Tumor RNA sequencing identifies a group of Mycosis Fungoides patients with failure of skin-directed therapies

Mycosis fungoides (MF), the most common type of primary cutaneous T-cell lymphoma, has a heterogeneous clinical course, and the factors determining disease progression from the indolent to the aggressive disease stages are not well known. Our previous study described prognostic groups of MF cases according to the transcription factors Twist and Zeb1. Using Twist expression, we differentiated distinct groups based on their global RNA expression. Using the same dataset with 40 formalin-fixed paraffin-embedded MF samples, we performed RNA sequencing using bioinformatics tools. Based on the consensus clustering, we defined four clusters with different disease pathologies and prognostic groups. Cluster 3 showed the shortest time to skin-directed therapy failure (p=0.031), and with Cluster 1 they included all the MF-related deaths (2/40). Cluster 3 presented a higher proportion of lymphocytic infiltrates per sample (p<0.001), with the highest percentages (>60 %) and a higher rate of CD30+ tumor cells (over 10 %, p=0.048). Cluster 3 samples also showed upregulated genes related to keratinocyte, basal cell, and melanocyte activities. This is the first study to describe the association between gene expression, tumor cell density, reactive cell infiltration, and failure of skin-directed therapies Overall design: Using the same dataset with 40 formalin-fixed paraffin-embedded MF samples, we performed RNA sequencing using bioinformatics tools. Based on the consensus clustering, we defined four clusters with different disease pathologies and prognostic groups.