Document resource
Background Given the success of checkpoint inhibitor therapy in the advanced setting in Merkel cell carcinoma (MCC), there has been interest in exploring immunotherapy in the neoadjuvant setting where paired tissues obtained at pretreatment biopsy and at surgery permits molecular interrogation of mechanisms of response and resistance.Methods We conducted a single center, phase II open label trial ( NCT04869137) in patients (pts) with resectable stage II-IV Merkel cell carcinoma. All pts were to receive six weeks of neoadjuvant therapy with pembrolizumab 200mg IV q3 weeks plus lenvatinib 20mg PO daily before planned surgery ± adjuvant radiation therapy. Pathological complete response (pCR) rate was the primary endpoint of the study. Pre- and post-neoadjuvant treatment tissue collection was performed for analyses of molecular correlates of response.Results Twenty-six pts were enrolled between 06/2021 and 09/2024, including 5 (19.2%) with clinical stage II disease, 20 (76.9%) with stage III, and 1 (3.8%) with stage IV. Of the total study cohort, 15 of the 26 pts (57.7%) achieved a pCR. Here, we report preliminary molecular data based on bulk RNAseq and spatial transcriptomics on patients with FFPE material.Bulk RNA sequencing performed on 41 samples from 25 patients revealed upregulation of innate immune activation, pro-inflammatory macrophages, and presence of B-cells to be associated with response (figure 1). Angiopoietins were also significantly increased in responders. Pathway analysis (GSEA) showed upregulation of interferon and other immune pathways in responders and upregulation of E2F and MYC pathways (critical for MCC oncogenesis) in non-responders. (figure 2).Spatial transcriptomics (10X Xenium) was performed on 21 samples from 12 patients (8 matched pre-/post- pairs), with an average of 690,915 high-quality cells per sample, 13 major cell types (tumor and immune) and 27 distinct subtypes (including multiple B- and T-cell subsets) (figure 2). In pre-treatment tumors, non-responders were enriched in developmental stemness programs (SIX1, NKX2-1), oncogenic growth factor signaling (IGF/PI3K/MAPK, WNT, Hedgehog, FGF), and neuroendocrine/neuronal pathways. CD8+ T cell infiltration was higher in responders and increased after treatment in both groups, with CD8+ T cells positioned significantly closer to tumor cells in post-treatment samples and in responders.Conclusions Lenvatinib plus pembrolizumab demonstrated encouraging efficacy when used as neoadjuvant therapy for Merkel cell carcinoma. Transcriptomic profiling on matched pairs of pre- and post-therapy tissue was feasible in the majority of patients. Our preliminary findings implicate immune, stemness, growth factor, and angiogenic transcriptional modules as potentially targetable pathways to ameliorate resistance.Acknowledgements The authors acknowledge the support from Merck for support of the trial associated analysis of biospecimens. This work has been supported in part by the Molecular Genomics and Tissue Core Facility at the H. Lee Moffitt Cancer Center & Research Institute, an NCI designated Comprehensive Cancer Center (P30-CA076292).Trial Registration US Clinical Trials NCT04869137Ethics Approval The results are de-identified molecular data linked only to response to therapy and obtained in the course of protocols MCC 20773 (IRB Pro00049454) and 19191 (IRB Pro00022458) the H. Lee Moffitt Cancer Center & Research Institute.Consent The results are de-identified molecular data linked only to response to therapy and obtained in the course of protocols MCC 20773 (IRB Pro00049454) and 19191 (IRB Pro00022458) the H. Lee Moffitt Cancer Center & Research Institute.Abstract 1346 Figure 1Molecular immune profiling in neoadjuvant lenvatinib + pembrolizumab trial. Samples profiled are (top). Comparisons of responders (right) and non-responders (left) across PRE-therapy (middle) and POST-therapy specimens (bottom) are shown.Abstract 1346 Figure 2Pathway and spatial transcriptomics reveals key pathways driving response. GSEA analysis highlights key pathways associated with response (top). 10X Xenium profiling identifies major cell subtypes (middle) mapped in space (bottom).