PB1959: BCR::ABL1 TRANSCRIPT IN SMALL EXTRACELLULAR VESICLES ISOLATED IN ADULT CHRONIC MYELOID LEUKEMIA PATIENTS CORRELATES WITH MOLECULAR RESPONSE LEVEL AND THE ONGOING THERAPY

Background: The role of small extracellular vesicles (sEVs) in tumor development has been demonstrated and explored also in Chronic Myeloid Leukemia (CML). They support the leukemogenesis, the angiogenesis and a leukemia-friendly microenvironment. The analysis of sEVs and their cargo represent a non-invasive approach that could reveal messages released by residual leukemic cells. The BCR::ABL1 transcript has been already isolated in the sEVs isolated from CML patients under TKIs response, but its correlation with clinical and biological variables is still uncertain. Aims: The study wonders whether there may be a correlation between the BCR::ABL1 transcript levels in sEVs and the disease status as well as different clinical characteristics. Methods: sEVs were isolated from 42 samples of plasma collected from adult CML patients in molecular response: 12/42 (28%) in MMR and the remnant 30/42 (72%) in DMR assessed by conventional RT-qPCR following the international scale (IS%). At the moment of sampling, 19/42 (45%) were treated with imatinib, 9/42 (21%) with nilotinib and 5/42 (12%) were in experimental TFR. Digital PCR (dPCR) was used to detect the presence of the transcript in sEVs, taking advantage of dPCR high sensitivity. Results: The vesicular BCR::ABL1 transcript levels show a decrease at the deepening of the MR: median 0.358 copies/μl for MR3.0 (range 0.083 – 0.698); median 0.243 copies/μl for MR4.0 (range 0.083 – 0.472); median 0.209 copies/μl for MR4.5 (range 0.072 – 0.375); median 0.19 copies/μl for MR5.0 (range 0 – 0.306). There was a significant difference between MR3.0 and MR4.5 (p = 0.03), and MR3.0 and MR5.0 (p = 0.01) (Fig 1A). By gathering overall samples according to MMR and DMR definition, the BCR::ABL1 transcript levels in sEVs differ by a significant statistical proportion (p= 0.0027) (Fig 1B). The undetectable samples by RT-qPCR were 11/42 (26%), while by dPCR on PB cells and on sEVs were 2/42 (5%) and 1/42 (2.5%), respectively. The sample undetectable by dPCR on sEVs resulted detectable by the others quantification analysis. A significant difference was found between RT-qPCR and dPCR on PB cells (p=0.01), and between RT-qPCR and dPCR on sEVs (p=0.003) (Fig 1C). No significance was found between vesicular transcript quantifications and the variables of sex, age, transcript type, IS%, BCR::ABL1 quantification by dPCR on cells and treatment duration. However, the analysis revealed a significant difference when considering the ongoing therapy. The transcript levels in sEVs were significantly higher in samples from patients in TFR, compared to the ones from patients treated with just imatinib (p=0.007), whilst no statistical difference was observed between samples from patients treated with nilotinib (Fig 1D). Summary/Conclusion: In conclusion, dPCR on sEVs was able to properly follow transcript reduction at the deepening of the MR classes and to discriminate between MMR and DMR. As expected, dPCR confirmed its higher sensitivity to detect even low levels of MRD in comparison to RT-qPCR, especially when associated with sEVs isolation. In relation to the ongoing treatment at the moment of sampling, the differences in BCR::ABL1 levels detected by dPCR on sEVs may be due to many factors, like the duration of treatment.