APBI 2016

Accelerated Partial Breast Irradiation

Introduction

Philip Poortmans, MD, PhD 13 November 2016

Past-President

Department of Radiation Oncology

None of the teachers and others involved have a conflict of interest.

Advanced Multidisc. TC in Breast Cancer

Sandra Hol (NL)

Accelerated Partial Breast Irradiation

Course director : Philip Poortmans, Nijmegen (NL) Local organiser: Sofia Rivera Teachers:

Tibor Major, Budapest (HU) Birgitte Offersen, Aarhus (DK) Roberto Orrechia , Milano (IT) Vratislav Strnad, Erlangen (DE) Contouring administrator: Sandra Hol, Tilburg (NL) Guest lecturers: Corine Balleyguier; Magali Lacroix; Barbara Pistilli; Francoise Rimareix; Sofia Rivera ESTRO representative: Melissa Vanderijst

Accelerated Partial Breast Irradiation

Course aim:  Background: - increasing popularity of APBI

- results of prospective trials coming out

 Topics:

- focus on patient selection - define & train optimal target volume delineation - demonstrate and teach techniques

 Methods: - presentations

- clinical case discussions - target volume contouring exercises - debates

 Faculty: - specialists in the field

- completed with contribution from companies

Accelerated Partial Breast Irradiation

Accelerated Partial Breast Irradiation

Accelerated Partial Breast Irradiation

Accelerated Partial Breast Irradiation

Accelerated Partial Breast Irradiation

Accelerated Partial Breast Irradiation

Workshop volume delineation: We are sure you all did the homework! Laptops (own)  1-2 participants/laptop. Description clinical cases will be presented and are available on the Falcon website. Target volumes = GTV APBI; CTV APBI; both breasts. Delineation only all slices deemed necessary (for breast only every 3-4 slices).

Accelerated Partial Breast Irradiation

Structure of the clinical case discussions: 1. A clinical case will be introduced by one of the teachers. 2. A MCQ will be shown  voting by all. 3. The teacher leads the discussion. 4. This can be repeated (several times) for the course of the case. 5. The teacher concludes about the case, including a summary and some additive information.

Accelerated Partial Breast Irradiation

Accelerated Partial Breast Irradiation

Accelerated Partial Breast Irradiation

SOCIAL DINNER: Monday 15 th November 2016

Dinner Cruise on the Seine We will leave from the venue together directly at the end of the day.

For further information or if you wish to buy a ticket for an accompanying person please contact ESTRO staff

We expect you all there! Please inform the ESTRO staff if you can not join us.

Accelerated Partial Breast Irradiation

SOCIAL DINNER: Monday 15 th November 2016

Dinner Cruise on the Seine We will leave from the venue together directly at the end of the day.

For further information or if you wish to buy a ticket for an accompanying person please contact ESTRO staff

We expect you all there! Please inform the ESTRO staff if you can not join us.

Multidisciplinary Breast Cancer Course ENJOY Paris…

Accelerated Partial Breast Irradiation

Certificate of attendance upon completion of evaluation form

Please use the link to do the evaluation on-line.

Last versions of the presentations Available through the ESTRO website

Accelerated Partial Breast Irradiation

Multiple choice evaluation To be completed at the end of the course using the voting tool or afterwards– watch your e-mails. Only 25 questions!!!!

Certificate of participation to the examination with results upon completion.

Accelerated Partial Breast Irradiation

Thank you all for your active contribution! - Local organiser, Sofia Rivera and her team - Teachers - Contouring administrator

- ESTRO staff - Participants

Imaging for early Stage Breast Cancer

Corinne BALLEYGUIER Radiology Dpt Gustave Roussy, Villejuif, France

Disclosure of interest: Invited lectures for GEHC

New imaging techniques ?  Are they able to provide earlier detection?

 How to manage breast cancer staging for a personalized treatment or APBI?

(New) Imaging Techniques

 Mammography / Ultrasound  Digital Breast Tomosynthesis  Contrast-Enhanced Spectral Mammography  Breast MRI

Breast Cancer Screening  Mammography remains the best imaging tool for screening : results of 10 randomized trials  I

Stephen W. Duffy, Eugenio Paci2 2012 - 21%

Sept 2011

 1 saved life / 414 screened women during 7 years  Number of patients to screen during 10 years to avoid 1 death / breast cancer : 300  445043 50 yo women in France in 2016 …1483 avoided deaths at 10 years …  Most of avoided deaths are those which would occur after 10 years ….  Mandatory to follow patients at least 15 years …Overdiagnosis : 10%

CONCLUSIONS

BI-RADS 5 : Extensive DCIS

Breast Ultrasound ?

 Overall Se MG = 80%  Reduced Se in case of dense breast  Saarenmaa I. Breast Cancer Res Treat 2001  40 % of dense breast (C-D) in occidental population  Giger, AJR 2016  Debate on interest of complementary screening ultrasound in case of dense breast ……

Performances of screening breast ultrasound  Detection rate with normal MG :  4.4 % o ( 2.5-7 % o )  More invasive cancer  Similar detection rate with MG  Melnikow , AHRQ Publication No. 14-05201-EF-3, January 2016  Berg WA, JAMA 2012  Corsetti V, Eur J Cancer. 2011

 Recall rate: 13.9 % (MRI : 8-23 %)  Biopsy rate : 5-6 % (MRI : 7%)  Nevertheless, ultrasound induces FP and biopsies

Tomosynthesis

 Mammography limits:  Superposition of breast tissue  « False » images  Lesions seen on one view  Subtle images : distorsion  Overall Se = 80%  Aim of tomosynthesis :

 To avoid breast tissue overlapping  To improve detection of suspicious breast lesions

Principles

• 9 exposures • 25°

Reconstructions in tomographic views Plan // detector

2D Mammography

DBT

DBT and Screening

 Prospective studies :  3 European trials :  OTST : Norvegian Study  STORM : Italian Study  Malmö study : Swedish trial  Retrospective studies :  USA trials

Cancer detection rate /1000 Screening test European countries

Country

2D

2D + 3D

Multiplication rate

Norway

7,1

9,4

1,31

Italy

5,3

8,1

1,51

Sweden

6,3

8,9

1,41

Recall rate/1000 Screening test: Europe

Gain percentage

2D

2D + 3D

Country

Norway FP

103

85

17%

Italy FP

45

35

18%

Sweden FP

26

38

-43%

Tomosynthesis : Non indications

 Microcalcification analysis :  Micro analysis is changed

 Look brighter, round shape for amorphous calcifications  Look more benign  Breast cancer staging :  Accuracy not proved in comparison with other contrast examinations

Breast Cancer staging?

 Imaging techniques with contrast injection are more accurate as non contrast techniques :  MRI  Contrast-enhanced spectral mammography

 Detection of neo-angiogenesis  Breast lesion detection and characterization

Breast MRI  Indications in case of breast cancer :  Discordance > 1 cm between MG and ultrasound  Discordance between physical examination and conventional breast imaging  Women < 40 yo  High-risk women for breast cancer  Lobular invasive carcinoma

 To decide specific treatment such as :  Oncoplasty  Conservative treatment / Mastectomy  APBI  Neoadjuvant CT

HAS 2010

EUSOMA 2013

Breast MRI Staging

 Additional lesions detection rate :  Multifocal lesions are found in 20-60 % of mastectomy specimens  Contralateral occult carcinoma : 5-10%  Increased risk in case of lobular carcinoma and genetic risk  Mean additional rate of true lesions detected with MRI > MG + US  20-30% homolateral  4-6% contralateral  Change of treatment strategy with MRI : 28 %  Sardanelli, Breast 2010, Liberman AJR 2003, Lehman NEJM 2007, Mann RM. Breast Cancer Res Treat 2008, Sardanelli F. Radiology 2007

In case of suspicious Breast MRI?  Second look US  Additional MG views  In case of BI-RADS 4 and 5 breast lesions detected on MRI, occult on mammogram or second look ultrasound  Particularly in selected population :  High-risk women for breast cancer  After breast cancer treatment or breast cancer staging :  Suspicion of a recurrence after conservative treatment for breast cancer  MR-Guided Breast Biopsy

DCIS : MRI Staging?

DCIS : MRI Staging?

 16-47 % of DCIS are non calcified  Not seen on MG  US?

 Jansen, Semin Ultrasound CT MRI, 2011

 Resurgical rate for invaded margins : 30-70 % according studies  Staging +++

MRI and DCIS: controversies  Over and underestimation rate of tumor size on MRI

 Nearly only retrospective studies  Need of large prospective studies

Accurate Over

Under estimation

estimation

Hata 2004 13 % 39 %

48 % 24 % 17 % 31 %

Schouten 2006 Kim 2007

38 % 38 %

72 % 11 %

Santamaria 2008 Nori 2014

n.a

9 %

51.2 % 28.6%

30.4 %

53 yo DCIS 15 mm MG MRI: 7 cm lesion Surgery : multifocal

DCIS > 8 cm Mastectomy

Contrast-Mammography

 CESM :  MG + Iodine contrast injection  Spectral mammography  Alternative to MRI in case of contra-indication of breast MRI:  Same indications of MRI excepted High Risk Screening  Pace-Maker  Claustrophobia  Advantages :  Faster examination  Feasible immediately after CESM / US  Analysed by the same radiologist, in same conditions

CESMTechnique

Dual Energy approach : Spectral mammography

Low energy

High energy

∼ 20s

time

IV Injection

Radiation exposure The total X-ray dose for a pair of low and high-energy images corresponds to about 1.2 times the dose delivered for a standard mammogram

55 yo, distorsion left breast US: not conclusive

CESM : positive Carcinoma

Multifocal IDC + DCIS

Clinical Performance : CESM vs MG

Authors

N Analysis MX CESM

Dromain, 2011 110 Per

78 % 92%

patient

Fallenberg, 2014

107 Per

77,9% 94,7%

patient

Mokhtar, 2014 60 Per

93,2% 97,7%

patient

Lobbes, 2014 113 Per

96,9% 100%

patient

Tardivel, RSNA 2014

195 Per lesion

94%

Fallenberg et al Breast Cancer Res Treat 2014 107 patients, 56 dense (BI-RADS C / D) 3 radiologists

Jochelson, 2012

52 Per

81% 59%

96% 83%

patient Per lesion

Dromain, RSNA 2011

53 Per

NA 94% 93%

patient Per lesion

Se CESM 92-100% > Se MG Se does not vary with breast density (≠ MG)

Fallenberg, 2013

80 Per

81% 100%

patient

CESM vs MRI

 CESM is nearly comparable to MRI for cancer staging  Same sensitivity for primary cancer, slightly lower for 2 nd ary foci  Slighty higher specificity  Good estimation of lesion size similar to MRI  Better patient experience than MRI***  Faster time procedure and greater comfort  Lower noise level and rates of anxiety

Authors Jochelson, 2012

N Analysis MX CESM MRI

52 Per patient Per lesion

81 % 59 %

96% 83%

96% 93%

Dromain, RSNA 2011

53 Per patient Per lesion 80 Per patient

NA 94% 93%

100% 98%

Fallenberg, 2013

81 %

100% 97%

Luczynca, 2015 102 Per patient

NA 100% 93%

*Lobbes MB et al. J Cancer. 2015 Jan 5;6(2):144-50 **Cheung YC et al Eur Radiol, 2015; DOI 10,1007 ***Hobbs MM et al. J Med Imaging Radiat Oncol. 2015;59(3): 300-5

Conclusion

 Screening :  MG first +/- Ultrasound according breast density  Tomosynthesis may increase cancer detection  To be systematically integrated in general screening?  MRI :  Recommended in specific conditions  Particularly before complex surgery or APBI  CESM :  Alternative to Breast MRI

Pathology with an emphasis on early stage breast cancer

Map of Paris, detail, by Roland Castro

Dr Magali LACROIX-TRIKI Pathologist Gustave Roussy, Villejuif

ESTRO School, Paris (Villejuif) 13 November 2016

Plan

1. Introduction

2. Histopathological parameters

3. Molecular classification

4. Conclusion

(or try to…)

Local recurrence • Surgical margins Metastatic risk • Age • pTNM • Histological types • Grade • Vascular invasion

T N M

Response to treatment • ER/PR and HER2 status

Local risk (and beyond)

12 x 10 mm → macroscopic tumor size

minimal margin 8 mm → first margin assessment intra-operative exam: lateral margins < 5 mm considered as insufficient

Surgical margins

Ink the lateral margins

• DCIS: margin ≥ 2mm (JCO 2016) • Invasive: no ink on tumor (JCO 2014)

microscopic exam: lateral margins < 2 mm considered as insufficient

Histological type Adenoid cystic ca. Mucinous ca.

WHO 2012

- good prognosis : • tubular • mucinous • papillary

Secretory ca.

Papillary ca.

• adenoic cystic • secretory

- intermediate : medullary

Micropapillary

- poor prognosis : invasive micropapillary

→ IDC-NST and lobular carcinomas : variable !

Histological grade

- Elston and Ellis grade :

• architecture (tubular formation) (1-3) • nuclear atypia (1-3) • mitotic activity (per mm²) (1-3)

→ grade I (score 3-5) grade II (score 6-7) grade III (score 8-9)

Lymphovascular invasion

- at the tumor periphery - present in 20-30% - independent prognostic factor

Predictive biomarkers

HER2

ER/PGR

HR cut-off

HER2 cut-off

• 10% in France • 1-9%= 1.4% BC • 1% ASCO

• > 10%, ratio ≥2 • ASCO 2013 / GEFPICS 2014 • Increase of 2+ category

Positive predictive value

Negative predictive value

30-50%

(<5% chance to respond to anti-estrogens or trastuzumab) HIGH 95%

September 2016

September 2016

Monosomy 17

Ki67/Mib1

- Expressed in G1 , S , G2 et M - Semi-quantitative IHC

- Prognostic factor Mib1 ⇔ size (+)

grade (+) mitosis(+) ER(-)

- Predictive in neoadjuvant - Luminal A vs B, help to CT decision in ER+ BC (20% cut- off) - …but lack of reproducibility, especially for intermediate values 10-30%

Towards a simplified taxonomy of breast cancer? « definition of intrinsic subtypes has proven efficient in defining prognosis for breast cancer patients »

RE pos

RE neg

C Perou & T Sorlie

New paradigm: Treatment on tumor biology and not only on tumor stage

THERANOSTIC & DECISION MAKING

ANTI-HER2

Patients ER+ HER2-

ENDOCRINE THERAPY

CHEMOTHERAPY

CHEMOTHERAPY

Courtesy N Robin

ADJUVANT - Résultats

DBCG77B trial

100

100

HR = 0,50

HR = 1,07

80

80

60

60

40

40

DFS (%)

DFS (%)

20

20

Non-lum A + CT Non-lum A - CT

Lum A strict + CT Lum A strict - CT

0

0

0 1

2 3

4

5

6

7

8

9

10

0 1

2 3

4

5

6

7

8

9

10

Années

Années

LA+ LA−

Non-LA+ Non-LA−

354 114

264 69

217 46

194 36

174 33

45 14

354 114

264 69

217 46

194 36

174 33

45 14

SABCS 2015 – Nielsen TO et al., abstr. S1-08

ER+ HER2- tumors: Luminal A vs B ???

PROLIFERATION → Luminal ER+ HER2- BC is a spectrum !

70-gene signature

21-gene signature PAM50

Genomic grade HOXB13:IL17BR 2-gene ratio

11-gene assay

MammaPrint™ (Agendia)

Oncotype DX® (Genomic health)

Prosigna (Nanostring)

MapQuant DX (Ipsogen/Haliodx) Microarray / qRT- PCR

BCI (biotheranostics)

Endopredict® (Sividon)

Microarray

qRT-PCR

qRT-PCR

RT-PCR

qRT-PCR

RS = 0 -100 High / Low / Intermediate

ROR= 0 - 100 High / Low / Intermediate Molecular subtype

0 - 10 High / Low / Intermediate

0 -15 High vs Low

High vs Low

High / Low Equivocal

50 genes

97 genes Simplified : MYBL2, KPNA2, CDC2, CDC20

ER, PR, BCL2, SCUBE2, Ki67, STK15, BIRC5, CCNB1, MYBL2, HER2, GRB7, MMP11, CTSL2, GSTM1, CD68, BAG1 → 10 yr recurrence ER+, N0, with HT ER+, 1-3 N+ FFPE

HOXB13, IL17BR, BUB1, CENPA, NEK2, RACGAP1, RRM2

DHCR7, AZGP1, MGP, STC2, BIRC5, UBE2C, RBBP8, IL6ST

Cryo / FFPE → M+ 5 yr ER+ / ER- N- / N+ (1-3)

FFPE

Cryo / FFPE

FFPE

FFPE

→ 10 yr recurrence HR+ with tamoxifen → 15 yr recurrence

→ recurrence ER+ with tamoxifen

→ 5 yr / 10 yr recurrence

→ 10 yr recurrence ER+, HER2-, with HT N+

adjuvant CT (high)

CT adjuvant / neoadjuvant (high)

HT adjuvant and CT neoadjuvant (high)

HT adj > 5 yr (high)

MINDACT

TAILORx RxPONDER

ASTER 70s

MammaPrint (Agendia, NL)

HR+ and HR - / HER2- , T < 5cm, N ≤ 3 Fresh/frozen tissue (recently adapted to FFPE) 70 GENES CELL CYCLE / PROLIFERATION SIGNAL TRANSDUCTION INVASION, METASTASIS, ANGIOGENESIS CENTRALIZED

RECURRENCE (5 years) GOOD SIGNATURE : LOW RISK NO CHEMOTHERAPY POOR SIGNATURE : HIGH RISK CHEMOTHERAPY

Cardoso F et al. N Engl J Med 2016; 375:717-729

• Level of evidence 1A for clinical utility of MammaPrint® in the clinical – high group • C-High/g-Low (48% N+ !) show a 5-yr DMFS >94% (with or without CT) • In the whole population: 14% reduction in CT prescription • In the c-High subgroup: 46% reduction in CT prescription

OncotypeDX (Genomic Health, USA)

HR+ / HER2- , T1-3, N-/N+ FFPE samples 21 GENES PROLIFERATION, ESTROGEN, HER2, INVASION (16 GENES) + REFS (5 GENES) CENTRALIZED RECURRENCE (10 years) BENEFIT TO ADJUVANT CT LOW RISK (RS<18):

+ HT / NO CHEMOTHERAPY INTERMEDIATE: DISCUSSION HIGH RISK (RS≥31): +HT / + CHEMOTHERAPY

Prosigna (PAM50) (NanoString Technology, USA ) MOLECULAR SUBTYPES (LumA, LumB, HER2, Basal) FFPE SAMPLES 50 GENES LOCAL TEST (providing dedicated equipment and learning phase)

RECURRENCE (10 years) RISK OF RELAPSE (ROR) (node status needed)

LOW RISK + HT / NO CHEMOTHERAPY INTERMEDIATE HIGH RISK + HT / + CHEMOTHERAPY

ROR score (cut-off values depend of node status)

Molecular subtype

Probability of distant recurrence at 10 years (adapted from 2 clinical trials designed for HR+ menopausal women)

EndoPredict (Myriad genetics USA)

HR+ / HER2- , T1-2, N0 FFPE samples 11 GENES PROLIFERATION, ESTROGEN (8 genes) + 3 REF LOCAL TEST (providing dedicated equipment and learning phase)

STC2 AZGP1 IL65T RBBP8 MGP

UBE2C BIRC5 DHCR7

RECURRENCE (at 5 and 10 years)

LOW RISK: + HT / NO CHEMOTHERAPY HIGH RISK: + HT / + CHEMOTHERAPY

CONCLUSION

Grade Histological type

Phenotype

11

Margins

LVI

For a defined breast cancer population, ER+ / HER2- Intermediate characteristics, node negative (France)

From Sotiriou C, Piccart M. Nat Cancer Review 2007

THANK YOU…

Map of Paris, detail, by Roland Castro

ESTRO School, Paris (Villejuif) 13 November 2016

The basics behind the rise of APBI

Background

Philip Poortmans, MD, PhD 13 November 2016

Past-President

Department of Radiation Oncology

No conflict of interest.

The basics behind the rise of APBI: background

Why not ?

Background: practical issue

Rate BCS

1.0

0.76

0.55

0.14 (42% RT)

Distance:

25

100

230

km 300

Athas WF et al JNCI 2000

Background: practical issue USA

EU

 + 15 satellites

Background: changing concepts in LR treatment

± 1970 Maximal tolerable treatment

± 2000 Minimal effective treatment

Background: local tumour spread

Holland R. Cancer 1985

Background: local tumour spread

≤ 2cm - ≤ 4 cm @ PA margin

@ 1 cm: 59-61% @ 2 cm: 42-41% @ 3 cm: 17-18% @ 4 cm: 10-11%

5

0

10

15

Holland R. Cancer 1985

Background: recurrence pattern

Milan III

• N = 273 • Lumpectomy • No RT • ● = recurrence or new tumour

Mannino M & Yarnold J. Radiother Oncol 2009

Background: there are protagonists!

Don’t try this at home: expertise required!

Background: patient wish…

Background: or marketing?

Background: real life…..

Many centers already use APBI on a regular basis

Background: guidelines

Background: target volume delineation

Background: target volume delineation

Background: target volume delineation

Target volume delineation of primary tumour bed: - by dedicated RTO’s - no clips - no seroma

van Mourik AM et al. Radiother Oncol. 2010;94:286-91.

Background: target volume delineation

Background: target volume coverage

Bartelink H, et al. Radiother Oncol. 2012;104:139-42.

Background: target volume coverage

340 cGy

= tumour

= TE

= 4 cm line

Background: target volume coverage

1.0 cm

1.7 cm

= tumour

= TE

= 1 & 4 cm line

PHP, 15 November 2012

The basics behind the rise of APBI: background

Why (not) ?

The basics behind the rise of APBI: background

Background: role of RT

0 Gy

0 Gy

± 50 Gy

± 50 Gy

-15.4%

- 3.3%

5:1

EBCTCG Lancet 2011; 378: 1707–1716

Background: role of RT

0 Gy

0 Gy

± 50 Gy

± 50 Gy

-21.2%

- 8.5%

2.5:1

EBCTCG Lancet 2011; 378: 1707–1716

Background: role of RT

Dutch population based cancer registry 2000-2004 cohort: 37,207 patients

- 58.4% BCT - 41.6% MRM

Background: role of RT

Dutch population based cancer registry 2000-2004 cohort: 37,207 patients

- 58.4% BCT - 41.6% MRM

Background: stage migration

Percentages invasieve tumoren naar lokalisatie, stadium en incidentiejaar

C50, Borst

2003 2004 2005 2006 2007 2008 2009 % % % % % % % 38,6 37,7 39,4 39,8 41,9 41,2 41,8

Stadium*

1 2 3 4

41,3 41,5 41

40,1 40

40,4 39,5

14,7 14,8 13,8 14,2 12,8 12,2 12,7

4,5 0,9

5 1

5

5

4,3 1,1 100

5,1 1,1 100

4,9

0,7

0,9

1

Onbekend

100

100

100

100

100

Totaal

* Postchirurgische TNM (pTNM) aangevuld met de klinische TNM (cTNM)

Background: Interaction S and LR treatments

1/1.5

1/4

1/2-3

1/4

1/ ∞

Poortmans P. Lancet. 2014 Jun 21;383(9935):2104-6.

Background: open questions

• Patient selection criteria • Optimal techniques • Optimal schedules • Volume definition (QA !) • Long term FU: – local control – long term toxicity

Background: open questions

Background: possible approaches

• Follow up is to short

• Definition risk groups

• Practical issues

• Doesn ’ t fit me: volumes, techniques; collaboration; …

Background: possible approaches

• Level of evidence increases

• Longer follow up data

• Risk groups are defined

• Logistical benefits

• Most trials are closed

Background: possible approaches

• Discuss in multidisciplinary team

• At least consider starting with low risk group

Background: possible approaches

Target definition in partial breast irradiation GEC-ESTRO consensus V.Strnad

University Hospital Erlangen

3 pillars of APBI Target definition and delineation Key to success of APBI

Target definition 3 pillars

Patient selection

Appropriate technique

University Hospital Erlangen

GEC-ESTRO guidelines for target definition in breast brachytherapy

GUIDELINES

Target localization

Target definition

University Hospital Erlangen

GEC-ESTRO guidelines for target definition in breast brachytherapy

GUIDELINES

BACKGROUND

Target localization

Target definition

AVAILABLE INFORMATIONS

University Hospital Erlangen

GEC-ESTRO guidelines for target definition in breast brachytherapy

GUIDELINES

BACKGROUND

Target localization

Target definition

AVAILABLE INFORMATIONS

University Hospital Erlangen

Target definition and delineation

What we can use? 1. Scar - skin

2. Imaging (Ultrasound, Mammography, MRI, CT)

3. Surgical report

4. Surgical clips

5. Scar inside of breast

University Hospital Erlangen

Skin scar

Scar

Scar

x x

x

Clips

Clips

Scar

Clips

x

x x

Clips

Scar

University Hospital Erlangen

. Mammography before surgery

 Tumor size  Localization (Quadrant)

…but:

 Squeezing of tissue must be respected  Distances – only limited value

nipple

Medio-lateral view

University Hospital Erlangen

Mammography before surgery

 No absolute distance values!

University Hospital Erlangen

Ultrasound before surgery

University Hospital Erlangen

Ultrasound and mammography after surgery

Scare

?

University Hospital Erlangen

CT after surgery

University Hospital Erlangen

CT after surgery

Cavity Visualization Score

DM Landis et al., IJROBP, 2007, 67, 1299-1308

University Hospital Erlangen

Surgical report

Wide excission

Where and how …

Surgical report: 

Closed cavity Open cavity

  

Plastic reconstructions

Clips – yes/no

Plastic reconstruction

Quadrantectomy

University Hospital Erlangen

GEC-ESTRO guidelines for target definition in breast brachytherapy

GUIDELINES

BACKGROUND

Target definition

Target localization

AVAILABLE INFORMATIONS

Recommendations from GEC ESTRO Breast Cancer Working Group (I): Target definition and target delineation for accelerated or boost Partial Breast Irradiation using multicatheter interstitial brachytherapy after breast conserving closed cavity surgery.

University Hospital Erlangen

Vratislav Strnad 1 , Erik Van Limbergen 2 , Jean-Michel Hannoun-Levi 3 , Jose-Luis Guinot 4 , Kristine Lössl 5 , Daniela Kauer-Dorner 6 , Alexandra Resch 6 , György Kovács 7 , Tibor Major 8 , Csaba Polgár 8 on behalf of

Target delineation – needs:

Basic informations Surgical report:

1. 2. 3. 4.

Closed cavity Open cavity

Plastic reconstructions

Clips – yes/no

Pathology: 1.

Tumor size

2.

Resection margins in all 6 directions !

University Hospital Erlangen

Target definition

TWO KEY QUESTIONS

1. How large should be the size of the safety margin ? (existing resection margin + „brachytherapy“ safety margin)

2. What we know about the accuracy of estimation of position of the resection margins in corresponding directions. Consequently, how should we respect the possible inaccuracy by our definition of size of safety margins?

University Hospital Erlangen

Target definition

Margin size ? (resection margin + „brachytherapy“ safety margin)??

- Faverly / Holland,

Europe

- Ohtake,

Japan

- Imamura,

Japan

- Vicini /Goldstein

USA

University Hospital Erlangen

Resection margin and residual disease

European point of view

0 10 20 30 40 50 60 70 residual disease probability

47 %

39 %

Pts.

22 %

1 cm 2 cm 3 cm

Theoretical resection Margin

135 pts., mastectomy specimens, tumor size : 44% ≤ 2 cm, 56% >2cm & ≤ 4cm

APBI relevant summary:

Breast Carcinoma of Limited Extent ( ≤2cm, no EIC, L0, pN0, absent of calcifications ) ...

...might be a potential candidate for breast surgery alone if the tumor is excided with a macroscopically free margins of approximately 2 cm.

University Hospital Erlangen

Resection margin and residual disease

Japanese point of view

APBI relevant summary of computer assisted tumor mapping:

1.

Distance of intraductal tumor extension:

22,7 mm for pts. <30 y. 14,3 mm for pts. 30-40 y. 6,7-7,7 mm for pts. >40 y.

2. The average maximum distance of extension was 11.9 mm . (Patients 50 years of age had a maximum extension of 8 mm.)

3. In contrast to Holland’s study, the mean tumor size in this study was only 1.7 cm.

Ohtake T, Abe R, Kimijima I, et al. Intraductal extension of primary invasive breast carcinoma treated by breast-conserving surgery. Computer graphic three-dimensional reconstruction of the mammary duct-lobular systems. Cancer 1995;76: 32–45.

University Hospital Erlangen

Target definition - BACKGROUND

- The differences in extension of the intraductal components - 324 cases with invasive ductal breast carcinoma Japanese point of view Resection margin and residual disease

University Hospital Erlangen

Resection margin and residual disease

Japanese point of view

Despite median extension of 9 mm, 35% of cases had a ductal tumour spread in the distance between 10-30 mm ! APBI relevant summary:

Median

35%

34+24=60~60/253= 24%

34

18+9=27~27/253= 11%

24 18 9

Maximum distance of ductal spread (MXDS). [mm]

University Hospital Erlangen

Target definition

Margin size ? (resection margin + „brachytherapy“ safety margin)

Answers:

pts.

needed safety margins

(n)

Faverly:

135

20 mm

Ohtake:

20

8 mm

Imamura:

324

30 mm

Vicini/Goldstein: 333

10 mm

University Hospital Erlangen

Target definition - BACKGROUND

Target definition

Margin size ? (resection margin + „brachytherapy“ safety margin)

The size of „safety margin´s“ ( existing resection margin + „brachytherapy“ safety margin) should be by 2 cm in all directions. Summary:

University Hospital Erlangen

Target definition

TWO KEY QUESTIONS

1. How large should be the size of the safety margin (existing resection margin + „brachytherapy“ safety margin)?? √

20 mm

2. What we know about the accuracy of estimation of position of the resection margins in corresponding directions. Consequently, how should we respect the possible inaccuracy by our definition of size of safety margins?

University Hospital Erlangen

Value of type of surgery

…and the precision of surgical clips…?

clips

CTV in clinical practice

Chest wall fascia

Precision of clips 1. Variability during the time. 2. Random-variability.

University Hospital Erlangen

Clips

Resection margins

Value of type of surgery

…and the precision of surgical clips…?

clips

Precision of clips:

Chest wall fascia

Random-variability – no data

min. ± 3-5mm

University Hospital Erlangen

Clips

Resection margins

Clips

University Hospital Erlangen

Clips

University Hospital Erlangen

Guidelines

University Hospital Erlangen

Guidelines

1. The target delineation is to made CT based – without contrast agent.

2. Please visualize the skin scar and delineate surgical clips .

3. Delineate Whole Surgical Scar – WS ( between skin scar and surgical clips ).

4. Delineate Imaging related target volme – ImTV.

5. Delineate Estimated tumour bed - ETB.

6. Delineate Clinical target volume - CTV.

7. Delineate, only if necessary and usefull ( ! ), Planning target volume – PTV.

University Hospital Erlangen

Guidelines

1.

Perform a CT with a mark on the middle of the skin scar

Delineation of clips Two kinds of clips: intraparenchymal and on thoracic wall  The clinician decides which clips are relevant for delineating the surgical bed and which clips are considered as part of the target clinically “target relevant surgical clips”.  We need a precise surgical report in which the surgeon describes the number and the position of the clips and what are the “relevant” and “non relevant” .

University Hospital Erlangen

Guidelines

Delineation of the surgical bed (Whole Scar, WS) inside the breast  Delineate the surgical bed that is the visible scar of the closed cavity, including the whole scar and all the clips.  Whenever the scar is not visible, the space between the skin scar and clips on the thoracic wall can be delineated as virtual scar (depending of the breast size).

2.

 If there are no clips and the scar is not visible, tumor bed cannot be delineated.

University Hospital Erlangen

WHOLE SCARE Guidelines

University Hospital Erlangen

Definition and delineation of ImTV (Imaging correlated Target Volume) by using preoperative mammography, ultrasound, MRI. 1. Take the relative distance between the center of the visible tumor on the mammograms to the skin and chest wall . 2. In this relative distance and position delineate the projection of the largest tumor size on the scan. 3. Guidelines

18 : 42 = 0,42

25 : 60 = 0,42

25 mm

18 mm

42 mm

60 mm

University Hospital Erlangen

Guidelines

Definition and delineation of ImTV (Imaging correlated Target Volume) by using preoperative mammography, ultrasound, MRI. 1. Take the relative distance between the center of the visible tumor on the mammograms to the skin and chest wall . 2. In this relative distance and position delineate the projection of the largest tumor size on the scan. 3. Imaging correlated T rget Volume

COR/ROTATED

SAG

ImTV

University Hospital Erlangen

3-D

Guidelines

Definition and delineation of ETB (Estimated Tumour Bed) In consideration of three factors : 1. clinically “target relevant surgical clips”, 2. whole scar (WS) and 3. ImTV The physician delineate ETB as part of the whole surgical scar that is considered as related to the tumor position and tumor size. 4.

WS

WS

ImTV

ETB

University Hospital Erlangen

Guidelines

ESTIMATED TUMOUR BED

University Hospital Erlangen

Guidelines

5.

Definition and delineation of CTV (Clinical Target Volume)

 CTV = ETB plus adapted safety margins : 20 mm minus surgical margin, but at least 10 mm (a surgical margin of 3 mm requires a safety margin of 17mm).

Thoracic wall and skin (thickness 5 mm) are not the part of CTV.



Lesion 1,0 cm



Surgical safety margins:  Lat. 5 mm  Cranial 2 mm  Other > 1cm



University Hospital Erlangen

Guidelines

CTV

University Hospital Erlangen

Guidelines

6.

Delineation of PTV (Planning target volume) - clinical decision, if reasonable PTV = CTV + 5 (10) mm

 To respect the random inaccuracy of positioning of surgical clips.  Whenever there is uncertainty in the definition of the scar (CVS1, Landis)  For External Beam Radiation Therapy another safety margin will be required, as usual.

WS

ImTV

ETB

CTV

PTV

University Hospital Erlangen

Guidelines

PTV

University Hospital Erlangen

Target size

Brachy vs. EBRT

PTV (EBRT)

PTV (EVAL)

PTV (Brachy)

University Hospital Erlangen

Guidelines Special case of closed cavity – ONCOPLASTIC SURGERY

1. CTV is defined as the sum of the clipped area (CA) and the distance of 20 mm minus the smallest surgical free margin (SFM) defined by the pathologist (CTV= CA + (20- SFM).

2. PTV is defined as the CTV + 10 mm. The additional distance of 10 mm resulting from the interobserver delineation variability is not influenced by setting clip markers.

University Hospital Erlangen

Summary

Recommendations for daily routine

1.

Discuss with pathologists : 1.

Please respect margins in all directions.

2.

Discuss with surgeons : 1.

Please reconstruct the breast tissue.

Make the scar at the location of the tumor, and/or place clips.

2. 3.

Please cautiously orient the specimen.

3.

Explore possibilities for further imaging (e.g. MRI, US).

4.

Follow guidelines and use common sense !

University Hospital Erlangen

3 pillars of APBI Target definition and delineation Secret of success of APBI

Target definition 3 pillars

Patient selection

Appropriate technique

University Hospital Erlangen

Target volume delineation for APBI with open cavity: GEC-ESTRO guidelines Tibor Major

Accelerated Partial Breast Irradiation, 13-16 November 2016, Paris

Techniques for tumor bed localisation in APBI

• Clinical estimation (preoperative mammography, scar location, palpation, surgery report) • Surgical clip (or fiducial) positions on radiographic/CT images • CT (preoperative, postoperative - preimplant, postimplant) • Ultrasound (needle guidance , 3D planning) • MRI (combination with CT imaging)

Quantification of contour comparisons

V 2

V 1

second contour or reference isodose

first contour

V 1

∩V 2

common part

2 VV

1 VV 2

2 VV  



1

1

index Conformal

index Coverage

=

⋅

=

V

V

V

1

2

1

2 VV VV 2

2 VV

2

⋅



 

1

1

t coefficien Dice

index Conformity

=

=

2 VV

+

1

1

union

Jaccard coefficient

Multi-institutional study on target volume delineation variation in breast radiotherapy in the presence of guidelines

13 radiation oncologists (observers), 8 patients

CT slices of 3 patients with delineated excision cavity contours of all 13 observers

- even in the presence of delineation guidelines considerable delineation variation is present (0.24 < SD < 1.22 cm) - presence of clips or seroma reduced interobserver variation (0.24 < SD < 0.62 cm)

van Mourik et al. Radiother Oncol 94:286-91. 2010

Studies with interobserver variations of lumpectomy cavity contouring

Author

No. of observers

No. of patients Contours

Comments

Landis Wong

4

33

cavity, PTV

4 + 4

5 + 5

seroma, CTV, PTV

Petersen Kosztyla Boersma

3 3 5

30 21 30 12 70 10 15 10 30

seroma seroma

GTV, CTV cavity, PTV

pre- and postop. CT-s

Ford

1 + 1

CT and PET/CT

Jolicoeur Hurkmans

3 4

cavity

boost volume

4 CT series

Giezen Kirby 1 Kirby 2

2 + 2

cavity cavity cavity

4 1 9

prone position

Li

3 8

cavity, PTV

van Mourik Struikmans

13

cavity, CTV, PTV

5 2

18 19

boost CTV

Yang

seroma

CBCTs

GEC-ESTRO

4 + 6

9 + 5

cavity, CTV

pre- and postimplant CT-s

GEC-ESTRO Breast Cancer Working Group study

GEC-ESTRO BCWG contouring study

Phase 1

4 observers 5 patients 4 contours (pre- and postimplant cavity + CTV) no visibility score no guidelines

Phase 2

4 observers 4 patients

4 contours (pre- and postimplant cavity + CTV) 4 visibility scores (CVS2, CVS3, CVS4, CVS5) simplified rules for contouring

Interobserver variation between delineations of cavity and PTV (Phase 1)

Metrics for quantitative evaluation of comparisons between contours

- volumes and standard deviations from the mean - V max , V min , V max / V min - common and encompassing volumes (V common , V union ) - conformity index (CI common = V common / V union ) - generalized CI (CI gen ) - distance between center points of V common and V union

Patient 4 – preimplant

Patient 2 – postimplant

British Columbia Cancer Agency Seroma Clarity Scale

0 = no visible seroma

1 = scar/shadow

2 = seroma identifiable with significant uncertainties

3 = seroma identifiable with minor uncertainties

4 = seroma easily identifiable, generally homogeneous with slightly blurred margin

5 = seroma easily identifiable, homogeneous with sharp boundaries

Preimplant cavity contours with different cavity visualisation score

Distance between center points of

and V union

averaged for all volumes

V common

V max

/V min

averaged for all volumes

0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5

0,00 0,20 0,40 0,60 0,80 1,00 1,20

Distance (cm)

CVS2

CVS3

CVS4

CVS5

CVS2

CVS3

CVS4

CVS5

CI Conformity index averaged for all volumes

0,00 0,10 0,20 0,30 0,40 0,50 0,60 0,70

CVS2

CVS3

CVS4

CVS5

Mean V max

/V min

Phase_1 Phase_2 Mean conformity index (CI common )

0,70

0,00 0,50 1,00 1,50 2,00 2,50 3,00 3,50

0,60

Phase_1 Phase_2

0,50

0,40

CI

0,30

0,20

0,10

preimplant cavity

preimplant CTV

postimplant cavity

postimplant CTV

0,00

pre_cavity

pre_CTV post_cavity

post_CTV

Different CI-s for postimplant contours

cavity Phase_1 Phase_2

CTV Phase_1 Phase_2

0.32

0.36

0.41

0.54

CI common

< <

< <

0.50

0.56

0.61

0.70

CI pairs

0.49

0.56

0.61

0.70

CI gen

<

<

0.67

0.70

0.76

0.81

Dice

<

<

Conclusions from the contouring study

Simple guidelines on defining the lumpectomy cavity significantly increased the consistency of contouring

Reliable consistency of target volume definition can be expected only for good visibility of cavity

Recommendations for cavity contouring

use consistent windowing for cavity visualization (WL=0, WW=500)

only patients with CVS ≥ 3 are recommended for implantation

the visible seroma with homogeneous background has to be outlined only

and surgical disturbances such as breast tissue protrusions/flanges around

the excision cavity have to be exluded from the contour

if surgical clips are present they have to be surrounded by the contour with

close contact

CT number distribution along a line

HU = 0 (water)

Effect of windowing on visuality

WL=0, WW=400

WL=0, WW=500

WL=0, WW=600

WL=+50, WW=500

WL=-50, WW=500

WW

WL

Effect of windowing on visuality

WL=0, WW=300456 7

WL=0, WW=700

Recommendations for cavity contouring

use consistent windowing for cavity visualization (WL=0, WW=500)

only patients with CVS ≥ 3 are recommended for implantation

the visible seroma with homogeneous background has to be outlined only

and surgical disturbances such as breast tissue protrusions/flanges around

the excision cavity have to be exluded from the contour

if surgical clips are present they have to be surrounded by the contour with

close contact

CVS=2

CVS=3

CVS=4

CVS=5

Recommendations for cavity contouring

use consistent windowing for cavity visualization (WL=0, WW=500)

only patients with CVS ≥ 3 are recommended for implantation

the visible seroma with homogeneous background has to be outlined only

and surgical disturbances such as breast tissue protrusions/flanges around

the excision cavity have to be exluded from the contour

if surgical clips are present they have to be surrounded by the contour with

close contact

Seroma with no distinct border

 

?

What to include in seroma delineation

?

?

 

?

?

Recommendations for cavity contouring

use consistent windowing for cavity visualization (WL=0, WW=500)

only patients with CVS ≥ 3 are recommended for implantation

the visible seroma with homogeneous background has to be outlined only

and surgical disturbances such as breast tissue protrusions/flanges around

the excision cavity have to be exluded from the contour

if surgical clips are present they have to be surrounded by the contour with

close contact