Breast cancer risk in relation to different types of hormone replacement therapy : Update of the E3N

1
Breast cancer risk in relation to different
types of hormone replacement therapy Update of
the E3N results. A. FOURNIER, F.
CLAVEL-CHAPELON (Inserm, Villejuif, France
94805), F. BERRINO (Istituto Nazionale Tumori,
Milan, Italy 20133).Breast Cancer Research and
Treatment 2007, online
2
An important public health question

• In France, 10 millions of women are
  postmenopausal
• In 2000-2002, around 1/3 of those aged 65 or less
  were using HRTs
• Breast cancer 1st incident cancer in women (40
  000 in year 2000 in France) and 1st cause of
  cancer death (10 000 in year 2000)

3
Introduction

• Estrogen-progestagen postmenopausal hormone
  replacement therapy (HRT) is associated with an
  increase in breast cancer risk
• The effect of estrogen alone is still the subject
  of intense debate
• Stefanick ML et al, Effects of conjugated
  equine estrogens on breast cancer and mammography
  screening in postmenopausal women with
  hysterectomy, JAMA 2006
• Collaborative Group on Hormonal Factors in
  Breast Cancer. Breast cancer and hormone
  replacement therapy collaborative reanalysis of
  data from 51 epidemiological studies of 52,705
  women with breast cancer and 108,411 women
  without breast cancer, Lancet 1997
• Role of the progestagen component ?

4
HRTs studied
CEE or estradiol
CEE MPA
MPA or norethisterone
or norgestrel, levonorgestrel
Estradiol
progesterone or dydrogesterone or
nomegestrol or promegestone or
norethisterone or chlormadinone or
medrogestone or cyproterone or MPA
5
MWS (1) Material and methods

• Prospective cohort
• Around 1 million British women, aged 50-64,
  participants to BC screening program
• Information on HRT use collected before
  mammography
• BC cases flagged on the National Health Service
  central registers

6
MWS (2) Results

• Increase of BC risk with duration of use, limited
  to the period of use of HRT

Million Women Study Collaborators, Lancet, 2003
7
MWS (3) Results

• Similar increase whatever
• -the type of progestogen
• -the type of regimen

Million Women Study Collaborators, Lancet, 2003
8
Aim

• To investigate the association between different
  HRTs and breast cancer risk among postmenopausal
  women, especially according to the progestagen
  component

9
Study population

• E3N Cohort Study established in 1990 with 98
  995 women from a health insurance scheme covering
  French teachers, born 1925-1950
• Part of the European Prospective Investigation
  into Cancer and Nutrition (EPIC)
• Follow-up questionnaires sent in 1992, 1993,
  1995, 1997, 2000 and 2002
• Exclusions premenopausal women, prevalent cancer
  cases
• Analytical sample 80 377 postmenopausal women

10
Identification of breast cancer cases

• Through follow-up questionnaires and causes of
  death
• Pathology reports obtained for 96 of the
  incident cases identified
• lt5 of false-positive self-reports
• 2 354 postmenopausal invasive breast cancer cases
  were identified,
• Mean duration of post-menopausal follow-up 8.1
  years

11
Ascertainment of HRT use

• Age at start
• Brand name
• Duration
• Update in each follow-up questionnaire
• Complete history of HRT use

12
Statistical analysis

• Relative risk (RR) estimates were obtained using
  Coxs proportional hazard model with time since
  menopause as the time scale.
• Multivariate adjustment included anthropometry,
  family history, personal history of benign breast
  disease, mammographic surveillance, use of
  hormones before menopause, and reproductive
  variables.
• HRT as a time-dependant variable
• Women who did not use the same type of HRT
  throughout follow-up contributed to a mixed use
  category from the time they changed HRT

13
HRT use in the E3N cohort
current use among postmenopausal women
Age
14
Relative Risk

• Given a baseline risk of breast cancer among
  women aged 55-59 in France in year 2000 of 300
  per 100 000,
• A RR of 1.4 means that among 100 000 HRT users of
  the same age, we will observe
• 300 x 1.4 420 breast cancer cases,
• That is, 420-300120 additional breast cancer
  cases

15
Results (1)
16
Results (2)
17

Strengths and limitations

• Prospective design, high follow-up rates and
  adjustment for several potential confounders
• Data on HRT updated throughout follow-up,
  enabling
• To isolate the effect of each type of HRT (mixed
  use category)
• To decrease misclassification of duration of use
• Data on HRT are self-reported

18

Conclusion

• Consistent with other studies for
  estrogen-progestagens HRTs containing MPA or
  testosterone-derived progestins increase in
  breast cancer risk
• First epidemiologic study evaluating HRTs
  containing progesterone or dydrogesterone (its
  isomer)
• Estrogen-progesterone and estrogen-dydrogesterone
  associated with an RR significantly lower than
  for other HRTs (except dydrogesterone)
• Other HRTs yielded RRs that did not differ
  significantly from one another
• Need for other studies on estrogen-progesterone
  and estrogen-dydrogesterone combinations