therapeutic strategies for patients with localized triple

Therapeutic Strategies for Patients with Localized Triple-Negative Breast Cancer (TNBC)

Sara Hurvitz, MDProfessor of Medicine

David Geffen School of Medicine at UCLADirector, Breast Cancer Clinical Research Program

Co-Director, Santa Monica-UCLA Outpatient Oncology PracticeSanta Monica, California

88%83%

T1aN0

85%80%

T1bN0

82%

75%

T1cN0 T2N0(3cm)

82%

75% 76%

67%

T2N0(4cm)

73%

63%

T2N0(5cm)

T2N0

77%

69%

T1cN1(1LN)

73%

62%

T1cN1(2LN)

69%

57%

T1cN1(3LN)

T1N1T1N0100%

20%

40%

60%

80%

100%

10-Y

ear O

vera

ll Su

rviv

al

10y Survival as per PREDICT NHS, 50y, G2 IDC, ER/HER2/negative, Ki67+

Do all patients with TNBC need chemotherapy?

(Neo)adjuvantchemotherapy

adjuvantchemo

?

Survival with/without chemotherapy

88%83%

T1aN0

85%80%

T1bN0

82%

75%

T1cN0

100%

20%

40%

60%

80%

100%

10-Y

ear O

vera

ll Su

rviv

al

10y Survival as per PREDICT NHS, 50y, G2 IDC, ER/HER2/negative, Ki67+

T1N0

OS 15 years % pts

sTILs < 30% 59% (53-65) 51%

sTILs 30%-75% 76% (69-84) 26%

sTILs ≥ 75% 93% (88-98) 22%

sTIL prognostic in young T1N0 TNBC patients (n=481)

Independent prognostic factors:

- TILs HR 0.75- LVI HR 2.52

De Jong et al, ESMO 2020

Do all patients with TNBC need chemotherapy? Survival with/without chemotherapy

No further therapy

Chemo or Trial

OP Chemo

Adjuvant Treatment

OPChemo

Neoadjuvant Treatment

Historically if inoperable

Preoperative or Postoperative Chemotherapy in TNBC?

DownstagingFacilitates less

invasive surgery (BCS, SLN, TAD)

RCTs confirm same outcome with adjuvant/neo-

adjuvant therapy (eg NSABP18)

Metaanalysis: pCR predictive of good outcome

Articles

www.thelancet.com Published online February 14, 2014 http://dx.doi.org/10.1016/S0140-6736(13)62422-8 5

anthracycline-based and taxane-based regimens were done, with the exception of the NOAH16 and TECHNO22 trials, which were limited to patients with HER2-positive locally advanced or infl ammatory breast cancer. Patients in the NOAH16 trial were randomly assigned to preoperative chemotherapy with or without trastuzumab, and those in the TECHNO22 trial received trastuzumab and chemo therapy followed by 1 year of adjuvant trastuzumab. In the GeparQuattro trial,11,12 patients with HER2-positive tumours received trastu-zumab con comitantly with chemotherapy. Overall, 1087 (55%) of 1989 patients with HER2-positive tumours included in the pooled analysis did not receive 1 year of adjuvant trastuzumab because they were treated before adjuvant trastuzumab trials were reported. All

patients with hormone-receptor-positive tumours were supposed to receive at least 5 years of endocrine therapy. Breast cancer subtype was established by clinic-opathological criteria, such as hormone-receptor status, HER2 overexpression, and histological grade as assessed by local pathologists. The Ki-67 labelling index had not been routinely assessed in the included studies.

11 955 patients were included in our pooled responder analysis (fi gure 1). Baseline characteristics are shown in the appendix. Median age was 49 years (IQR 43–57). 7328 patients (61%) had T2 tumours, 482 (4%) had infl ammatory breast cancer (ie, T4d tumours), and 5487 (46%) had clinically involved lymph nodes. 3572 (30%) patients had hormone-receptor-negative breast cancer, and 1989 (17%) had HER2-positive

Figure 5: Association between pCR and event-free survival, by breast cancer subtypepCR=pathological complete response. HR=hazard ratio.

Number at riskpCR

No pCR

0

20

40

60

80

100

Even

t-free

survi

val (%

)

0 1 2 3 4 5 6 7 8 9

2702491

2442226

2241978

1841616

1131017

69658

21247

684

220

21

0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9

0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9

Hormone-receptor-positive, HER2-negative

1481838

1341653

1231493

1021236

55790

33517

10198

568

215

21

Hormone-receptor-positive, HER2-negative, grade 1/2

102528

92458

83376

71290

49173

30111

938

114

05

Hormone-receptor-positive, HER2-negative, grade 3

Number at riskpCR

No pCR

0

20

40

60

80

100

Even

t-free

survi

val (%

)

5861403

5271157

454918

371713

212436

120269

37106

433

23

11

HER2-positive

HR 0·49 (95% CI 0·33–0·71) HR 0·63 (95% CI 0·38–1·04) HR 0·27 (95% CI 0·14–0·50)

HR 0·39 (95% CI 0·31–0·50) HR 0·58 (95% CI 0·42–0·82) HR 0·25 (95% CI 0·18–0·34)

HR 0·24 (95% CI 0·18–0·33)

Number at riskpCR

No pCR

0

20

40

60

80

100

Even

t-free

survi

val (%

)

0 1 2 3 4 5 6 7 8 9

389768

349604

310429

250317

166198

88125

2950

1113

11

Time since randomisation (years)

Triple negative

247839

224723

194617

157484

91306

50198

1779

224

11

23


HER2-positive, hormone-receptor-positive

325510

293392

250269

205200

115111

6559

1922

26


HER2-positive, hormone-receptor-negative

pCR No pCR

EFS, HR 0.24OS, HR 0.16

Cortazar, Lancet 2014

Path CR ypT0/is ypN0

Residual Disease DFS in TNBC (n=296)

Adjuvant Capecitabine improves outcome

EFS, HR 0.58OS, HR 0.52

Masuda N, NEJM 2017

CreateX trial Capecitabine(6–8 cycles)

Observation

Stage I–IIIB breast cancer (ER+ or TNBC)

Neoadjuvant chemotherapy

(95% A + T)

Non pathCR

Survival in TNBC (n=296)

Masuda N, et al. N Engl J Med. 2017;376:2147–2159.; Tutt et al, NEJM 2021

Can postoperative treatment improve cure rates in patients with residual disease after preoperative chemo?

Yes, in TNBC Yes, in patients with gBRCA1/2 mutations

Recurrence-free Survival

Olympia trialOlaparib

(1 year)

Placebo(1 year)

Stage I–IIIB breast cancer

NACT Non pathCR

Adjuvant (T2 or N+)

OlympiA: 3-Year Invasive DFS

Tutt ANJ et al. N Engl J Med 2021;384:2394-405.

OlympiA: Distant Disease-Free Survival


3-year dDFS difference: 7.1%HR: 0.0.57, p < 0.001N = 1,836

OlympiA: Overall Survival

3-year OS difference: 3.7%HR: 0.68, p < 0.02N = 1,836


OlympiA: Summary of Adverse Events


pCR Rates without Platinum around 35%

von Minckwitz G, SABCS 2015. #S2-04; von Minckwitz G. Lancet Oncol. 2014; Castrellon AB, Oncol Rev 2017, Sikov, JCO 2015, Sikov, SABCS 2015 S2-05; Loibl, S, et al. Lancet Oncol. 2018.

41%

54%

37%

53%

CALGB 40603

Paclitaxel +/-Bevacizum.

Paclitaxel +/-Bevacizum.+Carboplatin

Paclitaxel 80 mg/m2 q1w x 12 + Carboplatin AUC 6 q3w x 4

AC x 4

AC x 4

SURGERY

GeparSixto

Paclitaxel +NPL Doxo

Paclitaxel +NPL Doxo +Carboplatin

Paclitaxel 80 mg/m2 q1w x18 + NPLD 20 mg/m2

q1w x18 + Carboplatin AUC 1.5 q1w

SURGERY

71.6%

76.5%

Ø Cb

+Cb

3a-EFS

0.84 (0.58-1.22)

Paclitaxel 80 mg/m2 q1w x 12 + Carboplatin AUC 6 q3w x 4

BrighTNess

Paclitaxel

Paclitaxel +Carboplatin

AC x 4

AC x 4

SURGERY

31%

58%

76.1%

85.8%

3a-DFS

Ø Cb

+Cb0.56 (0.33-0.96)

Addition of carboplatin increases pCR rate in TNBC to >50% with improvement of EFS/OS

Can we increase response to NACT? Role of Platinum for TNBC

68.5%

79.3%

Ø Cb

+Cb

4a-EFS

0.57 (0.36–0.91), P=0.02

Neoadjuvant Immunotherapy in early TNBC Phase 3 trials of Immunotherapy in Stage II/III TNBC

Placebo

Pembrolizumab

- PDL1+ (22C3 CPS1) 82%- N+ 52%, T3/4 26%- Carbo QW 41%

- PDL1+ (SP142≥1%) 53%- N+ 38% (34% Ate; 43% Pla)- T3/4 28%

IMpassion031

Nab-Paclitaxel + Placebo

Nab-Paclitaxel + Atezolizumab

accAC + Placebo

accAC + Atezolizumab

SURGERY

R1:1

n = 333

Keynote 522Paclitaxel +

Carboplatin + Placebo

Paclitaxel + Carboplatin +

Pembrolizumab

AC/EC + Placebo

AC/EC + Pembrolizumab

SURGERY

R2:1

n = 1174

Co-primary endpoints:• pCR (ypT0/Tis ypN0) • Event-free Survival

Co-primary endpoints:• pCR (ypT0/Tis ypN0) in ITT & PD-L1+

Schmid P, et al NEJM 2020; Mittendorf E, et al Lancet 2020.

Atezolizumab: 840 mg given IV q2w (neoadjuvant); 1200 mg IV q3w x 11 (adjuvant)Nab-paclitaxel: 125 mg/m2 given IV qw for 12 weeksDoxorubicin: 60 mg/m2 given IV q2w/Cyclophosphamide: 600 mg/m2 given IV q2w

Pembrolizumab: 200 mg given IV q3w Paclitaxel: 80 mg/m2 given IV qw for 12 weeks;Carboplatin: AUC5 q3w x 4 or AUC1.5 qw x 12Doxorubicin: 60 mg/m2 given IV q2w/Cyclophosphamide: 600 mg/m2 given IV q2w

Atezolizumab

Unblinded F/U

Cross-trial comparison between Keynote 522 and IMpassion031 should be avoided

Addition of CIT significantly improves pCR in ITT Population

Neoadjuvant CIT in TNBC: Pathological complete response

0

10

20

30

40

50

60

70

80

90

100

ypT0/Tis ypN0

pCR,

% (9

5% C

I)

64.8%

51.2%

Δ 13.6%P=0.00055

260/401 103/201

Pembro + Chemo Placebo + Chemo

Keynote 522

0

10

20

30

40

50

60

70

80

90

100

pCR

(95%

CI)

(%)

pCR (ypT0/is ypN0)

57.6%

41.1%

95/165 69/168

∆ 16.5% P = 0.0044*

Atezo + Chemo Placebo + Chemo

Impassion 031

Schmid P, et al. ESMO 2019, Schmid, et al NEJM 2020, Harbeck et al, ESMO 2020This presentation is the intellectual property of the presenter. Contact [email protected] for permission to reprint and/or distribute

aHazard ratio (CI) analyzed based on a Cox regression model with treatment as a covariate stratified by the randomization stratification factors. bPrespecified P-value boundary of 0.00517 reached at this analysis. cDefined as the time from randomization to the data cutoff date of March 23, 2021.

84.5%

76.8%

Median follow-upc: 39.1 mo

Events HR (95% CI) P-value

Pembro + Chemo/Pembro 15.7% 0.63a

(0.48-0.82)

0.00031b

Pbo + Chemo/Pbo 23.8%

Addition of CIT significantly improves EFS in ITT population

Neoadjuvant CIT in TNBC: Event-free Survival

Schmid et al, ESMO 2021

Presenter: Peter Schmid, MD Content of this presentation is copyright and responsibility of the author. Permission is required for re-use.

0.1 1 10

Overall 123/784 (15.7) 93/390 (23.8) 0.63 (0.48 to 0.82)

Pembro + Chemo/Pembro Pbo + Chemo/PboHazard Ratio

(95% CI)

No. Events/No. Patients (%)

Positive 80/408 (19.6) 57/196 (29.1) 0.65 (0.46 to 0.91)Negative 43/376 (11.4) 36/194 (18.6) 0.58 (0.37 to 0.91)

T1/T2 64/581 (11.0) 59/290 (20.3) 0.51 (0.36 to 0.73)T3/T4 59/203 (29.1) 34/100 (34.0) 0.84 (0.55 to 1.28)

Every 3 weeks 50/334 (15.0) 37/167 (22.2) 0.65 (0.42 to 0.99)Weekly 71/444 (16.0) 56/220 (25.5) 0.60 (0.42 to 0.86)


<65 years 103/700 (14.7) 79/342 (23.1) 0.61 (0.45 to 0.82)³65 years 20/84 (23.8) 14/48 (29.2) 0.79 (0.40 to 1.56)

0 101/678 (14.9) 80/341 (23.5) 0.60 (0.45 to 0.80)1 22/106 (20.8) 13/49 (26.5) 0.81 (0.41 to 1.62)

Nodal status

Tumor size

Carboplatin schedule

PD-L1 status

Age category

ECOG PS

Subgroup

FavorsPbo + Chemo/Pbo

FavorsPembro + Chemo/Pembro

pCR by Nodal Status

EFS in Subgroups

0

10

20

30

40

50

60

07

80

90

100

Negative Positive

pCR

, % (9

5% C

I)

64.8%

44.1%

Δ 6.3%

58.6%

Δ 20.6%

64.9%

Nodal Status and Benefit from Immunotherapy

N0 and N+ patients have similar pCRwhen treated with CIT

N0 and N+ patients have similar EFS Benefit when treated with CIT


0.1 1 10

Overall 123/784 (15.7) 93/390 (23.8) 0.63 (0.48 to 0.82)

Pembro + Chemo/Pembro Pbo + Chemo/PboHazard Ratio

(95% CI)

No. Events/No. Patients (%)


T1/T2 64/581 (11.0) 59/290 (20.3) 0.51 (0.36 to 0.73)T3/T4 59/203 (29.1) 34/100 (34.0) 0.84 (0.55 to 1.28)

Every 3 weeks 50/334 (15.0) 37/167 (22.2) 0.65 (0.42 to 0.99)Weekly 71/444 (16.0) 56/220 (25.5) 0.60 (0.42 to 0.86)


<65 years 103/700 (14.7) 79/342 (23.1) 0.61 (0.45 to 0.82)³65 years 20/84 (23.8) 14/48 (29.2) 0.79 (0.40 to 1.56)

0 101/678 (14.9) 80/341 (23.5) 0.60 (0.45 to 0.80)1 22/106 (20.8) 13/49 (26.5) 0.81 (0.41 to 1.62)

Nodal status

Tumor size

Carboplatin schedule

PD-L1 status

Age category

ECOG PS

Subgroup

FavorsPbo + Chemo/Pbo

FavorsPembro + Chemo/Pembro

pCR by PD-L1 Score

0

10

20

30

40

50

60

70

80

90

100

CPS <1

pCR

, % (9

5% C

I)

30.3%

45.3%

CPS ≥1 CPS ≥10

68.9%

54.9%

77.9%

59.8%Δ 18.3%

Δ 14.2%Δ 17.5%

EFS in Subgroups

PD-L1 Status and Benefit from Immunotherapy

PD-L1+ and PD-L1- patients have similar EFS and pCR Benefit when treated with CIT

Distant Recurrence-Free Survival

aHazard ratio (CI) analyzed based on a Cox regression model with treatment as a covariate stratified by the randomization stratification factors. Data cutoff date: March 23, 2021.

87.0%

80.7%

Events HR (95% CI)

Pembro + Chemo 12.8% 0.61a

(0.46-0.82)Pbo + Chemo 20.3%

89.7%

86.9%

Events HR (95% CI)

P-value

Pembro + Chemo 10.2% 0.72a

(0.51-1.02) 0.03214bPbo + Chemo 14.1%

Neoadjuvant CIT in TNBC: Distant RFS and Overall SurvivalOverall Survival


108/201

94.4%

92.5%

56.8%

67.4%

pCR Yes

pCR No


Implications KN522:1. Presence of pCR benefit

predictive of long-term outcome2. Absence of pCR benefit does not

rule out substantial benefit

Does pCR-benefit with CIT translate into survival benefit?Is there a benefit beyond achieving a pCR? Event-free Survival by pCR

KEYNOTE-522 Study: EFS(IA4)

Neoadjuvant CIT: Treatment-related side effects

Schmid P, et al. ESMO 2019, Schmid, et al NEJM 2020

0

10

20

30

40

50

60

70

80

90

100

Inci

denc

e, %

Nause

a

Alo

pecia

Anem

ia

Neutropenia

Fatig

ue

Dia

rrhea

ALT

↑

Vom

iting

Ast

henia

Const

ipatio

n

Rash

Neutrophil

count ↓

1-2Grade

3-5

Pembro + Chemo

Placebo + Chemo

Periphera

l

neuro

path

y

Treatment-Related AEs With Incidence ≥20%

62.7 63.260.3

56.655.1 55.3

46.7 47.0

41.1

37.8

29.4

23.725.5 24.7 25.5

21.924.5 25.4

23.721.1

23.7

28.8

21.8

15.2

19.721.1

Treatment-Related AEs With Incidence ≥20%Keynote 522 Neodjuvant Phase

0

10

20

30

40

50

60

70

80

90

100

Inci

denc

e, %

Arthra

lgia

Rash

Ast

henia

Fatigue

Pru

ritu

s

Dia

rrhea

ALT

↑

Neutropenia

Nause

a

Anem

ia

Hyp

oth

yroid

-

ism

Lym

phopenia

AST ↑

7.96.7

4.92.2

3.3 3.8 3.3 3.84.22.2

3.72.5 2.6 2.9 2.2 1.6 2.2 1.3 2.0 1.9 1.6

2.91.6

2.5 1.6 2.5

Treatment-Related AEs With Incidence ≥2%Keynote 522 Adjuvant Phase

Response to Discussion:

Safety:1. No new safety signals were identified2. AEs consistent with the known safety profiles3. The addition of pembrolizumab did not compromise

exposure to chemotherapy or increase the incidence of common chemotherapy-related toxicities


a1 patient from pneumonitis and 1 patient from autoimmune encephalitis. Considered regardless of attribution to treatment or immune relatedness by the investigator. Related terms included in addition to preferred terms listed. Data cutoff date: March 23, 2021.

Immune-Mediated AEs in Adjuvant Phase

Hyperth

yroidis

m

Hypophy

sitis

Severe

skin

reactio

ns

Infusi

on rea

ctions

Hypothy

roidis

mColit

is

Thyroid

itis

Adrenal

insuffi

ciency

Hepatitis

Pneumoni

tis

Immune-Mediated AEs and Infusion Reactions with Incidence ≥10 Patients

0

2

4

6

8

10

12

14

16

18

20

Inci

denc

e, %

18.0

11.6

15.1

5.7 5.7

1.0

5.2

1.82.6

0

2.21.5 2.0

1.3 1.9

0.31.7

0.81.4

0.8

1-2Grade

3-5Pembro + Chemo/Pembro

Pbo + Chemo/Pbo Pembro + Chemo/Pembro

(N = 783)

Pbo + Chemo/Pbo

(N = 389)Any grade 43.6% 21.9%Grade 3-5 14.9% 2.1%Led to death 0.3%a 0Led to discontinuation of any drug

10.9% 2.6%

Pembro + Chemo/Pembro

(N = 588)

Pbo + Chemo/Pbo

(N = 331)Any grade 10.2% 6.0%Grade 3-5 2.9% 0.3%Led to death 0.2%a 0Led to discontinuation 1.4% 0.3%

0

2

4

6

8

10

12

14

16

18

a1 patient from autoimmune encephalitis. Considered regardless of attribution to treatment or immune relatedness by the investigator. Related terms included in addition to preferred terms listed. Data cutoff date: March 23, 2021.

Immune-Mediated AEs and Infusion Reactions with Incidence ≥2 Patients

20

Inci

denc

e, %

2.93.6

1.91.2

1.9

01.2

0.60

0.9 0.6 0.5 0.3 0 0.3 0

Hyperth

yroidis

m

Myocard

itis

Severe

skin

reactio

ns

Infusi

on rea

ctions

Hypothy

roidis

mAdre

nal

insuffi

ciency

Pneumoni

tisColit

is

Response to Discussion:

Immune-Mediated AEs:

1. Higher incidence of immune-mediated AEs primarily driven by endocrinopathies and skin reactions

2. AEs mostly occurred during the neoadjuvant phase with a very low incidence during the adjuvant phase

3. AEs generally low-grade, and successfully managed with treatment interruption, steroid administration, and/or hormone replacement, underscoring the importance of early identification and intervention to minimize risk and ensure continued treatment benefit

4. Although some immune-mediated AEs may be irreversible, analyses from other cancer types support the long-term safety of pembrolizumab, with no signal for late toxicities

5. Additional follow-up will inform the long-term safety of this regimen

Neoadjuvant CIT: Immune-mediated side effectsImmune-Mediated AEs in Combined Phases

Newly diagnosed TNBC • T1c/T2 N1 or • T3N0 or• locally advanced

Carboplatin (AUC2) + nab-Paclitaxel (125 mg/m2) days 1&8 q3 weeks x 8

Neoadjuvant Chemo + anti-PDL1 in TNBC: NeoTRIP Study

Primary endpoint: Event-free survival at 5 yearsKey secondary endpoints: pCR rates (ypT0/TisypN0), safety, predictive markers

OPCarboplatin +

Nab-Paclitaxel +/-Atezolizumab

R 2:1 AC/EC x 4

40.8%

43.5%

Control (no CIT)

Immunotherapy

- PDL1+ 56%- Locally advanced 49%- N+ 87%, N2/3 29%- T3/4 43%

Gianni, L, SABCS 2019

CIT trial designs in Early TNBC

OPNeoadjuvant Chemo

Immunotherapy Immunotherapy +/-

Keynote 522, Impassion031, NSABP B-59/GBG96

- Optimal antigen release- Possible over-treatment (pCR with CT)- pCR with CIT predictive of EFS?- Contribution of adjuvant therapy unclear

OPNeoadjuvant Chemo Immunoth. If suboptimal

response

Immunotherapy +/-- Reduced antigen release- Selection of high risk group- No validated definition of subop response

OPNeoadjuvant Chemo

Immunotherapy Immunotherapy +/-

Impassion030

- No/minimal antigen release- Possible over-treatment - NACT increasingly standard

OPNeoadjuvant Chemo non-path CR Immunotherapy - No/minimal antigen release- Single CIT agent efficacy limited

A-BRAVE, SWOG

De-escalation strategiesStage I

Stratified by TILs?

Stage II/III

Role of Carboplatin

with CIT

Identify optimal

responders(TILs? PDL1?)

OPChemo-free combinations?

OPChemo + CIT

Path CR ypT0/is ypN0

Residual Disease

Chemo in non pathCR?

Tailor strategies Chemo + CT?

ADC + CIT?

Role of adjuvant CIT?

Future directions in early TNBC

Based on personal communication from Prof. P. Schmid, Barts Cancer Institute.

therapeutic strategies for patients with localized triple

Documents