Immune activity in Merkel cell carcinoma

Analyses of tertiary lymphoid structures observed in cases of Merkel cell carcinoma showing spontaneous regression

Merkel cell carcinoma (MCC) is a high-grade skin cancer, but spontaneous regression is observed at a markedly higher frequency than in other carcinomas. Although spontaneous regression is a phenomenon that greatly impacts treatment planning, we still cannot predict it. We previously reported on the prognostic impact of the presence or absence of tertiary lymphoid structures (TLS) and of Merkel cell polyomavirus (MCPyV) infection. To learn more about the spontaneous regression of MCC, detailed analyses were performed focusing on spontaneous regression cases. We collected 71 Japanese patients with MCC including 6 cases of spontaneous regression. Samples were analysed by immunostaining, spatial single-cell analysis using PhenoCycler, and RNA sequencing using the next-generation sequencer (NGS). All 6 cases of spontaneous regression were positive for MCPyV. TLS was positive in all 5 cases analysed. Spatial single-cell analyses revealed that PD-L1-positive tumour cells were in close proximity to CD20-positive B cell and CD3-, 4-positive T cells. Gene set enrichment analysis between MCPyV-positive and TLS-positive samples and other samples showed significantly high enrichment of "B-cell-mediated immunity" gene sets in the MCPyV-positive and TLS-positive groups. In conclusion, TLS may play an important role in the spontaneous regression of MCC.

Tertiary Lymphoid Structures and Chemokine Landscape in Virus-Positive and Virus-Negative Merkel Cell Carcinoma

Tertiary lymphoid structures (TLSs) are used as biomarkers in many cancers for predicting the prognosis and assessing the response to immunotherapy. In Merkel cell carcinoma (MCC), TLSs have only been examined in MCPyV-positive cases. Here, we examined the prognostic value of the presence or absence of TLSs in 61 patients with MCC, including MCPyV-positive and MCPyV-negative cases. TLS-positive samples had a significantly better prognosis than TLS-negative samples. MCPyV-positive samples had a good prognosis with or without TLSs, and MCPyV-negative/TLS-positive samples had a similarly good prognosis as MCPyV-positive samples. Only MCPyV-negative/TLS-negative samples had a significantly poor prognosis. All cases with spontaneous regression were MCPyV-positive/TLS-positive. We also performed a comprehensive analysis of the chemokines associated with TLS formation using next-generation sequencing (NGS). The RNA sequencing results revealed 5 chemokine genes, CCL5, CCR2, CCR7, CXCL9, and CXCL13, with significantly high expression in TLS-positive samples compared with TLS-negative samples in both MCPyV-positive and MCPyV-negative samples. Only 2 chemokine genes, CXCL10 and CX3CR1, had significantly different expression levels in the presence or absence of MCPyV infection in TLS-negative samples. Patients with high CXCL13 or CCL5 expression have a significantly better prognosis than those with low expression. In conclusion, the presence of TLSs can be a potential prognostic marker even in cohorts that include MCPyV-negative cases. Chemokine profiles may help us understand the tumor microenvironment in patients with MCPyV-positive or MCPyV-negative MCC and may be a useful prognostic marker in their own right.