Abstract
Previous studies from our group and others have shown that increased circulatory levels of the liegend Insulin-Like Growth Factor 1 (IGF-1) and decreased levels of the predominant IGF-1 binding protein (IGFBP3) are associated with increased incidence of breast cancer and poor outcome. Some studies suggest that in addition to the influence of environmental factors on the levels of IGF-I and IGFBP3, alterations in their gene polymorphisms could play a significant role toward the risk for cancer. In this study, we investigated the association between gene polymorphisms along the IGF axis and breast cancer, including the IGF-1 (CA) dinucleotide repeat, IGFBP3 A-202C single nucleotide polymorphism, and the 2-bp deletion and (AGG)n repeat polymorphisms in the IGF Type I Receptor (IGF-IR). A total of 654 subjects, including both African-American and Hispanic/Latina subjects, were screened for various gene polymorphisms. IGF gene polymorphism genotyping was performed by PCR-GeneScan and PCR-RFLP methods. Our results demonstrate a significant association between the non-19/non-19 IGF-1 (CA)n polymorphism and breast cancer (OR = 1.75; 95% CI = 1.07–2.88; P=0.027). Furthermore, absence of the wildtype-19 allele and alleles <(CA)19 were strongly associated with breast cancer ( OR = 1.82; 95% CI = 1.20–2.77; P=0.005 and OR= 1.70; 95% CI = 1.19–2.43; P-value= 0.003, respectively). The association of the non-19/non-19 polymorphism with breast cancer was also more significant in premenopausal women (P=0.04). We did not find any significant association of the IGFBP-3 polymorphism with breast cancer. In the case of IGF-IR polymorphisms, the only significant trend was in the (AGG)5 allele, however, the frequency of this allele was very rare. In summary, our study demonstrated a significant association of IGF-1 polymorphisms and breast cancer. Future studies are necessary to understand the mechanistic value of these polymorphisms in breast cancer risk.
Keywords: insulin-like growth factor I, IGF-1, insulin-like growth factor binding protein 3, IGFBP3, insulin-like-growth factor I receptor, IGF-IR, breast cancer, genetic polymorphism, African Americans, Hispanic Americans
Introduction
Insulin-like growth factor-1 (IGF-1) is a critical growth factor that plays diverse biological roles. Most IGF-1 is found in circulation with the liver being the major source, but in vitro studies have clearly shown that almost all tissues express IGF-1 and the IGF-1 receptor (IGF-1R) (1). The bioavailability of IGF-1 in circulation is regulated by IGF binding proteins (IGFBPs) with the most highly associated IGF-1 binding protein being IGFBP-3. The IGFBPs enact a role in IGF-1 action by increasing the half-life of IGF-1 and sequestering IGF-1 from binding IGF-IR as a free ligand (1,2). Due to its important role in cellular proliferation and apoptosis, IGF-1 is an important mediator in the oncogenic processes (3,4).
Multiple studies have shown an association between high levels of IGF-1 and elevated risk of breast cancer (5–13). In addition, IGF-1 and IGFBP-3 circulatory levels have been associated with progression, recurrence, and survival of breast cancer (5,6). Though the levels of IGF-1 may be affected by environmental and lifestyle factors, there is evidence that a significant amount of IGF-1 expression may be influenced by genetic variables (14). Studies have suggested that one such genetic variable may be a (CA)n dinucleotide repeat polymorphism 969 base pairs upstream from the transcription start site which may affect transcriptional levels of IGF-1 (15). Polymorphic simple sequence repeats (SSR’s), including (CA)n repeats, have been extensively studied and implicated in the etiology of many diseases, exerting regulatory roles over transcription often in an inverse manner (16). Thus, the (CA)n promoter polymorphism in IGF-1 may modulate tumorogenesis via increased serum levels (17–20). The repeats of the CA dinucleotide sequence have been reported to range from 13 to 25 times with the 19 repeat being the most common (18). In the IGFBP-3 gene, the −202 locus has a single nucleotide polymorphism adenine to cytosine (A>C) that can produce three genotypes; AA (wild type), AC (Heterozygous) and CC (mutant Homozygous). The IGFBP-3 polymorphism is implicated in breast cancer because it could affect the transcriptional levels of IGFBP-3 and consequently regulate IGF-1 availability and function. Studies have found that the serum levels of IGFBP-3 are inversely associated with the polymorphic status (19, 20). The IGF-1 receptor is a well studied tyrosine receptor kinase receptor which stimulates both the pro-growth MAPK signal transduction pathways and the anti-apoptotic PI3K/Akt pathway (1). Furthermore, IGF-IR has been identified as being an important component in tumor transformation and promoting tumor growth via interactions with oncogenes and tumor suppressor genes like p53 (21, 22). Higher levels of IGF-IR have been associated with some tumors (23, 24) where the increased levels of the receptor could presumably provide increased IGF-1 mediated signaling. Polymorphisms that affect the non-coding regulatory regions of genes, such as untranslated regions (UTR’s) which contribute to the stability of mRNA, may therefore contribute to tumorogenesis and progression by increasing the half-life of the mRNA transcript resulting in increased protein translation. There are two gene polymorphisms of interest in the non-coding regions of IGF-IR which were investigated in this study. One polymorphism is a microsatellite trinucleotide repeat of AGG, which can vary from being repeated five to nine times, with seven repeats (AGG)7 considered the most common and “Wildtype.” The second polymorphism is a 2 base pair deletion (2bp-del) in the 3’UTR region. Thus, our current study aims to assess the potential association between genetic polymorphisms in IGF-1, IGFBP-3, and IGF-IR genes, and breast cancer.
Methods and Materials
Subject Selection
The women from this study were recruited from South Los Angeles County, CA. Women visiting the Mammography Clinic or the Hematology/Oncology Clinic at the Martin Luther King Ambulatory Care Center (MACC, formerly known as King-Drew Medical Center) between 1995 and 2005 were given the option to participate in an ongoing breast cancer study conducted in the Division of Cancer Research and Training at Charles Drew University of Medicine and Science and MACC. Recruitment was conducted by bi-lingual interviewers who obtained an informed and signed consent from subjects approved by the Internal Review Board at Charles Drew University. Upon consent, participating women donated a 5–7 ml blood sample that was separated into plasma, serum, red blood cells, and buffy coat components and stored at −80° C until experimental use.
The women included in this study were subjects from our existing database in the Division of Cancer Research and Training at Charles Drew University. The majority of the subjects were self-reported as African-American or Hispanic/Latina, and only these two ethnic groups were included in the study. Ethnic composition of the study reflected the patient population at the MACC medical center in South Central Los Angeles which is comprised primarily of underserved African-American and Hispanic/Latina populations. The 5% of subjects who were self-reported as Caucasian and Asian were not included in the study since it was not possible to perform meaningful analysis with the small number of subjects from these ethnicities. Personal history and medical information was abstracted from the subject’s medical chart.
Subject Category Definitions
Women categorized as “cases” or “women with breast cancer” were women for whom breast cancer status was determined by biopsy/pathology confirmed neoplasm of the breast. Only subjects who had clear documentation of biopsy/pathology information were included in the study. The women categorized as “controls” or “women without breast cancer” included normal subjects who came to the Mammography Clinic for routine mammogram and had a normal mammography result; and women who had radiology/pathology confirmed benign breast disease. Only subjects who had documentation of either disease-free mammography or biopsy with benign results were included in the study. Only African-American and Hispanic subjects older than 30 years old were included in the study in order to closer match controls with the cancer patients in the study who were usually older. An additional criterion for inclusion into the study was the availability of a blood sample from the subject and whether the quality of DNA extracted from the buffy coat of the donated blood samples was amplifiable for polymorphic analysis. We excluded women for whom breast cancer was not the primary cancer; or their breast cancer was not the first cancer in their lifetime; or subject who were undergoing current evaluation for cancer.
Clinicopathological Definitions
Clinicopathological tumor characteristics (stage, hormone receptor status, etc) for breast cancer patient information was extracted from the patient medical records Thus availability of complete medical records for each patient was also an inclusion requirement. For the breast cancer patients, the histology classification of the breast tumor was directly extracted from the patient’s Surgical Pathology Report in the medical chart. The TNM Staging is according to AJCC definitions. The lymph node status is defined “Negative” if N = N0, “Positive” if N = N1, N2, N3. Tumor size is according to the AJCC definitions. The receptor subject is defined as : ER+/PR+/HER2-; ER+/PR+/HER2+; ER−/PR−/HER2+; and Triple negative (ER−/PR−/HER2-). ER/PR status is considered “Positive” if >1% of tumor cell nuclei immunoreactive, or “Negative” if otherwise. HER2 status is considered “Positive” if HER2= 3+, “Negative” if HER2 = 0, 1+, 2+.
Genotyping of the IGF-1 Polymorphism
Genomic DNA of the subjects was extracted from buffy coat using a DNA extraction kit (Gentra Systems, MN). PCR-GeneScan method was used to genotype the dinucleotide repeats in the promoter region of IGF-1 gene. PCR amplification was done using the following primers:
Forward- 5’ ACCACTCTGGGAGAAGGGTA 3’, Reverse- 5’ GCTAGCCAGCTGGTGTTATT 3’ (Retrogen Inc., USA). The forward primer was Fam labeled for fluorescence detection. The PCR mixture had a total volume of 10µl and consisted of 100 ng of genomic DNA, 6 pmol of each primer, 0.2 µmol/L dNTPs, 2.0 mmol/L MgCl2, 1.5 U Taq polymerase (Qiagen, USA) and 1µl 10X PCR buffer (pH 8). The cycling conditions began with an initial denaturation at 94°C for 5 minutes which was then followed by 35 cycles at 94°C for 30 s, 61°C for 30 s, 72°C for 30 s and a final extension for 7 minutes at 72°C. PCR products were sent to UCLA Sequencing & Genotyping Core for GeneScan. Of the PCR products, 3µl of each sample were mixed with 0.3 µl of LIZ500 and 10 µl of HIDI formamide. Samples were then denatured for 5 min at 95 °C and cooled on ice before loading on ABI3730XL. Analysis was done with GeneMapper (version 4.0) software. We received the fragment size, height, and area of peaks of each sample and calculated the number of dinucleotide CA repeats. Furthermore, we confirmed GeneScan results by sequencing random samples in order to ascertain the accuracy of the resulting number of repeats calculated in this manner.
Genotyping of the IGFBP-3 Polymorphism
The IGFBP-3 PCR amplification was performed using the forward primers 5'CCACGAGGT ACACACGAATG- 3' and reverse primers 5'AGCCGCAGTGCTCGCATCTGG −3'. The PCR mixture had a total volume of 25 µL and consisted of 100ng genomic DNA, 1 pmol of primers, 1.5 mmol/L MgCl2, 0.1 mmol/L of each dNTPs, 2 U Taq DNA polymerase (Qiagen, USA), 5 µl Q-solution and 2.5µl 10X PCR buffer. The reaction was carried out using a 10 minute denaturation at 94°C, followed by 36 cycles of 40s at 94°C, 40s at 60.4°C, 40s at 72°C and a final extension of 7 min at 72°C. The IGFBP3 A-202C polymorphism was identified using restriction fragment length polymorphism (RFLP) analysis. From the PCR product, 8µl was digested with 5U of Alw21I (Fermentas, Canada) in a total volume of 15µl for 16–18 h at 37°C. Digested products were visualized on a 1.5% agarose gel stained with ethidium bromide. The amplified IGFBP-3 PCR product contains three Alw21I sites, one of which is eliminated when there is a C at −202. Thus, the AA genotype results in 242 and 162 bp products, AC genotype results in 280, 242, and 162 bp products, and the CC genotype results in 280 and 162bp products.
Genotyping of the IGF-IR Polymorphisms
Identification of (AGG)n and 2bp-del polymorphisms in IGF-IR were performed using the PCR-GeneScan method as described by Chen et al (25). The PCR primers used to amplify the region of interest for the AGG repeat were the following: Forward (Fam Labeled): 5’-GCTGAGGGAGGAGGCGGC-3’ and Reverse: 5’ – GGCGAGGGGCAGAAACGC – 3’: Nested Forward: 5’- CCTGGATTTGGGAAGGAGCTCG-3’; Nested Reverse: 5’-GAAGTCCGGGTCACAGGCGA-3’. The primers used to amplify the region of interest for the 2bp deletion were the following: Forward: 5’-CTCCTCTCTGCTTCATAACG-3’: Reverse (TET-Labeled) 5’-TCCGGACACGAGGAATCAGC-3’. The cycling conditions were according to the conditions used by Chen et al(25). PCR products were sent to UCLA Sequencing & Genotyping Core for GeneScan as described above. We received the fragment size, height, and area of peaks of each sample and calculated the number of trinucleotide repeats and identified the presence/absence of the 2 base pair deletion.
Analysis
The parameters considered are age, ethnicity, breast cancer status, IGF-1, IGFBP-3, and IGF-IR polymorphic genotypes. IGF-I genotypes were compared as (1) 19 CA repeat status in both alleles (19/19, 19/non-19, and non-19/non-19); or (2) CA repeats in one allele. Based on previous studies of the IGF-1 (CA)n reflective of the postulated transcriptional consequences exerted by the length of the repeats (references 15,17, 26–37], the CA repeats in both alleles less or more than 19 CA repeats were also compared to those with equal 19 CA repeats (<(CA)19, =(CA)19, >(CA)19 ). When IGF-1 (CA)n repeats in one allele were used for comparison, the CA repeat information from each subject was used twice (i.e. a subject with a IGF-1 (CA)n genotype of 19/20, would be represented in both the IGF-1 (CA)19 percentage and the IGF-1 (CA)20 percentage). The purpose of the Group 1 comparison is to assess the effect of the wildtype vs. non-wildtype genotype in the association of this polymorphism with breast cancer. The addition of the Group 2 comparison was to assess whether the proposed functional relevance of the polymorphic dinucleotide repeat length was significant in the context of cancer. As mentioned in the introduction, the length of the repeats may be inversely associated with transcriptional levels of the gene; since levels of IGF-1 have been associated with breast cancer and outcomes, this was an important comparison to consider. Group 3 assesses the presence of the wildtype allele compared to the absence of the wildtype allele. The IGF-IR 2 bp-del polymorphism was grouped by the presence of the wildtype allele (non-del) compared to the presence of the polymorphic 2 bp-del. The variables in this resulting group were non-del/non-del, non-del/ 2bp-del, and 2bp-del/2bp-del. The (AGG) polymorphism in IGF-IR was categorized in two ways. In the first, called Group 1, the polymorphism grouping was based on the presence or absence of the wildtype (AGG)7 genotype compared to the non-wildtype genotypes; (AGG)7/ (AGG)7, (AGG)7/(AGG)non, (AGG)non/(AGG)non. The second method, as in Group 2, was the actual number of repeats; (AGG)n/(AGG)n. These two grouping methods were chosen such that both the overall affect of the presence or the absence of the wildtype allele could be assessed, as well as the affect of the specific polymorphic genotypes.
Statistical analysis was performed using SPSS software (SPSS, Inc., Chicago, IL). Subject’s characteristics in different groups were compared by χ2-test. The distributions of CA repeats in both alleles, and the variation of 19 CA repeats of IGF-I were compared between breast cancer and non-cancer using χ2 –test. Similarly, the differences of IGFBP-3 genotypes in breast cancer and non-breast cancer were compared by χ2-test. The distribution of both the IGF-IR polymorphisms, in all grouping methods, among breast cancer vs. non-cancer subjects also used the χ2-test. The association of breast cancer with IGF-1, IGFBP-3, and IGF-IR genotypes were determined by logistic regression with univariate and multivariate analysis. Furthermore, the association between the IGF genotypes and the clinicopathological features were assessed by the χ2-test. All multivariate analysis performed were adjusted for age and BMI category (<18.5 Underweight, 18.5–24.9 Normal Weight, 25–29.9 Overweight, >30 Obese). Only the two-sided P-value less than 0.05 was considered statistically significant.
Results
Characteristics of Study Population
A total of n = 654 subjects are assessed in this study, with n = 268 breast cancer patients and n = 386 women with no breast cancer. A description of the menopausal status and ethnicity of the subjects is shown in Table I. When assessing the self-reported ethnic description of the study subjects, there are similar number of cases from both ethnicities (n = 138 for African American, n = 130 for Latinas). However, there is a significant difference (P<0.01) in the distribution of controls, with more Latina women in the study than African-American women. The clinicopathological features of the breast cancer patients are also described in Table I. The majority of cancer patients in this study have a breast tumor histology described as infiltrating ductal carcinoma (75%), followed by infiltrating lobular carcinomas (18%), with the remainder having infiltrating adenocarcinoma, ductal carcinoma in situ, and lobular carcinoma in situ.
Table I.
Characteristics of Study Population
| Case N (%) |
Control N (%) |
P-value | |
|---|---|---|---|
| Total Subjects in Study (n = 654) | 268 (41) | 386 (59) | |
| Ethnicity | |||
| African American | 138 (51) | 85 (22) | < 0.01 |
| Hispanic/Latina | 130 (49) | 301 (78) | |
| Histology (n=227) | |||
| Infiltrating ductal carcinoma | 170 (75) | --- | --- |
| Infiltrating lobular carcinoma | 41 (18) | --- | |
| Infiltrating adenocarcinoma | 9 (04) | --- | |
| Ductal carcinoma in situ | 2 (01) | --- | |
| Lobular carcinoma in situ | 5 (02) | --- | |
| Stage (n=218) | |||
| 0-II | 146 (67) | --- | --- |
| III-IV | 72 (33) | --- | |
| Lymph node (n=225) | |||
| Negative | 94 (42) | --- | --- |
| Positive | 131 (58) | --- | |
| Tumor size (n=221) | |||
| T0/Tis/T1 | 54 (24) | --- | --- |
| T2 | 102 (46) | --- | |
| T3/T4 | 65 (30) | --- | |
| ER/PR status (n=226) | |||
| ER/PR positive | 127 (56) | --- | --- |
| ER/PR negative | 99 (44) | --- | |
| HER2 status (n=195) | |||
| HER2 negative | 45 (23) | --- | --- |
| HER2 positive | 150 (77) | --- | |
| Subtype (n=194) | |||
| Luminal A | 85 (44) | --- | --- |
| Luminal B | 16 (08) | --- | |
| ER/PR−/HER2+ | 28 (14) | --- | |
| Triple negative | 65 (34) | --- | |
Age cut-off was for subject selection as > age 30.
Distribution of IGF-1 (CA)n Genotype in the Study Population
The IGF-1 genotype distributions in the total cohort and in subcategories based on menopausal status are shown in Table II. In the IGF-1 genotype distribution, alleles characterized by the number of CA repeats ranged from (CA)14 to (CA)25. The (CA)19 allele was most common across all categories and subdivisions (> 47%). Some general trends in the different distribution of alleles between cases and controls were observed. When grouping the subjects by the polymorphic genotypes (19/19, 19/non-19, non-19/non-19) as in Group 1, there was a significant difference between cases and controls in the total cohort (P = 0.03). The non-19/non-19 genotype was present in 31.7% of cases compared to only 20.3% of controls. Furthermore, when assessing the difference in the genotype distribution by menopausal status, there was a statistically significant association observed with the non-19/non-19 genotype in premenopausal women with breast cancer. The non-19/non-19 genotype was present in 32.1% of premenopausal cases and only 18.9% of premenopausal controls (P=0.04). Stratification according to the specific length of the genotype also revealed that the <(CA)19 alleles were more frequently found in cases than in controls. This trend was observed when grouping the frequency of alleles by <(CA)19, =(CA)19 and >(CA)19 as in Group 2. The percentage of cases with the <(CA)19 allele was 21% compared to 14% of controls. When assessing the “presence of the 19 allele” vs the “absence of the 19 allele” in the study subjects as in Group 3, there was a statistically significant association of the latter genotypes with breast cancer in the total cohort (P<0.01) as well as in premenopausal cases (P=0.01). In sum, the distribution of the (CA)n alleles revealed that there was an association of the <(CA)19 allele as well as the non-(CA)n alleles with breast cancer, especially in premenopausal women.
Table II.
Distribution of IGF-1 (CA)n Genotype in the Study Population
| Total Subjects (n = 475) |
Total |
Premenopausal |
Postmenopausal |
|||
|---|---|---|---|---|---|---|
| Case (n=199) |
Control (n=276) |
Case (n=106) |
Control (n=164) |
Case (n=93) |
Control (n=112) |
|
| IGF-1 (CA)n genotype | ||||||
| (CA)14 | 0.3 % | ---- | ---- | ---- | 0.5 % | ---- |
| (CA)15 | 0.5 % | 0.4 % | 0.9 % | 0.6 % | 0.0 % | ---- |
| (CA)16 | 2.0 % | 2.4 % | 1.9 % | 2.7 % | 2.2 % | 1.8% |
| (CA)17 | 4.3 % | 1.6 % | 4.7 % | 2.1 % | 3.8 % | 0.9% |
| (CA)18 | 14.4 % | 10.2 % | 13.7 % | 9.1 % | 15.1 % | 11.6% |
| (CA)19 | 48.0 % | 55.3 % | 47.9 % | 57.0 % | 47.8 % | 52.2% |
| (CA)20 | 14.6 % | 11.8 % | 15.2 % | 9.5 % | 14.0 % | 15.2% |
| (CA)21 | 14.1 % | 17.1 % | 15.2 % | 16.5 % | 12.9 % | 17.9% |
| (CA)22 | 1.5 % | 1.1 % | 0.5 % | 1.5 % | 2.7 % | 0.4% |
| (CA)23 | 0.3 % | 0.4 % | ---- | 0.6 % | 0.5 % | ---- |
| (CA)24 | 0.3 % | ---- | ---- | ---- | 0.5 % | ---- |
| (CA)25 | ---- | 0.2 % | ---- | 0.3 | 0.0 % | ---- |
| Range (CA)n | 14–24 | 15–25 | 15–22 | 15–25 | 14–24 | 16–22 |
| IGF-1 (CA)n genotype groups | ||||||
| Group 1† | ||||||
| 19/19 | 27.1 % | 30.4 % | 27.4 % | 32.9 % | 26.9 % | 26.8 % |
| 19/non-19 | 41.2 % | 49.3 % | 40.6 % | 48.2 % | 41.9 % | 50.9 % |
| non-19/non-19 | 31.7 % | 20.3 % | 32.1 % | 18.9 % | 31.2 % | 22.3 % |
| P-value = 0.03 | P-value = 0.04 | P-value = 0.39 | ||||
| Group 2 | ||||||
| < (CA)19 | 21.4 % | 14.5 % | 21.5 % | 14.6 % | 21.5 % | 14.3 % |
| = (CA)19 | 47.9 % | 55.1 % | 47.8 % | 57.0 % | 47.8 % | 52.2 % |
| > (CA)19 | 30.7 % | 30.4 % | 30.7 % | 28.4 % | 30.6 % | 33.5 % |
| P-value = 0.19 | P-value = 0.13 | P-value = 0.86 | ||||
| Group 3†† | ||||||
| 19 Allele Present | 68.3 % | 79.7 % | 67.9 % | 81.1 % | 67.9 % | 77.7 % |
| 19 Allele Absent | 31.7 % | 20.3 % | 32.1 % | 18.9 % | 32.1 % | 22.3 % |
| P-value = < 0.01 | P-value = 0.01 | P-value = 0.10 | ||||
(CA)n=19 repeats is the Wildtype (Non-polymorphic Genotype). All other repeats (CA)n<19 or (CA)n >19 are considered non-19.
The “19 Allele Present” is either the 19/19 or 19/non-19 genotype. The “19 Allele Absent” is the non-19/non-19 genotype.
Distribution of the IGFBP-3 Genotype in the Study Population
The distribution of IGFBP-3 genotypes is shown in Table III. In the total cohort, the most common genotype was the doubly polymorphic CC (present in 42.4 – 46.4% of study subjects) and the least common genotype was the wildtype AA genotype (present in 13.8–18.6% of study subjects). There were not any statistically significant trends in the distribution of the IGFBP-3 polymorphic genotype between cases and controls in any of the categories.
Table III.
Distribution of the IGFBP-3 Genotype in the Study Population
| Total Subjects (n=418) |
Total |
Premenopausal |
Postmenopausal |
|||
|---|---|---|---|---|---|---|
| Case (n=172) |
Control (n=246) |
Case (n=103) |
Control (n=147) |
Case (n=69) |
Control (n=99) |
|
| IGFBP-3 Genotype† | ||||||
| AA | 18.6 % | 13.8 % | 19.4 % | 13.6 % | 17.4 % | 14.1 % |
| AC | 39.0 % | 39.8 % | 41.7 % | 40.1 % | 34.8 % | 39.4 % |
| CC | 42.4 % | 46.4 % | 38.9 % | 46.3 % | 47.8 % | 46.5 % |
| P-value = 0.227 | P-value = 0.153 | P-value = 0.868. | ||||
The Wildtype (Non-polymorphic Genotype) is AA.
Distribution of the IGF-IR Genotype in the Study Population
The distribution of the IGFIR polymorphisms is shown in Table IV. The distribution of the IGFIR 2- base-pair deletion polymorphic genotype in the study population showed the wildtype genotype (non-del/non-del) present in >60% of the subjects, with the heterozygous genotype (non-del/2bp-del) present in ∼30% of subjects, and the doubly polymorphic genotype (2bp-del/2bp-del) being present in <10% of the subjects. There was a modest trend of the 2bp-del/2bp-del genotype being more prevalent in cases (7.7%) than in controls (5.1%), however these differences were not statistically significant. This modest trend was also observed in premenopausal cases and controls (6.6% vs. 4.2%, respectively) as well as in postmenopausal cases and controls (8.6% vs. 6.5%, respectively).
Table IV.
Distribution of the IGF-IR Genotype in the Study Population
| Total Subjects (n=496) |
Total |
Premenopausal |
Postmenopausal |
|||
|---|---|---|---|---|---|---|
| Case (n=222) |
Control (n=274) |
Case (n=106) |
Control (n=166) |
Case (n=116) |
Control (n=108) |
|
| IGF-IR 2bp deletion† | ||||||
| non-del/non-del | 64.9 % | 64.2 % | 65.1 % | 66.9 % | 64.7 % | 60.2 % |
| non-del/ 2bp-del | 27.4 % | 30.7 % | 28.3 % | 28.9 % | 26.7 % | 33.3 % |
| 2bp-del/2bp-del | 7.7 % | 5.1 % | 6.6 % | 4.2 % | 8.6 % | 6.5 % |
| P-value = 0.73 | P-value =0.57 | P-value = 0.78 | ||||
| IGF-IR (AGG)n†† | ||||||
| (AGG)5 | 3.0 % | 0.9 % | 3.1 % | 1.3 % | 2.9 % | 0.5 % |
| (AGG)6 | 26.6 % | 31.7 % | 29.7 % | 30.4 % | 23.6 % | 33.6 % |
| (AGG)7 | 69.8 % | 67.4 % | 66.7 % | 68.3 % | 72.7 % | 65.9 % |
| (AGG)8 | 0.4 % | ---- | 0.5 % | ---- | 0.4 % | ---- |
| (AGG)9 | 0.2 % | ---- | ---- | ---- | 0.4 % | ---- |
| Range (AGG)n | (AGG)5–9 | (AGG)5–7 | (AGG)5–8 | (AGG)5–7 | (AGG)5–9 | (AGG)5–7 |
| Group 1 | ||||||
| (AGG)7 / (AGG)7 | 47.8 % | 44.6 % | 45.9 % | 47.8 % | 49.6 % | 40.0 % |
| (AGG)7 / (AGG)non | 43.9 % | 45.4 % | 41.4 % | 41.0 % | 46.2 % | 51.8 % |
| (AGG)non / (AGG)non | 8.3 % | 10.0 % | 12.6 % | 11.2 % | 4.2 % | 8.2 % |
| P-value = 0.40 | P-value = 0.69 | P-value = 0.09 | ||||
| Group 2 | ||||||
| (AGG)5 / (AGG)6 | 0.4 % | ---- | 0.9 % | ---- | ---- | ---- |
| (AGG)5 / (AGG)7 | 5.7 % | 1.8 % | 5.4 % | 2.5 % | 5.9 % | 0.9 % |
| (AGG)6 / (AGG)6 | 7.8 % | 10.0 % | 11.7 % | 11.2 % | 4.2 % | 8.2 % |
| (AGG)6 / (AGG)7 | 37.0 % | 43.5 % | 35.1 % | 38.5 % | 38.7 % | 50.9 % |
| (AGG)7 / (AGG)7 | 47.8 % | 44.6 % | 45.9 % | 47.8 % | 49.6 % | 40.0 % |
| (AGG)7 / (AGG)8 | 0.9 % | ---- | 0.9 % | ---- | 0.8 % | ---- |
| (AGG)7 / (AGG)9 | 0.4 % | ---- | ---- | ---- | 0.8 % | ---- |
| P-value = 0.30 | P-value = 0.98 | P-value = 0.10 | ||||
The Wildtype genotype is the “non-del” genotype which signifies there is no 2 base pair deletion present. The term “2bp-del” signifies the polymorphic genotype with the 2 base pair deletion.
The (AGG)7 is considered the Wildtype genotype.
When assessing the IGFIR (AGG)n repeat polymorphism, the distribution of the (AGG)n alleles was around 45% for the wildtype genotype (AGG)7/(AGG)7, around 45% for the heterozygous repeat genotype (AGG)7/(AGG)non, and around 10% for the polymorphic genotype (AGG)non/(AGG)non. The specific (AGG)non genotype seen with a higher frequency in the cases was the (AGG)5 repeat with ∼3.0% of cases having this genotype compared to ∼1.0% of controls. Furthermore, the >(AGG)7 alleles, (AGG)8 and (AGG)9 were only seen in cases. The (AGG)6 genotype was more prevalent in the controls than in cases, with the distribution difference at its greatest (by 10%) in postmenopausal women – (23.6% in cases vs. 33.6% in controls). When stratifying the genotype distribution by (AGG)7 vs. (AGG)non, the (AGG)non/(AGG)non genotype was overall more present in controls than in cases. In postmenopausal women 8.2% of controls had the (AGG)non genotype compared to 4.2% of cases, though the difference was not statistically significant (P=0.09).
Odds ratios and 95% confidence intervals for the IGF-1, IGFBP-3, IGF-IR genotypes with breast cancer
The results for univariate and multivariate analysis of the association between IGF polymorphisms and breast cancer are shown in Table V. For the IGF-1 (CA)n repeat, there was a significant association of the non-19/non-19 genotype with breast cancer by univariate analysis (OR = 1.75; 95% CI = 1.07–2.88; P=0.027). When adjusting for age and BMI, there was still a statistically significant association of breast cancer with the non-19/non-19 genotype (OR=1.98; 95% CI = 1.01–3.89; P=0.047). Considering the association of single alleles with breast cancer, the <(CA)19 allele was significantly associated with breast cancer both in univariate analysis (OR = 1.70; 95% CI = 1.19–2.43; P=0.003) and multivariate analysis (OR=1.97; 95% CI = 1.21–3.22; P=0.006). Furthermore, the “absence of the 19 allele” was strongly associated with breast cancer in both univariate (OR=1.82; 95% CI 1.20–2.77; P=0.005) and multivariate analyses (OR = 1.87; 95% CI=1.05–3.32; P=0.005). The IGFBP-3 polymorphisms showed no significant association with breast cancer. The IGFIR 2bp deletion polymorphism analysis did not reveal any significant association with breast cancer. The IGF-IR (AGG)n repeat analysis did show some significant association of the polymorphic genotypes with breast cancer. In the univariate analysis, the (AGG)5 genotype was significantly associated with breast cancer (OR=3.18; 95% CI = 1.31–8.94; P=0.028). However, when the analysis was adjusted for age and BMI, the significance was lost (OR = 1.64; 95% CI= 0.39–6.89; P = 0.501). The rarity of the allele in the study population may account for this finding or it may be due to the influence of age and/or BMI on the affects from this allele. When discretely grouping the (AGG)n polymorphism by wildtype (AGG)7 vs. (AGG)non, there were no statistically significant associations. However, when considering the distribution of the various allele combinations, the (AGG)5/(AGG)7 allele was bordering statistical significance by univariate analysis (OR = 2.86; 95% CI = 0.99–8.28; P = 0.053). This association observation was once again lost when adjusting for age and BMI in the multivariate analysis (OR = 1.33; 95% CI 0.29–6.07; P = 0.716).
Table V.
Odds ratios and 95% confidence intervals for the IGF-1, IGFBP-3, IGF-IR genotypes with breast cancer
| Univariate |
Multivariate1 |
|||
|---|---|---|---|---|
| OR (95% CI) | P-value | OR (95% CI) | P-value | |
| IGF-1 (CA)n Genotype | ||||
| Group 1 | ||||
| 19/19 | 1.00 (ref) | ---- | 1.00 (ref) | ---- |
| 19/non-19 | 0.94 (0.61–1.45) | 0.774 | 1.10 (0.63–1.93) | 0.739 |
| non-19/non-19 | 1.75 (1.07–2.88) | 0.027 | 1.98 (1.01–3.89) | 0.047 |
| Group 2 | ||||
| < (CA)19 | 1.70 (1.19–2.43) | 0.003 | 1.97 (1.21–3.22) | 0.006 |
| = (CA)19 | 1.00 (ref) | ---- | 1.00 (ref) | ---- |
| > (CA)19 | 1.16 (0.87–1.56) | 0.319 | 1.16 (0.79–1.70) | 0.449 |
| Group 3 | ||||
| 19 Allele Present | 1.00 (ref) | ---- | 1.00 (ref) | ---- |
| 19 Allele Absent | 1.82 (1.20–2.77) | 0.005 | 1.87 (1.05–3.32) | 0.005 |
| IGFBP-3 Genotype | ||||
| AA | 1.00 (ref) | ---- | 1.00 (ref) | ---- |
| AC | 0.73 (0.41–1.29) | 0.275 | 0.59 (0.27–1.27) | 0.177 |
| CC | 0.68 (0.39–1.20) | 0.182 | 0.73 (0.34–1.55) | 0.409 |
| IGF-IR 2bp deletion | ||||
| non-del/non-del | 1.00 (ref) | ---- | 1.00 (ref) | ---- |
| non-del/ 2bp-del | 0.90 (0.62–1.30) | 0.560 | 1.04 (0.59–1.83) | 0.902 |
| 2bp-del/2bp-del | 1.57 (0.78–3.16) | 0.212 | 1.16 (0.39–3.48) | 0.796 |
| IGF-IR (AGG) n | ||||
| (AGG)5 | 3.18 (1.31–8.94) | 0.028 | 2.22 (0.43–11.4) | 0.338 |
| (AGG)6 | 0.81 (0.61–1.06) | 0.434 | 0.93 (0.62–1.38) | 0.703 |
| (AGG)7 | 1.00 (ref) | ---- | 1.00 (ref) | ---- |
| (AGG)8† | ---- | ---- | ---- | ---- |
| (AGG)9† | ---- | ---- | ---- | ---- |
| Group 1 | ||||
| (AGG)7 / (AGG)7 | 1.00 (ref) | ---- | 1.00 (ref) | ---- |
| (AGG)7 / (AGG)non | 0.90 (0.63–1.31) | 0.589 | 0.94 (0.54–1.64) | 0.831 |
| (AGG)non / (AGG)non | 0.77 (0.41–1.47) | 0.434 | 1.05 (0.46–2.43) | 0.905 |
| Group 2 | ||||
| (AGG)5 / (AGG)6† | ---- | ---- | ---- | ---- |
| (AGG)5 / (AGG)7 | 2.86 (0.99–8.28) | 0.053 | 1.33 (0.29–6.07) | 0.716 |
| (AGG)6 / (AGG)6 | 0.77 (0.38–1.41) | 0.350 | 0.99 (0.43–2.31) | 0.984 |
| (AGG)6 / (AGG)7 | 0.79 (0.54–1.16) | 0.230 | 0.86 (0.49–1.52) | 0.600 |
| (AGG)7 / (AGG)7 | 1.00 (ref) | ---- | 1.00 (ref) | ---- |
| (AGG)7 / (AGG)8† | ---- | ---- | ---- | ---- |
| (AGG)7 / (AGG)9† | ---- | ---- | ---- | ---- |
Adjusted by Age and BMI category (Underweight, Normal Weight, Overweight, or Obese).
Insufficient number to calculate OR
Odds ratios and 95% confidence intervals for the IGF-1, IGFBP-3, IGF-IR polymorphic genotypes with clinicopathological features
The clinicopathological details from breast cancer patients were available for about n = 200 patients out of 268 total cancer patients in this study. There were no significant associations between IGF polymorphisms and tumor characteristics and receptor subtypes (data not shown).
Discussion
Our current study provides additional support for the association between IGF-1 polymorphic genotypes and breast cancer. Our findings show that breast cancer patients, especially premenopausal women, were more likely to be carriers of the non-19 (CA)n alleles and these alleles were more likely to be the <(CA)19 genotype. The association of the <(CA)19 genotype with breast cancer found in our study as well as data from a much larger case-control study (31), imply that the length of polymorphic repeats could inversely affect transcription. There was no association found with >(CA)19 repeats with a decreased risk of cancer; suggesting that the possible transcriptional effects resulting from the length of the polymorphism may be limited to the <(CA)19 repeats.
Breast cancer patients were also more likely to have the non-19/non-19 genotype when compared to women with no breast cancer. This finding was most significant in premenopausal breast cancer patients. We had previously published findings of a significant association of increased IGF-1 serum levels, in minority premenopausal women, with breast cancer, decreased response to treatment, and poor prognosis and outcome (6). The findings in this current study, which includes the same study population, found a significant association of the IGF-1 polymorphism with breast cancer, especially in premenopausal but not postmenopausal women. Consequently, these results suggest, to a modest extent, a potential role that IGF-1 levels and promoter polymorphisms may play in breast cancer in premenopausal women. The role of IGF-1 in premenopausal women with breast cancer is further supported by us and the findings of other investigations (20, 31, 38– 41).
One of the most convincing studies that demonstrated a direct link between IGF-1 (CA)n repeat and risk for breast cancer was the Long Island Breast Cancer Study Project by Cleveland et al (31). These investigators examined 1028 breast cancer cases and 1086 controls. They observed an increased risk of breast cancer for genotypes that included alleles with fewer than (CA)19 repeats when compared to (CA)19 repeat carriers. This association was stronger among premenopausal women. No significant association was observed between an IGF-1 genotype with no (CA)19 repeat compared to (CA)19 repeat genotypes in either pre or postmenopausal women. However, when traditional breast cancer risk factors were considered, premenopausal women with genotypes that lacked a (CA)19 repeat had a nearly 60% increased risk of breast cancer among those who had ever used hormonal birth control, while never users had a significantly reduced risk. Among postmenopausal women, those with genotypes lacking a (CA)19 repeat allele had significantly increased breast cancer risk among subjects with a lower than median body mass index (BMI), while no association for IGF-1 genotype was seen among women with a higher than median BMI.
Several other groups have suggested that polymorphisms of the IGF-1 gene and of genes encoding for the major IGF-1 carriers may predict circulating levels of IGF-1 and have an impact on cancer risk. Canzian et al (19) tested this hypothesis with a case–control study of 807 breast cancer patients and 1588 matched control subjects, nested within the European Prospective Investigation into Cancer and Nutrition (EPIC). The authors found a weak but significant association of polymorphisms at the 5’ end of the IGF-1 gene with breast cancer risk, particularly among women younger than 55 years, and a strong association of polymorphisms located in the 5’ end of IGFBP3 with circulating levels of IGFBP-3. Although, we observed a direct correlation with breast cancer risk in premenopausal African American and Hispanic women with increase in IGF-1 plasma levels, together with decrease in serum IGFBP3 levels (6), we have not determined these associations with their IGF gene polymorphic status.
Cheng et al (42) examined a large multiethnic cohort for an association between IGFBP1 and IGFBP3 genotypes with either prostate or breast cancer risk. However, unlike our current study, their breast cancer cohort was primarily postmenopausal. Their data did not observe a strong association between IGFBP1 and IGFBP3 genotypes with either prostate or breast cancer risk. The number of African Americans and Latinos in their study comprised of cases/controls as follows: AA 284/418; and Latinos 330/380. Although our current study has fewer subjects, our data also confirm no association between IGFBP3 polymorphism (A-202C) and risk for breast cancer in either population. Other investigators have reported similar findings primarily on Caucasian women (19). Al-Zaharani et al (43) observed similar risks for breast cancer between pre and postmenopausal women with either IGF-1 or IGFBP3 SNPS.
However, it is necessary to note that a limitation of our study was a smaller sample size (∼500 women) than the aforementioned studies, and this may play a role in assessing the associations of this genotype with breast cancer. Nonetheless, the suggested importance of this genotype playing a role in premenopausal women, especially perhaps minority women, does merit some consideration. These data, however, must be interpreted noting that this was a hospital based comparison study, where there was a trend difference in age distribution between cases and controls. Although a comparison study is a limitation, the fact that there is an association of the IGF-1 genotype with cancer in younger women, while cancer is predominately considered later-age malady, does somewhat support the strength of the relevance of these findings.
To date there have been only a few studies which have looked at the distribution of the IGF (CA)n alleles in African-American and Hispanic/Latina women, with breast cancer (36, 41–46). Slattery and colleagues (42) reported that Non-Hispanic White (NHW) women in Southwestern United States, who were not recently exposed to hormones, and did not have the 19 CA repeat of IGF-1 gene, showed an association with breast cancer. The R allele of G972R of IRS polymorphism was less common in Hispanic women and is associated with breast cancer risk in women not recently exposed to hormones. These authors demonstrated several interesting additional associations. For instance, among post-menopausal Hispanic women recently exposed to hormones the A allele of the −202 C > A IGFBP3 polymorphism increased risk of breast cancer. The IGF-1 19 CA repeat polymorphism interacted with hormone replacement therapy (HRT) among NHW post-menopausal women. Women who had the 19/19 IGF-1 genotype were at reduced risk of breast cancer if they did not use HRT. Their study also demonstrated association between body mass index (BMI) and IGF-1 19 CA repeat (p=0.06) and between weight gain and the −202 C > A IGFBP3 polymorphism (p=0.05) in NHW postmenopausal women not recently exposed to hormones. In our current study, due to limited sample size, we were not able to examine with statistical significance, the breast cancer risks associated with BMI and the use of HRT in either IGF-1 and/or IGFBP3 genotypes. However, these are important risk factors and warrant further studies on African American and Hispanic women.
Similar to our study which includes a fair number of African-American women, other studies assessing IGF-1 genotype and IGF-1 levels found that African-American women were more likely to be carriers of the non-19 allele, especially the <(CA)19 alleles, when compared to mostly Non-Hispanic White women (36,41). Furthermore, our study supports previous findings that the most common IGF-1 genotype is the heterozygous 19/non-19 IGF-1 genotype (36,41). An important study to consider was one performed by DeLellis et al (36). This study included African-American and Hispanic/Latina subjects from Los Angeles, Southern California. The authors found that the non-19/non-19 genotype was more prevalent in African-American women compared to the Hispanic/Latina women. In addition, the distribution of the IGF-1 alleles between the two ethnic groups showed a pattern similar to ours, with the highest prevalence once again being the heterozygous genotype, followed by the homozygous wildtype genotype, and finally the homozygous non-wildtype genotype (36). The distribution of the IGFBP-3 genotype in our study was similar to other studies that had included Hispanic/Latina women (45,46). There is a greater prevalence of the heterozygous AC and the homozygous CC genotypes than the homozygous wildtype genotype AA. Here it should be noted that in our study, pooling both African American and Hispanic/Latina women into one “total subjects” category was done in order to gain an understanding of the distribution of comorbidities and IGF gene polymorphisms and to have enough numbers of women to perform meaningful statistical analyses. A possible limitation of our current study may be that the subjects were not matched by ethnicity, and there were more Hispanic/Latina women represented than African-American in this clinic-based comparison study.
Perhaps the strongest association between IGF-1 CA repeats and risk for colorectal cases is found in the hereditary form of CRC (47,48). Zecevic et al (47) investigated the relationship between IGF-1 promoter cytosine-adenine (CA) dinucleotide –repeat polymorphism length and CRC risk in 121 MMR gene mutation carriers using Cox regression and Kaplan-Meier analysis. This study was primarily on Caucasian patients. Their data demonstrated a statistically significant association between shorter IGF-1 CA-repeat lengths and increased risk for CRC in HNPCC (Hereditary Non-polyposis Colon Cancer) carriers. Similar observations were made recently by Reeves et al (48), who examined both Australian and Polish patients with MMR gene mutations (equally in MLH1 or MSH2). The authors concluded that IGF CA repeat is an important modifier of disease onset in HNPCC and the first polymorphism to yield consistent results across different populations.
In case of the prostate study in African Americans, Hernandez et al (44) showed that the two polymorphisms, rs7965399 C/T SNP and IGF-1 (CA)n repeat, do not affect IGF-1 serum levels nor prostate cancer risk. In addition, these authors showed that although the IGFBP-3 serum levels are not associated with prostate cancer risk but the C allele of the 202 A/C SNP increases risk and lowers IGFBP-3 serum levels. The authors caution that the impact of these genotypes may not be significant, given the high rates of aggressive disease in their prostate cancer population. This possibility was recently examined by Hoyo et al (49) in albeit a small population of African Americans and Whites from the Durham Veterans Administration Hospital. Essentially, they grouped 47 cases with Gleason sum >7, 50 cases with Gleason sum <7 and compared to 93 Controls. Their data confirmed that the inverse association between carrying the IGF (CA)19 repeat variant did not vary with grade or ethnicity. However, the association between IGFBP3 C allele and prostate cancer is grade specific in both ethnic groups.
Recent studies on non-small cell lung cancer (50); ovarian cancer (51) and pancreatic cancer (52) provide further support for studying IGF family member SNPs as predictors for cancer risk and prognosis (49). In the case of pancreatic cancer, Suzuki et al (52) the IGF-1 −177 GC/CC genotype was related to lower frequency of diabetes in controls and higher frequency of diabetes in cases among the genotype carriers. We plan to examine this SNP variant in our patient population.
The two IGF-IR polymorphisms assessed in our study population have only been examined in the context of cancer by only one other study (25). The authors examined the influence of the IGF-IR polymorphism on IGF-1 levels and prostate cancer risk. The study found that having a polymorphic genotype composed of fewer than two copies of the (AGG)7 polymorphisms (i.e. non(AGG)7/(AGG)7 or non(AGG)7/non(AGG)7) was associated with an increased risk of prostate cancer, and interestingly, lower levels of IGF-1. The association with lower IGF-1 levels and increased prostate risk found in the study is an unexpected finding since increased IGF-1 levels have been associated with increased risk of multiple cancers in large meta-analysis studies (40). We identified a modest trend towards significance of the (AGG)5 genotype and (AGG)5/(AGG)7 genotype with an increased OR with breast cancer. The association with the <(AGG)7 repeats in the aforementioned prostate study and our finding of the (AGG)5 repeat association with breast cancer both link this IGF-IR polymorphic genotype with cancer. In our study, however, upon adjusting for age and BMI, the significance of this association was lost. These results warrant careful interpretation since this allele was found in low frequency in our study cohort and may present a chance finding. Furthermore, the functional relevance of this polymorphism is still unclear; therefore, understanding the contribution of the polymorphism to gene expression and ultimately cancer is required before this polymorphism may be useful as a risk-associated biomarker.
Conclusions
Our study confirms that the IGF-1 (CA) repeat is associated with breast cancer, and this polymorphism may play a functional role in the transcription of IGF-1 and influence plasma or circulating IGF-1 levels. This study provides further support for a link between IGF-1 serum levels (6) and breast cancer in pre-menopausal minority women, and that this cohort of minority women are more likely to express non19/non19 IGF-1 genotypes. Additionally, our study has found a significant association of the IGF-IR (AGG)5 polymorphic genotype and breast cancer. These findings modestly suggest that polymorphisms in the IGF axis (ranging from the ligand to receptor) may be potentially important cancer risk-association biomarkers. Additional studies will analyze combined polymorphic panels along the IGF pathway which will assess whether combinations of risk-associated IGF polymorphisms may reveal stronger predictive associations with cancer.
Further consideration into the ethnic-specific differences, if any, requires additional investigation and expansion of the numbers of women. More studies are necessary to elucidate the underlying mechanisms which can explain the associations found between the IGF gene polymorphisms and breast cancer.
Acknowledgements
We would like to thank all of the amazing women who are participating in the ongoing breast cancer studies conducted by the Division of Cancer Research and Training at Charles Drew University. We could not have done this without their generous cooperation. This work was supported in part by grants from NIH/National Cancer Institute U54CA14393; U56 CA101599; CA15083-25S3; R25DK067015-01; and Department of Defense Breast Cancer Research Program grant BC043180 to J.V. Vadgama.
List of Abbreviations
- IGF-1
insulin-like growth factor 1
- IGFBP-3
insulin-like growth factor binding protein 3
- IGF-IR
insulin-like growth factor 1 receptor
- OR
odds ratio
- CRC
Colorectal Cancer
- HNPCC
Hereditary Non-Polyposis Colon Cancer
- CI
confidence interval
- BMI
body mass index
- PCR
polymerase chain reaction
- SNP
single nucleotide polymorphism
- RFLP
restriction fragment length polymorphism
- BI-RADS
breast imaging reporting and data system
Footnotes
Competing Interests
The authors declare that they have no competing interests.
References
- 1.LeRoith D, Raizada MK, editors. Adv Exp Med Biol; Woods Hole, Massachusetts. April 20–23, 1993; Proceedings of the 4th International Symposium on Insulin, IGFs, and their Receptors; 1993. pp. 1–417. [Google Scholar]
- 2.Jones JI, Clemmons DR. Insulin-like growth factors and their binding proteins: biological actions. Endocr Rev. 1995;16:3–34. doi: 10.1210/edrv-16-1-3. [DOI] [PubMed] [Google Scholar]
- 3.Macaulay VM. Insulin-like growth factors and cancer. Br J. Cancer. 1992;65:311–320. doi: 10.1038/bjc.1992.65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Karey KP, Sirbasku DA. Differential responsiveness of human breast cancer cell lines MCF-7 and T47D to growth factors and 17 beta-estradiol. Cancer Res. 1988;48:4083–4092. [PubMed] [Google Scholar]
- 5.Pollak M, Costantino J, Polychronakos C, et al. Effect of tamoxifen on serum insulin like growth factor I levels in stage I breast cancer patients. J. Natl Cancer Inst. 1990;82:1693–1697. doi: 10.1093/jnci/82.21.1693. [DOI] [PubMed] [Google Scholar]
- 6.Vadgama JV, Wu Y, Datta G, Khan H, Chillar R. Plasma insulin-like growth factor-I and serum IGF-binding protein 3 can be associated with the progression of breast cancer, and predict the risk of recurrence and the probability of survival in African American and Hispanic women. Oncology. 1999;57(4):330–340. doi: 10.1159/000012052. [DOI] [PubMed] [Google Scholar]
- 7.Hankinson SE, Willett WC, Colditz GA, et al. Circulating concentrations of insulin-like growth factor-I and risk of breast cancer. Lancet. 1998;351:1393–1396. doi: 10.1016/S0140-6736(97)10384-1. [DOI] [PubMed] [Google Scholar]
- 8.Rinaldi S, Peeters PHM, Berrino F, et al. IGF-1, IGFBP-3 and breast cancer risk in women: The European Prospective Investigation into Cancer and Nutrition (EPIC) Endocrine-Related Cancer. 2006;13(2):593–605. doi: 10.1677/erc.1.01150. [DOI] [PubMed] [Google Scholar]
- 9.Baglietto L, English DR, Hopper JL, Morris HA, Tilley WD, Giles GG. Circulating insulin-like growth factor-I and binding protein-3 and the risk of breast cancer. Cancer Epidemiol Biomarkers Prev. 2007;16(4):763–768. doi: 10.1158/1055-9965.EPI-06-0960. [DOI] [PubMed] [Google Scholar]
- 10.Toniolo P, Bruning PF, Akhmedkhanov A, et al. Serum insulin-like growth factor-I and breast cancer Int. J. Cancer. 2000;88:828–832. doi: 10.1002/1097-0215(20001201)88:5<828::aid-ijc22>3.0.co;2-8. [DOI] [PubMed] [Google Scholar]
- 11.Byrne C, Colditz GA, Willett WC, Speizer FE, Pollak M, Hankinson SE. Plasma insulin-like growth factor (IGF) I, IGF-binding protein 3, and mammographic density. Cancer Res. 2000;60(14):3744–3748. [PubMed] [Google Scholar]
- 12.Boyd NF, Stone J, Martin LJ, et al. The association of breast mitogens with mammographic densities. Br J Cancer. 2002;87(8):876–882. doi: 10.1038/sj.bjc.6600537. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Maskarinec G, Williams AE, Kaaks R. A cross-sectional investigation of breast density and insulin-like growth factor I. Int J Cancer. 2003;107(6):991–996. doi: 10.1002/ijc.11505. [DOI] [PubMed] [Google Scholar]
- 14.Harrela M, Koistinen H, Kaprio J, et al. Genetic and environmental components of interindividual variation in circulating levels of IGF-1, IGF-1I, IGFBP-1, and IGFBP-3. J Clin Invest. 1996;98(11):2612–2615. doi: 10.1172/JCI119081. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Rosen CJ, Kurland ES, Vereault D, et al. Association between serum insulin growth factor-I (IGF-1) and a simple sequence repeat in IGF-1 gene: implications for genetic studies of bone mineral density. J Clin Endocrinol Metab. 1998;83:2286–2290. doi: 10.1210/jcem.83.7.4964. [DOI] [PubMed] [Google Scholar]
- 16.Li YC, Korol AB, Fahima T, Nevo E. Microsatellites within genes: structure, function, and evolution. Mol Biol Evol. 2004;21(6):991–1007. doi: 10.1093/molbev/msh073. [DOI] [PubMed] [Google Scholar]
- 17.Yu H, Li BD, Smith M, Shi R, Berkel HJ, Kato I. Polymorphic CA repeats in the IGF-1 gene and breast cancer. Breast Cancer Res Treat. 2001;70:117–122. doi: 10.1023/a:1012947027213. [DOI] [PubMed] [Google Scholar]
- 18.Chen X, Guan J, Song Y, et al. IGF-1 (CA) repeat polymorphisms and risk of cancer: a meta-analysis. J Hum Genet. 2008;53:227–238. doi: 10.1007/s10038-007-0241-3. [DOI] [PubMed] [Google Scholar]
- 19.Canzian F, McKay JD, Cleveland RJ, et al. Polymorphisms of genes coding for insulin-like growth factor 1 and its major binding proteins, circulating levels of IGF-1 and IGFBP-3 and breast cancer risk: results from the EPIC study. Br J Cancer. 2006;94:299–307. doi: 10.1038/sj.bjc.6602936. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Jernstrom H, Deal C, Wilkin F, et al. Genetic and nongenetic factors associated with variation of plasma levels of insulin-like growth factor-I and insulin-like growth factor-binding protein-3 in healthy premenopausal women. Cancer Epidemiol Biomarkers Prev. 2001;10:377–384. [PubMed] [Google Scholar]
- 21.Larsson O, Girnita A, Girnita L. Role of insulin-like growth factor 1 receptor signaling in cancer. Br J Cancer. 2005;92:2097–2101. doi: 10.1038/sj.bjc.6602627. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.LeRoith D, Roberts CT., Jr The insulin-like growth factor system and cancer. Cancer Lett. 2003;195(2):127–137. doi: 10.1016/s0304-3835(03)00159-9. [DOI] [PubMed] [Google Scholar]
- 23.Almeida A, Muleris M, Dutrillaux B, Malfoy B. The insulin-like growth factor I receptor gene is the target for the 15q26 amplicon in breast cancer. Genes Chromosomes Cancer. 1994;11:63–65. doi: 10.1002/gcc.2870110110. [DOI] [PubMed] [Google Scholar]
- 24.Xie Y, Skytting B, Nilsson G, Brodin B, Larsson O. Expression of insulin-like growth factor-1 receptor in synovial sarcoma: association with an aggressive phenotype. Cancer Res. 1999;59:3588–3591. [PubMed] [Google Scholar]
- 25.Chen C, Freeman R, Voigt LF, Fitzpatrick A, Plymate SR, Weiss NS. Prostate cancer risk in relation to selected genetic polymorphisms in insulin-like growth factor-I, insulin-like growth factor binding protein-3, and insulin-like growth factor-I receptor. Cancer Epidemiol Biomarkers Prev. 2006;15(12):2461–2466. doi: 10.1158/1055-9965.EPI-06-0541. [DOI] [PubMed] [Google Scholar]
- 26.Figer A, Karasik YP, Baruch RG, et al. Insulin-like growth factor I polymorphism and breast cancer risk in Jewish women. Isr Med Assoc J. 2002;4:759–762. [PubMed] [Google Scholar]
- 27.Yu H, Jin F, Shu XO, et al. Insulin-like growth factors and breast cancer risk in Chinese women. Cancer Epidemiol Biomarkers Prev. 2002;11:705–712. [PubMed] [Google Scholar]
- 28.Morimoto LM, Newcomb PA, White E, Bigler J, Potter JD. Variation in Plasma Insulin-Like Growth Factor-1 and Insulin-Like Growth Factor Binding Protein-3: Genetic Factors. Cancer Epidemiol Biomarkers Prev. 2005;14(6):1394–1401. doi: 10.1158/1055-9965.EPI-04-0694. [DOI] [PubMed] [Google Scholar]
- 29.Wagner K, Hemminki K, Israelsson E, et al. Polymorphisms in the IGF-1 and IGFBP-3 promoter and the risk of breast cancer. Breast Cancer Res Treat. 2005;92:133–140. doi: 10.1007/s10549-005-2417-x. [DOI] [PubMed] [Google Scholar]
- 30.Wen W, Gao YT, Shu XO, et al. Insulin-like growth factor-I gene polymorphism and breast cancer risk in Chinese women. Int. J. Cancer. 2005;113:307–311. doi: 10.1002/ijc.20571. [DOI] [PubMed] [Google Scholar]
- 31.Cleveland RJ, Gammon MD, Edmiston SN, et al. IGF-1 CA repeat polymorphisms, lifestyle factors and breast cancer risk in the Long Island Breast Cancer Study Project. Carcinogenesis. 2006;27:758–765. doi: 10.1093/carcin/bgi294. [DOI] [PubMed] [Google Scholar]
- 32.Bågeman E, Ingvar C, Rose C, Jernström H. Absence of the common Insulin-like growth factor-1 19-repeat allele is associated with early age at breast cancer diagnosis in multiparous women. Br J Cancer. 2007;96(5):712–717. doi: 10.1038/sj.bjc.6603632. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Patel AV, Cheng I, Canzian F. et al. IGF-1, IGFBP-1, and IGFBP-3 polymorphisms predict circulating IGF levels but not breast cancer risk: findings from the Breast and Prostate Cancer Cohort Consortium (BPC3).Breast and Prostate Cancer Cohort Consortium. PLoS ONE. 2008;3(7):e2578. doi: 10.1371/journal.pone.0002578. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Missmer SA, Haiman CA, Hunter DJ, et al. A sequence repeat in the insulin-like growth factor-1 gene and risk of breast cancer. Int. J. Cancer. 2002;100:332–336. doi: 10.1002/ijc.10473. [DOI] [PubMed] [Google Scholar]
- 35.Kato J, Eastham B, Li M, Smith, Yu H. Genotype–phenotype analysis for the polymorphic CA repeat in the insulin-like growth factor-I (IGF-1) gene. European Journal of Epidemiology. 2003;18:203–209. doi: 10.1023/A:1023379100539. [DOI] [PubMed] [Google Scholar]
- 36.DeLellis K, Ingles S, Kolonel L, et al. IGF-1 genotype, mean plasma level and breast cancer risk in the Hawaii/Los Angeles multiethnic cohort. Br J Cancer. 2003;88:277–282. doi: 10.1038/sj.bjc.6600728. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Gonza´lez-Zuloeta Ladd AM, Liu F, Houben MP, et al. IGF-1 CA repeat variant and breast cancer risk in postmenopausal women. Eur J Cancer. 2007;43:1718–1722. doi: 10.1016/j.ejca.2007.04.026. [DOI] [PubMed] [Google Scholar]
- 38.Fehringer G, Ozcelik H, Knight JA, Paterson AD, Boyd NF. Association between IGF1 CA microsatellites and mammographic density, anthropometric measures, and circulating IGF-I levels in premenopausal Caucasian women. Breast Cancer Res Treat. 2009;116(2):413–423. doi: 10.1007/s10549-008-0146-7. [DOI] [PubMed] [Google Scholar]
- 39.Fehringer G, Boyd NF, Knight JA, et al. Family-based genetic association study of insulin-like growth factor I microsatellite markers and premenopausal breast cancer risk. Breast Cancer Res Treat. 2009;118(2):415–424. doi: 10.1007/s10549-009-0336-y. [DOI] [PubMed] [Google Scholar]
- 40.Renehan AG, Zwahlen M, Minder C, O'Dwyer ST, Shalet SM, Egger M. Insulin- like growth factor (IGF)-I, IGF binding protein-3, and cancer risk: systematic review and metaregression analysis. Lancet. 2004;63(9418):1346–1353. doi: 10.1016/S0140-6736(04)16044-3. [DOI] [PubMed] [Google Scholar]
- 41.Jernström H, Chu W, Vesprini D, et al. Genetic factors related to racial variation in plasma levels of insulin-like growth factor-1: implications for premenopausal breast cancer risk. Mol Genet Metab. 2001;72(2):144–154. doi: 10.1006/mgme.2000.3130. [DOI] [PubMed] [Google Scholar]
- 42.Cheng I, Penney Kathryn L, Stram DO, et al. Haplotype-Based Association Studies of IGFBP1 and IGFBP3 with Prostate and Breast Cancer Risk: The Multiethnic Cohort. Cancer Epidemiol Biomarkers Prev. 2006;15(10):1993–1997. doi: 10.1158/1055-9965.EPI-06-0361. [DOI] [PubMed] [Google Scholar]
- 43.Al-Zahrani A, Sandhu MS, Luben RN, et al. IGF-1 and IGFBP3 tagging polymorphisms are associated with circulating levels of IGF-1, IGFBP3 and risk of breast cancer. Hum Mol Genet. 2006;15:1–10. doi: 10.1093/hmg/ddi398. [DOI] [PubMed] [Google Scholar]
- 44.Hernandez W, Grenade C, Santos ER, Bonilla C, Ahaghotu C, Kittles RA. IGF-1 and IGFBP-3 gene variants influence on serum levels and prostate cancer risk in African Americans. Carcinogenesis. 2007;28(10):2154–2159. doi: 10.1093/carcin/bgm190. [DOI] [PubMed] [Google Scholar]
- 45.Slattery ML, Sweeney C, Wolff R, et al. Genetic variation in IGF-1, IGFBP3, IRS1, IRS2 and risk of breast cancer in women living in Southwestern United States. Breast Cancer Res Treat. 2007;104(2):197–209. doi: 10.1007/s10549-006-9403-9. [DOI] [PubMed] [Google Scholar]
- 46.Sweeney C, Murtaugh MA, Baumgartner KB, et al. Insulin-like growth factor pathway polymorphisms associated with body size in Hispanic and non-Hispanic white women. Cancer Epidemiol Biomarkers Prev. 2005;14(7):1802–1809. doi: 10.1158/1055-9965.EPI-05-0149. [DOI] [PubMed] [Google Scholar]
- 47.Zecevic M, Amos CI, Gu X, et al. IGF-1 Gene Polymorphism and Risk for Hereditary Nonpolyposis Colorectal Cancer. J Natl Cancer Inst. 2006;98:139–143. doi: 10.1093/jnci/djj016. [DOI] [PubMed] [Google Scholar]
- 48.Reeves SG, Rich D, Meldrum CJ, et al. IGF-1 is a modifier of disease risk in hereditary non-polyposis colorectal cancer. Int. J. Cancer. 2008;123:1339–1343. doi: 10.1002/ijc.23668. [DOI] [PubMed] [Google Scholar]
- 49.Hoyo C, Grubber J, Demark-Wahnefried W, et al. Grade-Specific Prostate Cancer Associations of IGF-1 (CA)19 Repeats and IGFBP3-202A/C in Blacks and Whites. JNMA. 2007;99(7):718–722. [PMC free article] [PubMed] [Google Scholar]
- 50.Zhang M, Hu Z, Huang J, et al. A 3′-Untranslated Region Polymorphism in IGF-1 Predicts Survival of Non–Small Cell Lung Cancer in a Chinese Population. Clin Cancer Res. 2010;6(4):1236–1244. doi: 10.1158/1078-0432.CCR-09-2719. [DOI] [PubMed] [Google Scholar]
- 51.Terry K, Tworoger SS, Gates MA, et al. Common genetic variant in IGF-1, IGFBP1 and IGFBP3 and ovarian cancer. Carcinogenesis. 2009;30(12):2042–2046. doi: 10.1093/carcin/bgp257. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Suzuki H, Li Y, Dong X, Hassan MM, et al. Effect of Insulin-Like Growth Factor Gene Polymorphisms Alone or In Interaction with Diabetes on the Risk of Pancreatic Cancer. Cancer Epidemiol Biomarkers Prev. 2008;17(12):3467–3473. doi: 10.1158/1055-9965.EPI-08-0514. [DOI] [PMC free article] [PubMed] [Google Scholar]