Figures
Abstract
Background
Tumor-infiltrating CD8+ T cells are correlated with prolonged progression-free and overall survival in epithelial ovarian cancer (EOC). A significant fraction of EOC patients mount autoantibody responses to various tumor antigens, however the relationship between autoantibodies and tumor-infiltrating T cells has not been investigated in EOC or any other human cancer. We hypothesized that autoantibody and T cell responses may be correlated in EOC and directed toward the same antigens.
Methodology and Principal Findings
We obtained matched serum and tumor tissue from 35 patients with high-grade serous ovarian cancer. Serum samples were assessed by ELISA for autoantibodies to the common tumor antigen NY-ESO-1. Tumor tissue was examined by immunohistochemistry for expression of NY-ESO-1, various T cell markers (CD3, CD4, CD8, CD25, FoxP3, TIA-1 and Granzyme B) and other immunological markers (CD20, MHC class I and MHC class II). Lymphocytic infiltrates varied widely among tumors and included cells positive for CD3, CD8, TIA-1, CD25, FoxP3 and CD4. Twenty-six percent (9/35) of patients demonstrated serum IgG autoantibodies to NY-ESO-1, which were positively correlated with expression of NY-ESO-1 antigen by tumor cells (r = 0.57, p = 0.0004). Autoantibodies to NY-ESO-1 were associated with increased tumor-infiltrating CD8+, CD4+ and FoxP3+ cells. In an individual HLA-A2+ patient with autoantibodies to NY-ESO-1, CD8+ T cells isolated from solid tumor and ascites were reactive to NY-ESO-1 by IFN-γ ELISPOT and MHC class I pentamer staining.
Conclusion and Significance
We demonstrate that tumor-specific autoantibodies and tumor-infiltrating T cells are correlated in human cancer and can be directed against the same target antigen. This implies that autoantibodies may collaborate with tumor-infiltrating T cells to influence clinical outcomes in EOC. Furthermore, serological screening methods may prove useful for identifying clinically relevant T cell antigens for immunotherapy.
Citation: Milne K, Barnes RO, Girardin A, Mawer MA, Nesslinger NJ, Ng A, et al. (2008) Tumor-Infiltrating T Cells Correlate with NY-ESO-1-Specific Autoantibodies in Ovarian Cancer. PLoS ONE 3(10): e3409. https://doi.org/10.1371/journal.pone.0003409
Editor: Aric Gregson, University of California Los Angeles, United States of America
Received: June 6, 2008; Accepted: September 17, 2008; Published: October 15, 2008
Copyright: © 2008 Milne et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: British Columbia Cancer Foundation, Canary Foundation, U.S. Department of Defense (OC000018). None of the sponsors had any role in the design and conduct of the study, in the collection, analysis, and interpretation of the data, or in the preparation, review, or approval of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Epithelial ovarian cancer (EOC) is a challenging disease that affects more than 190,000 women worldwide each year (International Agency for Research on Cancer). The high mortality rate is attributed to the fact that most patients are diagnosed with disseminated disease, often with extensive ascites. Standard treatment involves cytoreductive surgery followed by taxane- and platinum-based chemotherapy [1]. Over 80% of patients are highly responsive to frontline treatment, but 60–70% experience disease recurrence within 2–5 years and ultimately succumb to their disease [2], [3].
Despite these unfortunate statistics, 20–30% of EOC patients survive five years or more after diagnosis. Favorable prognostic factors include early stage, non-serous histology, low grade, good performance status, and optimal surgical debulking [4], [5]. In addition, several recent studies have shown a correlation between tumor-infiltrating CD3+CD8+ T cells and favorable outcomes [6], [7]. Zhang et. al. first reported that patients with CD3+ T cell infiltrates in tumor epithelium had increased progression-free and overall survival [8]. This has been confirmed by two other studies [9], [10], and two groups have extended this finding to the CD8+ T cell subset in particular [11], [12]. In addition, the presence of CD3+CD56+ T cells in ascites has been linked to platinum sensitivity [13]. These findings are in agreement with earlier studies showing a positive correlation between survival and expression of interferon-γ (IFN-γ) [14], [15], the IFN-γ receptor [16], IL-18 [17], and MHC class I [18], [19], all of which are characteristic of CD8+ T cell responses. In contrast, the presence of tumor-infiltrating CD25+FoxP3+ T cells in EOC is correlated with inferior survival [11], [20]–[22]. Thus, it appears that the balance of CD8+ effector T cells to CD25+FoxP3+ regulatory T cells is an important determinant of clinical outcomes in EOC.
In addition to tumor-infiltrating T cells, many EOC patients mount serum autoantibody responses to a variety of tumor antigens, including NY-ESO-1, HOXA7, Ep-CAM, HSP-90, MUC-1 and p53 [23]–[25]. In Type I diabetes and other autoimmune conditions, the development of autoantibody responses portends tissue infiltration and destruction by autoreactive T cells [26]. We therefore hypothesized that EOC patients may show a similar relationship between tumor-specific autoantibody responses and tumor-infiltrating lymphocytes. This hypothesis was tested in a cohort of 35 advanced stage, high grade serous EOC cases for which matched serum and tumor specimens were available. Using NY-ESO-1 as a test antigen, we demonstrate for the first time a correlation between tumor-specific autoantibodies and tumor-infiltrating T cells. Our findings raise the possibility that autoantibodies may play a role in the previously recognized relationship between tumor-infiltrating T cells and clinical outcomes in EOC.
Results
Study cohort
We investigated the relationship between tumor-specific autoantibodies and tumor-infiltrating lymphocytes using matched tumor and serum specimens from a retrospective cohort of 35 patients with high-grade serous EOC (Table 1). We elected to focus on a single histological subtype, as other subclasses of EOC exhibit distinct biological and clinical properties that might have confounded the analysis [27]. All blood samples were collected prior to surgery or chemotherapy, and all tumor specimens were obtained at the time of primary cytoreductive surgery prior to chemotherapy. Control blood samples were obtained from 60 age-matched women with no known personal history of cancer.
Analysis of tumor-infiltrating lymphocytes
Tumor-infiltrating lymphocytes were assessed by immunohistochemistry (IHC) using antibodies to a variety of immunological markers (Figure 1, Table 2). All raw IHC data can be found in Supplementary Tables S1 and S2. Only 23% (8/35) of evaluable tumors showed significant CD20+ B cell infiltrates. In contrast, 94% (33/35) of tumors had detectable CD3+ T cell infiltrates. Staining with antibodies to CD4 and CD8 revealed that 59% (20/34) and 69% (24/35) of evaluable tumors had significant CD4+ and CD8+ cellular infiltrates, respectively. CD4+ and CD8+ cells were strongly correlated (r = 0.69, p<0.0001). All evaluable tumors (27/27) expressed MHC class I to some degree, indicating they could theoretically present antigen to the infiltrating CD8+ T cells. Seventy-two percent (18/25) of tumors expressed MHC class II and hence could theoretically present antigen to CD4+ T cells.
Since many tumors had dense CD8+ T cell infiltrates, we analyzed tissues for TIA-1 and Granzyme B, both of which are expressed by CD8+ cytotoxic T cells as well as natural killer cells [28]–[30]. Seventy-three percent (25/34) of tumors had significant TIA-1+ cellular infiltrates, which showed a positive correlation with CD8+ infiltrates (r = 0.83, p<0.0001) (Figure 2). In contrast, only 20% (7/35) of the tumors had significant Granzyme B+ cellular infiltrates (data not shown); as expected, 6 of the 7 positive tumors were also positive for CD8+ cells.
Tumors were also analyzed for cellular infiltrates expressing FoxP3, a marker of activated T cells and regulatory T cells [31]. Sixty-six percent (23/35) of evaluable tumors had significant FoxP3+ infiltrates (Figure 2), which were strongly correlated to CD4+ cells (r = 0.73, p<0.0001). We also evaluated expression of CD25, an additional marker of activated and regulatory T cells [32]. Fifty percent (17/34) of tumors had significant CD25+ cellular infiltrates (Figure 2), which were strongly correlated with CD4+ cells (r = 0.58, p = 0.0003) and FoxP3+ cells (r = 0.75, p<0.0001).
We further stratified T cell infiltrates according to their epithelial or stromal location within the tumor, as intraepithelial CD3+/CD8+ T cells in particular have been correlated with increased survival in EOC [8]–[12]. For this analysis, we first measured the tumor composition as defined by the epithelial∶stromal ratio in each tissue core and then calculated the density of T cells per unit of epithelium or stroma. As summarized in Table 3, the density of CD3+ T cells per unit of tumor epithelium ranged from 0–17, with a median of 4.6. In comparison, the density of CD3+ T cells per unit of tumor stroma ranged from 0–138, with a median of 16.3. In general, the density of CD3+ T cells in tumor epithelium and stroma were only weakly correlated (r = 0.34, p = 0.048), and there were many examples of tumors with dense CD3+ infiltrates in epithelium but not stroma, and conversely, in stroma but not epithelium. Similar to CD3+ cells, CD8+, CD4+, FoxP3+ and TIA-1+ cells were generally denser in tumor stroma than tumor epithelium (Table 3).
Composition of tumor-infiltrating lymphocytes after neoadjuvant chemotherapy
The TMA used in the above analyses also contained an additional cohort of 15 tumors from women who, as part of a clinical trial, had undergone neoadjuvant platinum/taxane-based chemotherapy prior to their primary surgery. As with the 35-case cohort, these women had high-grade serous EOC. Tumors had been resected after three cycles of carboplatinum/taxol-based chemotherapy. Although the sample size was small, this provided a unique opportunity to evaluate the effects of chemotherapy on tumor-infiltrating lymphocytes. By most parameters, lymphocytic infiltrates were similar between treated and untreated tumors. However, treated tumors showed a uniform trend towards increased infiltration by all subsets of T cells assessed, and this increase was significantly higher for CD3+ (median 80 vs 35, p = 0.02) and CD8+ (median 78 vs 30, p = 0.013) cells (data not shown).
Serum autoantibody responses to NY-ESO-1
Although ovarian cancer patients demonstrate autoantibody responses to a broad repertoire of antigens [23]–[25], we focused on one of the most immunogenic antigens, NY-ESO-1 [24], [33], [34]. Sixty control sera were assayed for IgG autoantibodies to recombinant NY-ESO-1, and the mean and standard deviation of the OD values were calculated (Supplementary Tables S1 and S2). Individual sera were scored as positive if their OD value was equal to or greater than two standard deviations from the mean of control subjects. Consistent with published results, 26% (9/35) of ovarian cancer cases demonstrated IgG autoantibodies to NY-ESO-1, compared to only 5% (3/60) of controls (Figure 3A).
(A) Serum autoantibody responses to NY-ESO-1 in the 35-patient cohort. Autoantibody responses are reported as the number of standard deviations from the mean of 60 age- and gender-matched controls with no known personal history of cancer. (B,C) Immunohistochemical analysis of NY-ESO-1 expression in two representative serous ovarian tumors with high (B) and negative (C) expression of the antigen.
To determine whether autoantibody responses to NY-ESO-1 correlated with expression of the corresponding antigen, matched tumor specimens were analyzed by IHC for expression of NY-ESO-1. Of 34 evaluable tumors, 5 (14.7%) scored positive for NY-ESO-1 antigen (Figure 3B). All five of these cases were also positive for autoantibodies to NY-ESO-1. In contrast, 4 cases were positive for NY-ESO-1-specific autoantibodies but negative for NY-ESO-1 antigen expression.
Correlations between autoantibody responses and lymphocytic infiltrates
We next investigated whether serum autoantibodies to NY-ESO-1 were correlated with tumor-infiltrating lymphocytes. For this analysis, we considered epithelial and stromal T cell infiltrates separately, as described above. We classified cases as being positive or negative for autoantibodies to NY-ESO-1 using a cut-point of two standard deviations from the mean of the control group (Supplementary Tables S1 and S2). By Mann Whitney t test, patients with autoantibodies to NY-ESO-1 had a significantly greater stromal density of CD8+ cells (p = 0.011), FoxP3+ cells (p = 0.013), and CD4+ cells (p = 0.026). Thus, autoantibodies to NY-ESO-1 showed a significant correlation to T cell infiltrates, especially in tumor stroma.
Recognition of NY-ESO-1 by autoantibodies and tumor-infiltrating CD8+ T cells from the same patient
The correlation between autoantibodies to NY-ESO-1 and tumor-infiltrating T cells suggested the possibility that tumor-infiltrating T cells may recognize NY-ESO-1 in seropositive patients. To address this issue, we collected blood, tumor and ascites specimens from a prospective cohort of 15 newly diagnosed serous EOC patients. Two patients were positive for serum autoantibodies to NY-ESO-1. Of these, one case (IROC013) was HLA-A2+, allowing T cells to be enumerated by flow cytometry with HLA-A2 pentamers loaded with a known CD8+ epitope from NY-ESO-1 (Figure 4A). NY-ESO-1-specific CD8+ T cells were rare in peripheral blood from this patient (0.22% of CD8+ cells), but were enriched in ascites and solid tumor (1.53% and 6.64% of CD8+ cells, respectively). Furthermore, T cells from ascites and solid tumor produced IFN-γ in response to NY-ESO-1 peptide (Figure 4B) but not control peptides derived from p53, HER-2/neu or WT-1 (data not shown). Indeed, in ascites the response to NY-ESO-1 was almost as strong as that seen to the CEF viral control peptides. Intriguingly, tumor tissue from this patient showed dense CD3+ and CD8+ T cell infiltration of tumor stroma but not epithelium (Figure 4C), similar to the pattern commonly seen with autoantibody-positive patients in the retrospective cohort. Despite mounting a strong humoral and T-cell response to NY-ESO-1, solid tumor from this patient stained negative for expression of NY-ESO-1 antigen; however, ascites from this patient contained NY-ESO-1-positive cells (Figure 4C).
(A) MHC class I pentamer analysis demonstrating enrichment of NY-ESO-1-specific CD8+ T cells in ascites and solid tumor compared to peripheral blood. The boxed areas and associated numbers represent the percentage of pentamer-positive cells relative to total CD8+ cells. (B) ELISPOT analysis of IFN-γ production by T cells after stimulation with an HLA-A2-binding peptide from NY-ESO-1. Data is presented as the number of IFN-γ-producing cells per 1×106 bulk cells from the indicated tissue compartments. (C) Immunohistochemical analysis of tumor-infiltrating CD3+ and CD8+ T cells in tumor stroma, and expression of NY-ESO-1 antigen. While the solid tumor was negative for expression of NY-ESO-1, a fraction of cells from ascites were positive. The cellular fraction of ascites also contained cytokeratin-positive epithelial cells, presumably of tumor origin (data not shown).
Discussion
Using matched serum and tumor specimens from 35 patients with high-grade serous EOC, we have demonstrated a correlation between tumor-infiltrating T cells and tumor-specific autoantibodies. To our knowledge, this is the first study to identify such a correlation in any human cancer. Specifically, the presence of NY-ESO-1-specific IgG autoantibodies in serum correlated with infiltration of tumor stroma by cells expressing CD8, CD4 and FoxP3. Moreover, in an individual patient with autoantibodies to NY-ESO-1, corresponding ascites and solid tumor specimens were shown to be enriched for NY-ESO-1-reactive CD8+ T cells, as assessed by MHC class I pentamer staining and IFN-γ ELISPOT. These findings raise the possibility that autoantibody responses may collaborate with tumor-infiltrating T cells to influence clinical outcomes in EOC.
In addition to NY-ESO-1-specific autoantibodies, several factors expressed by ovarian tumors have shown a positive correlation with tumor-infiltrating T cells, including the chemokines CXCL9, CCL21, CCL22 [8], CCL2 and CCL5 [35]; p53 mutations [36]; and MHC class I [37]. Conversely, tumor-infiltrating T cells show a negative correlation with VEGF [8], B7-H1/PD-L1 [12], CD68+ macrophages [38] and the endothelin B receptor [39]. Thus, multiple factors influence the composition of tumor-infiltrating T cells in EOC.
While this study focused on a single, commonly recognized tumor antigen, NY-ESO-1, a large number of other autoantibody target antigens have been identified in EOC, including HOXA7, Ep-CAM, HSP-90, MUC-1 and p53 [23]–[25]. Indeed, tumor-specific autoantibody responses are common among EOC patients. For example, Stone et al. found that 44% of newly diagnosed EOC patients had an autoantibody response to at least 1 of a panel of 12 tumor antigens [24]. Future work will determine whether autoantibody responses to other tumor antigens also correlate with the presence of tumor-infiltrating T cells.
In contrast to serological responses, less is known about the target antigens of tumor-infiltrating T cells in EOC. T cell receptor (TCR) spectratyping studies have shown that tumor-infiltrating T cells in EOC represent oligoclonal populations, which is consistent with antigen-induced clonal expansion [40]–[44]. Many studies have shown recognition of autologous tumor cells by tumor-infiltrating T cells [45]–[50]. Moreover, previous work has identified T cells specific for HER2/neu [44], [51]–[53], p53 [54] and folate-binding protein [55], [56] among tumor-infiltrating or tumor-associated T cells. To our knowledge, the current study is the first to demonstrate recognition of NY-ESO-1 by tumor-infiltrating and tumor-associated (i.e., ascites derived) T cells, as evidenced by the results for patient IROC013. Intriguingly, T cell responses to NY-ESO-1 were undetectable in peripheral blood from this patient, indicating a strong enrichment of this T cell subpopulation in tumor and ascites. Indeed, the IFN-γ ELISPOT response to NY-ESO-1 was similar in magnitude to that seen with CEF control peptides. Nonetheless, the NY-ESO-1-specific subpopulation represented only 6.6% of all tumor-infiltrating CD8+ T cells in this subject, indicating there are likely many other antigens recognized by infiltrating T cells. Notably, over 80% of ovarian cancers exhibit loss of BRCA1 and/or BRCA2 function, leading to compromised DNA repair[57], [58]. One can speculate that the resulting genetic instability leads to the expression of abnormal proteins that may serve as neoantigens to the host immune system.
Although tumor infiltrating T cells are associated with prolonged progression-free and overall survival in EOC, it remains unclear whether this reflects an active or passive role of T cells. In other words, do T cells actively oppose tumor growth or are they simply markers of some other feature of the tumor that drives outcomes? There are several lines of evidence in support of the former possibility. First, not all T-cell subsets are associated with prognosis; rather, favorable outcomes are linked to the CD8+ subset and poor outcomes to the CD25+FoxP3+ subset [11], [12], [20], [21]. Second, several factors that are associated with active CD8+ cytolytic responses are also linked to favorable outcomes, including expression of IFN-γ [14], [15], the IFN-γ receptor [16] and IL-18 [17]. MHC class I has also been linked to increased survival in ovarian cancer [18], [19]. In the present study, which focused exclusively on the serous subtype, all evaluable tumors expressed MHC class I. Similar results have been reported by others for serous EOC [59], suggesting the vast majority of serous tumors are MHC class I positive and have the capacity to present antigen to tumor-infiltrating CD8+ T cells. Third, tumor-infiltrating T cells show significant cytotoxicity against ovarian tumors in vitro [43], [45], [47], [60], [61]. However, it is noteworthy that only a minority of tumors in this and another study [59] had significant Granzyme B-positive immune infiltrates, suggesting that cytotoxic function may be suppressed in vivo. And finally, as discussed above, tumor-infiltrating T cells show evidence of clonal expansion [40]–[44] and recognition of specific antigens [44], [51]–[56], as shown here for NY-ESO-1 (Figure 4). Thus, evidence is accumulating in favor of the concept that tumor-infiltrating T cells play an active role in promoting favorable clinical outcomes in EOC. Similarly, in colorectal cancer, functional markers associated with active Th1-like cytolytic T cell responses are linked to favorable clinical outcomes [62]. This suggests that immune modulatory strategies that enhance these naturally occurring T cell responses may improve clinical outcomes further.
In contrast to T cell infiltrates, the relationship between tumor-specific autoantibodies and clinical outcomes is less clear. This issue has been studied most extensively for the tumor antigen p53, which elicits autoantibody responses in 20–25% of EOC patients. Goodell et. al. reported a positive correlation between autoantibodies to p53 and increased overall survival in EOC [63]. However, other studies have found a negative correlation [64], [65] or no correlation [66], [67]. As for NY-ESO-1, we found a trend toward poor outcomes among patients with autoantibodies to NY-ESO-1 in the 35-case cohort studied here (data not shown), but this trend was not seen in an independent cohort of 35 patients from a prior study [24] (data not shown). As discussed above, the immune response to EOC involves multiple antigens, therefore we believe that the prognostic significance of autoantibodies is best addressed using an extended panel of tumor antigens and larger patient cohorts.
The autoantibody response to NY-ESO-1 correlated to T cell infiltration of tumor stroma as opposed to tumor epithelium. This may reflect a statistical issue, as the absolute number of T cells was higher in tumor stroma compared to epithelium, allowing the stromal values to achieve statistical significance. Indeed, Spearman rank correlation analysis showed a positive relationship between autoantibodies to NY-ESO-1 and infiltration of tumor epithelium by CD3+ and CD8+ cells, but this did not reach statistical significance (data not shown). Alternatively, there may be a biological explanation. In the Th1/Th2 paradigm, immune responses are thought to polarize toward humoral or cytolytic effector mechanisms [68]. Applied to our results, this would suggest that patients with strong autoantibody responses would have weak cytolytic responses, which may result in incomplete (i.e., stromal) infiltration of tumor tissue by T cells. A second possibility is suggested by murine models where autoantibody responses have been linked to weak CD8+ cytolytic T cell responses [69]. In this case, it was proposed that autoantibodies facilitate uptake and presentation of tumor antigens by B cells at the expense of dendritic cells. Since B cells are less potent antigen presenting cells than dendritic cells, the net result is an inferior T cell response against the tumor [69]. A final consideration is that, in the present study, autoantibody responses to NY-ESO-1 were correlated with infiltration of tumor by not only CD8+ T cells, but also by CD4+ and FoxP3+ T cells. The latter cells may represent regulatory T cells, which could inhibit cytolytic T cells responses and limit the extent of tumor infiltration [32]. We are currently collecting matched blood and tumor specimens from a larger, prospective cohort of EOC patients to better understand the relationship between tumor-specific autoantibodies, tumor-infiltrating T cells and clinical outcomes.
Although not the primary focus of this paper, we had the opportunity to assess the effect of neoadjuvant chemotherapy on tumor-infiltrating T cells in 15 patients. In general, treated tumors showed increased infiltration by all subsets of T cells, and this reached statistical significance for CD3+ and CD8+ cells. This is reminiscent of studies in breast cancer which showed increased T cell infiltration after chemotherapy with taxanes [70] and other agents [71] and an association with favorable clinical responses. Taxanes may also promote tumor immunity by enhancing T and NK cell activity [72]–[75] and antigen presentation [76]. Likewise, platinum agents can enhance cytokine synthesis by human T cells [77], abrogate suppressor T cell activity [78] and sensitize EOC cells to Fas-mediated apoptosis [79]. Thus, platinum/taxane-based chemotherapy may have favorable effects on host immunity to EOC, a hypothesis we are currently investigating in a prospective patient cohort.
Several groups have attempted adoptive immunotherapy of ovarian cancer using T cells expanded from tumor-infiltrating T cells [50], [80]–[89]. Although promising anecdotal responses have been reported, these efforts have generally met with limited success. It is noteworthy that solid tumor tissue from patient IROC013 was largely negative for expression of NY-ESO-1, despite having NY-ESO-1-specific T cells in tumor and ascites. If this scenario is representative of other EOC patients, it would suggest that many tumor-infiltrating T cells may recognize antigens that are poorly expressed by tumor tissue, possibly due to immune selection during tumor development. If we are to realize the promise of immunotherapy for EOC, there is a pressing need to identify target antigens that are essential to the growth and survival of recurrent, chemotherapy-resistant tumors. Our results suggest that autoantibody responses hold practical value for antigen identification and warrant further study with respect to their role in host tumor immunity and clinical outcomes.
Materials and Methods
Study subjects
All specimens and clinical data were obtained with informed consent under protocols approved by the Research Ethics Board of the BC Cancer Agency and the University of British Columbia. The retrospective case cohort consisted of 35 women with high-grade serous ovarian cancer from whom matched serum and tumor tissue was available (OvCaRe Ovarian Tumour Bank, Vancouver, BC, Canada). Tumor tissue was obtained at the time of primary surgery prior to any other treatment. Table 1 shows the general clinical characteristics of the 35-case cohort. The retrospective cohort also included tissue from an additional 15 women who received neoadjuvant chemotherapy prior to primary surgery; these cases are discussed separately in Results. Blood, ascites and tumor samples were also collected from a prospective cohort of 15 patients through the BC Cancer Agency's Tumour Tissue Repository. Control serum samples were obtained from 60 women with no known personal history of ovarian cancer or other cancers. All control subjects self reported receiving a negative mammographic result within the past year. The age distribution of the control cohort (mean 62.0 years, standard deviation 12.3 years, range 45.9 to 88.9 years) was similar to that of the case cohort.
Tumor and serum specimens
Retrospective cohort.
Tumor tissue was obtained during primary cytoreductive surgery. Tissue had an ischemia time of less than 30 minutes and spent less than 48 hours in formalin prior to being processed in paraffin. A tissue microarray (TMA) was constructed by taking duplicate 0.6 mm cores from tumor blocks after review of hematoxylin- and eosin-stained sections by a pathologist. Cores were selected from regions of tumor containing representative proportions of epithelium and stroma, while avoiding highly necrotic regions. TMAs were assembled using a Pathology Devices tissue arrayer (Westminster, MD). Serum samples from cases were collected prior to surgery or chemotherapy. Blood samples were processed by standard laboratory procedures.
Prospective cohort.
Blood was collected prior to surgery in heparanized Vacutainer tubes, and peripheral blood mononuclear cells (PBMC) were isolated by Ficoll density centrifugation. HLA-A2 status was determined by flow cytometry on a FACSCalibur (BD Biosciences, San Jose, CA) after surface staining PBMC with anti-HLA-A2 antibody (clone BB7.2, BD Pharmingen, San Diego, CA). Ascites collected during surgery was centrifuged (1200 rpm for 10 min), and red blood cells (RBCs) were removed by treatment with ACK lysis buffer (Sigma, St. Louis, MO). Solid tumor removed during surgery was minced with scalpels to approximately 2 mm2 and was then digested overnight at 4°C in RPMI 1640 (Invitrogen, Carlsbad, CA) containing collagenase Type I and IV (each at 0.05 mg/ml), 0.025 mg/ml hyaluronidase and 0.01 mg/ml DNAse I (all from Sigma, St. Louis, MO). After digestion, material was passed through a 100 µm sterile cell strainer to remove clumps, and the resulting single-cell suspension was pelleted as described above.
Immunohistochemistry and immunocytochemistry
TMAs were sectioned at 5 µm onto Superfrost plus slides (Fisher Scientific, Ottawa, ON) and incubated overnight at 37°C. Following deparaffinization, the slides were placed in a Ventana Discovery XT autostainer (Ventana, Tucson, AZ) for immunohistochemical staining. Ventana's standard CC1 protocol was used for antigen retrieval. Primary antibodies are listed in Table 2.
TMAs were incubated with primary antibodies for 60 minutes, and the appropriate cross-adsorbed, biotinylated secondary antibody (Jackson Immunoresearch, West Grove, PA) was applied for 32 minutes. Bound antibodies were detected using the DABMap kit (Ventana), counterstained with hematoxylin (Ventana), and coverslipped manually with Cytoseal-60 (Richard Allan, Kalamazoo, MI).
Ascites from serous ovarian cancer patients was collected and centrifuged using a Cytospin III cytocentrifuge (Thermo Shandon, Waltham, MA). The slides were then fixed in acetone and stored at −80°C until subjected to immunohistochemistry using standard protocols. The ascites sample from patient IROC013 was stained with antibodies to NY-ESO-1, CD3, CD8 and pan-cytokeratin (Table 2).
Histopathological analysis
Immunostained TMAs were examined independently by two pathologists and a high degree of inter-observer concordance was achieved (r>0.7, p<0.0001) as well as intra-observer concordance on different sessions (r = 0.79, p<0.0001). Although the TMAs showed reasonable core retention, some cores were lost during the sectioning process and were not evaluable for one or more markers. Immunostaining was scored using three semi-quantitative IHC scoring systems. For NY-ESO-1 antigen, a modified H score approach was used [IHC score = (% positive neoplastic epithelial cells)×(staining intensity ranked from 0 to 3)] that ranged from 0–300. Positive was defined as an H score greater than 10; in practice, negative cases had H scores of 0–5, whereas positive cases had H scores ranging from 105–250. For MHC class I and II, a simplified four category scale was used (0 to 3+); a score of ≥2 was defined as positive. For immune cells, scoring was undertaken using a Chalkley 25 point array and methods similar to those used to assess vascular density. Briefly, each immunostained tumor was reviewed at low magnification and the core with the highest density of positive cells was selected. This core was then assessed at higher magnification (×20 objective) with a 25 cross hair grid overlaid on the image. Under a ×20 objective magnification, this grid defines an area of 0.56 mm2. The proportion of the core occupied by tumor epithelium was estimated, as was the total number of positive immune cells within the area of the grid. The number of grid points that coincided with positively staining immune cells within both epithelial and stromal areas was then determined. Positive immune cells that touched or overlapped with tumor epithelial compartments were counted as intraepithelial. All other positive cells were counted as intrastromal. For subsequent statistical analyses, an epithelial or stromal region was considered positive for a given immune cell population if there were more than five cells per unit area.
ELISA to detect serum autoantibodies to NY-ESO-1
A cDNA encoding NY-ESO-1 was amplified by reverse-transcriptase PCR from the ovarian cancer cell line OVCAR-3. The C-terminal 25 amino acids were truncated to improve solubility of the protein. After sequencing, the cDNA was subcloned into the prokaryotic expression vector pDEST17, which adds a six-residue histidine tag at the N-terminus, and expressed in the E. coli strain BL21AI (Invitrogen, Carlsbad, CA). Urea-soluble recombinant NY-ESO-1 was purified on a HisTrap™ column (GE Healthcare, Fairfield Conn), eluted in urea buffer containing 500 mM imidazole, dialyzed in phosphate-buffered saline (PBS) and quantified by bicinchoninic acid (BCA) protein assay (Sigma, St. Louis, MO).
Maxisorp 96-well plates (Nunc, Roskilde, Denmark) were coated with 0.5 µg/well of purified NY-ESO-1 in 0.1 M carbonate buffer (33.5 mM Na2CO3, 0.1 M NaHCO3, pH 9.6) and incubated overnight at 4°C with gentle rocking. Plates were blocked with 3% bovine serum albumin (US Biological, Swampscott MA) in Tris buffered saline (TBS) containing 0.05% Tween-20 (3% BSA/TBST) for 2 hours at room temperature on a rapid shaker. All washes were performed with TBS/0.1% Tween-20 using a Skanwasher plate washer (Molecular Devices, Union City, CA). Plates were washed and incubated with patient and control serum diluted 1∶100 in 3% BSA/TBST for 1 hour at room temperature on a rapid shaker. All sera were assayed in triplicate. Plates were washed and incubated with goat anti-human IgG conjugated to horseradish peroxidase (Jackson, West Grove, PA) at 1∶10,000 in 3% BSA/TBST for 1 hour on a shaker at room temperature. Plates were developed with tetramethylbenzidine (TMB) (Neogen, Lansing, MI) for 3 minutes at room temperature and the reaction was stopped by addition of 1N HCl. The optical density of each well was analyzed at 450 nm on a Versamax plate reader (Molecular Devices, Union City, CA) and analyzed using Softmax Pro 4.8.
IFN-γ ELISPOT analysis
ELISPOT plates (MSIP, Millipore, Billerica, MA) were pre-coated overnight with 10 µg/ml anti-IFN-γ capture antibody (1-D1K-Mabtech, Cincinnati, OH) and then blocked for 2 hours at 37°C with cRPMI (RPMI 1640, 10% FBS, 2 mM L-glutamine, 50 uM 2-mercaptoethanol, 10 mM HEPES, 10 mM MEM non-essential amino acids, 10 mM sodium pyruvate and 50 ug/ml gentamicin). Single cell suspensions of PBMC, ascites cells or solid tumor-derived cells were prepared in 10 ml cRPMI and plated in triplicate at 3×105 cells per well. Cells were either left unstimulated (i.e., media only) or stimulated with HLA-A2-restricted tumor antigen peptides (NY-ESO-1157–165, p53264–272, WT-1126–134, HER-2/neu654–662; each at 10 µg/ml); a CEF (Cytomegalovirus, Epstein Barr, Flu) virus positive control peptide pool (10 µg/ml, Anaspec, San Jose, CA); or the T-cell mitogen phytohemagglutinin (PHA) (5 µg/ml). After overnight incubation at 37°C, ELISPOT plates were washed and incubated for 2 hours at 37°C with 1 µg/ml biotinylated anti-human IFN-γ (mAb 7-B6-1, Mabtech) followed by development with Vectastain ABC Elite kit and Vectastain AEC substrate reagent according to the manufacturer's instructions (Vector Labs, Burlingame, CA). Spots were quantified using a Zeiss automated ELISPOT reader and reported as the number of spot-forming cells (SFC) per 106 PBMC.
MHC class I pentamer analysis
Single-cell suspensions of ascites cells were depleted of red blood cells and stained with APC-conjugated HLA-A2-NY-ESO-1157–165 pentamer (Proimmune) according to the manufacturer's instructions. Pentamer staining was followed by surface staining with PerCP-conjugated anti-CD8 (53-6.7) (BD Pharmingen). Cells were analyzed on a BD FACSCalibur, and a minimum of 50,000 events were collected.
Statistical analysis
Spearman correlation, Mann Whitney t-tests, and log rank tests were performed as appropriate to test statistical significance using Graphpad Prism v4.2 (Graphpad Software, San Diego, CA). Additional un-paired t-test analysis was performed using JMP statistical software (v7.0) (SAS Institute, Cary, NC).
Acknowledgments
We thank the following for advice and technical assistance: Sindy Babinszky, Kristy Dillon, Xiaobo Duan, Sara Hahn, Jodi Leblanc, Heather Lockyer, Michele Martin, Erika Mehl, Jorg Michels, Melanie Olson, Lisa Szegedi, Dr. Nicholas van der Westhuizen, Erika Wall, Laura Walsh, Josh Wang, Nathan West, Taimei Yang and Siao Yong.
Author Contributions
Conceived and designed the experiments: KM RB AG MM NN RS ET JRW MQW DAW AW SEK PHW BN. Performed the experiments: KM RB AG NN AN RS ET JRW MQW DAW AW MK BG. Analyzed the data: KM RB AG MM NN AN JSN RS ET JRW MQW DAW AW MK SEK BG PHW BN. Contributed reagents/materials/analysis tools: EM. Wrote the paper: KM RB JSN PHW BN. Recruited patients to this study: EM.
References
- 1. Bookman MA (2005) Standard treatment in advanced ovarian cancer in 2005: the state of the art. Int J Gynecol Cancer 15: Suppl 3212–220.
- 2. Ozols RF, Bundy BN, Greer BE, Fowler JM, Clarke-Pearson D, et al. (2003) Phase III trial of carboplatin and paclitaxel compared with cisplatin and paclitaxel in patients with optimally resected stage III ovarian cancer: a Gynecologic Oncology Group study. J Clin Oncol 21: 3194–3200.
- 3. du Bois A, Luck HJ, Meier W, Adams HP, Mobus V, et al. (2003) A randomized clinical trial of cisplatin/paclitaxel versus carboplatin/paclitaxel as first-line treatment of ovarian cancer. J Natl Cancer Inst 95: 1320–1329.
- 4. Holschneider CH, Berek JS (2000) Ovarian cancer: epidemiology, biology, and prognostic factors. Semin Surg Oncol 19: 3–10.
- 5. Ozols RF (2000) Management of advanced ovarian cancer consensus summary. Advanced Ovarian Cancer Consensus Faculty. Semin Oncol 27: 47–49.
- 6. Nelson BH (2008) The impact of T-cell immunity on ovarian cancer outcomes. Immunol Rev 222: 101–116.
- 7. Gimotty PA, Zhang L, Alagkiozidis I, Cadungog M, Adams S, et al. (2007) Immune prognostic factors in ovarian cancer: lessons from translational research. Dis Markers 23: 445–452.
- 8. 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.
- 9. Raspollini MR, Castiglione F, Rossi Degl'innocenti D, Amunni G, Villanucci A, et al. (2005) Tumour-infiltrating gamma/delta T-lymphocytes are correlated with a brief disease-free interval in advanced ovarian serous carcinoma. Ann Oncol 16: 590–596.
- 10. Tomsova M, Melichar B, Sedlakova I, Steiner I (2008) Prognostic significance of CD3+ tumor-infiltrating lymphocytes in ovarian carcinoma. Gynecol Oncol 108: 415–420.
- 11. 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.
- 12. Hamanishi J, Mandai M, Iwasaki M, Okazaki T, Tanaka Y, et al. (2007) Programmed cell death 1 ligand 1 and tumor-infiltrating CD8+ T lymphocytes are prognostic factors of human ovarian cancer. Proc Natl Acad Sci U S A 104: 3360–3365.
- 13. Bamias A, Tsiatas ML, Kafantari E, Liakou C, Rodolakis A, et al. (2007) Significant differences of lymphocytes isolated from ascites of patients with ovarian cancer compared to blood and tumor lymphocytes. Association of CD3(+)CD56(+) cells with platinum resistance. Gynecol Oncol.
- 14. Marth C, Fiegl H, Zeimet AG, Muller-Holzner E, Deibl M, et al. (2004) Interferon-gamma expression is an independent prognostic factor in ovarian cancer. Am J Obstet Gynecol 191: 1598–1605.
- 15. Kusuda T, Shigemasa K, Arihiro K, Fujii T, Nagai N, et al. (2005) Relative expression levels of Th1 and Th2 cytokine mRNA are independent prognostic factors in patients with ovarian cancer. Oncol Rep 13: 1153–1158.
- 16. Duncan TJ, Rolland P, Deen S, Scott IV, Liu DT, et al. (2007) Loss of IFN gamma receptor is an independent prognostic factor in ovarian cancer. Clin Cancer Res 13: 4139–4145.
- 17. Akahiro J, Konno R, Ito K, Okamura K, Yaegashi N (2004) Impact of serum interleukin-18 level as a prognostic indicator in patients with epithelial ovarian carcinoma. Int J Clin Oncol 9: 42–46.
- 18. Rolland P, Deen S, Scott I, Durrant L, Spendlove I (2007) Human leukocyte antigen class I antigen expression is an independent prognostic factor in ovarian cancer. Clin Cancer Res 13: 3591–3596.
- 19. Moore DH, Fowler WC Jr, Olafsson K (1990) Class I histocompatibility antigen expression: a prognostic factor for aneuploid ovarian cancers. 38: 458–461.
- 20. 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.
- 21. 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.
- 22. Woo EY, Chu CS, Goletz TJ, Schlienger K, Yeh H, et al. (2001) Regulatory CD4(+)CD25(+) T cells in tumors from patients with early-stage non-small cell lung cancer and late-stage ovarian cancer. Cancer Res 61: 4766–4772.
- 23. Luborsky JL, Barua A, Shatavi SV, Kebede T, Abramowicz J, et al. (2005) Anti-tumor antibodies in ovarian cancer. Am J Reprod Immunol 54: 55–62.
- 24. Stone B, Schummer M, Paley PJ, Thompson L, Stewart J, et al. (2003) Serologic analysis of ovarian tumor antigens reveals a bias toward antigens encoded on 17q. International Journal of Cancer 104: 73–84.
- 25. Chatterjee M, Mohapatra S, Ionan A, Bawa G, Ali-Fehmi R, et al. (2006) Diagnostic markers of ovarian cancer by high-throughput antigen cloning and detection on arrays. Cancer Res 66: 1181–1190.
- 26. Wasserfall CH, Atkinson MA (2006) Autoantibody markers for the diagnosis and prediction of type 1 diabetes. Autoimmun Rev 5: 424–428.
- 27. Gilks CB, Ionescu DN, Kalloger SE, Kobel M, Irving J, et al. (2008) Tumor cell type can be reproducibly diagnosed and is of independent prognostic significance in patients with maximally debulked ovarian carcinoma. Human Pathology. In press.
- 28. Pipkin ME, Lieberman J (2007) Delivering the kiss of death: progress on understanding how perforin works. Curr Opin Immunol 19: 301–308.
- 29. Kawakami A, Tian Q, Duan X, Streuli M, Schlossman SF, et al. (1992) Identification and functional characterization of a TIA-1-related nucleolysin. Proc Natl Acad Sci U S A 89: 8681–8685.
- 30. Kanavaros P, Boulland ML, Petit B, Arnulf B, Gaulard P (2000) Expression of cytotoxic proteins in peripheral T-cell and natural killer-cell (NK) lymphomas: association with extranodal site, NK or Tgammadelta phenotype, anaplastic morphology and CD30 expression. Leuk Lymphoma 38: 317–326.
- 31. Ziegler SF (2007) FOXP3: not just for regulatory T cells anymore. Eur J Immunol 37: 21–23.
- 32. Zou W (2006) Regulatory T cells, tumour immunity and immunotherapy. Nat Rev Immunol 6: 295–307.
- 33. Odunsi K, Jungbluth AA, Stockert E, Qian F, Gnjatic S, et al. (2003) NY-ESO-1 and LAGE-1 cancer-testis antigens are potential targets for immunotherapy in epithelial ovarian cancer. Cancer Res 63: 6076–6083.
- 34. Yakirevich E, Sabo E, Lavie O, Mazareb S, Spagnoli GC, et al. (2003) Expression of the MAGE-A4 and NY-ESO-1 cancer-testis antigens in serous ovarian neoplasms. Clin Cancer Res 9: 6453–6460.
- 35. Negus RP, Stamp GW, Hadley J, Balkwill FR (1997) Quantitative assessment of the leukocyte infiltrate in ovarian cancer and its relationship to the expression of C-C chemokines. Am J Pathol 150: 1723–1734.
- 36. Shah CA, Allison KH, Garcia RL, Gray HJ, Goff BA, et al. (2008) Intratumoral T cells, tumor-associated macrophages, and regulatory T cells: association with p53 mutations, circulating tumor DNA and survival in women with ovarian cancer. Gynecol Oncol 109: 215–219.
- 37. Kooi S, Zhang HZ, Patenia R, Edwards CL, Platsoucas CD, et al. (1996) HLA class I expression on human ovarian carcinoma cells correlates with T-cell infiltration in vivo and T-cell expansion in vitro in low concentrations of recombinant interleukin-2. 174: 116–128.
- 38. Helal Tel A, Alla AE, Laban MA, Fahmy RM (2004) Immunophenotyping of tumor-infiltrating mononuclear cells in ovarian carcinoma. Pathol Oncol Res 10: 80–84.
- 39. Buckanovich RJ, Facciabene A, Kim S, Benencia F, Sasaroli D, et al. (2008) Endothelin B receptor mediates the endothelial barrier to T cell homing to tumors and disables immune therapy. Nat Med 14: 28–36.
- 40. Hayashi K, Yonamine K, Masuko-Hongo K, Iida T, Yamamoto K, et al. (1999) Clonal expansion of T cells that are specific for autologous ovarian tumor among tumor-infiltrating T cells in humans. Gynecol Oncol 74: 86–92.
- 41. Pappas J, Jung WJ, Barda AK, Lin WL, Fincke JE, et al. (2005) Substantial proportions of identical beta-chain T-cell receptor transcripts are present in epithelial ovarian carcinoma tumors. Cell Immunol 234: 81–101.
- 42. Halapi E, Yamamoto Y, Juhlin C, Jeddi-Tehrani M, Grunewald J, et al. (1993) Restricted T cell receptor V-beta and J-beta usage in T cells from interleukin-2-cultured lymphocytes of ovarian and renal carcinomas. Cancer Immunol Immunother 36: 191–197.
- 43. Peoples GE, Davey MP, Goedegebuure PS, Schoof DD, Eberlein TJ (1993) T cell receptor V beta 2 and V beta 6 mediate tumor-specific cytotoxicity by tumor-infiltrating lymphocytes in ovarian cancer. J Immunol 151: 5472–5480.
- 44. Peoples GE, Yoshino I, Douville CC, Andrews JV, Goedegebuure PS, et al. (1994) TCR V beta 3+ and V beta 6+ CTL recognize tumor-associated antigens related to HER2/neu expression in HLA-A2+ ovarian cancers. J Immunol 152: 4993–4999.
- 45. Goedegebuure PS, Douville CC, Doherty JM, Linehan DC, Lee KY, et al. (1997) Simultaneous production of T helper-1-like cytokines and cytolytic activity by tumor-specific T cells in ovarian and breast cancer. Cell Immunol 175: 150–156.
- 46. Dadmarz RD, Ordoubadi A, Mixon A, Thompson CO, Barracchini KC, et al. (1996) Tumor-infiltrating lymphocytes from human ovarian cancer patients recognize autologous tumor in an MHC class II-restricted fashion. Cancer J Sci Am 2: 263–272.
- 47. Ioannides CG, Platsoucas CD, Rashed S, Wharton JT, Edwards CL, et al. (1991) Tumor cytolysis by lymphocytes infiltrating ovarian malignant ascites. Cancer Res 51: 4257–4265.
- 48. Peoples GE, Schoof DD, Andrews JV, Goedegebuure PS, Eberlein TJ (1993) T-cell recognition of ovarian cancer. Surgery 114: 227–234.
- 49. Peoples GE, Goedegebuure PS, Andrews JV, Schoof DD, Eberlein TJ (1993) HLA-A2 presents shared tumor-associated antigens derived from endogenous proteins in ovarian cancer. J Immunol 151: 5481–5491.
- 50. Freedman RS, Platsoucas CD (1996) Immunotherapy for peritoneal ovarian carcinoma metastasis using ex vivo expanded tumor infiltrating lymphocytes. Cancer Treat Res 82: 115–146.
- 51. Yoshino I, Peoples GE, Goedegebuure PS, Maziarz R, Eberlein TJ (1994) Association of HER2/neu expression with sensitivity to tumor-specific CTL in human ovarian cancer. J Immunol 152: 2393–2400.
- 52. Peoples GE, Goedegebuure PS, Smith R, Linehan DC, Yoshino I, et al. (1995) Breast and ovarian cancer-specific cytotoxic T lymphocytes recognize the same HER2/neu-derived peptide. Proc Natl Acad Sci USA 92: 432–436.
- 53. Linehan DC, Peoples GE, Hess DT, Summerhayes IC, Parikh AS, et al. (1995) In vitro stimulation of ovarian tumour-associated lymphocytes with a peptide derived from HER2/neu induces cytotoxicity against autologous tumour. Surg Oncol 4: 41–49.
- 54. Lambeck A, Leffers N, Hoogeboom BN, Sluiter W, Hamming I, et al. (2007) P53-specific T cell responses in patients with malignant and benign ovarian tumors: Implications for p53 based immunotherapy. Int J Cancer.
- 55. Peoples GE, Anderson BW, Fisk B, Kudelka AP, Wharton JT, et al. (1998) Ovarian cancer-associated lymphocyte recognition of folate binding protein peptides. Ann Surg Oncol 5: 743–750.
- 56. Peoples GE, Anderson BW, Lee TV, Murray JL, Kudelka AP, et al. (1999) Vaccine implications of folate binding protein, a novel cytotoxic T lymphocyte-recognized antigen system in epithelial cancers. Clin Cancer Res 5: 4214–4223.
- 57. Geisler JP, Hatterman-Zogg MA, Rathe JA, Buller RE (2002) Frequency of BRCA1 dysfunction in ovarian cancer. J Natl Cancer Inst 94: 61–67.
- 58. Hilton JL, Geisler JP, Rathe JA, Hattermann-Zogg MA, DeYoung B, et al. (2002) Inactivation of BRCA1 and BRCA2 in ovarian cancer. J Natl Cancer Inst 94: 1396–1406.
- 59. Nijman HW, van Diest PJ, Poort-Keesom RJ, von Mensdorff-Pouilly S, Verstraeten RA, et al. (2001) T cell infiltration and MHC I and II expression in the presence of tumor antigens: An immunohistochemical study in patients with serous epithelial ovarian cancer. Eur J Obstet Gynecol Reprod Biol 94: 114–120.
- 60. Santin AD, Bellone S, Palmieri M, Bossini B, Cane S, et al. (2004) Restoration of tumor specific human leukocyte antigens class I-restricted cytotoxicity by dendritic cell stimulation of tumor infiltrating lymphocytes in patients with advanced ovarian cancer. Int J Gynecol Cancer 14: 64–75.
- 61. Ioannides CG, Freedman RS, Platsoucas CD, Rashed S, Kim YP (1991) Cytotoxic T cell clones isolated from ovarian tumor-infiltrating lymphocytes recognize multiple antigenic epitopes on autologous tumor cells. J Immunol 146: 1700–1707.
- 62. Galon J, Costes A, Sanchez-Cabo F, Kirilovsky A, Mlecnik B, et al. (2006) Type, density, and location of immune cells within human colorectal tumors predict clinical outcome. Science 313: 1960–1964.
- 63. Goodell V, Salazar LG, Urban N, Drescher CW, Gray H, et al. (2006) Antibody immunity to the p53 oncogenic protein is a prognostic indicator in ovarian cancer. J Clin Oncol 24: 762–768.
- 64. Vogl FD, Frey M, Kreienberg R, Runnebaum IB (2000) Autoimmunity against p53 predicts invasive cancer with poor survival in patients with an ovarian mass. Br J Cancer 83: 1338–1343.
- 65. Mayerhofer K, Tempfer C, Kucera E, Hefler L, Zeisler H, et al. (1999) Humoral p53 antibody response is a prognostic parameter in ovarian cancer. Anticancer Res 19: 875–878.
- 66. Hogdall EV, Hogdall CK, Blaakaer J, Heegaard NH, Glud E, et al. (2002) P53 autoantibodies in sera from Danish ovarian cancer patients and their correlation with clinical data and prognosis. Apmis 110: 545–553.
- 67. Gadducci A, Ferdeghini M, Buttitta F, Cosio S, Fanucchi A, et al. (1999) Assessment of the prognostic relevance of serum anti-p53 antibodies in epithelial ovarian cancer. 72: 76–81.
- 68. Crane IJ, Forrester JV (2005) Th1 and Th2 lymphocytes in autoimmune disease. Crit Rev Immunol 25: 75–102.
- 69. Qin Z, Richter G, Schuler T, Ibe S, Cao X, et al. (1998) B cells inhibit induction of T cell-dependent tumor immunity. Nat Med 4: 627–630.
- 70. Demaria S, Volm MD, Shapiro RL, Yee HT, Oratz R, et al. (2001) Development of tumor-infiltrating lymphocytes in breast cancer after neoadjuvant paclitaxel chemotherapy. Clin Cancer Res 7: 3025–3030.
- 71. Ladoire S, Arnould L, Apetoh L, Coudert B, Martin F, et al. (2008) Pathologic complete response to neoadjuvant chemotherapy of breast carcinoma is associated with the disappearance of tumor-infiltrating foxp3+ regulatory T cells. Clin Cancer Res 14: 2413–2420.
- 72. Tsavaris N, Kosmas C, Vadiaka M, Kanelopoulos P, Boulamatsis D (2002) Immune changes in patients with advanced breast cancer undergoing chemotherapy with taxanes. Br J Cancer 87: 21–27.
- 73. Carson WE 3rd, Shapiro CL, Crespin TR, Thornton LM, Andersen BL (2004) Cellular immunity in breast cancer patients completing taxane treatment. Clin Cancer Res 10: 3401–3409.
- 74. Chan OT, Yang LX (2000) The immunological effects of taxanes. Cancer Immunol Immunother 49: 181–185.
- 75. Fitzpatrick FA, Wheeler R (2003) The immunopharmacology of paclitaxel (Taxol), docetaxel (Taxotere), and related agents. Int Immunopharmacol 3: 1699–1714.
- 76. Tsuda N, Chang DZ, Mine T, Efferson C, Garcia-Sastre A, et al. (2007) Taxol increases the amount and T cell activating ability of self-immune stimulatory multimolecular complexes found in ovarian cancer cells. Cancer Res 67: 8378–8387.
- 77. Riesbeck K (1999) Cisplatin at clinically relevant concentrations enhances interleukin-2 synthesis by human primary blood lymphocytes. Anticancer Drugs 10: 219–227.
- 78. Tsuda H, Kitahashi S, Umesaki N, Kanaoka Y, Kawabata M, et al. (1994) Abrogation of suppressor cells activity by cis-diamminedichloroplatinum (CDDP) treatment using therapeutic doses in ovarian cancer patients. Gynecol Oncol 52: 218–221.
- 79. Bagnoli M, Balladore E, Luison E, Alberti P, Raspagliesi F, et al. (2007) Sensitization of p53-mutated epithelial ovarian cancer to CD95-mediated apoptosis is synergistically induced by cisplatin pretreatment. Mol Cancer Ther 6: 762–772.
- 80. Stewart JA, Belinson JL, Moore AL, Dorighi JA, Grant BW, et al. (1990) Phase I trial of intraperitoneal recombinant interleukin-2/lymphokine-activated killer cells in patients with ovarian cancer. Cancer Res 50: 6302–6310.
- 81. Steis RG, Urba WJ, VanderMolen LA, Bookman MA, Smith JW 2nd, et al. (1990) Intraperitoneal lymphokine-activated killer-cell and interleukin-2 therapy for malignancies limited to the peritoneal cavity. J Clin Oncol 8: 1618–1629.
- 82. Urba WJ, Clark JW, Steis RG, Bookman MA, Smith JW 2nd, et al. (1989) Intraperitoneal lymphokine-activated killer cell/interleukin-2 therapy in patients with intra-abdominal cancer: immunologic considerations. J Natl Cancer Inst 81: 602–611.
- 83. Aoki Y, Takakuwa K, Kodama S, Tanaka K, Takahashi M, et al. (1991) Use of adoptive transfer of tumor-infiltrating lymphocytes alone or in combination with cisplatin-containing chemotherapy in patients with epithelial ovarian cancer. Cancer Res 51: 1934–1939.
- 84. Ikarashi H, Fujita K, Takakuwa K, Kodama S, Tokunaga A, et al. (1994) Immunomodulation in patients with epithelial ovarian cancer after adoptive transfer of tumor-infiltrating lymphocytes. Cancer Res 54: 190–196.
- 85. Fujita K, Ikarashi H, Takakuwa K, Kodama S, Tokunaga A, et al. (1995) Prolonged disease-free period in patients with advanced epithelial ovarian cancer after adoptive transfer of tumor-infiltrating lymphocytes. Clin Cancer Res 1: 501–507.
- 86. Freedman RS, Edwards CL, Kavanagh JJ, Kudelka AP, Katz RL, et al. (1994) Intraperitoneal adoptive immunotherapy of ovarian carcinoma with tumor-infiltrating lymphocytes and low-dose recombinant interleukin-2: a pilot trial. J Immunother Emphasis Tumor Immunol 16: 198–210.
- 87. Freedman RS, Platsoucas CD, Deisseroth AB (1996) Use of a retroviral vector to study the trafficking patterns of purified ovarian tumor infiltrating lymphocytes (TIL) used in intraperitoneal adoptive immunotherapy of ovarian cancer patients. A pilot study. Hum Gene Ther 7: 1351–1365.
- 88. Freedman RS, Kudelka AP, Kavanagh JJ, Verschraegen C, Edwards CL, et al. (2000) Clinical and biological effects of intraperitoneal injections of recombinant interferon-gamma and recombinant interleukin 2 with or without tumor-infiltrating lymphocytes in patients with ovarian or peritoneal carcinoma. Clin Cancer Res 6: 2268–2278.
- 89. Fujimoto K, Tomonaga M, Goto S (2006) A case of recurrent ovarian cancer successfully treated with adoptive immunotherapy and lentinan. Anticancer Res 26: 4015–4018.