- Email: [email protected]
Antigen Specific T-Cell Responses Against Tumor Antigens are Controlled by Regulatory T Cells in Patients With Prostate Cancer
Antigen Specific T-Cell Responses Against Tumor Antigens are Controlled by Regulatory T Cells in Patients With Prostate Cancer Boris Hadaschik,*,† Yun Su,† Eva Huter, Yingzi Ge, Markus Hohenfellner and Philipp Beckhove From the Departments of Urology (BH, YS, MH) and Dermatology (EH), University Hospital Heidelberg and Translational Immunology Unit, The German Cancer Research Center (YG, PB), Heidelberg, Germany, and Department of Urology, Jiangsu Province Hospital of Traditional Chinese Medicine (YS), Nanjing, People’s Republic of China
Abbreviations and Acronyms BPH ⫽ benign prostatic hyperplasia CRPC ⫽ castration resistant PC DC ⫽ dendritic cell EGFR ⫽ epidermal growth factor receptor ELISPOT ⫽ enzyme-linked immunospot HM ⫽ healthy male donor hu ⫽ human IFN-␥ ⫽ interferon-␥ IL ⫽ interleukin MAGE-3 ⫽ melanoma antigenencoding gene-3 MUC ⫽ mucin PAP ⫽ prostatic acid phosphatase PC ⫽ prostate cancer PSA ⫽ prostate specific antigen TAA ⫽ tumor antigen Treg ⫽ regulatory T cell Submitted for publication January 25, 2011. Study received University of Heidelberg ethical committee approval. Supported by the Deutsche Forschungsgemeinschaft (German Research Foundation) (YS). Supplementary material for this article can be obtained at http://www.dkfz.de/de/translationaleimmunologie/images/data/Prostatecancerantigens.gif. * Correspondence: Department of Urology, University Hospital Heidelberg, Im Neuenheimer Feld 110, 69120 Heidelberg, Germany (telephone: 01149 6221 56 36454; FAX: 01149 6221 56 5366; e-mail: [email protected]). † Equal study contribution.
1458
www.jurology.com
Purpose: Immunotherapy is a promising approach in an effort to control castration resistant prostate cancer. We characterized tumor antigen reactive T cells in patients with prostate cancer and analyzed the suppression of antitumor responses by regulatory T cells. Materials and Methods: Peripheral blood samples were collected from 57 patients with histologically confirmed prostate cancer, 8 patients with benign prostatic hyperplasia and 16 healthy donors. Peripheral blood mononuclear cells were isolated and antigen specific interferon-␥ secretion of isolated T cells was analyzed by enzyme-linked immunospot assay. T cells were functionally characterized and T-cell responses before and after regulatory T-cell depletion were compared. As test tumor antigens, a panel of 11 long synthetic peptides derived from a total of 8 tumor antigens was used, including prostate specific antigen and prostatic acid phosphatase. Results: In patients with prostate cancer we noted a 74.5% effector T-cell response rate compared with only 25% in patients with benign prostatic hyperplasia and 31% in healthy donors. In most patients 2 or 3 tumor antigens were recognized. Comparing various disease stages there was a clear increase in the immune response against prostate specific antigens from intermediate to high risk tumors and castration resistant disease. Regulatory T-cell depletion led to a significant boost in effector T-cell responses against prostate specific antigen and prostatic acid phosphatase. Conclusions: Tumor specific effector T cells were detected in most patients with prostate cancer, especially those with castration resistant prostate cancer. Since effector T-cell responses against prostate specific antigens strongly increased after regulatory T-cell depletion, our results indicate that immunotherapy efficacy could be enhanced by decreasing regulatory T cells. Key Words: prostate; prostatic neoplasms; T-lymphocytes, regulatory; immunotherapy; antigens PROSTATE cancer is the most common noncutaneous cancer in men in Western countries and the second leading cause of male cancer death.1 Although early detection has increased with serum PSA testing, mortality has only decreased slightly. For patients with
metastatic disease androgen withdrawal is the mainstay of systemic therapy. Cancer progression and death occur despite castrate testosterone levels within a few years in most cases.2 In the last decade treatment options for men with progressive CRPC
0022-5347/12/1874-1458/0 THE JOURNAL OF UROLOGY® © 2012 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION
Vol. 187, 1458-1465, April 2012 Printed in U.S.A. DOI:10.1016/j.juro.2011.11.083
AND
RESEARCH, INC.
ANTIGEN SPECIFIC T-CELL RESPONSES AGAINST TUMOR ANTIGENS
have changed considerably. Besides docetaxel chemotherapy and improved androgen deprivation therapy immunotherapeutic approaches are a promising regimen.3,4 The host immune system can recognize tumor cells, as indicated by autoantibodies reactive to prostate derived peptides in patients with PC.5,6 However, progressive tumor growth despite T-cell infiltration shows the failure of the immune system to effectively combat PC.7 Cancer immunotherapy aims at reinforcing the patient T-cell antitumor immune response to mediate tumor regression. Therapeutic cancer vaccines have been investigated for a long time but, aside from sipuleucel-T and PROSTVAC®-VF, most vaccine strategies have had only limited success in clinical settings.8 –11 This might be explained by recent insights into the nature of regulatory elements of the immune system. Suppression of T-cell responses by Tregs was suggested to prevent the effective generation of immunity to tumor antigens.12 In patients with PC increased tumor Tregs and enhanced Treg functionality in peripheral blood were described.13,14 A correlation was recently noted of decreased Treg function and overall CRPC survival.15 We characterized the natural repertoire of tumor antigen reactive T cells in patients with PC. Our results have potential relevance for T-cell immune diagnostics, tumor vaccine design and the prediction of antitumor immune responsiveness.
1459
myeloid precursors with magnetic beads using a Dynal® T Cell Negative Isolation Kit, as recommended by the manufacturer. Before use in functional assays T cells were transferred into cytokine-free medium for 24 hours. To generate DCs adherent cells (80% to 90% monocytes) were incubated for 5 to 8 days in serum-free X-VIVO 20 containing 50 ng/ml recombinant human granulocyte macrophage colony-stimulating factor (Behringwerke, Marburg, Germany) and 1,000 U/ml IL-4. DCs were purified by negative depletion of contaminating cells using anti-CD3, anti-CD56 and anti-CD19 coupled magnetic beads (Dynal) and pulsed for 14 hours with 200 g/ml test or control polypeptides. To deplete Tregs the T cells were isolated for CD4 and CD25 expression by magnetic bead isolation with the MACS® CD4⫹CD25⫹ Regulatory T-cell Isolation Kit according to the manufacturer protocol. CD4⫹CD25- and CD8⫹ T cells depleted of CD4⫹CD25⫹ cells were mixed and used as the Treg depleted T-cell fraction.
Antigens As TAAs, we used 11 long synthetic polypeptides, including pure PSA, PSA plus, PAP3, PAP5, MUC-1 signal sequence (MUC-100), MUC-1 tandem repeat (MUC 20), MAGE-3, Her-2/neu, survivin, EGFR and p53. All peptides were designed to contain a previously described immunogenic HLA-A*0201 T-cell epitope.18 –22 In addition to TAAs, we used huIgG (CSL Behring, King of Prussia, Pennsylvania) and polypeptides derived from nonmalignancy associated antigens (IgG1 and collagen IV) as control antigens.
Flow Cytometry
MATERIALS AND METHODS Patient Details Peripheral blood samples (50 to 100 ml) were collected from 57 patients with histologically confirmed PC, 8 patients with BPH before transurethral resection and 16 HMs younger than 45 years. The mean age of patients with PC was 65.6 years (range 46 to 84 years). Of these men 19 (33%), 25 (44%) and 4 (7%) were classified into the D’Amico high, intermediate and low risk group, respectively.16 All 48 men had blood samples drawn immediately before radical prostatectomy without neoadjuvant hormonal therapy. Nine men (16%) in the PC cohort had CRPC. The mean age of patients with BPH was 71.3 years (range 57 to 77). Informed written consent was obtained from all study participants. The protocol was approved by the University of Heidelberg ethical committee.
Cell Purification and Culture DCs and T cells were generated as described previously.17,18 Briefly, mononuclear cell populations were separated over a Ficoll gradient (Amersham Pharmacia, Little Chalfont, United Kingdom) and cultured in cytokine-free X-VIVO™ 20 medium. T cells were obtained by washing off nonadherent cells. They were cultivated in RPMI medium (Life Technologies™) supplemented with 10% serum, 60 U/ml IL-4 (PromoCell, Heidelberg, Germany) and 100 U/ml IL-2 (Chiron, Emeryville, California) for 7 days. T lymphocytes were isolated after depleting B and NK cells, and
Mononuclear cells (5 ⫻ 105), 105 enriched CD4⫹ T cells or 105 T cells depleted of CD4⫹CD25⫹ Treg were blocked with Endobulin® polyclonal huIgs for 15 minutes on ice, followed by surface molecule staining using mouse antihuman monoclonal antibodies, including anti-CD3-PE (1:100), anti-CD4-PerCP-Cy5.5 (1:50) and anti-CD25-APCCy7 (BD Pharmingen™) (1:25) for 20 minutes. Intranuclear Foxp3 staining was then done according to the manufacturer anti-Foxp3-APC protocol (eBioscience®). Data were acquired on a FACSCanto™ II flow cytometer. Analysis was done using FlowJo 8.7 (Tree Star, Ashland, Oregon).
IFN-␥ ELISPOT Assay Cytokine secreting T cells were determined as previously described.17,18 Briefly, peptide pulsed DCs were co-incubated in triplicate wells with autologous T cells at a 1:5 ratio for 40 hours in ELISPOT plates and visualized using the ELISpot kit (Mabtech, Cincinnati, Ohio). Spots were measured automatically using an Axioplan 2 microscope and KS-ELISPOT software (Carl Zeiss Vision, Aalen, Germany). Spots induced by control peptides were considered unspecific background. Individuals were considered responders if the number of spots in the presence of test peptides was significantly greater than in the pooled wells of negative controls (p ⬍0.05). The frequency of tumor reactive T lymphocytes was calculated using the equation, (spots in test wells ⫺ spots in negative control wells)/ number of T cells per well. Each test group consisted of 3 wells. Results are shown as the mean ⫾ SEM.
1460
ANTIGEN SPECIFIC T-CELL RESPONSES AGAINST TUMOR ANTIGENS
Statistical Analysis Statistical analysis was done with Prism® 4.03 and InStat® 3.06. ELISPOT results and differences between test groups were calculated by the 2-sided Student t test, and repeated measures ANOVA and the Kruskal-Wallis test with post tests. Statistical significance was considered at p ⬍0.05.
RESULTS We used a broad panel of long synthetic peptides of established tumor antigens to identify and compare the antigen specificity of effector T cells in peripheral blood of patients with PC, patients with BPH and healthy donors. We synthesized 11 polypeptides derived from a total of 8 TAAs that are frequently over expressed in PC, including PSA and PAP.18 –22 To analyze TAA specific T-cell responses we performed 40-hour ELISPOT assays and measured the antigen specific secretion of IFN-␥ after co-culture of autologous peptide pulsed DCs and T cells.18 In this first set of experiments we tested 47 patients with histologically confirmed PC, 8 patients with BPH and 16 healthy controls. Overall 35 of 47 patients with PC (74.5%) had significant TAA specific T-cell responses compared with only 25% of those with BPH and 31% of healthy donors. We detected antigen specific IFN-␥ secretion to all PC associated peptides (pure PSA, PSA plus, PAP3 and PAP5) in patients with PC. Figure 1, A shows data on a representative patient. Compared to IgG this patient showed statistically significant responses against PSA plus and PAP3. Other known universal tumor antigens also induced specific IFN-␥ responses in patients with PC, indicating that antigen specific T cells exist in these patients. For these universal TAAs we chose huIgG as the background control antigen, which generally induced slightly higher background spot numbers than pure IgG1. If TAA induced spot numbers are below those of the control antigen, this can be explained by the presence of Tregs, which are also stimulated by the specific TAA. Low background spot numbers generally did not correlate with a higher probability of achieving positive (significant) results. In contrast to patients with PC, in most healthy men we detected no significant antigen specific IFN-␥ secretion to prostate derived peptides or to universal tumor antigens. Figure 1, B shows a representative example. Figure 2, A lists cumulative ELISPOT data on all patients with PC, patients with BPH and healthy controls. When focusing on the 4 prostate derived tumor antigens, there was a clear-cut difference in the response rate of up to 13% in patients with PC vs no significant response in men with BPH and healthy donors. In patients with PC and controls there were similarly high response rates toward
Figure. 1. Primary data from 40-hour IFN-␥ ELISPOT assays on T-cell (TC) response against TAA. Peripheral blood T cells were stimulated by DCs pulsed with polypeptides derived from tumor antigens or from negative control antigens IgG and huIgG. As background control antigens, we used IgG for pure PSA, PSA plus, PAP3 and PAP5, and huIgG for MUC-100, MUC 20, MAGE-3 (MAG), Her-2/neu (HER), survivin, EGFR and p53. Tumor reactive T lymphocyte frequency was calculated as (spots in test wells ⫺ spots in negative control wells)/number of T cells per well. Data represent mean ⫾ SEM number of spots in triplicate wells per antigen. Difference between spot number in test wells vs control antigens was calculated by 2-sided Student t test. A, representative patient with CRPC. B, representative healthy donor.
some other known TAAs, eg MUC-1, indicating that in all groups there were individuals with functional T-cells responses against certain TAA. However, we observed a markedly increased effector T-cell response rate in patients with PC compared to that in patients with BPH and healthy donors. To analyze whether differences in the number of antigen specific T cells existed among the groups,
ANTIGEN SPECIFIC T-CELL RESPONSES AGAINST TUMOR ANTIGENS
1461
Figure. 2. A, proportion of patients with PC, patients with BPH and healthy donors among all men tested who had significant T-cell reactivity against TAAs. Data represent percent of 8 to 47 participants with TAA specific T cells (spot number in test wells vs spot number of negative control antigens) by 2-tailed Student t test. Differences among test groups were calculated by repeated measures ANOVA (p ⫽ 0.0004) with Tukey-Kramer multiple comparisons post test. HER, Her-2/neu. Asterisk indicates p ⫽ 0.01 to 0.05. Triple asterisk indicates p ⬍ 0.001. MAG, MAGE-3. B, frequency of TAA specific T cells in peripheral blood from patients and healthy donors. Nonresponders were set at zero and are not visible in log scale format. Dots indicate T-cell frequency in samples containing significantly increased number of cytokine secreting cells in test wells vs control wells (responders). Lines indicate mean ⫾ SEM of all 88 to 497 tested samples per group. Differences among test groups were calculated by Kruskal-Wallis test (p ⫽ 0.0002) with Dunn multiple comparisons post test. C, number of different TAAs recognized in 35 responding patients with PC.
which could have explained differences in the overall responder rate, we compared the overall frequency of tumor antigen specific T cells (fig. 2, B). As expected, we found a significantly higher frequency
of responsive T cells in patients with PC. Across all patients and all TAAs the TAA specific T-cell responses showed a mean of 15.6/105 total T cells in patients with PC, 9.3/105 in those with BPH and
1462
ANTIGEN SPECIFIC T-CELL RESPONSES AGAINST TUMOR ANTIGENS
2.8/105 in healthy controls (Kruskal-Wallis test p ⫽ 0.0002). Leaving out nonresponders, responders among patients with PC, patients with BPH and healthy controls showed a mean of 87.1/105, 102.5/ 105 and 44.1/105 total T cells, respectively (PC vs HM Student t test p ⫽ 0.048). There was a trend toward an increased immune reaction in patients with BPH vs healthy men (fig. 2, A and B). However, the small sample size of these groups did not allow us to draw any conclusions. To determine whether a single TAA or several TAAs accounted for the strong response rates in patients with PC we then analyzed the number of tumor antigens recognized. Similar to controls (data not shown), in most patients with PC 2 or 3 tumor antigens were recognized (fig. 2, C). We next determined whether the observed antitumor T-cell responses varied by risk group, including advanced tumor stage or D’Amico risk group. We used the D’Amico classification for surgical patients to combine Gleason score and PSA as well as T stage. Accordingly we subclassified our cohort of 47 patients with PC into 3 groups, including 15 with high risk disease, 23 with low to intermediate risk tumors and 9 with CRPC. In these cohorts there was an overall increase in the immune response from localized disease to metastatic CRPC. While 8 of 9 CRPC samples (89%) contained a significantly increased number of cytokine secreting cells in test wells compared with control wells (responders), in the high and intermediate to low risk groups only 80% and 65% of patients showed significant T-cell responses. Especially the response rate to the prostate specific tumor antigens PSA and PAP were markedly higher in the CRPC group than in the other groups (fig. 3, A). For the universal TAAs differences were not as striking but cumulative differences in responder rates were significant among clinical stages. Similarly the frequency of TAA specific T cells was highest for CRPC (fig. 3, B). The overall variation in the number of TAAs recognized in each responding subgroup was not significant (fig. 3, C). The observation of a clear increase in immune responses against prostate derived antigens for intermediate vs high risk tumors and CRPC led to the question of why these strong antitumor T-cell responses were not protective in patients with CRPC. A possible explanation is the presence of Tregs, which suppress effective antitumor effector T-cell responses. To test the potential influence of Tregs on antigen specific TAA immune responses we performed ELISPOT assays with total T cells and T cells after Treg depletion in 10 patients. CD4⫹CD25⫹Foxp3⫹ Tregs accounted for approximately 5% of the total number of CD3⫹CD4⫹ cells (mean 5.1% ⫾ 2.56%). Treg depletion by negative selection with magnetic beads resulted in less than 0.1% Foxp3⫹ Tregs. In a representative patient TAA spe-
cific T-cell responses increased after Treg depletion. The differences in this patient were significant for pure PSA and PAP5 (fig. 4, A). Figure 4, B shows ELISPOT data on prostate derived antigens before and after Treg depletion, normalized to the antigen specific IFN-␥ responses of total T cells. Treg depletion led to a significant increase in effector T-cell responses against all PSA and PAP peptides. This clearly demonstrates the suppression of antitumor specific T-cell responses by Tregs in patients with PC. Similarly for most universal TAAs tested we saw an increase in the response rate after Treg depletion (data not shown).
DISCUSSION We provide a comparative analysis of effector T-cell responses against common tumor antigens in the peripheral blood of patients with PC. We tested long synthetic peptides that contain a previously described immunogenic HLA-A*0201 T-cell epitope and detected TAA reactive T cells in 74.5% of patients with PC (fig. 2). The use of synthetic long peptides for vaccination circumvents HLA restrictions since long peptides are processed by antigen presenting cells and presented as peptide fragments by various HLA I and HLA II alleles.18,23 T-cell responses against TAAs were individual and oligovalent. T-cell frequency was significantly higher in patients with PC than in men with BPH or healthy young controls. Subgroup analysis of patients with PC revealed the highest antigen specific T-cell frequency in men with advanced disease (fig. 3). Antitumor vaccines are designed to activate antigen specific adaptive immune responses against tumors in targeted fashion. For instance, sipuleucel-T, the first Food and Drug Administration approved autologous cellular immunotherapy, is an investigational immunotherapy approach based on DC vaccination that is designed to stimulate T-cell immunity against PAP, an antigen expressed in most PCs.24 Since the effector T-cell response against prostate specific antigens such as PSA or PAP strongly increased after Treg depletion (fig. 4, B), our study provides a strong rationale for additional Treg targeted therapy to enhance vaccine mediated antitumor immunity. For example, tumor mediated immunosuppression might be reversed by treatment with cyclophosphamide, a chemotherapeutic agent with a dose dependent bimodal effect on the immune system, which decreases the number of Tregs in tumor bearing animals. Consequently immune responses after vaccination could be augmented by a combination regimen.25,26 Similarly treatment with denileukin difitox for Treg depletion improves the antitumor T-cell response.27 Treg depletion might cause autoimmune disease. For example, vitiligo is induced by DC vaccination against melanoma derived tumor antigens. However,
ANTIGEN SPECIFIC T-CELL RESPONSES AGAINST TUMOR ANTIGENS
1463
Figure. 3. A, proportion of patients with CRPC, and high and low/intermediate risk PC among all men tested who had significant T-cell reactivity against TAAs. Data represent percent of 9 to 23 patients with TAA specific T cells (spot number in test wells vs spot number of negative control antigens) by 2-tailed Student t test. Differences among test groups were calculated by repeated measures ANOVA (p ⫽ 0.014) with Tukey-Kramer multiple comparisons post test. HER, Her-2/neu. Asterisk indicates p ⫽ 0.01 to 0.05. MAG, MAGE-3. B, frequency of TAA specific T cells in individuals. Nonresponders were set at zero and are not visible in log scale format. Dots indicate T-cell frequency in samples containing significantly increased number of cytokine secreting cells in test wells vs control wells (responders). Lines indicate mean ⫾ SEM of all 90 to 238 tested samples per group. Differences among test groups were calculated by Kruskal-Wallis test (p ⫽ 0.0003) with Dunn multiple comparisons post test. Double asterisk indicates p ⫽ 0.001 to 0.05. Triple asterisk indicates p ⬍ 0.001. C, number of TAAs recognized in 8 to 15 responders with PC. int., intermediate.
in the context of a deadly tumor we regard the relevance of potential autoimmune reactions as negligible. Alternatively selecting antigens for vaccination according to the preexisting T-cell response and Treg specificity of an individual may improve the efficacy of immunotherapy.18 While peptide based vaccines boost
T-cell activity, at the same time expansion or reactivation of preexisting antigen specific Tregs should be avoided. Thus, more detailed information on which tumor antigens are under Treg control is needed. Similar to previously published data from our group on patients with colorectal carcinoma,18 Treg depletion in patients
1464
ANTIGEN SPECIFIC T-CELL RESPONSES AGAINST TUMOR ANTIGENS
Figure. 4. Data from 40-hour IFN-␥ ELISPOT assays. A, primary data on T-cell responses against TAA in representative patient with PC before and after Treg depletion. Peripheral blood T cells were stimulated with polypeptides derived from tumor antigens or negative control antigens, including IgG for pure PSA, PSA plus, PAP3 and PAP5, and huIgG for MUC-100, MUC 20, MAGE-3, Her-2/neu (HER), survivin, EGFR and p53. Tumor reactive T-lymphocyte frequency was calculated as (spots in test wells ⫺ spots in negative control wells)/number of T cells per well. Data represent mean ⫾ SEM spot number of triplicate wells per antigen. Differences between spot number in test wells vs control antigens were calculated by 2-sided Student t test. B, pooled data on prostate derived antigens of 10 patients with PC before and after Treg depletion, normalized to antigen specific IFN-␥ response before Treg depletion as 1. Differences between normalized TAA specific T cells were calculated by Wilcoxon signed rank test.
with PC only led to a negligible increase in the response rate against survivin and MAGE-3, suggesting that they might be Treg independent TAAs (data not shown). A limitation of our study and many others is that besides sipuleucel-T (PAP) and PROSTVAC-VF (PSA) the clinical relevance of common tumor antigens remains
to be confirmed in clinical trials. A proven T-cell response to certain TAAs does not necessarily translate into clinical benefit since there might be other, even more relevant tumor antigens that we are not aware of today. Also, during the process of boosting T-cell responses toward known TAAs certain tumor clones might be se-
ANTIGEN SPECIFIC T-CELL RESPONSES AGAINST TUMOR ANTIGENS
lected to lose expression of these tumor antigens, thereby advancing to a secondary level of immune evasion.28 Concerning the optimal timing of immunotherapy during the course of PC progression we detected more TAA specific effector T cells and increased strength of spontaneous T-cell responses in patients with castration resistant disease than in men with lower risk disease (fig. 3). Thus, CRPC seems to be a valid target for recent trials such as IMPACT (IMunotherapy Prostate Adeno Carcinoma Treatment).10 However, in our setting we have not yet had the opportunity to test patients with androgen sensitive metastatic disease. At least in animals the combination of androgen ablation and Treg targeted immune therapy or PROSTVAC seems promising.29,30
1465
CONCLUSIONS Tumor specific effector T cells were detected in most patients with PC, especially in men with castration resistant disease. Since T-cell responses against prostate specific antigens strongly increased after Treg depletion, our results indicate that immunotherapy could be enhanced by preventing Treg accumulation.
ACKNOWLEDGMENTS Mariana Bucur, The German Cancer Research Center, provided assistance. Jesco Pfitzenmaier and Dirk Jäger provided advice.
REFERENCES 1. Jemal A, Siegel R, Xu J et al: Cancer statistics, 2010. CA Cancer J Clin 2010; 60: 277.
12. Shevach EM: Fatal attraction: tumors beckon regulatory T cells. Nat Med 2004; 10: 900.
2. Oh WK and Kantoff PW: Management of hormone refractory prostate cancer: current standards and future prospects. J Urol 1998; 160: 1220.
13. Miller AM, Lundberg K, Ozenci V et al: CD4⫹CD25high T cells are enriched in the tumor and peripheral blood of prostate cancer patients. J Immunol 2006; 177: 7398.
3. Matera L: The choice of the antigen in the dendritic cell-based vaccine therapy for prostate cancer. Cancer Treat Rev 2010; 36: 131. 4. Harzstark AL and Small EJ: Immunotherapeutics in development for prostate cancer. Oncologist 2009; 14: 391. 5. Wang X, Yu J, Sreekumar A et al: Autoantibody signatures in prostate cancer. N Engl J Med 2005; 353: 1224. 6. Tan EM and Zhang J: Autoantibodies to tumorassociated antigens: reporters from the immune system. Immunol Rev 2008; 222: 328. 7. Mercader M, Bodner BK, Moser MT et al: T cell infiltration of the prostate induced by androgen withdrawal in patients with prostate cancer. Proc Natl Acad Sci U S A 2001; 98: 14565. 8. Sanda MG, Smith DC, Charles LG et al: Recombinant vaccinia-PSA (PROSTVAC) can induce a prostate-specific immune response in androgen-modulated human prostate cancer. Urology 1999; 53: 260. 9. Vieweg J, Su Z, Dahm P et al: Reversal of tumor-mediated immunosuppression. Clin Cancer Res 2007; 13: 727s.
21. Xue BH, Zhang Y, Sosman JA et al: Induction of human cytotoxic T lymphocytes specific for prostate-specific antigen. Prostate 1997; 30: 73. 22. Correale P, Walmsley K, Nieroda C et al: In vitro generation of human cytotoxic T lymphocytes specific for peptides derived from prostate-specific antigen. J Natl Cancer Inst 1997; 89: 293.
14. Yokokawa J, Cereda V, Remondo C et al: Enhanced functionality of CD4⫹CD25(high)FoxP3⫹ regulatory T cells in the peripheral blood of patients with prostate cancer. Clin Cancer Res 2008; 14: 1032.
23. Melief CJ and van der Burg SH: Immunotherapy of established (pre)malignant disease by synthetic long peptide vaccines. Nat Rev Cancer 2008; 8: 351.
15. Vergati M, Cereda V, Madan RA et al: Analysis of circulating regulatory T cells in patients with metastatic prostate cancer pre- versus post-vaccination. Cancer Immunol Immunother 2011; 60: 197.
24. Drake CG: Prostate cancer as a model for tumour immunotherapy. Nat Rev Immunol 2010; 10: 580.
16. D’Amico AV, Whittington R, Malkowicz SB et al: Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer. JAMA 1998; 280: 969. 17. Feuerer M, Beckhove P, Bai L et al: Therapy of human tumors in NOD/SCID mice with patientderived reactivated memory T cells from bone marrow. Nat Med 2001; 7: 452. 18. Bonertz A, Weitz J, Pietsch DH et al: Antigenspecific Tregs control T cell responses against a limited repertoire of tumor antigens in patients with colorectal carcinoma. J Clin Invest 2009; 119: 3311.
10. Kantoff PW, Higano CS, Shore ND et al: Sipuleucel-T immunotherapy for castration-resistant prostate cancer. N Engl J Med 2010; 363: 411.
19. Peshwa MV, Shi JD, Ruegg C et al: Induction of prostate tumor-specific CD8⫹ cytotoxic T-lymphocytes in vitro using antigen-presenting cells pulsed with prostatic acid phosphatase peptide. Prostate 1998; 36: 129.
11. Kantoff PW, Schuetz TJ, Blumenstein BA et al: Overall survival analysis of a phase II randomized controlled trial of a Poxviral-based PSA-targeted immunotherapy in metastatic castration-resistant prostate cancer. J Clin Oncol 2010; 28: 1099.
20. Machlenkin A, Azriel-Rosenfeld R, Volovitz I et al: Preventive and therapeutic vaccination with PAP-3, a novel human prostate cancer peptide, inhibits carcinoma development in HLA transgenic mice. Cancer Immunol Immunother 2007; 56: 217.
25. Di Paolo NC, Tuve S, Ni S et al: Effect of adenovirus-mediated heat shock protein expression and oncolysis in combination with low-dose cyclophosphamide treatment on antitumor immune responses. Cancer Res 2006; 66: 960. 26. Salem ML, Kadima AN, El-Naggar SA et al: Defining the ability of cyclophosphamide preconditioning to enhance the antigen-specific CD8⫹ T-cell response to peptide vaccination: creation of a beneficial host microenvironment involving type I IFNs and myeloid cells. J Immunother 2007; 30: 40. 27. Dannull J, Su Z, Rizzieri D et al: Enhancement of vaccine-mediated antitumor immunity in cancer patients after depletion of regulatory T cells. J Clin Invest 2005; 115: 3623. 28. Melief CJ: Cancer immunotherapy by dendritic cells. Immunity 2008; 29: 372. 29. Akins EJ, Moore ML, Tang S et al: In situ vaccination combined with androgen ablation and regulatory T-cell depletion reduces castration-resistant tumor burden in prostate-specific pten knockout mice. Cancer Res 2010; 70: 3473. 30. Arredouani MS, Tseng-Rogenski SS, Hollenbeck BK et al: Androgen ablation augments human HLA2.1restricted T cell responses to PSA self-antigen in transgenic mice. Prostate 2010; 70: 1002.
Abbreviations and Acronyms BPH ⫽ benign prostatic hyperplasia CRPC ⫽ castration resistant PC DC ⫽ dendritic cell EGFR ⫽ epidermal growth factor receptor ELISPOT ⫽ enzyme-linked immunospot HM ⫽ healthy male donor hu ⫽ human IFN-␥ ⫽ interferon-␥ IL ⫽ interleukin MAGE-3 ⫽ melanoma antigenencoding gene-3 MUC ⫽ mucin PAP ⫽ prostatic acid phosphatase PC ⫽ prostate cancer PSA ⫽ prostate specific antigen TAA ⫽ tumor antigen Treg ⫽ regulatory T cell Submitted for publication January 25, 2011. Study received University of Heidelberg ethical committee approval. Supported by the Deutsche Forschungsgemeinschaft (German Research Foundation) (YS). Supplementary material for this article can be obtained at http://www.dkfz.de/de/translationaleimmunologie/images/data/Prostatecancerantigens.gif. * Correspondence: Department of Urology, University Hospital Heidelberg, Im Neuenheimer Feld 110, 69120 Heidelberg, Germany (telephone: 01149 6221 56 36454; FAX: 01149 6221 56 5366; e-mail: [email protected]). † Equal study contribution.
1458
www.jurology.com
Purpose: Immunotherapy is a promising approach in an effort to control castration resistant prostate cancer. We characterized tumor antigen reactive T cells in patients with prostate cancer and analyzed the suppression of antitumor responses by regulatory T cells. Materials and Methods: Peripheral blood samples were collected from 57 patients with histologically confirmed prostate cancer, 8 patients with benign prostatic hyperplasia and 16 healthy donors. Peripheral blood mononuclear cells were isolated and antigen specific interferon-␥ secretion of isolated T cells was analyzed by enzyme-linked immunospot assay. T cells were functionally characterized and T-cell responses before and after regulatory T-cell depletion were compared. As test tumor antigens, a panel of 11 long synthetic peptides derived from a total of 8 tumor antigens was used, including prostate specific antigen and prostatic acid phosphatase. Results: In patients with prostate cancer we noted a 74.5% effector T-cell response rate compared with only 25% in patients with benign prostatic hyperplasia and 31% in healthy donors. In most patients 2 or 3 tumor antigens were recognized. Comparing various disease stages there was a clear increase in the immune response against prostate specific antigens from intermediate to high risk tumors and castration resistant disease. Regulatory T-cell depletion led to a significant boost in effector T-cell responses against prostate specific antigen and prostatic acid phosphatase. Conclusions: Tumor specific effector T cells were detected in most patients with prostate cancer, especially those with castration resistant prostate cancer. Since effector T-cell responses against prostate specific antigens strongly increased after regulatory T-cell depletion, our results indicate that immunotherapy efficacy could be enhanced by decreasing regulatory T cells. Key Words: prostate; prostatic neoplasms; T-lymphocytes, regulatory; immunotherapy; antigens PROSTATE cancer is the most common noncutaneous cancer in men in Western countries and the second leading cause of male cancer death.1 Although early detection has increased with serum PSA testing, mortality has only decreased slightly. For patients with
metastatic disease androgen withdrawal is the mainstay of systemic therapy. Cancer progression and death occur despite castrate testosterone levels within a few years in most cases.2 In the last decade treatment options for men with progressive CRPC
0022-5347/12/1874-1458/0 THE JOURNAL OF UROLOGY® © 2012 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION
Vol. 187, 1458-1465, April 2012 Printed in U.S.A. DOI:10.1016/j.juro.2011.11.083
AND
RESEARCH, INC.
ANTIGEN SPECIFIC T-CELL RESPONSES AGAINST TUMOR ANTIGENS
have changed considerably. Besides docetaxel chemotherapy and improved androgen deprivation therapy immunotherapeutic approaches are a promising regimen.3,4 The host immune system can recognize tumor cells, as indicated by autoantibodies reactive to prostate derived peptides in patients with PC.5,6 However, progressive tumor growth despite T-cell infiltration shows the failure of the immune system to effectively combat PC.7 Cancer immunotherapy aims at reinforcing the patient T-cell antitumor immune response to mediate tumor regression. Therapeutic cancer vaccines have been investigated for a long time but, aside from sipuleucel-T and PROSTVAC®-VF, most vaccine strategies have had only limited success in clinical settings.8 –11 This might be explained by recent insights into the nature of regulatory elements of the immune system. Suppression of T-cell responses by Tregs was suggested to prevent the effective generation of immunity to tumor antigens.12 In patients with PC increased tumor Tregs and enhanced Treg functionality in peripheral blood were described.13,14 A correlation was recently noted of decreased Treg function and overall CRPC survival.15 We characterized the natural repertoire of tumor antigen reactive T cells in patients with PC. Our results have potential relevance for T-cell immune diagnostics, tumor vaccine design and the prediction of antitumor immune responsiveness.
1459
myeloid precursors with magnetic beads using a Dynal® T Cell Negative Isolation Kit, as recommended by the manufacturer. Before use in functional assays T cells were transferred into cytokine-free medium for 24 hours. To generate DCs adherent cells (80% to 90% monocytes) were incubated for 5 to 8 days in serum-free X-VIVO 20 containing 50 ng/ml recombinant human granulocyte macrophage colony-stimulating factor (Behringwerke, Marburg, Germany) and 1,000 U/ml IL-4. DCs were purified by negative depletion of contaminating cells using anti-CD3, anti-CD56 and anti-CD19 coupled magnetic beads (Dynal) and pulsed for 14 hours with 200 g/ml test or control polypeptides. To deplete Tregs the T cells were isolated for CD4 and CD25 expression by magnetic bead isolation with the MACS® CD4⫹CD25⫹ Regulatory T-cell Isolation Kit according to the manufacturer protocol. CD4⫹CD25- and CD8⫹ T cells depleted of CD4⫹CD25⫹ cells were mixed and used as the Treg depleted T-cell fraction.
Antigens As TAAs, we used 11 long synthetic polypeptides, including pure PSA, PSA plus, PAP3, PAP5, MUC-1 signal sequence (MUC-100), MUC-1 tandem repeat (MUC 20), MAGE-3, Her-2/neu, survivin, EGFR and p53. All peptides were designed to contain a previously described immunogenic HLA-A*0201 T-cell epitope.18 –22 In addition to TAAs, we used huIgG (CSL Behring, King of Prussia, Pennsylvania) and polypeptides derived from nonmalignancy associated antigens (IgG1 and collagen IV) as control antigens.
Flow Cytometry
MATERIALS AND METHODS Patient Details Peripheral blood samples (50 to 100 ml) were collected from 57 patients with histologically confirmed PC, 8 patients with BPH before transurethral resection and 16 HMs younger than 45 years. The mean age of patients with PC was 65.6 years (range 46 to 84 years). Of these men 19 (33%), 25 (44%) and 4 (7%) were classified into the D’Amico high, intermediate and low risk group, respectively.16 All 48 men had blood samples drawn immediately before radical prostatectomy without neoadjuvant hormonal therapy. Nine men (16%) in the PC cohort had CRPC. The mean age of patients with BPH was 71.3 years (range 57 to 77). Informed written consent was obtained from all study participants. The protocol was approved by the University of Heidelberg ethical committee.
Cell Purification and Culture DCs and T cells were generated as described previously.17,18 Briefly, mononuclear cell populations were separated over a Ficoll gradient (Amersham Pharmacia, Little Chalfont, United Kingdom) and cultured in cytokine-free X-VIVO™ 20 medium. T cells were obtained by washing off nonadherent cells. They were cultivated in RPMI medium (Life Technologies™) supplemented with 10% serum, 60 U/ml IL-4 (PromoCell, Heidelberg, Germany) and 100 U/ml IL-2 (Chiron, Emeryville, California) for 7 days. T lymphocytes were isolated after depleting B and NK cells, and
Mononuclear cells (5 ⫻ 105), 105 enriched CD4⫹ T cells or 105 T cells depleted of CD4⫹CD25⫹ Treg were blocked with Endobulin® polyclonal huIgs for 15 minutes on ice, followed by surface molecule staining using mouse antihuman monoclonal antibodies, including anti-CD3-PE (1:100), anti-CD4-PerCP-Cy5.5 (1:50) and anti-CD25-APCCy7 (BD Pharmingen™) (1:25) for 20 minutes. Intranuclear Foxp3 staining was then done according to the manufacturer anti-Foxp3-APC protocol (eBioscience®). Data were acquired on a FACSCanto™ II flow cytometer. Analysis was done using FlowJo 8.7 (Tree Star, Ashland, Oregon).
IFN-␥ ELISPOT Assay Cytokine secreting T cells were determined as previously described.17,18 Briefly, peptide pulsed DCs were co-incubated in triplicate wells with autologous T cells at a 1:5 ratio for 40 hours in ELISPOT plates and visualized using the ELISpot kit (Mabtech, Cincinnati, Ohio). Spots were measured automatically using an Axioplan 2 microscope and KS-ELISPOT software (Carl Zeiss Vision, Aalen, Germany). Spots induced by control peptides were considered unspecific background. Individuals were considered responders if the number of spots in the presence of test peptides was significantly greater than in the pooled wells of negative controls (p ⬍0.05). The frequency of tumor reactive T lymphocytes was calculated using the equation, (spots in test wells ⫺ spots in negative control wells)/ number of T cells per well. Each test group consisted of 3 wells. Results are shown as the mean ⫾ SEM.
1460
ANTIGEN SPECIFIC T-CELL RESPONSES AGAINST TUMOR ANTIGENS
Statistical Analysis Statistical analysis was done with Prism® 4.03 and InStat® 3.06. ELISPOT results and differences between test groups were calculated by the 2-sided Student t test, and repeated measures ANOVA and the Kruskal-Wallis test with post tests. Statistical significance was considered at p ⬍0.05.
RESULTS We used a broad panel of long synthetic peptides of established tumor antigens to identify and compare the antigen specificity of effector T cells in peripheral blood of patients with PC, patients with BPH and healthy donors. We synthesized 11 polypeptides derived from a total of 8 TAAs that are frequently over expressed in PC, including PSA and PAP.18 –22 To analyze TAA specific T-cell responses we performed 40-hour ELISPOT assays and measured the antigen specific secretion of IFN-␥ after co-culture of autologous peptide pulsed DCs and T cells.18 In this first set of experiments we tested 47 patients with histologically confirmed PC, 8 patients with BPH and 16 healthy controls. Overall 35 of 47 patients with PC (74.5%) had significant TAA specific T-cell responses compared with only 25% of those with BPH and 31% of healthy donors. We detected antigen specific IFN-␥ secretion to all PC associated peptides (pure PSA, PSA plus, PAP3 and PAP5) in patients with PC. Figure 1, A shows data on a representative patient. Compared to IgG this patient showed statistically significant responses against PSA plus and PAP3. Other known universal tumor antigens also induced specific IFN-␥ responses in patients with PC, indicating that antigen specific T cells exist in these patients. For these universal TAAs we chose huIgG as the background control antigen, which generally induced slightly higher background spot numbers than pure IgG1. If TAA induced spot numbers are below those of the control antigen, this can be explained by the presence of Tregs, which are also stimulated by the specific TAA. Low background spot numbers generally did not correlate with a higher probability of achieving positive (significant) results. In contrast to patients with PC, in most healthy men we detected no significant antigen specific IFN-␥ secretion to prostate derived peptides or to universal tumor antigens. Figure 1, B shows a representative example. Figure 2, A lists cumulative ELISPOT data on all patients with PC, patients with BPH and healthy controls. When focusing on the 4 prostate derived tumor antigens, there was a clear-cut difference in the response rate of up to 13% in patients with PC vs no significant response in men with BPH and healthy donors. In patients with PC and controls there were similarly high response rates toward
Figure. 1. Primary data from 40-hour IFN-␥ ELISPOT assays on T-cell (TC) response against TAA. Peripheral blood T cells were stimulated by DCs pulsed with polypeptides derived from tumor antigens or from negative control antigens IgG and huIgG. As background control antigens, we used IgG for pure PSA, PSA plus, PAP3 and PAP5, and huIgG for MUC-100, MUC 20, MAGE-3 (MAG), Her-2/neu (HER), survivin, EGFR and p53. Tumor reactive T lymphocyte frequency was calculated as (spots in test wells ⫺ spots in negative control wells)/number of T cells per well. Data represent mean ⫾ SEM number of spots in triplicate wells per antigen. Difference between spot number in test wells vs control antigens was calculated by 2-sided Student t test. A, representative patient with CRPC. B, representative healthy donor.
some other known TAAs, eg MUC-1, indicating that in all groups there were individuals with functional T-cells responses against certain TAA. However, we observed a markedly increased effector T-cell response rate in patients with PC compared to that in patients with BPH and healthy donors. To analyze whether differences in the number of antigen specific T cells existed among the groups,
ANTIGEN SPECIFIC T-CELL RESPONSES AGAINST TUMOR ANTIGENS
1461
Figure. 2. A, proportion of patients with PC, patients with BPH and healthy donors among all men tested who had significant T-cell reactivity against TAAs. Data represent percent of 8 to 47 participants with TAA specific T cells (spot number in test wells vs spot number of negative control antigens) by 2-tailed Student t test. Differences among test groups were calculated by repeated measures ANOVA (p ⫽ 0.0004) with Tukey-Kramer multiple comparisons post test. HER, Her-2/neu. Asterisk indicates p ⫽ 0.01 to 0.05. Triple asterisk indicates p ⬍ 0.001. MAG, MAGE-3. B, frequency of TAA specific T cells in peripheral blood from patients and healthy donors. Nonresponders were set at zero and are not visible in log scale format. Dots indicate T-cell frequency in samples containing significantly increased number of cytokine secreting cells in test wells vs control wells (responders). Lines indicate mean ⫾ SEM of all 88 to 497 tested samples per group. Differences among test groups were calculated by Kruskal-Wallis test (p ⫽ 0.0002) with Dunn multiple comparisons post test. C, number of different TAAs recognized in 35 responding patients with PC.
which could have explained differences in the overall responder rate, we compared the overall frequency of tumor antigen specific T cells (fig. 2, B). As expected, we found a significantly higher frequency
of responsive T cells in patients with PC. Across all patients and all TAAs the TAA specific T-cell responses showed a mean of 15.6/105 total T cells in patients with PC, 9.3/105 in those with BPH and
1462
ANTIGEN SPECIFIC T-CELL RESPONSES AGAINST TUMOR ANTIGENS
2.8/105 in healthy controls (Kruskal-Wallis test p ⫽ 0.0002). Leaving out nonresponders, responders among patients with PC, patients with BPH and healthy controls showed a mean of 87.1/105, 102.5/ 105 and 44.1/105 total T cells, respectively (PC vs HM Student t test p ⫽ 0.048). There was a trend toward an increased immune reaction in patients with BPH vs healthy men (fig. 2, A and B). However, the small sample size of these groups did not allow us to draw any conclusions. To determine whether a single TAA or several TAAs accounted for the strong response rates in patients with PC we then analyzed the number of tumor antigens recognized. Similar to controls (data not shown), in most patients with PC 2 or 3 tumor antigens were recognized (fig. 2, C). We next determined whether the observed antitumor T-cell responses varied by risk group, including advanced tumor stage or D’Amico risk group. We used the D’Amico classification for surgical patients to combine Gleason score and PSA as well as T stage. Accordingly we subclassified our cohort of 47 patients with PC into 3 groups, including 15 with high risk disease, 23 with low to intermediate risk tumors and 9 with CRPC. In these cohorts there was an overall increase in the immune response from localized disease to metastatic CRPC. While 8 of 9 CRPC samples (89%) contained a significantly increased number of cytokine secreting cells in test wells compared with control wells (responders), in the high and intermediate to low risk groups only 80% and 65% of patients showed significant T-cell responses. Especially the response rate to the prostate specific tumor antigens PSA and PAP were markedly higher in the CRPC group than in the other groups (fig. 3, A). For the universal TAAs differences were not as striking but cumulative differences in responder rates were significant among clinical stages. Similarly the frequency of TAA specific T cells was highest for CRPC (fig. 3, B). The overall variation in the number of TAAs recognized in each responding subgroup was not significant (fig. 3, C). The observation of a clear increase in immune responses against prostate derived antigens for intermediate vs high risk tumors and CRPC led to the question of why these strong antitumor T-cell responses were not protective in patients with CRPC. A possible explanation is the presence of Tregs, which suppress effective antitumor effector T-cell responses. To test the potential influence of Tregs on antigen specific TAA immune responses we performed ELISPOT assays with total T cells and T cells after Treg depletion in 10 patients. CD4⫹CD25⫹Foxp3⫹ Tregs accounted for approximately 5% of the total number of CD3⫹CD4⫹ cells (mean 5.1% ⫾ 2.56%). Treg depletion by negative selection with magnetic beads resulted in less than 0.1% Foxp3⫹ Tregs. In a representative patient TAA spe-
cific T-cell responses increased after Treg depletion. The differences in this patient were significant for pure PSA and PAP5 (fig. 4, A). Figure 4, B shows ELISPOT data on prostate derived antigens before and after Treg depletion, normalized to the antigen specific IFN-␥ responses of total T cells. Treg depletion led to a significant increase in effector T-cell responses against all PSA and PAP peptides. This clearly demonstrates the suppression of antitumor specific T-cell responses by Tregs in patients with PC. Similarly for most universal TAAs tested we saw an increase in the response rate after Treg depletion (data not shown).
DISCUSSION We provide a comparative analysis of effector T-cell responses against common tumor antigens in the peripheral blood of patients with PC. We tested long synthetic peptides that contain a previously described immunogenic HLA-A*0201 T-cell epitope and detected TAA reactive T cells in 74.5% of patients with PC (fig. 2). The use of synthetic long peptides for vaccination circumvents HLA restrictions since long peptides are processed by antigen presenting cells and presented as peptide fragments by various HLA I and HLA II alleles.18,23 T-cell responses against TAAs were individual and oligovalent. T-cell frequency was significantly higher in patients with PC than in men with BPH or healthy young controls. Subgroup analysis of patients with PC revealed the highest antigen specific T-cell frequency in men with advanced disease (fig. 3). Antitumor vaccines are designed to activate antigen specific adaptive immune responses against tumors in targeted fashion. For instance, sipuleucel-T, the first Food and Drug Administration approved autologous cellular immunotherapy, is an investigational immunotherapy approach based on DC vaccination that is designed to stimulate T-cell immunity against PAP, an antigen expressed in most PCs.24 Since the effector T-cell response against prostate specific antigens such as PSA or PAP strongly increased after Treg depletion (fig. 4, B), our study provides a strong rationale for additional Treg targeted therapy to enhance vaccine mediated antitumor immunity. For example, tumor mediated immunosuppression might be reversed by treatment with cyclophosphamide, a chemotherapeutic agent with a dose dependent bimodal effect on the immune system, which decreases the number of Tregs in tumor bearing animals. Consequently immune responses after vaccination could be augmented by a combination regimen.25,26 Similarly treatment with denileukin difitox for Treg depletion improves the antitumor T-cell response.27 Treg depletion might cause autoimmune disease. For example, vitiligo is induced by DC vaccination against melanoma derived tumor antigens. However,
ANTIGEN SPECIFIC T-CELL RESPONSES AGAINST TUMOR ANTIGENS
1463
Figure. 3. A, proportion of patients with CRPC, and high and low/intermediate risk PC among all men tested who had significant T-cell reactivity against TAAs. Data represent percent of 9 to 23 patients with TAA specific T cells (spot number in test wells vs spot number of negative control antigens) by 2-tailed Student t test. Differences among test groups were calculated by repeated measures ANOVA (p ⫽ 0.014) with Tukey-Kramer multiple comparisons post test. HER, Her-2/neu. Asterisk indicates p ⫽ 0.01 to 0.05. MAG, MAGE-3. B, frequency of TAA specific T cells in individuals. Nonresponders were set at zero and are not visible in log scale format. Dots indicate T-cell frequency in samples containing significantly increased number of cytokine secreting cells in test wells vs control wells (responders). Lines indicate mean ⫾ SEM of all 90 to 238 tested samples per group. Differences among test groups were calculated by Kruskal-Wallis test (p ⫽ 0.0003) with Dunn multiple comparisons post test. Double asterisk indicates p ⫽ 0.001 to 0.05. Triple asterisk indicates p ⬍ 0.001. C, number of TAAs recognized in 8 to 15 responders with PC. int., intermediate.
in the context of a deadly tumor we regard the relevance of potential autoimmune reactions as negligible. Alternatively selecting antigens for vaccination according to the preexisting T-cell response and Treg specificity of an individual may improve the efficacy of immunotherapy.18 While peptide based vaccines boost
T-cell activity, at the same time expansion or reactivation of preexisting antigen specific Tregs should be avoided. Thus, more detailed information on which tumor antigens are under Treg control is needed. Similar to previously published data from our group on patients with colorectal carcinoma,18 Treg depletion in patients
1464
ANTIGEN SPECIFIC T-CELL RESPONSES AGAINST TUMOR ANTIGENS
Figure. 4. Data from 40-hour IFN-␥ ELISPOT assays. A, primary data on T-cell responses against TAA in representative patient with PC before and after Treg depletion. Peripheral blood T cells were stimulated with polypeptides derived from tumor antigens or negative control antigens, including IgG for pure PSA, PSA plus, PAP3 and PAP5, and huIgG for MUC-100, MUC 20, MAGE-3, Her-2/neu (HER), survivin, EGFR and p53. Tumor reactive T-lymphocyte frequency was calculated as (spots in test wells ⫺ spots in negative control wells)/number of T cells per well. Data represent mean ⫾ SEM spot number of triplicate wells per antigen. Differences between spot number in test wells vs control antigens were calculated by 2-sided Student t test. B, pooled data on prostate derived antigens of 10 patients with PC before and after Treg depletion, normalized to antigen specific IFN-␥ response before Treg depletion as 1. Differences between normalized TAA specific T cells were calculated by Wilcoxon signed rank test.
with PC only led to a negligible increase in the response rate against survivin and MAGE-3, suggesting that they might be Treg independent TAAs (data not shown). A limitation of our study and many others is that besides sipuleucel-T (PAP) and PROSTVAC-VF (PSA) the clinical relevance of common tumor antigens remains
to be confirmed in clinical trials. A proven T-cell response to certain TAAs does not necessarily translate into clinical benefit since there might be other, even more relevant tumor antigens that we are not aware of today. Also, during the process of boosting T-cell responses toward known TAAs certain tumor clones might be se-
ANTIGEN SPECIFIC T-CELL RESPONSES AGAINST TUMOR ANTIGENS
lected to lose expression of these tumor antigens, thereby advancing to a secondary level of immune evasion.28 Concerning the optimal timing of immunotherapy during the course of PC progression we detected more TAA specific effector T cells and increased strength of spontaneous T-cell responses in patients with castration resistant disease than in men with lower risk disease (fig. 3). Thus, CRPC seems to be a valid target for recent trials such as IMPACT (IMunotherapy Prostate Adeno Carcinoma Treatment).10 However, in our setting we have not yet had the opportunity to test patients with androgen sensitive metastatic disease. At least in animals the combination of androgen ablation and Treg targeted immune therapy or PROSTVAC seems promising.29,30
1465
CONCLUSIONS Tumor specific effector T cells were detected in most patients with PC, especially in men with castration resistant disease. Since T-cell responses against prostate specific antigens strongly increased after Treg depletion, our results indicate that immunotherapy could be enhanced by preventing Treg accumulation.
ACKNOWLEDGMENTS Mariana Bucur, The German Cancer Research Center, provided assistance. Jesco Pfitzenmaier and Dirk Jäger provided advice.
REFERENCES 1. Jemal A, Siegel R, Xu J et al: Cancer statistics, 2010. CA Cancer J Clin 2010; 60: 277.
12. Shevach EM: Fatal attraction: tumors beckon regulatory T cells. Nat Med 2004; 10: 900.
2. Oh WK and Kantoff PW: Management of hormone refractory prostate cancer: current standards and future prospects. J Urol 1998; 160: 1220.
13. Miller AM, Lundberg K, Ozenci V et al: CD4⫹CD25high T cells are enriched in the tumor and peripheral blood of prostate cancer patients. J Immunol 2006; 177: 7398.
3. Matera L: The choice of the antigen in the dendritic cell-based vaccine therapy for prostate cancer. Cancer Treat Rev 2010; 36: 131. 4. Harzstark AL and Small EJ: Immunotherapeutics in development for prostate cancer. Oncologist 2009; 14: 391. 5. Wang X, Yu J, Sreekumar A et al: Autoantibody signatures in prostate cancer. N Engl J Med 2005; 353: 1224. 6. Tan EM and Zhang J: Autoantibodies to tumorassociated antigens: reporters from the immune system. Immunol Rev 2008; 222: 328. 7. Mercader M, Bodner BK, Moser MT et al: T cell infiltration of the prostate induced by androgen withdrawal in patients with prostate cancer. Proc Natl Acad Sci U S A 2001; 98: 14565. 8. Sanda MG, Smith DC, Charles LG et al: Recombinant vaccinia-PSA (PROSTVAC) can induce a prostate-specific immune response in androgen-modulated human prostate cancer. Urology 1999; 53: 260. 9. Vieweg J, Su Z, Dahm P et al: Reversal of tumor-mediated immunosuppression. Clin Cancer Res 2007; 13: 727s.
21. Xue BH, Zhang Y, Sosman JA et al: Induction of human cytotoxic T lymphocytes specific for prostate-specific antigen. Prostate 1997; 30: 73. 22. Correale P, Walmsley K, Nieroda C et al: In vitro generation of human cytotoxic T lymphocytes specific for peptides derived from prostate-specific antigen. J Natl Cancer Inst 1997; 89: 293.
14. Yokokawa J, Cereda V, Remondo C et al: Enhanced functionality of CD4⫹CD25(high)FoxP3⫹ regulatory T cells in the peripheral blood of patients with prostate cancer. Clin Cancer Res 2008; 14: 1032.
23. Melief CJ and van der Burg SH: Immunotherapy of established (pre)malignant disease by synthetic long peptide vaccines. Nat Rev Cancer 2008; 8: 351.
15. Vergati M, Cereda V, Madan RA et al: Analysis of circulating regulatory T cells in patients with metastatic prostate cancer pre- versus post-vaccination. Cancer Immunol Immunother 2011; 60: 197.
24. Drake CG: Prostate cancer as a model for tumour immunotherapy. Nat Rev Immunol 2010; 10: 580.
16. D’Amico AV, Whittington R, Malkowicz SB et al: Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer. JAMA 1998; 280: 969. 17. Feuerer M, Beckhove P, Bai L et al: Therapy of human tumors in NOD/SCID mice with patientderived reactivated memory T cells from bone marrow. Nat Med 2001; 7: 452. 18. Bonertz A, Weitz J, Pietsch DH et al: Antigenspecific Tregs control T cell responses against a limited repertoire of tumor antigens in patients with colorectal carcinoma. J Clin Invest 2009; 119: 3311.
10. Kantoff PW, Higano CS, Shore ND et al: Sipuleucel-T immunotherapy for castration-resistant prostate cancer. N Engl J Med 2010; 363: 411.
19. Peshwa MV, Shi JD, Ruegg C et al: Induction of prostate tumor-specific CD8⫹ cytotoxic T-lymphocytes in vitro using antigen-presenting cells pulsed with prostatic acid phosphatase peptide. Prostate 1998; 36: 129.
11. Kantoff PW, Schuetz TJ, Blumenstein BA et al: Overall survival analysis of a phase II randomized controlled trial of a Poxviral-based PSA-targeted immunotherapy in metastatic castration-resistant prostate cancer. J Clin Oncol 2010; 28: 1099.
20. Machlenkin A, Azriel-Rosenfeld R, Volovitz I et al: Preventive and therapeutic vaccination with PAP-3, a novel human prostate cancer peptide, inhibits carcinoma development in HLA transgenic mice. Cancer Immunol Immunother 2007; 56: 217.
25. Di Paolo NC, Tuve S, Ni S et al: Effect of adenovirus-mediated heat shock protein expression and oncolysis in combination with low-dose cyclophosphamide treatment on antitumor immune responses. Cancer Res 2006; 66: 960. 26. Salem ML, Kadima AN, El-Naggar SA et al: Defining the ability of cyclophosphamide preconditioning to enhance the antigen-specific CD8⫹ T-cell response to peptide vaccination: creation of a beneficial host microenvironment involving type I IFNs and myeloid cells. J Immunother 2007; 30: 40. 27. Dannull J, Su Z, Rizzieri D et al: Enhancement of vaccine-mediated antitumor immunity in cancer patients after depletion of regulatory T cells. J Clin Invest 2005; 115: 3623. 28. Melief CJ: Cancer immunotherapy by dendritic cells. Immunity 2008; 29: 372. 29. Akins EJ, Moore ML, Tang S et al: In situ vaccination combined with androgen ablation and regulatory T-cell depletion reduces castration-resistant tumor burden in prostate-specific pten knockout mice. Cancer Res 2010; 70: 3473. 30. Arredouani MS, Tseng-Rogenski SS, Hollenbeck BK et al: Androgen ablation augments human HLA2.1restricted T cell responses to PSA self-antigen in transgenic mice. Prostate 2010; 70: 1002.