SEARCH

SEARCH BY CITATION

Keywords:

  • insulin-like growth factor;
  • insulin-like growth factor binding protein-3;
  • breast cancer;
  • fibrocystic breast conditions;
  • Chinese

Abstract

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. References

We investigated whether circulating insulin-like growth factor-I (IGF-I) and insulin-like growth factor binding protein-3 (IGFBP-3) levels are associated with the risk of fibrocystic breast conditions (FBC), in a case-control study nested within a randomized trial of breast self-examination conducted in Shanghai, China. Participants were enrolled during 1989–1991 and were followed over 10 years for the development of breast diseases. Controls (n = 897) were frequency-matched by age to cases (n = 451), who were diagnosed with FBC between 1995 and 2000. Circulating IGF-I and IGFBP-3 levels and their molar ratio were positively associated with risk of FBC. The odds ratios (ORs) and 95% confidence intervals (CI) for the upper fourth of the distribution compared to the lowest fourth for IGF-I, IGFBP3 and their molar ratio were 3.02 (2.02–4.52), 1.92 (1.37–2.71) and 2.26 (1.52–3.36), respectively. The strength of the association between IGF-I levels and FBC was attenuated after adjustment for IGFBP-3 and that for IGFBP-3 was largely eliminated after adjustment for IGF-I. Increasing levels of IGF-I were particularly associated with increasing risk of FBC with proliferative elements (ORs and 95% CIs for the 2nd, 3rd and upper fourth of the distribution of IGF-I: 3.13 (1.50–6.53), 4.57 (2.22–9.39) and 6.30 (3.08–12.89), compared with the lowest fourth. Our results suggest that elevated levels of IGF-I may contribute to the development of FBC. © 2005 Wiley-Liss, Inc.

Insulin-like growth factors (IGFs) are potent mitogenic and anti-apoptotic factors that regulate cell proliferation, differentiation and apoptosis.1, 2, 3 The effect of IGFs is modulated by at least 6 high affinity IGF-binding proteins (IGFBPs);4, 5 greater than 95% of circulating IGF-I is complexed with IGFBP-3 and an acid-labile subunit.6 The majority of IGF-I and IGFBPs are produced by the liver, the rest by a wide variety of tissues.2

There are several reasons to suspect that the circulating levels of IGF-I and IGFBP-3 could influence breast cancer risk. In in vitro studies, IGF-I modulates gene expression and growth of the breast cancer cell line MCF-7.7, 8 Expression of IGF-I receptors affect responsiveness of MCF-7 cells to IGF-I and estradiol.9 Evidence from breast cancer cell lines, experimental animal models and epidemiologic studies suggests that there is a synergistic effect between IGF-I and estrogens and androgens on cell proliferation and breast cancer risk.10, 11, 12, 13, 14 Breast epithelial cells and stromal tissues have been shown to contain IGF-I and IGF-II proteins,15, 16, 17, 18 IGFBPs,19 as well as IGFs mRNA20, 21 and IGF-I receptor.22 In addition to modulation of IGF availability and action, IGFBPs have been shown to inhibit cellular growth23, 24, 25, 26 and IGF-I receptor activation,27 and enhance apoptosis.28, 29 In a mouse mammary model, a reduced circulating IGF-I level delayed the onset of chemically and genetically induced mammary tumors.30

There have been a number of epidemiologic studies, both prospective and retrospective, investigating the association of circulating levels of IGFs and IGFBPs with breast cancer risk. Results of 2 meta-analyses showed that increased circulating IGF-I levels are associated with a 40–70% increased risk of breast cancer among premenopausal but not postmenopausal women.31, 32 Included in these meta-analyses was a study of 300 Chinese women with and 300 without breast cancer. That study observed a 2- and 3-fold increased risk of breast cancer among women in the upper third of the respective IGF-I and IGFBP-3 distributions, compared with women in the lower third.33 This association was slightly more evident among premenopausal women.

Benign epithelial breast disorders include a wide variety of histologic conditions. Little is known about their etiology. Although non-proliferative benign breast conditions are not known to be associated with an increased risk of breast cancer, proliferative benign fibrocystic conditions without or with atypia have been shown to be associated with a 2- and 4-fold increased risk of breast cancer, respectively.34, 35, 36, 37, 38, 39, 40 Few studies have investigated the association of IGFs and IGFBPs with the risk of benign epithelial breast disorders. A study on Viennese women observed an elevation of IGF-I levels in women with benign breast lesions relative to controls without breast disease.41 Another study showed higher circulating IGFBP-3 levels (3.6 ± 0.7 mg/L) in women with benign breast disease than in age-matched control women (2.7 ± 0.6 mg/L) or women with breast cancer (2.7 ± 0.5 mg/L).42

We report here results from a case-control study that was nested in a large cohort of Chinese women43 enrolled in a trial of breast self examination. We examined whether levels of IGF-I and IGFBP-3 are associated with risk of fibrocystic breast conditions.

Material and methods

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. References

Study Population

Participants in this case-control study have been described in more detail previously.44 Briefly, women who participated in this study were selected from participants in a randomized trial of breast self-examination (BSE) in Shanghai, China.43, 45 Current and retired female employees of the Shanghai Textile Industry Bureau, born between 1925 and 1958, were eligible for the trial and were enrolled between October 1989 and October 1991. For the 266,064 participants, a trained medical worker administered a baseline questionnaire at recruitment. All of the trial participants were followed for vital status, continued employment in the textile industry, and residence in Shanghai, as well as for the development of malignant and benign breast conditions by active follow up of the cohort up to July 31, 2000. Case finding for breast cancer cases was supplemented by computerized matching of the cohort with the Shanghai Cancer Registry.43 Diagnoses of all breast conditions were made by hospital workers in 3 major hospitals serving the textile industry, as well as several other hospitals with contractual agreements with individual factories. Study personnel reviewed the pathology reports and other medical records of all women found to have a histologically confirmed benign breast lesion, recorded tumor size and histologic classification on a standardized form, and obtained slides of all lesions for shipment to Seattle, WA.

Participants in this study are from 2 sequential case-control studies of benign and malignant breast conditions, one conducted between September 1995 and August 1997, and the other between September 1997 and July 2000.44 Women enrolled in these 2 studies, who received a breast biopsy and, as a result, were diagnosed with a fibrocystic breast condition by a local pathologist, were eligible for the present study. For women who had adequate tissue for pathology review (at least 5 scanning power fields), diagnoses were confirmed by a single reference pathologist who had no knowledge of their original diagnosis. The reference pathologist recorded the number of scanning power fields examined and classified the women's conditions into 3 different types of fibrocystic conditions, according to the scheme developed by Stalsberg.46 Cases with atypia included those with atypical ductal hyperplasia, atypical lobular hyperplasia or moderate apocrine atypia. Cases with proliferative changes were those with moderate or florid ductal hyperplasia or moderate or predominant sclerosing adenosis and no atypia. Subjects with mild or no ductal hyperplasia and mild or no sclerosing adenosis were classified as having nonproliferative fibrocystic breast conditions. This classification is similar to Page's classification for benign breast disease.34 Of the 622 eligible women with fibrocystic conditions, 551 (89%) were interviewed (262 were diagnosed and interviewed between September 1995 through August 1997 and 289 between September 1997 and July 2000). Of the 551 interviewed women, 389 (71%) had enough tissue for further histologic classification by the reference pathologist. We obtained either a serum or plasma sample at the time of diagnosis of benign breast disease and measured levels of IGF-I and IGFBP-3 for 453 (82%) interviewed women with fibrocystic conditions, 327 (72%) of whom also had enough tissue for further histologic classification by the reference pathologist.

As described previously,44 controls were selected from all women in the BSE trial who did not have a breast biopsy and were frequency-matched to cases on age. Briefly, for each benign and malignant case enrolled in the 1995–1997 study, 20 potential controls with the same year of birth were randomly selected and listed on a form. Women from this list were contacted in the order listed, until 2 women were recruited (n = 367; 64% of eligible). For the benign and malignant cases diagnosed between September 1997 and July 2000, randomly selected control women were frequency matched to them by 5-year age group and hospital affiliation of their factories at baseline. In-person interviews were completed for 704 (82%) of 862 controls selected in this manner. Of the 1,071 interviewed controls who were eligible for the present study, 906 (85%) had either a serum or plasma sample at the time of interview.

Cases were interviewed and had their blood drawn before the breast biopsy that resulted in their diagnosis, and controls were interviewed and had blood drawn during the same general time period as the cases. Cases and controls were interviewed by the same team of interviewers. The questionnaire collected information on demographic characteristics, reproductive and gynecologic history, smoking and alcohol habits, medical history, family history of breast cancer and occupational and recreational physical activity. Informed consent was obtained from each woman before interview. The study was approved by the Institutional Review Boards of the Fred Hutchinson Cancer Research Center and the Station for Prevention and Treatment of Cancer of the Shanghai Textile Industry Bureau, in accordance with assurance filed with the Office for Human Research Protections of the US Department of Health and Human Services.

Because use of hormones may affect IGF levels, current users of postmenopausal hormone therapy (6 controls) or oral contraceptives (2 cases and 3 controls) were excluded, leaving a total of 451 cases and 897 controls included in analyses.

Laboratory assays

IGF-I and IGFBP-3 concentrations were determined for each case and control, in batches balanced by case-control status, using 2-site immunoradiometric assay (IRMA) kits from Diagnostic Systems Laboratories, Inc. (Webster, TX), by a technologist blinded to case-control status. Assay performance for each run, as assessed by percent coefficient of variation (CV), was monitored by analysis of 2 kit controls and in-house pooled controls that are of various concentrations to cover the entire assay range. Samples were reanalyzed if the duplicate measures yielded a %CV greater than 10%. For IGF-I, the intra-assay CVs were 3.0, 3.5 and 3.8% at 59, 214 and 275 ng/mL, respectively. The inter-assay CVs were 9.8, 10.3, 6.7, 6.2 and 10.6% at 64, 199, 52, 191 and 205 ng/mL, respectively. For IGFBP-3, the intra-assay CVs were 2.1, 2.9 and 5.8% at 7.8, 16.4 and 4347 ng/mL, respectively. The inter-assay CVs were 4.6, 2.4, 4.9, 3.8 and 7.3% at 5, 51, 4, 42 and 3962 ng/mL, respectively. For a few participants without serum aliquots, measurements were conducted using plasma. Plasma levels were multiplied by a factor of 1.08, according to the assay kit package insert, to be comparable to the serum results.

Statistical analyses

Because some of the controls for this study were also selected for studies of breast cancer, the aggregated controls used in this study tend to be older than the cases. The frequencies of reproductive and demographic characteristics were adjusted for the age distribution of the controls, using indirect adjustment methods.47 Odds ratios and 95% confidence intervals were generated using unconditional logistic regression models. Because of the known inverse association between IGF-I and IGFBP-3 levels and age,4 we adjusted for age by including it as a continuous variable in all models. We divided IGF-I, IGFBP-3 and their molar ratio into fourths based on their distributions among the controls. We examined whether there was an association between these variables and fibrocystic disease overall, and also in subgroups defined by menopausal status and proliferative status of the fibrocystic conditions. Because of small numbers of postmenopausal women in the proliferative and nonproliferative subgroups, we dichotomized the levels of IGF-I and IGFBP-3 and their molar ratios for these analyses.

We examined confounding influences of IGFBP-3 on IGF-I, and IGF-I on IGFBP-3; both were included in the final models. Confounding by other characteristics including education, oral contraceptive use, first degree family history of breast cancer, body mass index, physical activity, personal cigarette smoking history, exposure to spouse's cigarette smoke, age at first menstrual period, parity, duration of lactation, age at first live birth and menopausal status was assessed by examining whether their inclusion in these models altered the odds ratios for the main independent variables by 10% or more. We performed tests for trend by entering the categorical variables in regression models as ordinal variables.

Because of the different age distributions between cases and controls, we repeated our analyses restricted to ages 40–59, the group for which we had more comparable proportions of cases and controls. We also assessed whether inclusion of women who did not have enough tissue available for further histologic classification of the fibrocystic condition by the reference pathologist affected the results, by repeating our analyses restricted to women for whom a further histologic classification was assigned.

Results

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. References

As shown in Table I, women with fibrocystic conditions were younger than the control women, but the 2 groups of the same age were similar with respect to number of live births, age at menopause, smoking history and years lived with smoking spouse. There was a greater percentage of cases than controls with prior breast lumps or with a first degree relative with breast cancer. As expected, IGF-I and IGFBP-3 levels decreased with increasing age among the controls (r = −0.44, and r = −0.11, respectively; p < 0.001).

Table I. Selected Characteristics of Controls and Cases1
Characteristic/FactorControls (n = 897)Fibrocystic conditions (n = 451)
nn
  • 1

    For all characteristics except age, the indirect age-adjusted percentages based on the age distribution of controls are presented; columns may not add to total because of missing data.

  • 2

    Values in parentheses indicate percentages.

Age (y)
 <4013 (1.5)261 (13.5)
 40–44386 (43.0)193 (42.8)
 45–49174 (19.4)119 (26.4)
 50–5461 (6.8)21 (4.7)
 55–5940 (4.5)12 (2.7)
 60–64109 (12.2)13 (2.9)
 ≥65114 (12.7)32 (7.1)
Nulliparous36 (4.0)22 (4.9)
Number of live births
 1572 (64.1)350 (63.5)
 2111 (12.4)36 (11.2)
 ≥3174 (19.5)38 (20.3)
Age at first live birth (y)
 ≤24247 (27.7)77 (24.7)
 25–29484 (54.2)274 (56.5)
 ≥30126 (14.1)72 (13.9)
Age at menopause (y)
 Pre-menopausal544 (60.6)364 (65.1)
 <4559 (6.6)18 (4.9)
 45–49130 (14.5)34 (12.6)
 ≥50164 (18.3)35 (17.4)
Ever smoked
 No876 (97.7)444 (98.7)
 Yes21 (2.3)5 (1.3)
Years lived with smoking spouse
 None328 (36.8)151 (37.5)
 1–15250 (28.0)163 (28.3)
 16-20166 (18.6)85 (18.3)
 ≥21148 (16.6)43 (15.9)
Prior breast lumps
 Never846 (96.9)386 (90.1)
 Ever27 (3.1)45 (9.9)
First degree relative with breast cancer
 No882 (98.3)434 (96.6)
 Yes15 (1.7)17 (3.4)

Circulating IGF-I and IGFBP-3 levels and their molar ratio were positively associated with the risk of fibrocystic conditions (Table II). The strength of the association between IGF-I levels and fibrocystic conditions was slightly attenuated after adjustment for IGFBP-3. Further controlling for reproductive variables did not lead to substantial changes in the ORs. The association between IGFBP-3 and fibrocystic conditions was largely eliminated after adjustment for IGF-I.

Table II. Risk of Fibrocystic Conditions by Quartiles of IGF-I, IGFBP-3 and Their Molar Ratio
QuartileControls (n = 897)Fibrocystic conditions (n = 451)OR195% CIOR295% CIOR395% CI
nn
  • 1

    ORs adjusted for age.

  • 2

    OR adjusted for age and mutually adjusted for IGFBP-3 or IGF-I, respectively.

  • 3

    ORs adjusted for age, IGFBP-3 or IGF-I, respectively, number of live births, and age at first live birth (6 cases and 4 controls are missing data).

  • 4

    Values in parentheses in this column indicate percentages.

IGF-I (ng/mL)
 ≤12422444 (10)41.00 1.00 1.00 
 124.1–182.722488 (20)1.59(1.04, 2.41)1.51(0.98, 2.33)1.57(1.02, 2.43)
 182.8–247.4224123 (27)2.02(1.34, 3.05)1.87(1.20, 2.90)1.91(1.22, 2.98)
 ≥247.5225196 (43)3.02(2.02, 4.52)2.66(1.68, 4.22)2.63(1.65, 4.20)
 ptrend < 0.0001ptrend < 0.0001ptrend < 0.0001
IGFBP-3 (ng/mL)
 ≤319922474 (16)1.00 1.00 1.00 
 3200–366622597 (22)1.21(0.84, 1.74)1.03(0.71, 1.50)1.04(0.71, 1.51)
 3667–4165224128 (28)1.51(1.07, 2.14)1.10(0.75, 1.61)1.13(0.77, 1.66)
 ≥4166224152 (34)1.92(1.37, 2.71)1.24(0.84, 1.84)1.26(0.85, 1.87)
 ptrend < 0.0001ptrend = 0.24ptrend = 0.21
IGF-I/IGFBP-3
 ≤0.13222352 (12)1.00   1.00 
 0.133–0.18422579 (18)1.19(0.79, 1.79)  1.19(0.79, 1.80)
 0.185–0.238224136 (30)1.82(1.22, 2.70)  1.81(1.21, 2.70)
 ≥0.239225184 (41)2.26(1.52, 3.36)  2.22(1.49, 3.32)
 ptrend < 0.0001 ptrend < 0.0001

Increasing concentrations of circulating levels of IGF-I were particularly strongly associated with increasing risk of proliferative fibrocystic conditions (Table III). This association with IGF-I remained strong after controlling for levels of IGFBP-3 and reproductive variables. A weaker association between proliferative fibrocystic conditions and IGFBP-3 was not statistically significant after controlling for level of IGF-I and reproductive variables. The association between proliferative fibrocystic conditions and the molar ratio is, thus, largely due to the association with IGF-I. The associations with all of these variables were weaker for nonproliferative than for proliferative conditions.

Table III. Risk of Nonproliferative and Proliferative Fibrocystic Conditions by Quartiles of IGF-I, IGFBP-3 and their Molar Ratio
QuartileControls (n = 897)Nonproliferative fibrocystic conditions (n = 147)Proliferative fibrocystic conditions (n = 183)
nnOR195% CIOR295% CInOR195% CIOR295% CI
  • 1

    ORs adjusted for age.

  • 2

    ORs adjusted for age, number of live births, and age at first live birth (3 cases and 4 controls are missing data); ORs for IGF-I are additionally adjusted for IGFBP-3, and IGFBP-3 is additionally adjusted for IGF-I.

  • 3

    Values in parentheses in this column indicate percentages.

IGF-I (ng/mL)
 ≤12422420 (14)31.00 1.00 10 (5)31.00 1.00 
 124.1–182.722421 (14)0.79(0.41, 1.52)0.80(0.41, 1.58)36 (20)3.13(1.50, 6.53)3.10(1.46, 6.61)
 182.8–247.422438 (26)1.25(0.68, 2.31)1.29(0.67, 2.48)56 (31)4.57(2.22, 9.39)4.30(2.01, 9.18)
 ≥247.522568 (46)2.08(1.16, 3.72)2.02(1.03, 3.96)81 (44)6.30(3.08, 12.89)5.61(2.56, 12.31)
 ptrend = 0.0006ptrend = 0.005ptrend < 0.0001ptrend < 0.0001
IGFBP-3 (ng/mL)s
 ≤319922427 (18)1.00 1.00 27 (15)1.00 1.00 
 3200–366622533 (22)1.03(0.65, 1.95)1.02(0.58, 1.80)41 (22)1.41(0.83, 2.38)1.03(0.59, 1.78)
 3667–416522441 (28)1.31(0.77, 2.22)1.00(0.56, 1.78)52 (28)1.72(1.04, 2.85)1.08(0.63, 1.87)
 ≥416622446 (31)1.55(0.93, 2.61)1.02(0.56, 1.86)63 (34)2.23(1.36, 3.64)1.20(0.69, 2.09)
 ptrend = 0.07ptrend = 0.98ptrend = 0.0008ptrend = 0.48
IGF-I/IGFBP-3
 ≤0.13222320 (14)1.00 1.00 11 (6)1.00 1.00 
 0.133–0.18422520 (14)0.75(0.38, 1.46)0.76(0.39, 1.49)34 (19)2.72(1.33, 5.57)2.79(1.35, 5.73)
 0.185–0.23822443 (29)1.38(0.75, 2.53)1.42(0.77, 2.63)60 (33)4.46(2.21, 8.98)4.50(2.22, 9.12)
 ≥0.23922564 (44)1.87(1.03, 3.40)1.84(1.00, 3.36)78 (43)5.52(2.73, 11.16)5.43(2.67, 11.03)
 ptrend = 0.51ptrend < 0.004ptrend < 0.0001ptrend < 0.0001

The association between IGF-I and fibrocystic breast conditions was stronger among postmenopausal women than among premenopausal women (Table IV), and in these women the association was present only for the proliferative conditions (Table V). However, the confidence intervals were wide and overlapping, therefore, these results could potentially be due to chance. While there was no association between increasing levels of IGFBP-3 and fibrocystic conditions among postmenopausal women, in premenopausal women, there was a suggestion of an increasing trend in risk with IGFBP-3 level (Table IV). This suggestive association was of similar strength for proliferative and nonproliferative conditions (Table V).

Table IV. Associations Between Quartiles of IGF-I, IGFBP-3, and their Molar Ratio and Risk of Fibrocystic Conditions Among Pre- and Postmenopausal Women
Quartile1PremenopausalPostmenopausal
Controls (n = 542)Fibrocystic conditions (n = 360)Controls (n = 351)Fibrocystic conditions (n = 85)
 nnOR295% CInnOR295% CI
  • 1

    Quartile cutpoints are based on the overall control group.

  • 2

    ORs adjusted for age, number of live births, and age at first live birth (for premenopausal women, 4 cases and 2 controls are missing data, and for postmenopausal women, 2 cases and 2 controls are missing data); ORs for IGF-I are additionally adjusted for IGFBP-3, and IGFBP-3 is additionally adjusted for IGF-I.

  • 3

    Values in parentheses in this column indicate percentages.

IGF-I (ng/mL)
 ≤12470 (13)322 (6)31.00 153 (44)321 (25)31.00 
 124.1–182.7130 (24)69 (19)1.46(0.82, 2.60)92 (26)19 (22)1.56(0.76, 3.20)
 182.8–247.4159 (29)102 (28)1.66(0.93, 2.94)64 (18)20 (24)2.46(1.11, 5.43)
 ≥247.5183 (34)167 (46)2.07(1.15, 3.73)42 (12)25 (29)4.83(2.03, 11.51)
 ptrend = 0.01  ptrend = 0.0003
IGFBP-3 (ng/mL)
 ≤3199122 (23)55 (15)1.00 100 (28)18 (21)1.00 
 3200–3666138 (25)79 (22)1.11(0.72, 1.72)86 (25)16 (19)0.80(0.37, 1.73)
 3667–4165155 (29)108 (30)1.21(0.78, 1.89)68 (19)20 (24)1.04(0.48, 2.29)
 ≥4166127 (23)118 (33)1.55(0.97, 2.46)97 (28)31 (36)0.76(0.34, 1.71)
 ptrend = 0.05  ptrend = 0.63
IGF-I/IGFBP-3
 ≤0.13265 (12)25 (7)1.00 158 (45)27 (32)1.00 
 0.133–0.184124 (23)58 (16)1.13(0.65, 1.99)99 (28)20 (24)1.18(0.62, 2.23)
 0.185–0.238162 (30)116 (32)1.69(1.00, 2.86)60 (17)18 (21)1.73(0.87, 3.42)
 ≥0.239191 (35)161 (45)1.94(1.16, 3.26)34 (10)20 (24)3.26(1.53, 6.92)
 ptrend = 0.0009  ptrend = 0.003
Table V. Associations Between Quartiles of IGF-I, IGFBP-3, and their Molar Ratio and Risk of Fibrocystic Conditions Among Pre- and Postmenopausal Women, by Proliferative Status
PremenopausalControls (n = 542)1Nonproliferative (n = 121)Proliferative (n = 142)
nnOR295% CInOR295% CI
IGF-I (ng/mL)
 ≤157.213523 (19)31.00 22 (15)31.00 
 157.3–211.213520 (17)0.71(0.36, 1.41)26 (18)0.98(0.51, 1.87)
 211.3–269.613638 (31)1.34(0.71, 2.52)45 (32)1.65(0.89, 3.05)
 ≥269.713640 (33)1.27(0.64, 2.51)49 (35)1.65(0.86, 3.17)
   ptrend = 0.19 ptrend = 0.05
IGFBP-3 (ng/mL)
 ≤324813521 (17)1.00 23 (16)1.00 
 3249–369213629 (24)1.39(0.73, 2.64)33 (23)1.28(0.70, 2.36)
 3693–412813532 (26)1.33(0.68, 2.60)39 (27)1.37(0.73, 2.56)
 ≥412913639 (32)1.53(0.78, 3.03)47 (33)1.54(0.81, 2.93)
   ptrend = 0.28 ptrend = 0.20
IGF-I/IGFBP-3
 ≤0.16313519 (16)1.00 20 (14)1.00 
 0.164–0.20813326 (21)1.30(0.68, 2.48)30 (21)1.45(0.78, 2.69)
 0.209–0.26413630 (25)1.42(0.75, 2.67)45 (32)2.09(1.16, 3.76)
 ≥0.26513846 (38)2.14(1.18, 3.90)47 (33)2.14(1.19, 3.83)
   ptrend = 0.01 ptrend = 0.005
PostmenopausalControls (n = 351)4Nonproliferative (n = 24)Proliferative (n = 40)
nnOR295% CInOR295% CI
  • 1

    Quartile cutpoints are based on the distributions of IGF-I, IGFBP-3, and their molar ratio, respectively, among premenopausal controls only.

  • 2

    ORs adjusted for age, number of live births, and age at first live birth (for premenopausal women, 4 cases and 2 controls are missing data, and for postmenopausal women, 2 cases and 2 controls are missing data); ORs for IGF-I are additionally adjusted for IGFBP-3, and IGFBP-3 is additionally adjusted for IGF-I.

  • 3

    Values in parentheses in this column indicate percentages.

  • 4

    Cutpoints are based on the median values of IGF-I, IGFBP-3, and their molar ratio, respectively, among postmenopausal controls only.

IGF-I (ng/mL)
 ≤139.117510 (42)1.00 6 (15)1.00 
 >139.117614 (58)1.11(0.42, 2.90)34 (85)4.70(1.78, 12.45)
IGFBP-3 (ng/mL)
 ≤35961759 (38)1.00 12 (30)1.00 
 >359617615 (63)1.57(0.61, 4.05)28 (70)1.29(0.60, 2.80)
IGF-I/IGFBP-3
 ≤0.14117612 (50)1.00 7 (18)1.00 
 >0.14117512 (50)0.93(0.39, 2.22)33 (83)4.47(1.88, 10.64)

Discussion

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. References

Our results showed a strong positive association between fibrocystic conditions (particularly proliferative fibrocystic lesions) and IGF-I levels and the IGF-I/IGFBP-3 molar ratio. IGFBP-3 was not as strongly associated with risk of fibrocystic conditions. It is possible that high circulating levels of IGF-I promote proliferation of the breast tissue through increased IGF-I binding to its IGF-I receptors. These results are consistent with previously reported positive associations of elevated IGF-I levels and of IGF-I to IGFBP-3 ratios with benign breast disease,48 and also with the observation of elevated IGF-I concentrations in Type I breast cysts relative to normal breast tissue.49 Type I cysts originate from the terminal duct lobular units, the site where most proliferative breast lesions occur.

In our study, as is true for any case-control study, proper selection of controls is of paramount importance. Our study is comprised of 2 case-control studies, with participants derived from the well-defined BSE cohort. The controls were randomly selected from the cohort members who were free of a diagnosis of fibrocystic conditions or breast cancer. Based on the baseline questionnaire,50 the characteristics (e.g. number of live births, age at first birth and age at menarche) of the controls selected for the case-control studies are representative of the cohort overall.

There are several limitations to this study. First, we cannot rule out the possibility that a small percentage of the control women had undiagnosed fibrocystic breast conditions. However, only 1.4% of the entire cohort of 266,040 women in the BSE trial developed benign breast lumps. Even if some proportion of the controls had undiagnosed fibrocystic conditions, this misclassification would only bias our results towards the null. Second, our results are based on a single sample collected at the time of diagnosis of benign breast disease. Thus, it is unclear whether the elevation of IGF-I levels preceded or occurred after the development of the lesion. Third, the blood samples we used for the study were not collected at a fixed time of the menstrual cycle. However, this should have little or no impact on the results of the study since IGF-I levels differ little, if at all, during the menstrual cycle.51, 52, 53, 54, 55, 56 Even if IGF levels varied by time of the menstrual cycle, since there is no reason to believe that cases and controls would systematically differ on this characteristic, such non-differential misclassification would only cause the observed associations to be attenuated. Fourth, the age distribution of cases and controls differed, which is a concern because of the known inverse association between age and IGF-I levels. We addressed this limitation by controlling for age as a continuous variable. We also repeated our analyses restricted to ages 40–59, to evaluate whether the results were similar when restricted to a group that had more similar proportions of cases and controls. Indeed, the results were comparable. Fifth, even with a relatively large sample size overall, there were relatively few postmenopausal women, limiting the power of the study in this group of women. Lastly, only a portion of the women eligible for this study agreed to be interviewed, gave blood, and had lesions suitable for histologic classification. However, those excluded for these reasons, in the aggregate, did not differ from those included in this study with respect to risk factors for breast cancer as ascertained on the baseline questionnaire for the BSE trial.50 Furthermore, in order to assess whether exclusion of the 124 women for whom the reference pathologist could not assign a further histologic classification, we repeated the analyses shown in table IV, excluding these 124 women. The results were essentially the same as those presented, providing reassurance that our results are not affected by possible source of bias.

IGFBP-3 is the major binding protein for IGF-I in the circulation and as such, it modulates the level of IGF-I that is available to the tissue. One would posit that a high level of IGFBP-3 would be inversely associated with proliferative disease. However, we did not find that to be the case among premenopausal women. Our results showed increasing IGFBP-3 levels to be associated with a modest increasing risk of both proliferative and non-proliferative fibrocystic conditions in premenopausal women after adjustment for IGF-I levels, suggesting that IGFBP-3 could have an effect on the risk of fibrocystic conditions among premenopausal women other than modulating IGF-I actions. This is not inconceivable since a number of in vitro and in vivo studies have demonstrated that IGFBP-3 exerts IGF-I independent effect on the growth and apoptosis of a wide range of cells, including breast cancer cells, by mediating the actions of a diverse range of growth effectors, such as TGF-β, retinoic acid, 1,25-dihydroxy vitamin D3,5 and since elevated IGFBP-3 levels were associated with increased breast cancer risk in some, but not all, epidemiologic studies.31, 57

In conclusion, our results suggest that women who had high levels of circulating IGF-I may be at an increased risk of fibrocystic breast conditions, and particularly of such conditions with proliferative epithelial elements. If confirmed by others, further inquiry into the mechanism by which the IGF-I signaling pathway influences fibrocystic breast conditions is warranted.

References

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. References
  • 1
    Pollak M. Insulin-like growth factor physiology and neoplasia. Growth Horm IGF Res 2000; 10: S6S7.
  • 2
    Pollak M. Insulin-like growth factor physiology and cancer risk. Eur J Cancer 2000; 36: 12248.
  • 3
    Parrizas M, LeRoith D. Insulin-like growth factor-1 inhibition of apoptosis is associated with increased expression of the bcl-xL gene product. Endocrinology 1997; 138: 13558.
  • 4
    Rajaram S, Baylink DJ, Mohan S. Insulin-like growth factor-binding proteins in serum and other biological fluids: regulation and functions. Endocr Rev 1997; 18: 80131.
  • 5
    Firth SM, Baxter RC. Cellular actions of the insulin-like growth factor binding proteins. Endocr Rev 2002; 23: 82454.
  • 6
    Baxter RC. Characterization of the acid-labile subunit of the growth hormone-dependent insulin-like growth factor binding protein complex. J Clin Endocrinol Metab 1988; 67: 26572.
  • 7
    El-Tanani MK, Green CD. Interaction between estradiol and growth factors in the regulation of specific gene expression in MCF-7 human breast cancer cells. J Steroid Biochem Mol Biol 1997; 60: 26976.
  • 8
    Ngo TH, Barnard RJ, Leung PS, Cohen P, Aronson WJ. Insulin-like growth factor I (IGF-I) and IGF binding protein-1 modulate prostate cancer cell growth and apoptosis: possible mediators for the effects of diet and exercise on cancer cell survival. Endocrinology 2003; 144: 231924.
  • 9
    Daws MR, Westley BR, May FE. Paradoxical effects of overexpression of the type I insulin-like growth factor (IGF) receptor on the responsiveness of human breast cancer cells to IGFs and estradiol. Endocrinology 1996; 137: 117786.
  • 10
    Thorsen T, Lahooti H, Rasmussen M, Aakvaag A. Oestradiol treatment increases the sensitivity of MCF-7 cells for the growth stimulatory effect of IGF-I. J Steroid Biochem Mol Biol 1992; 41: 53740.
  • 11
    Ruan W, Catanese V, Wieczorek R, Feldman M, Kleinberg DL. Estradiol enhances the stimulatory effect of insulin-like growth factor-I (IGF-I) on mammary development and growth hormone-induced IGF-I messenger ribonucleic acid. Endocrinology 1995; 136: 1296302.
  • 12
    Yee D, Lee AV. Crosstalk between the insulin-like growth factors and estrogens in breast cancer. J Mammary Gland Biol Neoplasia 2000; 5: 10715.
  • 13
    Nicholson RI, Gee JM. Oestrogen and growth factor cross-talk and endocrine insensitivity and acquired resistance in breast cancer. Br J Cancer 2000; 82: 50113.
  • 14
    Yu H, Shu XO, Li BD, Dai Q, Gao YT, Jin F, Zheng W. Joint Effect of Insulin-like Growth Factors and Sex Steroids on Breast Cancer Risk. Cancer Epidemiol Biomarkers Prev 2003; 12: 106773.
  • 15
    Cullen KJ, Allison A, Martire I, Ellis M, Singer C. Insulin-like growth factor expression in breast cancer epithelium and stroma. Breast Cancer Res Treat 1992; 22: 219.
  • 16
    Cullen KJ, Smith HS, Hill S, Rosen N, Lippman ME. Growth factor messenger RNA expression by human breast fibroblasts from benign and malignant lesions. Cancer Res 1991; 51: 497885.
  • 17
    Rosen N, Yee D, Lippman ME, Paik S, Cullen KJ. Insulin-like growth factors in human breast cancer. Breast Cancer Res Treat 1991; 18: S55S62.
  • 18
    Ellis MJ, Singer C, Hornby A, Rasmussen A, Cullen KJ. Insulin-like growth factor mediated stromal-epithelial interactions in human breast cancer. Breast Cancer Res Treat 1994; 31: 24961.
  • 19
    McGuire WL,Jr, Jackson JG, Figueroa JA, Shimasaki S, Powell DR, Yee D. Regulation of insulin-like growth factor-binding protein (IGFBP) expression by breast cancer cells: use of IGFBP-1 as an inhibitor of insulin-like growth factor action. J Natl Cancer Inst 1992; 84: 133641.
  • 20
    Quinn KA, Treston AM, Unsworth EJ, Miller MJ, Vos M, Grimley C, Battey J, Mulshine JL, Cuttitta F. Insulin-like growth factor expression in human cancer cell lines. J Biol Chem 1996; 271: 1147783.
  • 21
    Gebauer G, Jager W, Lang N. mRNA expression of components of the insulin-like growth factor system in breast cancer cell lines, tissues, and metastatic breast cancer cells. Anticancer Res 1998; 18: 11915.
  • 22
    Ellis MJ, Jenkins S, Hanfelt J, Redington ME, Taylor M, Leek R, Siddle K, Harris A. Insulin-like growth factors in human breast cancer. Breast Cancer Res Treat 1998; 52: 17584.
  • 23
    Figueroa JA, Lee AV, Jackson JG, Yee D. Proliferation of cultured human prostate cancer cells is inhibited by insulin-like growth factor (IGF) binding protein-1: evidence for an IGF-II autocrine growth loop. J Clin Endocrinol Metab 1995; 80: 347682.
  • 24
    Wilson HM, Birnbaum RS, Poot M, Quinn LS, Swisshelm K. Insulin-like growth factor binding protein-related protein 1 inhibits proliferation of MCF-7 breast cancer cells via a senescence-like mechanism. Cell Growth Differ 2002; 13: 20513.
  • 25
    Butt AJ, Fraley KA, Firth SM, Baxter RC. IGF-binding protein-3-induced growth inhibition and apoptosis do not require cell surface binding and nuclear translocation in human breast cancer cells. Endocrinology 2002; 143: 26939.
  • 26
    Lee HY, Chun KH, Liu B, Wiehle SA, Cristiano RJ, Hong WK, Cohen P, Kurie JM. Insulin-like growth factor binding protein-3 inhibits the growth of non-small cell lung cancer. Cancer Res 2002; 62: 35307.
  • 27
    Ricort JM, Binoux M. Insulin-like growth factor-binding protein-3 activates a phosphotyrosine phosphatase. Effects on the insulin-like growth factor signaling pathway. J Biol Chem 2002; 277: 1944854.
  • 28
    Nickerson T, Huynh H, Pollak M. Insulin-like growth factor binding protein-3 induces apoptosis in MCF7 breast cancer cells. Biochem Biophys Res Commun 1997; 237: 6903.
  • 29
    Perks CM, McCaig C, Holly JM. Differential insulin-like growth factor (IGF)-independent interactions of IGF binding protein-3 and IGF binding protein-5 on apoptosis in human breast cancer cells. Involvement of the mitochondria. J Cell Biochem 2000; 80: 24858.
  • 30
    Wu Y, Cui K, Miyoshi K, Hennighausen L, Green JE, Setser J, LeRoith D, Yakar S. Reduced circulating insulin-like growth factor I levels delay the onset of chemically and genetically induced mammary tumors. Cancer Res 2003; 63: 43848.
  • 31
    Shi R, Yu H, McLarty J, Glass J. IGF-I and breast cancer: a meta-analysis. Int J Cancer 2004; 111: 41823.
  • 32
    Sugumar A, Liu YC, Xia Q, Koh YS, Matsuo K. Insulin-like growth factor (IGF)-I and IGF-binding protein 3 and the risk of premenopausal breast cancer: A meta-analysis of literature. Int J Cancer 2004; 111: 2937.
  • 33
    Yu H, Jin F, Shu XO, Li BD, Dai Q, Cheng JR, Berkel HJ, Zheng W. Insulin-like growth factors and breast cancer risk in chinese women. Cancer Epidemiol Biomarkers Prev 2002; 11: 70512.
  • 34
    Dupont WD, Page DL. Risk factors for breast cancer in women with proliferative breast disease. N Engl J Med 1985; 312: 14651.
  • 35
    Dupont WD, Page DL. Breast cancer risk associated with proliferative disease, age at first birth, and a family history of breast cancer. Am J Epidemiol 1987; 125: 76979.
  • 36
    Carter CL, Corle DK, Micozzi MS, Schatzkin A, Taylor PR. A prospective study of the development of breast cancer in 16,692 women with benign breast disease. Am J Epidemiol 1988; 128: 46777.
  • 37
    London SJ, Connolly JL, Schnitt SJ, Colditz GA. A prospective study of benign breast disease and the risk of breast cancer. JAMA 1992; 267: 9414.
  • 38
    Dupont WD, Parl FF, Hartmann WH, Brinton LA, Winfield AC, Worrell JA, Schuyler PA, Plummer WD. Breast cancer risk associated with proliferative breast disease and atypical hyperplasia. Cancer 1993; 71: 125865.
  • 39
    Baer HJ, Schnitt SJ, Connolly JL, Byrne C, Cho E, Willett WC, Colditz GA. Adolescent diet and incidence of proliferative benign breast disease. Cancer Epidemiol Biomarkers Prev 2003; 12: t-67.
  • 40
    Webb PM, Byrne C, Schnitt SJ, Connolly JL, Jacobs TW, Baer HJ, Willett WC, Colditz GA. A prospective study of diet and benign breast disease. Cancer Epidemiol Biomarkers Prev 2004; 13: 110613.
  • 41
    Singer CF, Mogg M, Koestler W, Pacher M, Marton E, Kubista E, Schreiber M. Insulin-like growth factor (IGF)-I and IGF-II serum concentrations in patients with benign and malignant breast lesions: free IGF-II is correlated with breast cancer size. Clin Cancer Res 2004; 10: 40039.
  • 42
    Holdaway IM, Mason BH, Lethaby AE, Singh V, Harman JE, MacCormick M, Civil ID. Serum levels of insulin-like growth factor binding protein-3 in benign and malignant breast disease. Aust N Z J Surg 1999; 69: 495500.
  • 43
    Thomas DB, Gao DL, Ray RM, Wang WW, Allison CJ, Chen FL, Porter P, Hu YW, Zhao GL, Pan LD, Li W, Wu C, et al. Randomized trial of breast self-examination in Shanghai: final results. J Natl Cancer Inst 2002; 94: 14457.
  • 44
    Wu C, Ray RM, Lin MG, Gao DL, Horner NK, Nelson ZC, Lampe JW, Hu YW, Shannon J, Stalsberg H, Li W, Fitzgibbons D, et al. A case-control study of risk factors for fibrocystic breast conditions: Shanghai nutrition and breast disease study, China, 1995–2000. Am J Epidemiol 2004; 160: 94560.
  • 45
    Thomas DB, Gao DL, Self SG, Allison CJ, Tao Y, Mahloch J, Ray R, Qin Q, Presley R, Porter P. Randomized Trial of Breast Self-examination in Shanghai: Methodology and Preliminary Results. J Natl Cancer Inst 1997; 89: 355365.
  • 46
    Aaman TB, Stalsberg H, Thomas DB. Extratumoral breast tissue in breast cancer patients: a multinational study of variations with age and country of residence in low- and high-risk countries. WHO Collaborative Study of Neoplasia and Steroid Contraceptives. Int J Cancer 1997; 71: 3339.
  • 47
    Szklo M, Nieto FJ. Epidemiology: beyond the basics. Gaithersburg, MD: Aspen Publishers, 2000. 275p.
  • 48
    Johansson H, Baglietto L, Guerrieri-Gonzaga A, Bonanni B, Mariette F, Macis D, Serrano D, Teresa SM, Decensi A. Factors associated with circulating levels of insulin-like growth factor-I and insulin-like growth factor binding protein-3 in 740 women at risk for breast cancer. Breast Cancer Res Treat 2004; 88: 6373.
  • 49
    Enriori PJ, Fischer CR, Gori JR, Etkin AE, Calandra RS, Luthy IA. Augmented serum levels of the IGF-I/IGF-binding protein-3 ratio in pre-menopausal patients with type I breast cysts. Eur J Endocrinol 2003; 148: 17784.
  • 50
    Li W, Ray RM, Lampe JW, Lin MG, Gao DL, Wu C, Nelson ZC, Fitzgibbons ED, Horner N, Hu YW, Shannon J, Satia JA, et al. Dietary and other risk factors in women having fibrocystic breast conditions with and without concurrent breast cancer: a nested case-control study in Shanghai, China. Int J Cancer 2005; 115: 98193.
  • 51
    Thierry van Dessel HJ, Chandrasekher Y, Yap OW, Lee PD, Hintz RL, Faessen GH, Braat DD, Fauser BC, Giudice LC. Serum and follicular fluid levels of insulin-like growth factor I (IGF-I), IGF-II, and IGF-binding protein-1 and -3 during the normal menstrual cycle. J Clin Endocrinol Metab 1996; 81: 122431.
  • 52
    Juul A, Scheike T, Pedersen AT, Main KM, Andersson AM, Pedersen LM, Skakkebaek NE. Changes in serum concentrations of growth hormone, insulin, insulin-like growth factor and insulin-like growth factor-binding proteins 1 and 3 and urinary growth hormone excretion during the menstrual cycle. Human Reprod 1997; 12: 21238.
  • 53
    Ovesen P, Vahl N, Fisker S, Veldhuis JD, Christiansen JS, Jorgensen JO. Increased pulsatile, but not basal, growth hormone secretion rates and plasma insulin-like growth factor I levels during the periovulatory interval in normal women. J Clin Endocrinol Metab 1998; 83: 16627.
  • 54
    Helle SI, Anker GB, Meadows KA, Holly JM, Lonning PE. Alterations in the insulin-like growth factor system during the menstrual cycle in normal women. Maturitas 1998; 28: 25965.
  • 55
    Lukanova A, Toniolo P, Akhmedkhanov A, Hunt K, Rinaldi S, Zeleniuch-Jacquotte A, Haley NJ, Riboli E, Stattin P, Lundin E, Kaaks R. A cross-sectional study of IGF-I determinants in women. Eur J Cancer Prev 2001; 10: 44352.
  • 56
    Lavigne JA, Wimbrow HH, Clevidence BA, Albert PS, Reichman ME, Campbell WS, Barrett JC, Hursting SD, Judd JT, Taylor PR. Effects of alcohol and menstrual cycle on insulin-like growth factor-I and insulin-like growth factor binding protein-3. Cancer Epidemiol Biomarkers Prev 2004; 13: 22647.
  • 57
    Gronbaek H, Flyvbjerg A, Mellemkjaer L, Tjonneland A, Christensen J, Toft Sorensen H, Overvad K. Serum insulin-like growth factors, insulin-like growth factor binding proteins, and breast cancer risk in postmenopausal women. Cancer Epidemiol Biomarkers Prev 2004; 13: 175964.