Transcript Axillary Node Surgery After Neoadjuvant Chemotherapy

NSABP
Neoadjuvant Chemotherapy
and Axillary Staging
Thomas B. Julian, MD, FACS
Professor of Surgery - Drexel University College of Medicine
Senior Surgical Director-Medical Affairs - NSABP
Director-Breast Surgical Oncology - WPAHS
Pittsburgh, PA
NC for Operable Breast Cancer
Results from First Generation RCTs
• No difference in outcome between NC and
adjuvant chemotherapy
• NC increases the rate of lumpectomy
• NC decreases the rate of axillary positivity
• Achievement of pCR correlates with
improved long-term outcome
Neoadjuvant Chemotherapy
Potential Advantages
• Increase in rate of breast-conserving surgery
• Ability to correlate tumor response to
outcome
• Potential to correlate biomarker expression
and changes in biomarkers with tumor
response and outcome
NC for Operable Breast Cancer
Continuing Clinical Rationale
• Evaluation of more effective regimens in
order to further reduce the extent of locoregional therapy:
• In the breast
• In the axilla (SNB)
• Use of primary tumor response as a guide
of further loco-regional and systemic
therapy
Neoadjuvant Anthracycline/Taxane Trials
Pathologic Complete Response (pCR)
• 4(A+C):
• 8 (CVAP)
• 6(A+C):
13-14%
15%
24%
Combo A/E + Taxane
• 4(E+TXL): 10%
• 4(A+TXT): 12%
• 6(A+TXT): 21%
Sequential A/E Taxane
• AC – TXT
26%
• 4CVAP – 4TXT: 31%
• 4ATXL – 4CMF: 23%
• 3E – 3TXL:
18%
• 4AC – 4TXT:
22%
• 12TXLw – 4FAC: 29%
• 4TXL -- 4FAC: 14%
Untch M: ASCO, 2002von Minckwitz G, et al. ASCO, 2002
Bear H: San Antonio, 2001, Gianni L: ASCO, 2002,
Evans T: ASCO 2004, Green MC: ASCO 2002
NSABP B-18
Neoadjuvant vs Adjuvant AC
Operable Breast Cancer
Stratification
• Age
• Clinical Tumor Size
• Clinical Nodal Status
Operation
AC x 4
AC x 4
Operation
•
•
•
•
Clinical Response: 79%
cCR: 36% cPR: 43%
pCR: 13%
Increase in lumpectomy
rate: 68% vs 60%
• Downstaging of (+) axillary
nodes: 58% vs 40%
• No difference in DFS and S
• Significant correlation
between pCR and outcome
NSABP B-18
Neoadjuvant vs. Adjuvant AC
Disease-Free Survival
Operable Breast Cancer
Stratification
• Age
100
80
• Clinical Tumor Size
• Clinical Nodal Status
60
40
Operation
AC x 4
20
pCR
pINV
P=0.00005
cPR
cNR
0
AC x 4
Operation
Year 0
2
4
6
8
Wolmark N: JNCI Monogr, 2001
B-18: 16-Year Update
DFS
OS
Rastogi P et al: J Clin Oncol 2008
B-18: Overall Survival by Age
≥50yrs
<50yrs
100
100
90
90
80
Qualitative
Treatment70 by Age
Interaction
60
p=0.01
80
70
60
50
40
50
N Ev HR
Post 388 167
Pre
P
40
381 139 .81 0.06
30
N Ev HR P
Post 363 148
Pre 361 171 1.23 0.07
30
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Wolmark et al: NCI State of the Science Conference 2007
NSABP B-27 Schema
Operable Breast
Cancer
(2411 pts)
Randomization
AC x 4
Tam X 5 Yrs
Surgery
AC x 4
Tam X 5 Yrs
AC x 4
Tam X 5 Yrs
Docetaxel x 4
Surgery
Surgery
Docetaxel x 4
B-27
Response in the Breast
Pathologic
Response
Clinical Response
cPR
cCR
P < 0.001
100
40%
80
%
60
40
20
0
cNR
85
%
45
%
14%
AC
(1502 pts)
NonInvasive
No Tumor
65%
91
%
30
P < 0.001
18.7%
20
26%
9%
AC Docetaxel
(687 pts)
10
0
9.8%
13.7%
3.9%
AC
(1,492 pts)
25.6%
6.9%
AC
Docetaxel
(718 pts)
Bear H, et al: JCO 2003
B-27: 8-Year Update
DFS
RFS
Rastogi P et al: J Clin Oncol 2008
B-27: 8-Year Update
Overall Survival
Wolmark et al: NCI State of the Science Conference 2007
Ki-67 Staining Can be Used to Identify
High-Risk Group Among no-pCR
Patients
Paik S: Unpublished data
NSABP B-45:
Concept Under Development
Residual Invasive Cancer in Breast or Axillary Nodes
Following a Minimum of Six Cycles of NC
Triple-Negative, Clinical Stage II or Stage III
STRATIFICATION
• Pathological nodal staging (ypNo or ypN1; ypN2;
ypN3)
• Postoperative radiation (yes; no)
Randomization
*
Arm 1
No adjuvant
chemotherapy
*
Arm 2
2
Eribulin mesylate 1.4 mg/m
IV Days 1 and 8
Every 21 days X 4 cycles
Combined Analysis of B-18/B-27
Independent Predictors of LRF
Lumpectomy + XRT
Mastectomy
(1890 Pts, 190 Events)
(1070 Pts, 128 Events)
Age
Clinical Tumor Size
(>50 years vs. <50 years)
(>5 cm vs. <5 cm)
Clinical Nodal Status
Clinical Nodal Status
(+) vs. (-)
(+) vs. (-)
Breast/Nodal Path Status Breast/Nodal Path Status
Node(-)/No pCR vs. Node()/pCR
Node(+) vs. Node(-) /pCR
Node(-)/No pCR vs. Node(-)/pCR
Node(+) vs. Node(-) /pCR
Mamounas et al: ASCO Breast 2010, Abstr. 90
10-Year Cum. Incidence of LRF
Lumpectomy Patients, >50 years
20
I B TR
R e gi ona l
n=122
15
Clin. Node (-)
Clin. Node (+)
10
n=348
n=90
5
0
n=58
n=212
n=31
10-Year Cum. Incidence of LRF
Lumpectomy Patients, <50 years
25
I B TR
20
R e gi ona l
Clin. Node (-)
0
8.7
n=223
n=376
n=135
6.9
n=57
0.5
8.3
2.4
2.3
0.7
5
Clin. Node (+)
n=84
15
10
n=154
10.5
1.8
5.3
11.4
13.6
10-Year Cum. Incidence of LRF
Mastectomy Patients, < 5 cm
20
C h e st Wa l l
15
Clin. Node (-)
R e gi ona l
Clin. Node (+)
n=46
5
4.3
2.2
0
6.4
n=37
n=183
10
n=143
2.8
n=178
8.1
10.6
2.3
7.3
4
n=21
0
2.7
10-Year Cum. Incidence of LRF
Mastectomy Patients, > 5 cm
25
C h e st Wa l l
n=128
R e gi ona l
20
4.8
Clin. Node (+)
Clin. Node (-)
n=179
15
n=95
1.7
n=33
3.2
10
0
n=16
12.3
5
0
9.2
8.6
6.2
n=11
0
0
17.6
Development of a Nomogram to
Predict LRF Following Neoadjuvant
Chemotherapy
• Results of multivariate analyses in which
age and tumor size were used as
continuous variables were similar to those
in which age and tumor size were used as
discrete variables
• The independent predictors of LRF were
then incorporated into two separate
nomograms:
– Lumpectomy + breast XRT
– Mastectomy
30
Lumpectomy + XRT
pos, Node
pos, Node
pos, Node
neg, Node
neg, Node
neg, Node
(+)
(-), No pCR
(-), pCR
(+)
(-), No pCR
(-), pCR
5
10
15
20
25
CNS
CNS
CNS
CNS
CNS
CNS
0
10-Year Probability of LRF
10-year probability (%) of being Local Failure Free
Nomogram for Prediction of
10-Year Rate of LRF After NC
40
45
50
55
60
Age
atatEntry
(Years)
Age
Entry (Years)
65
70
Nomogram for Prediction of
10-Year Rate of LRF After NC
30
pos, Node
pos, Node
pos, Node
neg, Node
neg, Node
neg, Node
(+)
(-), No pCR
(-), pCR
(+)
(-), No pCR
(-), pCR
2
3
5
10
15
20
25
CNS
CNS
CNS
CNS
CNS
CNS
0
10-year probability
(%) of being Local Failure
Free
of LRF
Probability
10-Year
Mastectomy
0
1
4
ClinicalClinical
Tumor
SizeSize
at Entry
(cm)
Tumor
(cm)
5
Summary I
• In patients with operable breast cancer, the 10year cum. incidence of LRF after neoadjuvant
chemotherapy was 10-12%
• Despite worse patient characteristics in B-27,
LRF with AC in B-27 was lower than in B-18 and
there was an effect of docetaxel (about 25%
reduction)
• Overall, the ratio of local vs. regional failure is
about 3:1 but this ratio is influenced by type of
surgery and other independent predictors of
LRF
Summary II
• Independent predictors of LRF in lumpectomy +
breast XRT patients include: age, clinical nodal
status (before NC) and pathologic breast/nodal
response
• Independent predictors of LRF in mastectomy
patients include: clinical tumor size (before NC),
clinical nodal status (before NC) and pathologic
breast/nodal response
• The effect of age (in lumpectomy patients),
clinical tumor size (in mastectomy patients) and
clinical nodal status on LRF appears to diminish
with increasing pathologic response in the
breast and axillary nodes
Summary III
• These independent predictors of LRF
after neoadjuvant chemotherapy can be
incorporated to separate nomograms for
lumpectomy + breast XRT and
mastectomy patients to provide guidance
on the need for regional XRT after
lumpectomy or loco-regional XRT after
mastectomy
• Further development and validation of
these nomograms with inclusion of
treatment effect is planned
NSABP
New Directions with
Neoadjuvant Chemotherapy
• Use pCR as a correlate of chemotherapy
efficacy to test new drugs and regimens
• Utilize micro-array technology to identify
genomic profiles associated with pCR to
specific drugs or combinations
• Candidates:
• Sequential anthracycline/taxane combinations
• New targeted therapies in combination with
chemo
NSABP
New Directions with
Neoadjuvant Chemotherapy
• Use pCR as a correlate of chemotherapy
efficacy to test new drugs and regimens
• Utilize micro-array technology to identify
genomic profiles associated with pCR to
specific drugs or combinations
• Candidates:
• Sequential anthracycline/taxane combinations
• New targeted therapies in combination with
chemo
Agents Targeting the VEGF Pathway
Anti-VEGF
antibodies
VEGF
Soluble VEGF
receptors
(eg, bevacizumab)
(eg, VEGF-Trap)
Anti-VEGFR
antibodies
Endothelial cell
(eg, IMC-1121b)
P
P
P
P
VEGFR-1
P
P
P
P
VEGFR-2
Survival Proliferation Migration
ANGIOGENESIS
Small-molecule
VEGFR inhibitors
(eg, vatalanib, sunitinib,
ZD6474, AZD2171)
NSABP B-40
Tissue for
Biomarkers
Tissue for
Biomarkers
T
A A A A
C C C C
T T T T
X X X X
A A A A
C C C C
T T T T
G G G G
A A A A
C C C C
B B B
B B
T
Operable
Breast
Cancer
R
+/-
T
T
B
S
U
R
G
E
R
Y
Accrual Completed: 1205
+/B
X 10
% pCR (Breast + Nodes)
NSABP B-40
Pathologic Complete Response
(Breast and Nodes)
N=588
N=583
OR = 1.27
Chi-square test: p=0.09
% pCR (Breast + Nodes)
NSABP B-40
Pathologic Complete Response (Breast and
Nodes) for HR+ and TN Breast Cancer
50
45
40
35
30
25
20
15
10
5
0
N=239
40.6
N=349
N=232
44
W/O BEV
N=351
BEV
17.1
11.5
HR+
OR = 1.59
p=0.033
TNBC
OR = 1.15
p=0.458
Interaction p value = 0.256
Capecitabine +/- Lapatinib:
Time to Progression (Intent-to-Treat)
Lapatinib +
Capecitabine Capecitabine
% of Patients Progression Free*
100
9
0
80
No. of pts
160
161
Progressed or died* 45 (28%)
69 (43%)
Median TTP, wk
19.7
36.9
Hazard ratio (95% CI)
0.51 (0.35, 0.74)
P-value (log-rank, 1-sided)
0.00016
7
0
60
5
0
4
0
3
0
2
0
10
0
*Censors 4 patients who died due to causes
other than breast cancer
0
1
0
2
0
3
4
Time
0 (weeks) 0
5
0
6
0
7
0
Geyer et al: N Engl J Med
NSABP B-41: Neoadjuvant Study
with Lapatinib vs. Trastuzumab vs. Combo
Tissue for
Biomarkers
Tissue for
Biomarkers
AC  TH
Operable
Breast
Cancer
HER-2 neu
Positive
R
AC  TL
AC  THL
S
U
R
G
E
R
Y
Trastuzumab
for a total of 1
year
Endpoints: pCR, cardiac events, DFS, OS
NeoALTTO
Study Design
Invasive operable
HER2+ BC
T > 2 cm
(inflammatory BC
excluded)
LVEF  50%
N=450
Stratification:
• T ≤ 5 cm vs. T > 5 cm
•ER or PgR + vs.
ER & PgR –
• N 0-1 vs. N ≥ 2
•Conservative surgery
or not
lapatinib
lapatinib
paclitaxel
R
A
N
D
O
M
I
Z
E
trastuzumab
paclitaxel
lapatinib
S
U
R
G
E
R
Y
F
E
C
trastuzumab
X
3
trastuzumab
lapatinib
trastuzumab
paclitaxel
6 wks
+ 12 wks
52 weeks of anti-HER2 therapy
34 weeks
NeoALTTO
Efficacy - pCR and tpCR
L: lapatinib; T: trastuzumab; L+T: lapatinib plus trastuzumab
pCR pathologic complete response
GEPARquinto Trial
Neoadjuvant Therapy Questions in
Hormonally Sensitive BC
• Role of neoadjuvant chemotherapy in certain ERpositive/HER-2 negative breast cancers has been
questioned
• Value of down-staging with neoadjuvant endocrine
therapy has been shown
• Potential for genomic profiling to assign ER-positive
patients to neoadjuvant endocrine therapy vs.
neoadjuvant chemotherapy
• Is pCR the most appropriate endpoint for these
patients?
43
21-Gene Recurrence Score Predicts Degree of Benefit from
Chemotherapy or Hormonal Therapy
Greater chemotherapy benefit
Distant Recurrence at 10 Years
Greater hormonal therapy benefit
Dotted lines represent 95% CI
Recurrence Score
Paik S, et al. N Engl J Med. 2004;351:2817. Paik S, et al. J Clin Oncol. 2006;24:3726.
Gianni L, et al. J Clin Oncol. 2005;23(29):7265-7277. Chang JC, et al. Breast Cancer Res Treat. 2008;108 (2):233-240.
Akashi-Tanaka S et al. Breast. 2009 Jun:171-174
Neoadjuvant Trial Proposal for HR+ BC
HR+/Her-2 Neg. Breast Cancer
Needing Neo Rx to Achieve BCT
Core BX for 21-Gene
RS
< 11
11-25
> 25
Neoadjuvant
Hormonal Tx
Randomize
Neoadjuvant
Chemotherapy
Neoadjuvant
Hormonal Tx
Neoadjuvant
Chemotherapy
SURGERY
Endpoints: Clinical Response, BCT, RCB, pCR
Axillary Node Surgery After Neoadjuvant
Chemotherapy
• Assuming clinical nodal exam, MRI
and/or axillary US pre-treatment are
negative, what axillary node staging
recommended?
• Axillary node dissection?
• Sentinel node biopsy?
• Before NCT?
• After NCT?
What if ?-MRI Prior to Neoadjuvant Therapy
Axillary nodes
largest 2.1 x 2.3 x 3.5 cm
Positive for cancer on
US-guided biopsy
Role for Sentinel Lymph Node Biopsy
in Patients Receiving NCT?
• NCCN and ASCO guidelines:
• NCT is a contraindication to use of SLN
biopsy
• Many advocate SLN biopsy prior to
NCT
• What are potential advantages and
pitfalls to SLN biopsy AFTER NCT?
Axillary Node Down-Staging with NCT
More
50 than 40% of initially node-positive
women could potentially avoid ALND!
%
40
Conversion
From
30
Node (+)
20
To
Node (-)
10
43
37
30
19
0
AC
NSABP B-18
FEC
EORTC
AT→CMF
ECTO
AC→TXT
NSABP B-27*
*Assuming 30% nodal down-staging with neoadjuvant AC
Arguments Against SLN Biopsy After NCT
1. SLN mapping after NCT may fail
and/or may not be accurate (i.e,
high false negative rate)
(Failure to map is really NOT an issue – if
can’t find SLN, then do ALND, which
is what SLN opponents recommend in
the first place)
Pathologic Status of Sentinel Nodes and
Non-Sentinel Nodes (N=343) - NSABP B-27
Identification Rate: 85% (343/428)
Sentinel
Node(s)
Positive
125 pts
Non-Sentinel
Node(s)
Negative
70 pts
Positive
55 pts
Negative
218 pts
Negative
203 pts
Positive
15 pts
False negative rate = 10.7% (15/140)
No significant difference between clinically
node negative (12.4%) vs. node positive (7.0%)
Mamounas et al JCO, 2005
Comparison of False Negative Rates
Between SN Multicenter Studies
Study
Multicenter SB-2 Trial
Italian Randomized Trial
Ann Arundel
University of Louisville
NSABP B-32 Randomized Trial
NSABP B-27 (After NC)
Meta-Analysis (Xing, 2006)
Meta-Analysis (Kelly, 2009)
FNR
11%
9%
13%
7%
10%
11%
12%
8%
(SN-/N+)
(13/114)
(8/91)
(25/193)
(24/333)
(75/766)
(15/140)
(65/540)
(~64/758)
Krag DN: N Engl J Med 1998
Veronesi U: N Engl J Med 2003 McMasters KM: J Clin Oncol 2000
Mamounas EP: J Clin Oncol 2005 Tafra L: Am J Surg 2001 Xing Y:Br J Surg 2005 Julian JB: SABCS
2004. Kelly, AM: Acad Radiol, 2009
M.D. Anderson Comparison of SLN After
Chemotherapy to Primary Surgical Patients
Clinically node negative: Negative on PE and US
GROUP
Planned SLN
+ ALND
Surgery 1st
(n = 542)
Chemo 1st
(n = 84)
FALSE NEGATIVE RATES (%)
Overall
After 2000
4.1
2.7
5.9
(p = 0.39)
5.5
K. Hunt et al, Ann. Surg, 2009
M.D. Anderson Comparison of SLN After
Chemotherapy to Primary Surgical Patients
GROUP
Surgery
1st
(n = 3171)
Chemo 1st
(n=575)
Axillary Dissections (%)
Regional
Recurrence
Rates (%)
T2
T3
40.6
65.7
0.9
27.1
(p < 0.001)
45.1
(p<0.045)
1.2
K. Hunt et al, Ann.Surg,, 2009
Arguments Against SLN Biopsy After NCT
2. Loss of important prognostic
information derived from prechemo nodal pathology
(Leads to recommendation to do SLN
prior to treatment)
80
100
B-27 DFS By Nodes
(without pCR in Breast - 1881)
60
40
1-3
4-9
20
#N
0
1-3
4-9
10+
0
% Disease-Free
0
0
%
47
32
16
5
10+
2
4
Years after Randomization
6
8
SLN Biopsy After Neoadjuvant
Chemotherapy – Do We Lose Prognostic
Information?
• Post-treatment nodal status is at least
as powerful as pre-treatment nodes
• In fact, by possibly removing the only
positive nodes with SLNB prior to
treatment, we lose even more important
information
SLN Before Neoadjuvant
Chemotherapy
• Two Major Disadvantages for
patients with positive nodes at
diagnosis:
• Two separate axillary surgical
procedures
• Many women (~ 40%) will undergo
unnecessary ALND
Eradication of LN Metastases with
Chemotherapy + Trastuzumab
• 109 consecutive patients with HER2+ BC and biopsy-proven LN
metastases
• NCT + trastuzumab
• All had complete ALND at surgery
• 81 (74%) had all negative nodes
Dominici, et al., MDACC, SABCS 2009, Abstract # 1086
Caveat – Potentially High FN Rate for
Node-Positive Patients
• 69 patients with biopsy- proven
axillary LN metastases – 25% FNR
(N2 & N3 included)
• 47 patients with clinical N1-N2
disease – 30% FNR
J. Shen et al., Cancer, 2007
P. Gimbergues et al., ASO, 2008
ACOSOG Z1071 Schema
Accuracy of SLN After NCT in Node Positive
Breast Cancer
T1-4, N1-2 invasive breast cancer
(pretreatment axillary ultrasound with FNA or core
biopsy documenting axillary metastases)
REGISTER*
Patients receive neoadjuvant chemotherapy
(stratify patients by age, stage and
number of cycles and type of chemotherapy)
REGISTER*
SLN and ALND
*Patients can be registered pre or post chemotherapy
ASBD Consensus Statement on SLN in
NCT Patients - 2010
• Clinically node negative initially • SLN acceptable before or after NCT
• Factor in impact of initial node status on
treatment
• Clinically node positive initially –
• Pathologic confirmation recommended
• Z1071 results pending
• ALND “standard of care” at definitive
surgery
Summary
• In pts with operable BC, NC results in equivalent
outcomes to those achieved with adjuvant chemotherapy
but has several potential advantages
• Information on outcome based on response to NC can be
obtained on an individualized basis
• Loco-regional therapy can be tailored based on tumor
response in the breast and axillary nodes
• This approach holds great promise as NC regimens with
targeted biologics become considerably more effective
and as genomic and imaging technology allows for more
accurate prediction and identification of pathologic
complete responders