Leukemia-associated antigen-specific T-cell responses following combined PR1 and WT1 peptide vaccination in patients with myeloid malignancies

The graft-versus-leukemia (GVL) effect following allogeneic stem-cell transplantation (SCT) is evidence that T lymphocytes can eliminate leukemia. The successful identification of a range of leukemia antigens such as proteinase 3 (PR3) and Wilms tumor-1 (WT1) has stimulated efforts to induce leukemia-specific T-cell responses to these antigens using peptide vaccines. We previously showed that PR1, a 9–amino acid HLA-A*0201–restricted peptide derived from PR3, induced myeloid leukemia–specific cytotoxic T-lymphocyte (CTL) responses that selectively killed leukemic CD34⁺ cells.^1,2  PR1-specific CD8⁺ T cells with a memory phenotype occur at low frequencies in healthy individuals and at higher frequencies in patients with leukemia,^3,,–6  suggesting that it should be relatively easy to boost these immune responses with vaccination. Highly encouraging preliminary data from a phase 1/2 study evaluating PR1 vaccination in patients with myeloid leukemias were presented at the annual meeting of the American Society of Hematology in 2004.⁷  WT1 protein, which is overexpressed in a wide range of malignancies including myeloid leukemias and myelodysplasia (MDS), is another attractive vaccine candidate.^8,,,,,,–15  CTLs specific for WT1 are selectively cytotoxic to myeloid leukemias.^16,–18  We and several investigators have reported the occurrence of WT1-specific CTLs in patients with cancers, myeloid and lymphoid leukemias, and healthy volunteers.^{4,5,19,,–22}  More recently, small clinical studies have demonstrated the feasibility and potential efficacy of WT1 peptide vaccination in humans.^23,–25 

Since immune responses against leukemia are often directed against multiple antigens,^4,5,19  there is a risk that targeting a single leukemia antigen may result in immunologic pressure against expression of the parent protein, resulting in the selection of antigen-loss variants. We therefore used a combined PR1 and WT1 peptide vaccine in an attempt to improve the probability of generating a sustained immune response against MDS and leukemia. We report here that following vaccination CD8⁺ T-cell responses against PR1 or WT1 were detected in all patients. The emergence of PR1- or WT1-specific CD8⁺ T cells was associated with a significant reduction in leukemia load as assessed by WT1 mRNA expression. We propose that an immunotherapeutic approach to vaccinate using a combination of PR1 and WT1 peptides will improve the likelihood of immune responses against MDS and leukemia.

HLA-A*0201–positive patients with acute myeloid leukemia (AML) in complete remission, chronic myeloid leukemia (CML) in chronic phase, and MDS (refractory anemia or refractory anemia with ringed sideroblasts) were eligible for this phase 1 study, assessing the safety and efficacy of a combination of PR1 and WT1 peptides in Montanide adjuvant, administered with granulocyte-macrophage colony-stimulating factor (National Institutes of Health [NIH] Protocol no. 06-H-0062). Additional inclusion criteria included (1) age ranging from 18 to 85 years, (2) unsuitable for allogeneic SCT, (3) marrow cellularity of 20% or higher, (4) no history of corticosteroid treatment within 14 days prior to enrolment, (5) absence of serologic antibody against proteinase-3 or antineutrophil cytoplasmic antibodies (ANCAs), (6) no previous history of Wegener granulomatosis, and (7) predicted survival of 28 days or more. The study was reviewed and approved by the Institutional Review Board of the National Heart, Lung, and Blood Institute. After written informed consent was obtained in accordance with the Declaration of Helsinki, patients received a single subcutaneous dose of PR1:169-177 (0.5 mg) and WT1 126-134 (0.2 mg) vaccination in Montanide ISA-51 VG (Seppic, Fairfield, NJ). Granulocyte-macrophage colony-stimulating factor (GM-CSF, sargramostim; Berlex Laboratories, Richmond, CA) was administered subcutaneously as 2 separate 100-μg injections in the same region as each peptide vaccine dose. Following vaccination, patients were reviewed weekly as outpatients for 4 weeks.

The dose of PR1 peptide was based on a phase 1 toxicity study of PR1 peptide vaccination at the M. D. Anderson Cancer Center (Houston, TX) where patients received 3 escalating PR1 peptide dose levels of 0.25, 0.5, and 1.0 mg. All peptide doses were well tolerated and a maximum tolerated dose was not reached. The investigators chose a PR1 peptide dose of 0.5 mg for their subsequent study, presented at the annual meeting of the American Society of Hematology in 2004.⁷  Similarly, in a phase 1 study of WT1 vaccination a dose of 0.2 mg WT1 peptide was administered with minimal toxicity.²³ 

WT1:126–134 (RMFPNAPYL) and PR1:169–177 (VLQELNVTV) peptides were synthesized to GMP grade by NeoMPS (San Diego, CA). The Pharmaceutical Development Section of the Pharmacy Department (NIH Clinical Center) reconstituted and vialed the peptides and provided quality assurance, IND no. 12632. Peptides were stored in dimethyl sulfoxide (DMSO) at −70°C and thawed on the day of injection. A water-in-oil emulsion vaccine was then prepared, consisting of the peptide (aqueous phase) and the adjuvant Montanide ISA 51 VG (oil phase), by combining equal parts of the peptide and the adjuvant. The emulsions formed were shown to be stable for at least a 3-hour time period. The vaccines were administered in the deep subcutaneous tissue at 2 different sites, and GM-CSF was administered subcutaneously as 2 separate 100-μg injections in the same region as each peptide vaccine dose.

At each outpatient visit, patients were evaluated for toxicities according to the National Cancer Institute Common Toxicity Criteria and independently reviewed by the Data Safety Monitoring Board (DSMB). The decision to discontinue therapy for each patient was based on the observation of grade 3 or higher toxicity.

To determine the immunogenicity of the vaccine protocol, peripheral blood mononuclear cells (PBMCs) obtained before and weekly after vaccination were stained with allophycocyanin (APC)–conjugated PR1/HLA-A*0201 (Beckman Coulter, Fullerton, CA), CMVpp65₄₉₅/HLA-A*0201 (Beckman Coulter), and WT1/HLA-A*0201 tetramers (NIH tetramer facility). Sample staining was performed using 1 × 10⁶ PBMCs in 50 μL 1% fetal calf serum/phosphate-buffered saline (FCS/PBS). Tetramers were added for 20 to 30 minutes at 37°C. Cells were washed once in 1% FCS/PBS and then stained with a titrated panel of directly conjugated antibodies to CD3, CD8, CD27, and CD45RO (all from Beckman Coulter, Miami, FL). Fluorescein isothiocyanate (FITC), phycoerythrin (PE), peridinin chlorophyll protein (PerCP), and PE-Cy7 were used as fluorophores. The lymphocytes were then washed in 1% BSA in PBS, and resuspended in 1% paraformaldehyde in PBS. A minimum of 0.5 × 10⁶ gated cells were acquired. Flow cytometry was performed on an LSR II flow cytometer (BD Biosciences, San Jose, CA) using FacsDiva software (BD Biosciences).

Intracellular cytokine detection was performed as described previously.²²  In brief, 1 × 10⁶ PBMCs were loaded with or without test peptides (0.1 μM). After 2 hours, 10 μg/mL brefeldin A (Sigma) was added. After an additional 4 hours, CD3⁺CD8⁺ T cells were stained with an anti-CD3 PerCP-conjugated antibody and anti-CD8 PE-conjugated antibody, fixed/permeabilized, and then stained with an anti–IFN-γ FITC conjugate (all BD/Pharmingen, San Diego, CA).

All samples for qRT-PCR were blinded. RNA was isolated from a minimum of 1 × 10⁶ PBMCs using RNeasy mini kits (Qiagen, Valencia, CA). cDNA was synthesized using the Advantage RT-for-PCR kit (Clontech, Mountain View, CA). ABL expression was used as the endogenous cDNA quantity control for all samples²⁶ ; its expression was measured using 300 nM primers and 200 nM probe.²⁷  Expression of WT1 was measured using 500 nM primers and 200 nM probe.¹⁵  All reactions by qRT-PCR using the ABI PRISM 7900 sequence detection system (Applied Biosystems, Foster City, CA) were performed in triplicate in 10 μL volume using standard conditions with 40 cycles of amplification. Two patients with known major breakpoint BCR-ABL fusion gene transcripts were assessed according to an established minimal residual disease (MRD) detection assay using recommended standardized probe and primers.²⁸  Both WT1 and BCR-ABL qRT-PCR reactions could consistently detect one leukemic cell in 1 000 000 nonleukemic cells.²² 

The relationship between CD8⁺ T-cell response to vaccination and MRD was analyzed by Fisher exact test. Spearman ρ was used to correlate antigen-specific CD8⁺ T-cell frequencies quantified by peptide/HLA-A*0201 tetramer and intracellular cytokine staining. Prism 4.00 for Windows software (GraphPad Software, San Diego, CA) was used for statistical analyses. P values of .05 or less were considered statistically significant.

Eight patients were enrolled in this study and their clinical characteristics are presented in Table 1. All patients completed 4 weeks of follow-up per protocol, to monitor toxicity and immunologic responses. All remain alive at a median follow-up of 252 days (range, 105-523 days). No grade 3 or higher toxicities were observed. Grade 1 erythema, pain, or swelling was noted in 8 of 8 patients at the PR1 and 2 of 8 patients at the WT1 injection site. Patient 5 developed systemic toxicity with transient chest pain within 30 minutes of vaccination, believed to be GM-CSF related and did not require intervention. Patient 4 was vaccinated 6 months following a second unrelated allogeneic SCT, and vaccination did not induce graft-versus-host disease. Of note, no cases of autoimmune reactions or cytopenias were seen. Patients were monitored for induction of anti–proteinase 3 and ANCA, and they all remained negative throughout the length of follow-up. These results indicated that the combined PR1 and WT1 vaccine is safe.

Unstimulated PBMC samples from 8 HLA-A*0201–positive patients with myeloid malignancies enrolled in this study were analyzed for circulating CD8⁺ T cells specific for PR1 and WT1 using fluorescence-activated cell sorting (FACS) staining with PR1/HLA-A*0201 and WT1/HLA-A*0201 tetramers. A significant CD8⁺ T-cell immune response to the vaccine was defined as the emergence of detectable PR1- or WT1-specific CD8⁺ T cells when the prestudy analysis found no response or a 2-fold increase in frequencies when responses were present before vaccination. This definition was a stringent modification of that used in previous vaccination studies.²⁵  The longitudinal tetramer analysis of a patient with a robust response to vaccination, patient 8, is illustrated in Figure 1. Prior to vaccination, we found CD8⁺ T-cell response to PR1 in 2 of 8 (median, 0.03%; range, 0%-0.21%) and to WT1 in 2 of 8 (median, 0.03%; range, 0%-0.16%) patients. Following vaccination, a significant CD8⁺ T-cell response to PR1 was noted in 7 of 8 patients (median, 0.34%; range, 0.04%-0.48%), to WT1 in 5 of 8 patients (median, 0.29%; range, 0%-0.42%), and to one or both antigens in 8 of 8 patients (Table 1). Responses to PR1 and WT1 vaccination were detectable as early as 1 week after vaccination.

We next analyzed CD8⁺ T-cell responses to PR1 and WT1 by intracellular interferon-γ (IC IFN-γ) staining. A response was considered positive if the percentage of peptide-specific IFN-γ–producing CD8⁺ T cells was 2-fold or higher compared with the percentage of IFN-γ–producing CD8⁺ T cells in the absence of peptide and if there was a minimum of 0.05% peptide-specific IFN-γ–producing CD8⁺ T cells in 10⁶ PBMCs (after subtracting the percentage of IFN-γ–producing CD8⁺ T cells in the absence of peptide). The longitudinal IC IFN-γ assay analysis of patient 8 is illustrated in Figure 2A-C. Prior to vaccination, CD8⁺ T cells specifically producing IFN-γ when exposed to PR1 were detected in 2 of 7 (median, 0.02%; range, 0%-0.14%) and to WT1 in 2 of 8 (median, 0.02%; range, 0%-0.10%) patients. Following vaccination, CD8⁺ T cells specifically producing IFN-γ when exposed to PR1 were noted in 7 of 8 patients (median, 0.18%; range, 0.0%-0.42%), to WT1 in 5 of 8 patients (median, 0.12%; range, 0.03%-0.45%). Frequencies of PR1- and WT1-specific CD8⁺ T cells by tetramer staining and IC IFN-γ assay were highly correlated (P < .001) (Figure 2D,E).

Three of 8 patients (patients 1, 4, and 7) taking part in this study were cytomegalovirus (CMV) seropositive. Following vaccination, the frequencies of CMVpp65 CD8⁺ T cells in these patients were unchanged and CMVpp65 CD8⁺ T cells remained undetectable in CMV-seronegative patients, suggesting that the observed PR1- and WT1-specific CD8⁺ T-cell responses were antigen driven and not a consequence of nonspecific inflammatory response to Montanide or GM-CSF (Figure 3).

To assess the potential antileukemia effect of vaccination, WT1 expression was used as a surrogate marker of MRD. WT1 expression was detected in 7 of 8 patients prior to vaccination (median, 0.0018 WT1/ABL; range, 0-0.0370 WT1/ABL). Patient 5 had a diagnosis of Philadelphia-positive AML, and BCR-ABL expression was used to assess his disease response to vaccination. The antileukemia effect of vaccination was assessed by monitoring the kinetics of WT1- and PR1-specific CD8⁺ T-cell response and MRD for each weekly time point after vaccination. In 3 of 6 evaluable patients (patients 1, 3, and 8), the emergence of PR1 or WT1⁺CD8⁺ T cells was associated with a 2-log or more reduction in WT1 expression. In 2 of these patients (patients 1 and 8) WT1transcripts became undetectable, suggesting a true PR1- or WT1-driven antileukemia effect. Conversely, loss of PR1 or WT1⁺CD8⁺ T-cell responses was associated with reappearance of WT1 transcripts (P < .01) (Figure 3). Patient 4 also had undetectable WT1 transcripts on weeks 2 and 3 after vaccination, associated with PR1 and WT1 CD8⁺ T-cell responses. However, we are interpreting the data in this patient with caution as in this case fewer than 1 million PBMCs were available for RNA extraction and qRT-PCR. In patients 5 and 6, the emergence of PR1⁺CD8⁺ T-cell response was not associated with a reduction in WT1 transcript (Figure 3). These patients had a clinical relapse 198 and 145 days following completion of vaccination, respectively. Patient 7, who was in complete molecular remission following imatinib therapy, did not have reliably quantifiable WT1 and BCR-ABL expression in samples obtained before or after vaccination, and the relationship between vaccine-specific CD8⁺ T-cell emergence and MRD could not be ascertained. Disease status and follow-up for each patient are included in Table 1.

In this phase 1 safety study, 8 HLA-A*0201–positive patients with myeloid malignancies received a single dose of a combination of PR1 and WT1 vaccines. The vaccine combination was found to be safe with minimal toxicity. Notably, there was no evidence of hematopoietic damage such as cytopenia, no autoimmune phenomena were seen, and all patients survived at a median of more than 8 months after treatment.

We used 2 independent assays to assess the immune response to PR1 and WT1 vaccination. A tetramer analysis was performed to enumerate the number and phenotype of T cells. We confirmed that these antigen-specific CD8⁺ T cells are functional using intracellular IFN-γ assay, which enables the determination of the number of PBMCs capable of secreting IFN-γ when stimulated with the relevant antigen. CD8⁺ T-cell responses against PR1 or WT1 could be detected prior to vaccination, supporting the natural immunogenicity of these antigens. We determined that the PR1:169-177 and WT1:126-134 epitopes are immunogenic and can elicit CD8⁺ T-cell responses that are detectable in the peripheral blood following vaccination. Immune responses against PR1 or WT1 could be detected in 8 of 8 patients, and in 4 patients CD8⁺ T-cell responses were detected against both antigens. Of note, preexisting immune response against one antigen did not appear to inhibit the induction of responses to the other epitope. The observed PR1- and WT1-specific CD8⁺ T-cell responses were antigen driven and not a consequence of nonspecific inflammatory response to Montanide or GM-CSF, as no significant change in the absolute lymphocyte count or frequencies of CMVpp65 CD8⁺ T cells was seen following vaccination. The benefit of using more than one epitope is therefore manifold. Naturally occurring leukemia-specific CD8⁺ T-cell responses are frequently multiepitopic, a feature that may compensate for the low frequency and amplitude of individual responses.^4,5,19,21  Such polyclonal responses would be valuable for maintaining a sufficient pool of T cells specific for leukemia-associated antigens. Moreover, since targeting a single leukemia antigen could induce antigen-negative variants that escape immune regulation, vaccination against more than one antigen should decrease this risk.

CD8⁺ T-cell responses to PR1 or WT1 were seen as early as 1 week after vaccination. These results are in keeping with work by Oka et al who observed, in a patient with MDS, a significant increase in WT1-specific CTLs within a week of a single dose of WT1 vaccination, followed by a rapid reduction in leukemic blast cells.²⁴  This suggests the expansion was derived from preexisting memory CD8⁺ T cells. Our phenotype data support a memory origin of PR1- and WT1-specific T-cell expansions (data not shown). It is interesting that vaccine-induced CD8⁺ T-cell responses were short lived. The decreases in response over time mayhave been due to the exhaustion of the CD27⁻ effector memory pool,^29,30  activation-induced cell death, or induction of regulatory T cells (T_REGs). However, we were unable to detect a significant difference in the frequency of T_REGs before and after vaccination (data not shown).

The ability of PR1- and WT1-specific CD8⁺ T cells detected by tetramer staining and IC IFN-γ production to mediate in vivo antileukemia or anti-MDS cytotoxicity was assessed indirectly by correlating the emergence of PR1- or WT1-specific CD8⁺ T cells with the leukemia load in the PB. WT1 has been shown to be a reliable marker of MRD in myeloid malignancies.^12,,–15,31  In 3 of 6 evaluable patients, emergence of PR1- or WT1-specific CD8⁺ T cells coincided with significant reduction or complete disappearance of WT1 gene expression in PB, whereas loss of PR1 or WT1 response was associated with an increase in WT1 expression in PB. Two patients who had detectable CD8⁺ T-cell responses to PR1 but failed to have a reduction in MRD relapsed 3 to 6 months following completion of vaccination. Although in this study we were unable to assess direct ex vivo cytotoxicity assays of vaccine-induced PR1- and WT1-specific CD8⁺ T cells due to limited sample availability, the direct cytotoxicity of PR1- and WT1-specific CTLs against leukemia CD34⁺ stem cells has been previously shown by our group and others.^2,3,16,18 

In conclusion, although this study was primarily designed to assess the safety of a combination of PR1 and WT1 vaccines in a small group of patients with limited clinical follow-up, it clearly shows that a vaccine approach using a combination of PR1 and WT1 vaccines is safe and can elicit immunologic responses associated with a reduction in WT1 expression in patients with myeloid leukemia. In addition, as the responses elicited seem to be short lived, maintenance of sustained or at least repetitive responses may require frequent boost injections. Based on these results, we have initiated a phase 2 study of repeated vaccination with PR1 and WT1 peptides in patients with myeloid malignancies.

Contribution: K.R. wrote the clinical protocol, supervised the clinical trial as principal investigator, recruited patients, provided clinical care, collected data, designed and performed experiments, analyzed data, and wrote the paper; A.S.M.Y. and S.M. performed experiments and commented on the paper; B.N.S. advised on statistical analysis and commented on the paper; L.M. provided clinical care and collected patient data; J.S. provided clinical care; B.J. performed experiments; C.B. oversaw the regulatory submission of the clinical protocol and collected data; A.J.B. supervised the clinical trial and laboratory study, and commented on the paper.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Katayoun Rezvani, National Heart Lung Blood Institute, Bldg 10, Hatfield CRC, Rm 3–5410, 10 Center Dr, MSC 1202, Bethesda, MD 20892-1201; e-mail: rezvanik@nhlbi.nih.gov.