Background: Acute myeloid leukemia (AML) is a heterogenous malignancy and efficient treatment strategies for high-risk patients are still lacking. To facilitate preclinical and translational leukemia research, e.g. to test novel therapeutics in vivo, we generate patient-derived xenograft (PDX) samples. We had transplanted more than 100 primary adult AML samples into immune-compromised NSG mice. Furthermore, we received PDX samples primografted in other labs of pediatric and adult patients. However, only around 20% of primary samples (i.e. “high score samples”) show a robust and reliable serial engraftment and can be genetically modified, facilitating repetitive, reproducible and convenient in vivo experiments.

Aims: (i) Identifying factors influencing reliable serial engraftment; (ii) Characterizing high score samples functionally and genetically; (iii) Performing in vivo (therapy) trials.

Methods: We generated 23 high score AML PDX samples characterized by (i) a serial, fast, reliable and systemic engraftment, and (ii) the capacity to be genetically engineered. These samples were characterized by DNA panel sequencing (seq), whole exome seq, low coverage whole genome seq, RNA seq, methylation profiling, karyotyping, and immunophenotyping. Furthermore, samples were transplanted repetitively, and capacity for serial engraftment was analyzed. Last, samples were used for diverse in vivo trials, e.g. long-term chemotherapy trials.

Results: High score samples originated mainly from relapsed AML cases (18/23, p<0.05). Overall survival (OS) of patients whose cells yielded high score samples was dramatically lower compared to OS of patients whose cells did not engraft or meet the criteria of high score samples (p<0.01). Analysis of AML-related mutations, gene fusions, translocations, copy number alterations and immunophenotype revealed that high score AML PDX samples resembled primary patient cells of both, initial diagnosis and relapsed disease, and - in contrast to established AML cell lines - covered many different genetic and genomic alterations. Within the WHO classification of 2022, most samples belonged to one of the three groups “AML with NPM1 mutation” (n=8), “AML with KMT2A rearrangement” (n=7), or “AML, myelodysplasia-related” (n=6).

High score samples were serially transplanted into recipient mice, with up to fifteen repetitive transplantations. Importantly, we did not lose a sample in serial re-transplantations, indicating that leukemia initiating cells proliferated within the murine niche for many months (median 412 days). On the contrary, under in vitro conditions, cells could be kept for some days to weeks, but ultimately died. With every engraftment, PDX cells can be amplified by a factor 30-fold or higher, depending on the sample, allowing the production of indefinite numbers of PDX cells.

Engraftment capacity and engraftment time were influenced by several factors; thawed PDX cells and injection of cell numbers below 500,000 cells had a reduced engraftment rate and a longer passaging time compared to freshly isolated cells or higher cells numbers. Serial re-passaging or expression of transgenes did, however, not influence engraftment. While murine gender influenced engraftment capacity of some samples, with male mice being less supportive, mouse age did not, and even older mice with more than 20 weeks of age allowed reliable engraftment.

High score AML PDX samples were further used in preclinical in vivo therapy trials. Stable luciferase expression allowed sensitive disease monitoring, facilitating real-time analysis of drug effects and of growth kinetics before and after treatment. Bioluminescence imaging revealed that azacitidine, in contrast to cytarabine, had a delayed effect on tumor burden. Both drugs, however, showed a sustained effect on the cells even after stop of long-term therapy, indicated by a reduced growth rate after treatment.

Conclusion and Outlook: High score AML PDX models represent highly aggressive primary patients' cells. By serial engraftment and genetic engineering, we can produce indefinite numbers of these cells, which represent a valuable and unique tool for sophisticated molecular and functional studies. Our well characterized AML PDX cohort allows repetitive and reproducible preclinical trials, and luciferase expression facilitates reliable monitoring of long-term therapy trials.

Disclosures

Vick:Tubulis GmbH: Patents & Royalties: pending patent application FLT3-mAb 20D9. Baldus:Janssen, Astellas, Pfizer, Astrazeneca, Servier, BMS: Consultancy, Honoraria. Götze:JAZZ: Honoraria; Otsuka: Honoraria; Abbvie: Honoraria; BMS: Honoraria. von Bergwelt-Baildon:TABBY: Membership on an entity's Board of Directors or advisory committees; AMGEN, Astellas, AstraZeneca, Bristol-Myers Squibb, Daiichi Sankyo, KITE/Gilead Mologen, Miltenyi, MSD Sharp + Dohme, Novartis, Priothera, Roche, TABBY: Consultancy, Honoraria, Research Funding, Speakers Bureau. Metzeler:Menarini Stem Line: Honoraria; Otsuka: Consultancy, Honoraria; Janssen: Consultancy, Honoraria; Servier: Honoraria; BMS/Celgene: Consultancy, Honoraria; AstraZeneca: Honoraria; Astellas: Honoraria; Abbvie: Honoraria, Research Funding; Sysmex: Honoraria. Subklewe:AbbVie, Amgen, Autolus, AvenCell, BMS, CanCell Therapeutics, Genmab US, Gilead, Ichnos Sciences, Incyte Biosciences, Interius BioTherapeutics, Janssen, Miltenyi Biomedicine, Molecular Partners, Nektar Therapeutics, Novartis, Orbital Therapeutics, Pfizer,: Honoraria; Amgen, BMS/Celgene, Gilead/Kite, Janssen, Miltenyi Biotec, Molecular Partners, Novartis, Roche, Seagen, Takeda: Research Funding; AstraZeneca, BMS, Gilead/Kite, GSK, Janssen, LAWG, Novartis, Pfizer, Roche, Springer Healthcare: Speakers Bureau. Herold:Servier Deutschland: Honoraria; Jazz Pharmaceuticals: Honoraria. Spiekermann:Tubulis GmbH: Patents & Royalties: pending patent application FLT3-mAb 20D9. Jeremias:Tubulis GmbH: Patents & Royalties: pending patent application FLT3-mAb 20D9.

This content is only available as a PDF.
2024
Sign in via your Institution