Top Banner
It takes a village to make a survivor Jamie L. Wagner, DO, FACOS Assistant Professor Division Chief, Breast Surgery Department of Surgery University of Kansas Health System
64

It takes a village to make a survivor

Apr 28, 2022

Download

Documents

dariahiddleston
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: It takes a village to make a survivor

It takes a village to make a survivorJamie L. Wagner, DO, FACOS

Assistant Professor

Division Chief, Breast Surgery

Department of Surgery

University of Kansas Health System

Page 2: It takes a village to make a survivor

A Century Long Unified

Paradigm of BC Surgical

Therapy

Page 3: It takes a village to make a survivor

Breast Cancer Progression

Alternate Hypothesis -

1960’s

Page 4: It takes a village to make a survivor

NSABP B-04

Page 5: It takes a village to make a survivor

NSABP B-06

Page 6: It takes a village to make a survivor

NSABP B-06

Effect of XRT and Systemic Therapy on

IBTR (20 Years)

IBTR (%)

Lump Lump + XRT

All Patients 39% 14%

Node Negative 36% 17%

Node Positive* 44% 17%

* Received adjuvant chemotherapy

Page 7: It takes a village to make a survivor

Breast Conserving Surgery vs.

Mastectomy

Invasive Breast Cancer

“Breast Conservation Treatment is an appropriate

method of primary therapy for the majority of women

with stage I and II breast cancer is preferable

because it provides survival equivalent to total

mastectomy while preserving the breast”

1990: NIH CDC Statement

Page 8: It takes a village to make a survivor

The Theme for the 1980’s and

1990’s:

Reduction in the Extent of Surgery

Page 9: It takes a village to make a survivor

SSO-ASTRO : Margins

Consensus Guideline

Study-level meta-analysis

Patients with invasive breast cancer

“No tumor on ink”

Dramatic decline in the rates of IBTR

5-year IBTR: 5.3%

Page 10: It takes a village to make a survivor

Nipple Sparing

Mastectomy

Before After

Page 11: It takes a village to make a survivor

Recurrence Risk

Page 12: It takes a village to make a survivor

Critical Dissection

Page 13: It takes a village to make a survivor

NSM Complications

0.5-7.3%

Page 14: It takes a village to make a survivor

Preventive Surgery

14

Page 15: It takes a village to make a survivor

Relative Risk Associated with

Breast Cancer Risk Factors

15

Page 16: It takes a village to make a survivor

Rationale for Sentinel Node

Biopsy

About 75% of patients with operable breast cancer have negative

axillary nodes upon dissection

Considerable morbidity from the procedure made an alternative

method for identifying involved axillary node desirable

Sentinel Node Concept:

Metastasis to regional lymph nodes follows an orderly

progression within the lymphatic system

Primary draining or sentinel node(s): first to contain metastases

and accurately predicts axillary node status

Page 17: It takes a village to make a survivor

Evolution of the SN

ConceptOriginal Concept Current Concept

Page 18: It takes a village to make a survivor

NSABP B-32Clinically Negative Axillary Nodes

N=5611

Stratification

Age

Clinical Tumor Size

Type of Surgery

Randomization

Group 1

SLNB + ALND

Group 2

SLNB

Positive SLN Negative SLN

ALND No ALND

Page 19: It takes a village to make a survivor

NSABP B-32

SizeTechnical Success

%

False Negative Rate

%

Overall 97.1 9.8

<2.0cm 96.9 10.3

2.1-4.0 cm 98.4 8.9

>4.0cm 98.2 7.4

p=0.03 p=0.86

Average # SLNs: 2.9

Page 20: It takes a village to make a survivor

NSABP B-32:

False-Negative Rate According to Number

of Removed SNs

Page 21: It takes a village to make a survivor

The 21st Century:

The Era of Loco-Regional Treatment

Individualization

Developments in breast imaging (digital

mammography, high resolution US, MRI)

Novel XRT techniques

Advances in systemic therapy

Molecular subtyping and genomic profiling

Developments in surgical therapy (NSM, expansion of

SLNB in node(+) its after NC)

Rates of Loco-regional recurrence have

steadily declined over the past 30 years

Page 22: It takes a village to make a survivor

ACOSOG Z0011

Primary Objective: To assess whether OS after SLNB

alone was not inferior to that for patients who

underwent completion ALND for a positive SLN

Page 23: It takes a village to make a survivor

Z0011: Loco-Regional

RecurrenceMedian follow-up = 6.3 years

Page 24: It takes a village to make a survivor

Z0011: Overall Survival

Page 25: It takes a village to make a survivor

AMAROS Trial:

Axillary Dissection vs. Axillary XRT After

(+) SLN

Primary Objective: To demonstrate non-inferiority in axillary

recurrence rate with axillary XRT vs. ALND

Page 26: It takes a village to make a survivor

AMAROS: Endpoints

Page 27: It takes a village to make a survivor

Effect of Neoadjuvant

Chemotherapy on Axillary Nodal

Metastasis

Neoadjuvant chemotherapy

down-stages axillary nodes in

20-40% of patients

Potential for decreasing

extent of axillary surgery with

SLNB

Page 28: It takes a village to make a survivor

Management of the Clinically Negative Axilla in

Patients Treated with Neoadjuvant Chemo

After a decade of fierce debate between SLNB before vs. after NC,

performing SLNB after NC has become arguably standard for

patients with operable BC

The approach capitalizes on the down staging effect of the NC in

sub-clinically involved axillary nodes

Its feasibility and accuracy has been shown in single-institution,

multi-center studies, and meta-analysis

SLN identification rate is lower than with upfront SLNB

No difference in FNR between the two approaches

Page 29: It takes a village to make a survivor

SLNB After Neoadjuvant Chemo in

Patients with Documented (+) Axillary

Nodes

Retrospective studies: Variability in SN identification rates

(78-98%) and SLN FNR (5-30%)

Three prospective trial were recently published (ACOSOG

Z0071, SESTINA, and SN FNAC)

Identification rates are lower with SLNB after neoadjuvant

chemotherapy (80-93%) compared to upfront SLNB

(>95%)

False negative rates range between 9.6-14% and are

mainly affects by the number of SNs removed

Page 30: It takes a village to make a survivor

SLNB After NC in Patients with + Nodes

FNR According to Number of Removed

Nodes

Page 31: It takes a village to make a survivor

The Next Nodal

Generation

Page 32: It takes a village to make a survivor

In 1989, Breast Cancer Was

So Much Simpler…

No subtypes

Half a dozen active chemotherapy agents

A few hormonal therapies, most of which did not work well

No non-hormonal therapy (in fact, not even a glimmer)

Doctors made most decisions

Cost of treatment was not a concern

Page 33: It takes a village to make a survivor

The Milan Trial: CMF vs. No

Chemotherapy

CMF x 12

Observation

Cyclophosphamide-

100mg/m2

Methotrexate-40mg/m2

IV days 1 and 8

Fluorouracil-600mg/m2

IV days 1 and 8 Q28

(20yrs) CMF Obs P

RFS 32% 25% 0.004

OS 34% 23% 0.04

Milan 2: 6 x CMF = 12 x CMF

Page 34: It takes a village to make a survivor

Chemotherapy in 1989

Adjuvant setting

CMF, CAF (mostly for node positive disease)

NCI alert issued informing medical oncologists of beneficial effect chemotherapy

in node negative disease

Metastatic setting

CMF

Doxorubicin

Vinblastine

Mitomycin

Thiotepa

Survival from time

chemotherapy started

usual no more than one

year

Page 35: It takes a village to make a survivor

Evolution of Adjuvant Chemotherapy

in 1990’s and Beyond

Anthracyclines became standard

Taxanes shown to be of incremental benefit

Dose dense AC-T became popular

Tens of thousands of women participated in trials showing

no difference between regimens or very small advantage

Clues began to emerge about which patients derived

greates benefit from chemotherapy

Page 36: It takes a village to make a survivor

2000 NIH Consensus

Conference

“Because adjuvant polychemotherapy improves

survival, it should be recommended to the majority of

women with localized breast cancer regardless of

nodal, menopausal, or hormone receptor status”

Bottom line: If tumor >1cm with or without nodal

involvment, chemo should be given

Page 37: It takes a village to make a survivor

Endocrine Therapy in

1989Adjuvant setting

Tamoxifen

Tamoxifen

Tamoxifen

Metastatic setting

Tamoxifen

Aminoglutethamide

High dose estrogen

Halotestin

Menace

But generally limited

to postmenopausal

women

Page 38: It takes a village to make a survivor

Overall and Relapse Free Survival

by Tumor Types Defined with Gene

Expression

Overall and Relapse Free Survival by Tumor Types Defined with Gene

Expression

Page 39: It takes a village to make a survivor

Breast Cancer is a Family of

DiseaseConvergence of clinical genomic data

Unclear how many distinct memebers of this family

At minimum

Her 2+

Basal-like or triple negative

ER+ (luminal A)

ER+ (luminal B)

“Basal-like”

Triple negative Her 2 positive ER positive luminal

B (high grade)ER positive luminal A

(low grade)

Page 40: It takes a village to make a survivor

Advances in the Past

Decade

Page 41: It takes a village to make a survivor

HER Family Signaling

Opportunities…and ChallengesHER Family Signaling

Opportunities…and Challenges

Page 42: It takes a village to make a survivor

CLEOPATRA: Dual Agent

Targeted Therapy

CLEOPATRA: Dual Agent Targeted TherapyThe n ew en gl an d j our n al of medi ci n e

n engl j med 372;8 nejm.org february 19, 2015728

Sensitivity Analyses

The 48 patients without disease progression who

opted to cross over from the control group to

receive pertuzumab had all been receiving treat-

ment for 2 years or longer. When their data were

censored at the time of the first pertuzumab

dose, the median overall survival was 56.5

months (95% CI, 49.3 to not reached) in the per-

tuzumab group and 39.6 months (95% CI, 35.0 to

45.1) in the control group (hazard ratio, 0.63;

B Subgroup Analysis of Overall Survival

A Overall Survival

0.6 1.0 2.0 3.0 4.0

Placebo BetterPertuzumab Better

All patients

Previous adjuvant or

neoadjuvant treatment

No

Yes

Region

Europe

North America

South America

Asia

Age

<65 yr

≥65 yr

Age

<75 yr

≥75 yr

Race or ethnic group

White

Black

Asian

Other

Disease type

Visceral

Nonvisceral

ER or PgR status

Positive

Negative

HER2 status

IHC 3+

FISH-positive

No. ofPatients Hazard Ratio (95% CI)Subgroup

0.71 (0.53–0.96)

1.11 (0.66–1.85)

0.61 (0.47–0.81)

0.66 (0.53–0.81)

0.69 (0.56–0.85)

0.59 (0.48–0.74)

0.82 (0.58–1.17)

0.37 (0.13–1.06)

0.68 (0.55–0.83)

0.63 (0.49–0.82)

0.41 (0.11–1.45)

0.85 (0.26–2.73)

0.53 (0.31–0.90)

0.70 (0.56–0.87)

0.82 (0.57–1.16)

0.50 (0.30–0.85)

0.63 (0.37–1.07)

0.65 (0.47–0.91)

0.70 (0.53–0.93)

0.67 (0.55–0.82)

0.2 0.4

0.64 (0.48–0.85)

P Value forInteraction

808

432

376

306

135

114

253

681

127

789

19

480

30

261

37

630

178

388

408

721

767

0.63

0.36

0.27

0.84

0.40

0.03

0.47

0.52

0.14

Ove

rall S

urv

ival (%

)

100

80

90

70

60

40

30

10

50

20

00 10 20 30 60 70 80

Months

Hazard ratio, 0.68 (95% CI, 0.56–0.84)

P<0.001

No. at RiskPertuzumabControl

402406

371350

318289

268230

40 50

226179

10491

2823

10

00

Pertuzumab, 168 events

Control, 221 events

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSITY OF KANSAS MEDICAL CENTER on October 12, 2015. For personal use only. No other uses without permission.

Copyright © 2015 Massachusetts Medical Society. All rights reserved.

Page 43: It takes a village to make a survivor

Progress in Triple Negative

Disease Has Been More

Limited

Page 44: It takes a village to make a survivor

TNT Trial: Docetaxel vs. Carboplatin in TNBC

Influence of BRCA Mutation Status

TNT Trial: Docetaxel vs. Carboplatin in TNBC Influence of BRCA Mutation Status

Interaction Term for treatment & BRCA 1/2 status p=0.03Interaction Term for treatment & BRCA 1/2 status p=0.03

Page 45: It takes a village to make a survivor

Advances Hormone Positive

Breast Cancer

Page 46: It takes a village to make a survivor

ATLAS Trial: Adjuvant Tamoxifen

Longer Against Shorter

Page 47: It takes a village to make a survivor

ATLAS Results Demonstrated

that 10 years of Tamoxifen is

Superior to 5 Years

ATLAS Results Demonstrated that 10 Years of Tamoxifen is Superior to 5

Years

Articles

www.thelancet.com Vol 381 March 9, 2013 809

tamoxifen has little net eff ect o n mortality not caused by

breast cancer (despite specifi c side-eff ects such as

endometrial cancer).1,2

Therefore, the main issue is how, in ER-positive disease,

10 years of treatment compares with 5 years of tamoxifen

in terms of main eff ects on recurrence and breast cancer

mortality, and how the specifi c side-eff ects of 10 years and

5 years of tamoxifen diff er.1,2,10

If the aim is to assess eff ects on breast cancer outcomes

in ER-positive disease, analyses need to be based either on

the fi ndings in patients known to have ER-positive disease

(which are straightforward to present and are provided in

full) or on a combination of the fi ndings in ER-positive

and ER-untested disease (which are provided as sensitivity

analyses). Emphasis on breast cancer outcomes only in

ER-positive disease was proposed by the data monitoring

committee statistician (RP) who knew the ATLAS results,

but the results from the main and sensitivity analyses

were much the same: appendix pp 14–18.

This study is registered, number ISRCTN19652633.

Role of the funding sourceOxford University (Oxford, UK) was the trial sponsor. The

study was designed, conducted, analysed, interpreted and

reported by the investigators independently of all funding

bodies (who saw the manuscript only after acceptance).

CD, HP, JG, RG, and RP had full access to all data and

had fi nal responsibility for the decision to submit for

publication.

ResultsFigure 1 describes the diff erent populations that were

analysed to assess the side-eff ects and the main eff ects

of continuing tamoxifen to 10 years versus stopping

tamoxifen at 5 years. After exclusion of 18 women who

had been entered in error and 2350 women who had

completed a median of only 2·4 years (IQR 2·0–3·1) of

adjuvant tamoxifen, 12 894 women remained who had

completed a median of 5·0 years (4·8–5·2) of adjuvant

tamoxifen. All were included in the analyses of side-

eff ects, regardless of ER status.

After exclusion of a further 6048 women with ER status

unknown or with ER-negative disease, 6846 women

with ER-positive disease remained for the main analy-

ses of the eff ects on breast cancer recurrence and breast

cancer mortality. Table 1 shows the characteristics of the

included patients.

Figure 2 shows compliance with the trial treatment

allocation. Among women who were without recurrence

2 years after entry (ie, at year 7 after diagnosis), 84% of

those allocated to continue were still on tamoxifen com-

pared with 4% of controls, a diff erence of 80%. Fewer

than 1% of women were receiving any adjuvant endocrine

treatment other than tamoxifen.

Figure 2 also shows that the completeness of follow-up

was similar in both treatment groups. In each group 91%

of the survivors were still being followed up 10 years after

diagnosis and 77% were still being followed up 15 years

after diagnosis; these proportions will increase as more

5–9 years: RR 0·90 (0·79–1·02)

≥10 years: RR 0·75 (0·62–0·90)

All years: log-rank p=0·002

5–9 years: RR 0·97 (0·79–1·18)

≥10 years: RR 0·71 (0·58–0·88)

All years: log-rank p=0·01

12·2%

15·0%

5·8%

6·0%

21·4%

25·1%

13·1%

14·5%

Continue tamoxifen to 10 years

Stop tamoxifen at 5 years

0

10

20

30

40

50

Cum

ulative

inciden

ce (%

)

A

0

(Diagnosis)

5

(ATLAS

entry)

10

(End of

treatment)

15

(10 years

since entry)

5–9 years

1·17%

(SE 0·09)

1·21%

(SE 0·09)

0·97 (SE 0·10)

–3·2/94·0

10–14 years

1·38%

(SE 0·12)

2·01%

(SE 0·15)

0·70 (SE 0·10)

–27·2/77·5

≥15 years

1·64%

(SE 0·39)

2·29%

(SE 0·47)

0·79 (SE 0·27)

–2·5/10·6

Continue tamoxifen to 10 years

Stop tamoxifen at 5 years

Rate ratio, from (O–E)/V

Log-rank O–E and variance V

5–9 years

2·83%

(428/15 115)

3·16%

(471/14 889)

0·90 (SE 0·06)

–24·8/224·7

10–14 years

1·96%

(165/8439)

2·66%

(214/8038)

0·74 (SE 0·09)

–29·1/94·7

≥15 years

2·54%

(24/945)

3·03%

(26/859)

0·85 (SE 0·26)

–2·1/12·5

B

0

(Diagnosis)

5

(ATLAS

entry)

10

(End of

treatment)

15

(10 years

since entry)

Figure 3: Recurrence (A) and breast cancer mortality (B) by treatment allocation for 6846 women with ER-positive disease

Bars show SE. Recurrence rates are percentage per year (events/patient-years of follow-up). Death rates (overall rate – rate in women without recurrence) are

percentage per year (SE). ATLAS=Adjuvant Tamoxifen: Longer Against Shorter.

Articles

www.thelancet.com Vol 381 March 9, 2013 809

tamoxifen has little net eff ect o n mortality not caused by

breast cancer (despite specifi c side-eff ects such as

endometrial cancer).1,2

Therefore, the main issue is how, in ER-positive disease,

10 years of treatment compares with 5 years of tamoxifen

in terms of main eff ects on recurrence and breast cancer

mortality, and how the specifi c side-eff ects of 10 years and

5 years of tamoxifen diff er.1,2,10

If the aim is to assess eff ects on breast cancer outcomes

in ER-positive disease, analyses need to be based either on

the fi ndings in patients known to have ER-positive disease

(which are straightforward to present and are provided in

full) or on a combination of the fi ndings in ER-positive

and ER-untested disease (which are provided as sensitivity

analyses). Emphasis on breast cancer outcomes only in

ER-positive disease was proposed by the data monitoring

committee statistician (RP) who knew the ATLAS results,

but the results from the main and sensitivity analyses

were much the same: appendix pp 14–18.

This study is registered, number ISRCTN19652633.

Role of the funding sourceOxford University (Oxford, UK) was the trial sponsor. The

study was designed, conducted, analysed, interpreted and

reported by the investigators independently of all funding

bodies (who saw the manuscript only after acceptance).

CD, HP, JG, RG, and RP had full access to all data and

had fi nal responsibility for the decision to submit for

publication.

ResultsFigure 1 describes the diff erent populations that were

analysed to assess the side-eff ects and the main eff ects

of continuing tamoxifen to 10 years versus stopping

tamoxifen at 5 years. After exclusion of 18 women who

had been entered in error and 2350 women who had

completed a median of only 2·4 years (IQR 2·0–3·1) of

adjuvant tamoxifen, 12 894 women remained who had

completed a median of 5·0 years (4·8–5·2) of adjuvant

tamoxifen. All were included in the analyses of side-

eff ects, regardless of ER status.

After exclusion of a further 6048 women with ER status

unknown or with ER-negative disease, 6846 women

with ER-positive disease remained for the main analy-

ses of the eff ects on breast cancer recurrence and breast

cancer mortality. Table 1 shows the characteristics of the

included patients.

Figure 2 shows compliance with the trial treatment

allocation. Among women who were without recurrence

2 years after entry (ie, at year 7 after diagnosis), 84% of

those allocated to continue were still on tamoxifen com-

pared with 4% of controls, a diff erence of 80%. Fewer

than 1% of women were receiving any adjuvant endocrine

treatment other than tamoxifen.

Figure 2 also shows that the completeness of follow-up

was similar in both treatment groups. In each group 91%

of the survivors were still being followed up 10 years after

diagnosis and 77% were still being followed up 15 years

after diagnosis; these proportions will increase as more

5–9 years: RR 0·90 (0·79–1·02)

≥10 years: RR 0·75 (0·62–0·90)

All years: log-rank p=0·002

5–9 years: RR 0·97 (0·79–1·18)

≥10 years: RR 0·71 (0·58–0·88)

All years: log-rank p=0·01

12·2%

15·0%

5·8%

6·0%

21·4%

25·1%

13·1%

14·5%

Continue tamoxifen to 10 years

Stop tamoxifen at 5 years

0

10

20

30

40

50

Cum

ulative

inciden

ce (%

)

A

0

(Diagnosis)

5

(ATLAS

entry)

10

(End of

treatment)

15

(10 years

since entry)

5–9 years

1·17%

(SE 0·09)

1·21%

(SE 0·09)

0·97 (SE 0·10)

–3·2/94·0

10–14 years

1·38%

(SE 0·12)

2·01%

(SE 0·15)

0·70 (SE 0·10)

–27·2/77·5

≥15 years

1·64%

(SE 0·39)

2·29%

(SE 0·47)

0·79 (SE 0·27)

–2·5/10·6

Continue tamoxifen to 10 years

Stop tamoxifen at 5 years

Rate ratio, from (O–E)/V

Log-rank O–E and variance V

5–9 years

2·83%

(428/15 115)

3·16%

(471/14 889)

0·90 (SE 0·06)

–24·8/224·7

10–14 years

1·96%

(165/8439)

2·66%

(214/8038)

0·74 (SE 0·09)

–29·1/94·7

≥15 years

2·54%

(24/945)

3·03%

(26/859)

0·85 (SE 0·26)

–2·1/12·5

B

0

(Diagnosis)

5

(ATLAS

entry)

10

(End of

treatment)

15

(10 years

since entry)

Figure 3: Recurrence (A) and breast cancer mortality (B) by treatment allocation for 6846 women with ER-positive disease

Bars show SE. Recurrence rates are percentage per year (events/patient-years of follow-up). Death rates (overall rate – rate in women without recurrence) are

percentage per year (SE). ATLAS=Adjuvant Tamoxifen: Longer Against Shorter.

Significant3.7%absolutereduc6oninrecurrence(p=.002)

Significant2.8%absolutereduc6oninbreastcancermortality(p=.01)

Recurrence Mortality

ATLAS Results Demonstrated that 10 Years of Tamoxifen is Superior to 5

Years

Articles

www.thelancet.com Vol 381 March 9, 2013 809

tamoxifen has little net eff ect o n mortality not caused by

breast cancer (despite specifi c side-eff ects such as

endometrial cancer).1,2

Therefore, the main issue is how, in ER-positive disease,

10 years of treatment compares with 5 years of tamoxifen

in terms of main eff ects on recurrence and breast cancer

mortality, and how the specifi c side-eff ects of 10 years and

5 years of tamoxifen diff er.1,2,10

If the aim is to assess eff ects on breast cancer outcomes

in ER-positive disease, analyses need to be based either on

the fi ndings in patients known to have ER-positive disease

(which are straightforward to present and are provided in

full) or on a combination of the fi ndings in ER-positive

and ER-untested disease (which are provided as sensitivity

analyses). Emphasis on breast cancer outcomes only in

ER-positive disease was proposed by the data monitoring

committee statistician (RP) who knew the ATLAS results,

but the results from the main and sensitivity analyses

were much the same: appendix pp 14–18.

This study is registered, number ISRCTN19652633.

Role of the funding sourceOxford University (Oxford, UK) was the trial sponsor. The

study was designed, conducted, analysed, interpreted and

reported by the investigators independently of all funding

bodies (who saw the manuscript only after acceptance).

CD, HP, JG, RG, and RP had full access to all data and

had fi nal responsibility for the decision to submit for

publication.

ResultsFigure 1 describes the diff erent populations that were

analysed to assess the side-eff ects and the main eff ects

of continuing tamoxifen to 10 years versus stopping

tamoxifen at 5 years. After exclusion of 18 women who

had been entered in error and 2350 women who had

completed a median of only 2·4 years (IQR 2·0–3·1) of

adjuvant tamoxifen, 12 894 women remained who had

completed a median of 5·0 years (4·8–5·2) of adjuvant

tamoxifen. All were included in the analyses of side-

eff ects, regardless of ER status.

After exclusion of a further 6048 women with ER status

unknown or with ER-negative disease, 6846 women

with ER-positive disease remained for the main analy-

ses of the eff ects on breast cancer recurrence and breast

cancer mortality. Table 1 shows the characteristics of the

included patients.

Figure 2 shows compliance with the trial treatment

allocation. Among women who were without recurrence

2 years after entry (ie, at year 7 after diagnosis), 84% of

those allocated to continue were still on tamoxifen com-

pared with 4% of controls, a diff erence of 80%. Fewer

than 1% of women were receiving any adjuvant endocrine

treatment other than tamoxifen.

Figure 2 also shows that the completeness of follow-up

was similar in both treatment groups. In each group 91%

of the survivors were still being followed up 10 years after

diagnosis and 77% were still being followed up 15 years

after diagnosis; these proportions will increase as more

5–9 years: RR 0·90 (0·79–1·02)

≥10 years: RR 0·75 (0·62–0·90)

All years: log-rank p=0·002

5–9 years: RR 0·97 (0·79–1·18)

≥10 years: RR 0·71 (0·58–0·88)

All years: log-rank p=0·01

12·2%

15·0%

5·8%

6·0%

21·4%

25·1%

13·1%

14·5%

Continue tamoxifen to 10 years

Stop tamoxifen at 5 years

0

10

20

30

40

50

Cum

ulative

inciden

ce (%

)

A

0

(Diagnosis)

5

(ATLAS

entry)

10

(End of

treatment)

15

(10 years

since entry)

5–9 years

1·17%

(SE 0·09)

1·21%

(SE 0·09)

0·97 (SE 0·10)

–3·2/94·0

10–14 years

1·38%

(SE 0·12)

2·01%

(SE 0·15)

0·70 (SE 0·10)

–27·2/77·5

≥15 years

1·64%

(SE 0·39)

2·29%

(SE 0·47)

0·79 (SE 0·27)

–2·5/10·6

Continue tamoxifen to 10 years

Stop tamoxifen at 5 years

Rate ratio, from (O–E)/V

Log-rank O–E and variance V

5–9 years

2·83%

(428/15 115)

3·16%

(471/14 889)

0·90 (SE 0·06)

–24·8/224·7

10–14 years

1·96%

(165/8439)

2·66%

(214/8038)

0·74 (SE 0·09)

–29·1/94·7

≥15 years

2·54%

(24/945)

3·03%

(26/859)

0·85 (SE 0·26)

–2·1/12·5

B

0

(Diagnosis)

5

(ATLAS

entry)

10

(End of

treatment)

15

(10 years

since entry)

Figure 3: Recurrence (A) and breast cancer mortality (B) by treatment allocation for 6846 women with ER-positive disease

Bars show SE. Recurrence rates are percentage per year (events/patient-years of follow-up). Death rates (overall rate – rate in women without recurrence) are

percentage per year (SE). ATLAS=Adjuvant Tamoxifen: Longer Against Shorter.

Articles

www.thelancet.com Vol 381 March 9, 2013 809

tamoxifen has little net eff ect o n mortality not caused by

breast cancer (despite specifi c side-eff ects such as

endometrial cancer).1,2

Therefore, the main issue is how, in ER-positive disease,

10 years of treatment compares with 5 years of tamoxifen

in terms of main eff ects on recurrence and breast cancer

mortality, and how the specifi c side-eff ects of 10 years and

5 years of tamoxifen diff er.1,2,10

If the aim is to assess eff ects on breast cancer outcomes

in ER-positive disease, analyses need to be based either on

the fi ndings in patients known to have ER-positive disease

(which are straightforward to present and are provided in

full) or on a combination of the fi ndings in ER-positive

and ER-untested disease (which are provided as sensitivity

analyses). Emphasis on breast cancer outcomes only in

ER-positive disease was proposed by the data monitoring

committee statistician (RP) who knew the ATLAS results,

but the results from the main and sensitivity analyses

were much the same: appendix pp 14–18.

This study is registered, number ISRCTN19652633.

Role of the funding sourceOxford University (Oxford, UK) was the trial sponsor. The

study was designed, conducted, analysed, interpreted and

reported by the investigators independently of all funding

bodies (who saw the manuscript only after acceptance).

CD, HP, JG, RG, and RP had full access to all data and

had fi nal responsibility for the decision to submit for

publication.

ResultsFigure 1 describes the diff erent populations that were

analysed to assess the side-eff ects and the main eff ects

of continuing tamoxifen to 10 years versus stopping

tamoxifen at 5 years. After exclusion of 18 women who

had been entered in error and 2350 women who had

completed a median of only 2·4 years (IQR 2·0–3·1) of

adjuvant tamoxifen, 12 894 women remained who had

completed a median of 5·0 years (4·8–5·2) of adjuvant

tamoxifen. All were included in the analyses of side-

eff ects, regardless of ER status.

After exclusion of a further 6048 women with ER status

unknown or with ER-negative disease, 6846 women

with ER-positive disease remained for the main analy-

ses of the eff ects on breast cancer recurrence and breast

cancer mortality. Table 1 shows the characteristics of the

included patients.

Figure 2 shows compliance with the trial treatment

allocation. Among women who were without recurrence

2 years after entry (ie, at year 7 after diagnosis), 84% of

those allocated to continue were still on tamoxifen com-

pared with 4% of controls, a diff erence of 80%. Fewer

than 1% of women were receiving any adjuvant endocrine

treatment other than tamoxifen.

Figure 2 also shows that the completeness of follow-up

was similar in both treatment groups. In each group 91%

of the survivors were still being followed up 10 years after

diagnosis and 77% were still being followed up 15 years

after diagnosis; these proportions will increase as more

5–9 years: RR 0·90 (0·79–1·02)

≥10 years: RR 0·75 (0·62–0·90)

All years: log-rank p=0·002

5–9 years: RR 0·97 (0·79–1·18)

≥10 years: RR 0·71 (0·58–0·88)

All years: log-rank p=0·01

12·2%

15·0%

5·8%

6·0%

21·4%

25·1%

13·1%

14·5%

Continue tamoxifen to 10 years

Stop tamoxifen at 5 years

0

10

20

30

40

50

Cum

ulative

inciden

ce (%

)

A

0

(Diagnosis)

5

(ATLAS

entry)

10

(End of

treatment)

15

(10 years

since entry)

5–9 years

1·17%

(SE 0·09)

1·21%

(SE 0·09)

0·97 (SE 0·10)

–3·2/94·0

10–14 years

1·38%

(SE 0·12)

2·01%

(SE 0·15)

0·70 (SE 0·10)

–27·2/77·5

≥15 years

1·64%

(SE 0·39)

2·29%

(SE 0·47)

0·79 (SE 0·27)

–2·5/10·6

Continue tamoxifen to 10 years

Stop tamoxifen at 5 years

Rate ratio, from (O–E)/V

Log-rank O–E and variance V

5–9 years

2·83%

(428/15 115)

3·16%

(471/14 889)

0·90 (SE 0·06)

–24·8/224·7

10–14 years

1·96%

(165/8439)

2·66%

(214/8038)

0·74 (SE 0·09)

–29·1/94·7

≥15 years

2·54%

(24/945)

3·03%

(26/859)

0·85 (SE 0·26)

–2·1/12·5

B

0

(Diagnosis)

5

(ATLAS

entry)

10

(End of

treatment)

15

(10 years

since entry)

Figure 3: Recurrence (A) and breast cancer mortality (B) by treatment allocation for 6846 women with ER-positive disease

Bars show SE. Recurrence rates are percentage per year (events/patient-years of follow-up). Death rates (overall rate – rate in women without recurrence) are

percentage per year (SE). ATLAS=Adjuvant Tamoxifen: Longer Against Shorter.

Significant3.7%absolutereduc6oninrecurrence(p=.002)

Significant2.8%absolutereduc6oninbreastcancermortality(p=.01)

Recurrence Mortality

ATLAS Results Demonstrated that 10 Years of Tamoxifen is Superior to 5

Years

Articles

www.thelancet.com Vol 381 March 9, 2013 809

tamoxifen has little net eff ect o n mortality not caused by

breast cancer (despite specifi c side-eff ects such as

endometrial cancer).1,2

Therefore, the main issue is how, in ER-positive disease,

10 years of treatment compares with 5 years of tamoxifen

in terms of main eff ects on recurrence and breast cancer

mortality, and how the specifi c side-eff ects of 10 years and

5 years of tamoxifen diff er.1,2,10

If the aim is to assess eff ects on breast cancer outcomes

in ER-positive disease, analyses need to be based either on

the fi ndings in patients known to have ER-positive disease

(which are straightforward to present and are provided in

full) or on a combination of the fi ndings in ER-positive

and ER-untested disease (which are provided as sensitivity

analyses). Emphasis on breast cancer outcomes only in

ER-positive disease was proposed by the data monitoring

committee statistician (RP) who knew the ATLAS results,

but the results from the main and sensitivity analyses

were much the same: appendix pp 14–18.

This study is registered, number ISRCTN19652633.

Role of the funding sourceOxford University (Oxford, UK) was the trial sponsor. The

study was designed, conducted, analysed, interpreted and

reported by the investigators independently of all funding

bodies (who saw the manuscript only after acceptance).

CD, HP, JG, RG, and RP had full access to all data and

had fi nal responsibility for the decision to submit for

publication.

ResultsFigure 1 describes the diff erent populations that were

analysed to assess the side-eff ects and the main eff ects

of continuing tamoxifen to 10 years versus stopping

tamoxifen at 5 years. After exclusion of 18 women who

had been entered in error and 2350 women who had

completed a median of only 2·4 years (IQR 2·0–3·1) of

adjuvant tamoxifen, 12 894 women remained who had

completed a median of 5·0 years (4·8–5·2) of adjuvant

tamoxifen. All were included in the analyses of side-

eff ects, regardless of ER status.

After exclusion of a further 6048 women with ER status

unknown or with ER-negative disease, 6846 women

with ER-positive disease remained for the main analy-

ses of the eff ects on breast cancer recurrence and breast

cancer mortality. Table 1 shows the characteristics of the

included patients.

Figure 2 shows compliance with the trial treatment

allocation. Among women who were without recurrence

2 years after entry (ie, at year 7 after diagnosis), 84% of

those allocated to continue were still on tamoxifen com-

pared with 4% of controls, a diff erence of 80%. Fewer

than 1% of women were receiving any adjuvant endocrine

treatment other than tamoxifen.

Figure 2 also shows that the completeness of follow-up

was similar in both treatment groups. In each group 91%

of the survivors were still being followed up 10 years after

diagnosis and 77% were still being followed up 15 years

after diagnosis; these proportions will increase as more

5–9 years: RR 0·90 (0·79–1·02)

≥10 years: RR 0·75 (0·62–0·90)

All years: log-rank p=0·002

5–9 years: RR 0·97 (0·79–1·18)

≥10 years: RR 0·71 (0·58–0·88)

All years: log-rank p=0·01

12·2%

15·0%

5·8%

6·0%

21·4%

25·1%

13·1%

14·5%

Continue tamoxifen to 10 years

Stop tamoxifen at 5 years

0

10

20

30

40

50

Cum

ulative

inciden

ce (%

)

A

0

(Diagnosis)

5

(ATLAS

entry)

10

(End of

treatment)

15

(10 years

since entry)

5–9 years

1·17%

(SE 0·09)

1·21%

(SE 0·09)

0·97 (SE 0·10)

–3·2/94·0

10–14 years

1·38%

(SE 0·12)

2·01%

(SE 0·15)

0·70 (SE 0·10)

–27·2/77·5

≥15 years

1·64%

(SE 0·39)

2·29%

(SE 0·47)

0·79 (SE 0·27)

–2·5/10·6

Continue tamoxifen to 10 years

Stop tamoxifen at 5 years

Rate ratio, from (O–E)/V

Log-rank O–E and variance V

5–9 years

2·83%

(428/15 115)

3·16%

(471/14 889)

0·90 (SE 0·06)

–24·8/224·7

10–14 years

1·96%

(165/8439)

2·66%

(214/8038)

0·74 (SE 0·09)

–29·1/94·7

≥15 years

2·54%

(24/945)

3·03%

(26/859)

0·85 (SE 0·26)

–2·1/12·5

B

0

(Diagnosis)

5

(ATLAS

entry)

10

(End of

treatment)

15

(10 years

since entry)

Figure 3: Recurrence (A) and breast cancer mortality (B) by treatment allocation for 6846 women with ER-positive disease

Bars show SE. Recurrence rates are percentage per year (events/patient-years of follow-up). Death rates (overall rate – rate in women without recurrence) are

percentage per year (SE). ATLAS=Adjuvant Tamoxifen: Longer Against Shorter.

Articles

www.thelancet.com Vol 381 March 9, 2013 809

tamoxifen has little net eff ect o n mortality not caused by

breast cancer (despite specifi c side-eff ects such as

endometrial cancer).1,2

Therefore, the main issue is how, in ER-positive disease,

10 years of treatment compares with 5 years of tamoxifen

in terms of main eff ects on recurrence and breast cancer

mortality, and how the specifi c side-eff ects of 10 years and

5 years of tamoxifen diff er.1,2,10

If the aim is to assess eff ects on breast cancer outcomes

in ER-positive disease, analyses need to be based either on

the fi ndings in patients known to have ER-positive disease

(which are straightforward to present and are provided in

full) or on a combination of the fi ndings in ER-positive

and ER-untested disease (which are provided as sensitivity

analyses). Emphasis on breast cancer outcomes only in

ER-positive disease was proposed by the data monitoring

committee statistician (RP) who knew the ATLAS results,

but the results from the main and sensitivity analyses

were much the same: appendix pp 14–18.

This study is registered, number ISRCTN19652633.

Role of the funding sourceOxford University (Oxford, UK) was the trial sponsor. The

study was designed, conducted, analysed, interpreted and

reported by the investigators independently of all funding

bodies (who saw the manuscript only after acceptance).

CD, HP, JG, RG, and RP had full access to all data and

had fi nal responsibility for the decision to submit for

publication.

ResultsFigure 1 describes the diff erent populations that were

analysed to assess the side-eff ects and the main eff ects

of continuing tamoxifen to 10 years versus stopping

tamoxifen at 5 years. After exclusion of 18 women who

had been entered in error and 2350 women who had

completed a median of only 2·4 years (IQR 2·0–3·1) of

adjuvant tamoxifen, 12 894 women remained who had

completed a median of 5·0 years (4·8–5·2) of adjuvant

tamoxifen. All were included in the analyses of side-

eff ects, regardless of ER status.

After exclusion of a further 6048 women with ER status

unknown or with ER-negative disease, 6846 women

with ER-positive disease remained for the main analy-

ses of the eff ects on breast cancer recurrence and breast

cancer mortality. Table 1 shows the characteristics of the

included patients.

Figure 2 shows compliance with the trial treatment

allocation. Among women who were without recurrence

2 years after entry (ie, at year 7 after diagnosis), 84% of

those allocated to continue were still on tamoxifen com-

pared with 4% of controls, a diff erence of 80%. Fewer

than 1% of women were receiving any adjuvant endocrine

treatment other than tamoxifen.

Figure 2 also shows that the completeness of follow-up

was similar in both treatment groups. In each group 91%

of the survivors were still being followed up 10 years after

diagnosis and 77% were still being followed up 15 years

after diagnosis; these proportions will increase as more

5–9 years: RR 0·90 (0·79–1·02)

≥10 years: RR 0·75 (0·62–0·90)

All years: log-rank p=0·002

5–9 years: RR 0·97 (0·79–1·18)

≥10 years: RR 0·71 (0·58–0·88)

All years: log-rank p=0·01

12·2%

15·0%

5·8%

6·0%

21·4%

25·1%

13·1%

14·5%

Continue tamoxifen to 10 years

Stop tamoxifen at 5 years

0

10

20

30

40

50

Cum

ulative

inciden

ce (%

)

A

0

(Diagnosis)

5

(ATLAS

entry)

10

(End of

treatment)

15

(10 years

since entry)

5–9 years

1·17%

(SE 0·09)

1·21%

(SE 0·09)

0·97 (SE 0·10)

–3·2/94·0

10–14 years

1·38%

(SE 0·12)

2·01%

(SE 0·15)

0·70 (SE 0·10)

–27·2/77·5

≥15 years

1·64%

(SE 0·39)

2·29%

(SE 0·47)

0·79 (SE 0·27)

–2·5/10·6

Continue tamoxifen to 10 years

Stop tamoxifen at 5 years

Rate ratio, from (O–E)/V

Log-rank O–E and variance V

5–9 years

2·83%

(428/15 115)

3·16%

(471/14 889)

0·90 (SE 0·06)

–24·8/224·7

10–14 years

1·96%

(165/8439)

2·66%

(214/8038)

0·74 (SE 0·09)

–29·1/94·7

≥15 years

2·54%

(24/945)

3·03%

(26/859)

0·85 (SE 0·26)

–2·1/12·5

B

0

(Diagnosis)

5

(ATLAS

entry)

10

(End of

treatment)

15

(10 years

since entry)

Figure 3: Recurrence (A) and breast cancer mortality (B) by treatment allocation for 6846 women with ER-positive disease

Bars show SE. Recurrence rates are percentage per year (events/patient-years of follow-up). Death rates (overall rate – rate in women without recurrence) are

percentage per year (SE). ATLAS=Adjuvant Tamoxifen: Longer Against Shorter.

Significant3.7%absolutereduc6oninrecurrence(p=.002)

Significant2.8%absolutereduc6oninbreastcancermortality(p=.01)

Recurrence Mortality

ATLAS Results Demonstrated that 10 Years of Tamoxifen is Superior to 5

Years

Articles

www.thelancet.com Vol 381 March 9, 2013 809

tamoxifen has little net eff ect o n mortality not caused by

breast cancer (despite specifi c side-eff ects such as

endometrial cancer).1,2

Therefore, the main issue is how, in ER-positive disease,

10 years of treatment compares with 5 years of tamoxifen

in terms of main eff ects on recurrence and breast cancer

mortality, and how the specifi c side-eff ects of 10 years and

5 years of tamoxifen diff er.1,2,10

If the aim is to assess eff ects on breast cancer outcomes

in ER-positive disease, analyses need to be based either on

the fi ndings in patients known to have ER-positive disease

(which are straightforward to present and are provided in

full) or on a combination of the fi ndings in ER-positive

and ER-untested disease (which are provided as sensitivity

analyses). Emphasis on breast cancer outcomes only in

ER-positive disease was proposed by the data monitoring

committee statistician (RP) who knew the ATLAS results,

but the results from the main and sensitivity analyses

were much the same: appendix pp 14–18.

This study is registered, number ISRCTN19652633.

Role of the funding sourceOxford University (Oxford, UK) was the trial sponsor. The

study was designed, conducted, analysed, interpreted and

reported by the investigators independently of all funding

bodies (who saw the manuscript only after acceptance).

CD, HP, JG, RG, and RP had full access to all data and

had fi nal responsibility for the decision to submit for

publication.

ResultsFigure 1 describes the diff erent populations that were

analysed to assess the side-eff ects and the main eff ects

of continuing tamoxifen to 10 years versus stopping

tamoxifen at 5 years. After exclusion of 18 women who

had been entered in error and 2350 women who had

completed a median of only 2·4 years (IQR 2·0–3·1) of

adjuvant tamoxifen, 12 894 women remained who had

completed a median of 5·0 years (4·8–5·2) of adjuvant

tamoxifen. All were included in the analyses of side-

eff ects, regardless of ER status.

After exclusion of a further 6048 women with ER status

unknown or with ER-negative disease, 6846 women

with ER-positive disease remained for the main analy-

ses of the eff ects on breast cancer recurrence and breast

cancer mortality. Table 1 shows the characteristics of the

included patients.

Figure 2 shows compliance with the trial treatment

allocation. Among women who were without recurrence

2 years after entry (ie, at year 7 after diagnosis), 84% of

those allocated to continue were still on tamoxifen com-

pared with 4% of controls, a diff erence of 80%. Fewer

than 1% of women were receiving any adjuvant endocrine

treatment other than tamoxifen.

Figure 2 also shows that the completeness of follow-up

was similar in both treatment groups. In each group 91%

of the survivors were still being followed up 10 years after

diagnosis and 77% were still being followed up 15 years

after diagnosis; these proportions will increase as more

5–9 years: RR 0·90 (0·79–1·02)

≥10 years: RR 0·75 (0·62–0·90)

All years: log-rank p=0·002

5–9 years: RR 0·97 (0·79–1·18)

≥10 years: RR 0·71 (0·58–0·88)

All years: log-rank p=0·01

12·2%

15·0%

5·8%

6·0%

21·4%

25·1%

13·1%

14·5%

Continue tamoxifen to 10 years

Stop tamoxifen at 5 years

0

10

20

30

40

50

Cum

ulative

inciden

ce (%

)

A

0

(Diagnosis)

5

(ATLAS

entry)

10

(End of

treatment)

15

(10 years

since entry)

5–9 years

1·17%

(SE 0·09)

1·21%

(SE 0·09)

0·97 (SE 0·10)

–3·2/94·0

10–14 years

1·38%

(SE 0·12)

2·01%

(SE 0·15)

0·70 (SE 0·10)

–27·2/77·5

≥15 years

1·64%

(SE 0·39)

2·29%

(SE 0·47)

0·79 (SE 0·27)

–2·5/10·6

Continue tamoxifen to 10 years

Stop tamoxifen at 5 years

Rate ratio, from (O–E)/V

Log-rank O–E and variance V

5–9 years

2·83%

(428/15 115)

3·16%

(471/14 889)

0·90 (SE 0·06)

–24·8/224·7

10–14 years

1·96%

(165/8439)

2·66%

(214/8038)

0·74 (SE 0·09)

–29·1/94·7

≥15 years

2·54%

(24/945)

3·03%

(26/859)

0·85 (SE 0·26)

–2·1/12·5

B

0

(Diagnosis)

5

(ATLAS

entry)

10

(End of

treatment)

15

(10 years

since entry)

Figure 3: Recurrence (A) and breast cancer mortality (B) by treatment allocation for 6846 women with ER-positive disease

Bars show SE. Recurrence rates are percentage per year (events/patient-years of follow-up). Death rates (overall rate – rate in women without recurrence) are

percentage per year (SE). ATLAS=Adjuvant Tamoxifen: Longer Against Shorter.

Articles

www.thelancet.com Vol 381 March 9, 2013 809

tamoxifen has little net eff ect o n mortality not caused by

breast cancer (despite specifi c side-eff ects such as

endometrial cancer).1,2

Therefore, the main issue is how, in ER-positive disease,

10 years of treatment compares with 5 years of tamoxifen

in terms of main eff ects on recurrence and breast cancer

mortality, and how the specifi c side-eff ects of 10 years and

5 years of tamoxifen diff er.1,2,10

If the aim is to assess eff ects on breast cancer outcomes

in ER-positive disease, analyses need to be based either on

the fi ndings in patients known to have ER-positive disease

(which are straightforward to present and are provided in

full) or on a combination of the fi ndings in ER-positive

and ER-untested disease (which are provided as sensitivity

analyses). Emphasis on breast cancer outcomes only in

ER-positive disease was proposed by the data monitoring

committee statistician (RP) who knew the ATLAS results,

but the results from the main and sensitivity analyses

were much the same: appendix pp 14–18.

This study is registered, number ISRCTN19652633.

Role of the funding sourceOxford University (Oxford, UK) was the trial sponsor. The

study was designed, conducted, analysed, interpreted and

reported by the investigators independently of all funding

bodies (who saw the manuscript only after acceptance).

CD, HP, JG, RG, and RP had full access to all data and

had fi nal responsibility for the decision to submit for

publication.

ResultsFigure 1 describes the diff erent populations that were

analysed to assess the side-eff ects and the main eff ects

of continuing tamoxifen to 10 years versus stopping

tamoxifen at 5 years. After exclusion of 18 women who

had been entered in error and 2350 women who had

completed a median of only 2·4 years (IQR 2·0–3·1) of

adjuvant tamoxifen, 12 894 women remained who had

completed a median of 5·0 years (4·8–5·2) of adjuvant

tamoxifen. All were included in the analyses of side-

eff ects, regardless of ER status.

After exclusion of a further 6048 women with ER status

unknown or with ER-negative disease, 6846 women

with ER-positive disease remained for the main analy-

ses of the eff ects on breast cancer recurrence and breast

cancer mortality. Table 1 shows the characteristics of the

included patients.

Figure 2 shows compliance with the trial treatment

allocation. Among women who were without recurrence

2 years after entry (ie, at year 7 after diagnosis), 84% of

those allocated to continue were still on tamoxifen com-

pared with 4% of controls, a diff erence of 80%. Fewer

than 1% of women were receiving any adjuvant endocrine

treatment other than tamoxifen.

Figure 2 also shows that the completeness of follow-up

was similar in both treatment groups. In each group 91%

of the survivors were still being followed up 10 years after

diagnosis and 77% were still being followed up 15 years

after diagnosis; these proportions will increase as more

5–9 years: RR 0·90 (0·79–1·02)

≥10 years: RR 0·75 (0·62–0·90)

All years: log-rank p=0·002

5–9 years: RR 0·97 (0·79–1·18)

≥10 years: RR 0·71 (0·58–0·88)

All years: log-rank p=0·01

12·2%

15·0%

5·8%

6·0%

21·4%

25·1%

13·1%

14·5%

Continue tamoxifen to 10 years

Stop tamoxifen at 5 years

0

10

20

30

40

50

Cum

ulative

inciden

ce (%

)

A

0

(Diagnosis)

5

(ATLAS

entry)

10

(End of

treatment)

15

(10 years

since entry)

5–9 years

1·17%

(SE 0·09)

1·21%

(SE 0·09)

0·97 (SE 0·10)

–3·2/94·0

10–14 years

1·38%

(SE 0·12)

2·01%

(SE 0·15)

0·70 (SE 0·10)

–27·2/77·5

≥15 years

1·64%

(SE 0·39)

2·29%

(SE 0·47)

0·79 (SE 0·27)

–2·5/10·6

Continue tamoxifen to 10 years

Stop tamoxifen at 5 years

Rate ratio, from (O–E)/V

Log-rank O–E and variance V

5–9 years

2·83%

(428/15 115)

3·16%

(471/14 889)

0·90 (SE 0·06)

–24·8/224·7

10–14 years

1·96%

(165/8439)

2·66%

(214/8038)

0·74 (SE 0·09)

–29·1/94·7

≥15 years

2·54%

(24/945)

3·03%

(26/859)

0·85 (SE 0·26)

–2·1/12·5

B

0

(Diagnosis)

5

(ATLAS

entry)

10

(End of

treatment)

15

(10 years

since entry)

Figure 3: Recurrence (A) and breast cancer mortality (B) by treatment allocation for 6846 women with ER-positive disease

Bars show SE. Recurrence rates are percentage per year (events/patient-years of follow-up). Death rates (overall rate – rate in women without recurrence) are

percentage per year (SE). ATLAS=Adjuvant Tamoxifen: Longer Against Shorter.

Significant3.7%absolutereduc6oninrecurrence(p=.002)

Significant2.8%absolutereduc6oninbreastcancermortality(p=.01)

Recurrence MortalityRecurrence Mortality

Page 48: It takes a village to make a survivor

TEXT/SOFTTEXT/ SOFT

Adj uvan t Ovar i an Suppr ession i n Br east Cancer

n engl j med 372;5 nejm.org january 29, 2015 443

Patients

(%

)

100

80

90

70

60

40

30

10

50

20

00 1 2 3 4 5 6

Years since Randomization

A Freedom from Breast Cancer

No. at RiskTamoxifenTamoxifen–OSExemestane–OS

TamoxifenTamoxifen–

OSExemestane–

OS

10181015

1014

140120

94

86.488.4

90.9

0.81 (0.63–1.03)

0.64 (0.49–0.83)

101810151014

956970957

900

932

912

855886869

728752766

533568550

314356342

No. ofPatients

No. ofPatients

withEvent

5-YrRate%

Hazard Ratio(95% CI)

Patients

(%

)

100

80

90

70

60

40

30

10

50

20

00 1 2 3 4 5 6

Years since Randomization

B Freedom from Distant Recurrence

No. at RiskTamoxifenTamoxifen–OSExemestane–OS

TamoxifenTamoxifen–

OSExemestane–

OS

10181015

1014

9689

70

90.791.3

93.0

0.88 (0.66–1.18)

0.71 (0.52–0.96)

101810151014

966977962

915943

920

875901882

755772783

559582562

333363352

No. ofPatients

No. ofPatients

withEvent

5-YrRate%

Hazard Ratio(95% CI)

Patients

(%

)

100

80

90

70

60

40

30

10

50

20

00 1 2 3 4 5 6

Years since Randomization

C No Chemotherapy, Freedom from Breast Cancer

No. at RiskTamoxifenTamoxifen–OSExemestane–OS

TamoxifenTamoxifen–

OSExemestane–

OS

476473

470

2423

14

95.895.1

97.1

0.95 (0.54–1.69)

0.59 (0.31–1.14)

476473470

461454443

445

447

425

429429414

377373374

277285278

169179176

No. ofPatients

No. ofPatients

withEvent

5-YrRate%

Hazard Ratio(95% CI)

Patients

(%

)

100

80

90

70

60

40

30

10

50

20

00 1 2 3 4 5 6

Years since Randomization

D No Chemotherapy, Freedom from Distant Recurrence

No. at RiskTamoxifenTamoxifen–OSExemestane–OS

TamoxifenTamoxifen–

OSExemestane–

OS

476473

470

67

3

98.698.7

99.3

1.16 (0.39–3.44)

0.52 (0.13–2.07)

476473470

465458444

449

453

429

436437419

386385381

284293283

176184180

No. ofPatients

No. ofPatients

withEvent

5-YrRate%

Hazard Ratio(95% CI)

Patients

(%

)

100

80

90

70

60

40

30

10

50

20

00 1 2 3 4 5 6

Years since Randomization

E Prior Chemotherapy, Freedom from Breast Cancer

No. at RiskTamoxifenTamoxifen–OSExemestane–OS

TamoxifenTamoxifen–

OSExemestane–

OS

542542

544

11697

80

78.082.5

85.7

0.78 (0.60–1.02)

0.65 (0.49–0.87)

542542544

494516514

455

485

487

426456455

352378391

255283273

144176166

No. ofPatients

No. ofPatients

withEvent

5-YrRate%

Hazard Ratio(95% CI)

Patients

(%

)100

80

90

70

60

40

30

10

50

20

00 1 2 3 4 5 6

Years since Randomization

F Prior Chemotherapy, Freedom from Distant Recurrence

No. at RiskTamoxifenTamoxifen–OSExemestane–OS

TamoxifenTamoxifen–

OSExemestane–

OS

542542

544

9082

67

83.684.8

87.8

0.87 (0.64–1.17)

0.72 (0.52–0.98)

542542544

501519518

466

490

491

439463463

369386401

274289280

156178172

No. ofPatients

No. ofPatients

withEvent

5-YrRate%

Hazard Ratio(95% CI)

Tamoxifen Tamoxifen–OS Exemestane–OS

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSITY OF KANSAS MEDICAL CENTER on October 14, 2015. For personal use only. No other uses without permission.

Copyright © 2015 Massachusetts Medical Society. All rights reserved.

Adjuvan t Ovar i an Suppr ession i n Br east Cancer

n engl j med 372;5 nejm.org january 29, 2015 443

Patien

ts (%

)

100

80

90

70

60

40

30

10

50

20

00 1 2 3 4 5 6

Years since Randomization

A Freedom from Breast Cancer

No. at RiskTamoxifenTamoxifen–OSExemestane–OS

TamoxifenTamoxifen–

OSExemestane–

OS

10181015

1014

140120

94

86.488.4

90.9

0.81 (0.63–1.03)

0.64 (0.49–0.83)

101810151014

956970957

900

932

912

855886869

728752766

533568550

314356342

No. ofPatients

No. ofPatients

withEvent

5-YrRate%

Hazard Ratio(95% CI)

Patien

ts (%

)

100

80

90

70

60

40

30

10

50

20

00 1 2 3 4 5 6

Years since Randomization

B Freedom from Distant Recurrence

No. at RiskTamoxifenTamoxifen–OSExemestane–OS

TamoxifenTamoxifen–

OSExemestane–

OS

10181015

1014

9689

70

90.791.3

93.0

0.88 (0.66–1.18)

0.71 (0.52–0.96)

101810151014

966977962

915943

920

875901882

755772783

559582562

333363352

No. ofPatients

No. ofPatients

withEvent

5-YrRate%

Hazard Ratio(95% CI)

Patien

ts (%

)

100

80

90

70

60

40

30

10

50

20

00 1 2 3 4 5 6

Years since Randomization

C No Chemotherapy, Freedom from Breast Cancer

No. at RiskTamoxifenTamoxifen–OSExemestane–OS

TamoxifenTamoxifen–

OSExemestane–

OS

476473

470

2423

14

95.895.1

97.1

0.95 (0.54–1.69)

0.59 (0.31–1.14)

476473470

461454443

445

447

425

429429414

377373374

277285278

169179176

No. ofPatients

No. ofPatients

withEvent

5-YrRate%

Hazard Ratio(95% CI)

Patien

ts (%

)

100

80

90

70

60

40

30

10

50

20

00 1 2 3 4 5 6

Years since Randomization

D No Chemotherapy, Freedom from Distant Recurrence

No. at RiskTamoxifenTamoxifen–OSExemestane–OS

TamoxifenTamoxifen–

OSExemestane–

OS

476473

470

67

3

98.698.7

99.3

1.16 (0.39–3.44)

0.52 (0.13–2.07)

476473470

465458444

449

453

429

436437419

386385381

284293283

176184180

No. ofPatients

No. ofPatients

withEvent

5-YrRate%

Hazard Ratio(95% CI)

Patients

(%

)

100

80

90

70

60

40

30

10

50

20

00 1 2 3 4 5 6

Years since Randomization

E Prior Chemotherapy, Freedom from Breast Cancer

No. at RiskTamoxifenTamoxifen–OSExemestane–OS

TamoxifenTamoxifen–

OSExemestane–

OS

542542

544

11697

80

78.082.5

85.7

0.78 (0.60–1.02)

0.65 (0.49–0.87)

542542544

494516514

455

485

487

426456455

352378391

255283273

144176166

No. ofPatients

No. ofPatients

withEvent

5-YrRate%

Hazard Ratio(95% CI)

Patients

(%

)

100

80

90

70

60

40

30

10

50

20

00 1 2 3 4 5 6

Years since Randomization

F Prior Chemotherapy, Freedom from Distant Recurrence

No. at RiskTamoxifenTamoxifen–OSExemestane–OS

TamoxifenTamoxifen–

OSExemestane–

OS

542542

544

9082

67

83.684.8

87.8

0.87 (0.64–1.17)

0.72 (0.52–0.98)

542542544

501519518

466

490

491

439463463

369386401

274289280

156178172

No. ofPatients

No. ofPatients

withEvent

5-YrRate%

Hazard Ratio(95% CI)

Tamoxifen Tamoxifen–OS Exemestane–OS

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSITY OF KANSAS MEDICAL CENTER on October 14, 2015. For personal use only. No other uses without permission.

Copyright © 2015 Massachusetts Medical Society. All rights reserved.

1. Women who had a significant risk of recurrence to warrant chemotherapy that remained

premenopausal benefit f rom Tamoxifen and OS

2. OS with an AI further reduced risk of recurrence compared to Tamoxifen-based therapy

1. Women who had a significant risk of recurrence to warrant

chemotherapy that remain premenopausal benefit from Tamoxifen and

OS

2. OS with an AI further reduced risk of recurrence compared to Tamoxifen-

based therapy

Page 49: It takes a village to make a survivor

Advances in Radiation

Page 50: It takes a village to make a survivor

Fisher et al., NEJM, 2002

NSABP B-06: Mastectomy

vs. Lumpectomy

Page 51: It takes a village to make a survivor

Fisher et al., NEJM, 2002

NSABP B-06: LRR with XRT

Page 52: It takes a village to make a survivor

EBCTCG: Meta-analysis

EBCTCG, Lancet 2011

Page 53: It takes a village to make a survivor

EBCTCG, Lancet 2011

EBCTCG

Page 54: It takes a village to make a survivor

Less is more?

Page 55: It takes a village to make a survivor

Randomized APBI trials

Page 56: It takes a village to make a survivor

Randomized APBI trials

Page 57: It takes a village to make a survivor

Whelan et al., NEJM, 2010

Hypofraction: LRR

Page 58: It takes a village to make a survivor

Whelan et al., NEJM, 2010

Hypofractionation: OS

Page 59: It takes a village to make a survivor

Haviland et al., Lancet Oncol, 2013

START Trial

Page 60: It takes a village to make a survivor

Haviland et al., Lancet Oncol, 2013

START Trial DFS

Page 61: It takes a village to make a survivor

Hypofraction RT: Here to

Stay

Standard of Care for early stage

Trials open for locally advanced disease

Trials assessing cosmesis

Page 62: It takes a village to make a survivor

Conclusion

The improved outcomes for breast caner patients has

come from a village of different sub-specialties all

making advances in their field

Critical importance to treating patients in a

comprehensive multidisciplinary team approach to

optimize patient outcomes

Page 63: It takes a village to make a survivor

Comprehensive Breast Program

Breast Cancer Clinics

Breast oncologyBreast surgery

Radiation oncology

Breast Imaging Screening breast imaging High risk breast imaging 3D mam (tomosynthesis) Diagnostic imaging Image guided biopsies

We have multidisciplinary tumor concordance conference for benign

and malignant disease.

Call 913-588-6804 (option 3) for scheduling.

High Risk ClinicAll patients with:

Family History of breast and/or ovarian cancer

Proliferative biopsy (atypia) Dense Breasts Mutation

Otherwise considered high risk

Consult includes risk assessment, genetic counseling

and testing if indicated, screening recs, prevention recs

and access to clinical trials.

Genetic Counseling

All patients with a family history concerning for a all

hereditary cancer syndromes or known

cancer mutation.

Visit includes taking a detailed history, family history and providing counseling on

genetic testing with genetic testing as indicated.

Call 913-588-5568 for scheduling genetic counseling.

Undiagnosed Clinic

All patients with a concern

for:

Breast mass

Breast rash

Nipple discharge

Nipple inversion

Needs workup for breast

problem.

Visit includes diagnostic assessment with clinical exam,

any necessary imaging and biopsy if needed.

Breast Construction

(Plastic Surgery)Both cancer related and

cosmetic. Depth of experience working with breast cancer patients and closely with our breast

surgeons.

Survivorship Clinic

For anyone with a history of breast cancer.

Visit includes review of treatments and long term

monitoring of breast cancer and treatment

related toxicities as well as future screening and

prevention with access to trials.

Page 64: It takes a village to make a survivor

Comprehensive Breast Cancer Clinical

Research Team

64