Molecular and clinical stratification of patients with angioimmunoblastic T-cell lymphoma (AITL) is unsatisfactory, which hinders the development of personalized therapies. This study aimed to identify molecular biomarkers for AITL based on peripheral cell-free DNA (cfDNA) that could be used to predict prognosis and guide treatment non-invasively. A customized panel containing 46 genes was used to study pretreatment cfDNA and paired tumour tissues in 64 Chinese AITL patients from three clinical centres, and gene mutations in cfDNA and tumour tissue were assessed for concordance (34 paired samples). Then, the association of gene mutations and prognosis was analysed, and a functional enrichment analysis was performed. The sequencing results showed good consistency between cfDNA samples and paired tissue samples. KDM5A, STAT1, FANCM, ERBB4, PIK3R5 and NSD1 were identified as novel recurrent mutations. Mutations in FANCM or combinations of RHOA, KDM5A and FAT1 were associated with poor prognosis. Additionally, functional analysis revealed that RHOAG17 might serve as a predictive biomarker of PD-1 blockade respondence. Our findings confirmed the role of cfDNA as a liquid biopsy in AITL, and revealed novel molecular determinants that can stratify patients and guide treatment options.
Keywords: angioimmunoblastic T-cell lymphoma; cell-free DNA; predictive biomarker; prognosis.
© 2023 British Society for Haematology and John Wiley & Sons Ltd.