Combination Gemcitabine and WT1 Peptide Vaccination Improves Progression-Free Survival in Advanced Pancreatic Ductal Adenocarcinoma: A Phase II Randomized Study

Pancreatic cancer remains one of the most lethal malignancies. Despite the development of novel diagnostic methods and therapeutic agents, most pancreatic cancer patients are diagnosed with unresectable advanced disease and succumb to disease within 1 year (1). Since 1997, gemcitabine (GEM) monotherapy has been the first-line therapy for unresectable, locally advanced and metastatic pancreatic ductal adenocarcinoma (PDAC; ref. 2). The combination of GEM with a variety of conventional drugs or novel molecular targeted agents has generally shown no substantial survival advantage compared with GEM alone (3). Two chemotherapeutic regimens of fluorouracil, leucovorin, irinotecan, and oxaliplatin (FOLFIRINOX) and GEM plus nanoparticle albumin-bound paclitaxel (nab-PTX), both of which showed longer survival benefit than GEM alone, changed the standard of care against advanced PDAC (4, 5). Survival of patients treated with these regimens, however, is marginal and has remained extremely poor over the past two decades, with 5-year survival under 10%. Novel therapeutic strategies are urgently needed to improve prognosis.

Cancer immunotherapy induces or enhances the preexisting host antitumor immune response. Since the discovery of the first tumor-associated antigen (TAA), MAGE-A1, many TAAs have been identified and used as targets for cancer immunotherapy. One of the most promising TAAs is the Wilms' tumor gene 1 (WT1), which was ranked as the top antigen among 75 TAAs (6). The WT1 gene was first isolated as a tumor suppressor gene responsible for Wilms' tumor (7). Subsequent studies, however, indicated that WT1 can play oncogenic roles during tumorigenesis, such as promotion of growth, inhibition of differentiation, resistance to cell death, and promotion of tumor angiogenesis (8–10, 11). Further, wild-type WT1 is overexpressed in various human cancers including PDAC and is a poor prognostic marker (12–15). In addition to these oncogenic functions, WT1 is immunogenic. The WT1 protein elicits cellular and humoral immune responses in vitro and in vivo (16, 17). Because WT1 is expressed in not only cancer cells but also vascular endothelial cells in the tumor microenvironment (10), WT1-expressing tumor vessels, as well as WT1-expressing cancer cells, can be targeted by WT1-specific immunity. We and others have demonstrated promising clinical effects of targeting WT1 for immunotherapy in advanced malignancies (11, 17–21).

GEM, a cytotoxic agent, also possesses immune-modulating functions, such as increase in antigen cross-presentation and selective inhibition of myeloid-derived suppressor cells (22–24), and enhances the expression of WT1 in pancreatic cancer cells, sensitizing them to WT1-specific cytotoxic T lymphocytes (WT1-CTL; ref. 25). We and others previously reported that GEM plus a WT1-targeting cancer vaccine is well tolerated and has promising and synergistic clinical effects in advanced pancreatic cancer (26–28).

In this study, we designed an open-label, randomized phase II study to investigate the clinical efficacy of GEM plus a WT1 peptide vaccine (WT1 vaccine) compared with GEM monotherapy as first-line therapy for patients with unresectable advanced PDAC. This study also aimed to demonstrate both the immunogenicity of the WT1 vaccine in pancreatic cancer patients and the synergistic effects of GEM plus the WT1 vaccine.

This open-label, randomized phase II study was conducted at seven medical centers in Japan. Randomization was centrally performed at a 1:1 ratio using the minimization method with the following balancing factors: extent of disease [UICC-stage III (locally advanced), stage IV (metastatic), or recurrent disease after surgery], primary tumor localization (head or body/tail), liver metastasis (yes or no), HLA-A locus typing (HLA-A*02:01 or HLA-A*24:02), and institution. The primary endpoint was overall survival (OS). Secondary endpoints were progression-free survival (PFS), disease control rate, safety, quality of life (QOL), and WT1-specific immune response. The study was approved by the independent ethics committee or institutional review board of each center, was conducted in accordance with the ethical principles of the Declaration of Helsinki, and was registered in the University Hospital Medical Information Network Clinical Trials Registry as UMIN000005248.

Human leukocyte antigen (HLA)-A*02:01– or A*24:02–positive patients with histologically or cytologically confirmed locally advanced or metastatic PDAC, or recurrent disease after surgery, were eligible. Patients were ages ≥ 20 years with a Karnofsky performance status (KPS) of 80% to 100%; had adequate hematologic, hepatic, and renal function; and had a life expectancy of at least 3 months. Patients with a KPS of 70% were also included but only if KPS was decreased by poorly controlled cancerous pain at enrollment. All patients provided written informed consent.

All patients received GEM at a dose of 1,000 mg/m² intravenously over 30 minutes on days 1, 8, and 15 of a 28-day cycle. Patients allocated to GEM plus WT1 vaccine (GEMWT1) were intradermally administered WT1 vaccine at six different sites (bilateral upper arms, lower abdomen, and femoral regions) on days 1 and 15 of a 28-day cycle. Patients received the study treatment until any of the following occurred: disease progression, discontinuation due to toxicity, withdrawal of consent, or loss to follow-up. Patients in the GEMWT1 group were permitted to continue the study treatment beyond initial progressive disease if they were considered to demonstrate clinical benefit by the investigators (for example, continuing symptom or disease control despite radiological progression). Patients in the GEM group were permitted to receive WT1 vaccine with or without GEM as the second-line or later treatment, but only after disease progression.

The WT1 vaccine was a water-in-emulsion product composed of 3 mg of HLA class I–restricted WT1 peptide and Montanide ISA51VG immune adjuvant (SEPPIC). The WT1 peptide sequence was as follows: HLA-A*02:01–restricted 9-mer WT1₁₂₆ peptide RMFPNAPYL (np126) and HLA-A*24:02–restricted modified 9-mer WT1₂₃₅ peptide CYTWNQMNL (mp235). Patients typed as HLA-A*02:01/any (including A*24:02) and HLA-A*24:02/any (except for A*02:01) were administered np126 and mp235, respectively. These peptides were produced according to the Good Manufacturing Practice Guidelines at Peptide Institute (Osaka, Japan). We previously reported details regarding the preparation of WT1 vaccine (17). For the preparation of np126 peptide solution, 3 mg of np126 was dissolved in 350 μL of 5% glucose. For the preparation of mp235 solution, 3 mg of mp235 peptide was dissolved in a small volume of dimethyl sulfoxide (DMSO; Sigma) and then diluted with 350 μL of 5% glucose. This peptide solution was finally emulsified with an equal weight of Montanide ISA51 adjuvant. The total volume of the WT1 vaccine was 700 μL.

Patients were assessed by physical examinations, complete blood count, and blood chemistry at each administration of GEM. All adverse events (AEs) were assessed according to the Common Terminology Criteria for Adverse Events (CTCAE) version 4.0. Computed tomography was performed every 6 weeks until disease progression. Tumor response was defined by investigator assessments according to the response evaluation criteria in solid tumors (RECIST) version 1.1. QOL was assessed using the Functional Assessment of Cancer Therapy-General (FACT-G) measurement system at baseline and after the second, third, and fourth treatment courses.

The WT1-specific immune response was assessed by (i) delayed-type hypersensitivity (DTH) to the WT1 peptide (assessed for patients in the GEMWT1 group) and (ii) WT1-specific CD8⁺ cytotoxic T lymphocytes (WT1-CTL). WT1-CTLs, which were defined as WT1-tetramer⁺ CD3⁺ CD8⁺ T lymphocytes, were assessed by an HLA-peptide tetramer staining assay (WT1-tetramer assay) with phycoerythrin (PE)-labeled WT1 tetramer for patients in both groups. DTH was assessed every course until discontinuation of study treatment. Ten micrograms of WT1 peptide (np126 for HLA-A*02:01 or mp235 for HLA-A*24:02) diluted by saline or saline alone were intradermally injected at the forearm, and the maximum diameter of erythema and other skin reactions were measured after 48 hours. DTH positivity was defined as a diameter of visible erythema of 1 mm or longer.

For the WT1-tetramer assay, peripheral blood mononuclear cells (PBMC) were collected on day 1 of each course and cryopreserved until use. Frozen PBMCs were thawed and incubated for 1 hour at 37°C in X-VIVO 15 medium (Lonza) supplemented with 10% AB serum (Gemini Bio-Products). Half of the PBMCs were used for the WT1-tetramer assay, and the other half were used for fluorescence minus one to determine background levels of tetramer staining. Thawed PBMCs were incubated with Clear Back (MBL) in phosphate-buffered saline containing 2% FBS and 0.02% sodium azide (FACS buffer) at room temperature for 5 minutes, then stained with WT1-tetramer for 1 hour at 4°C. These cells were then stained with monoclonal antibodies (mAB) against CD3, CD8, and CD4 for 25 minutes at 4°C in the dark, washed three times, and finally resuspended in appropriate quantities of FACS buffer and incubated with 7-AAD (eBioscience) for 5 minutes before analysis. The cells were analyzed on a FACSAria (BD Biosciences). The data were analyzed with FlowJo software (TreeStar). For the WT1-tetramer assay, the following PE-labeled WT1-tetramer and mAbs were used: WT1₁₂₆ peptide/HLA-A*02:01 tetramer or modified WT1₂₃₅ peptide/HLA-A*24:02 tetramer (MBL), anti-CD3-Pacific Blue, and anti-CD4-V500 (BD Biosciences), anti-CD8-FITC (Beckman Coulter).

OS was defined as the time from the date of randomization to the date of death by any cause. PFS was defined as the time from randomization to documented disease progression or death by any cause. Patients who started a second-line treatment before disease progression were censored at the last assessment. OS and PFS were analyzed with the use of the Kaplan–Meier method and the log-rank test. The α level for this study was set at 10% (two-sided). Adverse events were compared by Fisher exact test. FACT-G scores were summarized as mean and standard deviation and compared between groups using t tests. We used the intention-to-treat population, except for patients who were ineligible based on our criteria. The comparison of GEM dose intensity and the number of treatment courses between the two treatment arms was performed using the Mann–Whitney U test.

The study was projected to include 150 patients. Approximately 142 patients enrolled during the 3-year accrual and 1.5-year follow-up period provided 80% power to detect a hazard ratio (HR) of 0.65 compared with a 1-year survival rate of 25% in the GEM group. Interim analysis was prespecified 2 years after the initiation of the study or when 100 patients were enrolled. The objective of interim analysis was to assess the futility of the study using Bayesian posterior probability.

We performed HLA typing in 162 patients to screen for eligibility and identified 123 patients (74.5%) who had HLA-A*02:01 or HLA-A*24:02. A total of 91 patients were enrolled between May 31, 2011, and June 23, 2015. Interim analysis for futility was performed in September 2014, and the independent data monitoring committee (IDMC) recommended continuing the study. However, patient accrual was terminated according to the IDMC's recommendation before accrual of the planned number of patients because enrollment was slow. Forty-five patients were randomly allocated to receive GEMWT1, while the remaining 46 received GEM monotherapy. Eighty-five patients (GEMWT1: n = 42; GEM: n = 43) were finally included in the main efficacy and safety analyses (Fig. 1). Baseline characteristics were well balanced between the two groups, except for recurrent disease, which was not observed in the GEM group (Table 1).

Although the relative dose intensity of GEM was not different between the two groups, the median number of treatment courses in the GEMWT1 group and the GEM group were 6 (range, 1–19) and 3 (range, 1–18), respectively (P = 0.0040). The main reasons for treatment discontinuation were either disease progression [GEMWT1: 37 patients (88.1%); GEM: 34 (79.1%)] or AEs [GEMWT1: 4 (11.9%); GEM: 6 (13.9%)]. One patient in the GEMWT1 group (2.4%) underwent surgery for the complete resection, and three in the GEM group (7.0%) dropped out for reasons unrelated to AEs.

The analysis of OS was based on 85 patients. The median OS was 9.6 months in the GEMWT1 group and 8.9 months in the GEM group (HR, 0.82; 90% CI, 0.57–1.18, P = 0.363; Fig. 2A). GEM plus the WT1 vaccine, however, improved the 1-year OS% compared with GEM monotherapy (35.7% vs. 20.9%). In patients with metastatic disease, the 1-year OS% in the GEMWT1 group was 27.3%, approximately 2.5 times higher than that in the GEM group (11.8%), although the median OS was similar in both groups (GEMWT1: 9.3 months; GEM: 8.7 months; HR, 0.93; 90% CI, 0.61–1.40, P = 0.759; Fig. 2B).

GEM plus the WT1 vaccine improved PFS significantly. The median PFS was 5.2 months in the GEMWT1 group and 3.3 months in the GEM group (HR, 0.66; 90% CI, 0.44–0.98, P = 0.084; Fig. 2C). The PFS rate at 6 months (6-mo PFS%) in the GEMWT1 group was 40.5%, three times higher than that in the GEM group (12.4%). PFS was significantly improved in patients with metastatic disease in the GEMWT1 group. The median PFS and 6-mo PFS% were 3.7 months and 37.8% in the GEMWT1 group, and 2.2 months and 3.9% in the GEM group, respectively (HR, 0.51; 90% CI, 0.32–0.82, P = 0.017; Fig. 2D).

The disease control rate was 52.4% [partial response (PR): n = 6; stable disease (SD): n = 16) in the GEMWT1 group and 37.2% (PR: n = 5; SD: n = 11) in the GEM group (P = 0.194). In most patients with SD, the tumor burden was lower in the GEMWT1 group than in the GEM group (the latter showing slightly increased tumor burden but not reaching progressive disease). However, these differences were not significant (P = 0.0725; Fig. 2E). The response or disease stabilization in the GEMWT1 group was more durable compared with that in the GEM group (Fig. 2F).

In the GEMWT1 group, the most commonly reported (≥50% of patients) AEs of any grade were hematologic toxicities, fatigue, fever, gastrointestinal symptoms, elevation of hepatic enzymes, and hypoalbuminemia (Supplementary Table S1), and the frequently reported (≥10% of patients) grade 3 to 4 clinically significant AEs were leukocytopenia, neutropenia, hepatic-and-biliary tract infection, nausea, and increased AST (Table 2). No treatment-related deaths were reported, and no significant differences in the frequencies of any AEs between the two groups were observed (Table 2). The most frequently reported AE related to the WT1 vaccine was local skin reactions at vaccine injection sites (erythema: 92.9%; induration: 78.6%; pruritus: 54.7%). Most of these were grade 1 or 2 and easily managed, except for a grade 3 ulceration in one case.

We assessed QOL using the FACT-G scale (Supplementary Table S2). At baseline, scores were similar between the two groups (mean total score: GEMWT1 73.1; GEM 72.4; P = 0.830). After treatment, the total score in the GEMWT1 group increased. At the fourth course, the mean total score in the GEMWT1 group was 80.2, which was higher than that in the GEM group (70.4), but the difference was not significant (P = 0.063).

DTH to the WT1 peptide was evaluated in the GEMWT1 group. No patients showed DTH before treatment. Of 42 patients, 38 (HLA-A*02:01: n = 12; HLA-A*24:02: n= 26) were evaluable for DTH at least once after treatment. The remaining four patients could not be assessed because they discontinued the study treatment within the first course due to rapid disease progression. About 50% of patients (HLA-A*02:01: 5/12; HLA-A*24:02: 13/26) showed DTH after treatment. All of the 18 DTH-positive patients showed DTH within four courses of treatment, and 12 (80.0%; HLA-A*02:01: 3/5; HLA-A*24:02: 9/13) exhibited DTH after at least one course of treatment.

Significant prolongation of PFS by GEM plus the WT1 vaccine suggested that WT1-specific immune effector cells induced by the WT1 vaccine affected the clinical outcome. We evaluated the association between the clinical effect, in terms of PFS, and the elicitation of WT1-specific immune responses. This analysis was performed in patients with metastatic disease who received at least two courses of a study treatment (GEMWT1: n = 30; GEM: n = 30) because posttreatment DTH could not be assessed in patients in the GEMWT1 group who had not completed the first course. Half of the patients in the GEMWT1 group (HLA-A*02:01: 5/10; HLA-A*24:02: 10/20) showed DTH.

The PFS curve in the GEMWT1 group could clearly be divided into two separate curves according to the presence or absence of DTH (Fig. 3A). The median PFS and 6-mo PFS% were 6.5 months and 60.0% in DTH-positive patients, and 2.9 months and 17.8% in DTH-negative patients, respectively. PFS in DTH-positive patients was significantly prolonged compared with that in the GEM group (median PFS: 2.8 months; 6-mo PFS%: 4.2%; HR, 0.27; 90% CI, 0.14–0.51), whereas PFS in DTH-negative patients was similar to that in the GEM group (HR, 0.86; 90% CI, 0.49–1.51) (P = 0.001).

We next analyzed the association between PFS and both DTH and well-known prognostic factors in PDAC (29). In the univariate analysis, DTH-positivity in the GEMWT1 group was associated with longer PFS than in the GEM group (HR, 0.25; 95% CI, 0.10–0.62, P = 0.003). These prognostic factors were, therefore, not confounders of DTH, suggesting that DTH-positivity was an independent predictive marker for better PFS following treatment with GEM plus the WT1 vaccine.

We also evaluated the association between OS and DTH (Fig. 3B). The median OS and 1-year OS% were 11.2 months and 40.0% in DTH-positive patients (HR, 0.69; 90% CI, 0.41–1.18), 7.8 months and 20.0% in DTH-negative patients (HR, 1.29; 90% CI, 0.76–2.19), and 8.9 months and 13.3% in the GEM group, respectively (P = 0.229).

We used a WT1 tetramer assay to analyze dynamic changes in WT1-CTLs in patients with metastatic disease. Before treatment, the median percentages of WT1-CTLs in the GEMWT1 and GEM groups were 0.011% and 0.014%, respectively, in HLA-A*02:01 patients (P = 0.880), and 0.035% and 0.044%, respectively in HLA-A*24:02 patients (P = 0.403). Most of the DTH-positive patients in the GEMWT1 group (n = 15) showed an increased percentage of WT1-CTLs after treatment (Fig. 3C and D), and 11 patients (73.3%; HLA-A*02:01: 3/5; HLA-A*24:02: 8/10) showed an increase of five-fold or higher, peaking at the second to fourth course of treatment (Supplementary Table S3). In contrast, in the 14 DTH-negative patients (HLA-A*02:01: 5; HLA-A*24:02: 9) and the 29 patients in the GEM group (HLA-A*02:01: 10; HLA-A*24:02: 19), almost no change in WT1-CTL percentages were seen (Fig. 3C and D), except for three DTH-negative patients who exhibited increases of 5-fold or higher (Supplementary Table S3). DTH positivity was statistically associated with the increase in WT1-CTLs induced by WT1 vaccination (Fisher exact test, P = 0.0092). We further evaluated the association between PFS and increased WT1-CTLs. The PFS curve in the GEMWT1 group was divided into two separate curves according to whether WT1-CTLs exhibited a 5 times or higher increase and less than 5 times increase (including no increase; Fig. 3E). In the patients with increased WT1-CTLs, PFS was significantly prolonged compared with others (P < 0.0001).

The goal of this study was to evaluate the efficacy of GEM plus a WT1 vaccine compared with GEM monotherapy, with OS as the primary endpoint. Unfortunately, the trial failed to meet the primary endpoint of a statistically significant improvement in OS. The 1-year OS% in the GEMWT1 group, however, was about 1.7 times higher than that in the GEM group. PFS in the GEMWT1 group was also longer (HR, 0.66) and 6-mo PFS% was 3.3 times higher than those in the GEM group. The tumor burdens in the GEMWT1 group were reduced, with a longer duration of disease stabilization. This study, however, should be considered underpowered as the sample size was statistically insufficient to evaluate some endpoints. The major reason for the delayed subject recruitment was the success of two novel regimens, FOLFIRINOX and GEM plus nab-PTX (4, 5). Although GEM monotherapy was the standard chemotherapy when this study started, approval of these new treatments in Japan resulted in a rapid change in first-line chemotherapy for advanced PDAC.

Cancer immunotherapy is a promising new strategy for pancreatic cancer. Unfortunately, immunotherapies in mid- to large-sized or late-phase clinical trials failed to show favorable clinical outcomes, which had been expected in some early-phase studies (30, 31). Immune checkpoint inhibitors have also failed to meet expectations thus far (32). Considering the difficulty in developing cancer immunotherapy for pancreatic cancer, our results in this midsized randomized study provide a way to increase antitumor responses in patients.

Compared with GEM monotherapy, GEM plus the WT1 vaccine was well tolerated without any additional serious treatment-related AEs, even in elderly individuals with poor performance status (PS), and did not impair QOL in patients with advanced PDAC. These results are consistent with those of previous studies in which various cancer vaccines were combined with GEM-based chemotherapies (33–35). According to the 2016 guidelines for pancreatic cancer (36), intensive regimens using FOLFIRINOX and GEM plus nab-PTX are recommended for patients with good PS, while GEM monotherapy is still recommended for those with poor PS. However, many patients might not tolerate these toxic multidrug chemotherapies because of multiple prior comorbidities and disease progression. Hence, developing alternative regimens, such as GEM plus the WT1 vaccine, are needed.

In the subgroup analysis of patients with metastatic disease, GEM plus WT1 vaccination significantly prolonged PFS, with the improvement of 1-year OS%, compared with GEM, although it failed to show a significant survival benefit. Therefore, we have to consider its failure mechanism. First, it is possible that the GEMWT1 group contained two subpopulations. In fact, patients in the GEMWT1 group could be divided into two groups based on DTH positivity. Compared with the GEM group, the DTH-positive subgroup demonstrated the significantly prolonged PFS with a much better HR of 0.27. The OS was improved with a HR of 0.69, although this was not statistically significant, probably due to the low power secondary to the small sample size. On the other hand, the PFS in the DTH-negative subgroup was identical to that in the GEM group, and the OS was relatively poor with an HR of 1.29. These results suggested that WT1-specific immune responses were required for prolonging PFS and, thus, improving OS, and that WT1-specific immunogenicity was not achieved in some patients with advanced disease. Another possibility is the crossover effect of WT1 vaccination. Approximately 45% of patients in the GEM group received GEM plus the WT1 vaccine as second-line therapy, while another 45% received other chemotherapy regimens, including S-1 and GEM+S-1 (Supplementary Table S4-1). In 5 patients who survived for 1 year or longer, 4 received GEM plus WT1 vaccination (Supplementary Table S4-2). These results suggest that the addition of the WT1 vaccine to GEM as second-line therapy could also elicit WT1-specific immune responses and contribute to improved survival. As a result, the OS benefit derived from prolonged PFS in the GEMWT1 group could be blurred by the crossover effect of GEM plus the WT1 vaccine as second-line therapy in the GEM group.

Considering the mechanism of action of cancer vaccines, our results are reasonable because the clinical effects of cancer vaccines, which do not directly affect cancer cells, can be expected only after vaccines induce and augment TAA-specific CTLs sufficient to eliminate cancer cells. To assess the T-cell responses elicited by the WT1 vaccine (37), we used the following two assays: (i) DTH to the WT1 peptide, which reflects the in vivo immunologic response and (ii) the tetramer assay, which reflects the frequency of WT1-specific T cells. Approximately 50% of patients in this study were DTH positive, which is consistent with other reports, as well as our own previous phase I study (19–21). In the clinical setting of GEM plus WT1 vaccination, DTH was a useful predictive marker for PFS, independent of other prognostic factors associated with PDAC. DTH positivity was significantly correlated with an increase in WT1-CTLs. These results suggest that DTH is a simple and useful method that qualitatively reflects the induction of functional WT1-CTLs. We did not, however, identify any immunologic biomarkers suitable for the pretreatment determination of either the clinical benefit of the WT1 vaccine or WT1-specific immune responses.

This study had three main limitations. First, patients were selected according to HLA-A loci. HLA molecules affect T cell–mediated immune responses, but the association between HLA-type and PDAC prognosis remains unclear. We selected the patients with specific HLA-A loci because np126 and mp235 of the WT1 peptide are restricted to HLA-A*02:01 and HLA-A*24:02, respectively. Hence, the clinical results of patients with other HLA-types are unknown. Second, observer and subject biases should be considered when interpreting our results because this was an open-label study with investigator assessment. Third, as already discussed, was the lack of statistical power.

Our findings show that WT1-specific immunity can improve the prognosis of PDAC patients. If WT1-specific immunity can be induced more effectively, PFS and OS can be improved in more patients. One method is to coadminister HLA-class II epitopes derived from WT1 (27, 38). Another option is the development of safe and highly effective immune adjuvants to enhance TAA-specific immunity. For example, beneficial effects on the survival of pancreatic cancer patients resulted from both the addition of IMM-101 to GEM (39), as well as the combination of GVAX pancreas and CRS-207 (40). Novel therapeutic strategies to overcome the desmoplastic and immunosuppressive microenvironment of PDAC, which prevents the infiltration of CTLs, are also important (41). Ultimately, the development and clinical application of multimodal cancer therapy that fully mobilizes both conventional cancer treatments and novel immunotherapies will improve the prognosis of more PDAC patients in the future.

In conclusion, the combination therapy of GEM plus WT1 vaccination significantly prolonged PFS and improved 1-year OS%, although not significantly, compared with GEM monotherapy in patients with advanced PDAC, especially those with metastatic disease. These clinical effects were associated with WT1-specific immune responses, providing a proof of concept for addition of WT1 vaccination to GEM. This combination was generally well tolerated without unexpected toxicities and did not impair QOL. These results encourage us to conduct further clinical trials of the WT1 vaccine containing class I and class II peptides in combination with a standard chemotherapy, such as GEM plus nab-PTX.

No potential conflicts of interest were disclosed.

Masanori Kon was not available to confirm coauthorship, but the corresponding author, Sumiyuki Nishida, affirms that Masanori Kon contributed to the article and thus confirms his coauthorship status.

Conception and design: S. Nishida, S. Egawa, S. Koido, S. Homma, Y. Oka, S. Morita, H. Sugiyama

Development of methodology: S. Nishida, F. Fujiki, S. Morita

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): S. Nishida, T. Ishikawa, S. Egawa, S. Koido, H. Yanagimoto, J. Ishii, Y. Kanno, S. Kokura, H. Yasuda, M. Sato, S. Morimoto, F. Fujiki, H. Eguchi, H. Nagano, H. Shimada, K. Ito

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): S. Nishida, T. Ishikawa, S. Egawa, M.S. Oba, S. Morimoto, F. Fujiki, Y. Oka, S. Morita

Writing, review, and/or revision of the manuscript: S. Nishida, S. Egawa, S. Koido, H. Yanagimoto, M.S. Oba, S. Morimoto, Y. Oka, S. Morita

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): S. Nishida, T. Ishikawa, S. Koido, M. Sato, S. Morimoto, A. Kumanogoh, M. Unno

Study supervision: M. Unno, S. Homma

This study was supported, in part, by the Japanese Ministries of Education, Culture, Sports, Science and Technology (grant numbers 21790665 and 15K09050) and the Japanese Foundation for Multidisciplinary Treatment of Cancer.

We thank all the patients who participated in this study, their supporting families, and all referring physicians and the supporting medical staffs at all clinical sites. We also thank S. Tada (Osaka University) for her data management; S. Kobayashi, Y. Oji, A. Tsuboi, N. Hosen, and H. Nakajima (Osaka University); T. Okayama (Kyoto Prefectural University); Y. Otsuka (Toho University); M. Ueno and S. Ohkawa (Kanagawa Cancer Center); S. Ohno, and K. Sano (Teikyo University) for their kind professional support; S. Umeda (Osaka University); S. Tsuchiya (Kyoto Prefectural University); and S. Kuramae, E. Shibuya, and K. Kawamura (Tohoku University) for their kind technical support.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.