Nuove Linee Guida ASCO/
CAP: Cosa Cambia. Esperienza del Centro Vincenzo Arena, M.D. Cagliari, 5 luglio 2014 SeC. 2013 To: Vincenzo Arena <vincenzo.arena@rm.unicaC.it> From: ken.kornfield@asco.org Subject: JCO/2013/520833 -­‐ Final Decision MS TITLE: HER2 Immunohistochemical Assessment: Is It Time To Refine The Scoring Criteria For The "2+" Category? -­‐ PLEASE DO NOT REPLY TO THIS EMAIL -­‐ Dear Dr. Arena: Thank you for your submission to Journal of Clinical Oncology. I have read your manuscript in full detail and JCO regrets to report that it is unable to accept your manuscript for publica2on. Many considera^ons factored into our decision. the appropriateness of your ar^cle for the broad readership of JCO in the absence of clinical outcomes (such as disease outcome), and our overall feeling that your paper would be more appropriate for a subspecialty pathology journal. Sincerely, We had concern regarding […] Antonio Wolff, MD Associate Editor Journal of Clinical Oncology 14.11.2013 Clinical outcome THE ALGORITHM Neoplasia
IHC
0
1+
3+
2+
ISH
Indeterminato –
+
Eq
Trastuzumab
No staining weak incomplete weak complete ≤ 10% weak complete at least 10% 0 1+ 1+ 2+ 3+ intense complete >30% 0 No staining 0 faint ≤ 10% 1+ faint incomplete > 10% 2+ weak/moderate e/o incomplete >10% 2+ Complete, intense and circumferen^al ≤ 10% 3+ Intense, complete, omogeneus,dark circumferen^al ≥10% Quale è la percentuale aCesa di casi nega^vi (sec. LG 2007) c h e p a s s e r a n n o n e l l a categoria “equivoca” (LG 2013)? LG 2007 13% di 0 LG 2013 52% di 0 16% 0 No Staining No Staining faint ≤ 10% Membrane staining <10% No staining Weak Incomplete 1+ Weak complete ≤10% 2+ Weak complete at least 10% Faint incomplete >10% Weak/ moderate e/o incomplete >10% Complete intense and circumferen^al ≤ 10% Faint/Barely percep^ble staining>10% (part of their membrane) Weak to moderate complete in >10% Intense complete >30% Intense, complete, omogeneus,dark circumferen^al ≥10% Strong complete in >10% ASCO/CAP 2007 ASCO/CAP 2013 FDA 3+ …sulla base del nuovo algoritmo… ↑ Casi 0 Casi 0 ↔ ↓ Casi 1+ Casi 1+ ↔↓ ↓ ↓ Casi 2+ Casi 2+ ↓ Casi 3+ Casi 3+ ↔ ↓ ↓ …Esperienza del centro… PRE ASCO/CAP 2013 Post ASCO/CAP 2013 Novembre 2012-­‐Maggio 2013 8 mesi Novembre 2013-­‐Maggio 2014 8 mesi 562 Determinazioni 592 Determinazioni 1154 Determinazioni di HER2 su K mammella invasivo …Esperienza del centro… “soDoEpi immunoistochimici” 300 266 250 233 216 212 200 Nov 2012-­‐Mag 2013 150 Nov 2013-­‐Mag 2014 100 49 50 35 41 37 41 24 0 Lum-­‐A Lum-­‐B Lum-­‐B H+ H+ TN …Esperienza del centro… Score di HER2 350 324 + 48 casi 302 300 250 200 Nov 2012-­‐Mag 2013 Nov 2013-­‐Mag 2014 150 134 121 112 100 64 53 44 50 0 0 1+ 2+ 3+ C …Esperienza del centro… 70,00% 60,00% 0 1+ 2+ 3+ N° casi Mar/Apr 2011 58,50% 20,40% 12% 9,20% 168 SeC/OC 2011 59,70% 17,90% 7,50% 14,90% 142 Mar/Apr 2012 63,30% 13,30% 17,20% 6,30% 128 SeC/OC 2012 50,90% 23,70% 16,70% 8,80% 114 Nov 2012/Mag 2013 57% 21% 11% 9,40% 562 Nov 2013/Mag 2014 51% 22% 18% 7,40% 592 0 1+ 2+ 3+ 50,00% Mar/Apr 2011 SeC/OC 2011 40,00% Mar/Apr 2012 SeC/OC 2012 30,00% Nov 2012/Mag 2013 Nov 2013/Mag 2014 20,00% 10,00% 0,00% 0 1+ 2+ 3+ C …Esperienza del centro… Distribuzione score 2 + 90 80 80 70 60 50 Nov 2012/ Mag 2013 38 40 Nov 2013/ Mag 2014 30 23 20 19 10 5 7 2 2 0 Lum-­‐A Lum-­‐B H+ TN C …Esperienza del centro… …E GLI ALTRI SCORES? 171 180 160 SCORE 0 80 152 70 138 140 72 62 60 120 105 54 54 50 100 80 40 60 30 34 40 SCORE 1 26 20 20 0 Lum-­‐A Lum-­‐B 0 TN Lum-­‐A SCORE 3 35 Lum-­‐B 32 30 25 21 8 5 10 22 22 20 15 10 5 0 Lum-­‐B H+ Nov 2012-­‐Mag 2013 H+ Nov 2013-­‐Mag 2014 TN THE ALGORITHM Neoplasia
IHC
0
1+
3+
2+
ISH
Indeterminato –
+
Eq
Trastuzumab
Single Probe Dual Probe Ra^o <1.8 HER<4 HER<4 Ra^o <2 HER<4 Ra^o 1.8/2.2 HER 4-­‐6 HER2 ≥ 4 <6 Ra^o < 2 HER2 ≥ 4<6 Ra^o <2 HER≤4 Non Amplificato Equivoco Ra^o≥2 HER ≥4 Ra^o >2.2 HER>6 ASCO/CAP 2007 HER>6 HER= 6 Ra^o≥2 HER <4 Ra^o ≥2 Her>4 Ra^o<2 HER >6 ASCO/CAP 2013 FDA Amplificato “Doppi equivoci” “Table 3 also shows that, CASI DOPPI EQUIVOCI “TEORICI” depending on w5hether à -­‐10% a single-­‐ or dual-­‐probe assay is used, 5% to 10% of the new IHC 2+ tumors would yield an equivocal FISH result, implying an increase in double-­‐
CASI Eequivocal SAMINATI = 126 results.” Correspondence
CASI 2+ = 20 Table 3. Comparison of IHC Scores and FISH Results According to the 2013 Guideline Update
FISH Result
Single Probe
Dual Probe
IHC
2013
Total
Negative
Equivocal
Positive
Negative
Equivocal
Positive
Score 0
Score 1!
Score 2!
Total
64
39
20
123
63
37
18
118
1
1
2
4
0
1
0
1
63
36
18
117
1
1
1
3
0
2
1
3
Abbreviations: FISH, fluorescent in situ hybridization; IHC, immunohistochemistry.
Single probe Her2 Copy number = 5 Dual Probe RaEo = 5/1à 5 RaEo = 5/2à 2.5 …Esperienza del centro… A fronte dell’incremento dei casi e^cheCa^ come “2+” in immunoistochimica, quan^ sono risulta^ amplifica^ e quan^ non amplifica^? C …Esperienza del centro… …GUARDIAMO I 2+ Nov 2012-­‐Mag 2013 Nov 2013-­‐Mag 2014 29 19 PRE ASCO/CAP 2013 Post ASCO/CAP 2013 C …Esperienza del centro… …La posi^vità globale rela^va ai casi HER2+ ….. (2+A)+ (3+)= 20,00% 18,00% 18,50% 17,50% 12.4% NOV 2012-­‐MAG 2013 12.3% NOV 2013-­‐MAG 2014 17,20% 16,00% …Nel tempo… 14,00% 12,00% 10,50% 10,00% 8,00% 6,00% 4,00% 2,00% 0,00% Mar/Apr 2011 SeC/OC 2011 Mar/Apr 2012 SeC/OC 2012 Nov 2012/Mag Nov 2013/Mag 2013 2014 C …Esperienza del centro… …La posi^vità globale rela^va ai casi HER2+ ….. (2+A)+ (3+)= 12.4% NOV 2012-­‐MAG 2013 12.3% NOV 2013-­‐MAG 2014 …Se non avessi usato le nuove linee guida ASCO/CAP….??? 12.4% NOV 2 NOV2012-­‐MAG2013 (2+A)+ (3+)= 10.8% NOV 2013-­‐MAG2014 rces are strained, receptor status, p53 expression and tumour grade,
o minimise the
among others.8 It is well known that low-grade
Downloaded from jcp.bmj.com on April 24, 2014 - Published by group.bmj.com
re. Our study
tumours are more likely to be oestrogen receptor
le
ical
(ER)/progesterone receptor (PR) positive and
need for further
HER-2 negative, while half of high-grade tumours
7
Positive HER-2/neu
ER/PR receptor
HER-2 positive.
Changingarefrequency
of equivocal
scores
expression is more likely to coincide with absence
and factors
predictive of negative HER 2/neu
of HER-2 overexpression, and the opposite relation
8–10
fluorescent
in assitu
invasive
holds
well.hybridisation
The study by in
Taucher
et al9 goes
th factor
recep- as far
saying
that “the assessment of HER-2/neu
carcinomas
ofasthe
breast
the standard of status in this (ER/PR positive) subgroup of patients
1
Fouad
Boulos,1may
Chantal
Farra,1 Ehabunnecessary”.
M Saad Aldin,2,3
Cleo Masaad,
breast
cancer
be G
considered
Different
studies
2
1,4
2
Yasmin Hassoun,
Fedda,
Hassanand
Doumiati,
Ayman Nstatus
Tawil,1 to
and therapeutic
haveFaysal
linked
ethnicity
menopausal
2
5
2
Zeina Nasser, Zeina Nahleh, Arafat Tfayli
y and
erican
cal
ersity
ated
y,
r
ock,
, USA
ersity
13
13
013
Boulos F, et al. J Clin Pathol 2014;67:204–209. doi:10.1136/jclinpath-2013-201546
ABSTRACT
Background Analysis of human epidermal growth
factor receptor 2 (HER-2) status has become standard of
care in breast cancer patients due to its important
prognostic and therapeutic implications.
Immunohistochemistry (IHC) is the most commonly used
primary method for detection of HER-2 overexpression.
Controversy exists on the interpretation of samples that
are equivocal for HER 2 status (IHC 2+). Recent
guidelines state that samples equivocal for HER 2 status
require validation with fluorescent in situ hybridisation
(FISH). The use of FISH, however, despite higher
accuracy comes at a higher cost that is not affordable to
all patients.
Methods This study is a retrospective study conducted
at the American University of Beirut Medical Center,
including women diagnosed with breast cancer with
equivocal IHC scores presenting between 2009 and
2011. We attempted to correlate clinicopathological
characteristics of patients diagnosed with breast cancer
that can influence conclusions made on HER 2 status
when analysing IHC equivocal samples in an effort to
decrease the need for FISH testing. 113 patients in our
records were included; charts were reviewed for different
patient clinical characteristics and samples were analysed
for pathological characteristics.
Results Using logistic regression, progesterone receptor
status and HER-2 staining of the normal glands around
113 cases “It is well known that low-­‐grade tumours are more likely to be oestrogen receptor(ER)/progesterone receptor (PR) posiEve and HER-­‐2 negaEve, while half of high-­‐grade tumours are HER-­‐2 posiEve.” implications. The most commonly used methods
for the detection of HER-2 overexpression are
immunohistochemistry (IHC) detecting protein overexpression and fluorescent in situ hybridisation
(FISH) detecting gene amplification.1 There is still no
consensus on a gold standard for determining HER-2
status with both methods posing advantages and disadvantages.2–5 IHC has the compelling advantages of
affordability, ease of performance and wide availability in most pathology laboratories, although its accuracy and reproducibility may be affected by tissue
handling and fixation, and antibody sensitivity and
specificity.6 Alternatively, and while being more reliable and reproducible than IHC, FISH is more costly
and necessitates advanced technical skill and experience.6 The 2007 American Society for Clinical
Oncology/College of American Pathologists (ASCO/
CAP) guidelines proposed a testing algorithm (figure 1)
that takes into account the above factors.3 Accordingly,
IHC is sufficient for positive and negative HER-2
results, while equivocal (2+) results are settled by
FISH testing. Interestingly, since the introduction of
the FISH technique to our pathology department in
2009 to abide by this algorithm, it was noted that
the rate of equivocal IHC results increased from
16% to 37%. A higher rate of FISH testing translated into a potentially unnecessary financial burden
on the many Lebanese patients without health
“PosiEve ER/PR receptor expression is more likely to coincide with absence of HER-­‐2 overexpression, and the opposite relaEon holds as well.” Taucher S, Rudas M, Mader RM, et al. patient clinical characteristics and samples were analysed
for pathological characteristics.
Results Using logistic regression, progesterone receptor
status and HER-2 staining of the normal glands around
the tumour by IHC were the two statistically significant
variables that showed association with FISH results. The
strength of progesterone receptor status positivity and
HER-2 staining of the normal glands around the tumour
were proportional to the likelihood of a negative FISH.
Also, the presence of strong and diffuse hormone
receptor positivity in low-grade tumours was predictive
of negative HER-2 status.
Conclusions In countries where resources are strained,
oncologists need to think of measures to minimise the
increasing financial burden of cancer care. Our study
serves to highlight a few clinicopathological
characteristics that might eliminate the need for further
testing through FISH.
the rate of equivocal IHC results increased from
16% to 37%. A higher rate of FISH testing transPredictors of negaKve HER2 FISH lated into a potentially
unnecessary financial burden
tesKng on the many Lebanese patients without health
insurance.
1.  “non tumour with 100% ER/PR+ or scores for mstate
itosis othat
r The ASCO/CAP with 2007low guidelines
a
tubule information
formaKon was isHER2 careful review of patient
important
amplified following an IHC 7 score of to guide determination of HER-2 status as there
2+” have been several published results on the correlation between HER-2
byabsence both oFISH
and
2.  The cstatus
omplete f normal IHC and the clinicopathological
of
gland staining wcharacteristics
as associated with a high level of FISH posiKvity breast cancer. Such characteristics
include
hormone
while tand
he presence receptor status, p53 (67%), expression
tumourof grade,
land staining showed a among others.8 It isnormal well gknown
that low-grade
concomitant 20% FISH posiKvity. tumours are more likely to be oestrogen receptor
(ER)/progesterone3.  receptor
(PR)are positive
Younger woman more likely and
to HER-2 negative, while
half
of high-grade
tumours
have aggressive HER2 amplified 7
neoplasm Positive ER/PR receptor
are HER-2 positive.
expression is more likely to coincide with absence
of HER-2 overexpression, and the opposite relation
holds as well.8–10 The study by Taucher et al9 goes
as far as saying that “the assessment of HER-2/neu
status in this (ER/PR positive) subgroup of patients
may be considered unnecessary”. Different studies
have linked ethnicity and menopausal status to
“…But it should sEll be emphasized that in cases where neither coverage nor laboratory resources are an issue, it remains advisable to default to FISH tesEng in all cases with equivocal IHC results in order to give every paEent the best opportunity for well-­‐guided management” INTRODUCTION
Analysis of human epidermal growth factor receptor 2 (HER-2) status has become the standard of
care in patients presenting with breast cancer
because of its important prognostic and therapeutic
Boulos F, et al. J Clin Pathol 2014;67:204–209. doi:10.1136/jclinpath-2013-201546
…Esperienza del centro… Ci sono “amplifica^” ed “amplifica^” C …Esperienza del centro… PRE ASCO/CAP 2013 Nov-­‐2012/Mag 2013 16 15 14 12 10 8 6 4 4 2 0 need to count not need to count Nov 2012-­‐Mag 2013 PRE ASCO/CAP 2013 C …Esperienza del centro… Nel periodo Nov 2012-­‐Mag 2013 6 casi amplifica^ “inaCesi” Ki67 <20% ER/PR pos ER/PR pos ER/PR neg 3 CDI NOS G2 N0 1 CDI NOS G2 N0 mul^focale 1 CDI G2 (aspe~ apocrini) N+ (sn) 1 Apocrino G3 N0 Età media 56 aa (46-­‐70) PRE ASCO/CAP 2013 C …Esperienza del centro… C …Esperienza del centro… Post ASCO/CAP 2013 Nov-­‐2013/Mag 2014 18 16 16 14 13 12 10 8 6 4 2 0 need to count not need to count Nov 2013-­‐Mag 2014 C …Esperienza del centro… Nel periodo Nov 2013-­‐Mag 2014 ….Tra i 29 casi 2+ Amplifica^….. 9 casi erano “inaCesi” ER/PR Pos Ki67 <20% 2 CDI “gland-­‐forming” N0 1 CDI G2 N0 1 CDI mul^centrico 1 CDI G2 micropapillare 1 CDI G2 (recidiva) (primo nel 2009) 1 CLI pleomorfo ER/PR Neg Post ASCO/CAP 2013 1 CDI G2 N0 1 CDI G2 con aspe~ metaplas^ci N0 Età media 67 aa (48-­‐86) Post ASCO/CAP 2013 Post ASCO/CAP 2013 “ …note that there are some rare breast cancers (eg, gland forming tumors, micropapillary carcinomas) that show IHC 1+ staining that is intense but incomplete (basolateral or U shaped) and that are found to be amplified. [ASCO/CAP Guidelines 2013] C …Esperienza del centro… Ci sono “non amplifica^” e “ non amplifica^” C …Esperienza del centro… 50 44 45 40 35 30 25 20 15 10 5 1 0 need to count not need to count Nov 2012-­‐Mag 2013 56 60 50 40 30 28 1,85 1,65 1,725 1,8 1,843 20 10 0 need to count not need to count Nov 2013-­‐Mag2014 C …Esperienza del centro… 1,85 1,65 1,725 1,8 1,843 “No classifica^on is perfect nor is it likely that it will ever be. All classifica^ons depend on o u r k n o w l e d g e o f t h e pathology. Since this knowledge is far from perfect or complete, no classifica^on can be other than a reasonable compromise…” C …Esperienza del centro… Epitelio apocrino “normale” AR+/ER-­‐/PR-­‐ APOCRINI PURI APOCRINE-­‐LIKE 69% puri 57% HER+ 31% “Like” 47% HER+ 55 casi