory T cells: expression profiling and FoxP3 occupancy in Human. Nucleic Acids Res 39: 7946–7960.
  40. 40. (IMSGC) IMSGC (2010) IL12A, MPHOSPH9/CDK2AP1 and RGS1 are novel multiple sclerosis susceptibility loci. Genes Immun 11: 397–405.
  41. 41. Smyth DJ, Plagnol V, Walker NM, Cooper JD, Downes K, et al. (2008) Shared and distinct genetic variants in type 1 diabetes and celiac disease. N Engl J Med 359: 2767–2777.
  42. 42. Johnson BA, Wang J, Taylor EM, Caillier SJ, Herbert J, et al. (2010) Multiple sclerosis susceptibility alleles in African Americans. Genes Immun 11: 343–350.
  43. 43. Hunt KA, Zhernakova A, Turner G, Heap GAR, Franke L, et al. (2008) Newly identified genetic risk variants for celiac disease related to the immune response. Nat Genet 40: 395–402.
  44. 44. Wang R, Kozhaya L, Mercer F, Khaitan A, Fujii H, et al. (2009) Expression of GARP selectively identifies activated human FOXP3+ regulatory T cells. Proc Natl Acad Sci U S A 106: 13439–13444.
  45. 45. Probst-Kepper M, Balling R, Buer J (2010) FOXP3: required but not sufficient. the role of GARP (LRRC32) as a safeguard of the regulatory phenotype. Curr Mol Med 10: 533–539.
  46. 46. Stockis J, Colau D, Coulie PG, Lucas S (2009) Membrane protein GARP is a receptor for latent TGF-β on the surface of activated human Treg. Eur J Immunol 39: 3315–3322.
  47. 47. Tran DQ, Glass DD, Uzel G, Darnell DA, Spalding C, et al. (2009) Analysis of Adhesion Molecules, Target Cells, and Role of IL-2 in Human FOXP3+ Regulatory T Cell Suppressor Function. J Immunol 182: 2929–2938.
  48. 48. Andersson J, Tran DQ, Pesu M, Davidson TS, Ramsey H, et al. (2008) CD4+ FoxP3+ regulatory T cells confer infectious tolerance in a TGF-beta-dependent manner. J Exp Med 205: 1975–1981.
  49. 49. Ronchetti S, Zollo O, Bruscoli S, Agostini M, Bianchini R, et al. (2004) GITR, a member of the TNF receptor superfamily, is costimulatory to mouse T lymphocyte subpopulations. Eur J Immunol 34: 613–622.
  50. 50. Nocentini G, Riccardi C (2005) GITR: a multifaceted regulator of immunity belonging to the tumor necrosis factor receptor superfamily. Eur J Immunol 35: 1016–1022.
  51. 51. Shimizu J, Yamazaki S, Takahashi T, Ishida Y, Sakaguchi S (2002) Stimulation of CD25(+)CD4(+) regulatory T cells through GITR breaks immunological self-tolerance. Nat Immunol 3: 135–142.
  52. 52. McHugh RS, Whitters MJ, Piccirillo CA, Young DA, Shevach EM, et al. (2002) CD4(+)CD25(+) immunoregulatory T cells: gene expression analysis reveals a functional role for the glucocorticoid-induced TNF receptor. Immunity 16: 311–323.
  53. 53. Kitamura N, Murata S, Ueki T, Mekata E, Reilly RT, et al. (2009) OX40 costimulation can abrogate Foxp3+ regulatory T cell-mediated suppression of antitumor immunity. Int J Cancer 125: 630–638.
  54. 54. Vu MD, Xiao X, Gao W, Degauque N, Chen M, et al. (2007) OX40 costimulation turns off Foxp3+ Tregs. Blood 110: 2501–2510.
  55. 55. Lu L-F, Thai T-H, Calado DP, Chaudhry A, Kubo M, et al. (2009) Foxp3-dependent microRNA155 confers competitive fitness to regulatory T cells by targeting SOCS1 protein. Immunity 30: 80–91.
  56. 56. Karst AM, Drapkin R (2010) Ovarian cancer pathogenesis: a model in evolution. J Oncol 2010: 932371.