Transcript Breast.Karlsson.1009 - Lebanese Society of Medical Oncology

Discordance in hormone receptor status in breast
cancer during tumour progression
Eva Karlsson (1,2) Linda Lindström (1) Ulla Wilking (1) Lambert Skoog (3)
Ulla Johansson (4) Jonas Bergh (1,5)
1. Department of Oncology and Pathology Cancer Center Karolinska
Karolinska Institutet ,Stockholm, Sweden,
2. Department of Oncology Karlstad Hospital,Sweden
3. Department of Pathology/Cytology Karolinska University Hospital
4. Center of Oncology Stockholm
5. Dept of Medical Oncology,The Christie,Manchester University/Paterson Institute, UK
Discordance in hormone receptor status in breast cancer during
tumour progression
Background
•
Today therapy management of breast cancer relapse is almost always
based of primary tumour characteristics such as ER / PR and HER2 status,
sites of relapse and relapse free survival time
•
Diagnosis of breast cancer relapse is therefore based on a combination of
clinical and radiological examinations
•
In rare instances image diagnosis of relapse may actually represent a
benign lesion or a new primary cancer or metastases from another
malignancy
Discordance in hormone receptor status in breast cancer during
tumour progression
Background
• Lack of stability in hormone receptors between primary breast cancer and
corresponding relapse have been reported
• Despite repeated observations, management of metastatic breast cancer patients
has essentially been unchanged and based on primary tumour characteristics
Estrogen receptor status in primary breast cancers compared with the corresponding relapses
Publication / Abstract
Patients
Discordant Patients
Author
Number
Percent
Liedtke et al, 2009 Ann Oncol
228
18,4
Broom et al, 2009 Anticancer Res
62
17,7
Simmons et al, 2009 Ann Oncol
25
40
Amir et al, 2008 Clin Oncol
9
55,6
Guarneri et al, 2008 The Oncologist
75
22,7
Wu et al, 2008 Clin Cancer Res
10
20
Lower et al, 2005 Breast Res Treat
200
30
Wang et al, 2004 Ai Zheng
65
35,4
Nedergaard et al, 1995 APMIS
101
20,8
Kamby et al, 1989 Br J Cancer
62
37,1
Discordance in hormone receptor status in breast cancer during
tumour progression
Aims
• Determine if hormone receptors (ER / PR) change between primary breast
cancer and recurrence
Material and method
•
1095 breast cancer patients in Stockholm who relapsed during 1997-2007
•
All available information on these individuals were requested from the
Center of Oncology Stockholm
•
Primary cancers and corresponding relapses were retrospectively
compared for hormonal receptor status
Discordance in hormone receptor status in breast cancer during
tumour progression
Material and methods
•
Hormone receptor data were manually collected from the original pathology
reports
We gave first priority to immunohistochemistry (IHC) for ER/PR, if not
available we used immunocytochemistry (ICC) from the cytology aspirates,
which routinely is performed and if not available we used biochemical
receptor determination
•
Aspiration cytology was originally invented and described at Karolinska
Hospital some 50 years ago.
The technique requires considerable manual skills and is at Karolinska only
practiced by a few very experienced cytopathologist performing the tumour
aspirations and they have developed the techniques for ICC for ER, initially
in parallel with the biochemical receptor determinations.
•
This retrospective study was approved by the Ethical committee at the
Karolinska Institutet
Discordance in hormone receptor status in breast cancer during
tumour progression
Results
•
In 486 patients ER information were available from both primary and one or
more recurrent sites resulting in 679 patients pairs
ER changed in 27% from positive in primary tumour to negative in relapse
and 8% changed from negative to positive
•
In 456 patients PR information were available from both primary and one or
more recurrent sites resulting in 630 patients pairs
PR status changed in 38% from positive in primary tumour to negative in
relapse and 5% changed from negative to positive
Overall breast cancer survival from the time of primary tumour
Overall
cancer
from the
time of primary
tumour diagnosis to
or in
diagnosis
tobreast
death
or survival
censoring
contrasting
intra-individual
ERdeath
status
censoring contrasting intra-individual ER status in primary tumour and relapse (both local and
primary
tumour
and
relapse (both local and systemic relapses included)
systemic
relapses
included)
1
0,9
Survival proportion
0,8
0,7
0,6
0,5
log rank p<0.0001
0,4
Prim(+)-Rel(+)
0,3
Prim(+)-Rel(-)
0,2
Prim(-)-Rel(+)
0,1
Prim(-)-Rel(-)
0
0
2
4
6
8
10
131
52
13
25
96
41
11
15
Years since primary tumour diagnosis
Numbers at risk
Prim(+)-Rel(+)
213
Prim(+)-Rel(-)
123
Prim(-)-Rel(+)
32
Prim(-)-Rel(-)
109
207
116
28
93
189
86
25
54
150
69
18
37
Overall breast cancer survival from the time of relapse diagnosis to death or censoring
Overall breast cancer
the time oftumour
relapse diagnosis
to death or
censoring
contrasting intra-individual
ER survival
status from
in primary
and relapse
(both
local and systemic
contrasting intra-individual ER status in primary tumour and relapse (both local and systemic
relapses included)
relapses included)
1
0,9
Survival proportion
0,8
0,7
0,6
0,5
log rank p<0.12
0,4
Prim(+)-Rel(+)
0,3
Prim(+)-Rel(-)
0,2
Prim(-)-Rel(+)
0,1
Prim(-)-Rel(-)
0
0
2
4
6
8
10
18
6
1
6
8
2
0
0
Years since relapse diagnosis
Numbers at risk
Prim(+)-Rel(+)
213
Prim(+)-Rel(-)
123
Prim(-)-Rel(+)
32
Prim(-)-Rel(-)
109
125
61
19
55
71
40
10
26
44
20
3
15
Risk of death depending on intra-individual ER status in primary
tumour and relapse
Risk of death in breast cancer patients depending on intra-individual ER status in primary tumour and relapse
Overall survival from Overall survival from
breast cancer diagnosis breast cancer relapse
to death or censoring to death or censoring
Intra-individual primary tumour and relapse
Patients
Deaths
ER status
Local and systemic relapse
Prim(+)/Rel(+)
Prim(+)/Rel(-)
Prim(-)/Rel(+)
Prim(-)/Rel(-)
Number
Overall
Adjusted*
HR (95% CI)
Adjusted*
HR (95% CI)
217
125
33
109
93
66
12
55
1.0 ref.
1.49 (1.05-2.11)
0.86 (0.43-1.69)
1.26 (0.79-2.00)
1.0 ref.
1.27 (0.90-1.81)
0.89 (0.45-1.77)
0.97 (0.61-1.55)
145
108
23
82
57
63
9
46
1.0 ref.
1.91 (1.29-2.84)
1.38 (0.64-2.97)
1.89 (1.09-3.28)
1.0 ref.
1.26 (0.84-1.89)
1.11 (0.51-2.41)
1.22 (0.70-2.12)
Systemic relapse
Prim(+)/Rel(+)
Prim(+)/Rel(-)
Prim(-)/Rel(+)
Prim(-)/Rel(-)
*Adjusted for age and calender year of diagnosis, progesterone receptor,tumour classification,tumour stage,lymph node metastasis, hormonal treatment and chemotherapy
Discordance in hormone receptor status in breast cancer during
tumour progression
Methodological issues
•
Our results are based on both biochemical receptor determinations and IHC/ICC with very similar
data for concordant and discordant use of the methods, respectively
•
The concordance between the biochemical and IHC/ ICC methods are high (ER 82% to 93%)
•
New data at Departement of Pathology/Cytology Karolinska from 683 breast cancer patients with
ER status from both IHC and biochemical methodes , and the concordance between them was
high 88% ( manuscript Mahmoud R. Khoshnoud et al)
•
The pathology laboratory at Karolinska University Hospital has continuously participated in quality
assurance programmes for receptor analyses, both during the era of biochemical determinations
and the present IHC/ICC techniques and all these different techniques were run parallel for years
The usage of the different methods will therefore likely not explain our findings
ER*
Status
Concordant techniques Prim-Met
Prim(+)/Met(-)
Prim(-)/Met(+)
Discordant techniques Prim-Met
Prim(+)/Met(-)
Prim(-)/Met(+)
Number
Percent
72
17
26,8
6,3
114
34
28,2
8,4
*Cut-off value of 0.05 fmol/µg DNA and 10%, for monoclonal antibody based biochemical and IHC/ICC methods, respectively
Discordance in hormone receptor status in breast cancer during
tumour progression
Potential shortcomings
•
This study is retrospective with retrospectively collected information on
hormonal receptor status
•
Hormone receptor data were manually collected from the original pathology
reports
We gave first priority to immunohistochemistry (IHC ) for ER/PR, if not
available we used immunocytochemistry (ICC) from the cytology aspirates,
which routinely is performed and if not available we used biochemical
receptor determination
•
On one hand as for all immunohistochemical / immunocytochemical
techniques, they may be falsely negative and positive on the other hand not
systematically in one direction
Discordance in hormone receptor status in breast cancer during
tumour progression
Conclusions
•
Biopsy of a suspected ”metastatic” breast cancer lesion will improve the diagnostic
precision, for single patients offer alternative/better therapies and even occasionally
exclude recurrent breast cancer
•
Nearly every third patient with breast cancer change hormone receptor status during
tumour progression
•
Increased risk of dying were seen in patients loosing ER during tumour progression
compared with stable ER positive patients
•
Therapy management of metastatic disease is suboptimal when only based on
primary tumour characteristics
Our data together with multiple small and retrospective data even
from a prospective study (Simmons et al 2009 Ann Oncol) underlines
the needs for practice change
Discordance in hormone receptor status in breast cancer during
tumour progression
“Put simply, failure to biopsy recurrent or metastatic breast cancer carries a
significant risk that our management is inadequately informed and may be
inappropriate”
(Sharma et al Nature Reviews Clinical Oncology 2010)
Acknowledgement
All the patients
The clinical colleagues at Karolinska who had high ambitions in aiming at
securing the “correct” diagnosis of the breast cancer patients at time of
clinically and radiologically (suspected) relapse
Anders Höög (Karolinska), Anna-Lena Borg (CCK), Torsten Hägerström
(CCK) Mikael Bergenheim (Karlstad Hospital) Sten-Åke Lindahl (Karlstad
Hospital) Bo Nordenskjöld (Linköping University Hospital) Elisabet Lidbrink
(Karolinska University Hospital)
The Swedish Breast Cancer Association (BRO), FOU Värmland
Jonas Bergh’s research group is supported by grants from the Swedish
Cancer Society, the Stockholm Cancer Society, the King Gustav V Jubilee
Fund, the Swedish Research Council, the Stockholm City Council,
Karolinska Institutet and Stockholm County Council Research Strategy
Committee, The Swedish Breast Cancer Association (BRO), the Karolinska
Institutet Research Funds, Manchester University, Christie Hospital &
Paterson Institute
and Märit and Hans Rausing´s Initiative against Breast Cancer
Discordance in hormone receptor status in breast cancer during
tumour progression
The reported switch of hormonal receptors may partly represent
•
•
•
•
•
tumour progression and selection
influence from the microenvironment
intratumour heterogeneity
influence by given therapy
methodological shortcomings regarding
receptor determination
Discordance in hormone receptor status in breast cancer during
tumour progression
Relapse sites stratified on relapse ER status
Site of
relapse
All sites
ICD-9
Local relapse
1741,1742
Lymph node
1960-1969
Lung
1970
Pleura
1972
Abdomen
1976,1979
Liver
1977
Skin
1982
CNS
1983,1984
Skeleton
1985
Other
1951,1971,1981,1989
Unknown
ER*
PR*
ER negative
ER positive
Total
PR negative
PR positive
Total
Number Percent Number Percent Number Number Percent Number Percent Number
407
48.5
432
51.5
839
596
73,2
218
26,8
814
95
92
16
36
10
55
20
13
64
6
0
23.3
22.6
3.9
8.9
2.5
13.5
4.9
3.2
15.7
1.5
0
150
97
9
31
5
30
23
3
65
17
2
34.7
22.5
2.1
7.2
1.2
6.9
5.3
0.7
15.1
3.9
0.4
245
189
25
67
15
85
43
16
129
23
2
146
137
20
53
15
68
32
12
95
17
1
24,5
23
3,4
8,9
2,5
11,4
5,4
2
15,9
2,8
0,2
*Cut-off value of 0.05 fmol/µg DNA and 10%, for monoclonal antibody based biochemical and IHC/ICC methods, respectively
92
43
1
15
2
13
7
2
35
7
1
42,2
19,7
0,5
6,9
0,9
5,9
3,2
0,9
16,1
3,2
0,5
238
180
21
68
17
81
39
14
130
24
2
Discordance in hormone receptor status in breast cancer during
tumour progression
ER status in tumour relapse
Hormonal receptor
Status
ER status*
Prim(+)/Rel(+)
Prim(+)/Rel(-)
Prim(-)/Rel(+)
Prim(-)/Rel(-)
Total number
Local and systemic relapse
Number
Percent
213
123
32
109
477
44,65
25,79
6,71
22,85
100
Systemic relapse
Number
Percent
136
105
22
82
345
39,42
30,43
6,38
23,77
100
*Cut-off value of 0.05 fmol/µg DNA and 10%, for monoclonal antibody based biochemical and IHC/ICC methods, respectively