First-in-Class, First-in-Human Study Evaluating LV305, a Dendritic-Cell Tropic Lentiviral Vector, in Sarcoma and Other Solid Tumors Expressing NY-ESO-1

Successful solid tumor immunotherapy requires activation of a broad population of T cells that recognize tumor cells. T cells are activated by antigen-presenting cells, including dendritic cells (DC). DCs are uniquely equipped to process and present antigen to T cells in the context of MHC along with costimulatory molecules, and are effective for stimulating naïve T cells. The engineering of therapeutic technologies that selectively activate DCs and induce them to process and present tumor-specific antigens to CD8⁺ T cells could be an important step toward the development of more effective solid tumor immunotherapies (1).

Modified, third-generation, lentivirus-based vectors (LV) are a novel approach for in vivo DC-targeted vaccination. Unlike many other viral vectors, LV can transduce nondividing DCs, thus inducing the de novo transcription of the encoded antigen. Cytoplasmically expressed antigens can be directly presented to the immune system via MHC-I molecules for induction of CD8⁺ T cells; this direct presentation is hypothesized to be up to 10⁵ times more efficient than the loading of exogenous antigens onto MHC class I via cross-presentation (2). Furthermore, the RNA of LV is able to trigger toll-like receptor (TLR3, TLR7/TLR8) signaling which stimulates the maturation of DCs, increases their expression of costimulatory molecules (e.g., CD40, CD80, CD86), upregulates MHC Class II expression, and improves the interactions of DCs with T cells (3, 4). Thus, LV may be more effective at inducing T-cell responses than immunotherapeutic approaches that rely on exogenous antigens and cross-presentation.

The New York Esophageal Squamous Cell Carcinoma-1 [NY-ESO-1 (CTAG1B)] expressing LV305 is derived from the ZVex platform, which was engineered to deliver antigen-encoding gene(s) selectively to DCs in vivo. The envelope protein of ZVex is derived from the Sindbis virus envelope glycoprotein which interacts with the DC-specific, C-type lectin receptor DC-SIGN (CD209). ZVex is a third-generation LV that is replication-incompetent and integration-deficient due to extended deletion of the vector backbone and integrase inactivation (5). These redundant molecular safety features make persistence and integration of the vector very unlikely and enable safe parenteral injection compared with other third-generation LV currently used for ex vivo gene transfer in cell-based therapies.

LV305 induces the expression of full-length NY-ESO-1 protein (5–7), which is a highly immunogenic cancer testis antigen expressed in up to 40% of non–small cell lung, ovarian, and melanoma cancers, and in greater than 70% of certain sarcoma subtypes [e.g., synovial sarcoma (SS) and myxoid/round cell liposarcoma (MRCL); refs. 8–11]. NY-ESO-1–directed vaccination has obtained some promising clinical results in early phase I/II studies, and adoptive cell therapy has resulted in partial responses (PR) in 60% of NY-ESO-1–expressing sarcomas in small phase I trials (12). In preclinical studies, a single immunization with LV305 transduced DCs in vivo, generated multifunctional CD8 T-cell responses, and controlled the growth of NY-ESO-1–expressing murine tumors (7).

This first-in-human phase I study examined the clinical safety, immunogenicity, and preliminary efficacy of LV305 in patients with advanced sarcoma or other solid tumors expressing NY-ESO-1.

A phase I, open-label trial of LV305 was conducted to evaluate the safety and tolerability of LV305 administered intradermally (ID) in patients with locally advanced, relapsed, or metastatic cancers expressing NY-ESO-1.

Secondary objectives included evaluation of immunogenicity. Exploratory objectives included evaluation of responses every 8 weeks by immune-related response criteria (irRC) modified to use RECIST (v1.1), duration of response, progression-free survival (PFS), time to progression, and biomarkers. Three escalating doses of LV305 in four cohorts were assessed using a 3+3 sequential dose-escalation design, with expansion at the maximum safe dose.

The trial was conducted in accordance with the ethical principles in the Declaration of Helsinki, International Conference on Harmonization guidelines for Good Clinical Practice, and the code of Federal Regulations. The protocol was reviewed by the National Institutes of Health Recombinant DNA Advisory Committee and followed their approved guidelines and was approved by local Institutional Review Boards and Biosafety Committees. All patients provided written, informed consent. This trial is registered on Clinicaltrials.gov (NCT02122861).

Eligible patients were ≥18 years of age with locally advanced, relapsed, and/or metastatic cancer with low tumor burden as defined in Supplementary Table S1. During screening, tumor expression of NY-ESO-1 was determined by IHC (performed by Mosaic Laboratories) on tumor samples collected prior to screening using mouse monoclonal antibody E978 (Sigma). Eligible NY-ESO-1 expression was defined as ≥5% and later amended to include any tumor cell–specific positivity (as determined by a central lab pathologist) to allow enrollment of lower-expressing tumor types. Patients were not required to have HLA typing to determine eligibility.

During dose escalation, eligible NY-ESO-1–positive tumor types included sarcomas (any subtype), melanoma, non–small cell lung cancer (NSCLC), ovarian cancer (including fallopian tube carcinoma), and breast cancer. During expansion at the maximum safe dose, tumor types included sarcomas (SS and MRCL subtypes), melanoma, NSCLC, ovarian cancer (including fallopian tube carcinoma), and urothelial carcinoma of the urinary tract.

Eligible patients must have had an inadequate response, relapse, and/or unacceptable toxicity with one or more prior systemic, surgical, or radiation cancer therapies, except patients with NSCLC and breast cancer who must have two or more prior therapies. Inadequate response was defined as having persistent disease or ≥50% risk of recurrence per investigator assessment despite standard or curative intent treatment. An Eastern Cooperative Oncology Group performance status of 0 or 1, life expectancy of ≥6 months, and electrocardiogram without evidence of clinically significant arrhythmia or ischemia were required.

Exclusion criteria included prior treatment with other NY-ESO-1–targeting agents, uveal melanoma, unstable brain metastases, significant immunosuppression from corticosteroids or other immunosuppressive medications, significant autoimmune disease, myocardial infarction within 6 months or New York Heart Association Grade III or IV heart failure, inadequate organ function (marrow, hepatic, renal), and clinically significant infection.

LV305 (1 mL per dose) was administered ID once every 3 weeks using a MicronJet needle (Nanopass Technologies Ltd.). Each 1 mL dose was divided into 8 separate injections of 125 μL each; two 125 μL injections were administered to each upper arm and two 125 μL injections were administered to each thigh. Patients were enrolled in one of four cohorts: Cohort 1 received 1 × 10⁸ viral genomes (vg) x 3 doses, Cohort 1A received 1 × 10⁸ vg x 4 doses, Cohort 2 received 1 × 10⁹ vg x 4 doses, and Cohort 3 received 1 × 10¹⁰ vg x 4 doses. Concurrent treatment with corticosteroids or immunosuppressive medications (e.g., methotrexate, cyclosporine, azathioprine) was prohibited.

Adverse events (AEs) were assessed regularly during treatment, and AE severity was graded per NCI Common Terminology Criteria for Adverse Events v.4.03. During dose escalation, a dose-limiting toxicity (DLT) was defined as any treatment-related grade ≥3 AEs that occurred during the 28 days following the first dose of LV305, with the exception of alopecia, vomiting or diarrhea controlled by optimal anti-emetics or anti-diarrheals, grade 3 fatigue, asymptomatic grade 3 laboratory AEs, or grade 3 systemic reactions (e.g., fever, headache, influenza-like symptoms, myalgia, malaise, nausea) that returned to baseline or grade 1 within 3 days. Data were reviewed by an independent data monitoring committee for DLT during the specified time period for each dose cohort and then for all patients at least quarterly thereafter.

Tumor imaging and clinical assessments were performed by the investigator at screening (baseline) and then every 8 weeks until disease progression. After progression, long-term follow-up was conducted every 12 weeks for any serious AEs (SAE) possibly related to LV305, disease status, and survival. The presence or absence of detectable LV305 genome in peripheral blood mononuclear cells (PBMC) was assessed and monitored for up to 2 years from first dose of LV305.

Tumor response was assessed by the investigator using irRC (13, 14) modified to incorporate RECIST v 1.1 criteria (15). The objective response rate was calculated as the proportion of patients with complete response (CR) or PR, and the disease control rate (DCR) was calculated as the proportion of patients with CR, PR, or stable disease (SD) for a minimum of 2 months from start of therapy. Clopper–Pearson exact 95% confidence interval (CI) was provided.

Baseline tumor biopsies were obtained from all patients to screen for presence of NY-ESO-1 expression on tumor specimens by IHC. Posttreatment biopsies for exploratory biomarker analysis were optional in patients with accessible tumors. Peripheral blood samples were collected for immunologic testing at baseline and on days 14, 28, 63, and 84. These samples were processed into plasma and PBMC over Ficoll gradients and stored cryogenically for immunologic testing. Leukapheresis (1.5 to 2 blood volume process) were collected on each patient before and after treatment. PBMCs isolated from leukapheresis samples were cryopreserved for immunologic testing and exploratory biomarkers.

Details of the criteria for preexisting and induced antibody and T-cell responses have been published previously (16). For this study, plasma was analyzed for the presence of anti–NY-ESO-1 antibodies using an ELISA with recombinant NY-ESO-1 protein; a titer >1:100 was considered positive. The induction of anti–NY-ESO-1 antibody response was defined for this study as a ≥4-fold increase in the titer or the presence of a newly positive response after the first dose of LV305. Anti–NY-ESO-1 CD4⁺ and CD8⁺ T-cell responses were measured by IFNγ detecting enzyme-linked ImmunoSpot (ELISPOT) using isolated CD4⁺ and CD8⁺ T cells from PBMCs expanded in vitro with a NY-ESO-1 peptide pool (20-mer peptides, 10-mer overlap) and considered positive if there were >50 spot-forming units/50,000 cells observed for NY-ESO-1 peptides and a ≥2-fold increase in spot-forming units compared with a negative control. The induction of an anti–NY-ESO-1 CD4⁺ or CD8⁺ T-cell response was defined as de novo positive or ≥2-fold rise in the number of spot-forming units on the ELISPOT assay after the first dose of LV305. In addition, a FACS-based intracellular cytokine staining (ICS) assay for IFNγ or TNFα after stimulation with NY-ESO-1 peptide pool was used, and staining ≥2-fold above the baseline value was considered positive for T-cell responses.

T-cell receptor beta (TCRβ) repertoire in baseline and posttreatment PBMC were evaluated by TCRβ chain sequencing of complementarity-determining region (CDR)3 on PBMC and tumor biopsies. Immunosequencing of the CDR3 regions of human TCRβ chains was performed using the ImmunoSEQ Assay (Adaptive Biotechnologies). Extracted genomic DNA was amplified in a bias-controlled multiplex PCR, followed by high-throughput sequencing. Sequences were collapsed and filtered in order to identify and quantitate the absolute abundance of each unique TCRβ-CDR3 region for further analysis as previously described (17–19).

Shannon's Entropy, H, is defined as |- \sum\nolimits_{i = 1}^N {{p_i}{{\log }_2}( {{p_i}} )} $|⁠, where |{p_i}$| is the proportional abundance of clone i, and N is the total number of unique TCR receptor rearrangements. Clonality is related to Shannon's Entropy and defined by |${1 -{ \frac{{{H}}}{{{{\log }_2}{{N}}}}}$|⁠, where |\frac{{{H}}}{{{{\log }_2}{{N}}}}$| is also known as Pielou's evenness (20). Clonality values range from 0 to 1 and describe the shape of the immune repertoire's frequency distribution: clonality values approaching 0 indicate a very even distribution of frequencies, whereas values approaching 1 indicate an increasingly asymmetric distribution in which a few clones are present at high frequencies. Richness of a repertoire was calculated using the Daley–Smith Estimate (R package: preseqR, v3.0.1), which combines a rational function approximation with Good and Toulmin's nonparametric empirical Bayes power series estimator (21, 22). The potential association between TCR clonality and overall survival (OS) was analyzed using the Kaplan–Meier method.

PFS and OS were evaluated using the Kaplan–Meier method and compared using a log-rank test stratified by disease type. The OS rates at 1-year are presented, with P values for difference in 1-year OS calculated using Z statistics of difference in survival rates/square root of sum of variance. HR and its 95% CI were calculated using a stratified Cox model stratified by disease type.

Screening for NY-ESO-1 expression was performed on previously collected tumor samples from patients with eligible tumor types; the percentages of NY-ESO-1–positive specimens in screened patients are presented by tumor type in Supplementary Table S2; the highest percentages were seen in patients with MRCL (93%) and SS (68%–76%). A total of 39 patients were enrolled. The majority of enrolled patients had sarcoma (62%; n = 24); other tumor types enrolled were ovarian (21%; n = 8), melanoma (15%; n = 6), and NSCLC (3%; n = 1; Table 1). The most common sarcoma subtypes enrolled were SS (n = 13) and MRCL (n = 6). Eighty-seven percent of all enrolled patients had stage IV disease, and 28% had ≥3 prior systemic anticancer therapies (Table 1).

NY-ESO-1 expression was high (>50% of tumor cells by IHC; n = 18) or moderate (>25%–50% of tumor cells by IHC; n = 2) in 83% of sarcoma patients (20/24; Table 1). Among patients with other tumor types, high expression was detected in 50% of melanoma patients (3/6) and the only NSCLC patient. Ovarian cancer had the lowest NY-ESO-1 expression levels, with expression in ≤5% of tumor cells in all enrolled subjects.

Twelve patients received LV305 during dose escalation, 3 patients in each dose cohort. No DLTs were observed, and the MTD was not reached. Enrollment was expanded to enroll 27 additional patients at the highest dose level tested, 1 × 10¹⁰ vg x 4 doses.

Thirty-four patients (87%) experienced at least one treatment-related AE. The most common were fatigue (49%), injection site reaction (46%, includes the terms injection site pruritus, injection site pain, injection site erythema, and injection site reaction), and myalgia (21%; Table 2). All treatment-related events were grade 1 or 2 in severity. No grade ≥3 treatment-related AE and no treatment-related SAE were observed. Thirty patients (77%), including 20 of 24 (83%) sarcoma patients, completed all planned therapy, and no patients discontinued due to AE.

Seventy-four percent of patients had at least 1 evaluable posttreatment assessment for the persistence of the LV305 genome; 49% had 2 or more assessments, 38% were tested at approximately 12 months, and 26% were tested ≥22 months after baseline (including 2 patients at >3 years). The presence of LV305 genome was not detected in any patient at any timepoint on study.

The presence of anti–NY-ESO-1 T-cell responses was evaluable in 85% of patients, and the presence of anti–NY-ESO-1 antibodies in plasma was evaluable in 97%. Preexisting anti–NY-ESO-1 T-cell responses were detected at baseline in 36% of patients (39% of sarcoma patients), and anti–NY-ESO-1 antibodies were present at baseline in 42% of patients (39% of sarcoma patients). Figure 1A summarizes LV305-induced IR in all patients, and Fig. 1B summarizes LV305-induced IR in all sarcoma patients. LV305-induced CD4⁺ or CD8⁺ T cells were detected by ELISPOT and/or ICS in 52% of patients (57% of sarcoma patients) at ≥1 timepoint on study, and 18% at ≥2 timepoints. The survival rate at 1 year was 94.1% in patients with induction of anti–NY-ESO-1 T cells versus 62.5% in those without T-cell induction (P = 0.0181), and 91.7% in patients with preexisting anti–NY-ESO-1 T-cells versus 71.4% in those without (P = 0.1105; Fig. 1C and D). LV305-induced anti–NY-ESO-1 antibodies were detected by ELISA in 8% of patients (4% of sarcoma patients). The survival rate at 1 year was 100% in patients with induction of anti–NY-ESO-1 antibody versus 77.1% in those without antibody induction (P = 0.0013), and 93.8% in patients with preexisting anti–NY-ESO-1 antibody at baseline versus 68.2% in those without (P = 0.0279). Neither preexisting immunity (Ab or T cells) nor induced IR was associated with NY-ESO-1 expression in tumor in a logistic regression model that included immunity as the response variable and NY-ESO-1 expression as the independent variable (data not shown).

A durable PR was observed in 1 sarcoma patient, and an unconfirmed PR was observed in 1 lung cancer patient. In the patient with lung cancer, previously undetected lesions were identified in the central nervous system (an immune sanctuary site) within 1 month of the initial observation of PR and were associated with symptomatic deterioration and discontinuation from the study. An additional 20 patients (51.3%) had SD, based on modified irRC, for a DCR of 56.4%.

Among patients with sarcoma, 1 patient (4.2%; patient no. 006) achieved PR and 14 patients (58.3%) had SD, for a DCR of 62.5% (Fig. 2). The median duration of disease control in these 15 patients was 5.9 months (95% CI, 1.8–16.8 months). Among patients with SS, the DCR was 53.8%, with a median duration of 9.8 months (95% CI, 0 months–not assessable). In sarcoma patients who had evidence of any tumor growth tumor at study entry, as determined by site investigators, 61.5% (8 of 13) achieved disease control following LV305, compared with 63.6% of patients (7 of 11) who had SD or no evidence of disease (NED) with high risk of recurrence per investigator assessment.

The sarcoma patient who achieved a PR (patient no. 006) had evidence of tumor growth at study entry after having received extensive prior therapies, including multiple tumor-related surgeries, radiation, single-agent ifosfamide, ifosfamide/adriamycin, trabectedin, and pazopanib. This patient initially had SD upon completion of all dosing of LV305 with a slow tumor regression over time resulting in a PR approximately 1.5 years after completion of LV305 dosing (16). The tumor continued to regress to a near-CR state, measuring less than 1 cm at 42 months after the first dose of LV305 (Fig. 3A). This patient did not receive any tumor-related treatment after LV305 administration. This patient had both preexisting anti–NY-ESO-1 antibodies and T cells; however, the evidence of tumor growth at time of enrollment despite extensive prior therapies indicates the preexisting anti–NY-ESO-1 immune response was inadequate to control the tumor. Following LV305 dosing, the patient developed an induced and durable anti–NY-ESO-1 IR. We have previously shown that NY-ESO-1–specific CD4⁺ and CD8⁺ T cells were induced in this patient (16). TCRβ repertoire analysis using PBMC samples collected from this patient at baseline, day 84, and at 2-year follow-up showed increased tumor-infiltrating lymphocyte (TIL) sequences in PBMC post-LV305 dosing, and the tumor NY-ESO-1–specific sequences continued to be present 2 years after completion of LV305 dosing (Fig. 3B).

The median duration of follow-up was 27.9 months (range, 3.2–42.4 months). Median PFS was 4.6 months (95% CI, 2.7–11.7 months) in all patients and 4.6 months (95% CI, 2.5–8.6 months) in patients with sarcoma (Supplementary Fig. S1A). Median PFS was 2.8 months, 9.6 months, and 4.6 months in patients with SS, MRCL, and other sarcoma subtypes, respectively (Supplementary Fig. S1B).

Median OS was 31.1 months [95% CI, 18.8 months–not available (NA)] in all patients. In patients with sarcoma, the median OS was 33.9 months (95% CI, 18.8 months–NA) with OS rates of 83.3% and 66.7% at 12 and 24 months, respectively (Supplementary Fig. S1C). OS was similar across sarcoma subtypes (Supplementary Fig. S1D).

Sarcoma patients at high risk of recurrence per investigator assessment with NED at time of enrollment (n = 6) had longer PFS than patients with SD (n = 5) or evidence of tumor growth (n = 13) at enrollment; median PFS was 16.6 months (95% CI, 0.9 months–NA) in patients with NED; 2.6 months (95% CI, 2.2 months–NA) in patients with SD; and 2.8 months (95% CI, 1.7–4.6 months) in those with any evidence of tumor growth. Median OS was not reached in sarcoma patients with NED at time of enrollment, and was 31.11 months (95% CI, 7.03 months–NA) in patients with SD and 27.79 months (95% CI, 16.85–NA) in those with any evidence of tumor growth.

A post hoc exploratory analysis revealed significantly longer OS in patients with increased T-cell clonality (higher than median post-/preclonality ratio) in posttreatment PBMC among all patients (P = 0.0116; Fig. 4A) and sarcoma patients (P = 0.0184; Fig. 4B), indicating LV305-induced clonal expansion of T-cell anti–NY-ESO-1 responses may be associated with OS.

LV305 is the first vaccine designed to selectively induce expression of an antigen-encoding gene in DCs in vivo. This first-in-human study of this first-in-class agent demonstrates the potential of a LV vaccine to induce antigen expression in DCs and to promote IR and tumor responses in patients.

LV305 was administered ID in the outpatient setting and was well-tolerated with only grade 1 or 2 treatment-related AEs reported and no treatment-related SAEs. LV305 effectively induced an anti–NY-ESO-1 IR that was broad, tumor-specific, and durable. IR consisted of both anti–NY-ESO-1 antibody and CD4⁺/CD8⁺ T-cell responses, as measured by cell-based and molecular assays, with measurable responses ongoing as long as 2+ years after treatment. LV305 was associated with increases of preexisting immune responses to NY-ESO-1 as well as de novo IR, suggesting LV305 may act not only to prime, but also to boost IR. LV305 administration resulted in a greater induction of T-cell responses than antibody responses, which is not unexpected as NY-ESO-1 is expressed in relatively low amounts in DC and presumably not secreted.

The data indicate that LV305-induced anti–NY-ESO-1 IR can lead to tumor regression over time, as evident in one patient with synovial sarcoma who achieved a near-CR. In this patient, a relatively high level of preexisting T cells was boosted, with the durability of this T-cell response measured for over 2 years. Although no objective responses were observed in the other patients, 62.5% of all sarcoma patients and 61.5% of sarcoma patients with growing tumor at study entry maintained or achieved SD. These data suggest that the main mechanism of action of LV305 in this patient population may be slowing of tumor growth. A similar observation has been made with the ex vivo–generated DC-based prostate cancer vaccine sipuleucel-T, which also expresses a single tumor antigen (23).

In addition, in exploratory analyses, the induction of an anti–NY-ESO-1 response following LV305 (either a de novo response or an increase in a preexisting response) was associated with improved OS. Increased clonality (clonal expansion of antitumor T cells as measured by TCR beta CDR3 sequencing) also appeared to be associated with improved outcomes. This molecular evidence for a T-cell–mediated effect of LV305 on the clinical course of the disease warrants further study.

The induction of anti–NY-ESO-1 CD8⁺ T-cell IR can take several months to occur, and bulky tumors can be immunosuppressive, which is the reason enrollment to this study was limited to patients with low or moderate tumor burden. The favorable clinical outcomes of this group indicate that single-agent LV305 may have an optimal effect when given as a maintenance therapy. Alternatively, for patients with a larger, bulky tumor burden, LV305 may prove to play a role in combination immunotherapy strategies. Further investigation of LV305 as a solid tumor immunotherapy is merited, possibly in combination with a boosting vaccine and/or other agents, such as anti–PD-L1 checkpoint inhibitors. Further development of LV305 is planned in combination with G305, as the combination therapy “CMB305.” G305 is a cancer vaccine composed of a full-length recombinant NY-ESO-1 protein and the synthetic TLR4 agonist glucopyranosyl lipid A in stable emulsion (GLA-SE; ref. 24). LV305 and G305 have different but complementary mechanisms of action, and in preclinical models, priming DCs with LV305 and then boosting with G305 significantly increased NY-ESO-1–directed antitumor activity. Studies of CMB305 alone or in combination the anti–PD-L1 checkpoint inhibitor atezolizumab are ongoing (NCT02609984, NCT02387125). Additional opportunities for further development of LV305 continue to be explored.

Although limited by the exploratory nature and the small number of patients, the favorable safety profile, promising tumor outcomes, and durable IR observed in this first-in-class, first-in-human study demonstrate the feasibility and potential therapeutic impact of direct injection of a DC tropic LV vector into patients as a cancer vaccine.

N. Somaiah was a consultant/advisory board member for Immune Design at the time of this study. M.S. Block reports receiving commercial research grants from Immune Design, Marker Therapeutics, and Merck. J.W. Kim reports receiving commercial research grants from Immune Design, and is a consultant/advisory board member for Dendreon, AstraZeneca, and, Clovis. G.I. Shapiro reports receiving commercial research grants from Lilly, Merck EMD-Serono, Sierra Oncology, and Merck, and is a consultant/advisory board member for Pfizer, Lilly, G1 Therapeutics, Astex, Almac, Roche, Bicycle Therapeutics, Merck-EMD Serono, Fusion Pharmaceuticals, Sierra Oncology, Bayer, Ipsen, Cybrexa Therapeutics, Angiex, Daiichi Sankyo, and Seattle Genetics. K.T. Do is a consultant/advisory board member for Seattle Genetics. P. Hwu holds ownership interest (including patents) in Dragonfly and Immatics, and is a consultant/advisory board member for Dragonfly, GlaxoSmithKline, Immatics, and Sanofi. R.L. Jones is a consultant/advisory board member for Adaptimmune, Blueprint, Clinigen, Eisai, Epizyme, Daichii, Deciphera, Immune Design, Lilly, Merck, Pharmamar, and Tracon. J.H. ter Meulen and F.J. Hsu hold ownership interest (including patents) in Immune Design. S. Gnjatic reports receiving commercial research grants from Immune Design, Pfizer, Genentech, and Bristol-Myers Squibb, and is a consultant/advisory board member for Merck and OncoMed. S.M. Pollack reports receiving other commercial research support from Immune Design. H. Lu, J.H. ter Meulen, C. Bohac, M. Chen, and F.J. Hsu were employees of Immune Design during the study. J.H. ter Meulen and F.J. Hsu hold ownership interest (including patents) in Immune Design. No potential conflicts of interest were disclosed by the other authors.

Conception and design: N. Somaiah, P. Hwu, R.L. Jones, F.J. Hsu, S.M. Pollack

Development of methodology: N. Somaiah, R.L. Jones, H. Lu, F.J. Hsu, S. Gnjatic, S.M. Pollack

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): N. Somaiah, M.S. Block, J.W. Kim, G.I. Shapiro, K.T. Do, J.P. Eder, R.L. Jones, H. Lu, S. Gnjatic, S.M. Pollack

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): J.W. Kim, G.I. Shapiro, K.T. Do, P. Hwu, J.P. Eder, R.L. Jones, H. Lu, J.H. ter Meulen, C. Bohac, M. Chen, F.J. Hsu, S. Gnjatic, S.M. Pollack

Writing, review, and/or revision of the manuscript: N. Somaiah, J.W. Kim, G.I. Shapiro, K.T. Do, P. Hwu, J.P. Eder, R.L. Jones, H. Lu, J.H. ter Meulen, C. Bohac, F.J. Hsu, S. Gnjatic, S.M. Pollack

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): N. Somaiah, R.L. Jones, C. Bohac, F.J. Hsu, S. Gnjatic

Study supervision: N. Somaiah, R.L. Jones, C. Bohac, F.J. Hsu

Authors wish to thank Seunghee Kim-Schulze, Kevin Tuballes, and Mahlet Yishak for technical assistance, and Janis Leonoudakis, PhD, for medical writing assistance. S. Gnjatic acknowledges Mount Sinai's Cancer Center Support Grant P30CA196521. This study was funded by Immune Design.

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.