Lipids and cardiovascular/metabolic health

Administration of omega-3 fatty acids and Raloxifene to women at high risk of breast cancer: interim feasibility and biomarkers analysis from a clinical trial



The antiestrogen, Raloxifene (Ral) is an effective breast cancer chemopreventive agent. Omega-3 fatty acids (n-3FA) may inhibit mammary carcinogenesis. On the basis of their mechanisms of action, we test the hypothesis that a combination of n-3FA and Ral may be superior in reducing select biomarkers of breast cancer risk in women.


Postmenopausal women at increased risk for breast cancer (breast density 25%) were randomized to: (1) no intervention; (2) Ral 60 mg; (3) Ral 30 mg; (4) n-3FA (Lovaza) 4 g and (5) Lovaza 4 g+Ral 30 mg for 2 years. Reduction in breast density is the primary end point of the study. We report preliminary data on feasibility, compliance and changes in secondary end points related to IGF-I signaling, estrogen metabolism, oxidative stress and inflammation in the first group of 46 women who completed 1 year of the study.


All interventions were well tolerated with excellent compliance (96±1% overall) by pill count and also supported by the expected rise in both serum n-3FA and n-3FA/Omega-6 fatty acids (n-6FA) ratio in women randomized to groups 4 and 5 (P<0.05). Lovaza decreased serum triglycerides and increased high-density lipoprotein (HDL) cholesterol compared with control (P<0.05 for both). Ral reduced serum IGF-1 in a dose-dependent manner (P<0.05) while Lovaza did not. Lovaza had no effect on IGF-1 or IGFBP-3. None of the other biomarkers were affected by our treatment.


The combination of Lovaza and Ral is a feasible strategy that may be recommended in future breast cancer chemoprevention trials.


We are testing the hypothesis that a combination of antiestrogens and Omega-3 fatty acids (n-3FA) exerts a synergistic breast cancer chemopreventive effect due to the cross-talk of their downstream cellular effects that leads to decreased proliferation and increased apoptosis of premalignant mammary cells.1 While basic mechanisms are under investigation using preclinical models of mammary carcinogenesis,2, 3, 4 we are testing concomitantly the clinical relevance of this approach in postmenopausal women using surrogate markers of breast cancer development (NCT00723398). While the primary end point of our clinical trial is a reduction in breast density, a well-established risk factor for breast cancer,5 we are also examining the effects of the individual and combined administration of Ral and n-3FA on several biomarkers thought to be potentially involved in mammary carcinogenesis. These include markers of oxidative stress, estrogen metabolism, inflammation and insulin-like growth factor (IGF-1) signaling. In this report, we provide preliminary results of the treatment effects on these secondary end points in the first subset of 46 women who completed 1 year of the study. We also performed a detailed plasma fatty acid analysis in these same subjects to document adherence to their group assignments.

Subjects and methods


Healthy, postmenopausal women between the ages of 35–75 years were recruited into the study if their breast density was 25% at their yearly routine screening mammogram. Postmenopausal status was defined as a history of at least 12 months without spontaneous menstrual bleeding or having a documented hysterectomy with bilateral salpingo-oophorectomy. Additional eligibility criteria included no hormone replacement therapy (including oral contraceptives in younger women) for at least 6 months before entry into the study and being smoke free for >5 years. Ineligibility criteria included history of pulmonary embolism or deep vein thrombosis, stroke and atherosclerotic heart disease, presence of any known hypercoagulable state, either congenital (for example, protein S deficiency) or acquired (for example, corticosteroid treatment), diabetes mellitus, uncontrolled hypertension (blood pressure 140/90), previous history of breast cancer (including ductal carcinoma in situ and lobular carcinoma in situ), other prior malignancies except for satisfactorily treated basal cell or squamous cell skin cancer, in-situ cervical cancer, or any cancer for which the patient has been disease free for at least 5 years. Women were also excluded if they had a history of allergy to fish, were unwilling not to use n-3FA preparations outside of the protocol or were drinking more than one alcoholic beverage per day.

Trial design

The study was conducted with the approval of the Institutional Review Board of the Penn State College of Medicine. After signing the informed consent, each study participant was randomly assigned with equal probability to one of the following five treatment groups: group 1—control, no intervention; group 2—Ral 60 mg orally daily; group 3—Ral 30 mg orally daily; group 4—Lovaza 4 g per day orally with meals; group 5—Lovaza 4 g per day orally with meals plus Ral 30 mg orally daily. A blocked randomization scheme was used to ensure balanced treatment allocation during the course of enrolment. Upon entry, information was collected on parity, family history of breast cancer and personal history of benign breast disease. In addition, anthropometric measurements including weight, height, and waist-to-hip ratio were obtained. At study entry and follow-up visits, blood samples were obtained for a complete blood count, lipid panel and fatty acid analysis. Both blood and urine samples were collected for biomarker analysis.

Assessment of dietary habits

Study participants completed a modified version of the National Cancer Institute Diet History Questionnaire at baseline and at follow-up.6 The modified version includes the additions of items to capture consumption patterns of Pennsylvania residents.7 Participants reported their usual intake and portion size of 137 separate food items; 49 of which included embedded queries. Interviewers reviewed the returned Diet History Questionnaires in-person with study participants. Completed questionnaires were scanned and Diet*Calc version 14.3 (National Cancer Institute, Division of Cancer Control and Population Sciences, Risk Factor Monitoring and Methods Branch; the nutrient analysis program developed by the National Cancer Institute for this instrument and reconfigured for our modified questionnaires, was used to estimate energy and nutrient intakes.

Assessment of physical activity

The International Physical Activity Questionnaire was used to estimate energy expenditure due to physical activity at baseline and follow-up. It is publically available (, which has been validated and a scoring protocol is provided.8 The International Physical Activity Questionnaire asks about walking, moderate and vigorous-intensity physical activity in four domains (leisure-, domestic/yard-, work- and transport-related activities). Total physical activity is calculated as a continuous score expressed as metabolic equivalent task-min per week, which we used to estimate per day metabolic equivalent task-min (metabolic equivalent task level × minutes of activity × events per week)/7. A complete list of references for International Physical Activity Questionnaire is provided at

Fatty acid analysis

The fatty acid composition of the serum samples was determined at baseline and every 6 months thereafter as published by us.2, 3, 4 This measurement complemented pill counts to verify compliance.

Serum and urine marker analysis

Markers of oxidative stress and lipid peroxidation

Urinary 8-(isoprostane)-F2α was measured in urine samples using a competitive enzyme-linked immunosorbent assay (ELISA) from Cayman Biochemical (Cat. No. 516351; Ann Arbor, MI, USA). Before analysis, urine was purified by solid phase extraction using an ODS cartridge system (Cayman Biochemical) according to manufacturer’s instructions. Urinary 8-hydroxy-deoxyguansine was measured in urine samples using a competitive ELISA from Cayman Biochemical (Cat. No. 589320). Both 8-isoprostane and 8--hydroxy-deoxyguansine values were expressed on a per mg creatinine basis to account for urine dilution. Creatinine was determined by reaction with picrate as described previously.9

The urine metabolites of estrogen; 2-hydroxy estrone (2OHE1) and 16-α-hydroxy estrone (16αOHE1) were determined in spot first morning urine collections by enzyme immunoassay using reagents obtained from Immunacare Corporation (Blue Bell, PA, USA). Results were normalized to urine creatinine. An Estrogen Metabolite Index was calculated as 2OHE1/16αOHE1. Assay imprecision was <10% for both 2OHE1 and 16αOHE1 while assay sensitivity was 0.05 ng/ml for 2OHE1 and 0.15 ng/ml for 16αOHE1. The reference range for 2OHE1 for postmenopausal females is 1–20 ng/ml and 0.6–10 ng/ml for 16αOHE1. The average Estrogen Metabolite Index for postmenopausal women is 2.0.

Markers of inflammation: high sensitivity C-reactive protein and IL-6

High Sensitivity C-Reactive Protein (CRP) levels in serum were determined with a sandwich-based ELISA (ALPCO, Salem, NH, USA) that employs a specific capture antibody directed against a specific antigenic determinant on the CRP peptide. A rabbit antibody directed against another antigenic site on CRP conjugated to horseradish peroxidase was used as the signal antibody to form a sandwich of the CRP molecule. The ELISA color reaction was developed using tetramethylbenzidine that reacts with horseradish peroxidase and absorbance was measured spectrophotometrically at 450 nm. Method sensitivity is 1.0 ng/ml and within run and between run imprecision averaged 6% at a concentration of 9.0 ng/ml. High Sensitivity Human IL-6 was determined with an ELISA (R&D Systems, Minneapolis, MN, USA) that uses an E. coli-expressed recombinant human IL-6 standard in a sandwich-based enzyme immunoassay format. A monoclonal antibody specific for IL-6 is precoated on microtiter plates as capture antibody and an alkaline phosphatase-anti-IL-6 antibody conjugate provides the assay signal. The ELISA color reaction is developed using an NAD/NADH reaction system in an amplification reaction cascade where a tetrazolium salt is reduced by NADH to produce an intensely colored formazan dye. Method sensitivity is 0.5 pg/ml and within and between run imprecision averaged 7% at a concentration of 2.5 pg/ml.

Markers of IGF-I signaling: IGF-1 and insulin like growth factor binding protein (IGFBP)-3

IGF-1 was determined directly using a microtiter plate and sandwich-based ELISA with reagents obtained from ALPCO. The procedure employs an IGFBP-blocked ELISA in which IGF-I is disassociated from the IGFBPs, with an acidic buffer. The IGF-I antiserum is then dissolved in a buffer, which is able to neutralize the acidic samples. After the IGF-I antibody solution has neutralized the samples, the excess IGF-II occupies the IGF-binding sites of the binding proteins, thus allowing the measurement of the resulting free IGF-I. With this method, the IGFBPs are not removed, but their function and therefore their possible interference in the assay is neutralized. Due to the use of an IGF-1 antibody that has extremely low cross-reactivity with IGF-II, IGF-II does not hinder the interaction of the first antibody with IGF-I. With the use of a second specific anti-IGF-I-antibody that is biotinylated, a Streptavidin-Peroxidase-Enzyme conjugate binds to the second antibody to form a color complex proportional to the concentration of IGF-1. Analytical sensitivity for the assay is 0.1 ng/ml and within and between run imprecision averaged 7% at a concentration of 140 ng/ml.

IGFBP-3 was determined with a solid phase microtiter plate ELISA (R&D Systems) in a sandwich format immunoassay. Microtiter wells coated with a monoclonal antibody against IGFBP-3 capture the molecule and after incubation with a second anti-IGFBP-3 polyclonal antibody labeled with horseradish peroxidase, color development is obtained by the addition of the substrates hydrogen peroxide and tetramethylbenzidine that is read spectrophotometrically at a wavelength of 450 nm. Analytical sensitivity is 0.05 ng/ml and within and between run imprecision averaged 5% at a concentration of 12 ng/ml.

Statistical analysis

Summary statistics are provided for the demographic data shown in Supplementary Table 1 and for dietary and physical activity characteristics of study participants detailed in Table 1. The difference of these variables across the five treatment groups was tested using the analysis of variance or Fisher’s exact test as appropriate. The dependence of fatty acid profiles at month 3 and at month 6 on Ral, Lovaza, time and respective baseline values was analyzed using mixed effects regression model (Supplemental Table 2; Figure 1). For serum lipids data (Table 2), appropriate transformation (usually log) was first performed on the variable. The effect of Lovaza, Ral and their combination on the difference of these transformed variables between baseline and month 12 was then studied using linear regression. A similar analysis was performed to study the effects of Lovaza, Ral and their combination on biomarkers of estrogen metabolism, inflammation, oxidative stress and IGF-1 signaling as seen in Table 3 and Figure 2.

Table 1 Baseline and follow-up daily dietary and physical activity characteristics of study participants (n=46; mean values±s.e.)a
Figure 1

(ac) Treatment effects on serum n-3FA (a), n-6FA (b) and their ratio n-3FA/n-6FA (c). The number of subjects was 8 in the control, 11 in the Lovaza 4 g group and 8 in the Lovaza 4 g+Ral 30 mg group. Data represent mean values±s.e.m. *P<0.05 vs the other groups.

Table 2 Lipid panels at baseline and at 12 months
Table 3 Effects of Raloxifene and/or Lovaza on biomarkers of estrogen metabolism, inflammation and oxidative stress
Figure 2

(ac) Percent change in serum IGF-1 (a), IGFBP-3 (b) and IGF-1/IGFBP-3 (c) between baseline and at 12 months in the five groups. The number of subjects was 8 in the control group, 10 in Ral 30 mg group, 9 in Ral 60 mg group, 11 in Lovaza 4 g group and 8 in the combination group. Data represent means with the s.e. show in parenthesis. *P<0.02 vs control; +P<0.05 vs Ral 30 mg.


Demographic characteristics

Because of the demographics of the catchment area (rural Pennsylvania), the majority of subjects were Caucasian with only 1 out of 46 subjects African American. There were no significant differences across the five groups in age, body mass index (BMI), waist-to-hip ratio or history of prior breast biopsy. There was a difference across the five groups with respect to family history of breast cancer (Supplementary Table 1).


Compliance was monitored by pill counts by the nurse coordinator at the study visits and further verified by monitoring serum fatty acid analysis (see below). The average compliance for the four groups at 12 months was 96±1% (s.e.). No significant differences in compliance were found across groups. The average compliance for group 2 was 93±3% (s.e.) at 6 months and 92±3% (s.e.) at 12 months, for group 3 was 97±1% (s.e.) at 6 months and 94±2% (s.e.) at 12 months, for group 4 was 94±2% (s.e.) at 6 months and 97±2% (s.e.) at 12 months. For the combination group (group 5), compliance at 6 months was 97±3% (s.e.) for Ral and 92±2% (s.e.) for Lovaza. Compliance at 12 months in this group was 100±2% (s.e.) for Ral and 96±2% (s.e.) for Lovaza. Only four subjects had a compliance <85%, one in group 2 (68%), two in group 4 (84% and 81%) and one in group 5 (81%).

Adverse events

There were very few adverse effects. They included hot flashes in one patient in group 3 at 12 months, vaginal dryness in one subject in group 2 at 12 months, and leg cramps in one subject from group 2 and one from group 3 at 12 months. No severe adverse events were reported.

Diet and physical activity

Data on diet and physical activity at baseline and 12 months are summarized in Table 1. At baseline and follow-up, there were no statistically significant differences in total physical activity or dietary intake of total energy, macronutrients or fatty acids of interest across the five treatment groups (overall difference P-values all>0.05). In addition, changes in these variables over the follow-up period were not statistically significant for the group overall and in the individual groups except for the combined Ral/Lovaza group which reported significantly lower mean energy intake at follow-up compared with baseline. Of interest, both n-3FA and Omega-6 fatty acids (n-6FA) intake were reduced in this group with no change in the n-3FA/n-6FA ratio. However, there was no statistically significant change in BMI at 12 months in this group as well as any of the other groups.

Treatment effects on serum fatty acids

Plasma fatty acid analysis revealed that the subjects in the two groups receiving Lovaza had a statistically significant increase in n-3FA levels compared with the other groups at 6 and 12 months of intervention (P<0.05; Figure 1a). Conversely, serum n-6FA decreased in the same groups (Figure 1b), which resulted in a significant increase in the n-3FA to n-6FA ratio in the groups receiving Lovaza (Figure 1c). The effects of our interventions on the specific fatty acid composition of the serum are reported in detail in Supplementary Table 2. Specific Lovaza effects were observed for arachidonic acid, eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). Both EPA and DHA were increased in the Lovaza groups by 3-fold whereas levels of arachidonic acid were decreased 25–35% in these same groups. Ral had no effects on n-3FA and only small and inconsistent effects on two minor fatty acids, C16:2N4 and C20:1N9.

We then tested whether the baseline levels of these fatty acids as well as the BMI and waist-to-hip ratio influenced the changes observed in EPA and DHA levels after 1 year of Lovaza treatment. We found an inverse correlation between baseline EPA levels and the increase in this fatty acid at 12 months (P<0.02). There was no effect of baseline DHA on the increase in DHA levels observed at 1 year. There was a tendency for subjects with higher BMIs to exhibit a smaller increase in EPA levels at 12 months but this association did not reach statistical significance (P=0.06). Baseline BMI did not influence the change in DHA at 12 months. Waist-to-hip ratio did not affect the change in either EPA or DHA at 1 year.

Treatment effects on serum lipids

As can be seen in Table 2, the fasting lipid panels were essentially normal and not statistically different across groups. After 12 months of intervention, the triglycerides were significantly reduced in both groups receiving Lovaza (by 15% in the group receiving Lovaza alone and 22% in the combination group). High-density lipoprotein (HDL) levels were significantly elevated (by 7% in the group receiving Lovaza alone and up to 14% in the group receiving a combination of Lovaza and Ral). None of the treatments affected either total or high-density lipoprotein cholesterol.

Treatment effects on serum and urine biomarkers

As can be seen in Figure 2a, the administration of Ral for 1 year reduced the circulating level of IGF-I in a dose-dependent manner. While the effects of 30 mg of Ral were not statistically significant (P<0.1), the effects of the 60-mg dose were highly significant (P<0.002) and were superior to that of the lower dose of Ral (P<0.05). Lovaza did not affect serum IGF-I levels in the absence or the presence of Ral. None of the treatments affected the circulating levels of IGFBP-3 (Figure 2b). Consequently, the IGF-I/IGFBP-3 ratio (an index of IGF-I bioactivity) was significantly reduced by Ral in a dose-dependent manner (P=0.018; Figure 2c). None of the other biomarkers was significantly affected by any of the interventions (Table 3).


Preventive measures represent the best opportunity for reducing breast cancer incidence and mortality. Although antiestrogens like Tamoxifen and Ral have been found to be effective chemopreventive agents, they are ineffective in preventing the development of estrogen receptor-negative tumors and their use is associated with rare but potentially life-threatening adverse events such as a thromboembolism.10, 11 Consequently, these agents have limited acceptability in healthy subjects including women at high risk of developing breast cancer.12, 13 Therefore, safer and more effective chemopreventive strategies need to be developed to have a significant impact on the development of breast cancer. Since breast cancer requires the coordinated activation of multiple cellular mechanisms, a multi-targeted approach is needed to prevent the development of mammary carcinogenesis. We have been interested in testing the combination of antiestrogens with n-3FA, an attractive class of compounds that could potentiate the beneficial effects of antiestrogens by inhibiting multiple cellular pathways potentially involved in mammary carcinogenesis (for review, see Signori et al.1). We believe that a major attractive feature of this approach is its safety since it allows us to combine a lower and hence less toxic dose of antiestrogens with n-3FA, compounds that are known to have inherent health benefits (that is, reduction in cardiovascular risk) beyond their potential chemopreventive benefit in breast cancer.14

In parallel with our preclinical studies conducted in experimental models of mammary carcinogenesis,2, 3, 4 we have recently initiated a randomized clinical trial in postmenopausal women (NCT00723398) testing the combined effects of Lovaza and Ral at half its conventional dose (that is, 30 vs 60 mg) in reducing breast density, a well-established risk factor for breast cancer. The effects of this treatment combination will be compared with that of Ral alone at its conventional dose of 60 mg daily. The object of this report is to provide preliminary data on the feasibility and compliance with this combination approach as well as to describe the effects of our interventions on a number of potential biomarkers of mammary carcinogenesis in the first group of 46 women who have completed 1 year on the study. We show that compliance is very high as indicated by pill counts and plasma fatty acid analysis. As expected, we observed a significant increase in both serum n-3FA and the n-3FA/n-6FA ratio in the groups of women receiving Lovaza. This increase was present at 6 months and persisted at 12 months of treatment. In contrast, no change in serum fatty acid composition was observed in the control group and in the groups receiving Ral alone. These results indicate excellent compliance of our subjects with their treatment assignments for a period up to 1 year. The increase in serum n-3FA and the n-3FA/n-6FA ratio was comparable to that reported by Yee et al.,15 in their dose-response study with Lovaza administered to high-risk women (premenopausal and postmenopausal) for a 6-month duration. Since our study included a control group, we were able to exclude the possibility of ‘contamination’ (that is, women taking n-3FA supplements out of protocol). This is an important issue since n-3FA preparations are readily available to the general public and are appealing in view of their advertised health benefits. Yee et al.15 observed that a higher serum baseline of DHA and BMI was significantly associated with a lower incremental rise in serum DHA and EPA after Lovaza administration. Although we did not observe this, we did find an inverse relationship between baseline EPA levels and the increase in serum EPA levels after 1 year. It is likely that rather than true differences, these apparently discrepant results may be due to the relatively small number of subjects in both studies.

We observed a significant reduction in serum triglycerides in women receiving Lovaza (Table 2). This finding has been previously reported, although primarily in individuals with hypertriglyceridemia.16, 17 Our subjects, on the other hand, had normal serum triglyceride levels. We also observed a significant increase in HDL cholesterol following Lovaza administration (8% in women receiving Lovaza alone and 16% in those receiving the combination of Lovaza and Ral). This is at variance with other reports in the literature where Lovaza was reported to either cause no change16 or a minimal increase (1–3%) in HDL cholesterol.17

IGF-I is a potent mitogen for breast cancer and has been postulated to be involved in mammary carcinogenesis, tumor progression and metastasis.18, 19 Elevated levels of circulating IGF-I have been shown to be a significant risk factor for premenopausal breast cancer.20 In agreement with previous reports in the literature,21, 22 Ral administration reduced circulating IGF-I levels and the IGF-I/IGFBP-3 ratio, an index of IGF-I bioavailability. Lovaza administration, on the other hand, did not influence IGF-I levels as also observed by Yee et al.15 Whether increased oxidative stress, inflammation and preferential estrogen metabolism toward the pro-carcinogenic 16-α-hydroxylated metabolite have a role in breast cancer development remains uncertain at this time in view of conflicting reports in the literature.23, 24, 25 With regard to the role of oxidative stress and inflammation, a major issue is the appropriate selection of the most relevant biomarkers. Therefore, the lack of an effect of our interventions on select biomarkers of inflammation and oxidative stress does not necessarily rule out a role of cellular stress and inflammation on breast cancer development or an antitumor action of our combined treatments. It will obviously be important to evaluate the long-term effects of our interventions on the primary end point of this trial, that is, breast density, for which the study has been powered.


  1. 1

    Signori C, El-Bayoumy K, Russo J, Thompson HJ, Richie JP, Hartman TJ et al. Chemoprevention of breast cancer by fish oil in preclinical models: trials and tribulations. Cancer Res 2011; 71: 6091–6096.

  2. 2

    Manni A, Xu H, Washington S, Aliaga C, Cooper T, Richie JP et al. The impact of fish oil on the chemopreventive efficacy of tamoxifen against development of N-methyl-N-nitrosourea-induced rat mammary carcinogenesis. Cancer Prev Res 2010; 3: 322–330.

  3. 3

    Manni A, Richie JP, Xu H, Washington S, Aliaga C, Cooper TK et al. Effects of fish oil and Tamoxifen on preneoplastic lesion development and biomarkers of oxidative stress in the early stages of N-methyl-N-nitrosourea-induced rat mammary carcinogenesis. Int J Oncol 2011; 39: 1153–1164.

  4. 4

    Manni A, Xu H, Washington S, Aliaga C, Das A, Cooper T et al. The effects of Tamoxifen and fish oil on mammary carcinogenesis in polyoma middle T transgenic mice. Horm Cancer 2011; 2: 249–259.

  5. 5

    Boyd NF, Martin LJ, Rommens JM, Paterson AD, Minkin S, Yaffe MJ et al. Mammographic density: a heritable risk factor for breast cancer. Methods Mol Biol 2009; 472: 343–360.

  6. 6

    Subar AF, Thompson FE, Kipnis V, Midthune D, Hurwitz P, McNutt S et al. Comparative validation of the Block, Willett, and National Cancer Institute food frequency questionnaires: the Eating at America's Table Study. Am J Epidemiol 2001; 154: 1089–1099.

  7. 7

    Miller PE, Lazarus P, Lesko SM, Muscat JE, Harper G, Cross AJ et al. Diet index-based and empirically derived dietary patterns are associated with colorectal cancer risk. J Nutr 2010; 140: 1267–1273.

  8. 8

    Craig CL, Marshall AL, Sjostrom M, Bauman AE, Booth ML, Ainsworth BE et al. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc 2003; 35: 1381–1395.

  9. 9

    Slot C . Plasma creatinine determination. A new and specific Jaffe reaction method. Scand J Clin Lab Invest 1965; 17: 381–387.

  10. 10

    Fisher B, Costantino JP, Wickerham DL, Redmond CK, Kavanah M, Cronin WM et al. Tamoxifen for prevention of breast cancer: report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study. J Natl Cancer Inst 1998; 90: 1371–1388.

  11. 11

    Vogel VG, Costantino JP, Wickerham DL, Cronin WM, Cecchini RS, Atkins JN et al. Effects of tamoxifen vs raloxifene on the risk of developing invasive breast cancer and other disease outcomes: the NSABP Study of Tamoxifen and Raloxifene (STAR) P-2 trial. JAMA 2006; 295: 2727–2741.

  12. 12

    Ropka ME, Keim J, Philbrick JT . Patient decisions about breast cancer chemoprevention: a systematic review and meta-analysis. J Clin Oncol 2010; 28: 3090–3095.

  13. 13

    Waters EA, Cronin KA, Graubard BI, Han PK, Freedman AN . Prevalence of tamoxifen use for breast cancer chemoprevention among U.S. women. Cancer Epidemiol Biomarkers Prev 2010; 19: 443–446.

  14. 14

    Ruxton CH, Reed SC, Simpson MJ, Millington KJ . The health benefits of omega-3 polyunsaturated fatty acids: a review of the evidence. J Hum Nutr Diet 2007; 20: 275–285.

  15. 15

    Yee LD, Lester JL, Cole RM, Richardson JR, Hsu JC, Li Y et al. Omega-3 fatty acid supplements in women at high risk of breast cancer have dose-dependent effects on breast adipose tissue fatty acid composition. Am J Clin Nutr 2010; 91: 1185–1194.

  16. 16

    Harris WS . n-3 fatty acids and serum lipoproteins: human studies. Am J Clin Nutr 1997; 65 (5 Suppl), 1645S–1654SS.

  17. 17

    Skulas-Ray AC, Kris-Etherton PM, Harris WS, Vanden Heuvel JP, Wagner PR, West SG . Dose-response effects of omega-3 fatty acids on triglycerides, inflammation, and endothelial function in healthy persons with moderate hypertriglyceridemia. Am J Clin Nutr 2011; 93: 243–252.

  18. 18

    Lee AV . YD. Role of the IGF System. In Breast Cancer Proliferation and Progression. Manni A (ed)., Humana Press, Inc.: Totowa, NJ, 1999.

  19. 19

    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 meta-regression analysis. Lancet 2004; 363: 1346–1353.

  20. 20

    Hankinson SE, Willett WC, Colditz GA, Hunter DJ, Michaud DS, Deroo B et al. Circulating concentrations of insulin-like growth factor-I and risk of breast cancer. Lancet 1998; 351: 1393–1396.

  21. 21

    Torrisi R, Baglietto L, Johansson H, Veronesi G, Bonanni B, Guerrieri-Gonzaga A et al. Effect of raloxifene on IGF-I and IGFBP-3 in postmenopausal women with breast cancer. Br J Cancer 2001; 85: 1838–1841.

  22. 22

    Oleksik AM, Duong T, Pliester N, Asma G, Popp-Snijders C, Lips P . Effects of the selective estrogen receptor modulator, raloxifene, on the somatotropic axis and insulin-glucose homeostasis. J Clin Endocrinol Metab 2001; 86: 2763–2768.

  23. 23

    Eliassen AH, Missmer SA, Tworoger SS, Hankinson SE . Circulating 2-hydroxy- and 16alpha-hydroxy estrone levels and risk of breast cancer among postmenopausal women. Cancer Epidemiol Biomarkers Prev 2008; 17: 2029–2035.

  24. 24

    Osborne MPKR, Hershcopf RJ, Bradlow HL, Kourides IA, Williams WR, Rosen PP et al. Omega-3 fatty acids: modulation of estrogen metabolism and potential for breast cancer prevention. Cancer Investigation 1988; 6: 629–631.

  25. 25

    Rossner P, Gammon MD, Terry MB, Agrawal M, Zhang FF, Teitelbaum SL et al. Relationship between urinary 15-F2t-isoprostane and 8-oxodeoxyguanosine levels and breast cancer risk. Cancer Epidemiol Biomarkers Prev 2006; 15: 639–644.

Download references


We thank Glaxo Smith Kline and Eli-Lilly for their generous supply of Lovaza and Raloxifene, respectively. This work was supported by Susan G Komen for the Cure Grant no. KG081632.

Author information

Correspondence to A Manni.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies the paper on European Journal of Clinical Nutrition website

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Signori, C., DuBrock, C., Richie, J. et al. Administration of omega-3 fatty acids and Raloxifene to women at high risk of breast cancer: interim feasibility and biomarkers analysis from a clinical trial. Eur J Clin Nutr 66, 878–884 (2012).

Download citation


  • Raloxifene
  • Lovaza
  • breast cancer prevention
  • serum and urine biomarkers
  • serum fatty acid composition

Further reading