Personalized medicine has the ambition of no less than the paradigm shift from the era of one-size fits all therapy to a precision medicine in which each patient would benefit from individualized treatment adapted to the molecular data characterizing his or her pathology. In oncology, the revolution in OMICS-type technologies and pan-genomic and now single cell analyses are enabling considerable progress in this direction, in particular through the identification of molecular signatures characteristic of patient subgroups with variable prognostic values but which also testify to the heterogeneity of cancers and the complexity of treating them. However, the identification and characterization of molecular alterations, including at very high throughput, have not yet made it possible to orient therapeutic strategies with proven clinical benefit. It has therefore been proposed to complement the genomic data of a tumor with functional ones to better define the axes of tumor sensitivity (1). Various approaches for profiling sensitivity and resistance to drug panels (either FDA-approved or under clinical investigation) have been developed ex vivo, particularly in the context of hematological malignancies, demonstrating their feasibility and their interest such as for the repositioning of drugs and the role of specific gene networks in drug response. (2)(3)(4).
The interest in these approaches is particularly marked in Acute Myeloid Leukemia (AML) for which the majority of patients relapse and succumb to disease with a 5-year overall survival rate of around 20%. By focusing specifically on this population of refractory or relapsed (R/R) AML patients recruited in the formal framework of a clinical trial (NCT02619071, principal investigator: Pr VEY, Department of Hematology, Institut Paoli Calmettes), we evaluated the feasibility of producing in real time the relevant genomic (tNGS, aCGH: Dr BIRNBAUM, Predictive Oncology Team, Institut Paoli Calmettes and Cancer Research Center of Marseille) and functional (Drug Sensitivity Resistance Profile –DSRP- on a panel of 78 drugs: TrGET Preclinical platform, Institut Paoli Calmettes and Cancer Research Center of Marseille) data discussed by an institutional personalized committee with the aim to guide the therapeutic strategy chosen by the clinician (Tailored Treatment Strategy). The main objective of this study (TTS<21 days, a timing compatible with the management of such patients) was achieved for 58% of patients, 35% of whom were administered the TTS-guided treatment. Fifteen of these seventeen TTS having been guided by the combined genomic and DSRP data, whereas 2 TTS were guided by genomic or DSRP data only. Our results show that chemogenomic combining tNGS with DSRP to determine a TTS is a promising approach to propose patient-specific treatment options within 21 days. The short (≈3 months) overall survival in this cohort favors including patients earlier in their therapeutic course, yet 4 complete responses occurred in the TTS group. Expending this study to a growing number of patients and their compared chemogenomic analysis with patients at diagnosis is in progress ( , number NCT02320656), hopefully making it possible to better identify and target the specificity of these refractory patients while improving their personalized management., number