[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address 54.205.106.138. Please contact the publisher to request reinstatement.
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
Purchase Options:
[Skip to Content Landing]
Download PDF
Figure.
Identification of the OMEGA Study Cohort for Assessment of Breast Cancer Risk Following In Vitro Fertilization (IVF)
Identification of the OMEGA Study Cohort for Assessment of Breast Cancer Risk Following In Vitro Fertilization (IVF)

Adapted from Spaan et al.22

aAccording to the treatment center where these women were identified.

bWomen who were originally included in the non-IVF group but subsequently received IVF (eg, in another IVF clinic; n = 911 + 41 = 952) contributed person-time to both the non-IVF group and IVF group.

cWomen in this category contributed person-time from the first IVF treatment or first gynecological visit until date of death.

dFor unknown reasons, these women had originally not been identified as belonging in the IVF group; the women did not contribute person-time to the non-IVF group because IVF treatment was administered before 1989.

Table 1.  
Population Characteristics by IVF Exposure Status
Population Characteristics by IVF Exposure Status
Table 2.  
Incidence of First Breast Cancer and DCIS Compared With the General Population
Incidence of First Breast Cancer and DCIS Compared With the General Population
Table 3.  
Invasive Breast Cancer Risk According to Fertility Treatment and Reproductive Characteristicsa
Invasive Breast Cancer Risk According to Fertility Treatment and Reproductive Characteristicsa
Table 4.  
Invasive Breast Cancer Risk for IVF vs Non-IVF Treatment Within Risk Factor Subgroupsa,b
Invasive Breast Cancer Risk for IVF vs Non-IVF Treatment Within Risk Factor Subgroupsa,b
1.
Siegel  RL, Miller  KD, Jemal  A.  Cancer statistics, 2016. CA Cancer J Clin. 2016;66(1):7-30.PubMedArticle
2.
Bernstein  L.  Epidemiology of endocrine-related risk factors for breast cancer. J Mammary Gland Biol Neoplasia. 2002;7(1):3-15.PubMedArticle
3.
Collaborative Group on Hormonal Factors in Breast Cancer.  Breast cancer and hormonal contraceptives: collaborative reanalysis of individual data on 53 297 women with breast cancer and 100 239 women without breast cancer from 54 epidemiological studies. Lancet. 1996;347(9017):1713-1727.PubMedArticle
4.
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;350(9084):1047-1059.PubMedArticle
5.
Beckers  NG, Laven  JS, Eijkemans  MJ, Fauser  BC.  Follicular and luteal phase characteristics following early cessation of gonadotrophin-releasing hormone agonist during ovarian stimulation for in-vitro fertilization. Hum Reprod. 2000;15(1):43-49.PubMedArticle
6.
Joo  BS, Park  SH, An  BM, Kim  KS, Moon  SE, Moon  HS.  Serum estradiol levels during controlled ovarian hyperstimulation influence the pregnancy outcome of in vitro fertilization in a concentration-dependent manner. Fertil Steril. 2010;93(2):442-446.PubMedArticle
7.
Sunderam  S, Kissin  DM, Crawford  SB, Folger  SG, Jamieson  DJ, Barfield  WD; Centers for Disease Control and Prevention (CDC).  Assisted reproductive technology surveillance—United States, 2011. MMWR Surveill Summ. 2014;63(10):1-28.PubMed
8.
Stewart  LM, Hart  R.  Long-term cancer risks in women after treatment with IVF: do we have any answers yet? Womens Health (Lond Engl). 2015;11(1):7-10.PubMedArticle
9.
Sergentanis  TN, Diamantaras  AA, Perlepe  C, Kanavidis  P, Skalkidou  A, Petridou  ET.  IVF and breast cancer: a systematic review and meta-analysis. Hum Reprod Update. 2014;20(1):106-123.PubMedArticle
10.
Klip  H, Burger  CW, de Kraker  J, van Leeuwen  FE; OMEGA-project group.  Risk of cancer in the offspring of women who underwent ovarian stimulation for IVF. Hum Reprod. 2001;16(11):2451-2458.PubMed
11.
de Boer  EJ, den Tonkelaar  I, te Velde  ER, Burger  CW, van Leeuwen  FE; OMEGA-project group.  Increased risk of early menopausal transition and natural menopause after poor response at first IVF treatment. Hum Reprod. 2003;18(7):1544-1552.PubMedArticle
12.
van Leeuwen  FE, Klip  H, Mooij  TM,  et al.  Risk of borderline and invasive ovarian tumours after ovarian stimulation for in vitro fertilization in a large Dutch cohort. Hum Reprod. 2011;26(12):3456-3465.PubMedArticle
13.
Van den Brandt  PA, Schouten  LJ, Goldbohm  RA, Dorant  E, Hunen  PM.  Development of a record linkage protocol for use in the Dutch Cancer Registry for Epidemiological Research. Int J Epidemiol. 1990;19(3):553-558.PubMedArticle
14.
Breslow  NE, Day  NE. Statistical methods in cancer research: volume II—the design and analysis of cohort studies. IARC Sci Publ. 1987;(82):1-406.PubMed
15.
Forman  D, Bray  F, Brewster  DH,  et al, eds. Cancer Incidence in Five Continents. Vol X. IARC Sci Publ 164. Lyon, France: International Agency for Research on Cancer; 2014. http://www.iarc.fr/en/publications/pdfs-online/epi/sp164/CI5volX_Full.pdf. Accessed June 30, 2016.
16.
van Buuren  S, Boshuizen  HC, Knook  DL.  Multiple imputation of missing blood pressure covariates in survival analysis. Stat Med. 1999;18(6):681-694.PubMedArticle
17.
Raghunathan  TE, Lepkowski  JM, van Hoenwijk  J, Solenberger  P.  A multivariate technique for multiply imputing missing values using a sequence of regression models. Surv Methodol. 2001;27(1):85-95.
18.
White  IR, Royston  P.  Imputing missing covariate values for the Cox model. Stat Med. 2009;28(15):1982-1998.PubMedArticle
19.
Rubin  D. Multiple Imputation for Nonresponse in Surveys. Hoboken, NJ: Wiley; 2004.
20.
Li  F, Zaslavsky  AM, Landrum  MB.  Propensity score weighting with multilevel data. Stat Med. 2013;32(19):3373-3387.Article
21.
Hensher  D, Stopher  P, eds. Behavioral Travel Modelling. Hampshire, UK: Croom Helm; 1979.
22.
Spaan  M, van den Belt-Dusebout  AW, Schaapveld  M, Mooij  TM, Burger  CW, van Leeuwen  FE; OMEGA-project group.  Melanoma risk after ovarian stimulation for in vitro fertilization. Hum Reprod. 2015;30(5):1216-1228.PubMedArticle
23.
De Boer  EJ, Van Leeuwen  FE, Den Tonkelaar  I, Jansen  CA, Braat  DD, Burger  CW.  Methods and results of in-vitro fertilisation in the Netherlands in the years 1983-1994. Ned Tijdschr Geneeskd. 2004;148(29):1448-1455.PubMed
24.
Krul  IM, Groeneveld  E, Spaan  M,  et al; OMEGA-project group.  Increased breast cancer risk in in vitro fertilisation treated women with a multiple pregnancy: a new hypothesis based on historical in vitro fertilisation treatment data. Eur J Cancer. 2015;51(1):112-120.PubMedArticle
25.
Lo Russo  G, Spinelli  GP, Tomao  S,  et al.  Breast cancer risk after exposure to fertility drugs. Expert Rev Anticancer Ther. 2013;13(2):149-157.PubMedArticle
26.
Zreik  TG, Mazloom  A, Chen  Y,  et al.  Fertility drugs and the risk of breast cancer: a meta-analysis and review. Breast Cancer Res Treat. 2010;124(1):13-26.PubMedArticle
27.
Stewart  LM, Holman  CD, Hart  R, Bulsara  MK, Preen  DB, Finn  JC.  In vitro fertilization and breast cancer: is there cause for concern? Fertil Steril. 2012;98(2):334-340.PubMedArticle
28.
Venn  A, Watson  L, Bruinsma  F, Giles  G, Healy  D.  Risk of cancer after use of fertility drugs with in-vitro fertilisation. Lancet. 1999;354(9190):1586-1590.PubMedArticle
29.
Jensen  A, Sharif  H, Svare  EI, Frederiksen  K, Kjaer  SK.  Risk of breast cancer after exposure to fertility drugs: results from a large Danish cohort study. Cancer Epidemiol Biomarkers Prev. 2007;16(7):1400-1407.PubMedArticle
30.
Pappo  I, Lerner-Geva  L, Halevy  A,  et al.  The possible association between IVF and breast cancer incidence. Ann Surg Oncol. 2008;15(4):1048-1055.PubMedArticle
31.
Burkman  RT, Tang  MT, Malone  KE,  et al.  Infertility drugs and the risk of breast cancer: findings from the National Institute of Child Health and Human Development Women’s Contraceptive and Reproductive Experiences Study. Fertil Steril. 2003;79(4):844-851.PubMedArticle
32.
Reigstad  MM, Larsen  IK, Myklebust  TA,  et al.  Risk of breast cancer following fertility treatment—a registry based cohort study of parous women in Norway. Int J Cancer. 2015;136(5):1140-1148.PubMedArticle
33.
Katz  D, Paltiel  O, Peretz  T,  et al.  Beginning IVF treatments after age 30 increases the risk of breast cancer: results of a case-control study. Breast J. 2008;14(6):517-522.PubMedArticle
34.
Brinton  LA, Trabert  B, Shalev  V, Lunenfeld  E, Sella  T, Chodick  G.  In vitro fertilization and risk of breast and gynecologic cancers: a retrospective cohort study within the Israeli Maccabi Healthcare Services. Fertil Steril. 2013;99(5):1189-1196.PubMedArticle
35.
Russo  I, Russo  J.  Role of HCG and inhibin in breast-cancer  [review]. Int J Oncol. 1994;4(2):297-306.PubMed
36.
Russo  J, Mailo  D, Hu  YF, Balogh  G, Sheriff  F, Russo  IH.  Breast differentiation and its implication in cancer prevention. Clin Cancer Res. 2005;11(2 Pt 2):931s-936s.PubMed
37.
Boukaidi  SA, Cooley  A, Hardy  A, Matthews  L, Zelivianski  S, Jeruss  JS.  Impact of infertility regimens on breast cancer cells: follicle-stimulating hormone and luteinizing hormone lack a direct effect on breast cell proliferation in vitro. Fertil Steril. 2012;97(2):440-444.PubMedArticle
38.
Cooley  A, Matthews  L, Zelivianski  S, Hardy  A, Jeruss  JS.  Effect of infertility treatment and pregnancy-related hormones on breast cell proliferation in vitro. Hum Reprod. 2012;27(1):146-152.PubMedArticle
39.
Bernstein  L, Hanisch  R, Sullivan-Halley  J, Ross  RK.  Treatment with human chorionic gonadotropin and risk of breast cancer. Cancer Epidemiol Biomarkers Prev. 1995;4(5):437-440.PubMed
40.
Albrektsen  G, Heuch  I, Hansen  S, Kvåle  G.  Breast cancer risk by age at birth, time since birth and time intervals between births: exploring interaction effects. Br J Cancer. 2005;92(1):167-175.PubMedArticle
Original Investigation
July 19, 2016

Ovarian Stimulation for In Vitro Fertilization and Long-term Risk of Breast Cancer

Author Affiliations
  • 1Department of Epidemiology and Biostatistics, the Netherlands Cancer Institute, Amsterdam, the Netherlands
  • 2Department of Obstetrics and Gynecology, VU University Medical Center, Amsterdam, the Netherlands
  • 3Department of Reproductive Medicine and Gynecology, University Medical Center Utrecht, Utrecht, the Netherlands
  • 4Department of Obstetrics and Gynecology, Erasmus Medical Center, Rotterdam, the Netherlands
  • 5Department of Reproductive Medicine, Reinier de Graaf Hospital, Voorburg, the Netherlands
  • 6Department of Obstetrics, Gynecology, and Reproductive Medicine, Leiden University Medical Center, Leiden, the Netherlands
  • 7Department of Obstetrics and Gynecology, Isala Clinics, Zwolle, the Netherlands
  • 8Department of Obstetrics and Gynecology, Radboud University Nijmegen Medical Center, Nijmegen, the Netherlands
  • 9Department of Obstetrics and Gynecology, St Elisabeth Hospital, Tilburg, the Netherlands
  • 10Department of Obstetrics and Gynecology, University Medical Center Groningen, University Groningen, Groningen, the Netherlands.
  • 11Center for Reproductive Medicine, Department of Obstetrics and Gynecology, Academic Medical Center, Amsterdam, the Netherlands
  • 12Department of Obstetrics and Gynecology, Maastricht University Medical Center, Maastricht, the Netherlands
  • 13Department of Obstetrics and Gynecology, Catharina Hospital, Eindhoven, the Netherlands
JAMA. 2016;316(3):300-312. doi:10.1001/jama.2016.9389
Abstract

Importance  Previous studies of breast cancer risk after in vitro fertilization (IVF) treatment were inconclusive due to limited follow-up.

Objective  To assess long-term risk of breast cancer after ovarian stimulation for IVF.

Design, Setting, and Participants  Historical cohort (OMEGA study) with complete follow-up through December 2013 for 96% of the cohort. The cohort included 19 158 women who started IVF treatment between 1983 and 1995 (IVF group) and 5950 women starting other fertility treatments between 1980 and 1995 (non-IVF group) from all 12 IVF clinics in the Netherlands. The median age at end of follow-up was 53.8 years for the IVF group and 55.3 years for the non-IVF group.

Exposures  Information on ovarian stimulation for IVF, other fertility treatments, and potential confounders was collected from medical records and through mailed questionnaires.

Main Outcomes and Measures  Incidence of invasive and in situ breast cancers in women who underwent fertility treatments was obtained through linkage with the Netherlands Cancer Registry (1989-2013). Breast cancer risk in the IVF group was compared with risks in the general population (standardized incidence ratios [SIRs]) and the non-IVF group (hazard ratios [HRs]).

Results  Among 25 108 women (mean age at baseline, 32.8 years; mean number of IVF cycles, 3.6), 839 cases of invasive breast cancer and 109 cases of in situ breast cancer occurred after a median follow-up of 21.1 years. Breast cancer risk in IVF-treated women was not significantly different from that in the general population (SIR, 1.01 [95% CI, 0.93-1.09]) and from the risk in the non-IVF group (HR, 1.01 [95% CI, 0.86-1.19]). The cumulative incidences of breast cancer at age 55 were 3.0% for the IVF group and 2.9% for the non-IVF group (P = .85). The SIR did not increase with longer time since treatment (≥20 years) in the IVF group (0.92 [95% CI, 0.73-1.15]) or in the non-IVF group (1.03 [95% CI, 0.82-1.29]). Risk was significantly lower for those who underwent 7 or more IVF cycles (HR, 0.55 [95% CI, 0.39-0.77]) vs 1 to 2 IVF cycles and after poor response to the first IVF cycle (HR, 0.77 [95% CI, 0.61-0.96] for <4 vs ≥4 collected oocytes).

Conclusions and Relevance  Among women undergoing fertility treatment in the Netherlands between 1980 and 1995, IVF treatment compared with non-IVF treatment was not associated with increased risk of breast cancer after a median follow-up of 21 years. Breast cancer risk among IVF-treated women was also not significantly different from that in the general population. These findings are consistent with absence of a significant increase in long-term risk of breast cancer among IVF-treated women.

Introduction

Quiz Ref IDFor decades, breast cancer has been the most common malignancy among women worldwide.1 Both exogenous and endogenous estrogens and progestogens have been shown to affect breast cancer risk.24Quiz Ref ID Since in vitro fertilization (IVF) procedures temporarily cause decreased estradiol and progesterone levels (during down-regulation of the natural cycle),5 as well as strongly elevated levels6 (during stimulation phase), IVF might influence breast cancer risk.

Knowledge about possible long-term influences of ovarian stimulation for IVF on breast cancer risk is still scarce because use of IVF did not become widespread until the late 1980s.7,8 A meta-analysis reported inconclusive results regarding the effect of ovarian stimulation for IVF on breast cancer risk,9 but most studies had a relatively short follow-up and lacked subfertile comparison groups. Because of the high incidence of breast cancer and the large numbers of women undergoing ovarian stimulation for IVF, even a small risk increase would have important public health implications. Therefore, the aim of the present study was to quantify the long-term risk of breast cancer in a nationwide cohort of women treated with ovarian stimulation for IVF or other fertility treatments. A secondary aim was to examine the separate associations of subfertility diagnosis and of IVF treatment components with breast cancer risk.

Box Section Ref ID

Key Points

  • Question What is the long-term risk of breast cancer after ovarian stimulation for in vitro fertilization (IVF)?

  • Findings In this cohort study that included 25 108 women who underwent fertility treatments with a median follow-up of 21.1 years, breast cancer risk in IVF-treated women was not significantly different from that in the general population or in women who underwent other fertility treatments.

  • Meaning These findings are consistent with the absence of a significant increase in long-term risk of breast cancer among IVF-treated women.

Methods

In 1995 and 1996, a historical cohort (OMEGA study) of women treated for subfertility was identified, comprising women who started ovarian stimulation for IVF between 1983 and 1995 in one of the 12 IVF clinics in the Netherlands (IVF group) and a comparison group of women who started fertility treatments other than IVF between 1980 and 1995 in 4 clinics (non-IVF group). Most women in the non-IVF group underwent tubal surgery, (low-dose ovarian stimulation) intrauterine insemination (IUI), hormonal treatments (eg, clomiphene), or withdrew from the waiting list for IVF. Women in the non-IVF group were included with similar subfertility diagnoses as women in the IVF group. The institutional ethical committee of each of the IVF clinics approved the study procedures, which have been previously described.1012

Between 1997 and 2000, eligible women were invited to complete a risk factor questionnaire and to provide written informed consent for medical record data abstraction and future linkage with disease registries. The questionnaire ascertained each participant’s reproductive history, fertility treatments, use of exogenous hormones, lifestyle factors, and family history of cancer. Trained abstractors registered subfertility diagnoses, fertility-improving surgical procedures, and, for each IVF or IUI cycle, date, dosage, and type of fertility drugs, number of oocytes collected, and outcomes from medical records. In 2012, information on treatment cycles that women received after data collection in the period between 1997 and 2000 was added from electronically available data from all 12 IVF clinics. For approximately 23% of the cohort, subfertility diagnosis and number of IVF cycles were obtained from questionnaire data because data from medical records were not available. Both participating and nonresponding women were linked with the Dutch Municipal Personal Records Database, yielding information on vital status, parity, and age at first birth for 99% of the cohort through August 2013.

For 95.7% of the cohort, cancer incidence in the period 1989 through December 2013 was ascertained from the population-based Netherlands Cancer Registry (NCR)12,13 because 4.3% of women declined linkage. For cases of invasive breast cancer and ductal carcinoma in situ (DCIS), date of diagnosis and histology were obtained.

Statistical Analysis

Because the Netherlands Cancer Registry did not fully cover the Netherlands before 1989, observation time for each participant started on January 1989, the date of first IVF treatment (IVF group), or the date of the first clinic visit for subfertility evaluation (non-IVF group), whichever came last. Person-years of observation were calculated through December 31, 2013, date of any cancer diagnosis, or date of death, whichever came first. Women originally included in the non-IVF group, who subsequently received IVF, contributed person-time to the non-IVF group until the date of first IVF treatment, then switched to the exposed group from this date (according to standard cohort methodology regarding time-dependent allocation of person-years in case of changing exposure).14 Women with a cancer diagnosis before 1989 or cohort entry were excluded.

First invasive and in situ breast cancer incidence in the IVF group and the non-IVF group was compared with incidence among women in the Dutch population. Standardized incidence ratios (SIRs) were calculated as the ratios of observed and expected numbers in the cohort. Expected numbers of breast cancer cases were calculated by multiplying sex-, 5-year age-, and 3-year calendar-specific population incidence rates from the Netherlands Cancer Registry with the corresponding number of person-years observed among cohort members.15 Breast cancer incidence rates per 100 000 women were calculated for the IVF group and the non-IVF group by dividing the number of observed incident breast cancers by the number of person-years multiplied by 100 000. For the general population, breast cancer incidence rates per 100 000 women were calculated by dividing the observed numbers by the number of person-years, divided by 100 000, and multiplied by the observed:expected ratio in the full cohort.

Cumulative incidences of breast cancer at age 55 years were calculated for the IVF group and the non-IVF group using the life-table method. Cox proportional hazards models were used to compare breast cancer risk between the IVF and the non-IVF groups, adjusting for confounders. Hazard ratios (HRs) for breast cancer and 95% CIs were calculated with cumulative numbers of IVF cycles and births as time-dependent variables, using age (in years) as time scale. All other variables were included as fixed variables. None of the variables violated the proportional hazards assumption when evaluating log-minus-log plots. Confounders were defined as factors changing the risk estimate for the exposure of interest by 10% or more in a model including the potential confounder(s) and the variable of interest. Confounders were tested in models restricted to women who responded to the questionnaire since information on a number of potential confounders was missing for nonresponders. Selected confounders were subsequently included in all analyses including both responders and nonresponders.

Dose-response associations were examined with total number of IVF cycles, number of follicle-stimulating hormone (FSH) and human menopausal gonadotropin (hMG) ampules, and numbers of oocytes collected. Since women in the IVF group and the non-IVF group could have been treated with mild ovarian stimulation necessary for IUI, dose-response associations including the total number of IUI and IVF cycles were also investigated. Furthermore, the IVF group was also compared with a reference group of women who did not receive IVF, IUI, clomiphene, or any other fertility drug during the observation period. Effect modification of the association between IVF treatment and breast cancer risk by other breast cancer risk factors, especially parity, was examined. In all main analyses, missing values were included as a separate category. Four sensitivity analyses were performed: (1) with invasive breast cancer and DCIS as a combined outcome variable (75% of the cohort was ≥50 years at end of follow-up, the age at which population screening in the Netherlands starts); (2) in the cohort restricted to women who started fertility treatment from 1989 onward; (3) with multiple imputation for missing covariate data; and (4) with a propensity score–adjusted analysis.1620

The total numbers of IVF and IUI cycles and the subfertility diagnosis were imputed with multivariate imputation by chained equations.16,17 The imputation model was a negative binomial regression for numbers of IVF and IUI cycles and a nominal logistic regression for subfertility diagnosis. The model included year of first IVF treatment or first visit to a gynecologist; age at first IVF treatment or first visit to a gynecologist; and age at menarche, age at first birth, parity, number of births, body mass index, IVF clinic, total number of IUI cycles, subfertility diagnosis, breast cancer incidence, and the cumulative hazard calculated from a Cox proportional hazards model.18 For women with an unknown number of IVF cycles, the total number of IVF cycles was imputed and time since entry for each cycle was assigned based on the mean times among women with observed number of cycles and stratified by the number of births (0, 1, or >1). The imputation process was repeated 10 times to create 10 full data sets. The Rubin rule was used to combine results of the models fitted to these 10 data sets.19

The propensity to undergo IVF treatment was calculated using a mixed-effects logistic model20 with random intercept and fixed effects for the same variables included in the imputation model (previously reported in this section). The propensity score was added to the Cox model as a continuous covariate to adjust the IVF treatment effect for confounding. The goodness-of-fit of the propensity score model, based on the McFadden pseudo R2, was 0.31, which is an excellent fit.21 Trend tests were based on the P value of the category-specific mean as a continuous variable.14 All tests of statistical significance were 2-sided and a P value of less than .05 was considered statistically significant. Data were analyzed with Stata version 11.

Results

Of 19 861 women in the IVF group and 6604 women in the non-IVF group, 1015 were excluded because of invalid addresses and 71 were excluded because of cancer diagnoses before before 1989 or cohort entry (including 3 breast cancers). Thus, 19 275 women in the IVF group and 6078 women in the non-IVF group were sent questionnaires; 64.2% responded (Figure).22 Medical records of 15 007 women were abstracted. The analytic study cohort consisted of 25 108 women; 19 158 women in the IVF group and 5950 women in the non-IVF group (Figure).22 In the description of the cohort, 952 women who were originally included in the non-IVF group but subsequently received IVF in another clinic were included in the IVF group (Table 1).

Population Characteristics

Table 1 lists the population characteristics by IVF exposure status. Women in the non-IVF group had a slightly longer follow-up duration after start of treatment (median, 23.5 years) than women in the IVF group (median, 20.7 years) and were older at end of follow-up (median age 55.3 years) than women in the IVF group (53.8 years) due to the inclusion of women seeking fertility treatment before the introduction of IVF treatment to yield sufficient numbers of non-IVF controls. Thirty-eight percent of women who received IVF treatment remained nulliparous vs 28% of the non-IVF group. The mean age at baseline was 32.8 years and the mean number of IVF cycles was 3.6 (Table 1). IVF stimulation regimens used in the cohort have been described previously.12,23,24 Most women underwent stimulation with hMG or FSH preceded by down-regulation with gonadotropin-releasing hormone (eTable 1 in the Supplement). Responders had more often received IVF (84.2% vs 62.6%), were less often nulliparous (33.0% vs 39.4%), had fewer births (mean, 1.5 vs 2.9), an older age at first birth (31.4 vs 29.5 years) than nonresponders, but similar ages at start and end of follow-up. In total, there were 839 cases of first invasive breast cancer and 109 of DCIS (Table 2). Sixty-four percent of breast cancers were diagnosed before age 51 years.

Comparisons With External Reference Rates

Breast cancer risks were not increased in the IVF group (SIR, 1.01 [95% CI, 0.93-1.09]) or the non-IVF group (SIR, 1.00 [95% CI, 0.88-1.15]) compared with the Dutch general population (Table 2). The SIR did not increase with longer time since treatment, neither in the IVF group (SIR, 0.92 for ≥20 years [95% CI, 0.73-1.15]) nor in the non-IVF group (SIR, 1.03 for ≥20 years [95% CI, 0.82-1.29]). The actual rates of breast cancer per 100 000 women were 163.5 for the IVF group and 167.2 for the non-IVF group as compared with 163.3 for the general population. The SIRs did not significantly differ by age at first treatment. The SIR was significantly decreased for nulliparous women (SIR, 0.86 [95% CI, 0.76-0.97]) but significantly increased for parous women (SIR, 1.10 [95% CI, 1.02-1.20]) (P < .001). The SIRs for breast cancer and DCIS combined were 1.01 (95% CI, 0.93-1.08) for the IVF group, and 0.98 (95% CI, 0.86-1.11) for the non-IVF group (P = .70) (Table 2).

Internal Comparisons

The cumulative incidences of breast cancer at the age of 55 years were 3.0% for the IVF group and 2.9% for the non-IVF group (P = .85). As shown in Table 3, breast cancer risk in the IVF group was not significantly different from the risk in the non-IVF group (HR, 1.01 [95% CI, 0.86-1.19]), adjusting for parity and age at first birth. When comparing the IVF group with a reference group of women who did not receive IVF, IUI, clomiphene, or any other fertility drug, the HR was not significantly increased (HR, 1.08 [95% CI, 0.87-1.34] adjusted for parity). Within the IVF group, breast cancer risk decreased with more IVF cycles (P for trend = .001), with a significantly decreased risk with 7 cycles or more (HR, 0.55 [95% CI, 0.39-0.77]) compared with 1-2 IVF cycles. The HR for number of FSH and hMG ampules showed a similar significant trend, whereas the HR for number of IVF and IUI cycles combined did not. Breast cancer risk was significantly decreased (HR, 0.77 [95% CI, 0.61-0.96]) after poor response to the first IVF cycle (<4 collected oocytes) vs normal response (≥4 collected oocytes). Within normal responders, breast cancer risk did not significantly differ according to number of collected oocytes. Breast cancer risk was not associated with subfertility diagnosis or (type of) luteal phase support.

The risks according to parity showed a similar pattern as the SIRs. In the entire cohort, parous women had a significantly increased risk compared with nulliparous women (HR, 1.35 [95% CI, 1.16-3.73]). Women who were 35 years or older at first birth had a significantly increased risk compared with women younger than 25 years at first birth (HR, 1.73 [95% CI, 1.30-2.30] for age 35-39 years and HR, 2.52 [95% CI, 1.71-3.73] for age 40 years or older).

When stratifying by timing of first birth, the risks of breast cancer for IVF vs no IVF were equal among women who were already parous before IVF (HR,1.07 [95% CI, 0.80-1.41]), among women who had their first birth after IVF (HR, 1.06 [95% CI, 0.83-1.36]), and were not significantly different from women who remained nulliparous (Table 4). No significant differences in HRs for IVF vs no IVF were observed when stratifying analyses by attained age (>50 vs ≤50 years), by age at starting treatment (<30 vs 30-35 years and ≥36 years), or by time since treatment (<10 vs 10-19 and ≥20 years).

In analyses stratified by parity, breast cancer risk decreased with more IVF cycles among parous women (P for trend = .007), but not among nulliparous women (eTable 2 in the Supplement). Also, among parous women, breast cancer risk decreased significantly with fewer oocytes collected at first IVF cycle (P for trend = .01).

Four sensitivity analyses, using breast cancer and DCIS as a combined outcome (eTable 3 in the Supplement) among women who started treatment from 1989 onward (eTable 4 in the Supplement), multiple imputation for missing covariate data for those variables with nonnegligible proportions of missing values (eTable 5 in the Supplement), and propensity score–adjusted analysis all yielded HRs for IVF treatment (yes vs no) around unity. The HR for IVF treatment (yes vs no) in the propensity score–adjusted analysis was 0.98 (95% CI, 0.78-1.23).

Discussion

In this study of women undergoing fertility treatment in the Netherlands from 1980-1995 with a median follow-up of 21 years, breast cancer risk after IVF was not increased when compared with the general population or with a subfertile non-IVF comparison group. The risk did not differ by type of fertility drugs or subfertility diagnosis and was not increased at 20 or more years after IVF treatment. Women with 7 or more IVF cycles had a significantly decreased risk compared with women treated with 1 to 2 IVF cycles (HR, 0.55 [95% CI, 0.39-0.77]). Poor response to the first IVF cycle was associated with decreased breast cancer risk (HR, 0.77 [95% CI, 0.61-0.96] for <4 vs ≥4 collected oocytes). Parous women had a higher breast cancer risk than nulliparous women, especially when they were 35 years or older at first birth.

Quiz Ref IDThe results of the current study are consistent with recent reviews that reported no increased breast cancer risk after IVF,9,25,26 and are based on a median follow-up of 21.1 years compared with a mean of 8 years9 and 16 years27 in prior studies. The most recent review and meta-analysis comprised 576 incident breast cancer cases among women exposed to IVF from 8 studies, of which only 3 compared IVF-treated women with infertile women not treated with IVF.2729 In the largest previous study of 87 403 Danish subfertile women (mean follow-up, 8 years), breast cancer risk was not significantly different after treatment with or without ovarian stimulation.29 However, several studies observed increased breast cancer risk in subgroups, eg, after more than 4 cycles30 or 6 cycles of hMG31 or after more than 10 years since treatment.32 Stewart et al reported increased breast cancer risk among women who started IVF treatment at younger ages,27 whereas others reported a higher risk for women over age 30 years33 or 40 years30 at start of IVF. Increased breast cancer risks following IVF have also been reported for women who remained nulliparous34 and for parous women.32 Few epidemiologic studies have had sufficient power to study the effects of different fertility drugs used in IVF or effect modification by other risk factors such as parity.25

Quiz Ref IDSince IVF comprises several phases and treatment schedules changed over time, unraveling the potentially opposite effects of several components of the IVF procedure is challenging. Typically, the first phase in IVF treatment is down-regulation of a woman’s natural menstrual cycle with gonadotropin-releasing hormone, followed by ovarian stimulation with gonadotropins, causing up to 10-times higher estradiol and progesterone levels than in natural menstrual cycles.6 If sufficient follicles have developed, 10 000 IU human chorionic gonadotropin (hCG) is administered 36 hours prior to oocyte retrieval. Progestogens or hCG are given as luteal phase support for embryo implantation and are continued if pregnancy is achieved.

Previous studies have not addressed what the net effects on breast cancer risk are of the temporary decrease of estradiol and progesterone levels during the down-regulation phase and the strongly increased levels during ovarian stimulation, and whether the use of hCG might exert a protective effect through breast tissue differentiation.35,36 In 2 recent in vitro studies, FSH, luteinizing hormone, and clomiphene did not increase breast cell proliferation, whereas estrogen increased cell proliferation of estrogen receptor–positive breast cancer cells.37,38 Progesterone and hCG both decreased breast cell proliferation.38 A case-control study reported a decreased risk of breast cancer after hCG given during infertility or weight loss treatments.39 In our study, women with the highest number of IVF cycles had the lowest risk of breast cancer. Explanations could be that women treated with more IVF cycles received more hCG, had longer periods of down-regulation with low estradiol and progesterone levels, or because of some inherent characteristic of the women that required more IVF cycles. Women who remained nulliparous had a significantly lower risk of breast cancer than parous women in contrast with the established risk increase in nulliparous women in the general population. Compared with parous women, nulliparous women in the cohort more often had a poor response at first IVF cycle (23% vs 15%; P < .001), which in this study was associated with a decreased breast cancer risk. In a previous report on this cohort, poor response at first IVF cycle was predictive for early menopause,11 which reduces the risk of breast cancer. However, adjustment for poor response did not materially affect the association between parity and breast cancer risk (crude HR, 1.37 [95% CI, 1.18-1.58]; adjusted HR, 1.35 [95% CI, 1.17-1.57]). The higher risk for parous women may also be due to the temporary increase of breast cancer risk after giving birth.40 As this increased risk may last 20 to 30 years, many parous women in the cohort had not yet reached the ages at which parity would exert a protective effect on breast cancer risk. Furthermore, the late age at first birth in the cohort (median, 31.5 years; 75th quartile, 35.0 years) may have precluded the observation of a reduced risk associated with parity.

Strengths of this study include the large size of the cohort and the long and complete follow-up, providing sufficient power for subgroup analyses. Although IVF regimens changed after 1995, information about long-term cancer risk is important because many women received IVF treatment before 1995. Selection bias is unlikely since 96% of the cohort was linked with the population-based Netherlands Cancer Registry, enabling us to also evaluate the risk including DCIS. Furthermore, accurate information on the most important confounders (ie, number of births and age at first birth) was available for 99% of the cohort (both for responders and nonresponders to the questionnaire).

This study has several limitations. First, several potential confounding factors, such as subfertility diagnosis, had high rates of missing data. There was an imbalance with more missing data in the non-IVF group (33%) than in the IVF group (16%). However, near complete information was available on IVF exposure, and the most important potential confounders (age at first birth and parity). Furthermore, after adjusting for parity and age at first birth, adjustment for other potential confounders did not affect risk estimates. All comparisons of the IVF group with the general population and with the non-IVF group yielded similar results, including sensitivity analyses in subgroups, analyses with multiple imputation of missing covariate data, or propensity score adjustment.

Second, age at menopause and menopausal status at end of follow-up were unknown for most women because few were postmenopausal at questionnaire completion between 1997 and 2000. If IVF-treated women reach earlier menopause than women in the general population, breast cancer risk after IVF may have been underestimated. However, adjustment for number of collected oocytes at first IVF cycle (with low numbers of oocytes as a proxy for early menopause11) did not affect the HR estimates for IVF. Furthermore, results were comparable in analyses stratified by age (younger than and older than 50 years).

Third, for women who received fertility treatment before the Netherlands Cancer Registry fully covered the Netherlands (1989), person-years were included from 1989 onward because cancer incidence before 1989 was only known for responding women and not for nonresponding women. This lag between start of treatment and start of observation time for women treated between 1980 and 1988 might have influenced the results if cancers diagnosed between 1980 and 1989 were missed. However, with 71 responding women reporting any cancer (including 3 breast cancers) before 1989 or cohort entry and assuming equal breast cancer rates in nonresponding women, approximately 50 nonresponding women would have developed any cancer (including 2 breast cancers) before 1989 or cohort entry. Thus, the number of potentially missed early-onset breast cancers is too small to have influenced the results. A sensitivity analysis restricted to women starting treatment from 1989 onward, to examine whether the results were influenced by exclusion of person-years between exposure and 1989, had similar results to the main analysis.

Fourth, results are largely based on IVF treatment protocols used until 1995. During the study period, the number of IVF cycles and, consequently, the number of ampules of gonadotropins used increased until 1990 and decreased thereafter. Because more recent IVF regimens largely consist of protocols with antagonists and shorter periods of down-regulation (possibly associated with less risk reduction), and improved success rates (associated with fewer cycles), it is uncertain how study results generalize to more contemporary IVF treatment. Quiz Ref IDFurthermore, because only 14% of the cohort had reached age 60 years, follow-up is necessary to evaluate postmenopausal breast cancer risk after ovarian stimulation for IVF.

Conclusions

Among women undergoing fertility treatment in the Netherlands between 1980 and 1995, the use of IVF compared with non-IVF treatment was not associated with increased risk of breast cancer after a median follow-up of 21 years. These findings are consistent with the absence of a significant increase in the long-term risk of breast cancer among women treated with these IVF regimens.

Back to top
Article Information

Corresponding Author: Flora E. van Leeuwen, PhD, Department of Epidemiology, The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, the Netherlands (f.v.leeuwen@nki.nl).

Author Contributions: Drs van den Belt-Dusebout and van Leeuwen had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: van den Belt-Dusebout, Burger, van Leeuwen.

Acquisition, analysis, or interpretation of data: van den Belt-Dusebout, Spaan, Lambalk, Kortman, Laven, van Santbrink, van der Westerlaken, Cohlen, Braat, Smeenk, Land, Goddijn, van Rumste, Schats, Jóźwiak, Hauptmann, Rookus, van Leeuwen.

Drafting of the manuscript: van den Belt-Dusebout, Spaan, van Leeuwen.

Critical revision of the manuscript for important intellectual content: van den Belt-Dusebout, Spaan, Lambalk, Kortman, Laven, van Santbrink, van der Westerlaken, Cohlen, Braat, Smeenk, Land, Goddijn, van Golde, van Rumste, Schats, Jóźwiak, Hauptmann, Rookus, Burger, van Leeuwen.

Statistical analysis: van den Belt-Dusebout, Spaan, Jóźwiak, Hauptmann, Rookus, van Leeuwen.

Obtained funding: Burger, van Leeuwen.

Administrative, technical, or material support: van den Belt-Dusebout, Spaan, Lambalk, Kortman, van der Westerlaken, Braat, van Golde.

No additional contributions: Kortman, Laven, van Santbrink, Cohlen, Smeenk, Land, Goddijn, Schats.

Dr van den Belt-Dusebout and Ms Spaan contributed equally:

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Drs Land and Laven report receipt of unrestricted research grants from Ferring Pharmaceuticals outside the submitted work. Dr Braat reports receipt of unrestricted grants from Merck Serono and Ferring Pharmaceuticals outside the submitted work. Dr van Santbrink reports receipt of personal fees from Advisory Board Merck Serono outside the submitted work. Dr Smeenk reports serving on an advisory board and receipt of personal fees from Merck Serono, providing a lecture and receipt of personal fees from MSD, and a grant for quality management from Ferring Pharmaceuticals outside the submitted work.

Funding/Support: This work was supported by the Dutch Cancer Society (NKI 2006-3631); the Health Research and Development Counsel (28-2540); and the Dutch Ministry of Health.

Role of the Funder/Sponsor: The funding organizations were not involved in the design and conduct of the study, the collection, management, analysis, or interpretation of the data; preparation, review, or approval of the manuscript; nor the decision to submit the manuscript for publication.

Additional Contributions: The authors thank the participants of the OMEGA project, without whom this study would not have been possible. The authors thank the medical registries of the participating clinics for making patient selection possible, and all attending physicians for providing access to their patients’ medical records. The authors thank Helen Klip, PhD, Coordinator Scientific Research, Karakter Academy, Nijmegen, the Netherlands, for coordinating the data collection and building of the original database between 1996 and 2000 when employed at the Department of Epidemiology of the Netherlands Cancer Institute. Dr Klip did not receive any compensation for her role in the current study. The authors have obtained written permission to include her name in the Acknowledgment section. The authors also acknowledge the Netherlands Cancer Registry for providing the follow-up cancer data and the Dutch Municipal Personal Records Database for parity data.

References
1.
Siegel  RL, Miller  KD, Jemal  A.  Cancer statistics, 2016. CA Cancer J Clin. 2016;66(1):7-30.PubMedArticle
2.
Bernstein  L.  Epidemiology of endocrine-related risk factors for breast cancer. J Mammary Gland Biol Neoplasia. 2002;7(1):3-15.PubMedArticle
3.
Collaborative Group on Hormonal Factors in Breast Cancer.  Breast cancer and hormonal contraceptives: collaborative reanalysis of individual data on 53 297 women with breast cancer and 100 239 women without breast cancer from 54 epidemiological studies. Lancet. 1996;347(9017):1713-1727.PubMedArticle
4.
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;350(9084):1047-1059.PubMedArticle
5.
Beckers  NG, Laven  JS, Eijkemans  MJ, Fauser  BC.  Follicular and luteal phase characteristics following early cessation of gonadotrophin-releasing hormone agonist during ovarian stimulation for in-vitro fertilization. Hum Reprod. 2000;15(1):43-49.PubMedArticle
6.
Joo  BS, Park  SH, An  BM, Kim  KS, Moon  SE, Moon  HS.  Serum estradiol levels during controlled ovarian hyperstimulation influence the pregnancy outcome of in vitro fertilization in a concentration-dependent manner. Fertil Steril. 2010;93(2):442-446.PubMedArticle
7.
Sunderam  S, Kissin  DM, Crawford  SB, Folger  SG, Jamieson  DJ, Barfield  WD; Centers for Disease Control and Prevention (CDC).  Assisted reproductive technology surveillance—United States, 2011. MMWR Surveill Summ. 2014;63(10):1-28.PubMed
8.
Stewart  LM, Hart  R.  Long-term cancer risks in women after treatment with IVF: do we have any answers yet? Womens Health (Lond Engl). 2015;11(1):7-10.PubMedArticle
9.
Sergentanis  TN, Diamantaras  AA, Perlepe  C, Kanavidis  P, Skalkidou  A, Petridou  ET.  IVF and breast cancer: a systematic review and meta-analysis. Hum Reprod Update. 2014;20(1):106-123.PubMedArticle
10.
Klip  H, Burger  CW, de Kraker  J, van Leeuwen  FE; OMEGA-project group.  Risk of cancer in the offspring of women who underwent ovarian stimulation for IVF. Hum Reprod. 2001;16(11):2451-2458.PubMed
11.
de Boer  EJ, den Tonkelaar  I, te Velde  ER, Burger  CW, van Leeuwen  FE; OMEGA-project group.  Increased risk of early menopausal transition and natural menopause after poor response at first IVF treatment. Hum Reprod. 2003;18(7):1544-1552.PubMedArticle
12.
van Leeuwen  FE, Klip  H, Mooij  TM,  et al.  Risk of borderline and invasive ovarian tumours after ovarian stimulation for in vitro fertilization in a large Dutch cohort. Hum Reprod. 2011;26(12):3456-3465.PubMedArticle
13.
Van den Brandt  PA, Schouten  LJ, Goldbohm  RA, Dorant  E, Hunen  PM.  Development of a record linkage protocol for use in the Dutch Cancer Registry for Epidemiological Research. Int J Epidemiol. 1990;19(3):553-558.PubMedArticle
14.
Breslow  NE, Day  NE. Statistical methods in cancer research: volume II—the design and analysis of cohort studies. IARC Sci Publ. 1987;(82):1-406.PubMed
15.
Forman  D, Bray  F, Brewster  DH,  et al, eds. Cancer Incidence in Five Continents. Vol X. IARC Sci Publ 164. Lyon, France: International Agency for Research on Cancer; 2014. http://www.iarc.fr/en/publications/pdfs-online/epi/sp164/CI5volX_Full.pdf. Accessed June 30, 2016.
16.
van Buuren  S, Boshuizen  HC, Knook  DL.  Multiple imputation of missing blood pressure covariates in survival analysis. Stat Med. 1999;18(6):681-694.PubMedArticle
17.
Raghunathan  TE, Lepkowski  JM, van Hoenwijk  J, Solenberger  P.  A multivariate technique for multiply imputing missing values using a sequence of regression models. Surv Methodol. 2001;27(1):85-95.
18.
White  IR, Royston  P.  Imputing missing covariate values for the Cox model. Stat Med. 2009;28(15):1982-1998.PubMedArticle
19.
Rubin  D. Multiple Imputation for Nonresponse in Surveys. Hoboken, NJ: Wiley; 2004.
20.
Li  F, Zaslavsky  AM, Landrum  MB.  Propensity score weighting with multilevel data. Stat Med. 2013;32(19):3373-3387.Article
21.
Hensher  D, Stopher  P, eds. Behavioral Travel Modelling. Hampshire, UK: Croom Helm; 1979.
22.
Spaan  M, van den Belt-Dusebout  AW, Schaapveld  M, Mooij  TM, Burger  CW, van Leeuwen  FE; OMEGA-project group.  Melanoma risk after ovarian stimulation for in vitro fertilization. Hum Reprod. 2015;30(5):1216-1228.PubMedArticle
23.
De Boer  EJ, Van Leeuwen  FE, Den Tonkelaar  I, Jansen  CA, Braat  DD, Burger  CW.  Methods and results of in-vitro fertilisation in the Netherlands in the years 1983-1994. Ned Tijdschr Geneeskd. 2004;148(29):1448-1455.PubMed
24.
Krul  IM, Groeneveld  E, Spaan  M,  et al; OMEGA-project group.  Increased breast cancer risk in in vitro fertilisation treated women with a multiple pregnancy: a new hypothesis based on historical in vitro fertilisation treatment data. Eur J Cancer. 2015;51(1):112-120.PubMedArticle
25.
Lo Russo  G, Spinelli  GP, Tomao  S,  et al.  Breast cancer risk after exposure to fertility drugs. Expert Rev Anticancer Ther. 2013;13(2):149-157.PubMedArticle
26.
Zreik  TG, Mazloom  A, Chen  Y,  et al.  Fertility drugs and the risk of breast cancer: a meta-analysis and review. Breast Cancer Res Treat. 2010;124(1):13-26.PubMedArticle
27.
Stewart  LM, Holman  CD, Hart  R, Bulsara  MK, Preen  DB, Finn  JC.  In vitro fertilization and breast cancer: is there cause for concern? Fertil Steril. 2012;98(2):334-340.PubMedArticle
28.
Venn  A, Watson  L, Bruinsma  F, Giles  G, Healy  D.  Risk of cancer after use of fertility drugs with in-vitro fertilisation. Lancet. 1999;354(9190):1586-1590.PubMedArticle
29.
Jensen  A, Sharif  H, Svare  EI, Frederiksen  K, Kjaer  SK.  Risk of breast cancer after exposure to fertility drugs: results from a large Danish cohort study. Cancer Epidemiol Biomarkers Prev. 2007;16(7):1400-1407.PubMedArticle
30.
Pappo  I, Lerner-Geva  L, Halevy  A,  et al.  The possible association between IVF and breast cancer incidence. Ann Surg Oncol. 2008;15(4):1048-1055.PubMedArticle
31.
Burkman  RT, Tang  MT, Malone  KE,  et al.  Infertility drugs and the risk of breast cancer: findings from the National Institute of Child Health and Human Development Women’s Contraceptive and Reproductive Experiences Study. Fertil Steril. 2003;79(4):844-851.PubMedArticle
32.
Reigstad  MM, Larsen  IK, Myklebust  TA,  et al.  Risk of breast cancer following fertility treatment—a registry based cohort study of parous women in Norway. Int J Cancer. 2015;136(5):1140-1148.PubMedArticle
33.
Katz  D, Paltiel  O, Peretz  T,  et al.  Beginning IVF treatments after age 30 increases the risk of breast cancer: results of a case-control study. Breast J. 2008;14(6):517-522.PubMedArticle
34.
Brinton  LA, Trabert  B, Shalev  V, Lunenfeld  E, Sella  T, Chodick  G.  In vitro fertilization and risk of breast and gynecologic cancers: a retrospective cohort study within the Israeli Maccabi Healthcare Services. Fertil Steril. 2013;99(5):1189-1196.PubMedArticle
35.
Russo  I, Russo  J.  Role of HCG and inhibin in breast-cancer  [review]. Int J Oncol. 1994;4(2):297-306.PubMed
36.
Russo  J, Mailo  D, Hu  YF, Balogh  G, Sheriff  F, Russo  IH.  Breast differentiation and its implication in cancer prevention. Clin Cancer Res. 2005;11(2 Pt 2):931s-936s.PubMed
37.
Boukaidi  SA, Cooley  A, Hardy  A, Matthews  L, Zelivianski  S, Jeruss  JS.  Impact of infertility regimens on breast cancer cells: follicle-stimulating hormone and luteinizing hormone lack a direct effect on breast cell proliferation in vitro. Fertil Steril. 2012;97(2):440-444.PubMedArticle
38.
Cooley  A, Matthews  L, Zelivianski  S, Hardy  A, Jeruss  JS.  Effect of infertility treatment and pregnancy-related hormones on breast cell proliferation in vitro. Hum Reprod. 2012;27(1):146-152.PubMedArticle
39.
Bernstein  L, Hanisch  R, Sullivan-Halley  J, Ross  RK.  Treatment with human chorionic gonadotropin and risk of breast cancer. Cancer Epidemiol Biomarkers Prev. 1995;4(5):437-440.PubMed
40.
Albrektsen  G, Heuch  I, Hansen  S, Kvåle  G.  Breast cancer risk by age at birth, time since birth and time intervals between births: exploring interaction effects. Br J Cancer. 2005;92(1):167-175.PubMedArticle
×