Doi:10.1016/j.mehy.2006.09.039

http://intl.elsevierhealth.com/journals/mehy Personal care products that contain estrogens orxenoestrogens may increase breast cancer risk Maryann Donovan a, Chandra M. Tiwary a, Deborah Axelrod c,Annie J. Sasco b, Lovell Jones d, Richard Hajek d, Erin Sauber a,Jean Kuo a, Devra L. Davis a,* a Center for Environmental Oncology, University of Pittsburgh Cancer Institute, University of Pittsburgh,Graduate School of Public Health, Department of Epidemiology, UPMC Cancer Pavilion, Fourth Floor,5150 Centre Avenue, Suite 432 Pittsburgh, PA 15232, USAb University Victor Segalen Bordeaux 2, INSERM U 593, Team of Epidemiology for Cancer Prevention,146 rue Leo Saignat, 33076 Bordeaux cedex, Francec Community Cancer Education and Outreach, New York University Clinical Cancer Center, New York,New York, USAd M.D. Anderson Cancer Center, Center for Research on Minority Health, University of Texas, USA Received 15 September 2006; accepted 20 September 2006 Established models of breast cancer risk, such as the Gail model, do not account for patterns of the disease in women under the age of 35, especially in African Americans. With the possible exceptions of ionizing radiation orinheriting a known genetic mutation, most of the known risk factors for breast cancer are related to cumulativelifetime exposure to estrogens. Increased risk of breast cancer has been associated with earlier onset of menses orlater age at menopause, nulliparity or late first parity, use of hormonal contraceptives or hormone replacementtherapy, shorter lactation history, exposure to light at night, obesity, and regular ingestion of alcohol, all of whichincrease circulating levels of unbound estradiol. Among African Americans at all ages, use of hormone-containingpersonal care products (PCPs) is more common than among whites, as is premature appearance of secondary sexualcharacteristics among infants and toddlers. We hypothesize that the use of estrogen and other hormone-containingPCPs in young African American women accounts, in part, for their increased risk of breast cancer prior to menopause,by subjecting breast buds to elevated estrogen exposure during critical windows of vulnerability in utero and in earlylife. These early life and continuing exposures to estrogenic and xenoestrogenic agents may also contribute to theincreased lethality of breast cancer in young women in general and in African American women of all ages. Publicdisclosure by manufacturers of proprietary hormonally active ingredients is required for this research to move forward.
c 2006 Elsevier Ltd. All rights reserved.
* Corresponding author. Tel.: +1 412 623 3375; fax: +1 412 623 3201.
c 2006 Elsevier Ltd. All rights reserved.
Personal care products that contain estrogens or xenoestrogens may increase breast cancer risk older women , or at the least the relativeimportance of the risk factors may differ according Patterns of breast cancer are enigmatic. Most cases to age. Whether or not there are underlying avoid- occur in women with few established risk factors able risk factors that may account for differences The Gail model was developed from a nested in morbidity and mortality for breast cancer among case–control study conducted on a cohort of white different ages of ethnic and racial groups is a mat- women who were receiving regular screening mam- ter that merits serious examination.
mograms in order to calculate multivariate relative Accepted risk factors for breast cancer include: risks of breast cancer based on age at menarche, age at menarche, age at menopause, age at first age at first live birth, number of first-degree rela- live birth, number of first-degree relatives with tives (mother and sisters) with breast cancer, num- breast cancer, benign breast disease, breast biopsy ber of breast biopsies, and whether or not atypical outcomes, hormone replacement therapy, specific hyperplasia was present on any biopsy specimen.
gene mutations, survival of childhood cancers trea- This model was not intended to predict risk in wo- ted with radiation (particularly thyroid cancers and men under age 40, nor in African American women Hodgkin’s disease), adult weight gain, lack of exer- cise, obesity, radiation exposure, and excess alco- Breast cancer incidence, morbidity and mortal- ity vary across age, ethnic, geographic and racial however, cannot explain variations in breast can- groups, suggesting that there are different underly- cer incidence and mortality among different ethnic ing risks and susceptibilities. For all age groups groups. African American women under the age of combined, African American women are less likely 35 differ from white women in terms of breast can- to develop breast cancer than white women, but at cer risk factors that suggest they should have lower any age, they are more likely to die from it. In con- rates of the disease Thus, although African trast, African American women under age 40 have a American women tend to have more children ear- higher incidence of breast cancer than do whites lier in life, and this is protective against breast can- (15.5 new cases per 100,000 woman-years versus cer in white women, earlier parity may not afford 13.0 per 100,000, respectively). Incidence for Afri- this same protection for African Americans.
can American women ages 20–29 years is nearly Several reports have documented the more fre- 50% higher than for white women of the same age quent and longer term use of PCPs containing hor- (6.5 per 100,000 versus 4.4 per 100,000) . For mones or placenta by African Americans compared all ages combined, African American women also with white women. For treatment of hair and skin, suffer a higher breast cancer mortality rate com- African American females may begin using PCPs that pared to white women at 34.7 deaths per 100,000 contain hormones as infants and toddlers, and they woman-years versus 30.7 per 100,000, respectively also may be exposed in utero when their mothers Simply stated, compared with white women, use these products during pregnancy. These xeno- African American women have a higher incidence hormone-containing products are often used with of breast cancer at younger ages and greater mor- heat, which can enhance penetration and absorp- tion. If estrogens have been used as growth promot- In younger women, breast cancer is more aggres- ers in animal production, then exposure to estrogen sive, more deadly, and often less responsive to may also occur through dietary exposure by eating hormone-enhanced animal protein (beef, chicken).
breast cancer mortality rates among African Amer- Compared with other ethnic groups, most of the evi- ican women as a group can only partially be as- dence suggests that the use of hormone-containing PCPs by African Americans of all ages is more preva- percentage of localized disease, larger and possibly lent. Characterization of the amount and type of more aggressive tumors, and estrogen receptor products currently used by African Americans and negative tumors . In fact, African American wo- the biological consequences of exposure to hor- men appear to be more often diagnosed with a ba- mones contained in these PCPs may lead to the iden- sal subtype of breast cancer that are more tification of agents that contribute to African aggressive. Despite advances in screening and treatment for post-menopausal breast cancer, pro- Several reports have established that estrogen- gress against the disease in younger women has not containing or placenta-containing PCPs resulted in followed suit. Among other considerations, this indicates that cancer in younger women may be or toddlers. In one case report, the use of ointment caused by different factors compared to cancer in containing estrogens on the diaper area caused early sexual development in an 8.5-month-old child . Recent reports indicate that the premature sexual development in children may result fromthe use not only of estrogen-containing PCPs, but Placenta shampoo (contains placental extract)Queen Helene Placenta cream hair conditioner nous hormones, including contraceptives and hor- Placenta revitalizing shampoo (contains placental mone replacement therapy, are also relevant to breast cancer risk. These risk factors, however, cannot explain variations in breast cancer inci- dence and mortality among different ethnic groups Hask Placenta no rinse instant hair repair treatment Several case studies have established that chil- dren who show early sexual development following Nu skin body smoother (human placental extract) exposure to estrogen from products applied to the Nu skin NaPCA moisturizer (human placental extract) skin or ingested through foods have variable serum Nu skin pH balance (human placental extract)Nu skin enhancer (human placental extract) hormone values. For instance, although some chil- dren with premature thelarche show high basal val- Triple action Super Grow (contains hormones) ues of sex hormones at the time of diagnosis, the Supreme Vita-Gro (contains estrogen and allantoin) majority of children have basal values within a nor- Luster’s Sur Glo hormone (contains hormone mones with premature sexual development may B&B super Gro (contains estrogenic hormone be due to: the variable half-life of hormones in the blood, discontinuing the use of the hormone- Lekair natural Super Glo (contains hormones) containing PCPs prior to obtaining a blood speci- Lekair hormone hair treatment with vitamin E men, irregular use of the PCP, or the use of the Isoplus hormone hair treatment with Quinine PCP in quantities that would produce biological ef- Fermodyl with Placenta hair conditioner, no rinse fects but not cause changes in blood hormone lev- Supreme VITA-GRO with allantoin and estrogen plus hormones. In addition, constant exposure to low levels of exogenous hormones may produce biolog- b The presence or absence of a hormone was noted from ical effects. Occasional peak exposures that are the list of ingredients written on the label of the product.
not detectible months after they occur may also Quantitative analysis was performed only on some of the produce effects. It is clear that the biological con- above products (marked with a ‘‘b’’). The hormone analysis sequences of estrogen exposure may persist long was done by Leberco Testing/Inc., Roselle Park, NJ applying after the serum estrogen levels return to normal HPLC for estradiol and estrone and UV spectroscopy forestriol measurement using standard FDA approved methods.
. Furthermore, as infants or toddlers, African Reference: USP XXII, The United States Pharmacopeia, 22nd American girls may begin using hair products con- Revision. United States Pharmacopeia Convention Inc., 1990: taining hormonal substances and they also can be p. 530 (estrone and estradiol), p. 534 (estriol).
exposed in utero when their mothers use theseproducts during pregnancy. While dietary expo-sures to hormone or hormone-mimicking com- pattern are unclear but cannot solely be attributed pounds in animal proteins may also be relevant, the variability in exposure to hormones in food It is significant to note that the American Acad- eaten by African Americans compared with whites emy of Pediatrics lists 16 priority environmental requires further study. The use of estrogen- hazards that should be reduced or controlled to containing PCPs by African Americans, however, protect children . Although associated with may constitute a more specific preventable etio- early sexual development, PCPs are not on the logical risk factor for breast cancer.
Children, in particular African American girls, and its health consequences requires further are experiencing declining age of onset of second- ary sexual characteristics and earlier onset of pub-erty compared with their white counterparts. The proportion of girls who experience early sexual maturation is nearly four times greaterat age 8 for African American than for whites (48.3% It is well established that a woman’s lifetime risk of and 14.7%, respectively) The reasons for this breast cancer increases with greater duration and Personal care products that contain estrogens or xenoestrogens may increase breast cancer risk whether endogenous or exogenous. Estradiol andsynthetic estrogens are established carcinogens A literature review identified 10 reports describing cases of premature sexual development in children mercially available PCPs (). We hypothesize exposed to hormone containing products ).
that in utero and early life exposures to exogenous The development of premature sexual characteris- estrogens, including those added to PCPs and those tics in four African American girls aged 14 months derived from the widespread estrogenic environ- to almost 8 years is one example that shows the mental contaminant bisphenol A (BPA) that variation in age at which premature sexual devel- can be found in foods, make an overlooked and opment was observed following topical application underestimated contribution both to premature of PCPs containing hormones for as little as 2 sexual development and to breast cancer risk.
months after the use of a hair product containing Agents that stimulate breast bud development in placenta or 6 months after the use of a hair product early life may prime estrogen receptors in breast containing estrogens. Serum hormone levels were cells to undergo altered proliferative responses during young adulthood. These in utero, early life Most hair care products contain petrolatum, lan- and ongoing lifetime exposures to estrogens and olin, vegetable, mineral or animal oils Some of xenoestrogens and xenohormones could account these compounds may cause acne or papular erup- in part for the higher incidence of breast cancer tions on the temple or forehead but are not in young African American women. Continued reported to cause premature development of exposure to xenohormones in PCPs throughout life breasts and/or pubic hair. Hormone and/or pla- may also contribute to the increased lethality of centa are ingredients in certain hair products that breast cancer in both younger and older African have been associated with premature sexual devel- opment. Hormone analysis of several of these hair For premenopausal breast cancer, the earlier care products identified estrogens and measured routine ovulation is established the greater the estriol concentrations between 16 and 20 mg/g lifetime risk of breast cancer. In general, for each and concentrations of estradiol at 0.04 mg/g year that menarche is delayed, breast cancer risk Studies do not exist to specify the minimum declines between 10% and 20% . Younger age amount of estrogen that will result in early sexual at menarche translates into earlier estrogen expo- development. Direct comparison of estrogen con- sure, which may partly explain why African Ameri- centration in the published studies cannot be made can women develop breast cancer at younger ages because of the differences in the type, potency, than white women. Case-control studies have con- absorption characteristics and duration of action firmed that serum estrogen levels are higher in of the estrogens contained in these PCPs. For exam- breast cancer cases when compared to controls ple, the amount of estrogen absorbed from a hair care product depends on multiple variables such This hypothesis provides one explanation for as: the temperature of application; the quantity of the higher rates in young African Americans of the product applied; the place and area of applica- premature appearance of secondary sexual char- tion; the duration and frequency of application; var- acteristics and pre-menopausal breast cancer: iation in estrogenic potency and the concentration Xenoestrogenic environmental factors, including of the particular estrogen in the hair product; the those contained in PCPs such as cosmetics, sham- age and health of the child; the presence of any dis- poos, and styling aids, that appear to be more ease of the hair or skin; and individual variations in commonly used by the African American popula- estrogen metabolism. The concentration of a hor- tion, as well as dietary factors that can include mone in a PCP that was considered safe in 1962 , was shown to cause health effects in 1985 may account in part for these patterns. This . Thus, the use of PCPs containing even very hypothesis provides one explanation for the higher low amounts of estrogen during critical windows of rates of premature appearance of secondary sex- development during childhood is problematic.
ual characteristics in young African American girls and the excess incidence of premenopausal breast hypothesis is the fact that the effects of prenatal cancer in young African American women. Persist- or early-life estrogen exposure may be delayed ing exposures to xenohormones in PCPs throughout and subsequently appear many years after the life may also account for the increased lethality of estrogen exposure has been eliminated. Prenatal breast cancer for post-menopausal African Ameri- use of the synthetic hormone, diethylstilbestrol Published studies of hormone-containing products and sexual development in children and adults consistent with (endo)exogenous androgen exposure HCHPs on their childrenoccasionally but 33% of only one showed significant levelsof di-isooctyl phthalate have been classified as endocrinedisruptors (from a restricted sample) wentto a barber therefore exposureis unknown be considered in thedifferential diagnosisof early sexualdevelopment inchildren testosterone and gonadotropinswere depressed HCHP = hormone containing hair-care products.
Personal care products that contain estrogens or xenoestrogens may increase breast cancer risk Basal and LHRH-stimulated serum hormonal values among four african american girls with premature +Serum estradiol (pmol/L) after LHRH injection was 23.13 pmol/L. The value is in the range found in pubertal children.
LH = luteinizing hormone; FSH = follicle stimulating hormone; LHRH = luteinizing hormone-releasing hormone; ND = not done.
a Taken from Ref. b Normal basal serum estradiol value in pre-pubertal children is at or below the lower limit of the sensitivity of the assay.
c Range of the peak serum LH/FSH ratio in children with precocious puberty.
d Range of the peak serum LH/FSH ratio in children with premature adrenarche.
(DES) produced a range of reproductive abnormali- have been associated with premature sexual devel- ties in children that were not evident until early opment in African American girls. The reasons for adolescence. Female children were at risk of devel- this phenomenon are unknown. Body Mass Index oping vaginal adenomatosis during adolescence or (BMI) appears to make a less significant contribu- early adulthood and were at increased risk for tion to early pubertal development in African developing a rare adenocarcinoma of the vagina.
American girls compared with whites . Genetic As a result, DES has been the first recognized trans- and/or environmental factors specific to the placental carcinogen in humans ; males ex- African American population might be important posed prenatally to DES were at risk of impaired contributors to the early onset of puberty fertility Follow up of adolescent girls treated Studies in Ghana, South Africa and Nigeria report with DES and/or ethinyl estradiol,has shown re- that on average, girls reach menarche at between duced fertility in later life . Their mothers have a moderate increase in breast cancer risk from the Therefore, genetic factors are unlikely to be solely use of DES during pregnancy; a risk that their responsible because, as mentioned above, African blacks do not share the characteristic of prematuresexual maturation with African Americans.
Studies do not exist to specify the minimum amount of estrogens that will result in early sexualdevelopment. Doses of the estrogens that were in- The Gail Model and variations of it are commonly volved in the published case studies of premature used to assess an individual woman’s risk of devel- sexual development cannot be determined because of the differences in the type, potency, absorption served characteristics of women with breast characteristics and duration of action of the estro- cancer in a population that consisted chiefly of Some studies have found that African American widely used model does not predict breast cancer adults and children use PCP-containing hormones risk in young women generally and should not be about 6–10 times more frequently than do whites mates genetically inherited breast cancer because timate the frequency of use of the hormone or pla- it does not take into account paternal history.
centa containing hair products for a number of Other limitations of the Gail model are that it ne- reasons that include: (1) the surveys used to cap- glects family history information in second-degree ture product use did not include all the hair prod- relatives, it treats pre- and postmenopausal breast ucts that contain hormone or placenta; (2) some cancer as if they are the same disease, and it hair products may contain hormones but not list ignores personal histories of lobular neoplasia .
it as an ingredient preventing the user from being Early sexual maturation and/or pubic hair is made aware of their exposure to the hormone or more common among African American females placenta containing hair product; or (3) hair prod- than white females . Hormones and/or pla- ucts intended to be used by professionals (barbers centa are ingredients in certain hair products that and hair dressers) may not list the ingredients at all, therefore, a person visiting a barber will not research on this topic. In the Cosmetic Handbook know if he/she is exposed to a hormone or placenta published in 1992 the FDA describes proposed legis- lation to regulate estrogen concentration in over- use of PCPs from a young age combined with earlier the-counter products both as a drug and through estrogen exposure as a result of early age at men- misbranding. Estrogen concentration would be lim- arche may increase cumulative estrogen exposure ited to less than 10,000 IU per ounce and total and thereby stimulate the development of breast exposure for an individual is limited to 20,000 IU per month with no detail regarding size, age or The National Institutes of Environmental Health sex of consumers who might be exposed . In Sciences (NIEHS) Center grants on breast cancer 1992 the ‘‘panel’s recommendation [had] not yet and the environment are being used by researchers been accepted by the FDA as a basis for regulatory at several institutions to support epidemiologic decisions.’’ In 1994, the FDA began to officially reg- studies to evaluate the role of early estrogen expo- ulate products listing hormones as ingredients un- sure from a variety of sources including PCPs, in der their authority to regulate new drugs. ‘‘Any African American and white women. Other work ‘over-the-counter’ drug or a product that is la- being conducted as part of the Sister Study intra- beled, represented, or promoted as a topically ap- murally at NIEHS will evaluate the use of PCPs in plied hormone-containing product . . . is regarded 50,000 women. It will also examine the racial dis- as a new drug . . . for which an approved application tribution of breast cancer patients who use such . . . is required for marketing.’’ Since 1994 any PCP products, and will include specific questions on labeled to indicate that it is a ‘‘hormone cream’’ the use of hair products containing hormone or pla- or that it ‘‘contains hormones’’ is considered a centa, duration of use, estrogen content in the drug by the FDA and falls under their regulatory products and type of estrogen (estrone, estradiol, and estriol). In the case of placenta-containing It is important to note, however, that the biolog- products, where possible, this work will provide ical activity of estrogen varies depending on the additional analysis for human chorionic gonadotro- type of estrogen used as an ingredient. Of the nat- pin, luteinizing hormone, and follicle stimulating urally occurring estrogens, estradiol is most active, estrone has one-tenth the biological activity of To test this hypothesis it will be important for estradiol, and estriol has the lowest potency .
researchers to ascertain past use and exposure to The synthetic estrogen, DES, is more active than these products in women being studied. As men- all three natural estrogens. Most manufacturers tioned previously, definite regression or non-pro- do not list either the amount or the concentration gression of premature sexual characteristics has of estrogen on their product labels .
been reported in several children after the use of In addition, manufacturers are not currently re- hair product containing hormone or placenta has quired to disclose ingredients that they consider been discontinued. The resolution of the breast trade secrets nor are they required to report on may take months or years, however, in some cases past formulations of their products. It is a matter of considerable importance for public health to reason for the non-resolution is uncertain, but determining whether or not such estrogenic expo- may be related to onset of early puberty or activa- sures play a role in the unexplained patterns of tion of the hypothalamic pituitary system after breast cancer in young African American women, prolonged exposure to the estrogens . If it is or contribute to the poorer prognosis for African determined that exposure to estrogenic com- American women at all ages. Consistent with pounds contained in PCPs is a risk factor that af- established theories of breast cancer etiology, fects morbidity, mortality or initiation of breast early-life exposure to estrogenic agents is ex- cancer, it may be possible to modify the natural pected to increase lifetime risk of the disease, as history of breast cancer in exposed population by is continued exposures to hormonally active com- eliminating estrogenic ingredients from PCPs.
People have a right to know whether products they use on themselves or their children containcompounds that may increase their risk of disease, including cancer. Under current policy in manycountries that right is denied. As a matter of public The absence of public information on past expo- health policy, manufacturers should be required to sures to hormones and hormone-mimicking com- provide information on current and past inclusion pounds in PCPs hampers the ability to conduct of hormones in their PCPs. In the meantime, par- Personal care products that contain estrogens or xenoestrogens may increase breast cancer risk ents and guardians should be advised to avoid using [12] Swanson MG, Haslam SZ, Azzouz F. Breast cancer among any products on children that are known or sus- young African-American women: a summary of data andliterature and of issues discussed during the Summit pected to contain hormones and/or placenta. Ef- Meeting on Breast Cancer among African American women forts to promote voluntary release of ingredients in PCPs by manufacturers should be considered a priority so that consumers, who may use these [13] Armstrong K, Eisen A, Weber B. Assessing the risk of breast products on themselves and their children, can [14] Wolff MS, Britton JA, Wilson VP. Environmental risk factors make better and more informed choices.
for breast cancer among African-American women. Cancer2003;97(Suppl. 1):289–310.
[15] Halperin DS, Sizonenko PC. Prepubertal gynecomastia following topical inunction of estrogen containing oint- ment. Helv Paediatr Acta 1983;38:361–6.
[16] Lewis R. A study of the effects of theelin on gonorrheal Heinz Endowments, DSF Charitable Trust, and the vaginitis in children. Am J Obstet Gynecol 1933;26:593–9.
University of Pittsburgh Cancer Institute provided [17] Hertz R. Accidental ingestion of estrogens by children.
[18] Edidin DV, Levitsky LL. Prepubertal gynecomastia associ- ated with estrogen-containing hair cream. Am J Dis Child1982;136:587–8.
[19] Whittle CL. Precocity in a girl aged 5 due to stilboestrol inunction. Proc R Soc Med 1948;41:760.
[1] Bernstein L, Teal CR, Joslyn S, Wilson J. Ethnicity-related [20] Healy CE. Precocious puberty due to a diaper ointment.
[21] Zimmerman PA, Francis GL, Poth M. Hormone-containing [2] Gail MH, Costantino JP. Validating and improving models personal care products may cause signs of early sexual for projecting the absolute risk of breast cancer. JNCI development. Mil Med 1995;160:628–30.
[22] Tiwary CM. Premature sexual development in children [3] Surveillance, Epidemiology, and End Results (SEER) Pro- following the use of estrogen- or placenta-containing hair products. Clin Pediatr (Phila) 1998;37:733–9.
dence – SEER 9 Regs Public-Use, November 2005 Sub [23] Moore RJ, Chamberlain RM, Khuri FR. Apolipoprotein E and (1973–2003), National Cancer Institute, DCCPS, Surveil- the risk of breast cancer in African-American and non- lance Research Program, Cancer Statistics Branch, released Hispanic white women. A review. Oncology 2004;66:79–93.
April 2006, based on the November 2005 submission.
[24] Cook CD, McArthur JW, Berenberg W. Pseudoprecocious Calculations used SEER Cancer Query Systems (1973– puberty in girls as a result of estrogen ingestion. N Engl J [4] Surveillance, Epidemiology, and End Results (SEER) Pro- [25] Ramos AS, Bower BF. Pseudo-sexual precocity due to personal care product ingestion. JAMA 1969;207:368–9.
– All COD, Public-Use With State, Total U.S. (1969– [26] Freedman DS, Kettel Khan L, Serdula MK, Srinivasan SR, 2003), National Cancer Institute, DCCPS, Surveillance Berenson GS. The relation of age at menarche to race, time Research Program, Cancer Statistics Branch, released April period, and anthropometric dimensions: the Bogalusa heart [27] Herman-Giddens ME, Slora EJ, Wasserman RC, et al.
[5] Hall IJM, Patricia G, Millikan RC, Newman B. Comparative Secondary sexual characteristics and menses in young girls analysis of breast cancer risk factors among African- seen in office practice: a study from the Pediatric Research American women and white women. Am J Epidemiol in Office Settings network. Pediatrics 1997;99:505–12.
[28] Kaplowitz PB, Slora EJ, Wasserman RC, et al. Earlier onset [6] Aziz H, Hussain F, Sohn C, et al. Early onset of breast of puberty in girls: relation to increased body mass index carcinoma in African American women with poor prognostic and race. Pediatrics 2001;108:347–53.
factors. Am J Clin Oncol 1999;22:436–40.
[29] Saenz CA, Toro-Sola M, Conde L, Bayonet-Rivera NP, et al.
[7] Bonnier P, Romain S, Charpin C, et al. Age as a prognostic Premature thelarche and ovarian cyst probably secondary factor in breast cancer: relationship to pathologic and to estrogen contamination. Bol Asoc Med P R 1982;74:16–9.
biologic features. Int J Cancer 1995;62:138–44.
[30] Fara GM, Del Corvo G, Bernuzzi S, et al. Epidemic of breast [8] English WP, Cleveland KE, Barber WH. There is no differ- enlargement in an Italian school. Lancet 1979;2:295–7.
ence in survival between African-American and white [31] Styne DM. Growth. In: Greenspan FS, Baxter JD, editors.
women with breast cancer. Am Surg 2002;68:594–7.
Basic and clinical endocrinology. Norwalk, CT: Appleton & [9] Fowble BL, Schultz DJ, Overmoyer B, et al. The influence of young age on outcome in early stage breast cancer. Int J [32] American Academy of Pediatrics, Division of Health Policy Radiat Oncol Biol Phys 1994;30:23–33.
Research. Cigarette smoking tops lists of environmental [10] Golledge J, Wiggins JE, Callam MJ. Age-related variation in hazards: AAP Survey. AAP News 1999;15:17.
the treatment and outcomes of patients with breast [33] Estrogens in personal care products. Med Lett 1985;27:54– [11] Nixon AJ, Neuberg D, Hayes DF, et al. Relationship of [34] Report on Carcinogens, Eleventh Edition, 2005. Substance patient age to pathologic features of the tumor and Profiles: Estrogens, Steroidal*, U.S. Department of Health prognosis for patients with stage I or II breast cancer. J and Human Services, Public Health Service, National [35] Magdalena PA, Morimoto S, Ripoll C, Fuentes E, Nadal A.
[51] Gail MH, Benichou J. Validation studies on a model The estrogenic effect of Bisphenol A disrupts pancreatic b- for breast cancer risk. J Natl Cancer Inst 1994;86: cell function in vivo and induces insulin resistance. Environ Health Perspect 2006;114(1):106–12.
[52] Spiegelman D, Colditz GA, Hunter D, Hertzmark E. Valida- [36] Taylor SC. Skin of color: biology, structure, function, and tion of the Gail et al. model for predicting individual breast implications for dermatologic disease. J Am Acad Dermatol cancer risk. J Natl Cancer Inst 1994;86:600–7.
[53] Euhus DM. Understanding mathematical models for breast [37] Strauss JS. Hormones in personal care products. JAMA [38] Robboy SJ, Noller KL, O’Brien P, et al. Increased incidence [54] Cameron N, Wright CA. The start of breast development of cervical and vaginal dysplasia in 3,980 diethylstilbestrol- and age at menarche in South African black females. S Afr exposed young women. Experience of the national collab- [55] Adadevoh SW, Agble TK, Hobbs C, Elkins TE. Menarcheal age in Ghanaian school girls. Int J Gynaecol Obstet [39] Klip H, Verloop J, van Gool JD, Koster META, Burger CW, van Leeuwen FE. Hypospadias in sons of women exposed to [56] Fakeye O, Fagbule D. Age and anthropometric status of Nigerian girls at puberty: implication for the introduction of sex education into secondary schools. West Afr J Med [40] Venn A, Bruinsma F, Werther G, et al. Oestrogen treatment to reduce the adult height of tall girls: long-term effects on [57] Cameron N, Mitchell J, Meyer D, et al. Secondary sexual fertility. Lancet 2004;364:1513–8.
development of ‘Cape Coloured’ girls following kwashior- [41] Giusti RM, Iwamoto K, Hatch EE. Diethylstilbestrol revis- ited: a review of the long-term health effects. Ann Intern [58] Tiwary CM. A survey of use of hormone/placenta-containing hair preparations by parents and/or children attending [42] Hatch EE, Palmer JR, Titus-Ernstoff L, et al. Cancer risk in pediatric clinics. Mil Med 1997;162:252–6.
women exposed to diethylstilbestrol in utero. JAMA [59] Tiwary CM, Ward JA. Use of hair products containing hormone or placenta by U.S. military personnel. J Pediatr [43] DES and breast cancer: 30+ years later. Journal Watch [60] Li ST, Lozano P, Grossman DC, Graham E. Hormone- [44] Gail MH, Brinton LA, Byar DP, et al. Projecting individual- containing hair product use in prepubertal children. Arch ized probablities of developing breast cancer for white Pediatr Adolesc Med 2002;156:85–6.
[61] Donna Baird, personal communication; 2006.
[62] Gabrilove JL, Luria M. Persistent gynecomastia resulting [45] Bondy ML, Lustbader ED, Halabi S, Ross E, Vogel VG.
Validation of a breast-cancer risk assessment model in women with a positive family history. J Natl Cancer Inst [63] Kelch RP, Beitins IZ. Adolescent Sexual Development. In: Blizzard RM, Migeon CJ, Kappy MS, editors. Wilkins, The [46] Rockhill B, Spiegelman D, Byrne C, Hunter DJ, Colditz GA.
Diagnosis and treatment of endocrine disorders in child- Validation of the Gail et al. model of breast cancer risk hood and adolescence. Springfield, Illinois: CC Thomas; prediction and implications for chemoprevention. J Natl [64] U.S. Food and Drug Administration, Center for Food Safety [47] MacKarem G, Roche CA, Hughes KS. The effectiveness of and Applied Nutrition, FDA/Industry Activities Staff Book- the Gail model in estimating risk for development of breast let: 1992. Cosmetic Handbook: Products Containing Estro- cancer in women under 40 years of age. Breast J genic Hormones, Placental Extract of Vitamins. (available: [48] Costantino JP, Gail MH, Pee D, et al. Validation studies for models projecting the risk of invasive and total breast [65] U.S. Food and Drug Administration. Topically applied cancer incidence. J Natl Cancer Inst 1999;91:1541–8.
hormone-containing drug products for over-the-counter [49] Wingo PA, Ory HW, Layde PM, Lee NC. The evaluation of (OTC) human use. 1995, 21 Code of Federal Regulations the data collection process for a multicenter, population- based, case–control design. Am J Epidemiol 1988;128: [66] Strobl JS. Estrogens, progestins, and specific estrogen receptor modulators (SERMs). In: Craig Charles R, Stitzel [50] Gail MH, Benichou J. Assessing the risk of breast cancer in Robert E, editors. Modern pharmacology with clinical individuals. In: DeVita Jr VT, Hellman S, Rosenberg SA, applications. 6th ed. Philadelphia: Lippincott, Williams editors. Cancer prevention. Philadelphia (PA): Lippincott; and Wilkins; 2004. p. 704–15 (Chapter 61).
[67] Chandra Tiwary, personal communication; 2006.

Source: http://www.camerasante.org/doc/Donovan_cosmetics.pdf