In this open-label, randomized, phase IIa study we screened 44 chemotherapy-naive patients with CRPC. Patients with rising prostate-specific antigen (PSA) were closely monitored to detect early biochemical progression. Patients were screened for study eligibility as soon as patients met the criteria for CRPC [32]. Since there is no clear consensus om the correct timing for CRPC treatment, this window was used to screen asymptomatic or minimally symptomatic CRPC patients. Twenty-two of the screened patients were HLA-A-*0201. One of these patients was excluded because a second primary malignancy was detected (Additional file 1: Figure S1). All 21 included patients had histologically confirmed adenocarcinoma of the prostate. Eligible patients had not received any immunotherapy, docetaxel, cabazitaxel or treatment with the RANKL-inhibitor denosumab. Concurrent use of glucocorticoids up to 10 mg per day or a prednisone equivalent was permitted. Patients requiring opioids for cancer-related pain at screening were excluded. Patients had no visceral metastases. Other eligibility criteria were: an Eastern Cooperative Oncology Group (ECOG) performance status grade of 0 or 1; ongoing luteinizing hormone-releasing analogue therapy or status after bilateral orchidectomy; serum testosterone level of < 1.73/l (< 50 ng/dl); absence of active autoimmune diseases, absence of an active viral infection or allergy to shell fish; and no significant laboratory abnormalities (hemoglobin > 5.6 mmol/l (9.0 g/dl); white blood cell count of > 3.0 × 10⁹/l; platelets > 100 × 10⁹/l; serum creatinine < 150 μmol/l; AST/ALT < 3 x ULN, and serum bilirubin < 25 μmol/l, exception Gilbert’s syndrome). Baseline disease sites were assessed using ⁶⁸Ga- prostate-specific membrane antigen (PSMA) PET/CT scans [33], including thin-section diagnostic CT (3 mm) and ferumoxtran-10-enhanced MRIs [34, 35] and regular MRI of bones and lymph nodes. Response evaluation was assessed according to Response Evaluation Criteria In Solid Tumours (RECIST) version 1.1 [36] and the Prostate Cancer Clinical Trials Working Group 2 (PCWG2) criteria [37]. The immune-related response criteria and the iRECIST criteria were used to assess immune unconfirmed progressive disease [38–41]. Response evaluation was assessed by using contrast-enhanced ⁶⁸Ga-PSMA PET/CT scans and ferumoxtran-10-enhanced MRIs at 3 months, and for patients with long-term clinical benefit after 12 and 24 months. Regular follow-up MRI of lymph nodes and bones was performed at 6, 9, 15, 18 and 21 months. Measurable lesions were measured in at least one dimension with longest diameter ≥ 10 mm. Small lesions (longest diameter < 10 mm or pathological lymph nodes with < 15 mm short axis) are considered non-measurable disease. Bone metastases were documented and assessed according to PCWG2 criteria. Patients with absence of disease progression, defined as patients with a radiological complete or partial response or stable disease for > 6 months, were eligible for a maintenance cycle of three biweekly vaccinations. Patients without progressive disease after 12 months were eligible for a final vaccination cycle (Additional file 2: Figure S2A). Baseline characteristics and prior therapies are presented in Table 1.Tab1

Baseline characteristics of patients treated with blood-derived DC vaccinations

Patients with CRPC were randomly assigned in a 1:1:1 ratio to receive CD1c⁺ mDC vaccinations (2–5 × 10⁶ cells per injection; arm A), pDC vaccinations (1–3 × 10⁶ cells; arm B), or combined CD1c⁺ mDC and pDC vaccinations (combiDC; 3–8 × 10⁶ cells; arm C). One cycle of vaccinations consisted of three biweekly vaccinations administered intranodally in a clinically tumor-free lymph node by our expert radiologist or nuclear medicine physician. One to two weeks after the third vaccination a delayed-type hypersensitivity (DTH)-skin test was performed after intradermal administration of 1–10 × 10⁵ cells [42]. Adverse events were defined in accordance with the Common Terminology Criteria for Adverse Events (CTCAE) version 4.0. Primary endpoint of the study was the immunological response after DC vaccinations. Secondary objectives were safety, feasibility, quality of life and clinical efficacy (radiological progression-free survival (rPFS), OS, prostate-specific antigen doubling time (PSAdt), time to opiate use for cancer-related pain, time to SRE, time to decline in WHO/ECOG performance score by ≥1 point and time to the initiation of docetaxel chemotherapy). rPFS was defined as the time from apheresis to radiological progression of soft-tissue lesions or two or more new bone lesions or death from any cause. The event date of the unconfirmed progression was used for calculation of rPFS. OS was defined as the time from apheresis to death from any cause. The PSAdt was calculated according to the Memorial Sloan-Kettering Cancer Center guidelines (http://nomograms.mskcc.org/Prostate/PsaDoublingTime.aspx). An SRE was defined as a pathologic fracture, palliative radiotherapy to a bone lesion, spinal-cord compression or surgery involving bone.

Paired t-tests were performed to evaluate immunological responses before and after vaccination and independent-samples t-tests (Mann-Whitney U tests) were used to evaluate differences between groups. Statistical significance was defined as p < 0.05 (two-tailed significance level). Time-to-event data were evaluated using the Kaplan-Meier method. Statistical significance was evaluated using the two-sided log-rank test and was defined as p < 0.05. Differences between treatment arms were evaluated using one-way ANOVA. Statistical analysis was performed using SPSS® Statistics version 22 software (SPSS Inc., Chicago, IL, USA) and GraphPad Prism 5.03 (GraphPad Software, Inc., San Diego, CA, USA).

CD1c⁺ mDCs and pDCs were manufactured according to Good Manufacturing Practices (GMP). DCs were directly isolated from apheresis products using the fully automated and enclosed immunomagnetic CliniMACS Prodigy isolation system (Miltenyi Biotec, Bergisch-Gladbach, Germany). GMP-grade magnetic bead-coupled antibodies were used, following the manufacturer’s guidelines. For mDC isolation, first, CD19⁺ and CD14⁺ cells were depleted, followed by positive selection of BDCA1⁺ cells with biotin-coated CD1c (BDCA-1) antibodies and anti-biotin coated magnetic beads (arm A). PDC were selected with anti-CD304 (BDCA-4) coupled beads (arm B). When patients were randomized for vaccination with both mDCs and pDCs (arm C), first pDC were selected with anti-CD304 coupled beads, followed by depletion of CD19⁺ and CD14⁺ cells and positive selection of CD1c⁺ cells. mDCs were cultured overnight at a concentration of 1.5 × 10⁶ cells/ml with 800 IU/ml recombinant human GM-CSF in TexMACS GMP medium (both Miltenyi Biotec) supplemented with 2% human serum (Sanquin) and 10 μg/ml keyhole limpet hemocyan (KLH; Immucothel, Biosyn Arzneimittel GmbH) for immunomonitoring purposes. pDCs were cultured overnight at a concentration of 1.5 × 10⁶ cells/ml with 10 ng/ml recombinant human IL-3 in TexMACS GMP medium (both Milteny Biotec) supplemented with 2% pooled human serum. mDCs and pDCs were loaded with HLA-A*0201 binding peptides of NY-ESO-1:157–165 (SLLMWITQC) and MAGE-C2:336–344 (ALKDVEERV) [43] as well as NY-ESO-1 and MUC1 PepTivators (Miltenyi Biotec, Bergisch-Gladbach, Germany)) at a concentration of 1 μM. PepTivators consist of overlapping long peptides that cover the complete protein and bind multiple HLA-types, both MHC class I and II (Additional file 2: Figure S2B).

NY-ESO-1 and MUC1 PepTivators were added during overnight culturing. Thereafter, mDCs and pDCs were activated with premixed protamine HCl (Meda Pharma) and mRNA (gp100, Universitätsklinik Erlangen) for 6 h. Premix ratio 10 μg protamine + 5 μg mRNA, 10 μl premix per ml cell suspension [44]. During the last 3 h of maturation NY-ESO-1 and MAGE-C2 peptides were added at final concentration of 1 μM. This isolation and culture procedure gave rise to mature mDC and pDC meeting the release criteria: sterile, endotoxin level < 7 EU/ml, more than 50% viability, more than 50% purity, expression of CD80 > 50% on pDCs and expression of CD83 > 50% on mDCs. Expression of MHC class I, MHC class II, CD86 and CCR7 was reported, but no release criterium (Additional file 3: Figure S3A-D). Protamine/mRNA complex activated both mDCs and pDCs into mature functional DCs that secrete IFN-α, TNF-α (pDCs only), IL-12p70, and IL-6 (both mDCs and pDCs, but mainly pDCs) (Additional file 3: Figure S3E). Cells were frozen in TexMACS medium containing 10% dimethyl sulfoxide (DMSO; WAK Chemie Medical GmbH) and 40% Albuman (Sanquin), stored at < − 80 °C for max. 2 years and thawed on the day of vaccination. For combined pDC and mDC vaccines, both subsets were pooled in one syringe after thawing.

After apheresis, sufficient amounts of blood-derived DCs could be obtained for at least one vaccination cycle. In two patients randomized for treatment with combiDCs the final CD1c⁺ mDC product did not fulfill the release criteria. Therefore, these patients were vaccinated with pDCs only. Because the primary endpoint of the study was immunological, two extra patients were randomized within the combination arm. In patient pDC-06 the pDC purity was initially only 43%, which increased to 54% after overnight culture and maturation. In patient combiDC-06 CD1c⁺ mDC purity was 41% and thus out-of-specification. Nevertheless, the product was released and administered after accounting for the lower purity by administering at least 2.4 × 10⁶ cells. Consequently, patient received at least the minimum required dose of 2 × 10⁶ CD1c⁺ mDCs per vaccination.

Purity and phenotype of mDCs and pDCs after CliniMACS isolation were determined by flow cytometry with a FACSVerse (BD Biosciences, San Jose, CA, USA) or MACS Quant (Miltenyi Biotec). The following primary monoclonal antibodies and the appropriate isotype or fluorescence minus one controls were used: anti-CD1c-Viobright FITC, anti-BDCA-2-PE, anti-CD20-PE-Vio770, anti-CD123-APC, anti-CD45-APC-Vio770, anti-CD14-VioGreen, anti-FcεRI-VioBlue, anti-CD14-FITC, anti-CD15-PE, anti-CD56-APC, anti-CD3-BioBlue, anti-HLA-ABC-APC, anti-HLA-DR,DP,DQ-APC, anti-CCR7-APC, anti-CD80-APC, anti-CD83-APC and anti-CD86-APC (all Miltenyi Biotec).

DTH challenges were performed 2 weeks after each vaccination cycle to assess TAA-specific immune response in DC vaccinated patients [42, 45]. DCs used for the DTH-skin test were produced accordingly to the vaccinated cells, except that no KLH was added to the culture medium. At four different sites at the patient’s back maximally 5.0 × 10⁵ peptide-loaded blood-derived DCs were injected intradermally. After 48 h, 6 mm punch biopsies were taken. The biopsies were manually cut and half of the tissue was stored at − 150 °C; the other half was cultured as described previously [42]. After 2 to 4 weeks of culturing, skin-test infiltrating lymphocytes (SKILs) were tested for the presence of tumor antigen-specific T cells. SKILs and peripheral blood mononuclear cells (PBMCs) were stained with 1) anti-CD8-FITC and tetrameric PE- and APC-coupled MHC complexes containing NY-ESO-1 (SLLMWITQC), MAGE-C2 (ALKDVEERV) and MUC1 (LLLLTVLTV) HLA-A*0201 epitopes (all Sanquin, Amsterdam, The Netherlands); or 2) anti-CD8-BV421, anti-CD19-FITC and dextrameric PE- and APC-coupled MHC complexes containing the indicated epitopes (all Immudex, Copenhagen, Denmark). Dextrameric HLA-B*0801 (AAKGRGAAL) and tetrameric and dextrameric HIV (SLYNTVATL) were used as a negative control. Cells were analyzed by flow cytometry. To test peptide recognition, SKILs were challenged with autologous PBMCs loaded with the indicated peptides and PepTivators, phorbol myristate acetate (positive control), carcinoembryonic antigen peptide or no peptide (both negative control). Production of interferon-γ (IFN-γ), IL-2, IL-5 and IL-10 was measured in the supernatants after overnight co-culture by cytometric bead array according to the manufacturer’s instructions (BD Biosciences).

Cellular responses against KLH were measured in a proliferation assay. PBMCs were isolated from blood samples after each vaccination. 1 × 10⁵ PBMCs were plated per well of a 96-well tissue culture microplate either in the presence or absence of KLH. After 4 days of culture, 1 μCi/well of tritiated thymidine was added, incorporation of tritiated thymidine was measured in a beta-counter. A proliferation index (proliferation with KLH/proliferation without KLH) of > 2 was considered positive. Antibodies against KLH were measured in the serum of DC vaccinated patients by ELISA. KLH antibodies were detected with mouse anti-human IgG, IgA, or IgM antibodies labeled with horseradish peroxidase. 3,3′,5,5′-Tetramethylbenzidine was used as a substrate. Plates were measured with a microtiter plate reader at 450 nm. An isotype-specific calibration curve for the KLH response was included in each microtiter plate.

Formalin-fixed, paraffin-embedded tissue blocks of prostate biopsies or radical prostatectomy at time of diagnosis were collected from primary treatment centers located in the Netherlands and sections of 4-μm thickness were cut. Slides were deparaffinized using xylene and rehydrated with ethanol. Antigen retrieval was performed by boiling in EnVision™ FLEX target retrieval solution (pH 9, K8004, Dako) for 10 min for MUC1 staining or in Citrate buffer (pH 6, CBB999, ScyTek Laboratories) for 15 min for NY-ESO-1 and MAGE-C2 staining. After cooling down, endogenous peroxidase was blocked using 3% hydrogen peroxidase (76,051,800.1000, EMD Millipore) in PBS (4391.9010, Klinipath). Primary antibodies MUC1 (M0613, clone E26, Dako, dilution: 1/250), NY-ESO-1 (MABC1151, clone D8.38, Merck, dilution: 1/200) and MAGE-C2 (HPA062230, rabbit polyclonal, Merck, dilution: 1/200) were diluted in Normal Antibody diluent (VWRKBD09–999, Immunologic) and were incubated at room temperature for 1 h. Slides were washed between steps with EnVision™ FLEX Wash Buffer (DM831, Dako). Next, incubation with BrightVision poly-HRP-anti-Ms/Rb/Rt IgG (DPVO999HRP, ImmunoLogic) was performed at room temperature for 30 min. Visualization was performed with EnVision™ FLEX DAB Buffered Substrate and EnVision™ FLEX Substrate Buffer (K5207 and SM803; DAKO) for 7 min at room temperature. After dehydration, slides were counterstained with hematoxylin and were enclosed with Quick-D mounting medium (7281, Klinipath). Observed staining was cytoplasmic. Immunoreactivity was evaluated by a pathologist using a semi-quantitative, stepwise scoring system: negative (0% of cells stained), weak (1–10% of cells stained), moderate (11–50% of cells stained) and strong (51 to 100% of cells stained). Representative slides were scanned using the PerkinElmer Vectra (Vectra 3.0.4, PerkinElmer). Testicular or tonsil tissue (positive control) was used for antibody validation (Additional file 4: Figure S4).

In this prospective study twenty-one eligible patients with CRPC were enrolled. Participants were treated with blood-derived DC vaccines from November 2015 until May 2018. Baseline demographic, disease characteristics and prior therapies for hormone-sensitive PCa and CRPC are listed in Table 1. The described results are based on the cut-off date of 6th of March 2019. The median follow-up is 27.2 months (range 10.7–41.2* months). All twenty-one patients, seven per arm, received at least one cycle of three biweekly DC vaccinations and a DTH-skin test. Thirteen patients also received a second cycle and seven patients a third vaccination cycle.