Title: Dissecting phenotypic heterogeneity of human melanoma: building a rationale for active immunotherapies overcoming immunologically-induced dedifferentiation.

Project Coordinator:
Massimo Guidoboni (Italy) IRCCS IRST - Immunotherapy and Cell Therapy, Meldola (FC)

Project Partners:
Irwin DAVIDSON (France) IGBMC, CNRS/INSERM, Illkirch
Rosalie LUITEN (Netherlands) Dept. of Dermatology, Academic Medical Center, University of Amsterdam
Jon Amund KYTE (Norway) Dept. of Oncology Oslo University Hospital Radiumhospitalet, Oslo

Project Abstract:
Inter- and intralesional tumor heterogeneity is commonly seen in metastatic melanoma, likely playing a major role in resistance to therapy, including immunotherapy. During immunotherapy, many patients experience "mixed response",with some tumor lesions regressing and others progressing. Whereas vaccines targeting a few antigens can "passively" select for antigen-negative tumor cells, even broader spectrum immunotherapies (whole tumor vaccines or immunomodulating antibodies) fail in clearing the tumor in most cases. In this respect, rare tumor cells with stemness properties and distinct expression profiles likely exist in tumor tissue: these cells, being less immunogenic, would be hardly targeted by antitumor immunity. Finally, tumor-specific cytotoxic T lymphocytes can induce dedifferentiation of melanoma cells, possibly underlying tumor relapse after initial response to immunotherapy.
Current immunotherapies induce immune response against the "average" antigenic composition of tumor bulk, not being likely able to "see" antigens expressed by rare tumor cells. Moreover, DC vaccines loaded with tumor lysate elicit immunity against the pretherapy tumor tissue,not taking into account the dynamic changes induced by the treatment on tumor cells. Therefore, the identification of antigen sets expressed by distinct melanoma cell subpopulations will allow developing new immunotherapies that concurrently target different subpopulations, including rare or therapyinduced ones. This approach will thereby target the cellular heterogeneity of tumors and minimize the risk of immune-escape of antigen-loss variants.
Primary: to identify nonoverlapping sets of genes expressed by distinct melanoma cell subpopulations and their changes during immunotherapy.
Secondary: to validate as antigenic targets the proteins encoded by the genes overespressed by the distinct melanoma cell subpopulations identified in the primary aim.
- Melanoma cell collection: melanoma cells will be obtained by fine needle aspirates (FNA) of metastatic lesions before and after immunotherapy.
- Isolation of single cells: CD45- cells obtained by FNA will be isolated with a microchip-based cell isolation system.
- Single-cell transcriptomics: SMARTSeq RNA-Seq libraries will be generated for isolated single melanoma cells, tagmented using Nextera XT, and sequenced.
- Validation of candidate target antigens: expression of proteins encoded by genes overexpressed in the distinct melanoma cell subpopulations will be tested in metastatic melanoma lesions. Overexpressed proteins will be then tested for their ability to induce CTL cultures in vitro.
The identification of nonoverlapping gene sets stably expressed by different melanoma cell subpopulations and their validation as antigenic targets will allow designing new immunotherapies concurrently eliciting immune response against the majority of the patients' tumor cells.

Final Summary:
Inter- and intralesional tumor heterogeneity is commonly seen in metastatic melanoma, likely playing a major role in resistance to therapy, immunotherapy included. In particular, not all patients initially respond to immunotherapy (primary resistance) and even those who initially respond to treatment can later show disease progression (secondary resistance).

In this respect, both primary and secondary resistance to immunotherapy might be due to the presence, in the tumor bulk, of tumor cells with stemness properties and a phenotype different from their differentiated counterpart. These variants are less represented in most melanoma patients and are less immunogenic, being hardly targeted by specific immunity. In addition, it has been recently reported that secondary resistance may be sustained by immunologically-mediated dedifferentiation of tumor cells which could evade therapy-induced immune response.

In this line, spontaneous or therapy-induced antitumor immune response would be directed against the more common and immunogenic differentiated tumor cell subpopulations, not being likely able to "see" antigens expressed by the less common and less immunogenic dedifferentiated cells.

This research project is identifying, by RNA sequencing of single melanoma cells, the genes expressed by the distinct melanoma cell subpopulations and to validate their encoded proteins as antigenic targets. The identification of these targets will ultimately allow designing new and more effective immunotherapeutic approaches that, by concurrently targeting the different melanoma cell variants, will reduce both primary resistance to therapy and late relapse after initial response.

(Project funded under JTC 2014)