Caroline E. van der Minne

Neoantigen landscape dynamics during human melanoma–T cell interactions

Recognition of neoantigens that are formed as a consequence of DNA damage is likely to form a major driving force behind the clinical activity of cancer immunotherapies such as T-cell checkpoint blockade and adoptive T-cell therapy. Therefore, strategies to selectively enhance T-cell reactivity against genetically defined neoantigens are currently under development. In mouse models, T-cell pressure can sculpt the antigenicity of tumours, resulting in the emergence of tumours that lack defined mutant antigens. However, whether the T-cell-recognized neoantigen repertoire in human cancers is constant over time is unclear. Here we analyse the stability of neoantigen-specific T-cell responses and the antigens they recognize in two patients with stage IV melanoma treated by adoptive T-cell transfer. The T-cell-recognized neoantigens can be selectively lost from the tumour cell population, either by overall reduced expression of the genes or loss of the mutant alleles. Notably, loss of expression of T-cell-recognized neoantigens was accompanied by development of neoantigen-specific T-cell reactivity in tumour-infiltrating lymphocytes. These data demonstrate the dynamic interactions between cancer cells and T cells, which suggest that T cells mediate neoantigen immunoediting, and indicate that the therapeutic induction of broad neoantigen-specific T-cell responses should be used to avoid tumour resistance.

Successful treatment of metastatic melanoma by adoptive transfer of blood-derived polyclonal tumor-specific CD4+ and CD8+ T cells in combination with low-dose interferon-alpha

A phase I/II study was conducted to test the feasibility and safety of the adoptive transfer of tumor-reactive T cells and daily injections of interferon-alpha (IFNα) in metastatic melanoma patients with progressive disease. Autologous melanoma cell lines were established to generate tumor-specific T cells by autologous mixed lymphocyte tumor cell cultures using peripheral blood lymphocytes. Ten patients were treated with on average 259 (range 38–474) million T cells per infusion to a maximum of six infusions, and clinical response was evaluated according to the response evaluation criteria in solid tumors (RECIST). Five patients showed clinical benefit from this treatment, including one complete regression, one partial response, and three patients with stable disease. No treatment-related serious adverse events were observed, except for the appearance of necrotic-like fingertips in one patient. An IFNα-related transient leucopenia was detected in 6 patients, including all responders. One responding patient displayed vitiligo. The infused T-cell batches consisted of tumor-reactive polyclonal CD8+ and/or CD4+ T cells. Clinical reactivity correlated with the functional properties of the infused tumor-specific T cells, including their in vitro expansion rate and the secretion of mainly Th1 cytokines as opposed to Th2 cytokines. Our study shows that relatively low doses of T cells and low-dose IFNα can lead to successful treatment of metastatic melanoma and reveals a number of parameters potentially associated with this success.

Detection and Functional Analysis of CD8+ T Cells Specific for PRAME: a Target for T-Cell Therapy

Purpose: Preferentially expressed antigen on melanomas (PRAME) is an interesting antigen for T-cell therapy because it is frequently expressed in melanomas (95%) and other tumor types. Moreover, due to its role in oncogenic transformation, PRAME-negative tumor cells are not expected to easily arise and escape from T-cell immunity. The purpose of this study is to investigate the usefulness of PRAME as target for anticancer T-cell therapies. Experimental Design: HLA-A*0201-subtyped healthy individuals and advanced melanoma patients were screened for CD8+ T cells directed against previously identified HLA-A*0201-binding PRAME peptides by IFN-γ enzyme-linked immunosorbent spot assays and tetramer staining. PRAME-specific T-cell clones were isolated and tested for recognition of melanoma and acute lymphoid leukemia (ALL) cell lines. PRAME mRNA expression was determined by quantitative real-time reverse transcription-PCR. Results: In 30% to 40% of healthy individuals and patients, PRA100-108-specific CD8+ T cells were detected both after in vitro stimulation and directly ex vivo after isolation by magnetic microbeads. Although CD45RA− memory PRA100-108-specific T cells were found in some individuals, the majority of PRA100-108-tetramer+ T cells expressed CD45RA, suggesting a naive phenotype. PRA100-108-tetramer+ T-cell clones were shown to recognize and lyse HLA-A*0201+ and PRAME+ melanoma but not ALL cell lines. Quantitative real-time reverse transcription-PCR showed significantly lower PRAME mRNA levels in ALL than in melanoma cell lines, suggesting that PRAME expression in ALL is below the recognition threshold of our PRA100-108-tetramer+ T cells. Conclusion: These data support the usefulness of PRAME and in particular the PRA100-108 epitope as target for T-cell therapy of PRAME-overexpressing cancers.

A placebo-controlled randomized HPV16 synthetic long-peptide vaccination study in women with high-grade cervical squamous intraepithelial lesions

The aim of this study was to investigate the capacity of an HPV16 E6/E7 synthetic overlapping long-peptide vaccine to stimulate the HPV16-specific T-cell response, to enhance the infiltration of HPV16-specific type 1 T cells into the lesions of patients with HPV16+ high-grade cervical squamous intraepithelial lesion (HSIL) and HPV clearance. This was a placebo-controlled randomized phase II study in patients with HPV16-positive HSIL. HPV16-specific T-cell responses were determined pre- and post-vaccination by ELISPOT, proliferation assay and cytokine assays in PBMC and HSIL-infiltrating lymphocytes, and delayed-type hypersensitivity skin tests. Motivational problems of this patient group to postpone treatment of their premalignant lesions affected the inclusion rates and caused the study to stop prematurely. Of the accrued patients, 4 received a placebo and 5 received 1–2 vaccinations. Side effects mainly were flu-like symptoms and injection site reactions. A strong HPV-specific IFNγ-associated T-cell response was detected by ELISPOT in all vaccinated patients. The outcome of the skin tests correlated well with the ELISPOT analysis. The cytokine profile associated with HPV16-specific proliferation varied from robust type 1 to dominant type 2 responses. No conclusions could be drawn on vaccine-enhanced T-cell infiltration of the lesion, and there was no HPV clearance at the time of LEEP excision. Thus, vaccination of HSIL patients results in increased HPV16-specific T-cell immunity. Further development of this type of treatment relies on the ability to motivate patients and in the reduction in the side effects.

Identification of Multiple HLA-DR-Restricted Epitopes of the Tumor-Associated Antigen CAMEL by CD4+ Th1/Th2 Lymphocytes

CD4+ Th cells play an important role in the induction and maintenance of adequate CD8+ T cell-mediated antitumor responses. Therefore, identification of MHC class II-restricted tumor antigenic epitopes is of major importance for the development of effective immunotherapies with synthetic peptides. CAMEL and NY-ESO-ORF2 are tumor Ags translated in an alternative open reading frame from the highly homologous LAGE-1 and NY-ESO-1 genes, respectively. In this study, we investigated whether CD4+ T cell responses could be induced in vitro by autologous, mature dendritic cells pulsed with recombinant CAMEL protein. The data show efficient induction of CAMEL-specific CD4+ T cells with mixed Th1/Th2 phenotype in two healthy donors. Isolation of CD4+ T cell clones from the T cell cultures of both donors led to the identification of four naturally processed HLA-DR-binding CAMEL epitopes: CAMEL1–20, CAMEL14–33, CAMEL46–65, and CAMEL81–102. Two peptides (CAMEL1–20 and CAMEL14–33) also contain previously identified HLA class I-binding CD8+ T cell epitopes shared by CAMEL and NY-ESO-ORF2 and are therefore interesting tools to explore for immunotherapy. Furthermore, two CD4+ T cell clones that recognized the CAMEL14–33 peptide with similar affinities were shown to differ in recognition of tumor cells. These CD4+ T cell clones recognized the same minimal epitope and expressed similar levels of adhesion, costimulatory, and inhibitory molecules. TCR analysis demonstrated that these clones expressed identical TCR β-chains, but different complementarity-determining region 3 loops of the TCR α-chains. Introduction of the TCRs into proper recipient cells should reveal whether the different complementarity-determining region 3α loops are important for tumor cell recognition.

Influenza matrix 1‐specific human CD4+FOXP3+ and FOXP3− regulatory T cells can be detected long after viral clearance

Control and termination of infection with Influenza A virus is associated with increased IL‐10 production in mouse models. Notably, IL‐10 can be produced by Treg. Therefore, we investigated whether the population of IL‐10‐producing influenza‐specific CD4+ T cells comprised Treg as they are potent suppressors of the adaptive immune response. Influenza‐specific IL‐10‐producing T cells were detected in all human donors displaying influenza‐specific immunity. Isolation of Matrix 1 protein‐specific IL‐10‐producing T‐cell clones revealed that a substantial proportion of these T‐cell clones displayed the capacity to suppress effector cells, functionally identifying them as Treg. Both FOXP3+ and FOXP3− CD4+ Treg were isolated and all were able to exert their suppressive capacity when stimulated with cognate antigen, including influenza virus‐infected cells. In vitro suppression was not mediated by IL‐10 but involved interference with the IL‐2 axis. The isolated Treg suppressed amongst others the IL‐2 production of influenza‐specific T‐helper cells as well as partially prevented the upregulation of the high‐affinity IL‐2 receptor on CD8 effector cells. So far the induction of virus‐specific Treg has only been studied in the context of chronic viral infections. This study demonstrates that virus‐specific Treg can also be induced by viruses that are rapidly cleared in humans.

Induction of CAMEL/NY-ESO-ORF2-specific CD8+ T cells upon stimulation with dendritic cells infected with a modified Ad5 vector expressing a chimeric Ad5/35 fiber

Delivery of the full-length tumor antigen might be more successful in immunotherapy than single peptides and has the advantage that patients no longer need to be selected for their HLA type. In this study, we tested the in vitro induction of CAMEL/NY-ESO-ORF2-specific T cells by dendritic cells infected with an adenovirus (Ad) type 5 vector containing the fiber shaft and knob of human serotype Ad35 (Ad5F35 vector). Our data show induction of CD8+ T cells specific for the known HLA-A*0201-binding CAMEL/NY-ESO-ORF21–11 epitope by DC infected with Ad5F35-CAMEL, but not by DC pulsed with the recombinant CAMEL protein. In one healthy donor, even CD8+ T cells specific for a new HLA-B7-binding CAMEL/NY-ESO-ORF246–54 epitope were raised. In conclusion, the in vitro induction of CAMEL/NY-ESO-ORF2-specific CD8+ T cells in healthy donors by DC infected with Ad5F35-CAMEL strongly supports further investigation of the Ad5F35 vector as a vehicle for gene transfer into DC for the generation of tumor antigen-specific CD8+ T cell responses in vivo.