Determining the moment in which a tumor cell acquires a driver mutation granting it resistance to a therapy is crucial to optimize treatment regimes in cancer. In particular, assessing whether the resistant clone is already present (pre-existing) in the population of cancerous cells or if it originates after the treatment (de-novo), is paramount to decide whether the treatment itself would be beneficial to the patient. This task, however, is generally challenging to achieve through either bulk whole genome sequencing or single cell sequencing, as reaching sufficient coverage to detect mutations at very low frequency remains expensive with both technologies. In our work (Antiviral treatment triggers HLA loss associated immune-evasion in acute myeloid leukaemia undergoing allogeneic hematopoietic stem cell transplant), in preparation, we tackled this problem in the context of Acute Myeloid Leukemia (AML) resistance acquisition to immune-therapy with Allogeneic Hematopoietic Stem Cell Transplant (Allo-HSCT). In this case the re- sistance is due to the Copy Neutral Loss of Heterozygosity (CNLOH) of the genes of the Human Leukocyte Antigens (HLA) complex on chromosome 6. We built a model of cancer evolution under therapy, leveraging Probabilistic Bayesian modeling and con- cepts of population genetics theory, to track in real time the cell that first acquired the HLA-CNLOH causing disease relapse. We applied the model to a cohort of 11 AML patients treated with Allo-HSCT at the San Raffaele Hospital of Milan, and we were able to successfully time the birth of the resistant clones.
Determinare il momento nel quale una cellula tumorale acquisisce una mutazione driver che le conferisce resistenza a una terapia è fondamentale al fine di ottimizzare il trattamento dei pazienti oncologici. In particolare, sapere se la cellula resistente è già presente nella popolazione tumorale durante il trattamento o se si origina in seguito è vitale per capire quale sarà il risultato del trattamento stesso. Effettuare questa valutazione a livello sperimentale è tuttavia complesso, poiché sia nel caso di bulk whole genome sequencing che di single cell sequencing, il costo per raggiungere un coverage sufficiente per rilevare mutazioni a bassa frequenza rimane troppo alto per l'applicazione clinica. Il mio lavoro di dottorato affronta questo problema nel contesto della Leucemia Mieloide Acuta (AML) e dello sviluppo di resistenza a immunoterapia tramite trapianto allogenico di cellule staminali ematopoietiche (Allogeneic Hematopoietic Stem Cell Transplant, Allo-HSCT). In questo caso, l'insorgere della resistenza è dovuta alla perdita di eterozigosi tramite Copy Neutral Loss of Heterozygosity (CNLOH) dei geni facenti parte del sistema dell'antigene leucocitario umano (Human Leukocyte Antigens, HLA). Per tracciare il momento nel quale la leucemia acquista questa mutazione, abbiamo realizzato un modello probabilistico bayesiano e lo abbiamo applicato a una coorte di 11 pazienti affetti da AML e trattati con Allo-HSCT all'ospedale San Raffaele di Milano. Grazie al nostro modello, siamo riusciti a ricostruire la storia evolutiva di questi tumori e determinare il momento nel quale hanno sviluppato la resistenza.
Determinare l'origine della resistenza al trapianto allogenico di cellule staminali ematopoietiche nella Laucemia Mieloide Acuta tramite modelli Bayesiani
ANTONELLO, ALICE
2025
Abstract
Determining the moment in which a tumor cell acquires a driver mutation granting it resistance to a therapy is crucial to optimize treatment regimes in cancer. In particular, assessing whether the resistant clone is already present (pre-existing) in the population of cancerous cells or if it originates after the treatment (de-novo), is paramount to decide whether the treatment itself would be beneficial to the patient. This task, however, is generally challenging to achieve through either bulk whole genome sequencing or single cell sequencing, as reaching sufficient coverage to detect mutations at very low frequency remains expensive with both technologies. In our work (Antiviral treatment triggers HLA loss associated immune-evasion in acute myeloid leukaemia undergoing allogeneic hematopoietic stem cell transplant), in preparation, we tackled this problem in the context of Acute Myeloid Leukemia (AML) resistance acquisition to immune-therapy with Allogeneic Hematopoietic Stem Cell Transplant (Allo-HSCT). In this case the re- sistance is due to the Copy Neutral Loss of Heterozygosity (CNLOH) of the genes of the Human Leukocyte Antigens (HLA) complex on chromosome 6. We built a model of cancer evolution under therapy, leveraging Probabilistic Bayesian modeling and con- cepts of population genetics theory, to track in real time the cell that first acquired the HLA-CNLOH causing disease relapse. We applied the model to a cohort of 11 AML patients treated with Allo-HSCT at the San Raffaele Hospital of Milan, and we were able to successfully time the birth of the resistant clones.File | Dimensione | Formato | |
---|---|---|---|
Thesis.pdf
embargo fino al 02/02/2026
Dimensione
15.09 MB
Formato
Adobe PDF
|
15.09 MB | Adobe PDF | |
Thesis_1.pdf
embargo fino al 02/02/2026
Dimensione
15.09 MB
Formato
Adobe PDF
|
15.09 MB | Adobe PDF |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14242/189363
URN:NBN:IT:UNITS-189363