In vitro validation of neoantigen prediction algorithm for developing personalized cancer vaccine therapy

Background: The development of personalized neoantigen-based therapeutic cancer vaccines relies on computational algorithm-based pipelines. One of the critical issues in the pipeline is obtaining higher positive predictive value (PPV) performance, i.e., how many are immunogenic when selecting the top 5 to 20 candidate neoepitopes for the vaccination. We attempted to test the PPV of a neoepitope prediction algorithm Neopepsee. Methods: Six breast cancer patients and patient-derived xenografts from three lung cancer patients and their paired peripheral blood samples were subjected to whole-exome and RNA sequencing. Neoantigen was predicted using two different algorithms (Neopepsee and pVACseq). Response of induced memory T cells to neopeptide candidates was evaluated by IFN-γ Enzyme-linked immune absorbent spot (ELISpot) assays of peripheral blood mononuclear cell (PBMC) from three HLA-matched donors. Positive ELISpot response to a candidate peptide in at least 2 of 3 donor PBMC was regarded as an immunogenic response. Results: Neopepsee predicted 159 HLA-A matched neoepitope candidates out of 898 somatic mutations in nine patients (six breast and three lung cancer patients), whereas pVACseq predicted 84 HLA-A matched candidates. A total of 26 neopeptide candidates overlapped between the two predicted candidate pools. Among the candidates, 28 (20%, 28/ 137) and 15 (20%, 15/ 75) were positive by ELISpot assay, respectively. Among 26 overlapped candidates, 20 could be tested, and 7 of them (35%) were validated by ELISpot. Neopepsee identified at least one neoepitope in 7 of 9 patients (range 0-6), compared to 6 by pVACseq (range 0-5). Conclusion: As suggested by Tumor Neoantigen Selection Alliance (TESLA), our results demonstrate low PPV of individual prediction models as well as the complementary nature of the Neopepsee and pVACseq and may help design neoepitope targeted cancer vaccines. Our data contribute a significant addition to the database of tested neoepitope candidates that can be utilized to further train and improve the prediction algorithms.

Long-Term Follow-Up of Gemogenovatucel-T (Vigil) Survival and Molecular Signals of Immune Response in Recurrent Ovarian Cancer

Aim: To determine the relationship between gene expression profile (GEP) and overall survival (OS) by NanoString following treatment with Vigil. Patients and Methods: Recurrent ovarian cancer patients (n = 21) enrolled in prior clinical trials. Results: GEP stratified by TISHIGH vs. TISLOW demonstrated OS benefit (NR vs. 5.8 months HR 0.23; p = 0.0379), and in particular, MHC-II elevated baseline expression was correlated with OS advantage (p = 0.038). Moreover, 1-year OS was 75% in TISHIGH patients vs. 25% in TISLOW (p = 0.03795). OS was also correlated with positive γ-IFN ELISPOT response, 36.8 vs. 23.0 months (HR 0.19, p = 0.0098). Conclusion: Vigil demonstrates OS benefit in correlation with TISHIGH score, elevated MHC-II expression and positive γ-IFN ELISPOT in recurrent ovarian cancer patients.

Weak immunogenicity of SARS-CoV-2 vaccine in patients with hematologic malignancies

This study evaluated the safety and immunogenicity of BNT162b2 vaccine in patients with hematological malignancies. Antibodies blocking spike binding to immobilized ACE-2 (NAb) correlated with anti-Spike (S) IgG d42 titers (Spearman r = 0.865, p < 0.0001), and an anti-S IgG d42 level ≥3100 UA/mL was predictive of NAb ≥ 30%, the positivity cutoff for NAb (p < 0.0001). Only 47% of the patients achieved an anti-S IgG d42 level ≥3100 UA/mL after the two BNT162b2 inocula, compared to 87% of healthy controls. In multivariable analysis, male patients, use of B-cell targeting treatment within the last 12 months prior to vaccination, and CD19+ B-cell level <120/uL, were associated with a significantly decreased probability of achieving a protective anti-S IgG level after the second BNT162b2 inoculum. Finally, using the IFN-γ ELISPOT assay, we found a significant increase in T-cell response against the S protein, with 53% of patients having an anti-S IgG-positive ELISPOT after the second BNT162b2 inoculum. There was a correlation between the anti-S ELISPOT response and IgG d42 level (Spearman r = 0.3026, p = 0.012). These findings suggest that vaccination with two BNT162b2 inocula translates into a significant increase in humoral and cellular response in patients with hematological malignancies, but only around half of the patients can likely achieve effective immune protection against COVID-19.

Therapeutic Cancer Vaccination With a Peptide Derived From the Calreticulin Exon 9 Mutations Induces Strong Cellular Immune Responses in Patients With CALR-Mutant Chronic Myeloproliferative Neoplasms

BackgroundThe calreticulin (CALR) exon 9 mutations that are identified in 20% of patients with Philadelphia chromosome negative chronic myeloproliferative neoplasms (MPN) generate immunogenic antigens. Thus, therapeutic cancer vaccination against mutant CALR could be a new treatment modality in CALR-mutant MPN.MethodsThe safety and efficacy of vaccination with the peptide CALRLong36 derived from the CALR exon 9 mutations was tested in a phase I clinical vaccination trial with montanide as adjuvant. Ten patients with CALRmut MPN were included in the trial and received 15 vaccines over the course of one year. The primary end point was evaluation of safety and toxicity of the vaccine. Secondary endpoint was assessment of the immune response to the vaccination epitope (www.clinicaltrials.gov identifier NCT03566446).ResultsPatients had a median age of 59.5 years and a median disease duration of 6.5 years. All patients received the intended 15 vaccines, and the vaccines were deemed safe and tolerable as only two grade three AE were detected, and none of these were considered to be related to the vaccine. A decline in platelet counts relative to the platelets counts at baseline was detected during the first 100 days, however this did not translate into neither a clinical nor a molecular response in any of the patients. Immunomonitoring revealed that four of 10 patients had an in vitro interferon (IFN)-γ ELISPOT response to the CALRLong36 peptide at baseline, and four additional patients displayed a response in ELISPOT upon receiving three or more vaccines. The amplitude of the immune response increased during the entire vaccination schedule for patients with essential thrombocythemia. In contrast, the immune response in patients with primary myelofibrosis did not increase after three vaccines.ConclusionTherapeutic cancer vaccination with peptide vaccines derived from mutant CALR with montanide as an adjuvant, is safe and tolerable. The vaccines did not induce any clinical responses. However, the majority of patients displayed a marked T-cell response to the vaccine upon completion of the trial. This suggests that vaccines directed against mutant CALR may be used with other cancer therapeutic modalities to enhance the anti-tumor immune response.

Results of the randomized phase II study of sipuleucel-T (Sip-T) +/- Radium-223 (Ra-223) in men with bone-metastatic castration resistant prostate cancer.

5563 Background: It has been suggested that immune modulation can be augmented by radiation, possibly by enhancing tumor-antigen display. SipT-induced antigen-specific immune responses in mCRPC patients correlate with survival. We hypothesized that the combination of Ra223 and SipT would enhance SipT-related immune response and improve outcomes compared to SipT alone. Methods: Patients with asymptomatic, bone-predominant mCRPC, without visceral mets >1.0 cm, were randomized (1:1) to SipT alone or with 6 doses of Ra223 (NCT02463799). Men in the SipT+Ra223 arm started SipT between the 2nd and 3rd dose of Ra223. The primary immunologic endpoint was PA2024-specific T-cell proliferation 6 wks after the first SipT infusion. Secondary immune endpoints were PA2024-specific ELISPOT response, PAP-specific proliferation and ELISPOT, humoral responses against both antigens, and antigen spread. Clinical endpoints were radiographic PFS, PSA response (≥50% decline), AlkPhos response (≥30% decline), and safety. Results: 32 men were randomized, 16 per arm. Baseline characteristics in SipT+Ra223 and SipT arms were similar: age (median 71 vs. 70 yrs), Gleason (8-10: 69% vs. 69%), baseline PSA (med 25 vs. 33 ng/mL), AlkPhos (med 89 vs. 92 U/L) and ECOG score (≥1: 31% vs. 19%). There was no significant difference in prior use of abi/enza (38% vs. 44%), or chemo (0% vs. 25%). At 6 weeks, absolute PA2024-specific T-cell proliferation was 2.1-fold higher in the Sip-T arm compared to the SipT+Ra223 arm (35.6 vs. 16.6; P=0.03) and remained higher through week 26. Relative to baseline, the 6-week PA2024-specific T-cell proliferation change was 3.6 times greater in the Sip-T arm compared to the SipT+Ra223 arm ( P=0.007) and remained higher through week 14. There were no significant differences in antigen spread or humoral responses. Median radiographic PFS was longer in the SipT+Ra223 arm (9.3 vs. 3.2 months; HR 0.26, 95% CI 0.11–0.61; P=0.007). PSA and AlkPhos responses were better in the SipT+Ra223 arm (PSA50: 5/15=33% vs. 0/14=0%; P=0.04; AlkPhos30: 9/15=60% vs. 1/15=7%; P=0.01). There was no difference in SREs (13% vs. 7%). Conclusions: SipT+Ra223 was associated with improved clinical outcomes and a higher rate of PSA responses compared to SipT alone, although surprisingly, the SipT arm demonstrated higher peripheral PA2024-specific T-cell proliferation. Since neither agent reliably induces PSA responses alone, these data suggest a synergistic effect of the combination. Larger randomized studies of this combination are planned. Clinical trial information: NCT02463799 .

Randomized phase II study of sipuleucel-T (SipT) with or without radium-223 (Ra223) in men with asymptomatic bone-metastatic castrate-resistant prostate cancer (mCRPC).

130 Background: SipT-induced antigen-specific immune responses in mCRPC patients correlate with survival. Due to the immunomodulatory effects of radiopharmaceutical agents (e.g. enhancing tumor-antigen display), we hypothesized that combined use of Ra223 and SipT would augment SipT-related immune response and improve outcomes compared to SipT alone. Methods: Patients with asymptomatic mCRPC and bone-predominant mets, without visceral mets >1.0 cm, were randomized (1:1) to standard SipT alone or combined with 6 doses of Ra223 (NCT02463799). Men in the SipT+Ra223 arm received SipT between the 2nd and 3rd dose of Ra223. Clinical endpoints were radiographic/clinical PFS, PSA response (≥50% decline), AlkPhos response (≥30% decline), and safety. Immunologic endpoints were PA2024-specific T-cell proliferation 6 wks after the first SipT infusion, PA2024-specific ELISPOT response, PAP-specific proliferation and ELISPOT, humoral responses against both antigens, and antigen spread. Results: 32 men were randomized, 16 per arm. Baseline characteristics in SipT and SipT+Ra223 arms were matched with respect to age (median 70 vs 71 yrs), Gleason (8-10: 69% vs 69%), PSA (median 82 vs 72 ng/mL), AlkPhos (median 125 vs 125 U/L) and ECOG scores (≥1: 19% vs 31%). After median follow up of 5.3 (range 2.8–26.6) mo, median PFS was longer in the SipT+Ra223 arm (10.7 vs 3.1 mo; HR 0.35, 95% CI 0.15–0.81; P=0.02). Outcomes were also better in the SipT+Ra223 arm with respect to PSA responses (5/15=33% vs 0/14=0%; P=0.04) and AlkPhos responses (9/15=60% vs 1/15=7%; P=0.01). No safety concerns were observed with the combination (grade 3 AEs shown in the Table). Conclusions: SipT combined with Ra223 was associated with improved clinical outcomes compared to SipT alone. Since neither agent reliably induces PSA responses alone, these data suggest a synergistic effect of the combination. Immunologic endpoints will be presented at the meeting. Larger randomized studies of this combination are warranted. Clinical trial information: NCT02463799. [Table: see text]

Cellular and humoral cytomegalovirus immunity changes in one-year combined prophylaxis after lung transplantation: suggestions from and for clinical practice

Background: Immune responses, both cellular and humoral, against cytomegalovirus (CMV) are used to predict CMV manifestations in solid organ recipients. The aim of this study is to evaluate CMV enzyme-linked immunospot (ELISPOT) assay and serology during CMV infections, their concordance and variations after lung transplantation (LTx). Methods: We retrospectively analysed in one year the follow-up data of 43 patients receiving combined CMV prophylaxis with antiviral agents and CMV-specific immunoglobulin G (IgG). CMV infections were investigated by using molecular analyses on both 167 bronchoalveolar lavage and biopsy specimens and 1134 blood samples. Cellular CMV immunity was assessed with specific ELISPOT whereas the humoral one was assessed by quantifying specific immunoglobulins. Results: At the first month after LTx the majority of patients were ELISPOT responders (52.3%) and 30.9% were non-responders. ELISPOT responders had a lower incidence of CMV viremia ( p = 0.047), whereas neither effects on CMV pulmonary asymptomatic infection nor on acute rejection were observed. Responders had a higher CMV IgG titre ( p < 0.0001) in particular at the first month after LTx ( p = 0.0001). Concordance among CMV ELISPOT assay and IgG levels was moderate (Cohen’s K 0.524), with an agreement of 89.8%. All ELISPOT responders maintained their status and almost all non-responders became responders during follow-up (92.3%); the percentage of IgG seropositive subjects increased from 74.4% at the first month of follow-up to 97.4% after 1 year. Conclusions: Despite a moderate concordance with serology, ELISPOT response predicted a lower incidence of CMV viremia in LTx patients; no effects were reported on pulmonary clinical manifestations nor on acute rejection. The ELISPOT response as well as serology changed during the follow-up, not only after first CMV contact. The reviews of this paper are available via the supplemental material section.

2744. A Phase I Randomized, Observer-Blind, Controlled, Dose Escalation Trial of the Safety and Tolerability of a Single Intramuscular Dose of a PAL Adjuvant (Laboratory Code, FB-631) Co-administered with Seasonal TIV (2013–2014) to Healthy Adults ≥18–50 Years of Age

Background Inactivated influenza vaccines (IV) efficacy is variable and sometimes poor. In this phase 1 trial the safety and immunogenicity of a novel nanoparticle adjuvant (Papaya Mosaic Virus (PapMV or PAL) at different dose levels combined with inactivated trivalent IV (TIV; FLUVIRAL® 2013–2014, GSK, Kirkland PQ) was assessed. Nonpathogenic in mammals, PAL is recognized as a pathogen-associated molecular pattern (PAMP) which stimulates innate, cell-mediated immunity (CMI) and adaptive immunity in naïve mice through activation of toll like receptor 7 and 8. Methods Healthy persons 18–50 years of age were randomized to one of 6 study groups: 30 µg, 60 µg, 120 µg or 240 µg of PAL with 0.25 mL TIV, 240 µg of PAL with 0.125 mL TIV, or control (0.5 mL TIV). Solicited local and general adverse events (AE) were collected Day (D)0 to 6, unsolicited AE to D28, and serious AE to D1095. Hemagglutination-inhibition assays (HI), antibody to Influenza A virus nucleoprotein (NP), and peripheral blood mononuclear cells (PBMC) for measurement of interferon-gamma (IFNg) ELISPOT (response to PepMix influenza A H2N2 Ann Arbour NP, MP1, and an influenza peptide pool), granzyme B, and IFNg:IL:10 ratio were collected on D0, 7, 28, 120, and 180. Results The most common solicited AEs were transient mild-to-moderate local pain (62.5%–87.5% of participants/group), drowsiness (≤37.5% of participants/group) and generalized muscle aches (12.5–50% of participants/group). There was one unrelated SAE. All participants had HI and anti-NP titers at baseline. HI GMTs increased at D28 post vaccine in most groups (Figure 1) and waned over time. HI fold Ab (Far) responses to TIV strains were poor in all groups (≤37.5% of participants/group had 4-Far to any strain). CMI results were consistent with humoral responses. Conclusion The PAL adjuvant in doses of 30 to 240µg combined with reduced TIV dosages was safe with no signals up to 3 years after vaccine. Reduced doses of TIV co-presented with 240 µg PAL had similar HI GMTs as TIV. The CMI results suggest that the assessment of PAL efficacy on TIV immunization would have to be conducted in an influenza naïve population. Disclosures Scott Halperin, MD, GlaxoSmithKline: Ad hoc advisory boards, Other Financial or Material Support, Research Grant; Janssen Pharm: Research Grant; sanofi pasteur: Ad hoc advisory boards, Other Financial or Material Support, Research Grant.

Clinical outcomes model in renal cell cancer patients treated with modified vaccinia Ankara plus tumor-associated antigen 5T4

Aims We developed an outcomes model to select patients for renal cell cancer vaccine immunotherapy. Materials and methods We examined clinical data from 2 phase II studies of modified vaccinia Ankara as vector to express 5T4 (MVA-5T4), calculated progression-free survival (PFS) and overall survival (OS), and created risk groups based on the number of factors involved. Results Median OS was 12.4 months; median PFS was 3.6 months. Significant factors (p<0.05) included neutrophils (both), bone metastases (OS), ECOG performance status (OS), lactate dehydrogenase levels (both), prior therapy with tyrosine kinase inhibitors plus immunotherapy (OS), Fuhrman grade (OS), and 5T4-specific ELISPOT response (PFS). By group, median OS was not reached in patients with favorable risk (censored at cutoff), was 13.7 months in those with intermediate risk and 4.0 months in those with poor risk. Conclusions Further validation of this model will identify the patients most likely to respond to MVA-5T4 and provide a framework for outcomes models for other vaccine therapies.

Process of Assay Selection and Optimization for the Study of Case and Control Samples from a Phase IIb Efficacy Trial of a Candidate Tuberculosis Vaccine, MVA85A

ABSTRACTThe first phase IIb safety and efficacy trial of a new tuberculosis vaccine since that for BCG was completed in October 2012. BCG-vaccinated South African infants were randomized to receive modified vaccinia virus Ankara, expressing theMycobacterium tuberculosisantigen 85A (MVA85A), or placebo. MVA85A did not significantly boost the protective effect of BCG. Cryopreserved samples provide a unique opportunity for investigating the correlates of the risk of tuberculosis disease in this population. Due to the limited amount of sample available from each infant, preliminary work was necessary to determine which assays and conditions give the most useful information. Peripheral blood mononuclear cells (PBMC) were stimulated with antigen 85A (Ag85A) and purified protein derivative fromM. tuberculosisin anex vivogamma interferon (IFN-γ) enzyme-linked immunosorbent spot assay (ELISpot) and a Ki67 proliferation assay. The effects of a 2-h or overnight rest of thawed PBMC on ELISpot responses and cell populations were determined. Both the ELISpot and Ki67 assays detected differences between the MVA85A and placebo groups, and the results correlated well. The cell numbers and ELISpot responses decreased significantly after an overnight rest, and surface flow cytometry showed a significant loss of CD4+and CD8+T cells. Of the infants tested, 50% had a positive ELISpot response to a single pool of flu, Epstein-Barr virus (EBV), and cytomegalovirus (CMV) (FEC) peptides. This pilot work has been essential in determining the assays and conditions to be used in the correlate study. Moving forward, PBMC will be rested for 2 h before assay setup. The ELISpot assay, performed in duplicate, will be selected over the Ki67 assay, and further work is needed to evaluate the effect of high FEC responses on vaccine-induced immunity and susceptibility to tuberculosis disease.