High-Throughput Functional Ex-Vivo Drug Testing and Multi-Omics Profiling in Patients with Acute Myeloid Leukemia

• Published in: BLOOD Vol. 134; no. Supplement_1; p. 4641
• Main Authors: Erkers, Tom, Seashore-Ludlow, Brinton, Struyf, Nona, Marabita, Francesco, James, Tojo, Malani, Disha, Vesterlund, Mattias, Stalhl, Matthias, Pawitan, Yudi, Lehmann, Sören, Östling, Päivi, Lehtiö, Janne, Kallioniemi, Olli
• Format: Journal Article Conference Proceeding
• Language: English
• Published: Elsevier Inc 13.11.2019
• Subjects: Medicin och hälsovetenskap
• ISSN: 0006-4971; 1528-0020
• DOI: 10.1182/blood-2019-124332

▪ Acute myeloid leukemia (AML) is the most common leukemia in adults and the prognosis is dismal. The heterogeneity of driver mutations in AML from one patient to another is significant. Although new treatment targets and modalities have recently been introduced, a lack of therapies to cover the entire spectrum of molecularly distinct AML classes remains a challenge. Furthermore, often only a fraction of the patients carrying mutations actually respond to the drugs predicted to target such genomic subtypes in AML. Our question was if outcome can be improved by optimizing existing and emerging therapies? To address this, we applied direct functional drug testing ex vivo in patients with newly diagnosed AML, along with deep molecular profiling to identify new therapeutic and diagnostic opportunities for specific genetic subclasses. To date, 122 patient samples have been included in the study. We have collected AML patient mononuclear cells isolated from fresh bone marrow aspirates from the Karolinska University Hospital Huddinge and Uppsala University Hospital, or applied biobanked samples from the Swedish acute leukemia biobank. The cells from the patient are tested for 72h with up to 525 conventional and investigative oncology drugs in a 5-point concentration range to determine optimal treatment options. The test is based on bulk cell viability and the cells are cultured in bone marrow stroma-conditioned media. The data is compared to healthy bone marrow controls to filter out generally toxic drugs and combinations. Excess patient material is applied for multi-omics analysis, including genome sequencing, deep proteomic profiling, concentrations of soluble factors, bulk/single-cell RNA sequencing and mass cytometry. Our results indicate significant heterogeneity in functional drug responses across individual patients, even among those with the same founder mutations. We could identify potential targeted treatments for most patients based on the ex vivo testing. For instance, patients that show resistance to the Bcl-2 inhibitor Venetoclax can be associated with a distinct drug sensitivity response pattern that includes drugs with different mechanisms of actions. Based on the clustering of drug response data in all clinical samples, we could identify groups of patients with similar global drug responses, and drugs with similar patterns of efficacy across all patients. Lehmann:Pfizer: Membership on an entity's Board of Directors or advisory committees; Abbive: Membership on an entity's Board of Directors or advisory committees; TEVA: Consultancy, Membership on an entity's Board of Directors or advisory committees. Kallioniemi:Medisapiens: Other: Co-founder and stockholder; Sartar Therapeutics: Other: Co-founder and stockholder; Astra-Zeneca: Other: Joint Grant; Pelago: Other: Joint Grant; Takara: Other: Joint Grant; Abbot: Other: Licensing income.