Prolactin and Breast Cancer Etiology: An Epidemiologic Perspective
- 491 Downloads
A number of epidemiologic studies of prolactin and breast cancer etiology have recently become available. Retrospective case-control studies have suggested a modest positive or null relationship between circulating prolactin concentrations and risk of breast cancer. However these studies are limited by small sample sizes and the collection of blood after case diagnosis. Several large prospective studies, in which blood was collected prior to diagnosis, have observed modest positive associations between prolactin and risk. In a pooled analysis of ~80% of the world’s prospective data, the relative risk (RR) comparing women in the top vs bottom quartile of prolactin levels was 1.3 (95% confidence interval (CI): 1.1, 1.6, p-trend = 0.002). The results were similar for premenopausal and postmenopausal women. Most notably, high prolactin levels were associated with a 60% increased risk of estrogen receptor (ER) positive tumors, but not with ER negative tumors. Limited genetic data suggest a role of polymorphisms in the prolactin and prolactin receptor genes in risk of breast cancer. Studies of survival have suggested that high pretreatment prolactin levels were associated with treatment failure, earlier recurrence, and worse overall survival. Parity and certain medications are the only confirmed factors associated with prolactin levels in women. Overall, epidemiologic data suggest that prolactin is involved in breast cancer etiology. Further research to better elucidate these associations and their underlying mechanisms is warranted.
KeywordsProlactin Breast cancer risk Breast cancer survival Menopausal status
Nurses’ Health Study
Nurses’ Health Study II
single nucleotide polymorphism
insulin-like growth factor
Prolactin, a polypeptide hormone, is important for the proliferation and differentiation of normal mammary epithelium, and high levels are necessary to stimulate lactation, although levels return to baseline after several months . Further, during pregnancy prolactin levels increase and, in conjunction with estrogens and progesterone, lead to full lobuloalveolar development of the breasts . Due to the importance of prolactin in breast development and function, especially during pregnancy, it has been examined in relation to breast cancer in a number of epidemiologic studies.
Recent evidence suggesting that prolactin is locally produced within breast tissue [3, 4, 5] has further bolstered attention on its potential involvement in breast cancer etiology. Additionally, a number of animal and in vitro studies have suggested that prolactin may be involved in mammary tumorigenesis  by promoting cell proliferation and survival [6, 7, 8, 9], increasing cell motility , and supporting tumor vascularization [1, 11], which are discussed in more detail in other chapters. This review specifically focuses on epidemiologic studies of prolactin in relation to breast cancer risk, etiology, and survival; overall these studies provide strong support for an important role of prolactin. We further discuss potential correlates of prolactin levels in women, concentrating on lifestyle factors, reproductive factors, medications, mammographic density, and family history of breast cancer.
Prolactin and Breast Cancer Risk
Epidemiologic data on the relation between prolactin and breast cancer have increased over the last several years. Prolactinomas, a condition characterized by extremely high prolactin levels, may be associated with breast cancer development (see number 9 of Correlates of Prolactin Levels in Women section). One study examined the pituitary gland at autopsy of 125 women who died of breast cancer and 85 controls dying from other causes. There were no differences in the accumulation of hyperplastic or adenomatous prolactin secreting cells in cases vs controls . However, given that prolactin can be produced by breast cells, the pituitary gland may not be the site of most interest for breast cancer .
Retrospective case-control studies of the relationship between prolactin and breast cancer generally have been small and the results inconsistent. One serious limitation of this study design is that prolactin levels are assessed after the woman has been diagnosed with breast cancer; levels at this time may not reflect predisease exposures. Specifically, physical (e.g., surgery) and psychological stress (e.g., knowing one has cancer) among breast cancer patients can alter prolactin levels [13, 14]. This could bias results when assessing the relationship between prolactin and risk of breast cancer, possibly leading to a false positive result. Also, tamoxifen, a common treatment for breast cancer, may lower prolactin concentrations , although most studies were conducted when tamoxifen was not commonly prescribed.
To address this limitation, seven studies (n = 23–233 cases) collected blood samples from cases after diagnosis but prior to surgery to obtain a potentially better measure of prolactin exposure [16, 17, 18, 19, 20, 21, 22]. Of these, five reported higher levels of prolactin in cases vs controls in both premenopausal [16, 17, 21, 22] and postmenopausal [17, 18, 21, 22] women; the other studies reported no association [19, 20]. Interestingly, Rose et al. , observed that among control women, prolactin levels were inversely correlated with age, but this same association was not observed in cases, suggesting that cases had higher levels over time.
Among six studies collecting blood after breast cancer treatment, most [22, 23, 24, 25, 26], but not all , required a minimum time period after the end of treatment before blood collection (range of 1 to 6 months). This delay may allow prolactin concentrations to return to baseline levels [28, 29]. The number of cases ranged from 18 to 115. Half of the studies reported positive associations with a two-fold increase in breast cancer risk comparing high vs low prolactin levels [21, 22, 24]. The three other studies reported no associations [23, 25, 30], but two had 40 or fewer cases [23, 25]. Further Kwa et al.  while finding no association overall, did observe higher prolactin levels among familial cases of breast cancer vs controls. Four studies where the timing of blood draw was not clear, observed both positive [31, 32] and null results [33, 34].
The results of the retrospective studies are mixed, with 10 of 17 reporting a positive association between prolactin and breast cancer risk, and the remaining reporting no association; this suggests that if high prolactin levels are associated with breast cancer, they likely increase risk. However, given the small sample sizes and the methodologic issues of prolactin measurement in breast cancer cases, these studies only modestly contribute to the epidemiologic evidence for an association between circulating prolactin levels and breast cancer risk. Prospective studies, where blood samples are collected prior to diagnosis are important to establish a temporal association between increased prolactin levels and breast cancer risk.
Several prospective studies have been completed in both premenopausal and postmenopausal women. Three of five studies conducted in premenopausal or perimenopausal women with 71 or fewer cases did not find any significant associations [35, 36, 37]. One study of over 5,000 women living on the island of Guernsey had up to 22 years of follow-up and accrued 71 cases . The relative risk (RR) comparing the top vs bottom quintiles of prolactin levels was 1.1 [95% confidence interval (CI): 0.5, 2.2]. The Washington County Cohort with up to 17 years of follow-up and 21 cases, observed a RR comparing the top vs bottom tertile of prolactin of 1.1 (95% CI: 0.3, 4.1) . A third study, among Japanese women, accrued 46 cases over 13 years of follow-up, with a RR for a one unit increment in log10 prolactin concentrations of 1.0 (95% CI: 0.02, 47.4) . The number of cases in these studies was very small, precluding the ability to detect even a moderate to strong association between premenopausal prolactin levels and breast cancer risk.
Among postmenopausal women, two small prospective studies (n ≤ 40 cases) reported non-significant positive associations with prolactin [35, 36]. Wang et al.  observed a RR of 1.6 (95% CI = 0.6, 4.7) comparing the top vs bottom quintiles of prolactin levels, and Kabuto et al.  reported a RR of 6.5 (95% CI = 0.01, 43.9) for a unit increase in log10 prolactin. A larger study with 173 cases also reported a non-significant positive association (RR, top vs bottom quartile: 1.3, 95% CI: 0.8, 2.2) . These studies likely lack the statistical significance needed to detect modest associations due to the relatively small sample sizes. Despite this, these results suggest a modest positive association between prolactin and postmenopausal breast cancer.
Two previous reports from the NHS among postmenopausal women observed an increased risk of breast cancer with higher prolactin levels [41, 42]. Recently, we combined these 850 cases with 65 cases identified in the NHSII, representing over 10 years of follow-up  and 79% of the published postmenopausal cases. There was a 30% increase in breast cancer risk (p-trend = 0.01) for women with the highest vs lowest levels (Fig. 1). This point estimate is similar to that from Manjer et al. , which was not statistically significantly but had fewer cases. Perhaps more importantly, the association between prolactin and breast cancer did not appear to differ by menopausal status (p-interaction between premenopausal and postmenopausal women = 0.95). Thus for the analyses mentioned below, women of all menopausal statuses were combined.
In this combined study, the prolactin association was similar for invasive vs in situ (p-heterogeneity = 0.25) as well as ductal vs lobular cases (p-heterogeneity = 0.26 ). However there was a clear difference in the results by estrogen receptor (ER) and progesterone receptor (PR) status. Specifically the RR comparing the top vs bottom quartile of prolactin levels was 1.6 (95% CI = 1.3–2.0) for ER+/PR+, 1.7 (95% CI = 1.0–2.7) for ER+/PR−, and 0.9 (95% CI = 0.6–1.3) for ER−/PR− breast cancers. We had too few ER−/PR+ cases (n = 32) to consider separately. Given the similar RRs for ER+/PR+ and ER+/PR−, the association was assessed for all ER+ cases combined, with a 60% increased risk comparing the top vs bottom quartile of levels (Fig. 1). No association was observed for ER− tumors. Further the association with prolactin was largely unchanged after adjustment for either estradiol or testosterone levels, two known risk factors for breast cancer . Overall these results suggest that prolactin is linked specifically with hormone-sensitive tumors, but appears to act independently of other important sex hormones.
Further, there was a suggestion that prolactin was more strongly associated with risk of breast cancer among women who were diagnosed within 4 years (vs ≥4 years) of blood collection, although the difference was not statistically significant (p-heterogeneity = 0.40) . For cases diagnosed 0–2 and 2–4 years after blood collection the RRs comparing the top vs bottom quartile of levels was 1.9 and 1.8, respectively. For cases diagnosed 4–8 and 8+ years after blood collection the corresponding RRs were 1.4 and 1.5 respectively. All associations were statistically significant. These data suggest that prolactin may play a greater role in late-stage promotion of breast cancer or that preclinical breast tumors, many of which secrete prolactin, can increase plasma concentrations .
All prospective studies to date have only had one blood sample per participant—the prolactin concentration measured in this sample estimates exposure over the entire follow-up period. However, one measure of prolactin may not reflect long-term hormone levels well. For example the intraclass correlation of prolactin concentrations within the same woman over a 1- to 3-year period ranges from 0.40–0.64 in premenopausal women [45, 46, 47] and 0.18–0.76 in postmenopausal women [45, 47, 48]. Thus, because prolactin levels are only modestly stable within a woman over time, risk estimates from prospective studies are likely attenuated toward the null. It is possible to correct for this type of random measurement error by conducting a reproducibility study of women who give blood samples over several years. In the large NHS and NHSII study, we used a reproducibility study of 113 women with three blood samples collected over three years, to correct point and interval estimates for laboratory measurement error and random within-person variation [38, 39]. After correction, the RR comparing the median level in the top vs bottom quartile, increased from 1.3 to 1.7. Similarly, among ER+ cases, the RR increased from 1.5 to 2.1. The corrected relative risks are similar to those observed for testosterone and are only slightly weaker than those observed for estradiol . These data suggest that prolactin likely has a greater role in the risk of breast cancer than previously thought.
Other methodologic issues are important to consider when interpreting the epidemiologic data on prolactin and breast cancer risk. First, because of autocrine/paracrine production of prolactin in breast tissue , circulating levels are only an indirect marker of the most relevant exposure, i.e., that in the breast tissue. Unfortunately few data are available concerning the relationship between circulating and breast tissue levels. A small study of breast cancer patients treated with a hypophysectomy reported that many women regained near-normal circulating prolactin levels within several weeks of surgery, suggesting that at least some circulating prolactin may be derived from breast tissue . Further, immunohistochemical staining of prolactin in breast tumors was correlated (r = 0.41) with plasma prolactin concentrations .
Secondly, the prolactin assay used in nearly all epidemiologic studies to date measures multiple prolactin isoforms (e.g., big prolactin, glycosylated prolactin); these isoforms may have different biological activities [51, 52]. This increases the error in the exposure measure, attenuating the observed associations. Assays to specifically measure particular prolactin isoforms are difficult, time-intensive, and require large amounts of plasma and hence generally are not feasible in epidemiologic studies [51, 52]. However the Nb2 lymphoma cell bioassay is a sensitive measure of overall somatolactogenic (prolactin and growth hormone) activity in biological fluids . This measure and the ratio between the prolactin bioassay and immunoassay have been evaluated in several studies of systemic lupus erythematosus and found to be of importance [51, 52]. One breast cancer case-control study reported that prolactin levels measured by immunoassay were not different between cases and controls; however, mean bioassay levels and the bioassay/immunoassay ratio were significantly higher in cases vs controls , although another study reported no such association .
Less is known about whether genetic variability in the prolactin or prolactin receptor gene is important in regulating prolactin levels or breast cancer risk. The prolactin gene is composed of 5 exons and 4 introns that span approximately 10 kilobases , and has two independent promoter regions, controlling pituitary-specific (proximal) and extrapituitary (upstream) expression [56, 57]. The prolactin receptor gene contains ten exons with five alternative exon 1 sequences . A functional G/T SNP in the prolactin gene at position –1149 in the extrapituitary promoter leads to an extra GATA transcription factor binding site . However, the wild-type allele had a higher promoter activity and was associated with increased stimulation of prolactin in lymphocytes .
To our knowledge two studies has examined this and other SNPs in relation to breast cancer risk. Vaclavicek et al. , examined seven SNPs among 441 familial breast cancer cases and 522 age-matched controls. Two SNPs in the prolactin gene, including the functional -1149G/T SNP, were associated with a 70 to 110% increased risk of breast cancer. For the prolactin receptor gene, none of the three SNPs examined were individually associated with breast cancer risk; however one haplotype combining all three SNPs was associated with a reduced risk of breast cancer (RR = 0.7). Having 3 or 4 high risk alleles/haplotypes vs none was associated with a RR of 2.6 (95% CI: 1.4, 4.6, p-trend = 0.007). In the Multi-Ethnic Cohort study, examined 33 SNPs in the prolactin gene and 60 SNPs in the receptor among 1,615 breast cancer cases and 1,962 controls . They reported associations with breast cancer risk for one SNP in the prolactin gene (rs9466314, RR = 1.5) and one in the receptor gene (rs34024951, RR = 0.85); there was no association −1149G/T SNP or with haplotypes made from the SNP results. Further, among 362 postmenopausal controls, they examined these SNPs with circulating prolactin levels. Although several SNPs were suggestively associated with levels, no associations were statistically significant due to the relatively small sample size. Given these intriguing results, it will be important to continue studying the role of genetic variability in these genes both for breast cancer risk and prolactin levels.
Prolactin and Breast Cancer Survival
In the 1980s several clinical trials were conducted to examine whether drugs that reduced prolactin levels could be used as a treatment for breast cancer . The trials were unsuccessful, probably because the drugs only inhibited pituitary prolactin production, rather than breast tissue specific production. However a number of studies have examined both staining of prolactin and prolactin receptor in breast tumor tissue and prolactin levels pre- and posttreatment with survival.
While prolactin receptor staining in breast tumors generally has not been associated with tumor characteristics including tumor grade or stage [62, 63, 64, 65], it may be associated with tumor differentiation and age at diagnosis [66, 67, 68, 69]. However, positive receptor staining does not appear to be associated with disease-free survival or treatment response . Positive staining of the prolactin hormone also has been observed in about 80% of malignant breast tumors [50, 70]. Increased prolactin positivity was significantly associated with increased tumor size, higher stage, nodal involvement, and a worse overall survival in univariate analyses, although the association was attenuated after adjustment for stage .
Circulating levels of prolactin either pre- or posttreatment have been examined in relation to several tumor and patient characteristics. No consistent associations have been observed for tumor size [29, 50, 71], ER status , stage [50, 71, 72], or nodal status [13, 14, 28, 29, 50, 71]. One study of 433 cases reported higher pretreatment prolactin levels among women with grade 3 tumors vs grades 1 and 2 . Similarly, that study and a second study observed higher prolactin levels in cases diagnosed before age 50 [29, 72]. Perhaps, most interestingly is a consistent relationship between high prolactin levels and appearance of metastases [17, 18, 73]; these studies reported that prolactin levels appeared to rise before the detection of metastatic disease [17, 73].
High pretreatment prolactin concentrations also have been associated with treatment failure for both tamoxifen and aromatase inhibitors in most [74, 75, 76], but not all , studies. Interestingly, Frontini et al. , conducted a study of 24 metastatic breast cancer cases with hyperprolactinemia, randomizing 11 women to Taxotere and an antiprolactin drug and 13 women to Taxotere alone. Women on the combination therapy had a normalization of their prolactin levels, while those on Taxotere alone did not. Further, incidence of tumor regression was higher among women taking the antiprolactin drug versus not, suggesting that inhibition of prolactin may work as an adjuvant therapy in combination with Taxotere.
Correlates of Prolactin Levels in Women
Breast cancer risk factors associated with higher circulating prolactin levels in women: a summary of the evidence.
Breast cancer risk factors
Confirmed association with prolactin levels
Oral contraceptive use
Possible association with prolactin levels
Family history of breast cancer
Increased mammographic breast density
No evidence of association with prolactin levels
Age at menarche
Age at first birth
Age at menopause
Data too limited to draw conclusions
Low physical activity
Parity and age at first birth. A long lasting reduction in prolactin levels after a first pregnancy has been consistently observed [24, 25, 80, 81, 82, 83] in both premenopausal and postmenopausal women suggesting that the inverse association observed between parity and breast cancer risk may, at least in part, be due to a reduction in prolactin exposure. The percent reduction in levels has varied substantially by study with a range of 15–50% when comparing nulliparous to parous women. In the few studies with a large enough sample size for a detailed assessment [24, 80, 83], prolactin levels decreased either significantly or nonsignificantly with each additional pregnancy; however, the decrease associated with the first pregnancy appeared larger than subsequent decreases. Results from one recent study suggested that the reduction in prolactin levels attenuates slowly over time since a woman’s first birth, such that a larger reduction was observed in premenopausal versus postmenopausal women . No independent association between age at first birth and prolactin levels has been observed [80, 83]. Interestingly, circulating prolactin levels during the 2nd trimester of pregnancy (controlling for age and gestational age) were lower in women who had one or more prior pregnancies compared to those pregnant for the first time .
Family history of breast cancer. In studies of adolescents [87, 88, 89] or postmenopausal women [24, 80, 83], no difference in circulating prolactin levels according to family history of breast cancer has been observed. In studies of premenopausal women, the data are less consistent. In most studies [24, 80, 81, 90, 91, 92], investigators observed modestly higher prolactin levels (primarily luteal phase) in women with vs without a family history of breast cancer. However other studies [25, 90], including the largest study to date , no association was observed. Reasons for these inconsistencies are not clear, but the data suggest that any relationship, if it exists, likely is modest in magnitude and may be restricted to certain as yet poorly defined subgroups of women. To our knowledge, the relationship between prolactin levels and specific gene mutations (e.g., BRCA1) has not been assessed.
Mammographic density. Breast density, defined as the areas of opacity on a mammogram, reflects the amount of breast epithelial and stromal tissue, and is a strong positive risk factor for breast cancer . In the only assessment where premenopausal women were evaluated separately, no association was observed . In studies among postmenopausal women, no association [96, 97, 98] a significant positive association [95, 99], and a weak positive association  have been reported. In several of the null studies [96, 98], time of day when the blood was drawn and fasting status—two factors that substantially alter prolactin levels—were not considered in the analysis, thus complicating the interpretation. Hence further assessments in carefully controlled studies are warranted.
Ethnic differences. Prolactin levels have been assessed in adolescents or women defined as being at high or low risk of breast cancer according to breast cancer rates in their country of origin. In general, no substantial differences were observed when comparing average levels in women (or adolescents) from the United States or Britain (high risk countries) to those in rural Japan or China (low risk countries [101, 102, 103]). A recent study of African-American, Asian-American and Caucasian American premenopausal women observed no differences in circulating prolactin concentrations . Several studies also have assessed possible ethnic differences in prolactin levels during pregnancy. Prolactin levels were significantly lower in U.S. versus Chinese women at both week 16 and week 27 of pregnancy, which remained after control for maternal age and parity . Similarly, among U.S. women at 6 to 20 weeks gestation, Caucasians had lower prolactin levels during pregnancy compared to both African-American and Hispanic women after accounting for age, gestational age, education and parity . A third study also reported lower 2nd trimester prolactin levels in Caucasian versus Hispanic women , although in this report, levels in African-American and predominantly second generation Asian-American women were generally similar to levels in Caucasian women.
Body size and physical activity. Body mass index (a measure of adiposity, BMI) and physical activity are important factors in breast cancer etiology . To date, either no association [108, 109] or a weak positive association [30, 110] has been observed between BMI and circulating prolactin levels in premenopausal and postmenopausal women. Several [111, 112, 113, 114] although not all  studies have reported an acute but transient elevation of prolactin after an episode of vigorous exercise, but data on any long term influence on prolactin levels are sparse. In a 12-month exercise trial in postmenopausal women, moderate intensity exercise had no effect on prolactin levels, although among exercisers, increased physical fitness was associated with reduced prolactin levels . Hence, to date, prolactin seems to be only weakly related if at all to BMI, and the influence of physical activity is not yet well delineated.
Dietary intake. Relatively few dietary factors have been consistently associated with risk of breast cancer. Alcohol intake has been most consistently related to an increase in breast cancer risk, but in a single study, moderate intake was uncorrelated with postmenopausal prolactin levels . Several studies have evaluated prolactin levels and either dietary fat [24, 117, 118, 119, 120, 121] or soy intake , factors hypothesized to influence breast cancer risk, but consistent findings have yet to emerge.
Medication use. A number of medications are known to increase (e.g., oral contraceptives, reserpine, haldol, cimetidine and the phenothiazines) or decrease (e.g., levodopa) plasma prolactin levels. Long-term recent use of oral contraceptives increases risk of breast cancer . The increase in prolactin levels observed with their use  could conceivably play a role in this effect. Of the other medications known to influence prolactin levels, reserpine, an antihypertensive agent that is no longer commonly used, is one of the best studied. Reserpine initially causes an acute elevation of prolactin; however, long-term use results in about a 50% elevation in plasma levels . Although a positive association between reserpine use and breast cancer was noted in several studies [126, 127, 128], no association was observed in a number of subsequent evaluations [129, 130, 131, 132, 133, 134]. Possible reasons for this inconsistency include the small size of many of the studies and the exposure definition used (e.g., most investigators reported on “ever use” only). If prolactin is a promoter of breast cancer, only longer durations of reserpine use would be expected to have a discernible influence on risk . Cimetidine also has been reported to increase prolactin levels but the few studies published have not shown any meaningful link with breast cancer [136, 137, 138]. A number of recent studies have evaluated antidepressant use and breast cancer, with a focus on selective serotonin uptake inhibitors (SSRIs), a common class of antidepressants for which multiple small studies support a moderate (about 30–40%) increase in prolactin levels in some users [139, 140, 141, 142, 143]. However this does not appear to translate into an increased breast cancer risk [144, 145, 146, 147, 148, 149, 150, 151], except possibly in some subgroups (e.g. for PR- tumors ). Of note, most of the SSRI-breast cancer studies were limited in their ability to account for known breast cancer risk factors possibly biasing the observed association [148, 151], and generally had populations with a low prevalence and short durations of use [144, 145, 146, 149]. Thus current evaluations of medications known to influence prolactin levels do not indicate any important association with risk of breast cancer; however, further assessments which include a detailed evaluation of duration of use are warranted.
Prolactinomas and breast cancer risk. Women with prolactinomas have greatly elevated prolactin levels, thus rates of breast cancer in this group are of considerable interest. However, just a few case reports of breast cancer in women or men with prolactinomas [152, 153, 154, 155, 156] and a small cohort study of 67 women with prolactinomas  have been published to date, therefore additional data are needed. A limitation in using these data to infer the relationship between prolactin levels in the normal or modestly elevated range and breast cancer risk is the frequent occurrence of hypogonadism in women with prolactinomas . Lower exposure to estrogens and androgens premenopausally is hypothesized to decrease breast cancer risk hence potentially counterbalancing, at least in part, any increase in risk associated with elevated prolactin levels.
In summary, accumulating epidemiologic data indicate that prolactin is involved in the etiology of at least some breast tumors in premenopausal and postmenopausal women. Further follow-up in existing prospective cohort studies is needed to confirm and better define this relationship, especially for various tumor subtypes. Further, characterizing the relations between prolactin and various outcomes in breast cancer patients is an area requiring additional research; focus on effects of pre- vs post-surgical levels, as well as stratification by important tumor characteristics is essential. Also, better understanding how lifestyle factors are correlated with prolactin levels can provide insights in how women might better prevent breast cancer.
Little is known about whether genetic polymorphisms in either the prolactin or prolactin receptor genes affect levels, receptor expression, or risk of breast cancer. Further, no studies have evaluated whether high prolactin levels may only increase risk of prolactin receptor positive breast tumors. In addition, a circulating prolactin binding protein has been identified in human serum . Measurement of this binding protein, with prolactin, may provide a better measure of biologically active prolactin. The ultimate goals of future research should be to identify potential interventions to reduce risk of breast cancer through this pathway, refine risk prediction models, and to characterize targets for treatment of breast cancer.
Support for this project was from NIH grants P01 CA87969, CA49449, and CA119139.
- 2.Ganong WF. Review of medical physiology. 21st ed. New York: Lange Medical Books/McGraw-Hill Medical Publishing Division; 2001.Google Scholar
- 7.Gutzman JH, Miller KK, Schuler LA. Endogenous human prolactin and not exogenous human prolactin induces estrogen receptor alpha and prolactin receptor expression and increases estrogen responsiveness in breast cancer cells. J Ster Biochem Molec Biol 2004;88:69–77.Google Scholar
- 11.Struman I, Bentzien F, Lee H, Mainfroid V, D'Angelo G, Goffin V, et al. Opposing actions of intact and N-terminal fragments of the human prolactin/growth hormone family members on angiogenesis: an efficient mechanism for the regulation of angiogenesis. Proc Natl Acad Sci U S A 1999;96(4):1246–51.PubMedGoogle Scholar
- 61.Lee SA, Haiman CA, Burtt NP, Pooler LC, Cheng I, Kolonel LN, et al. A comprehensive analysis of common genetic variation in prolactin (PRL) and PRL receptor (PRLR) genes in relation to plasma prolactin levels and breast cancer risk: the Multiethnic Cohort. BMC Med Genet 2007;8(1):72.PubMedGoogle Scholar
- 63.De Placido S, Gallo C, Perrone F, Marinelli A, Pagliarulo C, Carlomagno C, et al. Prolactin receptor does not correlate with oestrogen and progesterone receptors in primary breast cancer and lacks prognostic significance. Ten year results of the Naples adjuvant (GUN) study. Br J Cancer 1990;62(4):643–6.PubMedGoogle Scholar
- 78.Patel DD, Bhatavdekar JM, Chikhlikar PR, Ghosh N, Suthar TP, Shah NG, et al. Node negative breast carcinoma: hyperprolactinemia and/or overexpression of p53 as an independent predictor of poor prognosis compared to newer and established prognosticators. J Surg Oncol 1996;62(2):86–92.PubMedGoogle Scholar
- 86.Hankinson SE, Colditz GA, Hunter DJ, Manson JE, Willett WC, Stampfer MJ, et al. Reproductive factors and family history of breast cancer in relation to plasma estrogen and prolactin levels in postmenopausal women in the Nurses’ Health Study (United States). Cancer Causes Control 1995;6(3):217–24.PubMedGoogle Scholar
- 97.Johansson H, Gandini S, Bonanni B, Mariette F, Guerrieri-Gonzaga A, Serrano D, et al. Relationships between circulating hormone levels, mammographic percent density and breast cancer risk factors in postmenopausal women. Breast Cancer Res Treat 2008 (in press).Google Scholar
- 110.McTiernan A, Wu L, Chen C, Chlebowski R, Mossavar-Rahmani Y, Modugno F, et al. Relation of BMI and physical activity to sex hormones in postmenopausal women. Obesity (Silver Spring) 2006;14(9):1662–77.Google Scholar
- 123.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):1713–27.Google Scholar
- 126.Boston Collaborative Drug Surveillance Program. Reserpine and breast cancer.1974;2(7882):669–71.Google Scholar
- 135.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):1047–59.Google Scholar
- 157.Popovic V, Damjanovic S, Micic D, Nesovic M, Djurovic M, Petakov M, et al. Increased incidence of neoplasia in patients with pituitary adenomas. The Pituitary Study Group. Clin Endocrinol (Oxf) 1998;49(4):441–5.Google Scholar