Physical Poster + E-Poster Presentation 34th Lorne Cancer Conference 2022

 Detection of circulating tumour DNA in high-risk paediatric acute lymphoblastic leukaemia (ALL) (#131)

Louise Doculara 1 , Richard Lock 1 , Narges Bayat 1
  1. Children's Cancer Institute, Kensington, NSW, Australia

Mixed lineage leukaemia-rearranged ALL (MLL-r ALL) is a high-risk malignancy occurring in 80% of infants with ALL, in which overall survival is poor. The strongest prognostic factor for ALL relapse is the persistence of minimal residual disease (MRD) throughout treatment in the bone marrow (BM). However, the sensitivity of current MRD techniques relies on the volume of BM sample that is collected at any one time. Also, BM biopsies are invasive and fail to predict relapse in extramedullary sites. Analysis of circulating tumour DNA (ctDNA) has shown promise for MRD detection and assessment of treatment response in adult and solid cancers. This is not the case for childhood cancer, partly due to the lack of preclinical models for ctDNA analysis. Therefore, we sought to assess the predictive value of ctDNA in infant MLL-r ALL. We aimed to design patient-specific assays for MLL fusion detection in MLL-r ALL and assess the dynamics of ctDNA during disease progression in a preclinical model of infant MLL-r ALL. Real-time quantitative polymerase chain reaction (RQ-PCR) patient-specific MLL fusion assays were validated in the genomic DNA of 5 MLL-r ALL patient-derived xenografts (PDXs). Sanger sequencing, melt curve analysis and gel electrophoresis confirmed their specificity and sensitivity. Immune-deficient mice were inoculated with MLL-r ALL PDXs to establish an orthotopic model of the disease and blood samples were collected throughout disease progression. TapeStation analysis of total cell-free DNA (cfDNA) showed significantly higher cfDNA levels in MLL-r plasma samples compared to naïve mice (p<0.001).  With droplet digital PCR (ddPCR), we showed a significant concordance between ctDNA concentration and leukaemia burden, measured by flow cytometry, in mice engrafted with all 5 MLL-r PDXs (r2 = 0.75, 0.77, 0.84, 0.87 and 0.88, p<0.001). MLL-r ctDNA was not detected in naïve mice and mice engrafted with a non-MLL-r PDX. In preclinical models, ctDNA allows for longitudinal measurement of disease progression, highlighting the value of ctDNA detection for disease monitoring in MLL-r ALL.