Microsoft PowerPoint - Talk IMW S Minvielle [Compatibility Mode]

The genetic progression
of
Myeloma
Stéphane Minvielle, PhD
I have no conflicts of interest
13th International Myeloma Workshop
Paris, Carrousel du Louvre
May 3-6 2011

The genetic progression
of
Myeloma
Stéphane Minvielle, PhD
Director, Team 11 Inserm U892
Nantes Cancer Center
Nantes University Hospital

Multiple Myeloma
·
Heterogeneous disease with some patients dying within a few
weeks of diagnosis, while others live for longer than 10 years
·
Nearly all patients relapse

Multiple Myeloma
·
Heterogeneous disease with some patients dying within a few
weeks of diagnosis, while others live for longer than 10 years
·
Nearly all patients relapse
·
Now evident that the underlying genetic features of the tumor
cells largely dictate the clinical heterogeneity of MM
t(4;14) or del(17p)
1.00
amp(5q) no del(12p)
and normal m
2
no amp(1q)
0.75
al
al
iv
iv
Others
0.50
Surv
Surv
Others
erall
erall
Ov
amp(5q) amp(1q)
Ov
0.25
del(12p)
P < .0001
or no amp(5q)
t(4;14) or del(17p) and
and[amp(1q) and/or del(12p)]
0.00
high m
2
0
20
406080
100
Months
Time (months)
Time (months)

Gradual evolution
·
Multiple myeloma development model
Adapted from Kuehl et al Nature Review Cancer 2002;2,175

Gradual evolution
·
Multiple myeloma development model
·
Multi-step process accumulating sequential genetic changes
Primary genetic events
· Translocations
t(11;14)
t (4;14) ...
· Hyperdiploidy ...
Adapted from Kuehl et al Nature Review Cancer 2002;2,175

Gradual evolution
·
Multiple myeloma development model
·
Multi-step process accumulating sequential genetic changes
Primary genetic events
Progression events
· Translocations
· Del(17p) TP53
t(11;14)
· Chr.13 deletion
t (4;14) ...
· Chr.1 abnormalities
· Hyperdiploidy ...
· c-MYC rearrangements
· NF-B activation ...
Adapted from Kuehl et al Nature Review Cancer 2002;2,175

How to study genetic progression ?
·
Ideally: matched MGUS, SMM, MM, relapse samples in
many patients
·
In practice, paired diagnostic and relapse samples in a
small cohort of patients
·
Available tools
Targeted abnormalities (FISH)
Genome-wide allele specific copy number (SNP array)
Genome-wide intra /inter-chromosome rearrangements
and point mutations (Whole-genome sequencing)

Genomic analyis from SNP array data
·
CN and SNP markers (1.8 milions, intermaker distance < 1kb)
·
Genome-wide copy number changes
·
Landscape of genomic abnormalities
192 newly diagnosed patients
Chromosomes
Avet-Loiseau et al, JCO 2009

Genomic analysis from SNP array data
·
Identification of focal lesions (~ 50kb)

Genomic analysis from SNP array data
·
Allelic copy number changes and allelic imbalance (0.9M SNPS)
·
Loss of heterozygosity (LOH)
·
Subpopulations identification
Normal diploid
status

Genomic analysis from SNP array data
·
Allelic copy number changes and allelic imbalance (0.9M SNPS)
·
Loss of heterozygosity (LOH)
·
Subpopulations identification
Deletion
Normal diploid
status
40%

Genomic analysis from SNP array data
·
Allelic copy number changes and allelic imbalance (0.9M SNPs)
·
Loss of heterozygosity (LOH)
·
Subpopulations identification
Deletion
Normal diploid
status
100%
40%

Analysis using genome-wide SNP arrays
MM Patients
· 24 patients; median age 59 years
· Matched diagnostic and relapse samples
· Induction treatment
· VAD (n=12)
· Bortezomib dex (n=12)
· Median follow-up (25 months)
Magrangeas et al, submitted

Analysis using genome-wide SNP arrays

Pathways targeted at relapse
· NF-b activation (25% of the MM)
Amplification of activator (CD40)
Homozygous deletion of repressors (CYLD,
TRAF3, cIAP1/2 )

NF-b signaling activation
· Amplification of CD40

NF-b signaling activation
· Homozygous deletion of TRAF3
MM_#11
DR

NF-b signaling activation
· Homozygous deletion of cIAP1/2

NF-b signaling activation
· Homozygous deletion of cIAP1/2
MM_#42
R*
D
R

NF-b signaling activation
· Homozygous deletion of CYLD

NF-b signaling activation
· Homozygous deletion of CYLD
MM_#07

Conclusion (I)
· NF-b pathway is frequently targeted by relapse-
associated CNAs
· Genomic instability persists at relapse:
· Significant increase in CNAs at relapse (15.8 vs. 19.1,
p= 0.002)
· Two patients acquired new rearrangements at
relapse generated by two different mechanisms of
DNA repair
· A minor subclone with biallelic CYLD deletion
outcompeted the predominant diagnostic clone
· Is selection of minor subclone a common
phenomenon at relapse ?

Selection of subclones after initial therapy
· Deletion at 16p

Selection of subclones after initial therapy
· Biallelic deletion of AJAP1

Selection of subclones after initial therapy
1q UPD

Conclusion (II)
· MM at diagnosis is often composed of genetically
disctinct subclones present in varying proportions
· Minor subclones at initial presentation are often the
source of major clones that recur after treatment
· Are relapse clones evolving from diagnostic clones or
from ancestral clones?

Loss of lesions after initial therapy
UPD loss

Loss of lesions after initial therapy
· Biallelic deletion loss

Loss of lesions after initial therapy
· Deletion loss

Conclusion (III)
· In one third of the patients, the dominant clone at
relapse originates from a subclone that shared most of
genetic lesions with the dominant diagnostic clone but
did not evolve from it
· The ancestral clone gave rise to different subclones
that evolve independently by acquiring new CNAs
· Is emergence of an evolutionary past clone associated
with a type of treatment?

Treatment
·
Expansion of evolutionary past clone is almost
exclusively identified in patients treated with
bortezomib (p= 0.009)
·
Ancestral minor clones survive bortezomib therapy,
evolve and expand leading to relapse
·
Two explanations
The clone is more aggressive in response to bortezomib
Bortezomib treatment specifically extinguishes the dominant
subclone carrying the "driver" mutation that manifests as the
symptomatic myeloma while other subclones persist, thus
minor subclones which are not initially competitive against the
dominant population cells have a chance to thrive and acquire
new anomalies.

Evolutionary relationship between
diagnostic and relapse MM samples
·
At least three evolutionary models

Genetic progression in MM
· Remarkable adaptive changes driven by two forces,
genomic instability and clonal selection in response
to drug selection pressure
· At diagnosis, genetically distinct subclones already
possess variably aggressive growth properties
· Suggests new treatment paradigm that would
combine targeted therapy and subpopulations control
to eradicate all myeloma subclones in order to obtain
long-term remissions

Research team
Collaborations
Florence MAGRANGEAS
Nikhil MUNSHI
Hervé AVET-LOISEAU
Kenneth ANDERSON
Loïc CAMPION
Philippe MOREAU
Olivier DECAUX
Catherine GUERIN
Wilfried GOURAUD