Cancer Recurrence in TransATAC Prognostic Signatures for ... TransATAC... · This presentation is the intellectual property of the author/presenter. Contact [email protected] for

San Antonio Breast Cancer Symposium – December 6-10, 2016

This presentation is the intellectual property of the author/presenter. Contact [email protected] for permission to reprint and/or distribute.

Comprehensive Comparison of Prognostic Signatures for Breast Cancer Recurrence in TransATAC

Ivana Sestak1 Richard Buus2, Jack Cuzick1, Peter Dubsky3, Ralf Kronenwett4,

Sean Ferree5, Dennis Sgroi6, Catherine Schnabel7, Rick Baehner8, Elizabeth Mallon2, Mitch Dowsett2

1. Centre for Cancer Prevention, Queen Mary University of London, London, UK 2. Ralph Lauren Centre for Breast Cancer Research, Royal Marsden, London, UK

3. Klinik St. Anna, Luzern, Switzerland 4. Sividon Diagnostics, Cologne, Germany 5. NanoString Technologies, Seattle, USA

6. Massachusetts General Hospital, Boston, USA 7. bioTheranostics, San Diego, USA

8. GenomicHealth, Redwood City, USA Confidential - D

o Not Distribute



Disclosures • Dr. Sestak has received speaker’s fees from Myriad. • Dr. Dubsky discloses grant support from Agendia/Sividon/Myriad/NanoString,

advisory role for Amgen, speaker’s fees from Myriad/Sividon/Amgen. • Dr. Kronenwett discloses that he is an inventor on the EndoPredict patent,

employee and previous shareholder of Sividon Diagnotics GmbH, a Myriad company.

• Dr. Ferree discloses that he is an employee of and shareholder in NanoString Technologies.

• Dr. Schnabel discloses that she is an employee of and shareholder in bioTheranostics, Inc.

• Dr. Baehner discloses that he is an employee of and shareholder in Genomic Health, Inc.

• Dr. Dowsett has received fees for advisory board from Genoptix and for lectures from Myriad.

• Drs. Buus, Cuzick, Sgroi, and Mallon have no conflict of interest to declare. Confidential - Do Not Distribute



Background • Multigene expression profiles useful for additional information on

prognosis in HR+ breast cancer patients treated with endocrine therapy • Growing number of prognostic signatures:

• Clinical Treatment Score • Mammaprint • Rotterdam Signatures • Genomic Grade Index • IHC4 • Oncotype Dx Recurrence Score • Breast Cancer Index • Prosigna • EndoPredict

• Important questions: 1. Comparative accuracy for prognosis Chemotherapy needed? 2. Prediction of late recurrence Extended endocrine therapy needed? Confidential - D

o Not Distribute



Aims 1. Prognostic performance of six signatures for distant recurrence in N- and N+ separately in transATAC:

2. Added prognostic value of signatures to clinical variables 3. Clinically useful risk groups

In years 0-10 (need for chemotherapy)

In years 5-10 (need for extended endocrine therapy)

Confidential - Do Not Distribute



Prognostic signatures Signature Information included

Clinical Treatment Score (CTS) Nodal status, grade, tumour size, age, treatment

Immunohistochemical markers (IHC4) ER, PgR, Ki67, HER2

Oncotype Recurrence Score (RS) 21 genes (oestrogen, proliferation, invasion, HER2 genes)

Breast Cancer Index (BCI) H/I and 5 proliferation genes (Molecular Grade Index)

Prosigna (ROR) 46 genes, proliferation score, tumour size (EU cut-offs from transATAC for N- and N+)

EndoPredict (EPclin) 8 genes (proliferation, differentiation, oestrogen); nodal status and tumour size Confidential - D

o Not Distribute



Prognostic signatures Signature Information included

Clinical Treatment Score (CTS) Nodal status, grade, tumour size, age, treatment

Immunohistochemical markers (IHC4) ER, PgR, Ki67, HER2

Oncotype Recurrence Score (RS) 21 genes (oestrogen, proliferation, invasion, HER2 genes)

Breast Cancer Index (BCI) H/I and 5 proliferation genes (Molecular Grade Index)

Prosigna (ROR) 46 genes, proliferation score, tumour size (EU cut-offs from transATAC for N- and N+)

EndoPredict (EPclin) 8 genes (proliferation, differentiation, oestrogen); nodal status and tumour size Confidential - D

o Not Distribute



Statistical analysis • 818 postmenopausal women with ER+/HER2-negative disease

• 5 years of tamoxifen or anastrozole, NO chemotherapy

• 10 year median follow-up

• Distant recurrence (DR) primary endpoint

• Cox regression models used to determine prognostic value (LR-χ2)

• Pre-defined cut-offs used to determine 10 year DR risk

All results presented for node-negative and node-positive patients separately Confidential - Do Not Distribute



Patient characteristics Node-negative

(N=591) Node-positive

(N=227)

Mean age, years (SD) 63.4 (7.9) 67.2 (8.2)

Mean BMI, kg/m2 (SD) 27.3 (4.9) 27.1 (5.0)

Grade

1 23.2% 18.9%

2 59.7% 61.2%

3 17.1% 19.8%

Mean tumour size, mm (SD) 17.6 (8.5) 25.7 (13.6)

Distant recurrence

0-10 years 60 (10.2%) 66 (29.1%)

5-10 years 34 (5.7%) 31 (13.7%) Confidential - Do Not Distribute



31.8

30.6

43.8

22.8

50.8

40.6

Likelihood Ratio χ2

EPclin

IHC4

CTS

ROR

RS

BCI

Prognostic value years 0-10 – node-negative

17.1

22.5

10.6

23.7

15.2

CTS

CTS

CTS

CTS

CTS

Likelihood Ratio ∆χ2

EPclin

IHC4

ROR

RS

BCI

% Improvement

53.8%

70.8%

33.3%

74.5%

47.8% Confidential - Do Not Distribute



40.2

6.3

9.6

6.4

15.5

35.6


EPclin

IHC4

CTS

ROR

RS

BCI

Prognostic value years 0-10 – node-positive

4.8

5.2

5.0

6.0

8.5

CTS

CTS

CTS

CTS

CTS


EPclin

IHC4

ROR

RS

BCI

% Improvement

11.9%

12.9%

12.3%

14.9%




Dis

tant

recu

rren

ce fr

ee (%

)

0 2 4 6 8 10 Follow-up time [years]

100

80

60

BCI

EPclin

100

80

60

3.9% 19.3% 27.3%

6.6%

22.1%

61.8% patients

72.6% patients

27.4% patients

24.2% patients 14.0% patients

100

80

60

ROR 3.0% 14.1% 33.4%

53.8% patients 30.1% patients 16.1% patients

100

80

60

RS

16.7% 27.2%

5.9% 63.3% patients 26.4% patients 10.3% patients

DR free (%) in years 0-10 – node-negative DR risk % Patients % in risk groups




Dis

tant

recu

rren

ce fr

ee (%

) 23.8% 33.1% 50.6%

5.6%

37.2%

49.3% patients

18.9% patients

81.1% patients



100

80

60

BCI

0.0% 20.7%

39.1%

6.6% patients


100

80

60

ROR

34.7% 48.8%


100

80

60

RS

100

80

60

EPclin

DR free (%) in years 0-10 – node-positive DR risk % Patients % in risk groups




Dis

tant

recu

rren

ce fr

ee (%

) 23.8% 33.1% 50.6%

5.6%

37.2%

49.3% patients

18.9% patients

81.1% patients



100

80

60

BCI

0.0% 20.7%

39.1%

6.6% patients


100

80

60

ROR

34.7% 48.8%


100

80

60

RS

100

80

60

EPclin





16.6

6.6

19.5

3.4

31.3

24.0


EPclin

IHC4

CTS

ROR

RS

BCI

Prognostic value years 5-10 – node-negative

3.3

11.2

1.9

18.4

10.3

CTS

CTS

CTS

CTS

CTS


EPclin

IHC4

ROR

RS

BCI

% Improvement

20.0%

67.5%

11.4%

111.0%




16.0

1.0

3.1

1.1

7.3

14.9


EPclin

IHC4

CTS

ROR

RS

BCI

Prognostic value years 5-10 – node-positive

1.2

1.8

1.1

4.1

4.4

CTS

CTS

CTS

CTS

CTS


EPclin

IHC4

ROR

RS

BCI

% Improvement

7.5%

11.3%

6.9%

25.6%




Dis

tant

recu

rren

ce fr

ee (%

)

Follow-up time [years] 5 6 7 8 9 10

80

90 1

00 EPclin

80

90 1

00 BCI

2.5% 14.4% 15.9%

4.3%

14.6%

63.6% patients

73.5% patients

26.5% patients


80

90 1

00 ROR

1.4%

10.0%

23.2%

54.6% patients

30.8% patients

14.6% patients

80

90 1

00 RS

9.6% 16.1%


DR free (%) in years 5-10 – node-negative DR risk % Patients % in risk groups




Dis

tant

recu

rren

ce fr

ee (%

)


14.3% 19.7% 36.5%

3.3%

23.6%

51.7% patients

22.0% patients

78.0% patients

32.4% patients 15.9% patients 60

80

10

0 BCI

0%

25.0% 13.0%

8.2% patients


60

80

100 ROR

19.5% 27.5%


60

80

100 RS

60

80

100 EPclin





Dis

tant

recu

rren

ce fr

ee (%

)


14.3% 19.7% 36.5%

3.3%

23.6%

51.7% patients

22.0% patients

78.0% patients

32.4% patients 15.9% patients 60

80

10

0 BCI

0%

25.0% 13.0%

8.2% patients


60

80

100 ROR

19.5% 27.5%


60

80

100 RS

60

80

100 EPclin





Conclusions • Unique cohort with well annotated samples, mature clinical

outcomes, and prognostic information from six signatures

• Prediction of recurrence in years 0-10: • Node-negative:

• All signatures good predictors for DR and identify patients with a low DR risk

no need for chemotherapy

• Node-positive: • ROR/EPclin identify low risk patients with good DR risk no need for chemotherapy




Conclusions II • Prediction of recurrence in years 5-10:

• Node-negative: • BCI, ROR and EPclin good predictors for late DR (above and beyond CTS) • All signatures identify low risk patients with a good late DR risk no need for extended endocrine therapy

• Node-positive:

• ROR/EPclin identify patients at low risk of late DR no need for extended endocrine therapy

• Incorporation of clinical variables important




Acknowledgements ATAC patients

TransATAC investigators and pathologists LATTE Steering Committee


Documents

Cancer Recurrence in TransATAC Prognostic Signatures for ... TransATAC... · This presentation is the intellectual property of the author/presenter. Contact [email protected] for