Paolo Marchetti

1
Chemioprevenzione
una nuova modalità di trattamento delle neoplasie
prima della loro insorgenza

• Paolo Marchetti
• Oncologia Medica
• Università di LAquila

2

• CANCER CHEMOPREVENTION
• A New Way to Treat Cancer Before It Happens
• Use of natural and/or synthetic compounds to
  prevent, inhibit or reverse carcinogenesis before
  the clinical event
• Lesson from cardiology drugs to correct risk
  factors have reduced mortality
• The disease is carcinogenesis (a process) rather
  than cancer (an event)

3
NeoplasiaProcess vs Event

• The recognition that carcinogenesis, not
  clinically evident cancer alone, is the disease
  of interest provides a compelling impetus
• for more sensitive measures of risk in
  asymptomatic populations,
• for identification and development of
  chemopreventive agents, and
• for more accurate and reliable tests of
  preventive efficacy.
• These elements can be combined to generate
  prevention trials that are more effective,
  efficient, and predictive of meaningful benefit.

4
The target of the hormonal therapy
Aromatase Activity
5
Mechanisms of Action of Hormonal Therapies

• Block oestrogen action
• tamoxifen, other SERMs
• Block oestrogen synthesis
• ovarian ablation (premenopausal)
• inhibition or inactivation of aromatase
  (postmenopausal)
• Other mechanisms?
• progestins
• androgens
• oestrogens

6
Selective Estrogen Receptor Modulators (SERMs)

• Estradiol
• Tamoxifen
• Toremiphene
• Droloxifene
• Idoxifene
• Raloxifene
• Faslodex (ICI 182,780)

• Pure agonist
• Mixed agonist/antagonist
• Pure antagonist

7
Breast Cancer Initiation and Promotion
NUMBER OF CELLS

INITIATION
PROMOTION
PREINVASIVE
INVASIVE
DCIS
ADH
EH
LIMIT OF CLINICAL DETECTION
HOST DEATH
Modified from Cavalieri et al., AACR 2002
8
Relative Risk of Breast Cancer in 238
Postmenopausal Women According to Serum Estradiol
Concentration
Reprinted from Clemons and Goss. N Engl J Med.
2001344276-285 Data from Thomas et al. Br J
Cancer. 199776401-405
9
ISSUES IN CANCER CHEMOPREVENTION

• at-risk cohorts
• surrogate biomarkers
• effective agents

10
AT-RISK SUBJECTS

• Genetic (BRCA1 and 2, BRCA?, p53)
• Pre-invasive breast lesions (DCIS, LCIS, AH)
• Previous breast cancer
• Gail Model

11
Breast Cancer Initiation and Promotion
NUMBER OF CELLS

INITIATION
PROMOTION
INVASIVE
PREINVASIVE
DCIS
ADH
EH
LIMIT OF CLINICAL DETECTION
HOST DEATH
Modified from Cavalieri et al., AACR 2002
12
Breast Cancer Initiation and Promotion
13
Pre-invasive breast lesions (DCIS, LCIS, AH)?
Gene expression profiles of human breast cancer
progression

• One surprising result from this study was the
  remarkable similarity in the expression profiles
  of the distinct pathological stages.

Xiao-Jun Ma et al., PNAS, 2003
14
Pre-invasive breast lesions (DCIS, LCIS, AH)?
Gene expression profiles of human breast cancer
progression

• As compared with patient-matched normal
  epithelium, significant global alterations in
  gene expression occur at ADH, the earliest
  phenotypically recognized stage of progression,
  and such alterations are maintained in the later
  stages of DCIS and IDC.

Xiao-Jun Ma et al., PNAS, 2003
15
Breast Cancer Initiation and Promotion
NUMBER OF CELLS

PROMOTION
INITIATION
INVASIVE
PREINVASIVE
DCIS
ADH
EH
LIMIT OF CLINICAL DETECTION
HOST DEATH
Modified from Cavalieri et al., AACR 2002
16
Breast cancer risk predictionGail model

• Age at menarche
• Age at first live birth
• Breast biopsies
• Presence of atypical hyperplasia
• Familial BC (mother/sisters)

17
SURROGATE ENDPOINT BIOMARKERS

• Precancerous lesions (LCIS,AH)
• Cytological atypia
• Circulating IGF-I
• Mx density

18
Plasma IGF- I and breast cancer risk
Cases/controls 397
breast cancers /620 age-matched controls
Time from blood collection to diagnosis
28 months (range 1-57)

• All, top vs bottom quintile of IGF-I

RR0.99 (0.65-1.50)

• Postmenopause, t vs b quintile

RR0.85 (0.53-1.39)

• Premenopause, t vs b tertile

RR2.88 (1.21-6.85)
RR7.28 (2.40-22.0)

• Premenopause lt50 yrs, t vs b tertile

adjusted for IGFBP-3 Hankinson SE et al, Lancet
1998
19
Effect of mammographic density on breast cancer
risk in prospective studies

• Author N. N. Adjusted 95 CI
• cases control density OR
• Saftlas 1991 67 58 45-65 3.8 2.1-3.6
• 45 33 gt65 4.3 2.1-8.8
• Boyd 1995 66 31 gt75 6.1 2.8-13
• Byrne 1995 194 136 gt75 4.4 3.1-6.1
• 576 554 50-74 2.8 2.1-3.6
• Kato 1995 37 99 (pre) gt66 3.6 1.7-7.9
• 48 81 (post) gt66 2.1 1.1-3.8

20
SERMs in Breast Cancer PreventionProven Efficacy
Trial Outcome NSABP B-14 ? Contralateral
cancers NSABP P1 ? invasive/preinvasive breast
cancer Royal Marsden No ? in risk Italian
Study Risk ? in HRT users only MORE Study ?
Invasive cancers IBIS 1 ? invasive/preinvasive
breast cancer STAR Ongoing IBIS 2 Ongoing
21
Chemoprevention trials
22
Breast cancer incidenceSurveillance and
Chemoprevention
23
DISCREPANCY OF RESULTS

• Compliance (26 women dropped out in ITPT).
• Study population.
• Duration of follow-up (average follow-up for
  NSABP-P1 was only 3.5 years compared with the
  median of nearly 6 years.
• Duration of tamoxifen treatment.

24
Controversial results of published prevention
studies.

• The two trials give insights concerning the rate
  of BC incidence in women selected according to
  epidemiological risk and familial risk,
  respectively, and it is equivalent in the placebo
  arms (0.46 and 0.48/100 women/y, respectively)
• Apparently, TAM shows beneficial effects in women
  selected mostly according to epidemiological
  criteria of risk than according to familial risk
  (0.23 and 0.46 /100 women/y, respectively)
• In the NSABP-P1 trial, mostly characterized by
  postmenopausal women, the advantage of the TAM is
  significant also in the subset of selected women
  showing familial risk of BC (0.48 vs 0.25 BCs/
  100 women/y in placebo and TAM arm,
  respectively)

25
Controversial results of published prevention
studies (ii).

• The TAM effect is clear also in the cohort with
  more consistent risk (gt3 familial BCs) of genetic
  predisposition (0.64 vs 0.35 BCs/ 100 women/y in
  placebo and TAM arm, respectively).
• The preliminary evaluation of the BRCA1/2 status
  of the 70 incidental BCs in RMH trial showed 7
  BRCA1/2 mutations only (10) in this very
  high-risk group due to FH and the occurrence of
  BC (Powles, 2nd European Breast Cancer
  Conference, Bruxelles00).

26
NSABP-P1Adverse effects of TMX

• Endometrial cancer (RR 2.5)
• Thromboembolic events
• pulmonary embolus
• deep vein thrombosis
• cerebral vascular accident

27
IBIS 1 Number of Breast Cancers and
Characteristics by Treatment Allocation
28
Discrepancies between P1 and IBIS 1

• The compliance rate in P-1 was 76 compared with
  64 in IBIS-1.
• The median follow-up in IBIS-1 was shorter (50 vs
  55 months) and a smaller proportion of patients
  (25 vs 37) reached the 5-year treatment
  threshold, creating a narrower window for
  observed effect.
• Whereas P-1 excluded women who were treated with
  hormone replacement therapy (HRT) within 3 months
  of the study, in IBIS-1, 40 of participants
  received HRT during the course of the trial.

Samuel Cykert, Lancet 2004
29
IBIS 1 a comment.

• Tamoxifen for breast-cancer prevention is a
  limited intervention that requires thoughtful
  decision-making. However, because of
  methodological and follow-up constraints, the
  benefits of this approach are probably
  underestimated and harms exaggerated in this
  report.
• Samuel Cykert, Lancet 2004

30
SERMs in Breast Cancer PreventionProven Efficacy
Overview of the four tamoxifen breast cancer
prevention trials (Lancet, 2003)

• The combined data showed a 38 reduction in
  breast cancer incidence (95 CI 2846, P0.001).
  
• There was an increase in endometrial cancer
  events (Relative Risk (RR) 2.4, CI 1.54.0) and
  thromboembolic events (RR 1.9, CI 1.42.6), but
  not in cardiovascular events.
• Overall, there was no significant effect on
  mortality (RR 0.91, CI 0.701.18).
• The data, together with the data from the Oxford
  metanalysis, indicate that it is very unlikely
  that mortality is increased by tamoxifen.

31
Low-dose tamoxifen and fenretinide
Placebo Placebo
Premenopausal women DCIS, LCIS
TAM 5mg P
R
2y
Gail gt 1.3 in 5 yrs
4-HPR 200mg P
TAM 4-HPR
I endpoint
? IGFs and Mx density
? endometrial and ovarian effects
II endpoint
? breast FNA (image analysis)
Sample size
300 subjects
32
The HOT (Hormone Replacement Therapy and
Tamoxifen) Study
HRT users Placebo/day R (de
novo or Tamoxifen 5 mg/day current
users)
Sample size 8500 subjects (4250 per
arm) Endpoint Breast cancer incidence (IBC
and DCIS)
33
Breast Cancer Prevention TrialsUnanswered
Questions.

• Type of preventive treatment
• Durability of the preventive effect
• Influence on mortality
• Subsets who benefit from treatment
• Interaction with HRT
• Preventive effect in BRCA1/2 carriers

34
Randomized phase III studiesof anti-Aromatase
Agents vs Tamoxifenas Initial Therapy of
Metastatic Breast Cancer
From EORTC 10951, ASCO 2004
35
Adjuvant Therapy Trials What Theyll Tell Us
About Anti-Aromatase Agents

• Efficacy in the adjuvant setting
• Long-term tolerability
• adverse events
• bone and lipid metabolism
• quality of life
• menopausal function
• Efficacy in prevention of contralateral breast
  cancer

36
Estimated reduction in contralateral breast
cancer in the ATAC trial
37
(No Transcript)
38
(No Transcript)
39
(No Transcript)
40
FINAL RESULTS OFA RANDOMIZED PHASE III
TRIALCOMPARING EXEMESTANE WITH TAMOXIFEN AS
FIRST-LINE HORMONE THERAPYFOR POSTMENOPAUSAL
WOMEN WITH METASTATIC BREAST CANCEREORTC
10951CONDUCTED BY THE EORTC BREAST GROUP in
collaboration withTHE EXEMESTANE WORKING
GROUPAND THE NCIC CLINICAL TRIALS GROUP
ASCO - New Orleans - june 5-8, 2004
41
PFS by Treatment Arm
Sensitivity test Wilcoxon p0.028
HR 0.84 95 CI (0.67, 1.05)
42
EORTC 10951 - PFS estimates

EXE TAM
182 patients
189 patients
158 events (87) 161 events
(85)
Median PFS (months)
9.9 5.8 (95
CI) (8.7 - 11.8)
(5.3 - 8.1) PFS at 6 months
66 49
(95 CI) (59 -
73) (42 - 57) PFS at 12 months
42
31 (95 CI)
(35 - 49) (24 - 38)
43
Considerations in the Selection of an
Anti-Aromatase Agent for Long-Term Therapy

• Efficacy
• there are no published direct comparisons of
  anti-aromatase agents, but trials are ongoing
• there is no evidence, to date, that one
  anti-aromatase agent is superior to another in
  efficacy
• Tolerability
• ?

44
Anti-aromatase AgentsThey Are Not All the Same

• Different estrogen-suppression potencies?
• Enzyme inactivation vs inhibition?
• Different effects or PK profile on tumour
  aromatase?
• Additional action though androgen receptor?

45
Working Mechanism of Action for Anti-aromatase
Agents
Type I inactivators (steroidal) (target the
substrate-binding site, formestane, exemestane)
Androgen
Oestrogen
Cytochrome P450
Aromatase molecule
Type II inhibitors (nonsteroidal) (target the
cytochrome P450 aromatase, aminoglutetimide,
fadrozole, anastrozole, letrozole)
NADP
NADPH
From Textbook of Breast Cancer. A Clinical Guide
to Therapy London, England. Martin Dunitz Ltd
1997281304
46
Bone Effects of Antiaromatase Agents Summary
1. ATAC Trialists Group, Lancet 2002. Vol 259
2131-39 2. Eastell et al. Annals of Oncology,
2002. 13 (Suppl. 5) page 32, Abs 113PD 3.
Harper-Wynne et al. Breast Cancer Res Treat.
200169225, 136 4. Heshmati et al. J Bone
Mineral Res. 2002 17 172-178 5. Goss et al.
Breast Cancer Res Treat. 2002 vol. 76, sup 1,
S76, Abs 267 (updated) 6. Johnell et al. J Clin
Endocrinol Metab 2002 Mar87(3)985-92
47
Effects of steroidal and nonsteroidal aromatase
inhibitors (AIs) on markers of bone turnover and
lipid metabolism in healthy volunteers. M. Subar,
P. E. Goss, T. Thomsen, J. Banke-Bochita.
AI Effects on Estrogens

• There were no significant differences noted in
  baseline estrogens among the 4 study arms.
• All 3 AIs effectively suppressed serum estrone.
• All 3 AIs effectively suppressed serum estradiol.

48
Effects of steroidal and nonsteroidal aromatase
inhibitors (AIs) on markers of bone turnover and
lipid metabolism in healthy volunteers. M. Subar,
P. E. Goss, T. Thomsen, J. Banke-Bochita.
49
Effects of steroidal and nonsteroidal aromatase
inhibitors (AIs) on markers of bone turnover and
lipid metabolism in healthy volunteers. M. Subar,
P. E. Goss, T. Thomsen, J. Banke-Bochita.
50
(No Transcript)
51
Clinical implications of familial breast cancer
screening
52
The problem.

• Women carrying BRCA1/2 mutations have a lifetime
  risk of BC of 56 and 80, respectively (Easton et
  , Am J H Gen95 Ford et al, Am J Hum Gen 98
  Struewing et al, NEJM 97).
• Both genes also confer an increased risk of
  ovarian cancer 60 for BRCA1 and 27 for BRCA2
  by the age of 80 years, respectively.
• Furthermore, it is now clear that both genes also
  confer an increased risk to develop different
  other cancer, equating to an absolute risk of
  6-14 (Ford, Lancet 94 Breast cancer Linkage
  Consortium, JNCI 99).

53
BRCA1/2 Breast Cancer Risk
Cumulative Risk
Age
Eeles and Powles, JCO Vol 18, 2000 pp93s-99s
54
BRCA1/2 Ovarian Cancer Risk
Cumulative Risk
Age
Eeles and Powles, JCO Vol 18, 2000 pp93s-99s
55
BRCA1/2 carriersPrevention strategies and risk
reduction

• Surveillance
• Prophylactic Surgery
• Hartmann et al., NEJM, 1999
• bPM 90 risk reduction.
• Meijers-Heijboer et al., NEJM, 2001
• in pts with BRCA1/2 mutations
• surveillance alone 8 BC/63
• bPM 0 BC/76
• Rebbeck et al., JNCI, 1999 PO 53 risk
  reduction.
• Breast Cancer Prevention

56
SECONDARY PREVENTION

• Although there were no significant differences
  between women with BRCA-associated breast cancer
  and those with non- BRCA-associated cancers in
• 5-year relapse-free survival (65 versus 69, P
  NS),
• 5-year event-free survival (57 versus 68, P
  NS),
• 5-year overall survival,
• women with germline BRCA mutations were
  significantly more likely to develop
  contralateral breast cancer at 5 years (31
  versus 4, P 0.0007).
• Robson et al, JCO 1998.

57
SECONDARY PREVENTION

• Breast cancer survivors who carry BRCA mutations
  remain at risk for second primary malignancies
  because of inherited susceptibility.
• What is the optimal clinical management for a
  young woman with a strong family history and a
  new diagnosis of unilateral breast cancer?
• Is breast conservation therapy an option?
• Is OOX necessary?
• What are the risks for secondary cancers?

58
BRCA1/2 carriersPrevention strategies and risk
reduction

• Surveillance
• Prophylactic Surgery
• Hartmann et al., NEJM, 1999
• bPM 90 risk reduction.
• Meijers-Heijboer et al., NEJM, 2001
• in pts with BRCA1/2 mutations
• surveillance alone 8 BC/63
• bPM 0 BC/76
• Rebbeck et al., JCO, 2004
• BPM reduces the risk of BC in women with BRCA1/2
  mutations by approximately 90.
• Rebbeck et al., JNCI, 1999 PO 53 risk
  reduction.
• Breast Cancer Prevention

59
Time to breast cancer diagnosis in female BRCA1
mutation carrierswith and without bilateral
prophylactic mastectomy (BPM).

• Kaplan-Meier analysis of breast cancer events by
  postsurgery follow-up time in cases compared with
  controls.
• (Rebbeck et al., JCO, 2004)

60
Bilateral prophylactic mastectomy

• We cannot make strong inferences about optimal
  type and timing of surgery or about risk factors
  that may influence post-bilateral prophylactic
  mastectomy breast cancers.

61
Bilateral prophylactic mastectomy

• Subcutaneous bilateral prophylactic mastectomy
  leaves substantial residual breast tissue intact,
  including the nipple-areolar complex and,
  therefore, is not optimal for a prophylactic
  procedure.
• Total mastectomy requires more extensive
  reconstruction and may result in an inferior
  cosmetic result, but it removes substantially
  more breast tissue.
• However, the recently developed skin-sparing
  mastectomy with immediate reconstruction combines
  adequate tissue removal with excellent cosmetic
  outcome.

62
Bilateral prophylactic mastectomy

• Regardless of the selected procedure, care should
  be taken to remove as much breast tissue as
  possible to maximize risk reduction.
• There are surgical and anesthetic risks that
  should be considered when offering prophylactic
  surgery to a healthy individual.

63
Bilateral prophylactic mastectomy risks

• In a recent series of 112 high-risk women (79
  with a BRCA1/2 mutation) who underwent
  prophylactic mastectomy (103 with immediate
  reconstruction), 21 had complications, including
  hematoma, infection, contracture, or implant
  rupture.
• Use of autologous tissue, such as with transverse
  rectus abdominis musculocutaneous (TRAM) or
  latissimus dorsi reconstruction, may eliminate
  the need for silicone implants, but complication
  rates may be even higher.
• In one series of 147 breast cancer patients with
  TRAM reconstruction after mastectomy, follow-up
  operations were necessary in 71 of patients,
  including intervention for complications such as
  abdominal hernia, full or partial TRAM ischemic
  loss, and fat necrosis.

64
BRCA1/2 carriersPrevention strategies and risk
reduction

• Surveillance
• Prophylactic Surgery
• Hartmann et al., NEJM, 1999
• bPM 90 risk reduction.
• Meijers-Heijboer et al., NEJM, 2001
• in pts with BRCA1/2 mutations
• surveillance alone 8 BC/63
• bPM 0 BC/76
• Rebbeck et al., JNCI, 1999 PO 53 risk
  reduction.
• Breast Cancer Prevention

65
Hormonal Preventive Effect in BRCA1/2 Carriers

• Early reports suggested that there is a loss of
  ER and PgR in tumors with BRCA1 mutations,
  whereas tumors with BRCA2 mutations are often ER
  positive1.
• The critical question is whether breast cancer
  prevention, specifically hormone-therapy, would
  also reduce incidence of invasive BC among
  cancer-free women with inherited BRCA1 or BRCA2
  mutations.

1Johannsson et al., Eur J Cancer 33 362-371 1997
66
Study participants who developed BC in 288
genotyped cases (NSABP-P1, JAMA, Nov 14, 2001)
Includes 288 genotyped cases and 32 cases without
DNA available
67
ER status of tumors (NSABP-P1, JAMA, Nov 14,
2001)
68
Preventive effect in BRCA1/2 carriers

• In BC treatment oophorectomy, tamoxifen, or
  anti-aromatase agents are effective.
• If oophorectomy, performed before 35 years, is
  effective in reducing BC incidence among women
  with BRCA1 mutations (Rebbeck et al., JNCI,
  1999), then TAM or anti-aromatase agents might be
  effective in cancer-free women with BRCA1
  mutations.
• It is possible that early in the course of BRCA1
  tumors, hormone-therapy might still have a role
  to play.

69
Chemopreventive trials in BRCA mutated carriers.

• Which treatment?
• Tamoxifen.
• LH-RH agonists aromatase inhibitors
  (premenopausal women).
• Aromatase inhibitors (postmenopausal women).

70
Treatment

• Substantial evidence support the concept that
  oestrogens contribute to cause BC by stimulating
  proliferation of breast cells.
• Either high oestrogen levels or enhanced
  aromatase activity associate with increased
  susceptibility to BC.
• The rationale of the BC preventive treatment with
  SERMs is to antagonize estrogen receptor
  activity, thus inhibiting the proliferative
  hormonal stimula.

71
Treatment (ii)

• In presence of a mutation of an oncosuppressor
  gene, as BRCA 1 or 2, is the ER activation, even
  if partial, safe?
• It seems reasonable to prevent the estrogen
  production with aromatase inhibitors than
  antagonize estrogen effect.

72
ExemestaneRationale for Use in BC Prevention

• Exemestane inhibits in situ aromatase by more
  than 95.
• It also reduces endogenous oestrogen
  concentrations in BC. The treatment with
  irreversible aromatase inhibitors has been
  demonstrated to completely abrogate estrogen
  production, at the level of mammary gland.
• Suppressing local estrogen production may be
  important, as suggested by the discovery of a
  unique transcriptional promoter of aromatase gene
  expression in breast adipose tissue.

73
ExemestaneRationale for Use in BC Prevention

• Preventive effect in preclinical models
• Decreased levels of aromatase enzyme (instead of
  the increase observed after non-steroidal
  anti-aromatase agents)
• Activity in advanced breast cancer
• Improved tolerability vs TAM
• No negative effects on lipids
• Preclinical and clinical favourable bone data

74
ApreS (Aromasin Prevention Study)

• Double-Blind, Placebo-Controlled Study of
  Exemestane for the Prevention of Breast Cancer in
  Postmenopausal Unaffected Carriers of BRCA1/2
  Mutations
• Participating Italian Institutions (partial
  list)
• Italian Consortium HB/OC (G. Bevilacqua)
• Cooperative group for the identification of
  families at BC risk in Italy (V. Silingardi, S.
  Venuta)
• IRE Rome (F. Cognetti, M. Lopez, E. Terzoli),
  University of Napoli (A.R. Bianco, S. De Placido,
  A. Contegiacomo), University of Modena (M.
  Federico), University of LAquila (C. Ficorella,
  P. Marchetti), University of Chieti (S.
  Iacobelli, R. Mariani Costantini), University of
  Padova (Chieco Bianchi, E. D'Andrea, Monfardini),
  University of Messina (M. Mesiti), University of
  Ancona (R. Cellerino, A. Piga), University of
  Torino (P. Sismondi), Catholic University, Roma
  (G. Scambia, D. Terribile), Medical Oncology,
  Terni (F. Di Costanzo).
• Participation of 4 more European cooperative
  groups is pending.

75
ApreS Primary End-Point

• The efficacy of the irreversible aromatase
  inhibitor exemestane in preventing breast cancer
  by significantly reducing the incidence rate of
  invasive breast cancer in unaffected
  postmenopausal women carriers of BRCA1/BRCA2
  inactivation.