Identification of immunotherapeutic targets in melanoma

Lycke Woittiez

S1418165

Research Internship

2 February 2009 – 26 June 2009

The Wistar Institute, Philadelphia, USA

Cancer immunology

Local supervisor: Shyam Somasundaram

(Senior scientist)

Principal Investigator: Professor D.V.M. Herlyn

Faculty supervisor: Dr. W.M. Smit

Contents

Summary

1.Introduction

1.1Melanoma

1.2Immunology and melanoma

1.3Immunotherapy in melanoma

1.4Modified peptides

1.5Improving the immune response

1.6BRAF mutation

1.7Goal and hypothesis

2Material and methods

2.1Patients

2.2Cell lines

2.3Determination of the HLA-A2 status.

2.4Determination of the BRAF mutation status

2.4.1DNA purification

2.4.2Polymerase Chain Reaction (PCR)

2.4.3Gel electrophoresis

2.4.4TOPO cloning

2.4.5Miniprep

2.4.6Restriction digestion

2.4.7Sequencing

2.5Peptide binding assay

2.6PLG-encapsulation.

2.7T cell proliferation assay.

2.8Phenotyping of T-cells

2.9Cytotoxicity assay

3Results

3.1Determination of the HLA-A2 status.

3.2Determination of the BRAF mutation status.

3.2.1DNA Purification

3.2.2Polymerase Chain Reaction

3.2.3Gel electrophoresis

3.2.4TOPO cloning

3.2.5Miniprep

3.2.6 Restriction digestion

3.2.7Sequencing

3.3Peptide binding assay

3.4T-cell proliferation assay

3.5Phenotyping of T-cells

3.6Cytotoxicity assay

4Discussion

4.1Determination of the HLA-A2 status.

4.2Detection of the BRAF mutation status.

4.3Peptide binding assay

4.4T cell proliferation assay

4.5Phenotyping of T-cells

4.6Cytotoxicity assay

4.7General conclusion

5.List of abbrevations

6.Citations

7.Word of thanks

8.Summary of research in Dutch

Summary

Introduction

Melanoma is a common malignant tumor of the skin, with a bad prognosis when disseminated disease is present. Conventional therapies are of little benefit becausemost tumor cells are relatively chemoresistant. Lymphocyte infiltrates in or around the melanoma are associated withprolonged survival, suggesting a cellular immune response against melanoma. Active immunotherapy has been shown to stimulate the immune response and has been subject of previous studies as therapeutic modality in malignant melanoma. Tumor antigenic peptides can stimulate the immune response. Modification of these peptides may result in a stronger binding to HLA molecules, resulting in an increased immune response.

Previous studies showed that peptides spanning the BRAFV600E mutation, a mutation in the Map-kinase pathway that is present in 70% of melanoma cells, were immunogenic. In this study we determined whether modified BRAFV600E peptides that bind stronger to HLA molecules, may induce a stronger immune response than unmodified BRAFV600E peptides in melanoma.

Material and methods

Twenty-three melanoma patients were included in this study. The HLA-A2 status was determined by flow cytometry. Genomic DNA was isolated from tumor cells and used to determine the BRAFV600E mutation status by amplification in a Polymerase Chain Reaction (PCR), followed by sequencing of the amplified DNA fragment using the primer SP6 and reverse primer T7, and M13 forward and M13 reverse primer.

A T2 peptide binding assay was performed to determine the binding ability of modified peptides to T2 cells, to select the peptides with a high binding ability.

A standard lymphocyte proliferation assay (3HT-thymidine incorporation assay) was used to determine the extent of proliferation of T-cells cultured with various peptides. Proliferating lymphocytes were phenotyped to determine whether it were CD4 (Th) or CD8 (CTL) cells.

The proliferating cells were expanded further in the presence of IL-2 and used in a cytotoxicity assay, to determine whether they were efficient in killing tumor cells.

Results

Lymphocytes of 14 out of 23 patients were tested for their HLA-A2 status. Four (29%) of those patients were HLA-A2 positive, and thus could be used for further experiments. Another nine patients were tested positive for HLA-A2 before and were thereforeincluded in this study.

Four of the 11 patients tested for their BRAF status, had the V600E mutation.

The T2 binding assay showed a strong binding of modified peptide 4, which is altered in both anchor residues, compared to the unmodified peptides. The lymphocyte proliferation assay showed a low proliferation in all tested cells, except WM35. The phenotype of the proliferating lymphocytes was mostly of the CD8 subtype. The cytotoxicity assay gave inconsistent results, with no significant difference in cytotoxicity of the modified peptides compared to the unmodified peptides.

Conclusion

Modified melanoma peptide 4 bound stronger in a peptide binding assay than the unmodified peptides and may therefore be a good candidate for further studies with specific cellular immunotherapy in patients with malignant melanoma.

1.Introduction

1.1Melanoma

Melanoma is a malignant tumor of the skin. It is the 6th most common malignancy in men and the 7th most common in women. The incidence of melanoma has risen markedly during the second half of the 20th century (1, 2).

Melanoma is divided in four stages. Stage I and II represent localized disease, stage III involves regional nodal metastases and stage IV indicates the presence of distant metastases. The stage of melanoma at presentation is an important determinant of the prognosis. If melanoma is detected early it is highly curable, with a 5-year survival rate of localized disease of 99%. When distant metastases are present, the 5 year survival rate is only 15%. Approximately 3% of the patients present with distant metastases.

The survival of melanoma patients has not changed in the past 35 years. This is mostly due to the fact that there have not been any true advances in the treatment strategy for melanoma (1, 2).

The current management options for melanoma are surgery, radiation, chemotherapy and immunotherapy. Those therapies can be used individually or in combination (2, 3).

Surgery is the treatment of choice for the resection of solitary lesions or solitary metastatic disease and can significantly improve survival in these patients. Radiotherapy can be effective for palliation in patients with brain metastases, spinal cord compression or painful bone metastases (3).

Dacarbazine is the only approved chemotherapeutic agent for metastatic melanoma in the USA. The response rate is 15-20%. The combination of different chemotherapeutic regimens has approximately the same response rate. Major disadvantages of chemotherapy are the toxic side effects that often occur and the chemoresistance of the cancer cells(2, 3).

Traditionally, anti-cancer drugs are designed to interfere with DNA-synthesis and the cell cycle. This method affects rapidly dividing cells, like cancer cells, more than normal cells. However, these therapies have many side effects because other rapid dividing cells in the body, such as bone marrow cells, are also affected. A new generation of experimental drugs targets signaling pathways and DNA repair enzymes. Drugs targeting the Map kinase pathway are currently tested in clinical trials. Another new treatment strategy for melanoma is the use of immunotherapy.

1.2Immunology and melanoma

The concept of immunosurveillance indicates that the immune system can recognize precursors of cancer. In most cases, the immune system will destroy those malignant precursor cells. However, sometimes cancer does develop, despite the immune response (4).

The immune system identifies tumor cells through the recognition of tumor associated antigens (TAA) that are presented as short peptides in the context of HLA class I or II. These complexes can be expressed on the tumor cell itself, resulting in direct priming of T cells upon recognition, or on professional antigen presenting cells (APCs) that have picked up antigens and processed it for indirect priming of T-cells.

Melanoma associated antigens (MAAs) can generate both humoral and cell-mediated immune responses (5, 6). The immune response in melanoma patients can cause spontaneous regression of melanomas, probably due to infiltrating T cells. T cell infiltrates in primary melanoma are prognostic of disease outcome, and tumor infiltrating lymphocytes (TILs) have been associated with regression of established metastases in 15-30% of the patients. Antigen-specific CD8+ T-lymphocytes are the major subpopulation of TILs (5, 7-10). Disease free status in melanoma patients was associated with a strong T-cell reaction. Unfortunately, the spontaneous T-cell response against melanoma associated antigens is rather weak in most patients (9).

The above mentioned observations illustrate the important role of the immune system in the prevention and regulation of cancer. Therefore, investigators hypothesized that therapeutic modalities that induce or modulate the immune response might be beneficial for melanoma patients. This led to the introduction of immunotherapy as treatment modality for melanoma patients.

1.3Immunotherapy in melanoma

Immunotherapy is a relatively new treatment strategy for melanoma. Tumor-specific T cells, spontaneous regression of primary tumors and regression of metastatic lesions induced by immune-modulatory therapies have been shown in humans in clinical trials(8). Immunotherapy can be specifically targeted to tumor tissues, leaving normal tissues intact (11). The most attractive targets for immunotherapy are cell characteristics on which cancer cells are dependent for progression and survival(12).

Immunization can be performed both passive and active. The most common strategy used in melanoma is active immunization. There are two main forms of active immunization, i.e. the administration of whole tumor cells or tumor cell lysates, and vaccination with peptides.

Vaccination with whole tumor cells can be performed with both autologous and allogeneiccells. The tumor cells are irradiated to make them unable to replicate, and then re-administered to the patient with an immune adjuvant. The advantage of this method is that both humoral and cellular immunity are generated. However, developing the vaccine is a very time-consuming and difficult procedure (13). Anti melanoma-associated antigen response is generated in a fraction of the vaccinated patients. The response rate in early clinical trials was less than 10% without a clear correlation with the induced immune response (5, 7).

A more common form of vaccination, and the one used in this study, is the use of peptide or protein based vaccines. These vaccines are made of well defined melanoma associated antigens (MAA) peptide epitopes (5). After administration of the peptide, it is endocytosed by professional antigen presenting cells. Those cells present the peptides on MHC class II molecules to CD4+ cells and on MHC class I molecules to CD8+ cells. The T cells are activated and recognize antigens presented by MHC class I molecules on tumor cells. After recognition of the tumor cells, the cytotoxic T cell will attack and kill the melanoma cells (figure 1).

Different tumor-optimized peptide analogs have been identified. However, each peptide based vaccine is limited to a particular HLA subtype. Furthermore, this therapy only targets a limited number of epitopes. Gene mutation leading to antigen loss can be a problem in this form of vaccination because tumor cells will escape the immune response. Therefore it is critical to identify and target tumor antigens that are vital to maintaining a malignant phenotype(9, 13).

Figure 1: mechanism of CTL activation after peptide vaccination. Source:

Vaccines haveshown effect in other forms of cancer, such as immunotherapy targeting EGFRvIII in patients with malignant melanoma (14), and a monoclonal antibody targeting HER2/neu in breast cancer patients (15).

In previous studies in melanoma the HLA-A2 molecule was shown to be important for the melanoma cell recognition by autologous T-lymphocytes. HLA-A2 is a MHC class I molecule, that is present in most Caucasian people (16, 17).

There have been many trials using peptide based vaccines in melanoma patients, but the immune responses were detected inconsistently(13, 18). 10-30% Of the patients receiving melanoma peptides had partial or complete tumor regression (19). In some patients a detectable peptide-specific immune response was associated with regression of single metastases (5, 18). Unfortunately, the tumor response does not correlate well with the T-cell response measured in peripheral blood lymphocytes (5, 13, 20).

Objective regression and long term stabilization of metastatic melanoma have been observed in a number of vaccine trials in individual patients. In most cases, tumor regression is associated with a detectable immune response to the vaccine. But the immune responses were also induced in the population without a measurable clinical benefit (18). Therefore, cancer vaccination in patients with melanoma needs further investigation.

1.4Modified peptides

Because peptide immunogenicity correlates well with the peptide binding affinity for MHC, an effective strategy to improve the immune response is converting low affinity peptides into high affinity forms (9, 19). These modified peptides can be considerably more immunogenic in vitro and in vivo due to enhanced binding affinity to HLA molecules and slower dissociation rate (9, 19, 21). Single amino acid changes can be enough to significantly alter the binding of peptides.

T cell responsiveness to an epitope is affected not only by the affinity for the presenting MHC molecule but also by the affinity of the MHC-peptide complex for the T-cell receptor (TCR). Natural tumor antigens elicit relatively weak T cell responses. Amino acid substitutions can increase the stability of the MHC-peptide-TCR complex and can make the immune response significantly more potent (18, 22-24).

In this study we will use both modified and unmodified peptides, to evaluate their effect on the immune response.

1.5Improving the immune response

Immunotherapy is not (yet) as effective as the initial studies reported. There have been many clinical trials with peptide vaccination, but only rarely a response rate higher than 20% was found (9). This might have several reasons. The first reason is the presence of regulatory T cells. There is an important role for CD4+CD25+FoxP3+ regulatory T cells (Tregs) in limiting the anti tumor immune response in vivo. Regulatory T cells can suppress the activation of tumor antigen-specific CD8+ effector T cells in vitro, and high levels of Treg cells correlate with poor survival. The tumor uses many other immune suppressive mechanisms (25, 26).

Cancer cells downregulate or losetheir expression of HLA class I molecules. This abrogates antigen recognition and tumor cell killing by CD8+ CTL, and is considered the most common strategy exploited by tumor cells to escape T-cell control. Partial or complete loss of class I HLA expression can be detected in up to 67% of metastatic melanomas (9). Furthermore, downmodulation of molecules involved in antigen processing and presentation, such as peptide transporters associated with antigen processing and proteasome subunits has been reported (4, 9, 27).

Other mechanisms of tumor cells to evade the immune response are the downregulation of tumor-associated antigens, the development of mechanisms to avoid being killed by T cells, the downregulation of death receptors, a lack of costimulatory molecule expression, the secretion of immunosuppressive substances, the expression of pro-apoptotic molecules and the blockage of the apoptosis pathway (4, 9, 25, 27).

The most important strategies to overcome the tolerance of cancer cells for the immune systemand thus to improve the efficacy of vaccination, are the co-injection of cytokines, the use of co-stimulatory molecules, depletion of the lymphocytes pre-vaccination, immunomodulation by antineoplastic drugs, or blockage of the negative co-stimulatory molecules with anti-CTLA-4 AB (13).

One of the main approaches is the combination of immunotherapy with chemotherapy, called biochemotherapy. Chemotherapy has direct effects on the tumor or host environment, such as induction of tumor cell death, elimination of Tregs and/or enhancement of tumor cell sensitivity to lysis by CTL. These may account for the enhanced effect of immunotherapy in combination with chemotherapy. However, in phase III studies, no clear improvement of overall survival was seen with biochemotherapy (3, 23, 28).

Other therapies that are being used are IL-2 and IFN-α, that are approved as immunotherapeutic targets for melanoma in the USA. Those therapies modulate the immune response against melanoma. Furthermore, CLTA4 blockade, tremelimumab, other anti-CTLA4 monoclonal antibodies and TLR9 agonsists are used in the treatment of melanoma (7). CLTA4 is a negative costimulatory molecule, so by blocking this molecule the immune response will be improved.

1.6BRAF mutation

The RAS-RAF-MEK-ERK-MAP kinase pathway mediates cellular responses to growth signals (29, 30). RAF is a point of regulation in the MAP kinase pathway. There are different RAF kinases, all are activated by the RAS small GTPase regulatory phosphorylation events and scaffolding proteins. Different RAF isoforms combine common ad unique mechanisms to regulate RAF kinase activity (31). Mutated BRAF proteins have an elevated kinase activity, which is a crucial step in the initiation of neoplasia (30). BRAF can act as a potent oncogene in the early stages of melanoma development (32). BRAF stimulates constitutive ERK signaling, stimulating proliferation and survival and providing essential tumor growth and maintenance functions. It contributes to neo-angiogenesis and is implicated in several aspects of melanoma induction and progression (12). Melanomas are dependent on the activation of the RAS-MAP kinase pathway for replication and survival (33).

The BRAF mutation is present in 66% of melanomas but also in 82% of benign nevi (12, 30, 32, 34). Thus BRAF mutation alone is insufficient for melanoma initiation (12). It is possible that the mutation in concert with other oncogenic events will promote tumor formation in nevi (29). A single substitution in the kinase domain (V600E) accounts for 80%-95%of these mutations(29, 30, 35). Somatic mutations in the BRAF proto oncogene are present in a wide variety of malignancies, namely melanoma, papillary thyroid cancer and colon cancer (36).

Because of its tumor specificity and expression in the majority of melanomas, BRAFV600E is a potential immunotherapeutic target in melanoma. Following knockdown with RNA interference methods, melanoma cells showed profound inhibition of the MAP kinase cascade and a diminished proliferative activity. Potential methods to target BRAF function are kinase inhibition and protein depletion. Knockdown of BRAF expression and inhibition of downstream signaling in human melanoma cells causes growth arrest, promotes apoptosis and prevents colony formation (36). Increased apoptosis when BRAF expression is downregulated supports a role for oncogenic BRAF driven MEK/ERK over activation(22).

Because it is expressed intracellularly and not on the surface of tumor cells, BRAF is a target for T cells but not for B cells (antibodies). Anchors for HLA class I and class II are expressed in close vicinity of the mutation, suggesting that both CD8+ and CD4+ T lymphocytes can be induced. There is a positive correlation between BRAFV600E mutation status in melanoma lesions and the immune responses to the mutated epitope (35).

Lymphoproliferative responses to BRAFV600E are HLA-A2 restricted. The proliferating lymphocytes are cytotoxic against HLA-A2+/BRAFV600E melanoma cells. The high prevalence (approximately 50%) of HLA-A2 among melanoma patients renders HLA-A2-restricted BRAFV600E peptides attractive candidate vaccines for these patients (35, 37).

Much researchhas been conducted with agents that specifically target BRAF. Unfortunately, they all met with limited success. In this study, we use modified BRAFV600E peptides to induce an immune response in HLA-A2+ patients.