poročilo eha 2012 s j12.pdf · ambaglio i, et al. poster presentation at eha 2010; abstract 0310...
TRANSCRIPT
1
Poročilo EHA 2012
Simon Bitežnik
2
Burden of chronic anaemia in patients
with MDS Anaemia is a major clinical problem in patients with
MDS
~80% patients are anaemic at diagnosis1
In patients with MDS, anaemia has a negative impact
on
Survival2
Risk of NLD2
HRQoL3
In patients with MDS, the severity of anaemia
correlates with IPSS category4
1. Kelaidi C, et al. Haematologica 2010;95:892–9
2. Ambaglio I, et al. Poster presentation at EHA 2010; Abstract 0310
3. Jansen AJG, et al. Br J Haematol 2003;121:270–4
4. Kao JM, et al. Am J Haematol 2008;83:765–70
3
Treated with EPO (n=403)
Untreated (n=628)
Overa
ll s
urv
iva
l (%
)
100
80
60
40
20
0
0 2 4 6 8 10 Years*
EPO responders
Non-responders
Untreated
P<0.001
EPO therapy may improve OS in
erythroid responders
Park et al. Blood 2008;111:574–82 *from diagnosis or from EPO treatment initiation
4
Not all MDS patients will respond to
EPO: prognostic factors
At multivariate analysis, factors associated with response were:
Transfusion-independence
Serum EPO levels
Karyotype
Baseline Hb levels
Latagliata et al. Acta Haematologica 2008;120:104–107
For each g/dl increase in pre-treatment Hb, l
the probability of response increased by
98% (p=0.02)
Variables
Hazard Ratio
(95% confidence internal) P
Age 1.011 (0.981–1.042) 0.464
Males 0.792 (0.385–1.628) 0.526
Time from diagnosis 0.995 (0.980–1.010) 0.507
Serum Epo levels 0.993 (0.986–1.000) 0.046
Hb 1.845 (1.235–2.756) 0.003
Cytogenetics 0.479 (0.244–0.938) 0.032
LDH 1.000 (0.999–1.002) 0.720
Ferritin 1.000 (1.000–1.000) 0.845
TD 2.867 (1.354–6.070) 0.006
Univariate Cox regression analysis of response
5
Poor response to EPO in patients with
del(5q) versus non-del(5q) MDS
Kelaidi C, et al. Leuk Res 2008;32:1049–53 DAR = darbepoietin
Retrospective analysis of 345 patients with MDS treated with EPO or DAR
p=0.03 p=0.02
6
MDS-004: significant improvements in RBC-TI in
patients randomised to lenalidomide versus
placebo
Fenaux P, et al. Blood 2011:118;3765–76
†mITT population defined as patients with centrally
confirmed MDS who received ≥1 dose
*
RB
C-T
I (%
)
30
50
60
70
10
0 Protocol defined IWG 2000
(≥26 weeks) (≥8 weeks)
40
20
Placebo (n=51)
LEN 5mg (n=47)
LEN 10mg (n=41)
*p<0.001 versus placebo
Bars represent 95% CI
mITT population†
6
43
56
51
61
8
*
*
*
7
Lenalidomide vs placebo
Erythroid response associated with better
survival
Fenaux P, et al. Blood 2011;118:3765-76
Time (days)
p=0.0085
Leu
kem
ia-f
ree s
urv
iva
l
0.4
0.6
0.8
1.0
0.2
0 0 1 2 3 4 5
Erythroid responders
Erythroid
non-responders
Time (days)
p=0.0028
Overa
ll s
urv
iva
l
0.4
0.6
0.8
1.0
0.2
0 0 1 2 3 4 5
Erythroid responders
Erythroid
non-responders
8
Azacitidine induces erythroid responses
in higher risk MDS
Silverman LR, et al. J Clin Oncol 2002;20:2429–40 BSC = Best supportive care
*P<0.0001; †P<0.01
AZA BSC
Responses (CR + PR) 23 (23%)* 0%
Complete response (CR) 7 (7%)† 0%
Partial response (PR) 16 (16%)† 0%
Haematological improvement 37 (37%) 5%
9
p≤0.0001
HR=0.23 [95% CI: 0.10–0.51]
Time from randomisation
0
0.2
0.4
0.6
0.8
1.0
Su
rviv
al
Patients experiencing haematological
improvement with azacitidine treatment have a
better OS
List AF, et al.Oral presentation at ASCO 2008; Chicago, IL, USA
0 5 10 15 20 25 30 35 40
AZA
CCR 24
71.7%
16.6 mo.s 27.1%
Median not reached
10
Conclusions
Iron chelation therapy in MDS patients receiving transfusions
may be associated with a better outcome
Treatment with ESAs in lower-risk MDS induces erythroid
responses in 40–70% of cases
Responders experience improvements in QoL and a better survival
Lenalidomide in lower-risk MDS with del(5q) induces erythroid
responses and cytogenetic changes
Responders experience improvements in QoL and a better survival
Patients with higher-risk MDS receiving azacitidine may
experience erythroid responses
Responders experience improvements in QoL and a better survival
11
Cytogenetic abnormalities are very common in
patients with MDS
• 52% of patients had clonal abnormalities; 44% of these patients had
>1 abnormality
• The most common abnormality was del(5q)
0
Nu
mb
er
of
cases
Cytogenetic abnormality Haase D, et al. Blood 2007;110:4385–95
Single abnormality
Abnormality present + one additional aberration
Abnormality present within complex karyotype
Analysis of a database of 2,124 patients with MDS
MDS = myelodysplastic syndrome
350
300
250
200
150
100
50
12
Del(5q) MDS is not 5q– syndrome
*del(5q) with ≥2 additional abnormalities
OS = overall survival
1. Greenberg P, et al. Blood 1997;89:2079–88
2. Haase D, et al. Blood 2007;110:4385–95
3. Mallo M, et al. Leukemia 2011;25:110–2
4. Giagounidis A, et al. Hematology 2004;9:271–7
MDS with chromosomal
aberrations
del(5q) MDS
isolated del(5q)
del(5q) + 1
additional
abnormality
complex
abnormalities*
All MDS
WHO MDS with
isolated del(5q) 5q– syndrome
OS = 4.8–68.4 months1
OS = 24 months2
OS = 7–80 months2
OS = 66 months3
OS = 107 months4
13
Additional risk factors in patients with del(5q)
MDS: further chromosomal abnormalities
Mallo M, et al. Leukemia 2011;25:110–20
Retrospective analysis of 541 del(5q) MDS patients assessed
in the pre-lenalidomide era
Cu
mu
lati
ve
su
rviv
al
Time (months)
● Two distinct risk groups: isolated
del(5q) and del(5q)+1; del(5q) + ≥2
● Three distinct risk groups: isolated
del(5q); del(5q) + 1; del(5q) + ≥2
1.0
0.8
0.6
0.4
0.2
0 0 100 200 300 400
n=148
n=275 n=89
p<0.001
p=0.131 (isolated vs + 1)
del(5q)
del(5q)+1
del(5q)+≥2
Pro
ba
bil
ity o
f A
ML
tran
sfo
rmati
on
Time (months)
1.0
0.8
0.6
0.4
0.2
0 0 50 100 150 200 250
n=96
n=43
n=160
p<0.001
p=0.001 (isolated vs + 1)
del(5q)
del(5q)+1
del(5q)+≥2
Additional chromosomal abnormalities
are associated with reduced survival
Additional chromosomal abnormalities
are associated with increased AML
transformation
14 Giagounidis A, et al. Hematology 2004;9:271–7
p=0.00005
Time (months)
Cu
mu
lati
ve
su
rviv
al
360 300 240 180 120 60 0
1.0
0.8
0.6
0.4
0.2
0
≥5% blasts
<5% blasts
del(5q) with <5% BM blasts
del(5q) with ≥5% BM blasts
Additional risk factors in patients with del(5q)
MDS: BM blast percentage Retrospective analysis of 60 patients with isolated
del(5q) MDS; median follow-up = 53 months
BM = bone marrow
15
Combined analysis of MDS-003 and MDS-004:
impact of karyotypic complexity on clinical outcomes
Giagounidis A, et al. Oral presentation at 11th International
Symposium on MDS 2011, Edinburgh, UK. Abstract 58
Karyotype complexity had a negative
impact on OS
Pro
ba
bil
ity o
f s
urv
iva
l
Time to death (years)
1.0
0.8
0.6
0.4
0.2
0
0 1 2 3 4 5 7 6
3.9 years
3.0 years
1.8 years
p<0.01 p<0.05
Time to AML (years)
Pro
bab
ilit
y o
f A
ML
pro
gre
ss
ion
Median OS
Log-rank
p=0.0016
Karyotype complexity had a negative
impact on AML progression
Isolated del(5q) del(5q) + 1 del(5q) + ≥2
Analysis included a total of 286 lenalidomide-treated patients with del(5q) MDS
AML = acute myeloid leukaemia
1.0
0.8
0.6
0.4
0.2
0
0 1 2 3 4 5 7 6
Log-rank
p=0.0044
16
TP53 mutations in MDS and their impact on
patient outcomes
Kulasekararaj AG, et al.
Oral presentation at ASH 2011. Abstract 792
Retrospective analysis of the incidence and prognostic impact of TP53 mutations
in patients with del(5q) using next-generation sequencing
Patient characteristics (n=318)
• Median age, years (range): 65 (17–72)
• IPSS risk, n (%)
– low: 71 (24)
– int-1: 101 (32)
– int-2: 58 (18)
– high: 29 (9)
• 40 patients (12%) received BM transplant, IC, azacitidine or lenalidomide
TP53 mutational status
• Patients with mutation, n (%): 30 (9.4)
• Median clone size, % (range): 42 (2.5–93)
TP53 mutations are an independent
prognostic marker in patients with
MDS and del(5q)
OS by TP53 mutational status*
All
patients
Isolated
del(5q) Complex
0
20
40
60
80
Mo
nth
s
9.7
NR
(>66) 66
23
14.1 9.6
p<0.0001 p<0.002
p<0.0001
Wild-type
TP53 mutation
Multivariate analysis†: TP53 mutational
status was the strongest predictor for OS
and PFS (p<0.0001 for both)
*Survival analysis was censored at treatment date †Co-variables: age, sex, WHO subtype, IPSS risk,
±mutations; progression-free survival
17
Commonly retained regions versus commonly
deleted regions in del(5q) MDS Analysis of 1,155 MDS/MPN/AML patients to characterise 5q disorders and
identify the impact of associated genomic lesions, by SNP-A and MC
25
20
15
10
5
0 MDS
(n=473)
MDS/
MPN (n=62)
AML (n=252)
23
8
13
Pati
en
ts w
ith
del(
5q
) b
y
SN
P-A
(%
)
Del(5q) frequency
MC SNP-A 0
50
100
150
Pa
tie
nts
wit
h 5
q le
sio
n, n
5q lesions
MC SNP-A 0
40
20
60
80
100
Pa
tie
nts
(%
)
Additional
abnormalities
5q by MC
5q UPD
No growth by MC
Isolated 5q
5q + additional lesions
Jerez A, et al. J Clin Oncol 2012; Epub ahead of print
MC = metaphase cytogenetics
SNP-A = single-nucleotide polymorphism array
UPD = uniparental disomy
18
Diagnosis of del(5q) MDS: methodology
Standard metaphase
cytogenetics (MC)1
● Standard MC is integral to
the diagnosis of del(5q)
MDS
● Guidelines stipulate that
≥25 metaphases should be
analysed2
Fluorescence in situ
hybridisation (FISH)1
● May play a supplementary
role to standard MC
especially when insufficient
metaphases are available
● FISH detected del(5q) in 6%
of 637 MDS patients who
had normal karyotype by
MC at diagnosis
SNP array (SNP-A)
karyotyping3
● SNP-A can detect
microdeletions and regions
of UPD that are
undetectable by MC or FISH
● Can improve detection of
del(5q) in conjunction with
MC and FISH
1. Mallo M, et al. Haematologica 2008;93:1001–8; 2. Steidl C, et al. Leuk Res
2005;29:987–93; 3. Makishima H, et al. Leuk Res 2010;34:447–53
UPD: uniparental disomy
SNP: single-nucleotide polymorphism
19
Technological advances: FISH for improving
diagnosis of primary MDS
Luciano L, et al. Poster presentation at EHA 2010,
Barcelona, Spain. Abstract 0917
Patients and methods
• 121 patients with suspected primary MDS
• Conventional cytogenetic analysis of BM samples
– informative karyotype (≥20 metaphases obtained) for 90 patients
• Samples from remaining 31 patients analysed by FISH
Normal
Del(5q)
Trisomy 8
–7/7q–
Other*
Failure
*hypo/hyperdiploid karyotype, complex karyotype,
unusual deletions and/or translocations
37%
27%
20% 9%
7%
70%
18%
6%
5%
Conventional cytogenetics FISH
20
FISH and del(5q)
FISH detected del(5q) in 38/637 (6%) cases
in which MC had not identified 5q–
Mallo M, et al. Haematologica 2008;93:1001–8
Group A
5q– not detected by MC (n=637)
Group B
5q– detected by MC (n=79)
Screening (n=716)
BM analysis by MC
Study design
Authors’ recommendation:
FISH may help in classification of cases with
● Suspected 5q– syndrome
● No metaphases or metaphases are not evaluable
● Abnormal karyotype without evidence of 5q–
21
SNP arrays can detect additional ‘cryptic’
abnormalities
Tiu RV, et al. Blood 2011;117:4552–60
Increased detection of cytogenetic
abnormalities
Newly detected lesions by SNP-A indicated a poorer patient prognosis
Conventional
karyotyping
Conventional
karyotyping +
SNP array
20
0
40
60
80
Ch
rom
oso
mal
ab
no
rmaliti
es d
ete
cte
d (
%)
44%
74%
p<0.0001 N=430
Novel lesions were
detected with SNP in:
• 54% of normal/non-
informative conventional
karyotyping results
• 62% of abnormal
conventional karyotyping
results
22
MC and SNP-A analyses of 5q disorders:
impact of lesions on prognosis
Jerez A, et al. J Clin Oncol 2012; Epub ahead of print
OS PFS
Loss of additional genes in the distal regions of 5q may contribute to a
poorer prognosis in patients with del(5q) abnormalities
1.0
0.8
0.6
0.4
0.2
0.0
0 10 20 30 40 50 60
Time (months)
Ove
rall
su
rviv
al
(pro
po
rtio
n)
5q-syndrome (n=29; events, 5)
Other forms of del(5q) MDS not
involving 5qSy-CRR (n=33; events, 10)
Other forms of del(5q) MDS involving
5qSy-CRR (n=42; events, 28)
P=0.06
P=0.04
1.0
0.8
0.6
0.4
0.2
0.0
0 10 20 30 40 50 60
Time (months)
Pro
gre
ss
ion
-fre
e s
urv
iva
l
(pro
po
rtio
n)
5q-syndrome (n=29; events, 3)
Other forms of del(5q) MDS not
involving 5qSy-CRR (n=33; events, 6)
Other forms of del(5q) MDS involving
5qSy-CRR (n=42; events, 19)
P=0.09
P=0.02
P≤0.001
PFS = progression-free survival
23
Considerations with cytogenetic techniques
Metaphase cytogenetics only Complement with FISH?
FISH is limited by the probes
used; it is too expensive to use
probes against entire genome
FISH must complement, not
replace, metaphase
cytogenetics
SNP-A is much more sensitive
than FISH, but it can miss
chromosomal translocations
If SNP-A becomes routinely
used, it should be in conjunction
with other techniques
24
FISH on enriched CD34+ PB cells to monitor
genetic response to lenalidomide: LE-MON-5 study
Shirneshan K, et al. Poster presentation at
ASH 2011, San Diego, CA. Abstract 1721
Preliminary cytogenetic findings in 37 patients with lower-risk
WHO-defined MDS and isolated del(5q) treated with lenalidomide
Every 6 months
Karyotyping and FISH on BM aspirates
*Based on time to cytogenetic remission †Each line represents one patient
FISH analysis of CD34+ PB cells, in combination with standard MK, could be a
powerful non-invasive tool to monitor response to lenalidomide therapy
Fast responders (n=14)* >50% clone size reduction in
≤2 months†
0 2 4 6 8 10 12 14
100
80
60
40
20
0
100
80
60
40
20
0
0 2 4 6 8 10 12 14 16 18
Months Months
Clo
ne s
ize
(%
)
Non-responders (n=9)* No cytogenetic response
detected
Clo
ne
siz
e (
%)
Slow responders (n=14)* >50% clone size reduction in
>2 months†
Resp
on
se
Study start
Karyotyping and FISH on BM aspirates
FISH on enriched CD34+ PB cells
Every 2–3 months
FISH on enriched CD34+ PB cells
25
1. Haase D, et al. Blood 2007;110:4385–95
2. Giagounidis A, et al. Hematology 2004;9:271–7; 3. Germing U, et al. Leukemia 2012; Epub ahead of print
4. Braulke F, et al. Leuk Res 2010;34:1296–301; 5. Tiu RV, et al. Blood 2011;117:4552–60
Summary and conclusions
• Del(5q) is the most common cytogenetic abnormality in patients
with MDS1
• Some patients with isolated del(5q) MDS have distinct features:
‘5q– syndrome’2
• Del(5q) MDS is highly heterogeneous: a number of factors have a
negative impact on prognosis, e.g.3
– Transfusion dependence
– Elevated blast count
– Additional genomic aberrations
– Thrombocytopenia
– Age >65 years
• Technological advances, e.g. FISH analysis on PB CD34+ cells and
SNP arrays, will facilitate the monitoring of disease evolution and
response to therapy in individual patients4,5
• When considering treatment options, clinicians need to assess the
risk profile of individual patients
26
Lenalidomide
Clinical effects
Cellular effects
Replacement of del(5q)
precursors by normal ones2
Pro-erythroid effect on
normal cells2
Initial worsening of
cytopenias3,4
Median time
to RBC-TI:
4.6 weeks3
Within days?
Median time to
CyR: 8 weeks
>1 year
Targeted apoptosis
of abnormal clone1
Increased normal
erythropoiesis2
Reduced probability of
progression to AML?4
Reduced RBC-transfusion
requirement3,4
Achievement of
RBC-TI3,4
Reduced probability of
non-leukaemic death
Cycle 1–2
RBC-TD patient with lower-risk del(5q) MDS:
severe anaemia; no other cytopenias
Possible link between mechanism of action of
lenalidomide and observed clinical data
1. Wei S, et al. Proc Natl Acad Sci USA 2009;106:12974–9; 2. Ximeri M, et al. Haematologica 2010;95:406–14
3. List AF, et al. N Engl J Med 2006;355:1456–65; 4. Fenaux P, et al. Blood 2011; 118:3765–76
High probability of
cytogenetic remission3
27
*MDS-003: >50,000/Μl †As defined by the IWG 2000 criteria ‡Double-blind phase; at 16 weeks, responders continued
double-blind treatment for 1 year, non-responders entered
open-label treatment or withdrew
MDS-0031
(n=148)
• Primary endpoint:
– erythroid response†
– RBC-TI (for ≥56
consecutive days)
• Secondary endpoints:
– RBC-TI duration
– cytogenetic response
– tolerability
Patient
characteristics
• IPSS low-/int-1-risk
MDS
• Del(5q) ± additional
abnormalities
• RBC transfusion
dependent
• ANC >500 cells/µL
• Platelets >25,000/µL* MDS-0042
(n=205)
Lenalidomide 10mg/day
21/28-day cycles
For 24 weeks
Lenalidomide‡
5mg/day
28-day cycles
Lenalidomide‡ 10mg/day
28-day cycles
Placebo‡
OR
OR
• Primary endpoint:
– RBC-TI for ≥26 weeks
• Secondary endpoints:
– erythroid response
– RBC-TI duration
– cytogenetic response
– time to AML
progression
– QoL
– tolerability
MDS-003 and MDS-004 study design
1. List A, et al. N Engl J Med 2006;355:1456–65
2. Fenaux P, et al. Blood 2011;118:3765–76
28
Achievement of CyR was associated
with significantly longer OS Risk of AML progression appeared to
be higher in patients without CyR
Time to death (years) 0 1 2 3 4 5 6
1.0
0.8
0.6
0.4
0.2
0
7
Pro
ba
bil
ity o
f s
urv
iva
l
CyR No CyR
0 1 2 3 4 5 6
1.0
0.8
0.6
0.4
0.2
0
7
Log-rank
p=0.0517
Log-rank
p<0.0001
Time to AML (years)
Pro
bab
ilit
y o
f A
ML
pro
gre
ss
ion
Analysis included a total of 286 lenalidomide-treated patients with del (5q) MDS
Combined analysis of MDS-003 and MDS-004:
impact of CyR on clinical outcomes
Giagounidis A, et al. Oral presentation at 11th International
Symposium on MDS 2011, Edinburgh, UK. Abstract 58 CyR = cytogenetic response
29
AML-free survival by RBC-TI for ≥8 weeks
in patients randomised to lenalidomide*
OS by RBC-TI for ≥8 weeks in patients
randomised to lenalidomide*
In patients treated with lenalidomide, achievement of RBC-TI for ≥8 weeks was
associated with improved OS and reduced risk of AML progression
AM
L-f
ree
su
rviv
al
1.0
0.8
0.6
0.4
0.2
0
Time from randomisation (years)
0 1 2 3 4 5
Log-rank p=0.0085
RBC-TI responders
RBC-TI non-responders
Ove
rall
su
rviv
al
1.0
0.8
0.6
0.4
0.2
0
Time from randomisation (years)
0 1 2 3 4 5
Log-rank p=0.0028
RBC-TI responders
RBC-TI non-responders
MDS-004: OS and progression to AML in
patients who achieved RBC-TI
Fenaux P, et al. Blood 2011;118:3765–76 *Landmark 6-month analysis
30
Karyotype complexity had a negative
impact on OS
Pro
ba
bil
ity o
f s
urv
iva
l
Time to death (years)
1.0
0.8
0.6
0.4
0.2
0 0 1 2 3 4 5 7 6
3.9 years
3.0 years
1.8 years
p<0.01 p<0.05
Time to AML (years)
1.0
0.8
0.6
0.4
0.2
0
Pro
ba
bil
ity o
f A
ML
pro
gre
ss
ion
0 1 2 3 4 5 7 6
Median OS
Log-rank
p=0.0016 Log-rank
p=0.0044
Karyotype complexity had a negative
impact on AML progression
Isolated del(5q) del(5q) + 1 del(5q) + ≥2
Analysis included a total of 286 lenalidomide-treated patients with del (5q) MDS
Combined analysis of MDS-003 and MDS-004: impact
of karyotypic complexity on clinical outcomes
Giagounidis A, et al. Oral presentation at 11th International
Symposium on MDS 2011, Edinburgh, UK. Abstract 58
31
Combined analysis of MDS-003 and MDS-004:
predictive factors for progression to AML
Giagounidis A, et al. Oral presentation at 11th International
Symposium on MDS 2011, Edinburgh, UK. Abstract 58
Covariate Univariate analysis
p-value
Multivariate analysis
Hazard ratio (95% CI) p-value
Transfusion burden,
units/8 weeks
<0.001 1.13 (1.06–1.21) <0.001
del(5q) ≥2 vs isolated del(5q) <0.001 3.97 (1.89–8.33) <0.001
RBC-TI ≥26 weeks (no vs yes) <0.001 NS
Patients with del(5q) MDS and ≥2 additional cytogenetic abnormalities were at ~97% increased
risk of progression to AML than patients with isolated del(5q) MDS
32
Objective: to retrospectively compare AML progression and OS in
lenalidomide-treated vs untreated RBC-TD patients with lower-risk del(5q) MDS
MDS-004
n=205
*Nine centres of MDS
registries in Europe, USA, Australia
Eligibility criteria for present study:
• IPSS low- or int-1 risk MDS with del(5q)
• RBC-TD (≥1 unit/8 weeks)
• ANC ≥500/μL
• Platelets ≥50,000/μL
Lenalidomide-
treated patients
n=295 Median follow-up:
4.3 years
(range 0.02–6.8)
Untreated
patients
n=125 Median follow-up:
4.6 years
(range 0.06–19.4)
Multicentre
registry*
n=459
MDS-003
n=148
Patients were selected from clinical trials and registries
Comparative study of AML progression and OS
with lenalidomide therapy: design
Kuendgen A, et al.
Oral presentation at ASH 2011; San Diego, CA. Abstract 119
33
Comparative study of AML progression and OS with
lenalidomide therapy: baseline characteristics
Kuendgen A, et al. Oral presentation at ASH 2011;
San Diego, CA. Abstract 119
*Nine centres of MDS
registries in Europe, USA, Australia
Baseline characteristics Lenalidomide treated Untreated
Median age, years 65 66
Median RBC transfusions,
units/8 weeks 6 2
BM blasts
5–10%, % 13 16
Median Hb, g/dL 8 8
Cytogenetics, %
isolated del(5q)
del(5q) +1 abn
del(5q) + >1 abn
76
18
5
83
13
4
IPSS risk, %
low-1
int-1
43
57
43
57
34
125 87 66 48 31 24 14
53 101 125 135 127 101
Number of patients at risk
Time (years)
0 1 2 3 4 5 6
1.0
0.8
0.6
0.4
0.2
0
Su
rviv
al p
rob
ab
ilit
y
Log-rank p= 0.755
Untreated
OS 3.8 years (95% CI: 2.9–4.8)
LEN-treated
OS 5.2 years (95% CI: 4.5–5.9)
Comparative study of AML progression and OS
with lenalidomide therapy: OS
Kuendgen A, et al. Oral presentation at ASH 2011;
San Diego, CA. Abstract 119
35
125 79 58 39 24 18 9
53 97 119 128 120 98
Number of patients at risk Time (years)
0 1 2 3 4 5 6
1.0
0.8
0.6
0.4
0.2
0
Cu
mu
lati
ve
in
cid
en
ce
of
AM
L p
rog
res
sio
n
Untreated
LEN-treated
Median time to AML progression: not reached
Comparative study of AML progression and OS
with lenalidomide therapy: progression to AML
Kuendgen A, et al. Oral presentation at ASH 2011;
San Diego, CA. Abstract 119
36
*Only significant when using multivariate
Cox proportional hazard model
Predictive factors for AML progression and OS
Hazard ratio (95% CI)
Age, years
LEN-treated vs untreated
Gender (female vs male)
Transfusion burden,
units/8 weeks
IPSS risk (int-1 vs low) p=0.041
p=0.741
p=0.017
p=0.003*
p=0.723
p=0.002
LEN-treated vs untreated
IPSS risk (int-1 vs low)
Transfusion burden,
units/8 weeks
p<0.001
p=0.003
AM
L p
rog
ressio
n
De
ath
Increased Decreased
Relative hazard (multivariate Cox proportional hazard model)
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Treated patients showed improved survival
Lenalidomide was not associated with increased risk of AML progression
Comparative study of AML progression and OS
with lenalidomide therapy: predictive factors
Kuendgen A, et al. Oral presentation at ASH 2011;
San Diego, CA. Abstract 119
37
Summary and conclusions in patients with
del(5q) MDS
Lenalidomide induces durable erythroid responses; ~65% achieved RBC-TI
Lenalidomide treated and untreated patients showed no difference in the risk of AML
Possible risk factors for progression for treated and untreated patients included
Increased medullary blast count
Complex karyotype
Transfusion need
No response to treatment (CyR/ RBC-TI)
P53 mutation?
Advice:
Response assessment at ~4 months, end of treatment in case of non-response
0 24 48 72 96 120 144
Intensive chemotherapy (IC) in patients with
int-2/high-risk MDS: low survival rate
1. Kantarjian H, et al. Cancer 2006;106:1099–109
2. National Cancer Institute. SEER Cancer Statistics Review 1975–2006, MDS
Su
rviv
al
(% p
ati
en
ts)
100
90
80
70
60
50
40
30
20
10
0
Months
OS after 5 years 8%
n=510; Median age 63 years1
CR rate 55%
Induction mortality 17%
Rate
per
10
0,0
00
po
pu
lati
on
Age at diagnosis (years)
MDS & age at diagnosis2
50
40
30
10
0
20
<0 40–49 50–59 60–69 70–79 >80
0.2 0.8 2.3
8.5
24.8
43.9
Survival with palliative treatment of high-risk MDS and
AML in patients aged >60 years not fit for IC
• An analysis of 36 studies (median age: 70 years)1
• Median survival was 4 months in a randomised study (low-dose cytarabine
[LDAC] vs hydroxyurea)2
1. Deschler B, et al. Haematologica 2006;94:1513–22; 2. Burnett AK, et al. Cancer 2007;109:1114–24
Treatment Median survival (weeks)
Low-intensity therapies 12
BSC alone 7.5
No. patients No. events Obs. – Exp.
LD Ara-C 103 100 −22.0
HU 99 99 22.0
p=0.0009
Pe
rce
nta
ge
sti
ll a
live
100
75
50
25
0 0 1 2 3
Years from entry
Total number of HSCTs*
25,563
Allogeneic
39% Autologous
61%
Indication for HSCT
Lymphoproliferative
disorders
Acute lymphoblastic
leukaemia
Solid tumours
Non-malignant
disorders
Chronic lymphocytic
leukaemia
Chronic myeloid
leukaemia
Myeloproliferative
syndromes
AML + MDS
20%
Gratwohl A, et al. Bone Marrow Transplant 2009;43:275–91
In 613 centres in 42 countries
RIC = reduced-intensity conditioning
1. Effective consolidation therapy [ex. haemopoieitc stem cell transplant
(HSCT) including RIC-HSCT]
How can survival be improved in elderly patients with
high-risk MDS and AML?
Median follow-up 42 months (n=256), median age 62 years
OS
(%
)
OS = 40%
Years after HCT Years after HCT
DF
S (
%)
DFS = 35%
Years after HCT
RI
(%)
RI = 48%
Years after HCT
NRM = 27%
NR
M (
%)
Al-Ali HK, et al. Oral presentation at EHA 2011
Outcome at 5 years after RIC-HCT in MDS and AML
1.0
0.8
0.6
0.4
0.2
0 0 5 10
1.0
0.8
0.6
0.4
0.2
0 0 5 10
1.0
0.8
0.6
0.4
0.2
0 0 5 10
1.0
0.8
0.6
0.4
0.2
0 0 5 10
Azacitidine prolongs overall survival in patients with
IPSS Int-2 or High-risk MDS
Cytogenetic
risk group Patients, % (n/N)
Median OS, months
Aza CCR HR
Good 47 (167/358) Not reached 17.1 0.59
Intermediate 21 (76/358) 26.3 17.0 0.44
Poor 28 (100/358) 17.2 6.0 0.53
CCR = conventional care regimen
Pati
en
ts s
urv
ivin
g (
%)
Time since randomisation (months)
Intent-to-treat population
0
20
40
60
80
100
0 5 10 15 20 25 30 35 40
Aza (OS 24.5 months; 82 deaths)
CCR (OS 15.0 months; 113 deaths)
p=0.0001
Fenaux P, et al. Lancet Oncol 2009;10:223–32
Gotze KS, et al. Cancer Manag Res 2009;1:119–30
LDAC
1. Kantarjian H, et al. Cancer 2006;106:1099–109; 2. Burnett AK, et al. Cancer 2007;109:1114–24
Only CR has a favorable impact on survival with
chemotherapy in MDS and AML
IC in MDS1
Median survival of 66 days in
non-remitters vs 575 days in patients
with CR (overall CR rate of 18%)2
1.0
0.8
0.6
0.4
0.2
0
Pro
po
rtio
n s
urv
ivin
g
0 24 48 72 96 120 144
Months
Total Dead
Complete
response 252 192
Other 154 135
p<0.0001
AML 14 non-intensive – overall survival
Favourable/intermediate
Pe
rce
nta
ge s
till a
live
100
75
50
25
0
0 1 2 3 Years from entry
AML 14 non-intensive – overall survival
Adverse
0 1 2 3 Years from entry
Pe
rce
nta
ge s
till a
live
100
75
50
25
0
p=0.4
p=0.004
LD Ara-C
HU
LD Ara-C
HU
The CR rate was similar in the AZA and CCR groups, suggesting that
AZA-mediated improvement in OS occurred independently of response
Fenaux P, et al. J Clin Oncol 2010;28:562–9 *WHO-defined AML
0
5
10
15
20
25
AZA
(n=55)
CCR
(n=58)
Pati
en
ts w
ith
CR
(%
)
18 16
p=0.8
0
20
40
60
80
100
0 5 10 15 20 25 30 35 40
Pati
en
ts s
urv
ivin
g (
%)
Time since randomisation (months)
CCR (n=58)
AZA (n=55)
Median OS
2-y
ear
OS
rate
AZA-001: response in elderly patients with
20–30% blasts*
Adapted from List AF, et al. Oral presentation at
ASCO 2008, Chicago, IL, USA
Time from randomisation (months)
Pati
en
ts s
urv
ivin
g (
%)
0 5 10 15 20 25 30 35 40 0
20
40
60
80
100
HI (p<0.0001)
PR (p=0.006)
CR
CCR
71.7%
78.4% (p<0.0001)
26.2%
2-year survival rates
AZA-001: OS by best response
OS by best response (IWG, 2000)
0 1 2 3
Survival according to
cytogenetics1
1. Knipp S, et al. Cancer 2007;110:345–52; 2. Burnett AK, et al. Cancer 2007;109:1114–24
Adverse cytogenetics has a strong negative impact on
survival after chemotherapy in MDS and AML
Pa
tie
nts
su
rviv
ing
(%
)
Duration (years)
0 1 2 3 4 5 6 7 8 9 10 11
100
Normal
Abnormal, non-complex
Complex
p=0.0001
No CR with LDAC in adverse
cytogenetics!!2
After IC
LDAC
AML 14 non-intensive – overall survival
Favourable/intermediate
Pe
rce
nta
ge s
till a
live
100
75
50
25
0
0 1 2 3 Years from entry
AML 14 non-intensive – overall survival
Adverse
Years from entry
Pe
rce
nta
ge s
till a
live
100
75
50
25
0
p=0.4
p=0.004
80
60
40
20
0
LD Ara-C
HU
LD Ara-C
HU
AZA 001: AZA vs LDAC in patients with IPSS
Int-2- or High-risk MDS regarding cytogenetics
AZA LD Ara-C
n 45 49
Median number of cycles 9 4.5
Median OS 24.5 15.3 p<0.001
OS good-risk cytogenetics NR 19 HR=0.46
OS poor-risk cytogenetics 24.5 2.9 HR=0.07
CR + PR 31% 12% p=0.047
HI 53% 25% p=0.006
Transfusion independent 45% 13% p=0.01
Severe infections/patient-year 0.44 1 p=0.017
Median days in hospital/patient-year 18 27 NR
Fenaux P, et al. Br J Haematol 2010;149:244–9; Fenaux P, et al. Lancet Oncol 2009;10:223–32
Old age is not a contraindication to treatment with AZA
0
0.2
0.4
0.6
0.8
1.0
0 10 20 30 40
Time from randomisation (months)
AZA
CCR
HR: 0.48 (95% CI: 0.26–089); p=0.0193
No. at Risk AZA 38 31 27 14 9 3 0 0 0
CCR 49 37 23 16 5 3 1 0 0
10.8 months
15%
55%
Pro
po
rtio
n s
urv
ivin
g
Seymour JF, et al. Crit Rev Oncol Hematol 2010;76:218–27
AZA-001 sub-group analysis ≥75 years of age
Azacitidine in patients with AML
0
20
40
60
80
100
0 5 10 15 20 25 30 35 40
24.5 months
16.0 months
Pati
en
ts s
urv
ivin
g (
%)
Time since randomisation (months)
p = 0.004
CCR (n=58)
Azacitidine (n=55)
50.2%
15.9%
*WHO-defined AML Fenaux P, et al. J Clin Oncol 2010;28:562–9
Azacitidine prolongs overall survival in elderly patients with
20–30% blasts*
AZA-001: analysis of patients with 20–30% blasts
(WHO AML) – infections and hospitalisation
Rates of
hospitalisation
Rates of infection requiring
i.v. antibiotics
0
0.4
0.8
1.2
AZA CCR
p=0.003
Ra
te p
er
pa
tie
nt
ye
ar
0.6
1.1
p=0.05
0
1
2
3
4
5
AZA CCR
Ra
te p
er
pa
tie
nt
ye
ar
3.4
4.3
Number of days
in hospital
Fenaux P, et al. J Clin Oncol 2010;28:562–9
0
20
40
60
AZA CCR
p<0.001
Da
ys
pe
r p
ati
en
t ye
ar
26
51
• AZA 75 mg/m2/day subcutaneously for 5 days every 4 weeks
• Median follow-up duration was 13 months (range: 9–16 months)
OSHO trial: AZA in patients with newly diagnosed or
relapsed/refractory AML
Characteristic
All patients
(n=40)
Newly
diagnosed
(n=20)
Relapsed or
refractory
(n=20)
Median age, years 72 78 67
Male, n (%) 19 (48) 11 (55) 8 (40)
Median BM blasts, % 42 44 40
Median WBC count × 109/L 3.6 3.4 3.6
Cytogenetics,† n (%)
Intermediate risk
High risk
28 (70)
12 (30)
15 (75)
5 (25)
13 (65)
7 (35)
Al-Ali HK, et al. Leuk Lymph. 2012;53:110–7
OS in patients with newly diagnosed
or relapsed/refractory AML
Median OS for patients with SD was 4 months, whereas median OS for patients with
CR, PR, or HI was not reached (p=0.045); (median bone marrow blasts 42%)
OS according
to haematological response
Pa
tien
ts s
urv
ivin
g (
%)
Duration (months)
18 12 6 0
100
80
60
40
20
0
SD censored
CR/PR/HI censored
SD (n=15)
PD or NR (n=2)
CR, PR, or HI (n=12)
NE (n=11) 100
80
60
40
20
0 0 2 4 6 8 10
2.9 months
7.7 months
Pa
tien
ts s
urv
ivin
g (
%)
Duration (months)
Newly diagnosed (n=20)
Relapsed or refractory (n=20)
p=0.04
AZA improves survival in AML patients with HI or
better and newly diagnosed patients have longer OS
Al-Ali HK, et al. Leuk Lymph 2012;53:110–7
Bories P, et al. ASH 2011. Abstract 2614
n=98
Median age 76 years
WBC >10 g/L 14 (14.2%)
Median BM blasts, % 35 (20–85)
Cytogenetic risk, n (%)
Favourable: 0
Intermediate: 48 (48.9)
Adverse: 44 (44.8)
AZA
• 75 mg/m2/day x 7
days, 28-day
cycles
– median 6
(1–27) cycles
AZA in elderly AML patients unfit for IC
13
56
26
0
5
10
15
20
25
30
Pati
en
ts (
n)
ORR (CR + CRi + PR + HI) = 51%
CR CRi PR HI
Toxicity
• 60 (61.2%) patients required
hospitalisation for infections
• Death rate in first 2 months: 17%
– from AML progression: 83%
– from infections: 17%
HR (CI) p value
Performance status (≥2 vs <2) 0.98 (0.49–1.96) 0.96*
LDH (>normal vs ≤normal) 1.98 (1.02–3.88) 0.045
Cytogenetics (adverse vs intermediate) 2.38 (1.29–4.41) 0.005
AZA appears to be a valuable option for AML patients considered unfit for CT
OS by patient characteristics: multivariate analysis
OS (n=98) OS by cytogenetics (n=92)
100
75
50
25
0 0 6 12 18 24 30 36 42 48 54 60
Months
Pa
tien
ts s
urv
ivin
g (
%)
Months
p=0.005
Intermediate
Adverse
1-year OS: 50%
2-year OS: 28%
Pa
tie
nts
su
rviv
ing
(%
)
AZA in elderly AML patients unfit for IC
100
75
50
25
0 0 6 12 18 24 30 36 42 48 54 60
Bories P, et al. ASH 2011. Abstract 2614
AZA versus IC or BSC in 182 elderly WHO-AML
patients
Serrano J, et al. ASH 2011. Abstract 2612
Retrospective study of AZA compared with IC or BSC on OS
in patients >60 years old
AZA†
(n=67)
IC
(n=47)
BSC†
(n=68) p value
Median age, years
(range)
71 (60–84) 65 (60–75) 76 (62–89) 0.04,
<0.01*
Median leucocyte count,
×109/L (range)
4.4
(0.2–17.5)
28.6
(0.8–166)
17.6
(0.4–211)
0.01,
<0.01*
Median BM blasts count,
% (range)
34 (20–84) 64 (24–93) 47 (21–90) 0.003,
<0.01*
ECOG ≥2, n (%) 45 (67) 7 (15) 60 (88) <0.01,
<0.01*
Cytogenetics, n (%)
Intermediate
Adverse
NT
46 (68.6)
15 (22.3)
6
32 (68)
12 (25.5)
3
39 (57.3)
12 (17.6)
17
0.86*
Patient characteristics
*IC vs AZA groups; †patients not eligible for IC
AZA group: • AZA 75–100 mg/m2/day s.c.
(7 days of 28-day cycle)
• Median cycles: 6 (1–24)
• Median follow-up:
7.4 (1–28.3) months
IC group: • Cytarabine
100–200 mg/m2/day c.i.v.
(7 days)
• + Idarubicin
9–12 mg/m2/day (3 days)
• + HSCT
• Median follow-up:
13 (2–46) months
BSC group: • Oral CT agents
• Transfusions
• Antibiotics with G-CSF
• Median follow-up:
5 (0–10) months
13 13
16
Response to AZA
Significantly better survival with AZA
was associated with:
CR, PR or HI (p=0.018)
ORR = 38%
1- and 2-year OS
Pa
tien
ts (
%)
p=0.75 AZA vs IC
OS
0 10 20 30 40 50
AZA or ICT vs BSC:
p<0.001
0.0
0.2
0.4
0.6
0.8
1.0
Pro
po
rtio
n a
live
BSC (n=68)
AZA (n=67)
IC (n=47)
2.03 mo
13.7 mo 11.2 mo
Months
In elderly patients with AML not
eligible for IC, AZA led to an OS
comparable to the one observed
in patients treated with IC
AZA versus IC or BSC in 182 elderly WHO-AML
patients: outcomes
0
5
10
15
20
CR PR HI
AZA IC BSC
1 year
2 years
0
20
40
60
Serrano J, et al. ASH 2011. Abstract 2612
Damaj G, et al. ASH 2011. Abstract 160
Prior to HSCT
Characteristics
BSC
(n=171)
AZA
(n=49)
Induction CT
(n=180)
AZA + induction CT
(n=17) p value
IPSS, n (%)
Low/Int-1
Int-2/High
108 (64)
60 (36)
16 (33)
32 (67)
73 (42)
102 (58)
7 (41)
10 (59)
0.0001
Retrospective analysis: 417 patients receiving allogeneic HSCT between 1999 and 2009
Characteristics BSC (n=171)
AZA
(n=49)
Induction CT
(n=180)
AZA + induction CT
(n=17) p value
Median age, years (range) 51 (44–57) 60 (30–62) 56 (18–61) 58 (32–61) 0.0001
Disease progression, n (%) 32 (19) 7 (15) 95 (53) 10 (59) 0.0001
BM blasts <5%, n (%) 77 (52) 33 (67) 131 (74) 9 (53) 0.0003
Responders, n (%) – 33 (69) 131 (73) 9 (53) 0.30
Characteristics at diagnosis, per pre-HSCT therapy
Characteristics at transplantation, per pre-HSCT therapy
AZA & allogeneic HSCT in MDS and AML
3-year NRM per prior treatment 3-year OS per prior treatment
p=0.036* 3-y
ea
r O
S
AZA
Induction CT
AZA + induction CT
p=0.027*
200 400 600 800 1,000 1,200 0
0.4
0
0.2
0.6
0.8
1.0
Time (days)
p=0.10
200 400 600 800 1,000 1,200 0
0.4
0
0.2
0.6
0.8
1.0
Time (days)
p=0.138
3-y
ea
r N
RM
AZA & allogeneic HSCT in MDS and AML
AZA monotherapy appears a valid pre-allogeneic HSCT treatment approach
*Azacitidine vs azacitidine + induction CT
AZA
Induction CT
AZA + induction CT
Prior to HSCT: OS and NRM by treatment
Damaj G, et al. ASH 2011. Abstract 160
1. Jabbour E, et al. Cancer. 2009;115:1089–905
2. Schroeder T et al. ASH 2011. Abstract 656; 3. Goodyear O et al. Blood, 2012;119:3361–9
AZA & allogeneic HSCT in MDS and AML After HSCT
● AZA alone or in combination with donor lymphocyte infusions may induce
durable remissions for patients with acute leukaemia who develop disease
recurrence after HSCT1,2
● AZA is well tolerated and not associated with exacerbation of GVHD
● AZA increases the
expansion of regulatory
T-cells after allogeneic
HSCT in patients with AML3
These data support the further examination of AZA after HCT as a mechanism
of augmenting a GVL effect without a concomitant increase in GVHD
2.0x102
1.5x102
1.0x102
5.0x102
0
CD
4+
CD
25
+C
D1
27lo
Fo
x
P3
+c
ell
s/L
P=0.0127
Conclusions
• Although high CR rates are achieved with standard chemotherapy,
long-term survival rates remain low in patients with high-risk MDS
& AML
• AZA improves OS versus CCR independent of response, adverse
cytogenetics, age and comorbidities in high-risk MDS & AML
(20–30% blasts)
• The exact role of AZA in AML (>30% blasts) in older patients is
being assessed in the AML-001 study
• AZA has a manageable safety profile and can often be administered
in an outpatient setting
• AZA appears to be a valid pre-allogeneic HSCT treatment approach
• Early data indicate that post-HSCT AZA may augment a GVL effect
without a concomitant increase in GVHD