Advanced

Oral contraceptive use and breast cancer risk: A cohort study of BRCA1 and BRCA2 mutation carriers

Schrijver, L ; H., Olsson LU ; Antoniou, A ; Milne, R ; Phillips, K ; Andrieu, N ; Easton, D ; Goldgar, D ; Engel, C and Kast, K , et al. (2017) American Association for Cancer Research (AACR) 108th Annual Meeting 2017 In Cancer Research 77(13 Suppl 1).
Abstract
Background: BRCA1 and BRCA2 mutation carriers are at high risk of breast and ovarian cancer. Oral contraceptive preparations (OCPs) may reduce ovarian cancer risk, but its effect on breast cancer risk remains unclear. Methods: Combined data from three cohorts of 5705 BRCA1 and 3521 BRCA2 mutation carriers (IBCCS, BCFR and kConFab) were analyzed using age-dependent Cox regression models stratified for study and birth cohort. We conducted the first prospective analyses on this topic. Our additional main retrospective analyses were leftcensored 5 years preceding date of baseline questionnaire to control for survival bias. The full-cohort retrospective analyses, without left-censoring, was performed to compare results with the literature.... (More)
Background: BRCA1 and BRCA2 mutation carriers are at high risk of breast and ovarian cancer. Oral contraceptive preparations (OCPs) may reduce ovarian cancer risk, but its effect on breast cancer risk remains unclear. Methods: Combined data from three cohorts of 5705 BRCA1 and 3521 BRCA2 mutation carriers (IBCCS, BCFR and kConFab) were analyzed using age-dependent Cox regression models stratified for study and birth cohort. We conducted the first prospective analyses on this topic. Our additional main retrospective analyses were leftcensored 5 years preceding date of baseline questionnaire to control for survival bias. The full-cohort retrospective analyses, without left-censoring, was performed to compare results with the literature. Prospective analyses were considered most valid, while retrospective analyses were most powerful. Results: For BRCA1 mutation carriers we found no association between ever OCP use and risk of breast cancer in the prospective analyses (HR=1.08, 95% CI 0.75-1.56), but 23% and 27% increased risks for ever OCP use in the left-censored and full retrospective analyses, respectively. Retrospectively, an increasing trend for longer duration of use, especially before first full-term pregnancy (FFTP) was found (left-censored analyses: 10 years HR 1.41 (95%CI 1.10-1.813), p-trend=0.001 for duration of use before FFTP). For BRCA2 mutation carriers we found a positive association between ever OCP use and risk of breast cancer prospectively (HR=1.75, 95% CI 1.03-2.97), but retrospectively findings were inconsistent (HR=1.06, 95% CI 0.85-1.33 and HR=1.52, 95% CI 1.28-1.81 for the left-censored and full analyses, respectively). Conclusion: For BRCA1 mutation carriers the discrepancy between results of prospective and retrospective analyses may be explained by time since last OCP use before FFTP. Thus, a temporal increased risk of breast cancer following longer duration of OCP use before FFTP cannot be ruled out. The discordant findings between prospective and retrospective analyses for BRCA2 carriers could not be explained. Because of the lack of scientific clarity it is too early to give an unequivocal advice on OCP use with respect to breast cancer risk to BRCA1 and BRCA2 mutation carriers. (Less)
Please use this url to cite or link to this publication:
author
, et al. (More)
(Less)
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
BRCA1 protein, BRCA2 protein, endogenous compound, oral contraceptive agent, adverse drug reaction, age, breast cancer, cancer risk, cancer survival, case report, cohort analysis, female, gene mutation, genetic susceptibility, human, oral contraceptive use, pregnancy, proportional hazards model, prospective study, questionnaire, retrospective study, side effect
in
Cancer Research
volume
77
issue
13 Suppl 1
article number
Abstract 4276
publisher
American Association for Cancer Research Inc.
conference name
American Association for Cancer Research (AACR) 108th Annual Meeting 2017
conference location
Washington, DC, United States
conference dates
2017-04-01 - 2017-04-05
ISSN
1538-7445
DOI
10.1158/1538-7445.AM2017-4276
language
English
LU publication?
yes
id
ee301d12-be2a-4aa3-b432-8c91b56e08c9
date added to LUP
2019-07-01 10:10:56
date last changed
2019-07-17 10:45:54
@misc{ee301d12-be2a-4aa3-b432-8c91b56e08c9,
  abstract     = {Background: BRCA1 and BRCA2 mutation carriers are at high risk of breast and ovarian cancer. Oral contraceptive preparations (OCPs) may reduce ovarian cancer risk, but its effect on breast cancer risk remains unclear. Methods: Combined data from three cohorts of 5705 BRCA1 and 3521 BRCA2 mutation carriers (IBCCS, BCFR and kConFab) were analyzed using age-dependent Cox regression models stratified for study and birth cohort. We conducted the first prospective analyses on this topic. Our additional main retrospective analyses were leftcensored 5 years preceding date of baseline questionnaire to control for survival bias. The full-cohort retrospective analyses, without left-censoring, was performed to compare results with the literature. Prospective analyses were considered most valid, while retrospective analyses were most powerful. Results: For BRCA1 mutation carriers we found no association between ever OCP use and risk of breast cancer in the prospective analyses (HR=1.08, 95% CI 0.75-1.56), but 23% and 27% increased risks for ever OCP use in the left-censored and full retrospective analyses, respectively. Retrospectively, an increasing trend for longer duration of use, especially before first full-term pregnancy (FFTP) was found (left-censored analyses: 10 years HR 1.41 (95%CI 1.10-1.813), p-trend=0.001 for duration of use before FFTP). For BRCA2 mutation carriers we found a positive association between ever OCP use and risk of breast cancer prospectively (HR=1.75, 95% CI 1.03-2.97), but retrospectively findings were inconsistent (HR=1.06, 95% CI 0.85-1.33 and HR=1.52, 95% CI 1.28-1.81 for the left-censored and full analyses, respectively). Conclusion: For BRCA1 mutation carriers the discrepancy between results of prospective and retrospective analyses may be explained by time since last OCP use before FFTP. Thus, a temporal increased risk of breast cancer following longer duration of OCP use before FFTP cannot be ruled out. The discordant findings between prospective and retrospective analyses for BRCA2 carriers could not be explained. Because of the lack of scientific clarity it is too early to give an unequivocal advice on OCP use with respect to breast cancer risk to BRCA1 and BRCA2 mutation carriers.},
  author       = {Schrijver, L and H., Olsson and Antoniou, A and Milne, R and Phillips, K and Andrieu, N and Easton, D and Goldgar, D and Engel, C and Kast, K and Blom, M-J and Mooij, T and Hopper, J and Van Leeuwen, F and Terry, M and Rookus, M},
  issn         = {1538-7445},
  language     = {eng},
  note         = {Conference Abstract},
  number       = {13 Suppl 1},
  publisher    = {American Association for Cancer Research Inc.},
  series       = {Cancer Research},
  title        = {Oral contraceptive use and breast cancer risk: A cohort study of BRCA1 and BRCA2 mutation carriers},
  url          = {http://dx.doi.org/10.1158/1538-7445.AM2017-4276},
  doi          = {10.1158/1538-7445.AM2017-4276},
  volume       = {77},
  year         = {2017},
}