Chapters Transcript Video Maria Capilla Guerra, MD discusses rapid diagnosis of acute leukemia with integrated epigentic and genetic profiling Acute leukemias are rapid progressing diseases, which diagnosis requires multiple modalities, including histology, flow cytometry, panel sequencing, and cytogenetics. And the problem with the current standard approaches is that they often take up to 2 weeks to be completed, which delays final diagnosis and treatment decisions. In prior work, our lab, in collaboration with the Hobson Lab has developed Marlin, a machine learning classifier that generates accurate acute leukemia diagnosis based on DNA methylation patterns when coupled to Oxford nanoport sequencing long replatform. However, methylation-based classifications alone do not provide complete information about genetic lesions, while acute leukemia classifications are based on genetic events. Oxford nanopore sequencing also provides a method called adaptive sampling, which allows for real-time active enrichment of genomic DNA molecules of interest in order to find mutations driving leukemia. In this context, by using nanopore sequencing, we established a clinical framework for rapid diagnosis of acute leukemia that integrates um disease profiling by modeling and genetic characterization of acute leukemia. To do so, we assemble a pipeline that performs active enrichment on 274 acute leukemia-related genes, looks for structural variations, looks for single nucleotide changes involved in disease, assess karyotypes, alterations, and provide marlin methylation-based diagnosis. Today, we have applied our pipeline on 20 retrospective cases and 5 real-time patient samples. Overall, our results demonstrate that we can get a million coverage of more than 75 eggs on our regions of interest, allowing us to perform further genetic analysis to explore point mutations, gene fusions, and digital karyotype. For all cases, we were able to achieve confident methylation-based predictions by Marling. Well, genetic analysis allowed us to identify the major genetic drivers in all cases, such as acute leukemia-defining mutations, for example, MPM1, P53 mutations, and acute leukemia-defining fusions, such as BCR-ABL in the case of BALL or inversion 16 in the case of AML. One of the key points of our research is that our approach provides higher resolution of genomic events such as rearrangement and large insertion deletions in comparison to current standard methods. that actually rely on share re-sequencing. As an example, with this approach, we've been able to identify two different cryptic rearrangements in two different patients not previously reported by conventional testing. Finally, the major highlight of our research is that we've been able to apply this workflow on a real-time basis, getting confident diagnosis within 24 hours upon sample receipt. We, we believe that our workflow has already demonstrated great results and anticipate huge impact, not only in terms of rapid diagnosis, but also in patient risk evaluation and decision making. Published December 18, 2025 Created by
