Neoadjuvant Endocrine Therapy for Hormone ReceptorPositive Breast Cancer

1
Neoadjuvant Endocrine Therapy for Hormone
Receptor-Positive Breast Cancer

• Hematology / Oncology Grand Rounds

2
Financial Disclosure

• None to disclose

3
Overview

• Breast cancer and rationale for neoadjuvant
  endocrine therapy
• Third generation AIs in neoadjuvant endocrine
  therapy
• Letrozole P024, IMPACT, and PROACT
• Endocrine vs chemo
• Who benefits?
• Estrogen receptor the Allred score
• Aromatase expression
• Her-2
• Neoadjuvant endocrine therapy Predictive tool?
• Ki-67
• PEPI score
• Unanswered questions and future directions
• References

4
Overview Breast Cancer

• Approximately 182,460 American women are
  diagnosed with breast cancer each year, and
  40,480 die from the disease
• Approximately half of all cases occur in women
  aged 65 years and older
• Approximately 80 of tumors arising in
  postmenopausal patients express hormone receptors
• Traditionally treated initially with surgery,
  with or without radiotherapy, and subsequent
  adjuvant therapy (endocrine therapy or
  chemotherapy) as indicated
• Neoadjuvant systemic chemotherapy has been used
  in locally advanced breast cancer to improve
  resection options

5
Agents for Endocrine Modulation

• Selective estrogen receptor modulators
• Tamoxifen (Nolvadex)
• Raloxifene (Evista)
• Toremifene (Fareston)
• Third Generation Non-steroidal AI
• Letrozole (Femara)
• Anastrozole (Arimidex)
• Third Generation Steroidal AI
• Exemestane (Aromasin)
• Pure estrogen receptor down-regulator
  (antagonist)
• Fulvestrant (Faslodex)
• LHRH agonists
• Leuprolide (Lupron)
• Groserelin (Zoladex)

6
Rationale for Neoadjuvant Endocrine Therapy in
Breast Cancer

• Early studies comparing tamoxifen alone to
  surgery with or without tamoxifen demonstrated
  the effects of tamoxifen as a primary treatment
  option for breast cancer
• Neoadjuvant endocrine therapy with third
  generation AIs is emerging as an alternative for
  postmenopausal women with HR positive breast
  cancer
• Potential of neoadjuvant endocrine therapy trials
  was first emphased by Eiermann et al in the
  letrozole P024 trial
• Other major neoadjuvant endocrine therapy trials
  include the IMPACT (Immediate Preoperative
  Anastrozole, Tamoxifen, or Combined with
  Tamoxifen) trial and the PROACT (Preoperative
  Arimedex Compared with Tamoxifen)

7
Rationale for Neoadjuvant Endocrine Therapy in
Breast Cancer

• Attractive features of neoadjuvant endocrine
  therapy for HR breast cancer
• Well-tolerated toxicity profile can potentially
  be used in the peri-operative period
  (chemotherapy may not be feasible in the
  peri-operative setting)
• Improve the likelihood of breast conservation
  surgery or to make an inoperable tumor resectable
• As a research tool, may allow in-vivo assessment
  of tumor response and molecular changes induced
  by treatment
• Development of molecular biomarkers to predict
  long-term outcomes, allow for risk stratification
  and individualization of therapy
• Elucidation of the biologic basis of estrogen
  pathway-targeting agents
• Drug development -gt can surrogate endpoints in
  neoadjuvant trials predict long-term outcome in
  adjuvant trials

8
Neoadjuvant Endocrine Therapy
The 3rd Generation AI's
9
Letrozole P024 Trial

• Multinational (16 countries) Phase IIb-III study
  enrolling patients from March 1998 to August 1999
• Double-blind, double-dummy, randomized,
  parallel-group study comparing letrozole 2.5 mg
  daily to tamoxifen 20 mg daily for four months
• Post-menopausal women with ER and/or PgR positive
  (10 nuclear staining) primary invasive breast
  cancer
• Untreated primary breast cancer confirmed by core
  biopsy
• Clinical stage T2-4a-c, N0-2, M0 tumors
  measurable by clinical exam, mammography and
  ultrasound
• Tumors are considered in operable or not amenable
  to BCS (breast conserving surgery)
• Patients with smaller tumors (i.e. T2) required a
  comment from the investigator to explain why they
  were not candidates for BCS (mostly non-favorable
  tumor breast size ratio)

10
Letrozole P024 Trial

• Correlative study component tumor biopsies /
  blood samples acquired prior to therapy and at
  the end of treatment
• Endpoints
• Primary efficacy endpoint Clinical response rate
  (CRPR by clinical exam)
• Secondary efficacy endpoints
• Percentage of patients who underwent BCS
• Response rate (CRPR) by mammography at 4 months
• Response rate (CRPR) by ultrasound at 4 months
• Powered to detect a difference of 15 between
  letrozole and tamoxifen at a 5 significance
  level (2-sided) with 80 power
• Response rate of 65 was anticipated in the
  tamoxifen arm
• Intention-to-treat analysis

11
Letrozole P024 Trial
Postmenopausal women with HR-positive breast
cancer N 337
13 patients (3) excluded, including 9 from 2 GCP
non-compliant centers, 3 patients w/o evidence of
breast cancer, and one who never took study meds
Randomization
Tamoxifen Arm Tamoxifen 20 mg daily x 4 mo N
170 (175 randomized)
Letrozole Arm Letrozole 2.5 mg daily x 4 mo N
154 (162 randomized)
23 discontinued, PD in 13
41 discontinued, PD in 21
Surgical resection in 82 BCS in 35
Surgical resection in 88 BCS in 45
12
Letrozole P024 Trial

• None of the patients were candidates for BCS at
  baseline and 14 were considered inoperable (20
  in letrozole vs 24 in tamoxifen)
• 50 patients (15) did not undergo resection
  following study meds (18 tamoxifen group and 12
  letrozole group)
• 20 with progressive disease, 6 were inoperable at
  baseline and remained inoperable, 13 refused
  surgery, 5 with contraindications, 6 withdrawn
  from study
• Treatment post-surgery is at the discretion of
  the investigator
• Letrozole was superior to tamoxifen
• Overall response (clinical) 55 vs 36
  (plt0.001)
• Overall response (US) 35 vs 25 (p 0.042)
• Overall response (Mammo) 34 vs 16 (p lt0.001)
• Rate of BCS 45 vs 35 (p 0.022)
• Similar rates of adverse events in both arms -
  adverse events mostly involved hot flashes (20
  vs 24 in letrozole vs tamoxifen)

13
Letrozole P024 Trial

• 5-year follow up for disease recurrence and
  survival
• In P024, ER/PR positivity was determined at study
  institution for the purposes of enrolment
• Study biopsies for tumor bank analyzed centrally
• No differences in the 2 arms in the frequency of
  ER and PgR expression
• Majority of tumors express ER in central
  analysis, but 12 were ER- and most of these were
  PgR- as well
• In view of discrepancies between study
  institution and central lab, supportive analysis
  was conducted that restricted the analysis of
  trial outcomes to study biopsy-confirmed ER and
  or PgR cases
• Response rate to neoadjuvant letrozole vs
  tamoxifen was 60 vs 41 ( p 0.004) and BCS
  rate was 48 vs 36 (p 0.036) respectively

14
IMPACT Anastrozole vs Tamoxifen

• Multi-centered randomized trial conducted in the
  UK and Germany
• Postmenopausal women with ER, invasive,
  non-metastatic, and operable or locally-advanced
  potentially operable breast cancer
• Tumor considered ER if more than 1 staining
  nuclei
• Neoadjuvant twin of the ATAC study to determine
  whether surrogate endpoints of 1) 12-week end
  point clinical response, and 2) 2-week biologic
  endpoint of change in proliferation index as
  assessed by Ki-67 staining
• Randomized 111 to three arms anastrozole
  alone, tamoxifen alone, or combination of
  anastrozole and tamoxifen, for 3 months
  pre-operatively
• Initial plans were for patients to continue on
  study medication for 5 years, but once results of
  ATAC trial became available, protocol amendment
  would be made to unblind the subjects, who would
  continue on single agent with either anastrozole
  or tamoxifen

15
IMPACT Anastrozole vs Tamoxifen

• Patients undergoing BCS and some patients with
  positive lymph nodes after mastectomies were
  given post-operative radiation
• Patients lt 70 years old were offered adjuvant
  chemotherapy if involved nodes or had other
  pathologic indicators of high risk disease at
  surgery
• All treatments were well-tolerated, but no
  differences in objective response as measured by
  caliper or ultrasound
• Treatment with anastrozole resulted in a BCS rate
  of 44 compared to 31 for tamoxifen (not
  statistically significant)
• Short-term clinical response in intent-to-treat
  population did not predict ATAC trial outcome

16
IMPACT Trial
Postmenopausal women with HR-positive breast
cancer N 330
Randomization
Anastrozole 1 mg placebo x 3 mo N 113
Tamoxifen 20 mg placebo x 3 mo N 108
Tamoxifen 20 mg anastrozole 1 mg x 3 mo N 109
Caliper OR 37 US OR 24 M to BCS 44
Caliper OR 36 US OR 20 M to BCS 31
Caliper OR 39 US OR 28 M to BCS 24
17
PROACT Anastrozole vs Tamoxifen

• Multinational study enrolling 451 patients from
  August 2000 to September 2002
• Double-blind, double-dummy, randomized,
  parallel-group study comparing anastrozole 1 mg
  daily plus tamoxifen placebo to tamoxifen 20 mg
  daily plus anastrozole placebo
• Post-menopausal women with ER and/or PgR positive
  (10 nuclear staining) primary invasive breast
  cancer with large operable, or potentially
  operable tumors T2-4b, N0-2, M0
• Patients are allowed to receive concomitant
  chemotherapy and both chemotherapy/radiotherapy
  after surgery if appropriate
• 137 patients received chemotherapy
• Eligible patients were to receive surgery at 3
  months, and can continue on study medications for
  up to 5 years or until recurrence

18
PROACT Trial
Postmenopausal women with large operable or
potentially operable HR-positive breast cancer
N 451 (386 inoperable at baseline or require
mastectomies)
Randomization
Tamoxifen 20 mg daily x 3 mo Anastrozole
placebo /- chemo N 223
Anastrozole 1 mg daily x 3 mo Tamoxifen placebo
/- chemo N 228
26 discontinued in pre-op phase
22 discontinued in pre-op phase
Feasible surgery in 66 of 184 (35.9) Actual
surgery improved in 55 of 184 (29.9)
Feasible surgery in 83 of 202 (41.1) Actual
surgery improved in 77 of 202 (38.1)
19
PROACT Anastrozole vs Tamoxifen

• Of 451 randomized patients, 386 would have either
  required a mastectomy or were considered
  inoperable at baseline
• 202 in anastrozole group and 184 in tamoxifen
  group
• In intent-to-treat population of patients
  assessed at baseline to be inoperable or
  requiring a mastectomy, improvement in feasible
  surgery at baseline vs actual surgery in
  anastrozole group was 38.1 vs 29.9
• In the endocrine therapy only patients, more
  patients treated with anastrozole alone had an
  improvement in surgery of 43.0 compared to 30.8
  for the tamoxifen group

20
Neoadjuvant Chemotherapy vs Endocrine Therapy

• Semiglaznov et al published a randomized,
  controlled, phase 2 study evaluating neoadjuvant
  chemotherapy to endocrine treatment with
  aromatase inhibitors
• Post-menopausal women with untreated,
  histologically confirmed, primary invasive breast
  cancer, ER and/or PgR, and life expectancy gt 6
  months
• Chemo arm received doxorubicin at 60 mg/m2 and
  paclitaxel 200 mg/m2 every 3 weeks x 4 cycles
• Endocrine arms received either exemestane 25 mg
  daily for 3 months or anastrozole 1 mg daily for
  3 months
• Surgery is scheduled for 3 months from the date
  of first treatment
• Treatment after surgery is at the discretion of
  the investigators -gt generally recommended
  tamoxifen for 5 years
• Patients were monitored for development of local
  recurrence, distant mets and survival at 5 years
  post surgery

21
Neoadjuvant Chemotherapy vs Endocrine Therapy

• Total of 239 women enrolled
• 118 received chemo, 60 received exemestane and 61
  received anastrozole
• Sample size chosen to provide 80 power to detect
  absence of a difference between the endocrine
  therapy and chemo groups at a 5 significance
  level (2 sided)
• Expected RR of 62 for AIs and 63 for chemo
• Clinical objective response was 64 in both
  endocrine therapy and chemo groups
• Rates of pathologic CR and disease progression
  did not differ significantly between the two
  groups
• Rates of BCS were slightly higher in the
  endocrine group 33 vs 24
• Common chemo toxicities included alopecia (79),
  grades 3 or 4 neutropenia (33) and grade 2
  neuropathy (30)

22
Neoadjuvant Endocrine Therapy
Who benefits?
23
Role of ER and PgR Expression in the Letrozole
P024 Trial

• Clinical Response Rate vs ER Allred Score

• Clinical Response Rate vs PgR Allred Score

24
Aromatase Expression in Letrozole P024 Trial

• IHC assay for aromatase applied to samples from
  the P024 trial
• Aromatase expression by tumor and stromal cells
  was correlated with tumor response,
  treatment-induced changes in Ki67, RFS, and BCSS
• Tumor and stromal aromatase expression were
  highly correlated
• Tumor aromatase also correlated with smaller
  tumor at presentation and higher baseline ER
  Allred score
• Presence of tumor aromatase expression at
  baseline sample remained a favorable independent
  prognostic biomarker for both BCSS (HR 3.76, 95
  CI 1.4-10.0) and RFS (HR 2.3, 95 CI 1.2-4.6)

25
Role of Receptor Tyrosine Kinases Erb-B2 and
Erb-B1

• Ellis et al reported on the relationship between
  Erb-B2 (Her2/Neu) expression and primary tumor
  regression using samples from the letrozole P024
  trial, as this RTK was thought to be a marker for
  tamoxifen resistance and was included as part of
  the correlative studies for the P024 trial
• Erb-B1 (EGFR) was also included as it had been
  previously linked to endocrine resistance
• 15.2 of tumors were ER (central designation)
  and Erb-B1 and/or Erb-B2 positive in the P024
  trial
• In the P024 trial, tumors that are ErbB-1 and/or
  ErbB-2 positive have a lower response rate to
  tamoxifen, with a response rate of only 14,
  compared to 41 for ErbB-1 and ErbB-2 negative
  tumors

26
Role of Receptor Tyrosine Kinases Erb-B2 and
Erb-B1

• In contrast, tumors that are ErbB-1 and/or ErbB-2
  positive in the P024 trial have a response rate
  of 58 with letrozole, and tumors that are ErbB-1
  and ErbB-2 negative have a response rate of 54
• In the IMPACT trial, 14 of assessed patients
  were HER-2 positive defined as 3 on IHC, or 2
  on IHC -gt confirmed by FISH
• Objective response for patients with HER-2
  positive tumors was 58 with anastrozole compared
  to 22 with tamoxifen

27
Neoadjuvant Endocrine Therapy
Predictive tool?
28
Ki-67 in Letrozole P024 Trial

• In the P024 trial, letrozole inhibited tumor
  proliferation as measured by Ki-67 to greater
  extent than tamoxifen
• Reduction in geometric mean Ki-67 level of 87 vs
  75
• Differences also observed in ER, HER-1 and/or
  HER-2 over-expressing tumors (88 vs 45)
  However,
• HER-2 FISH positive tumors showed higher
  histologic grade, higher pre-treatment Ki-67 and
  less Ki-67 suppression after letrozole compared
  to HER-2 neg
• More HER-2 FISH neg tumors met the definition of
  cell cycle CR at time of surgery (60 vs 12)
• Biomarkers discordant with clinical observations
  that tumor regression is unaffected by HER-2
  status in patients treated with neoadjuvant
  letrozole

29
Ki-67 in IMPACT Trial

• As part of the design of IMPACT, IHC analysis of
  Ki-67 was performed at baseline, 2 weeks after
  treatment and at surgery
• Greater reduction in Ki-67 in anastrozole group
  than tamoxifen or combination groups at 2 weeks
  and 12 weeks, but no significant relationship
  between Ki-67 changes and clinical response
• However, post-treatment Ki-67 appears to
  correlate with RFS as shown on next slide

30
Ki-67 and RFS in IMPACT

• Recurrence-free survival according to tertiles
  of tumor Ki67 expression
• At baseline (top panel)
• After 2 weeks of treatment (bottom panel).
• Divisions refer to the natural logarithm of the
  percentage of Ki67-positive cells at 2 weeks

31
Outcome Prediction The PEPI Score

• Ellis et al analyzed tumors from 228
  post-menopausal women in the P024 trial for
  post-treatment ER status, Ki-67 proliferation
  index, histologic grade, pathologic tumor size
  and node status, and treatment response
• Cox proportional hazards used to identify factors
  that were associated with relapse-free survival
  (RFS) and breast cancer-specific survival (BCSS)
• Risk groups according to PEPI score
• Group 1 score of 0
• Group 2 score of 1-3
• Group 3 score of 4
• Validated in IMPACT trial

32
The PEPI Score
33
PEPI Score as Predictor of RFS and BCSS

• RFS and BCSS by risk group
• Green line group 1
• Red line Group 2
• Purple line Group 3
• Data from the P024 trial is shown on the left
• Data from the IMPACT trial (validation) is shown
  on the right
• Bottom - heat maps summarizing distribution of
  individual components of risk score

34
Neoadjuvant Endocrine Therapy
Many unanswered questions...
35
Unanswered Questions

• Which AI to use?
• ACOSOG Z1031
• Randomized phase III study post-menopausal
  women with ER-positive breast cancers (T2-T4c,
  any N, M0 disease)
• Allred score of 6, 7 or 8
• 3 different arms
• Letrozole 2.5 mg daily for 16 weeks
• Anastrozole 1 mg daily for 16 weeks
• Exemestane 25 mg daily for 16 weeks
• Goal is to identify one of these three
  neoadjuvant AI treatment in this patient
  population for use in a phase III study comparing
  neoadjuvant AI with neoadjuvant chemotherapy
• Neoadjuvant endocrine vs neoadjuvant chemotherapy
• How do you select patients who would benefit the
  most?
• How do the two options compare?
• What can we learn about the biologic basis of ER
  pathway targeting agents?