Examining treatment responses of diagnostic marrow in murine xenografts to predict relapse in children with acute lymphoblastic leukaemia
Menée à l'aide d'échantillons de moelle osseuse prélevés par aspiration sur 30 enfants atteints d'une leucémie lymphoblastique aiguë à risque intermédiaire puis inoculés à un modèle murin traité par vincristine, dexaméthasone et L-asparaginase, cette étude évalue l'association entre la proportion de cellules CD45+ humaines présentes dans le sang périphérique des souris et le risque de récidive chez l'enfant
Background : While current chemotherapy has increased cure rates for children with acute lymphoblastic leukaemia (ALL), the largest number of relapsing patients are still stratified as medium risk (MR) at diagnosis (50–60%). This highlights an opportunity to develop improved relapse-prediction models for MR patients. We hypothesised that bone marrow from MR patients who eventually relapsed would regrow faster in a patient-derived xenograft (PDX) model after induction chemotherapy than samples from patients in long-term remission.
Methods : Diagnostic bone marrow aspirates from 30 paediatric MR-ALL patients (19 who relapsed, 11 who experienced remission) were inoculated into immune-deficient (NSG) mice and subsequently treated with either control or an induction-type regimen of vincristine, dexamethasone, and L-asparaginase (VXL). Engraftment was monitored by enumeration of the proportion of human CD45+ cells (%huCD45+) in the murine peripheral blood, and events were defined a priori as the time to reach 1% huCD45+, 25% huCD45+ (TT25%) or clinical manifestations of leukaemia (TTL).
Results : The TT25% value significantly predicted MR patient relapse. Mutational profiles of PDXs matched their tumours of origin, with a clonal shift towards relapse observed in one set of VXL-treated PDXs.
Conclusions : In conclusion, establishing PDXs at diagnosis and subsequently applying chemotherapy has the potential to improve relapse prediction in paediatric MR-ALL.
British Journal of Cancer , résumé, 2020