Anatomic Pathology / HER2 Heterogeneous Amplification
Heterogeneous HER2 Gene Amplification
Impact on Patient Outcome and a Clinically Relevant Definition
Alastair I. Bartlett,1 Jane Starcyznski, PhD,2 Tammy Robson, MSc,1 Alex MacLellan,3
Fiona M. Campbell, MSc,1 Cornelis J. H. van de Velde, MD,4 Annette Hasenburg, MD,5
Christos Markopoulos, MD,6 Caroline Seynaeve, MD,7 Daniel Rea, PhD,8
and John M. S. Bartlett, FRCPath1
Key Words: HER2; Amplification; Breast; Heterogeneity
DOI: 10.1309/AJCP0EN6AQMWETZZ
Abstract
Heterogeneous expression or amplification is a
challenge to HER2 diagnostics. A guideline defines
heterogeneity as the presence of between 5% and 50%
cells with HER2/CEP17 ratios of more than 2.20. We
audited the frequency of such cells and their clinical
impact in the results from 2 pathology laboratories
combined with data from the TEAM [Tamoxifen vs
Exemestane Adjuvant Multicentre] pathology study.
HER2 reports were scanned and the percentages of
amplified cells reported.
Of 6,461 eligible cases, 754 (11.7%) exhibited
50% or more cells with ratios of more than 2.20, which
is “amplified” by College of American Pathologists
guidelines. Of the cases, 2,166 (33.5%) exhibited
more than 5% but less than 50% of cells with HER2/
CEP17 ratios of more than 2.20, or “heterogeneous
amplification.” No prognostic impact was observed
when fewer than 30% of cells exhibited ratios of more
than 2.20. All amplified cases with 30% to 50% of cells
with ratios more than 2.20 were identified as such by
United Kingdom guidelines.
The percentage of tumor cells with HER2/
CEP17 ratios more than 2.20 does not identify cases
with heterogeneous amplification or poor outcome.
A modified approach for identification of true
heterogeneous amplification is suggested.
266
266
Am J Clin Pathol 2011;136:266-274
DOI: 10.1309/AJCP0EN6AQMWETZZ
Guidelines for breast cancer management indicate the
testing of all patients for HER2 status at initial diagnosis or at
recurrence. National and international guidelines and external
quality assurance schemes aim to ensure accurate and robust
diagnostic testing of HER2 expression and amplification to
support treatment decisions.1-4 Establishing HER2 status is
essential for predicting response to trastuzumab (Herceptin,
Genentech/Roche, Basel, Switzerland), lapatinib (Tykerb,
GlaxoSmithKline, Brentford, England), and other HER2targeted agents5,6 and may also inform treatment choice.7
A number of external quality assurance schemes exist to
monitor laboratory performance.8,9 However, discussion about
unusual patterns of HER2 gene amplification is increasingly
highlighting the need for additional information in cases in
which interpretation of HER2 positivity is not straightforward.
Cases with low-level or heterogeneous amplification or
coamplification of the centromeric marker present challenges
for diagnosticians and require an evidence-based approach
to support the final determination of HER2 status.10,11 These
challenges are, however, rarely addressed in guidelines, and,
despite the importance of robust evidence to support the
definition, reporting, and clinical importance of such unusual
cases, few studies address such issues.
Heterogeneous amplification, defined as the existence
of 2 distinct or intermixed clones of breast cancer cells
exhibiting different patterns of gene amplification (usually
1 clone amplified and 1 normal) is considered a significant
challenge to HER2 diagnostics,2 but at present no clear
definition is available.
By United Kingdom (UK) guidelines, HER2 heterogeneity
is assessed by scanning the entire tumor section before selecting
at least 3 separate tumor fields and counting the number of
© American Society for Clinical Pathology
Anatomic Pathology / Original Article
chromosome 17 (CEP17) and HER2 signals for up to 20
cells per field.1 However, a clear definition of heterogeneity
and evidence-based data regarding the frequency and clinical
relevance and significance of this finding are lacking. A recent
panel guideline, by the College of American Pathologists
(CAP), redefined heterogeneous amplification of HER2 as
the presence of between 5% and fewer than 50% of cells
with an HER2/CEP17 ratio more than 2.20. According to this
guideline, the presence of a single cell with an HER2/CEP17
ratio of more than 2.20, when 20 cells are counted, would lead
to a clinical diagnosis of heterogeneous amplification.12 We
are unaware of data resulting from auditing the frequency of
cells with this ratio in fluorescence in situ hybridization (FISH)
analyses or the clinical impact of this definition. Without such
data, it is difficult for clinicians to interpret the clinical impact
of a breast cancer with 5% HER2-amplified cells in the context
of therapeutic decision making.
We therefore performed an extensive audit of more
than 6,000 cases to explore the frequency of heterogeneous
amplification and the impact of this finding on patient
outcome.
Materials and Methods
Cases and Methods
An audit of FISH results, reported during 2009, from
the Birmingham Heartlands Hospital (BHH, Birmingham,
England; cohort 1) routine pathology laboratory was
combined with the results of FISH tests performed at Glasgow
Royal Infirmary (Glasgow, Scotland [GRI]) between 2000
and 2007 (cohort 2) and within the TEAM [Tamoxifen
vs Exemestane Adjuvant Multicentre] pathology study
(TPS; cohort 3). All FISH reports were reviewed, and the
percentage of amplified cells, as defined by CAP guidelines
(HER2/CEP17 ratio >2.20), was reported for all cases with
at least 20 cells scored according to UK guidelines. (A small
proportion of TEAM cases, assessed on tissue microarrays),
with 10 to 19 cells counted were also included for the
evaluation of prognosis).1,12
These cohorts were selected because cohort 1 (BHH)
reflects current UK practice in which FISH is performed
only on equivocal cases (immunohistochemical score, 2+),
whereas cohort 2 (GRI) included a significant series of
immunohistochemically negative (score, 0/1+) and positive
(score, 3+) cases. Cohort 3 (TPS) represents a clinical trial
cohort that is part of the TEAM trial13-15 and was included
to allow evaluation of the prognostic impact of increasing
percentages of “amplified cells” (HER2 heterogeneous
amplification) in a clinical trial cohort as outcome data
were available. Of note, owing to the inclusion period (mid
2001-2006) none of these patients received trastuzumab as
adjuvant therapy.
Eligible Cases
There were 1,051 eligible cases (data on at least 20 cells
per case) in cohort 1. Of 1,657 eligible cases from cohort 2,
542 were immunohistochemically negative (0/1+), 546 were
equivocal by immunohistochemical analysis (2+), and 93
were immunohistochemically positive (3+) according to UK
guidelines at the time of testing.3,4 Immunohistochemical
data for the remaining 476 cases were not available. Cohort
3 (TPS) included 3,753 eligible cases. Overall, data were
available for 6,461 eligible cases.
Samples were primarily stratified into 22 groups ❚Table
1❚ and later grouped into 12 ❚Figure 1❚ and then 6 ❚Figure 2❚
categories as follows: For ease of illustration the 22 groups
(5% increments of percentages of cells with ratio more than
2.20) were condensed into groups with 10% increments
(0%, 1%-9.9%, etc). These 12 groups were then further
condensed by combining all cases with less than 5% of cells
with ratio above 2.20 (Group A), cases with 5% to <30%
(group B), 30% to <50% (Group C), 50% to <75% (Group
D), 75% to <100% (Group E), and 100% to 110% (Group
F) of cells with ratios above 2.20. Outcome, irrespective of
adjuvant treatment (ie, aromatase inhibitor vs tamoxifen),
was estimated for each group.
Results
Data on HER2 Status
Cohort 1
All cases were referred owing to equivocal
immunohistochemical results (all immunohistochemically
2+). Of 1,051 eligible cases, 188 (17.9%) were HER2
amplified as defined by current UK guidelines (HER2/
CEP17 ratio ≥2.00).1 Of these cases, 32 would be regarded
as “borderline” by American Society of Clinical Oncology
(ASCO)/CAP guidelines.2
Analysis of Cases by CAP Guidelines.—Of the cases, 148
(14.1%) exhibited 50% or more cells with ratios more than
2.20 and would be considered “amplified” under CAP panel
guidelines ❚Table 2❚; 50 (33.8%) of the 148 cases exhibited
30% or more “nonamplified” cells (ratio, ≤2.20). Of the 1,051
cases, 381 (36.3%) exhibited between 5% and fewer than 50%
of cells with HER2/CEP17 ratios more than 2.20, regarded
under new CAP guidelines as exhibiting heterogeneous
amplification, of which 40 were diagnosed as amplified
by the central laboratory (BHH). Only 55 cases exhibited
between 30% and 50% amplified cells, and 37 of these were
considered amplified by the central laboratory.
© American Society for Clinical Pathology
Am J Clin Pathol 2011;136:266-274
267
DOI: 10.1309/AJCP0EN6AQMWETZZ
267
267
Bartlett et al / HER2 Heterogeneous Amplification
❚Table 1❚
Distribution of Cells With HER2/CEP17 Ratios More Than 2.20 in Audited Cases*
IHC Score
% Amplified Cells
0/1+ (123/3,840
[3.2%])
2+ (392/1,725
[22.7%])
3+ (315/345
[91.3%])
No IHC Testing
(80/551 [14.5%])
Total (910/6,461
[14.1%])
0
>0-<5
5-<10
10-<15
15-<20
20-<25
25-<30
30-<35
35-<40
40-<45
45-<50
50-<55
55-<60
60-<65
65-<70
70-<75
75-<80
80-<85
85-<90
90-<95
95-<100
100
0/1,910
0/641
0/616
0/284
0/132
2/57
3/43
9/34
10/19
12/16
11/12
12/12
8/8
9/9
7/7
5/5
4/4
9/9
2/2
6/6
2/2
12/12
0/377
0/325
0/257
0/134
2/92
3/76
0/39
9/38
19/23
26/29
25/27
25/25
23/23
25/25
23/23
17/17
19/19
25/25
20/20
21/21
32/32
78/78
0/8
0/3
0/6
0/7
0/0
0/1
0/2
0/2
0/1
4/4
6/6
3/3
9/9
6/6
5/5
17/17
13/13
12/12
23/23
36/36
61/61
120/120
0/183
0/94
0/84
0/33
0/35
0/17
2/14
1/7
2/7
6/8
4/4
6/6
3/3
3/3
6/6
3/3
4/4
5/5
1/1
4/4
9/9
21/21
0/2,478
0/1,063
0/963
0/458
2/259
5/151
5/98
19/81
31/50
48/57
46/49
46/46
43/43
43/43
41/41
42/42
40/40
51/51
46/46
67/67
104/104
231/231
1.0
1.0
0.9
0.9
Disease-Free Survival
Disease-Free Survival
IHC, immunohistochemical.
* Classification of IHC staining was as follows: 0/1+, weak; 2+, equivocal; 3+, positive by United Kingdom (UK) guidelines. % Amplified is the percentage of amplified cases
using UK guidelines, as follows: 0, cases with no cells exhibiting HER2/CEP17 ratio >2.20; >0-<5, cases with <5% cells exhibiting HER2/CEP17 ratio >2.20; 10-<15, cases
with 10 to <15.0% cells exhibiting HER2/CEP17 ratio, etc. Values in individual cells indicate the number of cases amplified using mean HER2/CEP17 ratios/total number
of cases for the category.
0.8
0.7
0.7
0.6
0.6
0
1
3
2
4
5
Years
❚Figure 1❚ Outcome for patients within the TEAM (Tamoxifen
vs Exemestane Adjuvant Multicentre Trial) cohort only
stratified by percentage of cells with HER2/CEP17 ratios
>2.20, as follows: group 0 (solid black line), 0%; group 1 (solid
blue line), 1%-9.9%; group 2 (solid green line), 10%-19.9%;
group 3 (yellow line), 20%-29.9%; group 4 (solid red line),
30%-39.9%; group 5 (orange line), 40%-49.9%; group 6
(turquoise line), 50%-59.9%; group 7 (dotted black line),
60%-69.9%; group 8 (dotted red line), 70%-79.9%; group 9
(dotted blue line), 80%-89.9%; group 10 (dotted green line),
90%-99.9%; and group 11 (mauve line), 100%.
268
268
0.8
Am J Clin Pathol 2011;136:266-274
DOI: 10.1309/AJCP0EN6AQMWETZZ
0
1
3
2
4
5
Years
❚Figure 2❚ Outcome for patients stratified by percentage
of tumor cells with HER2/CEP17 ratios >2.20, in simplified
groupings, as follows: group A (solid black line), <5% (n =
2,481); group B (green line), 5%-29.99% (n = 1,078); group
C (blue line), 30%-49.99% (n = 97); group D (magenta line),
50%-74.9% (n = 91); group E (red line), 75%-99.9% (n =
159); and group F (dotted black line), 100% (n = 110).
© American Society for Clinical Pathology
Anatomic Pathology / Original Article
❚Table 2❚
Cohort 1: BHH Samples*
Immunohistochemical Score
% Amplified 0/1+ (0/1,051
Cells
[0%])
2+ (188/1,051
[17.9%])
3+ (0/1,051
[0%])
Total
(188/1,051
[17.9%])
0
>0-<5
5-<10
10-<15
15-<20
20-<25
25-<30
30-<35
35-<40
40-<45
45-<50
50-<55
55-<60
60-<65
65-<70
70-<75
75-<80
80-<85
85-<90
90-<95
95-<100
100
0/296
0/226
0/152
0/69
2/46
1/33
0/26
5/19
9/11
10/10
13/15
15/15
9/9
15/15
11/11
8/8
3/3
7/7
6/6
8/8
12/12
54/54
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
0/296
0/226
0/152
0/69
2/46
1/33
0/26
5/19
9/11
10/10
13/15
15/15
9/9
15/15
11/11
8/8
3/3
7/7
6/6
8/8
12/12
54/54
*
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
For cohort 1, from Birmingham Heartlands Hospital (BHH), Birmingham, England.
Classification of immunohistochemical staining was as follows: 0/1+, weak; 2+,
equivocal; 3+, positive by United Kingdom (UK) guidelines. % Amplified is the
percentage of amplified cases using UK guidelines, as follows: 0, cases with no
cells exhibiting HER2/CEP17 ratio >2.20; >0-<5, cases with <5% cells exhibiting
HER2/CEP17 ratio >2.20; 10-<15, cases with 10 to <15.0% cells exhibiting HER2/
CEP17 ratio, etc. Values in individual cells indicate the number of cases amplified
using mean HER2/CEP17 ratios/total number of cases for the category.
Cohort 2
Of 1,657 eligible cases, 325 (19.6%) were classified by
the central laboratory (GRI) as amplified; 37 of these were
borderline amplified by ASCO/CAP guidelines. Of the cases,
263 were amplified under the new CAP panel guidelines, of
which 54 exhibited 30% or more cells with a ratio 2.20 or less
❚Table 3❚. Another 661 cases exhibited 5% to fewer than 50%
of cells with HER2/CEP17 ratios more than 2.20, exhibiting
heterogeneous amplification by CAP guidelines, of which
only 62 were regarded as amplified by the central laboratory.
Only 91 cases exhibited between 30% and 50% amplified
cells, and 58 of these cases were considered amplified by the
central laboratory.
Cohort 3
Of 3,753 eligible cases, all tested in a central research
laboratory, 397 (10.6%) were classified as amplified by UK
guidelines, of which 34 were defined as borderline by ASCO/
CAP guidelines. All were positive for estrogen receptor,
explaining the low frequency of HER2 amplification ❚Table
4❚ in this cohort. Of the cases, 343 exhibited 50% or more
cells with ratios more than 2.20, satisfying the new CAP
guidelines for amplification (nonheterogeneous). Of these,
69 cases exhibited at least 30% of cells with ratios 2.20 or
less. Another 1,124 cases exhibited 5% to fewer than 50%
of cells with HER2/CEP17 ratios more than 2.20, exhibiting
heterogeneous amplification according to CAP guidelines,
of which 54 were diagnosed as amplified by the central
laboratory. Only 91 cases exhibited between 30% and fewer
than 50% amplified cells, and 49 of these 91 cases were
diagnosed as amplified by the central laboratory.
Combined Results
In total, 6,461 cases were audited, of which 1,725
were known to be 2+ by immunohistochemical analysis.
In cases with 0 or 1+ immunohistochemical staining, 3.2%
were HER2-amplified; of 2+ cases, 22.7% exhibited HER2
amplification; and 91.3% of immunohistochemically 3+
cases exhibited HER2 amplification. Overall, 14.1% of cases
were HER2-amplified. Cohorts 1 and 2 represented routine
laboratory experience and were therefore markedly enriched
for 2+ cases (Table 1).
Frequency of Heterogeneous Amplification in
Immunohistochemically 2+ Cases
Of 1,725 immunohistochemically stained 2+ cases, 392
(22.7%) exhibited mean HER2/CEP17 ratios 2.0 or more,
and 308 (17.9%) were amplified according to UK and CAP
guidelines (>50% of cells with ratios >2.20 or mean ratio
≥2.0). Another 715 cases (41.4%) had between 5% and 50%
of cells with ratios more than 2.20. Of these, only 84 (11.7%)
were classified as amplified (ratio, ≥2.0) by UK guidelines.
Only 117 cases exhibited between 30% and fewer than 50%
amplified cells, and 79 of the 117 cases were amplified
according to the original central laboratory diagnosis (Table 1).
Overall Frequency of Heterogeneous Amplification
Of the 6,461 cases audited, 910 (14.1%) were classified
as amplified, of which 103 would be defined as borderline by
ASCO/CAP guidelines but amplified by UK guidelines (ratios
2.00 to <2.20). Of the cases, 754 exhibited 50% or more cells
with ratios more than 2.20, satisfying the CAP guidelines for
amplification (nonheterogeneous). Of these, 173 exhibited
at least 30% of cells with ratios 2.20 or less. Another 2,166
cases exhibited between 5% and fewer than 50% of cells
with an HER2/CEP17 ratio more than 2.20 and exhibited
heterogeneous amplification, of which only 156 were assessed
as amplified at diagnosis by central laboratories. Only 237
cases exhibited between 30% and fewer than 50% amplified
cells, and 144 of these cases were considered amplified at
original diagnosis by the central laboratories.
Impact of Heterogeneous Amplification on Outcome
No significant differences in 5-year survival rates were
observed in patients with up to 30% of tumor cells with ratios
© American Society for Clinical Pathology
Am J Clin Pathol 2011;136:266-274
269
DOI: 10.1309/AJCP0EN6AQMWETZZ
269
269
Bartlett et al / HER2 Heterogeneous Amplification
❚Table 3❚
Cohort 2: GRI Samples*
IHC Score
% Amplified Cells
0/1+ (18/542
[3.3%])
2+ (150/546
[27.5%])
3+ (84/93
[90%])
No IHC Testing
(73/476 [15.3%])
Total (325/1,657
[19.6%])
0
>0-<5
5-<10
10-<15
15-<20
20-<25
25-<30
30-<35
35-<40
40-<45
45-<50
50-<55
55-<60
60-<65
65-<70
70-<75
75-<80
80-<85
85-<90
90-<95
95-<100
100
0/207
0/147
0/92
0/41
0/18
1/12
0/5
2/5
4/4
1/1
3/3
1/1
0/0
0/0
2/2
1/1
0/0
0/0
0/0
1/1
0/0
2/2
0/66
0/81
0/97
0/48
0/37
1/40
0/11
4/18
8/10
13/14
8/8
6/6
9/9
8/8
9/9
8/8
11/11
14/14
10/10
9/9
14/14
18/18
0/2
0/1
0/0
0/4
0/0
0/0
0/1
0/1
0/0
1/1
2/2
1/1
2/2
1/1
1/1
3/3
4/4
1/1
6/6
8/8
22/22
32/32
0/136
0/93
0/76
0/28
0/33
0/15
2/12
1/6
2/7
6/8
3/3
5/5
2/2
2/2
5/5
3/3
3/3
5/5
1/1
4/4
9/9
20/20
0/411
0/322
0/265
0/121
0/88
2/67
2/29
7/30
14/21
21/24
16/16
13/13
13/13
11/11
17/17
15/15
18/18
20/20
17/17
22/22
45/45
72/72
IHC, immunohistochemical.
* For cohort 2, from the Glasgow Royal Infirmary (GRI), Glasgow, Scotland. Classification of IHC staining was as follows: 0/1+, weak; 2+, equivocal; 3+, positive by United
Kingdom (UK) guidelines. % Amplified is the percentage of amplified cases using UK guidelines, as follows: 0, cases with no cells exhibiting HER2/CEP17 ratio >2.20;
>0-<5, cases with <5% cells exhibiting HER2/CEP17 ratio >2.20; 10-<15, cases with 10 to <15.0% cells exhibiting HER2/CEP17 ratio, etc. Values in individual cells indicate
the number of cases amplified using mean HER2/CEP17 ratios/total number of cases for the category.
❚Table 4❚
Cohort 3: TPS*
IHC Score
Group
% Amplified
Cells
0/1+ (105/3,298
[3.2%])
2+ (54/128
[42.2%])
3+ (231/252
[91.7%])
No IHC Testing
(7/75 [9%])
Total (397/3.753
[10.6%])
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
0
>0-<5
5-<10
10-<15
15-<20
20-<25
25-<30
30-<35
35-<40
40-<45
45-<50
50-<55
55-<60
60-<65
65-<70
70-<75
75-<80
80-<85
85-<90
90-<95
95-<100
100
0/1,703
0/494
0/524
0/243
0/114
1/45
3/38
7/29
6/15
11/15
8/9
11/11
8/8
9/9
5/5
4/4
4/4
9/9
2/2
5/5
2/2
10/10
0/15
0/18
0/8
0/17
0/9
1/3
0/2
0/1
2/2
3/5
4/4
4/4
5/5
2/2
3/3
1/1
5/5
4/4
4/4
4/4
6/6
6/6
0/6
0/2
0/6
0/3
0/0
0/1
0/1
0/1
0/1
3/3
4/4
2/2
7/7
5/5
4/4
14/14
9/9
11/11
17/17
28/28
39/39
88/88
0/47
0/1
0/8
0/5
0/2
0/2
0/2
0/1
0/0
0/0
1/1
1/1
1/1
1/1
1/1
0/0
1/1
0/0
0/0
0/0
0/0
1/1
0/1,771
0/515
0/546
0/268
0/125
2/51
3/43
7/32
8/18
17/23
17/18
18/18
21/21
17/17
13/13
19/19
19/19
24/24
23/23
37/37
47/47
105/105
IHC, immunohistochemical.
* For cohort 3, from the TEAM (Tamoxifen vs Exemestane Adjuvant Multicentre Trial13-15) pathology study (TPS). Classification of IHC staining was as follows: 0/1+, weak; 2+,
equivocal; 3+, positive by United Kingdom (UK) guidelines. % Amplified is the percentage of amplified cases using UK guidelines, as follows: 0, cases with no cells exhibiting
HER2/CEP17 ratio >2.20; >0-<5, cases with <5% cells exhibiting HER2/CEP17 ratio >2.20; 10-<15, cases with 10 to <15.0% cells exhibiting HER2/CEP17 ratio, etc. Values
in individual cells indicate the number of cases amplified using mean HER2/CEP17 ratios/total number of cases for the category.
270
270
Am J Clin Pathol 2011;136:266-274
DOI: 10.1309/AJCP0EN6AQMWETZZ
© American Society for Clinical Pathology
Anatomic Pathology / Original Article
more than 2.20 (groups 2-4) (Figure 1) when compared with
the 1,966 cases without detectable amplified cells (group
1, Figure 1). Reduced 5-year, recurrence-free survival was
observed for patients whose tumors exhibited 30% to 100% of
cells with ratios more than 2.20; however, samples for some
groups were fewer than 20 cases. The outcome for the 6 groups
is shown (Figure 2), in which group A included tumors with
fewer than 5% cells with ratios more than 2.20 (n = 2,217);
group B, tumors with 5% to fewer than 30% cells with ratios
more than 2.20 (n = 1,003); group C, tumors with between
30% and fewer than 50% cells with ratios more than 2.20 (n =
90); group D, tumors with 50% to fewer than 75% of cells with
ratios more than 2.20 (n = 84); group E, tumors with 75% to
99.9% of cells with ratios more than 2.20 (n = 148); and group
F, tumors with 100% amplified cells (n = 102). As shown in
Figure 2, patients with tumors exhibiting few (<5%) or 5%
to fewer than 30% had similar outcome, whereas outcome
progressively worsened with tumors containing more than
30% of cells with HER2/CEP17 ratios more than 2.20.
Heterogeneous Amplification in Cases With 30% to 50%
Tumor Cells With HER2/CEP17 Ratios More Than 2.20
Of 91 cases with between 30% and fewer than 50%
amplified cells, for which outcome data were available in 90
cases (TPS), 49 were recorded as amplified by UK guidelines
(HER2/CEP17 ratios, 2.00-9.39). Of the remaining 41 cases
with mean HER2/CEP17 ratios less than 2.00, 21 were
identified when originally analyzed by the central laboratory
as exhibiting heterogeneous amplification according to UK
guidelines; these cases exhibited 2 discrete clones ❚Image
1A❚ or intermixed amplified and nonamplified cells ❚Image
1B❚. The remaining 20 cases were classified as nonamplified
(ie, not exhibiting heterogeneous or homogenous HER2
amplification by UK guidelines), despite a significant
proportion of cells with ratios more than 2.20 ❚Image 1C❚
and ❚Image 1D❚. Outcome data from these cases ❚Figure 3❚
suggest that the presence of an identifiable amplified HER2
clone, as defined by UK guidelines, is required for a negative
prognostic impact; however, this result was not significant at
the P = .05 level.
Discussion
Current guidelines for HER2 testing do not fully address
the reporting of complex cases, in particular cases with
true heterogeneous amplification in which 2 tumor clones
with different HER2 status are present1,2,12,16,17 (Images 1A
and 1B). Data on the clinical importance of heterogeneous
amplification are also lacking. Following the recent CAP
guideline12 suggesting that all cases with between 5% and
fewer than 50% cells with HER2/CEP17 ratios more than
2.20 be regarded as heterogeneously amplified, we performed
an audit of more than 6,000 cases to assess the impact of
this definition on routine clinical practice. Use of the CAP
guidelines identified almost 45% of 6,461 breast cancers as
exhibiting HER2 amplification or heterogeneous amplification
in a population in which the mean HER2/CEP17 ratio
according to UK guidelines identified 14.1% of cases as
amplified. Therefore, the CAP guideline more than tripled
the number of cases considered to exhibit abnormalities in the
HER2 oncogene.
By using the pathology substudy cohort originating
from the TEAM trial,15,18 we investigated the impact of an
increasing frequency of amplified cells on the outcome of
early postmenopausal breast cancer. No evidence of a negative
impact on prognosis was observed in patients with between 0%
and fewer than 30% of tumor cells exhibiting HER2/CEP17
ratios more than 2.20. Only in patients with more than 30%
of cells with ratios more than 2.20 was decreased disease-free
survival (DFS) observed relative to patients with no abnormal
tumor cells (based on HER2/CEP17 ratios; Figures 1 and 2).
All cases with more than 50% of abnormal tumor cells, using
UK and CAP guidelines,1,12 were diagnosed as exhibiting
HER2 gene amplification, even without consideration of
heterogeneous amplification. The patients in all such cases
would be considered for targeted HER2 therapies. Within the
TEAM cohort, which was the only cohort for which outcome
data were available, only a small subset of cases exhibited
evidence of true heterogeneous amplification that influenced
the prognosis (Figure 3).
The 2008 UK guidelines for HER2 testing1 included
this guidance for the detection of heterogeneous HER2
amplification: “The number of chromosome 17 (Ch17) and
HER2 signals are counted in 20 to 60 non-overlapping invasive
cancer cell nuclei, using at least three distinct tumor fields or
more if there is heterogeneity.” This definition, like the CAP
definition, is based on expert opinion rather than evidence. We
applied the UK definition, before the commencement of the
audit, to the TPS cases. Of the 91 TEAM cases with between
30% and 50% of abnormal tumor cells and not regarded as
amplified based on the mean HER2/CEP17 ratio, 1 was lost
to follow-up, 21 were defined as exhibiting heterogeneous
amplification using the UK guidelines, 49 were regarded as
amplified by both methods (of which 28 were heterogeneously
amplified by UK guidelines), and 20 were regarded as not
amplified by the mean HER2/CEP17 ratio and did not exhibit
heterogeneous amplification as defined by the UK guidelines.
In an exploratory analysis, patients with tumors exhibiting
heterogeneous or homogeneous HER2 amplification exhibited
decreased DFS compared with patients with tumors without
evidence of heterogeneous amplification, even in the presence
of significant numbers of cells with ratios more than 2.20
(30% to <50%).
© American Society for Clinical Pathology
Am J Clin Pathol 2011;136:266-274
271
DOI: 10.1309/AJCP0EN6AQMWETZZ
271
271
Bartlett et al / HER2 Heterogeneous Amplification
A
B
C
D
❚Image 1❚ Heterogeneous amplification of HER2 in breast cancer. There are several patterns of heterogeneous amplification
of HER2. A, A tumor that exhibits 2 distinct tumor clones, the first (arrowheads) showing amplification, and the second
(arrows) showing deletion of HER2. B, The presence of isolated highly amplified tumor cells on a background of nonamplified
cells. C and D, Tumors with an increase in HER2 and CEP17 copy number, with a nonamplified ratio. C, About 10% of cells
show an HER2/CEP 17 ratio >2.20 (arrowheads), with some clearly amplified cells noted, which would support intermixed
heterogeneous amplification of HER2. D, Fewer than 10% of cells show an HER2/CEP 17 ratio <2.20 (arrowhead).
The findings of the current analysis raise a number of
important questions, including the following: What is the
significance of a small number of apparently amplified cells
within an otherwise normal tumor? Are these cells not, in
fact, amplified? Most important, why is the definition of
amplification applied to a cell population apparently not
applicable to individual cells? The answers to these questions
lie in a clear technical understanding of the process of ISHbased analysis of gene amplification.
A number of important factors influence this analysis
and the interpretation of results. First, analysis of gene
272
272
Am J Clin Pathol 2011;136:266-274
DOI: 10.1309/AJCP0EN6AQMWETZZ
amplification is performed on thin tissue sections in situ
and, therefore, involves the analysis of incomplete nuclei.
Second, the analysis of neoplastic cells for gene and
chromosome copy imbalances takes place in the context of
duplication of part or all of the parent chromosome (in this
case, chromosome 17) during normal cellular proliferation
(S phase) or in aneuploid tumors. Third, hybridization
of the ISH probes is not 100% efficient. Each of these
factors affects the interpretation of ISH and can explain the
difference in ratios observed in single cells and across an
entire population.18-21
© American Society for Clinical Pathology
Anatomic Pathology / Original Article
Disease-Free Survival
1.0
0.9
0.8
0.7
0.6
0
1
3
2
4
5
Years
❚Figure 3❚ Outcome for patients with tumors exhibiting
between 30% and 50% of cells with HER2/CEP17 ratios
>2.20 stratified by original diagnosis according to United
Kingdom guidelines for identification of heterogeneous
amplification. Group A (black line), nonamplified (n = 20);
group B (red line), amplified (n = 49); group C (green line),
heterogeneous amplification (n = 21).
The cumulative effect of these technical realities is the
frequent observation of apparently amplified cells (eg, cells
with 1 CEP17 copy and 3 HER2 copies) in otherwise entirely
normal nonamplified tumors (Figure 2, Group B). In some
tumors, a proportion of cells exhibits apparent amplification
with an average HER2/CEP17 ratio less than 2.00, showing
that the tumor is, in fact, not amplified by conventional scoring
methods. In some cases, this represents a genuine intermixed
population of cells, some of which show amplification as
shown in Figure 3, representing heterogeneous amplification,
and although it is not amplified, the average HER2/CEP
17 ratio is raised. In others, there is apparent heterogeneity
with a subset of tumor cells showing a ratio more than 2.20
(<30%); however, in these cases, this most likely represents
an artifact. By using information based on the outcome data
from cohort 3, we have suggested a guideline for identifying
these different HER2 patterns.
Evidence from an extensive audit of diagnostic and
clinical trial–based materials suggests that a significant
proportion of tumors (33%) contains a proportion of cells
with HER2/CEP17 ratios more than 2.20. The prognostic
impact of this observation, in the absence of clear evidence
of an amplified clone of cells within the tumor, seems
negligible, and, therefore, the proportion of such cells within
breast cancers is not a valid definition of heterogeneous
amplification, in contrast with the suggestion of the recent
CAP guidelines. The presence of such cells is most often a
reflection of technical and biologic aspects of determination
of gene amplification by ISH in neoplastic tissues.
Based on the current data, we propose the following
method for determination of heterogeneous HER2
amplification: In all cases in which ISH is performed, the
entire slide should be scanned before counting, and areas of
apparent heterogeneity should be identified during this scan
and/or by reference to an immunohistochemically stained
slide. The number of CEP17 and HER2 signals should be
counted in 20 to 60 nonoverlapping invasive cancer cell
nuclei using at least 3 distinct tumor areas. If there is evidence
of heterogeneity between areas (or less frequently within
areas), additional cells (at least 20 per area) and/or areas (up
to 6) should be counted. The HER2/CEP17 ratio should be
calculated for each area individually. When the mean HER2/
CEP17 ratio in any area is 2.00 or more, the tumor should
be regarded as amplified. Cases containing amplified and
nonamplified areas (using this definition) should be reported
as exhibiting heterogeneous amplification. For all cases in
which the ratio is between 1.80 and 2.20, results should be
based on counting at least 60 tumor cells, and in cases in
which heterogeneity is suspected, this should be 60 cells per
area. In rare cases in which amplified and nonamplified tumor
cells are intermingled in a single area, interpretation is difficult
and evidence is lacking. We suggest that for such cases, only
the presence of clearly amplified cells, with multiple HER2
signals, is considered evidence of heterogeneity, although
evidence is lacking in this area. Current evidence does not
support using the existence of small numbers of apparently
amplified cells within an individual tumor area to identify
heterogeneous amplification.
From the 1Endocrine Cancer Group, Edinburgh University,
Edinburgh, Scotland; 2Department of Pathology Birmingham
Heartlands Hospital, Birmingham, England; 3Experimental
Cancer Medicine Centre, University of Edinburgh, Edinburgh;
4Leiden University Medical Center, Leiden, the Netherlands;
5University Hospital, Freiburg, Germany; 6Athens University
Medical School, Athens, Greece; 7Department of Medical
Oncology, Erasmus Medical Center, Rotterdam, the Netherlands;
and 8Cancer Research UK Clinical Trials Unit, University of
Birmingham, Birmingham.
Supported by Roche, Burgess Hill, England.
Address reprint requests to Dr J.M.S. Bartlett: Ontario
Institute for Cancer Research, MaRS Centre, South Tower, 101
College St, Suite 800, Toronto, Ontario, Canada M5G 043.
References
1. Walker RA, Bartlett JMS, Dowsett M, et al. HER2 testing
in the UK: further update to recommendations. J Clin Pathol.
2008;61:818-824.
2. Wolff AC, Hammond ME, Schwartz JN, et al. American
Society of Clinical Oncology/College of American
Pathologists guideline recommendations for human epidermal
growth factor receptor 2 testing in breast cancer. J Clin
Oncol. 2007;25:118-145.
© American Society for Clinical Pathology
Am J Clin Pathol 2011;136:266-274
273
DOI: 10.1309/AJCP0EN6AQMWETZZ
273
273
Bartlett et al / HER2 Heterogeneous Amplification
3. Ellis IO, Bartlett J, Dowsett M, et al. Updated
recommendations for HER2 testing in the UK. J Clin Pathol.
2004;57:233-237.
4. Ellis IO, Dowsett M, Bartlett J, et al. Recommendations for
HER2 testing in the UK. J Clin Pathol. 2000;53:890-892.
5. Press MF, Finn RS, Cameron D, et al. HER-2 gene
amplification, HER-2 and epidermal growth factor receptor
mRNA and protein expression, and lapatinib efficacy in
women with metastatic breast cancer. Clin Cancer Res.
2008;14:7861-7870.
6. Piccart-Gebhart MJ, Procter M, Leyland-Jones B, et al.
Trastuzumab after adjuvant chemotherapy in HER2-positive
breast cancer. N Engl J Med. 2005;353:1659-1672.
7. Ellis P, Barrett-Lee P, Johnson L, et al. Sequential docetaxel
as adjuvant chemotherapy for early breast cancer (TACT):
an open-label, phase III, randomised controlled trial. Lancet.
2009;373:1681-1692.
8. Bartlett JMS, Ibrahim M, Jasani B, et al. External quality
assurance of HER2 FISH and ISH testing: three years of the
UK National External Quality Assurance Scheme. Am J Clin
Pathol. 2009;131:106-111.
9. Miller KD, Ibrahim M, Barnett S, et al. Technical aspects
of predictive and prognostic markers in breast cancer: what
UKNEQAS data shows. Curr Diagn Pathol. 2007;13:135-149.
10. Bempt IV, Drijkoningen M, Wolf-Peeters C. The complexity
of genotype alterations underlying HER2-positive by breast
cancer: an explanation for its clinical heterogeneity. Curr
Opin Oncol. 2007;19:552-557.
11. Farabegoli F, Santini D, Ceccarelli C, et al. Clone
heterogeneity in diploid and aneuploid breast carcinomas
as detected by FISH. Cytometry. 2001;46:50-56.
12. Vance GH, Barry TS, Bloom KJ, et al. Genetic heterogeneity
in HER2 testing in breast cancer: panel summary and
guidelines. Arch Pathol Lab Med. 2009;133:611-612.
13. Bartlett JMS, Brookes CL, Robson R, et al. Estrogen
receptor and progesterone receptor as predictive biomarkers
of response to endocrine therapy: a prospectively powered
pathology study in the Tamoxifen and Exemestane Adjuvant
Multinational (TEAM) trial. J Clin Oncol. 2011 29;15311538.
274
274
Am J Clin Pathol 2011;136:266-274
DOI: 10.1309/AJCP0EN6AQMWETZZ
14. Jones SE, Seynaeve C, Hasenburg A, et al. Results of the
first planned analysis of the TEAM (tamoxifen exemestane
adjuvant multinational) prospective randomized phase Ill
trial in hormone sensitive postmenopausal early breast cancer
[abstract]. Cancer Res. 2009;69(suppl 2):67S. Abstract 15.
15. van de Velde CJH, Rea DW, Seynaeve C, et al. Adjuvant
tamoxifen and exemestance in early breast cancer (TEAM):
a randomised phase 3 trial. Lancet. 2011;377:321-331.
16. Bilous M, Dowsett M, Hanna W, et al. Current perspectives
on HER2 testing: a review of national testing guidelines. Mod
Pathol. 2003;16:173-182.
17. Bilous M. HER2 testing recommendations in Australia.
Pathology. 2001;33:425-427.
18. Bartlett JM, Munro AF, Dunn JA, et al. Predictive markers
of anthracycline benefit: a prospectively planned analysis
of the UK National Epirubicin Adjuvant Trial (NEAT/
BR9601). Lancet Oncol. 2010;11:266-274.
19. Sauter G, Lee J, Bartlett JMS, et al. Guidelines for human
epidermal growth factor receptor 2 testing: biologic and
methodologic considerations. J Clin Oncol. 2009;27:13231333.
20. Bartlett JMS, Campbell FM, Mallon EA. Determination of
HER2 amplification by in situ hybridization: when should
chromosome 17 also be determined? Am J Clin Pathol.
2008;130:920-926.
21. Watters AD, Going JJ, Cooke TG, et al. Chromosome
17 aneusomy is associated with poor prognostic factors
in invasive breast carcinoma. Breast Cancer Res Treat.
2003;77:109-114.
© American Society for Clinical Pathology