Nessun titolo diapositiva

1
Genomic and proteomic predictors of sensitivity
to preoperative chemotherapy
Stefano Iacobelli Università G. dAnnunzio
Chieti-Pescara
Dipartimento di Oncologia Neuroscienze
2
No. of patients needed to treat to obtain one
favorable event Adjuvant hormonotherapy 15-30
Adjuvant chemotherapy 2-15 Metastatic
disease 2-10 Neoadjuvant chemotherapy 2-15
pCR
3
Preoperative systemic treatment (neoadjuvant or
primary therapy)
1. Allows the tumor to be used as a measure of
response in vivo
2. Trials in this setting require fewer patients
and can be completed rapidly
4
No difference in survival between neoadjuvant or
adjuvant CT
5
80 (noninflammatory) breast cancers
mastectomy
6
Molecular analyses in pretreatment biopsy samples
TP53 status determined by the yeast functional
assay Mutant TP53 transcripts red colonies
Wild-type transcripts white colonies
IHC ESR1 (estrogen receptor 1) KRT5/6/17
(keratin 5/6/17) ? basal features CDKN2A
(cyclin-dependent kinase inhibitor 2A) ? cell
cycle arrest PROM1 (prominin 1) ? stem cell
RT-PCR ERBB2
7
80 pts
52 WT p53
28 mutated p53
0 pCR
15 pCR
Complete disappearance of invasive tumor cells
in the mastectomy specimen and in the lymph
nodes, or a single microscopic invasive focus, or
residual in situ carcinoma
8
pCR according to clinical or biological features
(dark yellow and bold type, p0.0001 light
yellow,p0.0001, or p0.01)
TP53 status significantly linked to pCR in
univariate analysis After stratification for
TP53 status, all other variables, with the
exception of KRT5/6/17, remained no longer
significant
9
Kaplan-Meier analysis of the event-free and
overall survival (stratified by TP53 status)
TP53 mutant tumors had a favorable, although not
significant, event-free survival (46 mo median
F-U, too short?)
Among patients who did not reach pCR, those with
WT-TP53 tumors had a more favorable overall
survival compared with those with TP53-mutant
tumors
10
The role of gene profiling in predicting pCR
77 genes correlated with pCR (54 were
p53-associated genes)
No profile predicts pCR among TP53 mutant tumors!
11
Conclusion TP53 mutations are highly predictive
of pCR to a dose-dense neoadjuvant
epirubicincyclophosphamide regimen.
Why did previous studies fail to identify this
relationship?
1. TP53 status, using yeast functional assay,
detects not only mutations and deletions, but
also abnormal splicing events 2. The dose
intensity of the regimen is 2-4 times higher than
previous ones.
12
IHC subtypes of breast cancer
Basal-like ER - PR - HER2 - HER2 ER - PR
- HER2 Luminal B ER/PR HER2 Luminal
A ER/PR HER2 -
13
107 breast cancers
q14 days (28 pts) q21 days (79 pts)
Doxorubicin 60 mg/m2 Cyclophosphamide 600
mg/m2
4 cycles
Taxanes (80 pts)
Surgery (27 pts)
Surgery
Adjuvant hormonal therapy for HR positive
14
The higher pCR (stage0) rate was observed in
basal-like and HER-2 subtypes
15
DDFS and OS according to breast cancer subtype
16
DDFS among patients with residual disease after
CT by breast cancer subtype
17
Conclusion
Patients with basal-like and HER2/ER- subtype
have higher rates of pathologic complete response
to neoadjuvant anthracycline-based CT than those
with luminal subtype Patients achieving a
pathologic complete response have a highly
favorable outcome. Patients with basal-like and
HER2/ER- subtype with residual disease have
early relapse and death compared with those with
luminal subtype
18
(No Transcript)
19
Caspase-3 and caspase-3s expression in breast
tumor cell lines (A and B) and the effect of
caspase-3 and/or caspase-3s transfection in MCF-7
(C) and HBL-100 (D) cells on cyclophosphamide
(CPA)-induced apoptosis.

• Overexpression of caspase-3 and caspase-3s
  inhibits CPA-induced apoptosis y (P lt 0.0001)
  both in MCF-7 and HBL-100 cell lines

20
130 pts with breast cancer (analyzed
retrospectively)
40 pts
90 pts with LABC
Surgery
Neodjuvant therapy with FEC100 (4-6 cycles)
Adjuvant therapy
Surgery
Pretreatment biopsy samples for RT-PCR evaluation
of Caspase-3 Caspase-3s
21

• Expression of caspase-3 is higher
• in cancer than in adjacent normal
• tissues
• No differences for caspase-3s

22
No relation between the expression of the
caspase-3, caspase-3s, or the ratio
caspase-3s/caspase-3 and patient characteristics
23
No relation between caspase-3, caspase-3s, or the
caspase-3s/caspase-3 and clinical outcome
24
90 pts with LABC
FEC100
30 Responders
60 Nonresponders
Responders (pCR) included patients with no
evidence of residual tumor in the breast or
axillary lymph nodes and patients with only
residual in situ carcinoma.
Caspase-3s/caspase-3 ratio significantly
increased in nonresponders
25
Increased caspase-3s expression is an independent
indicator of no response to chemotherapy
26
Background
formalin-fixed paraffin-embedded (FPE)
tumors (447 pts)
RT-PCR (259 candidate genes)
21-gene panel (Oncotype DX 21)
Recurrence Score (RS) algorithm.
The RS was clinically validated to quantify the
risk of distant recurrence in 668 women with ER,
N0 breast tumors treated with adjuvant tamoxifen
in a clinical trial conducted by the NSABP Is RS,
along with additional genes, associated with pCR?
27
102 locally advanced breast cancers
doxorubicin 60 mg/m2 and paclitaxel 200 mg/m2
q 21 days for 3 cycles
surgery
weekly paclitaxel (80 mg/m2) x 12
RT-PCR (384 genes, including 21 genes of RS)
formalin-fixed paraffin-embedded core
biopsy (95/102 pts)
?
Correlation with pCR
Validation series in a separate cohort of 82
patients treated at M.D. Anderson Cancer Center
receiving a similar chemotherapy
defined as the absence of invasive cancer in the
breast
28
(No Transcript)
29
ER-negative tumors were more likely to have a
pCR 7/31 (23) were ER-negative 4/52 (8)
were ERpositive There was no correlation between
patient age or tumor grade and pCR.
30
86 genes correlated with pCR (plt0.05)
These genes fell into categories defined by their
biologic function proliferation (eg, MCM6,
E2F1, MYBL2), apoptosis (BBC3, BAD, DR4,
TP53BP1), invasion metastasis (FYN, MMP12),
drug resistance/metabolism(ABCC5, ALDH1A1,
CYP3A4).
31
Supervised analysis
The cluster analysis of the 86 genes revealed 3
groups related to the likelihood of pCR 1.
ER-related gene cluster 2. Proliferation-relate
d gene cluster 3. Immune-related gene cluster.
32
Higher expression of the ER gene cluster
(including PR, SCUBE2, ER, NPD009, GATA3, IGF1R,
IRS1) correlated with a low likelihood of pCR.
Higher expression of the proliferation gene
cluster (including CDC20, E2F1, MYBL2, TOP2A,
FBXO5, MCM2, MCM6, CDC25B) or immune-related gene
cluster (including MCP1, CD68, CTSB, CD18, ILT-2,
CD3z, FasL, HLA.DPB1, GBP1) correlated with a
high likelihood of pCR.
33
Recurrence Score (RS) (Oncotype DX 21) and pCR
The RS was positively associated with pCR (P lt
0.005), suggesting that the pts who are at
highest recurrence risk are more likely to have
chemotherapy benefit
34
79 of the 86 genes identified as predictive in
the INT-Milan were used for correlation with pCR
in 82 patients treated in the MDA-CC
The association of gene expression with pCR was
similar in the two series
The observation that many of the discriminating
genes maintained their predictive value across
the 2 cohorts and the different gene expression
profiling platforms makes the findings compelling
enough to subject these genes to further clinical
evaluation
35
Topoisomerase 2-alpha (TOPO IIa) -marker for
anthracycline sensitivity
Microtubule-associated protein tau
(MAPT) -marker for taxane sensitivity
36
70 breast cancers
TAC
doxorubicin 50 mg/m2 cyclophosphamide 500
mg/m2 docetaxel 75 mg/m2
q 21 days for 2 cycles
Nonresponders
Responders
4 NX
4 TAC
6 TAC
4 TAC
Vinorelbine 25 mg/m2 d1, 8 q21 Xeloda 1000 mg/mq
x2 days 1-14 q21
Surgery
Surgery
Surgery
Surgery
pCR defined as no evidence of invasive tumor in
breast and node 50 biopsy specimens were
available for microarray
37
8 pCR 7 after TAC 1 after NX
38
Tumor stage, histological type, grading, and
menopausal status showed no power in predicting
pCR after CT
ER and PR negativity (determined by IHC) were
predictive of pCR
39
Gene expression evaluated by microarray analysis
HER-2 gene expression revealed to be strongly
predictive of pCR
TOPO IIa was not significant in predicting
pCR (2 different probes)
40
Gene expression evaluated by microarray analysis
MAP-Tau was not significant in predicting pCR (3
different probes)
41
Patient 1
Patient 2
Before CT
After CT
42
(No Transcript)
43
The data shown refer to the initial biopsy sample
44
(No Transcript)
45
(No Transcript)
46
Two previous studies correlating TP53 with
response to therapy
Doxorubicin study (51 pts)
FUMI study (30 pts)
(Doxorubicin weekly for 16 weeks)
(5-FU Mitomycin)
81 pts
Microarray analysis
47
Based on gene expression profile, breast cancers
is classified in 5 different subtypes 1.
Luminal A 2. Luminal B 3. ERBB2 4.
Basal-like 5. Normal-like
This classification showed prognostic effects on
DFS and OS
Thus far, limited value to predict response to
therapy
48
(No Transcript)
49
Both studies showed a lower PD for Luminal A
(2/36 vs 13/45, plt0.0089) Higher PD for Luminal B
in the doxorubicin group (5/8, plt0.0078) No other
associations between response and tumor subtype
No gene profile predicts response to either
treatment in supervised analysis
But, just clinical response!!
50
(Molecular) Imaging
51
(No Transcript)
52
(No Transcript)
53
(No Transcript)
54
(No Transcript)
55
After 6 cycles of CT, using an arbitrary cutoff
value of 60 of tumor size at baseline
Physical exam
US
Mammography
Sensitivity Specificity PPV NPV
100 31 67 100
64 43 53 55
31 56 42 45
56
FDG PET provides early information on tumor
response to CT!
57
(No Transcript)
58
(No Transcript)
59
Diffuse Optical Spectroscopy (DOS)
Provides functional informations about tumor
biochemical composition, by noninvasive
quantitative measurement of tissue concentration
of - oxygenated Hb ctO2Hb - deoxygenated
Hb ctHHb - water ctH2O - bulk lipid
1. Absorption spectra at near-infrared
absosrption (NIR) 2. Reduced scattered spectra
60
(No Transcript)
61
(No Transcript)
62
(No Transcript)
63
(No Transcript)
64

• CONCLUSION

• Tumor concentration of ctCHHb, ctO2Hb, and ctH2O
  dropped
• within one week of the neoadjuvant CT only for
  pathology
• confirmed responders

2. The best single predictor of response was
ctCHHb (83 sensistivity 100
specificity) Best combination ctCHHb
ctH20 (100 sensitivity and specificity)
3. The tumor/normal ratio of ctO2Hb was
significantly higher in responders than in
nonresponders
DOS as a new tool for predicting and monitoring
an individuals response to treatment?
65
Clinical decision making in medical oncology
Treatment selection
66
BC targets we want to direct therapy toward
Agent VEGF VEGFR PDGFR
EGFR HER-2 c-KIT MAPtau TopoII
mTOR
Anthracyclines X
X Taxanes X
Bevacizumab X Sunitinib X X
X Lapatinib X X
Trastuzumab X mTOR inhibitors
x
67
                 
68
Optimize Therapy through Pathway Evaluation(OTPE)
The expression pattern of activated signaling
molecules can be used to map pathways in biopsy
sample!
A CINBOs Project
69
CONCLUSION
In the future, it is hoped that neoadjuvant
therapy will increase our knowledge of the
molecular effects of treatment in vivo and
enable the development of robust surrogate
biological markers of response that predict
long-term outcome. This will enhance the
assessment of new treatments and enable the
development of tailored treatments for an
individual patient.