07/06/2013

1

With main focus on the Estrogen Receptor

Anne-Vibeke LænkholmDepartment of PathologyNæstved-SlagelseDenmark

Commentary”The best way to achieve optimal treatment of today´s patients is to ensure the availability of

li bl d ti l th l i l t ireliable and timely pathological assessment in routine practice including treatment targetidentification”

Three questions in the evaluation of biomarkers

Question Feature

Is it true? Analytical Validity

Is it meaningful? Clinical Validity

Is it useful? Clinical Utility

Analytic Validityof a biomarker refers to its accuracy in measuring what it is supposed to measure▪ sensitivity and specificity of the assay▪ Assay robustness

Clinical ValidityRefers to the predictive value of a biomarker for a given clinical outcome▪ ie ER

Clinical UtilityIs the biomarker helpful in improving or maintaining the health of patients▪ Treatment strategy

Luminal A: ER-positive, HER2-negative, Ki67- low and

Luminal B: ER-positive, HER2-negative, Ki67> 14% or

L minal B ER positi e HER2 o ere pressed or amplifiedLuminal B: ER-positive, HER2-overexpressed or amplified, any Ki67

HER2-positive (non luminal): HER2 overexpressed oramplified and ER negative

Triple-negative: ER-negative, (PR-negative), HER2-negative

07/06/2013

2

Risk of recurrence pr. year N = 3,562 patients

Cop yrigh t © A m erican Soc ie ty o f C lin ical O ncolog y

Lin, N. U. et al. J Clin Oncol; 26:798-805 2008

Positive in app. 80% (ER) of breast carcinomascut off ≥ 1%

1-9% low≥10% high

ER-pos./PR-pos. ER-pos./PR-neg.ER-neg./PR-pos. ER-neg./PR-neg.

Serving as control for false negative ER?

PR TOTAL

Unknown( %)

Negative ( %)

positive (%)

ER Unknown (%) 54 (0.71) 1 (0.01) 2 (0.03) 57 (0.75)

Negative (%) 55 (0.72) 1269 (16.65) 23 (0.3) 1347 (17.67)

Positive (%) 547 (7.18) 1154 (15.14) 4518 (59.27) 6219 (81.58)

TOTAL 656 (8.61) 2424 (31.80) 4543 (59.60) 7623 (100)

Age Tumor size Lymph node status

Grade ER % HER2 RiskGroupsI (low)II (high)

≥ 60 years

≤ 10 mm negative I: IDCI + II: ILC

≥ 10% negative I

P i i II

Risk Factors and Risk GroupsDanish Breast Cancer Cooperative Group (DBCG) 2013

www.dbcg.dk

6% (2012)

Positive II

0-9% II

II +III: IDCIII: ILC

II

≥1 positive

II

> 10 mm II

< 60 years

II

Christiansen P et al. J Natl Cancer Inst. 2011 Sep 21;103(18):1363-72.

Preanalytical standardizationFixation▪ 10% NBF within 1 hour▪ Fixation time: 6 (at least) – 72 (no more than) hours.

Analytical standardizationAntibody/Antigen Retrieval/Detection SystemsControl samples

Postanalytical standardizationInterpretationCut-off levelInternal quality controlTissue / MaterialImage Analysis?

Participation in quality assurance programs

07/06/2013

3

AntibodyERα

Dilution Detection system

SP1monoclonal rabbit (12)(Th l V )

Ready To Use (majority)Optiview-DABUl i DAB(Thermocycler or Ventana)

Clone 6F11Monoclonal mouse (1)(Novocastra)

1:100

Ultraview-DABEnVisionEnVision Flex+ Rabbit

The majority of the pathology laboratories in Denmark apply the Benchmark Ultra(Ventana Medical Systems) for ER staining. A few laboratories apply the Dakoautostainer (DAKO /Denmark).

70%

80%90%

100%

biopsi60%

80%

100%

unknown

Danish Breast Cancer Cooperative Group (DBCG)

0%

10%20%

30%40%

50%

60%

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

Lump

Mast

0%

20%

40%

All 1 2 3 4 5 6 7 8 9 10 11 12

10-100%

1-9%

0%

70%

80%90%

100%

biopsi70%

80%

90%

100%

k

Change of cut-off from ≥ 10% to ≥ 1%

Danish Breast Cancer Cooperative Group

0%

10%20%

30%40%

50%

60%

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

Lump

Mast

0%

10%

20%

30%

40%

50%

60%

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

unknown

10-100%

1-9%

0%

Scoring method

Algoritm type Definitions Range of scores

Allred score PS + IS PS: 0:none; 1: <1%; 2: 1-10%; 3: 10-33% ; 4: 33-66%; 5: >66%IS: 0: none; 1: weak; 2: moderate; 3:strong

0,2 - 8

H-score Sum of (PS at a given ISxIS)

PS: 1-100%; IS: o: none; 1: weak; 2: moderate; 3:strong

0-300

Semi-quantitativeBright FieldLM

Percentage positive tumor cellsCut-off value ≥ 1

0-100%

IHC4(ER, PR, KI67, HER2)

Mathematicalmodel

Creation of an IHC4 score

IHC subtype Variation of IHC4

Determination of ER:Biochemical method, ex. dextran coated charcoal (DCC) and

presently immunohistochemistry

Cut-off value ≥ 1%(range 1-100%)

Poor dynamic range

Thorpe et al. Eur J Cancer 1994. 730 -274% mesured ER positive by DCC (age<70), median value of ER is 42 fmol/mg cytosol protein (0-2542)

IHC: •No method to determine the exact amount of

protein by weight in IHC exists .•The influence of fixation and HIER on protein loss is largely unknown, hence estimation of

staining intensity can be unreliable. Shi et al. J Histochem Cytochem 2011;59:13-32

07/06/2013

4

Guideline recommendationsOnly methods with pre-analytical and analytical components conformed exactly to clinically validated assays or compared with the clinically validated assays should be used for prediction of response for endocrine treatment ▪ Positive agreement must be ≥90% ▪ Negative agreement ≥95% ▪ Positive results are defined as ≥1% immuno-reactive cells

Arch Pathol Lab Med 2010;134:48-72

Cheang MC et al. J Clin Oncol 2006; 24:5637-44The proportion of ER positive breast cancers increased from 63% to 71% when SP1 was used instead of 1D5 with superior prognostic information.

Higher antibody affinity causes increased ER positivity in other organs i.e. lung adenocarcinomas.

Lau SK et al. Appl Immunohistochem Mol Morphol 2006;14:83-7.

564 patients

500 TMA

Evaluable TMA

Is the prevalence of ER (IHC) expression in primary breast cancer dependent on the choice of antibody and HIER?

Grabau D et al. : Acta Oncologica, 2013; epub Jan 23 slide courtesy of Dorthe Grabau

1D5, SP1n=390

PharmDxn=361

Evaluable TMA

1D5, SP1PharmDx

n=321

1D5

Tissue Micro Array (TMA)

Grabau D et al. : Acta Oncologica, 2013; epub Jan 23 slide courtesy of Dorthe Grabau

SP1

1D5Cut-off 1% - +

SP1 - 95 3+ 29 263

Overall agreement (95+263)/390 92 (89-94)Negative agreement 95/124 77 (68-84)Positive agreement 263/266 99 (97-99)

Cut-off 10% - +

Results agreement

SP1 - 109 0+ 24 257

Overall agreement (109+257)/390 94 (91-96)Negative agreement 109/133 82 (74-88)Positive agreement 257/257 100 (99-100)

Grabau D et al. : Acta Oncologica, 2013; epub Jan 23 slide courtesy of Dorthe Grabau

GuidelinePositive agreement must be ≥90% Negative agreement ≥95% Positive results are defined as ≥1% immunoreactive cellscompared with a clinically validated assay

0.25

0.50

0.75

1.00

1D5+/SP1+1D5-/SP1+1D5-/SP1-

Recurrence-free survival, cut-off 1%, +TAMb

1D5-/SP1+ vs 1D5+/SP1+: p=0.371D5-/SP1+ vs 1D5-/SP1- : p=0.42

0.25

0.50

0.75

1.00

1D5+/SP1+1D5-/SP1+1D5-/SP1-

Recurrence-free survival, cut-off 10%, +TAMe

1D5-/SP1+ vs 1D5+/SP1+: p=0.071D5-/SP1+ vs 1D5-/SP1- : p=0.86

1D5 and SP1

Grabau D et al. : Acta Oncologica, 2013; epub Jan 23 slide courtesy of Dorthe Grabau

0.00

48 39 30 27 26 261D5-/SP1-15 13 11 9 8 81D5-/SP1+

127 123 111 103 97 921D5+/SP1+Number at risk

0 1 2 3 4 5Years

1D5 /SP1 0.00

55 45 35 32 31 311D5-/SP1-14 13 10 7 6 61D5-/SP1+

123 119 108 101 95 901D5+/SP1+Number at risk

0 1 2 3 4 5Years

1D5 /SP1

In conclusion: •The proportion of ER postive breast cancer increased from 68% to 75% (cut-off 1%) when SP1 was used compared to 1D5. •The choice of antibody and HIER influences the prevalence of ER-positivity.•When applying changes in the staining procedure caution must be taken towardsthe guideline recommendations concerning positive and negative agreement withclinically validated assays

Core Needle Biopsy

07/06/2013

5

1391000813910006

13910006 13910008

Slide courtesy of Eva Balslev

Östrogen receptor

NordiQC : Positive staining reaction of the stromal cells indicates that a high sensitive protocol is being applied.

slide courtesy of Dorthe Grabau

Surgical tumor specimensCore Needle Biopsies (CNB)Tissue Micro Array (TMA)

ll b fSmall biopsies from metastastic sites i.e. bone (decalcified tissue)

ER CNB < 1% CNB 1-100%

SS <1% 8 1

SS 1-100% 0 80

No Neoadjuvant ChemotherapyN=89, bimodal distributionConcordance 98%Ranked correlation 1 (rho)Sensitivity 0.909Specificity 1Conclusion: With a cut-off value ≥ 1%, ER on CNB is sufficientlyaccurate and reliable.

Histological malignancy grade, estrogen receptor and HER2 receptor status on core needle biopsy compared to surgical specimen inbreast cancer – how accurate is it? Munch-Petersen HD, Balslev E.APMIS: Special Issue: Abstracts of the Annual Meeting of the Danish Pathology Society, 14–16 March 2013, Randers, DenmarkMarch 2013, Volume 121, Issue Supplement s135, Page 8 In case of negative ER on Core Needle Biopsy – repeat analysis on surgical specimen

07/06/2013

6

Discrepancy of resultsA change of 8-33% in Hormone Receptor status was reportedPR more discordant than ERPR more discordant than ER

Retesting of Hormone Receptor is recommended

Bernsdorf M et al. Breast Cancer Res Treat 2011 Jul;128(1):165-70. Value of post-operative reassessment of estrogen receptor α expression following neoadjuvantchemotherapy with or without gefitinib for estrogen receptor negative breast cancer.

Study Number of casesER

Size (mm) and number of cores

Scoring Method Concordance(%)

Henriksen KL et al. J Clin Pathol 2007;60:397-404

54 2 x 2 Allred score 96

Rossing HH et al. APMIS 2012; 120:341-347

89 2 x 2 categorized groups (10) 98.9; 34 347

Thomas A et al. Am J Clin Pathol2009;132:899-905

119 0.6 x 4 4 methods includingintensity and number of positive tumor cells

97.7 (VentanaAb)99.2 (Dako Ab)

Sapino A et al. VirchowsArch 2006;449:288-296

237 4 x 1 Quick score/Allred >98

Recommended publication:Pinder SE et al. The manufacture and assessment of tissue microarrays: suggestions and criteria for analysis, with breast cancer as an example.J Clin Pathol 2013; 66:169-177.(On behalf of the Translational Subgroup of the NRCI Breast Clinical Studies Group)

• ER discrepancy: 12 – 29%, often with loss of receptor

• HER2 discrepancy: 6 – 20%, often with gain of HER2+

Limitations:

• Many ”pathology chartreview” studies, did not re-analyse tumor samples (methodological variation)

• Prospective studies:-Biopsy had treatmentconsequence in 15-20% -Benign disease/othermalignancies in 14%

Slide courtesy of Jeanette Dupont Jensen. Department of Oncology, Odense University Hospital, Denmark

Standardization of staining methodologyPreanalyticalAnalyticalPostanalytical▪ According to American Society of Clinical Oncology/College of American Pathologists Guideline Recommendations (Arch Pathol Lab

Med. 2010;134:e48-e72)

Staining protocols have to be adjusted to the tissue material of interest

When applying changes in the staining procedure caution must be taken towards the guidelineWhen applying changes in the staining procedure caution must be taken towards the guideline recommendations concerning positive and negative agreement with clinically validated assays

Application of controls on all ER, (PR) stained slides

Repeat staining on surgical specimen (whole section) in case of ER negative result on primaryCore Needle Biopsy or TMA.

Repeat staining on surgical specimens after neoadjuvant chemotherapy in any case of ER staining.

Attendance in external Quality Assurance Programs

”Routine ER, PR, HER2 and KI67 used in combination as binary categories are not the same as a (centrally dertermined) IHC score (based on continuous variables) or multivariate genomicpredictors.” Dr. Lajos Pusztai, St. Gallen 2013.

Departments of Pathology, Denmark

Giedrius Lelkaitis▪ Aalborg University Hospital

Tomasz Piotr Tabor▪ Holstebro Hospital

Morten Johansen▪ Vendsyssel Hospital

Henrik Mertz▪ Randers Hospital

Danish Breast Cancer Cooperative Group (DBCG)▪ Karsten Bjerre▪ Susanne Møller▪ Maj-Britt Jensen

ER receptor status in Core Needle Biopsy. How accurate is it?▪ Helga Duverger Munch-Petersen & Eva BalslevRanders Hospital

Gorm Søndergaard▪ Viborg Hospital

Vibeke Jensen▪ Aarhus University Hospital

Tina Di Catarino▪ Sydvestjysk Hospital, Esbjerg

Birthe Østergaard▪ Vejle Hospital

Thomas Poulsen▪ Sønderjylland Hospital, Sønderborg

Anne-Marie Bak Jylling▪ Odense University Hospital

Henrik Mygind▪ Slagelse Hospital

Maj-Lis Møller Talman▪ Copenhagen University Hospital

Eva Balslev and Birgitte Bruun Rasmussen▪ Herlev Hospital

Helga Duverger Munch Petersen & Eva Balslev ▪ Department of Pathology, Herlev Hospital, Denmark

The prevalence of ER positivity▪ Dorthe Grabau

▪ Department of Pathology, Skåne University Hospital, Lund, Sweden.

▪ Pär-Ola Bendahl and Mårten Fernö▪ Division of Oncology, Department of Clinical Sciences, Lund

University, Lund, Sweden

▪ Lisa Rydén▪ Department of Surgery, Skåne University Hospital, Lund,

Sweden.

▪ Olle Stål▪ Department of Clinical and Experimental Medicine, Division

of Oncology, Faculty of Health Sciences, LinköpingUniversity, Linköping, Sweden

Acta oncologica 2013; PMID: 23343224

Jeanette Dupont Jensen▪ Department of Oncology, Odense University

Hospital, Odense, Denmark