CRCT - Cancer Research Center of Toulouse
ResearchResearch Teams13 H. Avet-Loiseau: Pharmacogenomics of multiple myeloma

13 H. Avet-Loiseau: Pharmacogenomics of multiple myeloma

Team Research Project

Team 13: (H. Avet-Loiseau): to improve management and outcome of patients with multiple myeloma.

Our project is totally focused on Multiple Myeloma (MM). MM is a highly heterogeneous disease, from biological to clinical points of view. Regarding outcome, if the median overall survival of patients is about 10 years for the youngest ones (< 65 y), ~20% of these patients will die during the first 2 years after diagnosis, whereas ~15-20% are still alive more than 15 years later. This heterogeneous outcome is mainly due to molecular variability within the tumor plasma cells. Our first aim is to understand this variability, and to go back to bedside in proposing individualized therapy to patients. Thanks to our large tumor bank, and with a long follow-up, we are able to analyze the tumor variability, using different genetic techniques (FISH, SNParray, NGS). Based on these analyses, we are then able to test their prognostic values on large series of homogeneously treated patients. For example, we did identify several chromosomal abnormalities such as t(4;14), del(17p), or del(1p) that are now in the panel of tests analyzed at diagnosis for every patient worldwide. 
Using high-throughput genomic tools like SNParray, we did analyze more than 1,000 patients with clinical annotations. This large study is currently under analysis to (i) identify the most recurrent abnormalities, and (ii) correlate these abnormalities with response to treatment and outcome. The ultimate goal is to identify genomic profiles associated with specific response in order to propose individualized therapies in the near future.
Heterogeneity is also observed at the response to treatment level. With current therapies, the complete response (CR) rate is 60% to 70%, signing the major improvements observed in the past decade. However, these results hide a heterogeneity in the true response of the patients, some of them being in true CR, and possible cured, whereas other have still important residual disease. The only way to address this issue is to develop sensitive tools to discriminate the two types of patients. A US company, Sequenta, recently developed a technology based on NGS, to measure the level of minimal residual disease (MRD). This technique is based on the identification at diagnosis of a patient-specific clonal biomarker, the IGH and/or IGLκ rearrangements. NGS enables to quantify this biomarker with a high sensitivity, i.e., 10-6. We developed an  exclusive agreement with Sequenta to use their technology in our lab. By measuring the level of MRD after treatment, we will be able to identify the patients with a negative MRD, who may be cured, and patients with still positive MRD, for whom further treatments like maintenance could be useful. Furthermore, sequential MRD evaluations may allow the early detection of molecular relapses, enabling therapeutic strategies before the onset of clinical relapse.
Finally, it has been shown that MM is an oligoclonal disease at the molecular level. At the time of diagnosis, the tumor population is generally a mixture of a major subclone, plus different minor subclones. Interestingly, the relapse can be due to the major subclone, but also to one or several of the minor diagnostic subclones. Two hypotheses can be proposed: either a selection by the therapeutic pressure, or a natural history of the disease. Even though these subclones may differ by a single somatic mutation, this mutation can be important at the molecular level in conferring resistance to some specific drugs. By using exome sequencing, it is possible to identify these different subclones. Analysis of large cohorts of patients will enable to address these different issues, in order to try again to help the physician’s choice or treatment. This project requires important bio-informatic analyses. To address this important issue, we developed a collaboration with the Bio-informatic&Biostatistics department of Harvard in Boston.

Our second aim is to analyze the relationship between the tumor plasma cells and the microenvironment. MM is a highly microenvironment-dependent cancer. The microenvironment is a complex network that closely interacts with MM cells and consists of three components (cellular, extra-cellular matrix and soluble). These interactions allow the proliferation, survival, and chemo-resistance of MM cells. It has led to the development of new drugs which target not only MM cells but also their microenvironment. We plan to analyze two parts of the microenvironment, the bone marrow MSCs and the immune system.
We were among the first to report specific abnormalities of MSC from patients with MM. MM MSC showed impaired osteoblastic differentiation, induction of an overgrowth of MM cells, and a specific secretome and gene-expression profile. We also reported that one of these specific abnormalities, GDF15 overexpression, participates to tumor cells proliferation, survival, and chemo-resistance to conventional and new drugs. Furthermore, GDF15 plasma concentration is directly related to survival in MM patients.
We now aim to analyze the impact of drugs on MM MSC, but also the role of these MSC on the sensitivity or resistance of malignant plasma cells. We also plan to assess which abnormalities in MSC may impact on the genesis and natural course of the MM, but also in the relapse after treatment. In this purpose, we will perform longitudinal analysis of MSC from MGUS (monoclonal gammopathy of undetermined significance) to MM at diagnostic, but also during remission and at relapse in the same patients. Our goal is to identify: (i) biomarkers of MSC abnormalities predictive of MM progression and sensitivity to treatment; (ii) therapeutic targets specific to microenvironment, in order to design preventing or curative approach based on its normalization.
Like many neoplasms progressive immune suppression has been involved in the progression of MM. Additional proof of MM control by immune components derives from the clinical use of immuno-modulatory drugs (IMIDs) whose anti-myeloma properties rely, at least partially, on the co-stimulation of T and Natural Killer cells (NK). We showed that NK cells in MM express CD226, and that MM cells do express its ligands CD112 and CD155 (Nature Immunology, in press). Abrogation of CD226 led to a decreased response to different anti-MM drugs in a non-immunologically suppressed mouse model (Vκ-MYC). Our aim is to test the functionality of NK and T cells in MM patients. Thanks to the IFM network, we will analyze the expression of CD226, CD112, and CD155 in patients at diagnosis, but also during follow-up, and at relapse. Furthermore, we developed a collaboration with different pharma companies developing anti-CD38 monoclonal antibodies for MM therapy (Janssen, Sanofi, Celgene). In these research contracts, we will test the role of NK and T cells, in the response to these antibodies. 

Our third aim is to build a mouse model, and to generate radio-immunologic tools to detect and monitor the disease. 
No clinically valid animal model exists in MM. The most powerful one is the SCID-hu model, in which human fetal bones are subcutaneously graft to a SCID mouse, and then injected with tumor plasma cells. Because of ethics reasons, this model is not available in France. To mimic this SCID-hu model, we plan to inject in the NSG mouse tibia patient bone marrow MSCs, in order to “humanize” and “tumorigenize” the mouse bone marrow microenvironment. Secondary, or at the same time, tumor plasma cells will be injected in the same tibia. We are just starting this project, with very preliminary but encouraging results. Using myeloma cell lines, tumors are developing in the injected bone. Most importantly, these tumors disseminate in the contralateral tibia and in vertebras. These results are very preliminary, and we need to optimize and extend to patient’s tumor plasma cells. We need also to see if we can obtain a disseminated disease mimicking the human disease. To address this latter issue, we plan to develop an imaging approach enabling to follow the disease in the mice without killing them. Our project is to label a specific plasma cell antibody, i.e., anti-CD38, with radio-conjugate, zirconium. This radio-conjugated has the advantage of a long half-life (one week), enabling different times of imaging. Thanks to the IUC environment, small animal PET-CT will be available. If this model is working in mice, we plan to develop a human counterpart in order to follow the disease in patients. Finally, a much longer objective would be to use this approach for radio-immuno-therapy.


  • Multiple Myeloma
  • Translational research
  • Genomics
  • Microenvironment
  • Imaging





Team members

Hospital Practitioner - PU-PH

Hospital Practitioner - PU-PH


   +33 5 31 15 61 45 
Engineer - IE2

   +33(0)5 31 15 61 49 
Engineer - IE2


Post-doctoral fellow

Hospital Practitioner - MCU-PH


   05 31 15 63 88 

Hospital Practitioner - AH/AHU


Technician - AI

   +33(0)5 31 15 61 48 
Engineer - IE2


   +33(0)5 31 15 61 43 
Engineer - IE2

   +33 5 31 15 61 75 
Technician - other


   +33 5 31 15 61 47 
Researcher - CR1

Post-doctoral fellow


Hospital Practitioner - PH

PhD student


 12th GERLI international Lipidomics Meeting - Microbe and Host Lipids
12th GERLI international Lipidomics Meeting - Microbe and Host Lipids

 Cellular origin and mechanisms of pancreatic cancer
23/05/16 - Dr Axel Behrens - Auditorium IUCT-O - 10:00 - 11:00
Inserm Université Toulouse 3 - Paul Sabatier CNRS