Background
Merkel cell carcinoma (MCC) is a highly aggressive neuroendocrine skin cancer linked to Merkel cell polyomavirus (MCPyV) or ultraviolet radiation-induced mutations in RB1 and TP53. While immunotherapy targeting the programmed cell death-1 (PD-1) pathway has shown promise, response rates in single-agent anti-PD1-based therapies are over 50%. Nevertheless, resistance to immunotherapy remains a significant challenge. The lack of targeted agents or highly effective strategies for recurrent or immunotherapy-refractory MCC further complicates the situation. Developing an appropriately credentialed immunocompetent animal model is crucial to tackle these issues. Our novel mouse model provides a platform for investigating immunotherapy resistance and developing therapeutic strategies, thus offering a significant step forward in the field.
Methods and Results
We undertook a unique and successful approach to developing our novel mouse model. Based on Foundation One profiles testing over 148 UV-driven Merkel cell carcinomas, which revealed enriched loss or mutation of RB1 and amplification of L-MYC, we adapted the RPM model (Rb1fl/fl; Trp53fl/fl; LSL-MycT58A/T58A) previously used in small cell lung carcinoma (SCLC). This adaptation led to generating K14-CreERT2; Rb1fl/+; Trp53fl/+; LSL-MycT58A/+ compound mutant mice, to which we applied topical tamoxifen to newborns. The result was the development of tumors that histologically resembled human MCC and expressed MCC markers (CK20, CD56, and chromogranin A). RNA sequencing compared our mouse model-derived cell lines to human MCC and SCLC cell lines, demonstrating that mouse MCC cells clustered with human MCC, distinct from other skin cancers (figure 1). Cell lines derived from our model grew as spheroids in neural stem cell support media, and allograft models were assessed from two cell lines. Anti-PD-1 efficacy was tested, showing substantial tumor regression in a subset of animals with one cell line while the second cell line exhibited primary or acquired resistance (figure 2). It is related with the predominant cold phenotype of spontaneous mouse MCC tumors (figure 3). This finding opens avenues for research into MCC immune evasion and potential therapeutic strategies.
Conclusions
Our novel mouse model offers a valuable platform for investigating immunotherapy resistance in MCC, providing insights into tumor microenvironment dynamics and potential therapeutic strategies. The complexity of immune evasion in MCC underscores the need for tailored treatment approaches to overcome resistance. This model’s potential to guide novel immunotherapeutic interventions and strategies is promising, making it a crucial preclinical tool for advancing MCC treatment.
Abstract 1363 Figure 1
tSNE plot showing clustering of RNA-seq data from MCC mouse tumors (blue) and cell lines (turquoise) with human MCC (brown) and SCLC (purple), demonstrating distinct separation from HNSCC (green) and melanoma (pink) samples
Abstract 1363 Figure 2
Two cell lines from K14-RPM-Cre were injected into C57B/L6 mice. Cell line 439-4120R fully responded to anti-PD1 monotherapy, while 438-856 showed a partial response. A 50% complete response rate was observed (right), n=10
Abstract 1363 Figure 3
FACS-based immunophenotyping of spontaneous mouse MCC tumors reveals predominantly a cold phenotype. The heatmap shows the relative expression of immune checkpoint molecules and cell populations within the TME of n=3 spontaneous MCC tumors (K14-RPM-het)