解析皮肤T细胞淋巴瘤大细胞转化背后的基因调控网络与空间异质性

Cutaneous T cell lymphoma (CTCL) is a rare malignancy of skin-homing T cells, with Mycosis Fungoides (MF) representing 50-70% of all cases. Although MF typically follows an indolent clinical course, a subset of MF patients progresses to develop large cell transformation (LCT) that is associated with a worse outcome. Disease progression from patch/plaque (PP) lesions to transformed tumor (TT) is accompanied by profound genetic and transcriptional alterations across multiple signaling pathways, including oxidative phosphorylation and MYC, posing a significant challenge for single-agent therapeutic strategies. Epigenetic therapies such as HDAC inhibitors have shown limited efficacy, underscoring the need to identify key transcription factors (TFs) and novel gene regulatory networks (GRNs) that drive CTCL disease progression and may serve as targeted vulnerabilities. In this study, we first performed joint profiling of gene expression and chromatin accessibility using scMultiome on six MF samples (2 PP and 4 TT). By integrating TF expression, chromatin accessibility at targeting regions, and expression of nearby genes, we inferred cell-type specific GRNs across the tumor microenvironment cell types, including the malignant T cells in PP and TT lesions, and other benign immune and non-immune cells. Importantly, this analysis identified JUND-associated GRNs enriched in B cells and ATF6 networks in myeloid cells, demonstrating robust regulatory signal recovery. Applying the same framework, we identified RUNX2 and ATF2 as putative regulators of PP malignant T cells, and LEF1, IKZF2, and TCF7 as key drivers of transformed CTCL cells. We further validated the critical roles of these transcription factors in in-house and publicly available scRNAseq datasets. To investigate the spatially resolved chromatin accessibility landscape in CTCL, we generated the first spatial ATACseq dataset for this rare cancer and profiled paired PP and TT lesions from the same patient. Spatial clustering and inferred copy number analysis revealed distinct epigenetic clone structures, including PP-dominant clones, TT-dominant clones, and shared clones between PP and TT. Notably, GRNs enriched in PP or TT by scMultiome were recapitulated within PP-dominant and TT-dominant spatial ATAC clones, respectively, highlighting consistent regulatory programs across modalities. Together, these results uncover previously unrecognized GRNs associated with MF progression and provide the first spatially resolved epigenetic map of CTCL. Our findings demonstrated potential regulators and clonal heterogeneity that may inform the development of novel therapeutic strategies, including precision epigenetic or combination therapies.