Expression of immune-related genes and breast cancer recurrence in women with ductal carcinoma in situ

BACKGROUND AND AIM

Ductal carcinoma in situ (DCIS) is a non-obligate precursor of invasive breast cancer with highly variable clinical behavior, but risk stratification is still challenging. We sought to identify immune-related gene expression signatures of pure DCIS associated with different risks of breast cancer recurrence.

METHODS

A retrospective nested case-control study of 143 pure DCIS was performed including 70 women with subsequent ipsilateral breast event (IBE, in situ or invasive; cases) and 73 DCIS women with no IBE and matched for age, tumor size, treatment, hormone receptors/HER2 status, and follow-up time (controls). RNA was extracted from DCIS samples and subjected to next-generation sequencing gene expression analysis of 395 immune-related genes. Correlations between DCIS immune-related gene expression and IBE were analyzed using weighted Cox regression for nested case-control data.

RESULTS

Eight immune-related genes were differentially expressed between cases and controls. MAGEA10 expression (present vs. absent) and high expression levels of IFNA17 and CBLB (Q4 vs. Q1) were observed more frequently in DCIS of women with subsequent IBE, mainly invasive (p-valueFDR < 0.05). Conversely, expression of IL3RA1, TAGAP, TNFAIP8, and high expression levels of CCL2 and LRP1 were associated with a lower risk of IBE (p-valueFDR < 0.05).

CONCLUSION

This exploratory analysis of pure DCIS showed significant differences in immune-related gene expression profiles between women with and with no subsequent IBE, particularly as invasive IBE. These results, after additional validation, could improve risk stratification and management of DCIS patients.

Effect of metformin and lifestyle intervention on adipokines and hormones in breast cancer survivors: a pooled analysis from two randomized controlled trials

PURPOSE

We investigated the effect of metformin and lifestyle intervention on metabolic, inflammatory, and steroid biomarkers of breast cancer (BC) recurrence risk in two intervention trials among BC survivors with overweight or obesity.

METHODS

Baseline and follow-up serum samples collected during the two trials were analyzed and data pooled. The USA trial (Reach for Health) included postmenopausal BC survivors (n = 333) randomly assigned to 6-month metformin vs placebo and lifestyle intervention (LSI) vs control (2 × 2 factorial design). The Italian trial (MetBreCS) included BC survivors (n = 40) randomized to 12-month metformin vs placebo. Insulin resistance (HOMA-IR), adipokines, cytokines, and steroids were measured.

RESULTS

Metformin compared to placebo showed a favorable decrease in leptin (- 8.8 vs - 3.5 ng/mL; p < 0.01) and HOMA-IR (- 0.48 vs - 0.25; p = 0.03), and an increase in SHBG (2.80 vs 1.45 nmol/L; p < 0.01). Excluding women taking aromatase inhibitors, metformin (n = 84) compared to placebo (n = 99) decreased estradiol (- 4 vs 0 pmol/L; p < 0.01), estrone (- 8 vs 2 pmol/L; p < 0.01) and testosterone (- 0.1 vs 0 nmol/L-; p = 0.02). LSI favorably affected adiponectin (0.45 vs - 0.06 ug/mL; p < 0.01), leptin (- 10.5 vs - 4.4 ng/mL; p < 0.01), HOMA-IR (- 0.6 vs 0.2; p = 0.03), and SHBG (2.7 vs 1.1 nMol/L; p = 0.04) compared to controls. The strongest impact was observed combining metformin with LSI on adipokines, CRP, SHBG, and estrogens.

CONCLUSIONS

Supportive healthy lifestyle programs combined with metformin to achieve maximal risk reduction among BC cancer survivors are recommended, especially for those with obesity in menopause.

Prognostic Value of PCSK9 Levels in Premenopausal Women at Risk of Breast Cancer-Evidence from a 17-Year Follow-Up Study

BACKGROUND AND AIM

The involvement of cholesterol in cancer development remains a topic of debate, and its association with breast cancer has yet to be consistently demonstrated. Considering that circulating cholesterol levels depend on several concomitant processes, we tested the liability of plasma levels of proprotein convertase subtilisin/kexin type 9 (PCSK9), one of the key regulators of cholesterol levels, as a prognostic biomarker in the context of breast neoplastic events.

METHODS

Within a prospective randomized breast cancer prevention trial we measured baseline plasma levels of PCSK9. A total of 235 at-risk premenopausal women were randomized and followed up for 17 years. Participants enrolled in this placebo-controlled, phase II, double-blind trial were randomly assigned to receive either tamoxifen 5 mg/d or fenretinide 200 mg/d, both agents, or placebo for 2 years. The associations with breast cancer events were evaluated through competing risk and Cox regression survival models, adjusted for randomization strata (5-year Gail risk ≥ 1.3% vs. intraepithelial neoplasia or small invasive breast cancer of favorable prognosis), age, and treatment allocation. PCSK9 associations with biomarkers linked to breast cancer risk were assessed on blood samples collected at baseline.

RESULTS

The plasmatic PCSK9 median and interquartile range were 207 ng/mL and 170-252 ng/mL, respectively. Over a median follow-up period of 17 years and 89 breast neoplastic events, disease-free survival curves showed a hazard ratio of 1.002 (95% CI: 0.999-1.005, p = 0.22) for women with PCSK9 plasma levels ≥ 207 ng/mL compared to women with levels below 207 ng/mL. No differences between randomization strata were observed. We found a negative correlation between PCSK9 and estradiol (r = -0.305), maintained even after partial adjustment for BMI and age (r = -0.287). Cholesterol (r = 0.266), LDL-C (r = 0.207), non-HDL-C (r = 0.246), remnant cholesterol (r = 0.233), and triglycerides (r = 0.233) also correlated with PCSK9.

CONCLUSIONS

In premenopausal women at risk of early-stage breast cancer, PCSK9 did not appear to have a role as a prognostic biomarker of breast neoplastic events. Larger studies are warranted investigating patients in different settings.

CDH1 Genotype Exploration in Women With Hereditary Lobular Breast Cancer Phenotype

Importance

Pathogenic or likely pathogenic (P/LP) germline CDH1 variants are associated with risk for diffuse gastric cancer and lobular breast cancer (LBC) in the so-called hereditary diffuse gastric cancer (HDGC) syndrome. However, in some circumstances, LBC can be the first manifestation of this syndrome in the absence of diffuse gastric cancer manifestation.

Objectives

To evaluate the frequency of germline CDH1 variants in women with the hereditary LBC (HLBC) phenotype, somatic CDH1 gene inactivation in germline CDH1 variant carriers' tumor samples, and the association of genetic profiles with clinical-pathological data and survival.

Design, Setting, and Participants

This single-center, longitudinal, prospective cohort study was conducted from January 1, 1997, to December 31, 2021, with follow-up until January 31, 2023. Women with LBC seen at the European Institute of Oncology were included. Testing for germline CDH1, BRCA1, and BRCA2 genes was performed. Somatic profiling was assessed for germline CDH1 carriers.

Main Outcomes and Measures

Accurate estimates of prevalence of germline CDH1 variants among patients with HLBC and the association of somatic sequence alteration with HLBC syndrome. The Kaplan-Meier method and a multivariable Cox proportional hazards regression model were applied for overall and disease-free survival analysis.

Results

Of 5429 cases of primary LBC, familial LBC phenotype accounted for 1867 (34.4%). A total of 394 women with LBC were tested, among whom 15 germline CDH1 variants in 15 unrelated families were identified. Among these variants, 6 (40.0%) were P/LP, with an overall frequency of 1.5% (6 of 394). Of the 6 probands with P/LP CDH1 LBC, 5 (83.3%) had a positive family history of BC and only 1 (16.7%) had sporadic juvenile early-onset LBC. No germline BRCA1 and BRCA2 variants were identified in CDH1 carriers. An inactivating CDH1 mechanism (second hit) was identified in 4 of 6 explored matched tumor samples (66.7%) in P/LP germline carriers. The P/LP CDH1 LBC variant carriers had a significantly lower age at diagnosis compared with the group carrying CDH1 variants of unknown significance or likely benign (42.5 [IQR, 38.3-43.0] vs 51.0 [IQR, 45.0-53.0] years; P = .03).

Conclusions and Relevance

In this cohort study, P/LP germline CDH1 variants were identified in individuals not fulfilling the classic clinical criteria for HDGC screening, suggesting that identification of these variants may provide a novel method to test women with LBC with early age at diagnosis and/or positive family history of BC.

Fenretinide in young women at genetic or familial risk of breast cancer: A placebo-controlled biomarker trial

Fenretinide, a retinoid with a low toxicity profile that accumulates in the breast, has been shown to prevent second breast cancer in young women. Fenretinide exhibits apoptotic and anti-invasive properties and it improves insulin sensitivity in overweight premenopausal women with insulin resistance. The present study aimed to further characterize its role in cancer prevention by measuring circulating biomarkers related to insulin sensitivity and breast cancer risk. Sixty-two women, aged 20 to 46 years, healthy or who had already undergone breast cancer surgery, with a known BRCA1/2 mutation or a likelihood of mutation ≥ 20% according to the BRCAPRO model, were randomly assigned to receive fenretinide (200 mg/day) or placebo for 5 years (trial registration: EudraCT Number: 2009-010260-41). Fasting blood samples were drawn at baseline, 12 and 36 months, and the following biomarkers were analyzed: retinol, leptin, adiponectin, retinol-binding protein 4, total cholesterol, HDL and LDL cholesterol, triglycerides, glucose, insulin, IGF-I, IGFBP-3, SHBG, testosterone, and VEGF. After 12 months of treatment, we observed a favorable effect of fenretinide on glucose (decrease; P=0.005), insulin (decrease; P=0.03), HOMA index (decrease; P=0.004), HDL cholesterol (increase; P=0.002), even though these effects were less prominent after 36 months. Retinol and retinol-binding protein 4 markedly decreased (P<0.0001) throughout the study. None of the other measured biomarkers changed.

Alternative dosing regimen of exemestane in a randomized presurgical trial: the role of obesity in biomarker modulation

In a 3-arm presurgical trial, four-six weeks exemestane 25 mg three times/week (TIW) was non-inferior to 25 mg/day (QD) in suppressing circulating estradiol in postmenopausal women with ER-positive breast cancer. Since obesity may decrease exemestane efficacy, we analyzed changes in sex steroids, adipokines, Ki-67, and drug levels in relation to obesity. Postmenopausal women with early-stage ER-positive breast cancer were randomized to either exemestane 25 mg QD (n = 57), 25 mg TIW (n = 57), or 25 mg/week (QW, n = 62) for 4-6 weeks before breast surgery. Serum and tissue pre- and post-treatment biomarkers were stratified by body mass index (BMI)< or ≥30 kg/m2. Post-treatment median exemestane and 17-OH exemestane levels were 5-6 times higher in the QD arm compared to the TIW arm. For obese women, TIW maintained comparable reductions to QD in systemic estradiol levels, although the reduction in estrone was less with the TIW regimen. There was less suppression of SHBG with the TIW versus the QD dose schedule in obese women which should result in less systemic bioavailable estrogens. Metabolically, the effect of the TIW regimen was similar to the QD regimen for obese women in terms of leptin suppression and increase in the adiponectin-leptin ratio. Reduction in tissue Ki-67 was less for obese women on the TIW regimen than QD, although changes were similar for non-obese women. Our findings suggest that TIW exemestane should be explored further for primary cancer prevention in both normal weight and obese cohorts.

Inflammatory and Metabolic Biomarker Assessment in a Randomized Presurgical Trial of Curcumin and Anthocyanin Supplements in Patients with Colorectal Adenomas

Colorectal cancer prevention is crucial for public health, given its high mortality rates, particularly in young adults. The early detection and treatment of precancerous lesions is key to preventing carcinogenesis progression. Natural compounds like curcumin and anthocyanins show promise in impeding adenomatous polyp progression in preclinical models. We conducted a randomized, double-blind, placebo-controlled, phase II presurgical trial in 35 patients with adenomatous polyps to explore the biological effects of curcumin and anthocyanins on circulating biomarkers of inflammation and metabolism. No significant difference in biomarker changes by treatment arm was observed. However, the network analysis before treatment revealed inverse correlations between adiponectin and BMI and glycemia, as well as direct links between inflammatory biomarkers and leptin and BMI. In addition, a considerable inverse relationship between adiponectin and grade of dysplasia was detected after treatment (corr = -0.45). Finally, a significant increase in IL-6 at the end of treatment in subjects with high-grade dysplasia was also observed (p = 0.02). The combined treatment of anthocyanins and curcumin did not result in the direct modulation of circulating biomarkers of inflammation and metabolism, but revealed a complex modulation of inflammatory and metabolic biomarkers of colon carcinogenesis.

Vitamin D Supplementation and Adherence to World Cancer Research Fund (WCRF) Diet Recommendations for Colorectal Cancer Prevention: A Nested Prospective Cohort Study of a Phase II Randomized Trial

Vitamin D and a healthy diet, based on World Cancer Research Fund (WCRF) recommendations, are considered key elements for colorectal cancer (CRC) prevention. In a CRC case-control study, we observed that CRC cases were often significantly Vitamin D deficient while subjects following WCRF recommendations significantly decreased their risk of developing CRC. We conducted a randomized phase-II trial (EudraCT number-2015-000467-14) where 74 CRC patients showed differences in response to Vitamin D supplementation, 2000 IU in average per day, according to gender and microbiota. The aim of this nested study is to correlate Vitamin D (supplementation, serum level and receptor polymorphisms), circulating biomarkers, and events (polyp/adenoma, CRC relapse and other cancers) in concomitant to WCRF recommendation adherence. Vitamin D supplementation did not modulate circulating biomarkers or follow-up events. FokI and TaqI VDR were associated with 25-hydroxyvitamin D (25OHD) levels. Patients following the WCRF recommendations had significantly lower leptin, significantly lower IL-6 (only in females), and significantly lower risk of events (HR = 0.41, 95%CI: 0.18-0.92; p = 0.03; median follow-up 2.6 years). Interestingly, no WCRF adherents had significantly more events if they were in the placebo (p < 0.0001), whereas no influence of WCRF was observed in the Vitamin D arm. While one-year Vitamin D supplementation might be too short to show significant preventive activity, a healthy diet and lifestyle should be the first step for preventive programs.

Skeletonema marinoi Extracts and Associated Carotenoid Fucoxanthin Downregulate Pro-Angiogenic Mediators on Prostate Cancer and Endothelial Cells

The exploration of natural preventive molecules for nutraceutical and pharmaceutical use has recently increased. In this scenario, marine microorganisms represent an underestimated source of bioactive products endowed with beneficial effects on health that include anti-oxidant, anti-inflammatory, differentiating, anti-tumor, and anti-angiogenic activities. Here, we tested the potential chemopreventive and anti-angiogenic activities of an extract from the marine coastal diatom Skeletonema marinoi Sarno and Zingone (Sm) on prostate cancer (PCa) and endothelial cells. We also tested one of the main carotenoids of the diatom, the xanthophyll pigment fucoxanthin (Fuco). Fuco from the literature is a potential candidate compound involved in chemopreventive activities. Sm extract and Fuco were able to inhibit PCa cell growth and hinder vascular network formation of endothelial cells. The reduced number of cells was partially due to growth inhibition and apoptosis. We studied the molecular targets by qPCR and membrane antibody arrays. Angiogenesis and inflammation molecules were modulated. In particular, Fuco downregulated the expression of Angiopoietin 2, CXCL5, TGFβ, IL6, STAT3, MMP1, TIMP1 and TIMP2 in both prostate and endothelial cells. Our study confirmed microalgae-derived drugs as potentially relevant sources of novel nutraceuticals, providing candidates for potential dietary or dietary supplement intervention in cancer prevention approaches.

Colorectal cancer, Vitamin D and microbiota: A double-blind Phase II randomized trial (ColoViD) in colorectal cancer patients

BACKGROUND

Several studies suggest a role of gut microbiota in colorectal cancer (CRC) initiation and progression. Vitamin D (vitD) blood levels are also inversely correlated with CRC risk and prognosis. However, these factors' interplay remains unknown.

METHODS

74 CRC patients after standard treatment were randomized to 1-year 2000 IU/day vitD or placebo.  Baseline and post-treatment fecal microbiota for shotgun metagenomics sequencing was collected. Coda-lasso and Principal Component Analysis were used to select and summarize treatment-associated taxa and pathways. Associations between vitD and taxa/pathways were investigated with logistic regression. Mediation analysis was performed to study if treatment-associated taxa mediated the effect of supplementation on 25(OH)D levels. Cox proportional-hazards model was used for disease-free survival (DFS).

RESULTS

60 patients were analyzed. Change in alpha diversity (Shannon: p = 0.77; Simpson: p = 0.63) and post-treatment beta diversity (p = 0.70) were comparable between arms. Post-treatment abundances of 63 taxa and 32 pathways differed between arms. The 63 taxa also mediated the effect of supplementation on 25(OH)D (p = 0.02). There were sex differences in vitD levels, microbiota and pathways. Pathways of essential amino acids' biosynthesis were more abundant in supplemented women. Fusobacterium nucleatum presence at baseline was associated with worse DFS (p = 0.02). Those achieving vitD sufficiency (25(OH)D≥30 ng/ml) had lower post-treatment abundances (p = 0.05). Women were more likely to have F. nucleatum post-treatment (p = 0.02).

CONCLUSIONS

VitD supplementation may contribute shaping the gut microbiota and the microbiota may partially mediate the effect of supplementation on 25(OH)D. The observed sex-specific differences highlight the necessity of including sex/gender as a variable in microbiome studies.

A Novel Automated Immunoassay Platform to Evaluate the Association of Adiponectin and Leptin Levels with Breast Cancer Risk

Simple Summary

Adiponectin and leptin are adipokines secreted by the adipose tissue that have been associated with several chronic diseases including cancer. We compared two methods for their measurement and investigated their association with breast cancer. We measured adiponectin and leptin with the automated ELLA platform and a manual commercially available enzyme-linked immunosorbent assay (ELISA) kit on serum samples of women enrolled in two international breast cancer prevention trials. We found a good concordance between the two methods and our results support the association of low adiponectin levels with breast cancer, irrespective of the method used. The take-home message is that ELLA is a very robust platform that represents a step forward for the future use of adipokines, along with other biomarkers, in clinical cancer risk assessment and prevention. Its use should be taken into account whenever biomarkers should be measured in a large cohort of patients for clinical validation or cancer association studies.

Abstract

Adiponectin and leptin are adipokines secreted by the adipose tissue that are associated with several chronic diseases including cancer. We aimed to compare the immunoassay platform ELLA with an enzyme-linked immunosorbent assay (ELISA) kit and to assess whether the results of the association analyses with breast cancer risk were dependent on the assay used. We measured adiponectin and leptin with ELLA and ELISA on baseline serum samples of 116 Italian postmenopausal women enrolled in two international breast cancer prevention trials. Results were compared with Deming, Passing–Bablok regression and Bland–Altman plots. Disease-free survival was analyzed with the Cox model. There was a good correlation between the methods for adiponectin and leptin (r > 0.96). We found an increased breast cancer risk for very low adiponectin levels (HR for ELLA = 3.75; 95% CI: 1.37;10.25, p = 0.01), whereas no significant association was found for leptin levels. The disease-free survival curves were almost identical for values obtained with the two methods, for both biomarkers. The ELLA platform showed a good concordance with ELISA for adiponectin and leptin measurements. Our results support the association of very low adiponectin levels with postmenopausal breast cancer risk, irrespective of the method used. The ELLA platform is a time-saving system with high reproducibility, therefore we recommend its use for biomarker assessment.

Vitamin D Supplementation and Disease-Free Survival in Stage II Melanoma: A Randomized Placebo Controlled Trial

Patients with newly resected stage II melanoma (n = 104) were randomized to receive adjuvant vitamin D3 (100,000 IU every 50 days) or placebo for 3 years to investigate vitamin D3 protective effects on developing a recurrent disease. Median age at diagnosis was 50 years, and 43% of the patients were female. Median serum 25-hydroxy vitamin D (25OHD) level at baseline was 18 ng/mL, interquartile range (IQ) was 13–24 ng/mL, and 80% of the patients had insufficient vitamin D levels. We observed pronounced increases in 25OHD levels after 4 months in the active arm (median 32.9 ng/mL; IQ range 25.9–38.4) against placebo (median 19.05 ng/mL; IQ range 13.0–25.9), constantly rising during treatment. Remarkably, patients with low Breslow score (<3 mm) had a double increase in 25OHD levels from baseline, whereas patients with Breslow score ≥3 mm had a significantly lower increase over time. After 12 months, subjects with low 25OHD levels and Breslow score ≥3 mm had shorter disease-free survival (p = 0.02) compared to those with Breslow score <3 mm and/or high levels of 25OHD. Adjusting for age and treatment arm, the hazard ratio for relapse was 4.81 (95% CI: 1.44–16.09, p = 0.011). Despite the evidence of a role of 25OHD in melanoma prognosis, larger trials with vitamin D supplementation involving subjects with melanoma are needed.

A Meta-Analysis of Obesity and Risk of Colorectal Cancer in Patients with Lynch Syndrome: The Impact of Sex and Genetics

There appears to be a sex-specific association between obesity and colorectal neoplasia in patients with Lynch Syndrome (LS). We meta-analyzed studies reporting on obesity and colorectal cancer (CRC) risk in LS patients to test whether obese subjects were at increased risk of cancer compared to those of normal weight. We explored also a possible sex-specific relationship between adiposity and CRC risk among patients with LS. The summary relative risk (SRR) and 95% confidence intervals (CI) were calculated through random effect models. We investigated the causes of between-study heterogeneity and assessed the presence of publication bias. We were able to retrieve suitable data from four independent studies. We found a twofold risk of CRC in obese men compared to nonobese men (SRR = 2.09; 95%CI: 1.23–3.55, I² = 33%), and no indication of publication bias (p = 0.13). No significantly increased risk due to obesity was found for women. A 49% increased CRC risk for obesity was found for subjects with an MLH1 mutation (SRR = 1.49; 95%CI: 1.11–1.99, I² = 0%). These results confirm the different effects of sex on obesity and CRC risk and also support the public measures to reduce overweight in people with LS, particularly for men.

Association of Vitamin D Receptor and Vitamin D-Binding Protein Polymorphisms with Familial Breast Cancer Prognosis in a Mono-Institutional Cohort

Low 25-hydroxyvitamin D (25OHD) has been associated with an increased cancer incidence and poorer prognosis. Single nucleotide polymorphisms (SNPs) of vitamin D receptor (VDR) and vitamin D binding protein (GC gene) may interfere with vitamin D activity. This study assesses the role of VDR and GC SNPs on breast cancer (BC) recurrence and survival in a cohort of patients with a family history of breast cancer, without the pathogenic variant for BRCA1 and BRCA2. A consecutive series of patients who underwent genetic testing were genotyped for VDR and GC genes. Specifically, ApaI, FokI, TaqI, BsmI and rs2282679, rs4588, rs7041 SNPs were determined. A total of 368 wild type (WT) patients with BC were analyzed for VDR and GC SNPs. The GC rs2282679 minor allele was significantly associated with luminal subtype of the primary tumor compared to Her2+/TN breast cancer (p = 0.007). Multivariate Cox models showed that BmsI and TaqI are significantly associated with BC outcome. Patients with the major alleles showed more than 30% lower hazard of relapse (BsmI p = 0.02 and TaqI p = 0.03). Our study supports the evidence for a pivotal role of 25OHD metabolism in BC. GC SNPs may influence the hormone tumor responsiveness and VDR may affect tumor prognosis.

Association of CYP2D6 genotype and tamoxifen metabolites with breast cancer recurrence in a low-dose trial

Low-dose tamoxifen halves recurrence in non-invasive breast cancer without significant adverse events. Some adjuvant trials with tamoxifen 20 mg/day had shown an association between low endoxifen levels (9–16 nM) and recurrence, but no association with CYP2D6 was shown in the NSABP P1 and P2 prevention trials. We studied the association of CYP2D6 genotype and tamoxifen metabolites with tumor biomarkers and recurrence in a randomized phase III trial of low-dose tamoxifen. Median (IQR) endoxifen levels at year 1 were 8.4 (5.3–11.4) in patients who recurred vs 7.5 (5.1–10.2) in those who did not recur (p = 0.60). Tamoxifen and metabolites significantly decreased C-reactive protein (CRP, p < 0.05), and a CRP increase after 3 years was associated with higher risk of recurrence (HR = 4.37, 95% CI, 1.14–16.73, P = 0.03). In conclusion, endoxifen is below 9 nM in most subjects treated with 5 mg/day despite strong efficacy and there is no association with recurrence, suggesting that the reason for tamoxifen failure is not poor drug metabolism. Trial registration: ClinicalTrials.gov, Identifier: NCT01357772.

Abstract 1705: Identification of an immune gene expression signature for progression of breast ductal carcinoma in situ (dcis)

Introduction. DCIS is a potential precursor of invasive breast cancer (IBC). Immune escape might play a critical role in the transition from DCIS to IBC. We aim to identify an immune-related gene expression signature of DCIS to distinguish hazardous lesions, and to avoid the burden of overtreatment in women with less harmful DCIS. Since tumor-infiltrating lymphocytes (TILs) occur more frequently in hormone receptor-negative, Her-2 positive DCIS, we focused the analysis on this subtype.

Methods. We performed a prospective nested case-control study involving women with pure DCIS (hormone receptor-negative, Her-2 positive, TILs ≥ 20%), treated with breast conservative surgery with or without radiotherapy. Cases were defined as women with DCIS developing subsequent ipsilateral breast event (IBE), controls as women with DCIS without IBE at 8 years of follow-up. Next-generation sequencing gene expression assay (Oncomine™ Immune Response Research Assay) targeting 395 genes associated with tumor-immune systems interactions was performed on RNA extracted from DCIS samples of 25 cases and 25 age- and radiotherapy-matched controls. In order to obtain reliable results, we considered adequate samples with mapped reads &gt; 1 million and valid reads &gt; 800.000. Normalized gene-level count data generated from the run were further processed with Affymetrix™. A Volcano plot was also employed in order to distinguish genes differently expressed between cases and controls according to the log-gene-level fold change and the ANOVA test (p values &lt; 0.05). Univariate (Log-rank) and multivariate Cox proportional hazard models identified genes associated with the disease free survival.

Results. Thirty-one samples (15 cases; 16 controls) reached the quality parameters. The analysis identified 11 genes with a significant different expression between cases and controls. The upregulation of three immune-related genes (VTCN1, NT5E and CMKLR1) remained significantly associated with disease free survival. Notably, VTCN1 and NT5E expressions were significantly associated with the risk of invasive local recurrence.

Conclusions. This exploratory analysis of hormone receptor-negative, Her-2 positive DCIS with TILs ≥ 20% showed significant differences in immune-related gene expression profiles between women with subsequent IBE and controls. VTCN1 and NT5E may play an important role in immune surveillance mechanisms of DCIS and may be targeted by immune-based therapy to slow or prevent DCIS progression. These results are well worth further validation on future studies.

Citation Format: Matteo Lazzeroni, Caternia Fumagalli, Federica Corso, Alberto Ranghiero, Debora Macis, Elena Guerini Rocco, Valentina Aristarco, Harriet Johansson, Aliana Guerrieri-Gonzaga, Davide Serrano, Andrea DeCensi, Sara Gandini, Massimo Barberis, Bernardo Bonanni. Identification of an immune gene expression signature for progression of breast ductal carcinoma in situ (dcis) [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 1705.

Development of an HPLC-MS/MS Method for the Determination of Silybin in Human Plasma, Urine and Breast Tissue

Silybin is a flavonolignan extracted from Silybum marianum with chemopreventive activity against various cancers, including breast. This study was designed to develop an HPLC-MS/MS method for the determination of silybin in human plasma, urine and breast tissue in early breast cancer patients undergoing Siliphos^® supplementation, an oral silybin-phosphatidylcholine complex. The determination of silybin was carried out by liquid-liquid extraction with methyl-tert-butyl ether (MTBE); total silybin concentration was determined by treating the samples with β-glucuronidase, while for the determination of free silybin, the hydrolytic step was omitted. Naringenin and naproxen were selected as internal standards. The detection of the analyte was carried out by mass spectrometry and by chromatography. The HPLC-MS/MS method was evaluated in terms of selectivity, linearity, limit of quantification, precision and accuracy, and carryover. The method proved to be selective, linear, precise and accurate for the determination of silybin. To the best of our knowledge, this presents the first analytical method with the capacity to quantify the major bioactive components of milk thistle in three different biological matrices with a lower limit of quantification of 0.5 ng/mL for plasma. Silybin phosphatidylcholine, taken orally, can deliver high blood concentrations of silybin, which selectively accumulates in breast tumor tissue.

Relationship between CYP2D6 genotype, tamoxifen metabolites, and adverse events, tumor biomarkers and breast cancer recurrence in a low-dose phase III trial in noninvasive disease

1553

Background: Low dose tamoxifen (T, 5 mg/d) given for 3 years halved recurrence in 500 women with non-invasive disease (DeCensi et al. JCO 2019). Retrospective studies with 20 mg/d have shown an association between low levels of endoxifen (9-16 nM) or Z-4OHtam (3.26 nM) and recurrence, but recent prospective studies have not confirmed these findings. We measured CYP2D6 genotype and 8 metabolite levels to determine their associations with adverse events, tumor biomarkers (IGF-I, SHBG, C-reactive protein, CRP) and breast cancer recurrence. Methods: CYP2D6 genotyping was performed in the T arm (n = 183) as previously described (Johansson H et al. BCRT 2016). T and metabolites were measured at 1 (n = 169) and 3 y (n = 152) as previously described (Helland T et al. BCR 2017). We tested linear relationships between metabolite levels and biomarkers, adjusting for age, BMI, treatment compliance and baseline biomarker levels. Cumulative incidence of recurrence according to endoxifen levels was calculated by the Cox model. Results: Endoxifen concentrations were associated to CYP2D6 metabolizer status (p < 0.001). Median (IQR) endoxifen levels were 8.4 (5.2-11.3) and 8.8 (5.8-11.5) at 1 and 3 y, with only 42% and 47% of subjects reaching 9 nM. Median endoxifen levels were related to pill count (5.5, 7.1 and 9.0 nM/L for medication possession rate < 83.3%, 83.4-99.9%, 100%, respectively). There was no difference in metabolite levels and menopausal symptoms. There was an inverse relationship between endoxifen and endometrial thickness at 3 y (p = 0.04), and between endoxifen or tamoxifen levels and IGF-I levels at 3 y (p = 0.001). T levels were positively associated with SHBG levels in postmenopausal women (p-interaction = 0.04). Endoxifen, T and 4OHtam decreased CRP, with a greater effect in premenopausal women (p-interaction = 0.02). An increase in CRP after 3 years was associated with a HR of 4.37 (95% CI, 1.14-16.73, P = 0.03) of recurrence compared to women with no increase of CRP. Median (IQR) endoxifen levels at year 1 were 8.4 (5.3-11.4) in patients who recurred vs 7.5 (5.1-10.2) in those who did not recur (p = 0.6), although this comparison was underpowered. Conclusions: T levels themselves may contribute to clinical activity by decreasing IGF-I and increasing SHBG. Elevated CRP is a predictive factor for recurrence which is down-regulated by T and metabolites. Endoxifen is below 9 nM in the majority of subjects treated with 5 mg/day, although this threshold was obtained in studies up to 20 years. Clinical trial information: NCT01357772.

Abstract PD3-08: Effects of low dose tamoxifen on circulating risk biomarkers in a phase III trial in breast pre-invasive disease

Background: We recently showed that low dose (5mg/d) tamoxifen (babytam) for 3 years can halve the incidence of new breast neoplastic events in hormone sensitive or unknown breast pre-invasive neoplasia after surgery with limited toxicity (DeCensi et al JCO 2019; 37:1629-37). Here we report the results of circulating surrogate endpoint biomarkers of breast cancer risk with special attention to the risk of ovarian stimulation observed with the full dose in premenopausal women. Methods: Five hundred women 75 years old or younger were randomized to babytam or placebo (PLA). A subgroup of 406 women consented for collection of morning fasting serum at baseline (0), 1 (1Y) and 3 years (3Y) of treatment. There was a loss of about 25% of blood sampling at 3Y. Serum IGF-I, IGFBP-3, SHBG and C-reactive Protein (CRP) were performed on all available samples. Estradiol and testosterone were determined in a subsample (n=285) to determine the extent of ovarian stimulation by babytam in premenopausal women. We used Mann-Whitney test for univariate comparisons and linear regression modeling for multivariate analyses setting changes from baseline (1Y or 3Y minus baseline values) as dependent variable and treatment arm, age, BMI and baseline biomarkers values as explanatory factors. We tested treatment*menopausal status as interaction term. COX P-H model was used to calculate hazard ratio for CRP increase. Results: At baseline, all biomarkers were evenly balanced between arms (data not shown). IGF-I decreased significantly on babytam as compared to PLA. The difference of the changes between arms was -20 ng/mL after 1Y (p&lt;0.001) and -23 after 3Y (p&lt;0.001). A treatment by menopausal status interaction was observed after 1Y (p=0.017) and 3Y (p=0.058), with a steeper decrease in postmenopausal (-25, p&lt;0.001 both after 1Y and 3Y) as compared to premenopausal women (-15, p=0.007 after 1Y and -17, p=0.009 after 3Y). IGFBP-3 increased significantly on babytam, without effect modification by menopausal status. The difference between arms was +0.26 ug/mL after 1Y (p=0.006) and +0.19 after 3Y (p=0.024). SHBG increased markedly on babytam, irrespective of menopausal status. The difference between arms from 0 to 1Y and 3Y was 24 nMol/L (p&lt;0.001). CRP levels followed a trend towards a decrease with babytam (-0.172 mg/dL, p=0.057 after 1Y and -0.091, p=0.108 after 3Y), without effect modification by menopausal status. There was no significant increase in serum estradiol after 1Y with babytam overall (+12 pg/mL, p=0.106), but a trend to an interaction with menopausal status (p=0.079): +17 (p=0.361) in premenopausal women versus +3 (p=0.284) in postmenopausal women. Similarly, there was no overall effect of babytam on testosterone after 1Y (-0.008 ng/mL, p=0.518), but a significant effect modification by menopausal status (p=0.001), showing an increase of +0.056 (p=0.006) in premenopausal women, and a decrease of -0.045 (p=0.006) in postmenopausal women. Irrespective of treatment, the increase in CRP at 3Y was significantly higher in women who experienced recurrence compared to women who did not (Mann-Whitney p=0.009). An increase in CRP was associated with a HR of 2.9 (95% CI, 1.0-8.3, p=0.05) as compared to women with a decreased or stable CRP levels. Conclusions: Babytam for 3 years exhibits a favorable effect on sex hormones and IGFs with only a slight increase of estradiol in premenopausal women which is far below that observed with 20 mg and is well compensated by a significant increase in SHBG. These findings further support the use of babytam as an effective and safe treatment for high risk individuals. ClinicalTrials.gov Identifier: NCT01357772. Supported by the Italian Ministry of Health (RFPS-2006-339898), the Italian Association for Cancer Research (IG 2008 Grant No 5611) and the Italian League against Cancer (LILT 7-08).

Citation Format: Harriet Johansson, Matteo Puntoni, Debora Macis, Valentina Aristarco, Aliana Guerrieri-Gonzaga, Davide Serrano, Silvia Caviglia, Laura Cortesi, Cristiana Taverniti, Antonio Ponti, Maria Grazia Pacquola, Marcella Gulisano, Fabio Falcini, Maria Digennaro, Anna Carriello, Katia Cagossi, Graziella Pinotti, Tania Buttiron Webber, Matteo Lazzeroni, Bernardo Bonanni, Luca Boni, Andrea DeCensi. Effects of low dose tamoxifen on circulating risk biomarkers in a phase III trial in breast pre-invasive disease [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr PD3-08.

Quality of Life in a Randomized Breast Cancer Prevention Trial of Low-Dose Tamoxifen and Fenretinide in Premenopausal Women

Menopausal symptoms are the main reason for withdrawal in tamoxifen prevention trials. Here, we present Menopause Quality of Life (MenQoL) assessment within a randomized 2 × 2 phase II clinical trial of low-dose tamoxifen and the synthetic retinoid fenretinide. A total of 235 premenopausal women at higher risk for breast cancer were randomized to either tamoxifen 5 mg daily, fenretinide 200 mg daily, their combination, or placebo. Climacteric symptoms were investigated using the MenQoL questionnaire which was self-administered at each visit for 2 years of treatment and for 1 year of follow-up. CYP2D6 was genotyped in subjects taking tamoxifen to study the association with menopausal symptoms. The MenQoL effect size analysis showed no statistically significant difference among the four treatment arms for all four domains (vasomotor, physical, psychosocial, and sexual). Vasomotor symptoms only slightly increased under tamoxifen, with a score at year two of 1.45, 1.21, 0.58, and 1.17 in the combined, tamoxifen, fenretinide, and placebo arms, respectively. Compared with the slow metabolizers, a higher percentage of subjects with CYP2D6 extensive metabolizer genotype complained of a ≥3 score in the vasomotor, psychosocial, and sexual domain in the tamoxifen arms (P value = 0.01, 0.007, and 0.007, respectively). QoL in premenopausal or perimenopausal women was not significantly worsened by low-dose tamoxifen or fenretinide. Our findings suggest that a low dose of tamoxifen may increase its acceptability for breast cancer prevention.

Abstract 4216: Adiponectin, leptin and breast cancer in high risk postmenopausal women: Results from a nested case-control study

Introduction: Obesity is a risk factor for postmenopausal breast cancer (BC) and evidence supports the pivotal role of adipokines in the association between obesity and BC. In the present study, we aim to investigate the role of adiponectin and leptin as independent risk markers for postmenopausal BC in a prospective nested case-control study within the cohort of women from the International Breast cancer Intervention prevention Study II (IBIS-II). Methods and Results: In the IBIS-II Prevention trial, 3864 healthy postmenopausal women at increased risk for BC were randomized to receive 1 mg/day oral anastrozole or matching placebo for 5 years. We measured serum adiponectin and leptin at baseline and 12-months in 119 cases and 336 controls in the placebo arm, and 57 cases and 165 controls in the anastrozole arm. We performed the adipokines measurements on the automated immunoassay platform ELLA (ProteinSimple, Biotechne). Baseline characteristics (age, body mass index (BMI), smoking status, and oophorectomy) were not different between cases and controls in both trial arms, except for hormone replacement therapy (HRT) use (38.7% of cases vs 51.5% of controls) and hysterectomy (25.2% of cases vs 37.5% of controls) in the placebo arm. We did not observe any statistically significant difference in adiponectin and leptin at baseline between cases and controls in both trial arms. When we considered the change in adiponectin levels between baseline and 12-months of follow-up, we observed a borderline significant higher proportion of BC cases in subjects with a decrease in adiponectin (36.4%) compared to subjects with an increase in adiponectin (25.7%). The multivariate Cox model indicated that women on placebo with a decrease in adiponectin between baseline and 12-months had a 36% significant increased risk in BC (HR=0.64; CI 0.42-0.98; adjusted for age, BMI, previous HRT use, smoking, oophorectomy, hysterectomy). Changes in leptin were not significantly associated with BC incidence overall. However, the multivariate Cox model for the anastrozole arm, indicated a significant higher BC risk for women with the highest increase (&gt;10 ng/ml) vs the lowest decrease (&lt;-10 ng/ml) in leptin: HR=4.45; CI 1.08-18.30; adjusted for age, BMI, previous HRT use, smoking, oophorectomy, hysterectomy. Conclusion: Our results did not support a prognostic role of baseline adiponectin and leptin levels in postmenopausal BC development. However, we observed an increased BC risk in women with a 12-month adiponectin decrease and leptin increase. These data suggest further investigations of changes of adiponectin and leptin levels as attractive targets for BC prevention in postmenopausal women at high risk for BC.

Citation Format: Debora Macis, Ivana Sestak, Valentina Aristarco, Harriet Johansson, Aliana Guerrieri-Gonzaga, Andrea Decensi, Bernardo Bonanni, Jack Cuzick, Sara Gandini. Adiponectin, leptin and breast cancer in high risk postmenopausal women: Results from a nested case-control study [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 4216.

Hereditary lobular breast cancer with an emphasis on E-cadherin genetic defect

Recent studies have reported germline CDH1 mutations in cases of lobular breast cancer (LBC) not associated with the classical hereditary diffuse gastric cancer syndrome. A multidisciplinary workgroup discussed genetic susceptibility, pathophysiology and clinical management of hereditary LBC (HLBC). The team has established the clinical criteria for CDH1 screening and results' interpretation, and created consensus guidelines regarding genetic counselling, breast surveillance and imaging techniques, clinicopathological findings, psychological and decisional support, as well as prophylactic surgery and plastic reconstruction. Based on a review of current evidence for the identification of HLBC cases/families, CDH1 genetic testing is recommended in patients fulfilling the following criteria: (A) bilateral LBC with or without family history of LBC, with age at onset <50 years, and (B) unilateral LBC with family history of LBC, with age at onset <45 years. In CDH1 asymptomatic mutant carriers, breast surveillance with clinical examination, yearly mammography, contrast-enhanced breast MRI and breast ultrasonography (US) with 6-month interval between the US and the MRI should be implemented as a first approach. In selected cases with personal history, family history of LBC and CDH1 mutations, prophylactic mastectomy could be discussed with an integrative group of clinical experts. Psychodecisional support also plays a pivotal role in the management of individuals with or without CDH1 germline alterations. Ultimately, the definition of a specific protocol for CDH1 genetic screening and ongoing coordinated management of patients with HLBC is crucial for the effective surveillance and early detection of LBC.

Abstract P4-08-05: Impact of common polymorphisms of CYP19A1 and UGT2B17 gene deletion on early endocrine-responsive postmenopausal breast cancer

Background Polymorphisms of genes involved in estrogen production have been linked to breast cancer risk, prognosis and treatment response. Polymorphisms of the aromatase gene CYP19A1 influence its activity. The UGT2B17 catalyzes glucuronic acid transfer to a variety of substrates, including steroids and drugs like the aromatase inhibitor exemestane. We investigated the impact of two variants of CYP19A1 (rs10046, rs4646) and the UGT2B17 gene deletion on disease outcome in 125 postmenopausal women operated for ER-positive primary breast cancer enrolled in a randomized pre-surgical trial.

Patients Briefly, upon informed consent, postmenopausal patients with ER-positive breast cancer (stage T1–2, N0–1, M0) eligible for surgery were randomized to receive either exemestane (25 mg/day), or celecoxib (800 mg/day), or placebo for 6 weeks prior to surgery at the European Institute of Oncology (2004-2008). Exemestane showed a significant 10% absolute reduction in Ki67 labeling index compared to the other two arms. Serum and whole blood was taken at baseline and the day before surgery and stored at -80°C until assayed.

Methods DNA was extracted from blood by QIAamp DNA Blood Kits. The CYP19A1 rs1004/rs46466 were analyzed by Taqman genotyping assays in real-time PCR. The UGT2B17 deletion was estimated by copy number assay (Lifetechnologies). Serum estradiol (E2) and estrone (E1) levels were measured by gas chromatography tandem mass spectrometry detection (GS-MS/MS) after liquid-liquid extraction. The lower limit of quantitation were 0.625 pg/mL for estradiol and 1.56 pg/mL for estrone. The association of genetic polymorphisms with “any event” was assessed by the Cox proportional hazards models adjusted for confounders.

Results The genetic polymorphisms did not deviate from Hardy-Weinberg equilibrium (P-value &gt;0.41) and minor allele frequency of rs10046 (A/G), rs4646 (C/A), and UGT2B17Del were 0.45, 0.22, and 0.31, respectively. The rs10046 A and rs4646 C alleles were associated with higher estrogen levels. Carriers of rs10046 AA had median levels of 7.57 pg/ml E2 and 35.9 pg/mL E1 versus 3,9 pg/mL E2 and 27.4 E1 pg/mL in CA/AA genotypes (P&lt;0.003). Carriers of rs4646 CC had 5.6 pg/ml E2 and 30.45 pg/mL E1 versus 3,95 pg/mL E2 and 27.4 E1 pg/mL in CA/AA genotypes (P=0.05 only for E1). After 6 weeks treatment with exemestane, we observed steeper decreases in estrogen levels in the rs10046 AA/rs4646 CC carriers (P=0.02 for E2). After a median follow-up of 7 years we found that women carrying at least one SNP of rs10046 and one SNP of rs4646 had a better prognosis compared with women carrying homozygote wt SNPs (HR=0.44; 95% CI: 0.2-0.99 P=0.049). Similarly, the UGT2B17 deletion was associated with a better prognosis (HR= 0.43; 95% CI: 0.19-0.97; P=0.0439). There was no interaction with pre-surgical or adjuvant treatment.

Conclusions Our analysis confirms previous findings of an association of CYP19A1 rs10046/rs4646 with estrogen levels in postmenopausal women. Interestingly, the carriers of the variants associated with lower estrogen levels at diagnosis had better prognosis. Further genomic profiling in larger trials aimed to enhance tailored treatment efficacy in endocrine-responsive postmenopausal breast cancer are warranted.

Citation Format: Johansson H, Gandini S, Aristarco V, Macis D, Guerrieri-Gonzaga A, Serrano D, Pruneri G, Lazzeroni M, Viale G, Toesca A, Rajasekaran A, Bonanni B, DeCensi A. Impact of common polymorphisms of CYP19A1 and UGT2B17 gene deletion on early endocrine-responsive postmenopausal breast cancer [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P4-08-05.

Abstract P3-10-02: Presence of atypia in ductal lavage and risk of subsequent breast cancer in a prospective study

Background: Atypical hyperplasia is considered a nearly obligate precursor of breast cancer and is associated with a higher risk of developing breast cancer (BC). Attempts to improve early detection of breast cancer and to provide individualized breast cancer risk assessment would greatly benefit from sampling cellular material from the target tissue. Ductal lavage (DL) is a minimally invasive technique which provides adequate material to detect atypical cells in mammary ducts. However, long term data of the association between atypia from ductal lavage and BC risk are lacking. We studied the prevalence of atypia in DL in different risk categories and its ability to predict BC development in women at risk. Methods: From March 2000 to July 2012 we performed DL in a consecutive series of 348 women with median age of 45 years (range 19-74) at increased BC risk based on the following characteristics: 5 yrs Gail model &gt; 1.66% or &gt; 10% probability of BRCA mutation (n = 155), history of contralateral BC (CBC, n = 161), presence of a BRCA pathogenic variant (n = 32). We analyzed the presence of atypical cells in the baseline specimen of ductal lavage and in repeated lavage and observed their evolution during follow-up. Results: The procedure was safe and well tolerated in most women, with pain and disconfort preventing the procedure in 5.4% of subjects. Overall, 126 (36%) women had atypia at baseline, with a prevalence of 32%, 39%, and 41% in the Gail, CBC and BRCA groups, respectively (p = 0.38). The overall prevalence of atypia considering all visits was 44% (range 36-51). After a median follow up of 6 years, cumulative BC events were 8% in women without atypia versus 14% in those with atypia (log-rank p = 0.08). In the highest risk groups (CBC and BRCA pathogenic variants), the number of BC events was 16 (21%) in women with atypia versus 11 (10%) in women without atypia (p = 0.02 after adjustment for age). Conclusions: Our findings suggest that cytologic atypia in the fluid obtained by DL predicts subsequent BC in women at increased risk, providing individual risk assessment. The reversal of atypia in DL should be evaluated as a surrogate biomarker of BC therapeutic prevention.

Supported by: Associazione Italiana per la Ricerca sul Cancro (AIRC), Lega Italina per la Lotta contro i Tumori (LILT), AVON Foundation for Women.

Citation Format: De Censi A, Cazzaniga M, Gandini S, Casadio C, Chiapparini L, Guerrieri-Gonzaga A, Macis D, Veronesi P, Bonanni B. Presence of atypia in ductal lavage and risk of subsequent breast cancer in a prospective study [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P3-10-02.

Association of baseline plasma levels of 25OH vitamin D and breast cancer risk in a chemoprevention trial

1546

Background: Observational studies have shown a correlation of higher serum 25-hydroxyvitamin D (25OHD) concentrations and reduced cancer risk, including breast cancer (BC). We assessed the association of 25OHD and Vitamin D Receptor (VDR) polymorphisms (SNPs) with breast cancer events within a chemoprevention trial in premenopausal women at risk for BC. Methods: Premenopausal women with history of intraepithelial neoplasia (IEN) or at risk women according to the Gail model were included in a 4 arm phase II prevention trial (low dose tamoxifen vs fenrerinide vs their combination vs placebo). Level of 25OHD were measured at baseline. VDR SNPs ( FokI, BsmI, TaqI, ApaI, and Cdx2) were determined using the Applied Biosystems’ Taqman Allelic Discrimination Assay according to manufacturer’s instructions. Survival analysis for breast cancer events was performed using competing risk models, adjusted for BMI, age, season and risk strata. Results: Plasma samples were available for 228 subjects. At baseline the median plasma 25OHD concentrations were slightly higher in 53 unaffected women, 19.6 ng/ml (IQR 12.7-27.3 ng/mL) than in 175 women with IEN, 18.8 ng/ml (11.7-25.8). After a median follow up of 15 years, 79 women developed a breast cancer event, 12 had different neoplastic events. The median level of 25OHD was 19.7 ng/ml (IQR 12.8-26.1) in subjects free from oncological events and 17.8 ng/ml (9.7-23.9) in subjects with breast events. Women in the lowest 25OHD quartile (≤12 ng/ml) had an increased risk of breast cancer events: HR = 1.79 (95%CI, 1.07-2.99, p = 0.03). Considering all cancer events, the associations with 25OHD were confirmed (P = 0.005). Among the VDR SNPs the ApaI variant in homozygote or heterozygote versus wild type showed an HR of 1.83 (95%CI, 1.04-3.21 p = 0.03). Conclusions: Our results support a role of plasma vitamin D levels in breast cancer risk. Subjects in the lowest quartile had a two-fold increased risk of a cancer event. Our findings support trials of 25OHD supplementation to prevent breast cancer in 25OHD insufficient subjects.

Association of atypia in ductal lavage and breast cancer risk

e13040

Background: Atypical hyperplasia is considered a nearly obligate precursor of breast cancer and is associated with a higher risk of developing breast cancer (BC). Attempts to improve early detection of breast cancer and to provide individualized breast cancer risk assessment will greatly benefit from sampling cellular material from the target tissue. Ductal lavage (DL) is a minimally invasive technique which provides adequate material to detect atypical cells in mammary ducts. However, long term data of the association between atypia from and BC risk are lacking .We studied the prevalence of atypia in DL and its ability to predict BC development in women at risk. Methods: From March 2000 to July 2012 we performed DL in a consecutive series of 348 women with median age of 45 (range 19-74) at increased BC risk based on the following characteristics: 5 yrs Gail model > 1.66% or > 10% probability of BRCA mutation (n = 155), history of contralateral BC (CBC, n = 161), presence of a BRCA pathogenic variant (n = 32). We analyzed the presence of atypical cells in the baseline specimens and observed their evolution during follow-up. Results: The procedure was safe and well tolerated in most women, with pain and disconfort preventing the procedure in 5.4% of subjects. Overall, 126 (36%) women had atypia, with a prevalence of 32%, 39%, and 41% in the Gail, CBC and BRCA groups, respectively (p = 0.38). The overall prevalence of atypia considering all visits was 44% (range 36-51). After a median follow up of 6 years, cumulative BC events were 8% in women without atypia versus 14% in those with atypia (log-rank p = 0.08). In the highest risk groups (CBC and BRCA pathogenic variants), the number of BC events was 16 vs 11 in women with or without atypia (21% vs 10%; log-rank p = 0.05, Cox model adjusted for age p = 0.02). Conclusions: Our findings suggest that cytologic atypia in the fluid obtained by DL may predict the onset of BC in high-risk women. The reversal of atypia in DL as a surrogate biomarker of BC risk reduction warrants investigation.

A Presurgical Study of Lecithin Formulation of Green Tea Extract in Women with Early Breast Cancer

Epidemiologic data support an inverse association between green tea intake and breast cancer risk. Greenselect Phytosome (GSP) is a lecithin formulation of a caffeine-free green tea catechin extract. The purpose of the study was to determine the tissue distribution of epigallocatechin-3-O-gallate (EGCG) and its effect on cell proliferation and circulating biomarkers in breast cancer patients. Twelve early breast cancer patients received GSP 300 mg, equivalent to 44.9 mg of EGCG, daily for 4 weeks prior to surgery. The EGCG levels were measured before (free) and after (total) enzymatic hydrolysis by HPLC-MS/MS in plasma, urine, breast cancer tissue, and surrounding normal breast tissue. Fasting blood samples were taken at baseline, before the last administration, and 2 hours later. Repeated administration of GSP achieved levels of total EGCG ranging from 17 to 121 ng/mL in plasma. Despite a high between-subject variability, total EGCG was detectable in all tumor tissue samples collected up to 8 ng/g. Median total EGCG concentration was higher in the tumor as compared with the adjacent normal tissue (3.18 ng/g vs. 0 ng/g, P = 0.02). Free EGCG concentrations ranged from 8 to 65.8 ng/mL in plasma (P between last administration and 2 hours after <0.001). Free EGCG plasma levels showed a significant positive correlation with the Ki-67 decrease in tumor tissue (P = 0.02). No change in any other biomarkers was noted, except for a slight increase in testosterone levels after treatment. Oral GSP increases bioavailability of EGCG, which is detectable in breast tumor tissue and is associated with antiproliferative effects on breast cancer tissue. Cancer Prev Res; 10(6); 363-9. ©2017 AACR.

Impact of CYP19A1 and ESR1 variants on early-onset side effects during combined endocrine therapy in the TEXT trial

Background

Single nucleotide polymorphisms (SNPs) in the estrogen receptor 1 (ESR1) and cytochrome P450 19A1 (CYP19A1) genes have been associated with breast cancer risk, endocrine therapy response and side effects, mainly in postmenopausal women with early breast cancer. This analysis aimed to assess the association of selected germline CYP19A1 and ESR1 SNPs with early-onset hot flashes, sweating and musculoskeletal symptoms in premenopausal patients enrolled in the Tamoxifen and Exemestane Trial (TEXT).

Methods

Blood was collected from consenting premenopausal women with hormone-responsive early breast cancer, randomly assigned to 5-years of tamoxifen plus ovarian suppression (OFS) or exemestane plus OFS. DNA was extracted with QIAamp kits and genotyped for two CYP19A1 (rs4646 and rs10046) and three ESR1 (rs2077647, rs2234693 and rs9340799) SNPs by a real-time pyrosequencing technique. Adverse events (AEs) were recorded at baseline and 3-monthly during the first year. Associations of the genotype variants with grade ≥2 early-onset targeted AEs of hot flashes/sweating or musculoskeletal events were assessed using logistic regression models.

Results

There were 2660 premenopausal patients with breast cancer in the intention-to-treat population of TEXT, and 1967 (74 %) are included in this translational study. The CYP19A1 rs10046 variant T/T, represented in 23 % of women, was associated with a reduced incidence of grade ≥2 hot flashes/sweating (univariate odds ratio (OR) = 0.78; 95 % CI 0.63–0.97; P = 0.03), more strongly in patients assigned exemestane + OFS (TT vs CT/CC: OR = 0.65, 95 % CI = 0.48–0.89) than assigned tamoxifen + OFS (OR = 0.94, 95 % CI = 0.69–1.27, interaction P = 0.03). No association with any of the CYP19A1/ESR1 genotypes and musculoskeletal AEs was found.

Conclusion

The CYP19A1 rs10046 variant T/T favors lower incidence of hot flashes/sweating under exemestane + OFS treatment, suggesting endocrine-mediated effects. Based on findings from others, this SNP may potentially enhance treatment adherence and treatment efficacy. We plan to evaluate the clinical impact of this polymorphism during time, pending sufficient median follow up.

Trial registration

ClinicalTrials.gov NCT00066703, registered August 6, 2003.

A pooled analysis of CYP2D6 genotype in breast cancer prevention trials of low-dose tamoxifen

Decreased CYP2D6 activity is associated with lower levels of active tamoxifen metabolites. We examined the impact of CYP2D6 genotype on tamoxifen pharmacokinetics, biomarker activity, and efficacy in a pooled analysis of low-dose tamoxifen. Four randomized breast cancer prevention trials of very-low-dose (1 mg/day, n = 52 or 10 mg/week, n = 152) or low-dose tamoxifen (5 mg/day, n = 171) were pooled. DNA from 367 subjects was genotyped for CYP2D6 alleles associated with absent (PM allele: *3, *4, *5, *6, *7, *8, *12, and *14), reduced (IM allele: *9, *10, *17, *29, *41), normal (EM allele), or increased (UM: *XN) enzyme activity. Associations of tamoxifen, metabolites, activity biomarkers, and event-free survival with rapid (UM/EM, UM/IM, EM/EM, EM/IM, or EM/PM alleles) versus slow metabolizers (PM/IM or PM/PM) were investigated through random effects models, with 'study' as the random factor, and Cox regression models, adjusting for confounders. Rapid metabolizers had higher endoxifen levels than slow metabolizers: 15.3 versus 12.2 ng/mL (P = 0.018) with 5 mg/day, and 3.8 versus 2.8 ng/mL (P = 0.004) with 1 mg/day or 10 mg/week tamoxifen. The IGF-I decrease correlated with endoxifen (P = 0.002) and 4-hydroxytamoxifen levels, demonstrating steeper decreases at higher metabolite levels (P = 0.001). After a median follow-up of 12 years, rapid metabolizers with prior history of breast neoplasms allocated to tamoxifen 5 mg/day had a 60 % reduction of risk of recurrences (HR = 0.40, 95 % CI: 0.16-0.99) compared to slow metabolizers. CYP2D6 genotype may have an impact on tamoxifen efficacy at low doses. Trials investigating tamoxifen dose adjustments based on the woman's hormonal context and CYP2D6 genotype are warranted.

A Randomized, Placebo-Controlled, Phase II, Presurgical Biomarker Trial of Celecoxib Versus Exemestane in Postmenopausal Breast Cancer Patients

In breast cancer presurgical trials, the Ki-67 labeling index predicts disease outcome and offers clues to the preventive potential of drugs. We conducted a placebo-controlled trial to evaluate the activity of exemestane and celecoxib before surgery. The main endpoint was the change in Ki-67. Secondary endpoints were the modulation of circulating biomarkers. Postmenopausal women with histologically confirmed estrogen receptor-positive breast cancer were randomly assigned to exemestane 25 mg/day (n = 50), or celecoxib 800 mg/day (n = 50), or placebo (n = 25) for 6 weeks before surgery. Changes in biomarkers were analyzed through an ANCOVA model adjusting for baseline values. Exemestane showed a median absolute 10% reduction in Ki-67 [from 22 (interquartile range, IQR, 16-27), to 8 (IQR 5-18)], and a 15% absolute reduction in PgR expression [from 50 (IQR 3-90) to 15 (IQR -0-30)] after 6 weeks of treatment. Exemestane significantly increased testosterone [median change 0.21 ng/mL, (IQR 0.12-0.35)], decreased SHBG [median change -14.6 nmol/L, (IQR -23.1 to -8.6)], decreased total and HDL cholesterol by -10 mg/dL (IQR -21-2) and -7 mg/dL, (IQR -14 to -2), respectively. Triglycerides were reduced by both agents [median change -0.5 mg/dL (IQR -17.5-13.5) and -8 mg/dL (IQR -28-9) for celecoxib and exemestane, respectively]. Exemestane showed a remarkable antiproliferative effect on breast cancer, whereas celecoxib did not affect breast cancer proliferation. Given the proven preventive efficacy of exemestane, these findings support the use of Ki-67 to explore the optimal exemestane dose and schedule in the prevention setting. Cancer Prev Res; 9(5); 349-56. ©2016 AACR.

Abstract 1909: Associations of Vitamin D related polymorphisms with hereditary breast cancer in 1025 subjects undergoing BRCA 1/2 testing

Background: Vitamin D plays an important role in many normal body functions, including regulation of cell growth, bone formation, inflammation, neuromuscular and immune function. Lower serum levels of 25-hydroxyvitamin D, an index of vitamin D status, are associated with greater risk of cancer. Genetic variants in vitamin D related genes have been linked to breast cancer (BC) risk. These include the vitamin D receptor (VDR) and the GC gene that encodes for the vitamin D binding protein.

Aims: To determinate the frequency of 5 single nucleotide polymorphisms (SNPs) of VDR and GC in a large cohort of subjects candidate to BRCA 1/2 genetic testing and to assess any interaction between vitamin D and BRCA pathways on disease outcome.

Methods: From 2002 to 2012 blood samples from 1025 subjects with family history of ovarian cancer (OC) or BC were collected after informed consent was signed during the genetic counseling. DNA was extracted from whole EDTA treated blood and Allelic Discrimination (Applied Biosystems’ Taqman assays) of VDR (BsmI, FokI, TaqI, ApaI) and GC SNPs is ongoing.

Results: At last follow-up in February 2014 our cohort included 272 unaffected subjects (117 BRCA carriers, 37 WT and 118 true negative) and 753 affected subjects (276 BRCA1/2 carriers, 469 WT and 1 true negative subjects). Amongst BC patients (N = 675), 221 were BRCA carriers, 446 WT, and 8 true negative. Those affected ofBC and OC (N = 27), included 23 BRCA carriers and 4 WT, while OC patients (N = 51) included 32 BRCA carriers and 19 WT. Here we present the preliminary data obtained for VDR and GC SNPs in 833 and 710 subjects, respectively. None of the genotypes deviated from the Hardy-Weinberg equilibrium. At the AACR meeting the genotyping of all subjects will be completed.

Conclusions: Our results of vitamin D genotyping will be correlated with BC tumor molecular subtype in affected subjects, and any association with BRCA mutation and outcome will be assessed. The study was funded by IEO Foundation and AVON.

Description of vitamin D genetic polymorphisms and their frequenciesSNPsType of polymorphismMinor Allele Frequency (MAF)MAF HapMapHWE Pvalue(1df chi-square test)rs 1544410; VDR BsmITransition C/TT: 0.420.440.50rs2228570; VDR FokITransition A/GA: 0.360.410.72rs731236; VDR TaqITransition A/GC: 0.410.440.52rs7975232; VDR ApaITransversion A/CG: 0.430.430.91rs22282679; GCTransversion G/TG: 0.280.260.52

Citation Format: Valentina Aristarco, Harriet Johansson, Debora Macis, Aliana Guerrieri Gonzaga, Sara Gandini, Davide Serrano, Irene Feroce, Monica Barile, Antonella Puccio, Lorenzo Brocca, Bernardo Bonanni. Associations of Vitamin D related polymorphisms with hereditary breast cancer in 1025 subjects undergoing BRCA 1/2 testing. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1909. doi:10.1158/1538-7445.AM2015-1909

Abstract P3-06-43: A genetic deletion in the uridine diphosphate glucuronosyltransferase 2B17 affects the pharmacokinetics of exemestane

Background Exemestane (EXE) is an aromatase inactivator used in the prevention and treatment of breast cancer (BC). The majority of EXE and its active metabolite 17-dihydroEXE are excreted as glucuronide conjugates by uridine diphosphate glucuronosyltransferase (UGT). The UGT2B17 enzyme is the most expressed in human liver. A deletion spanning the entire UGT2B17 gene (*2) decreased the EXE glucuronidation process by 14-fold in human liver microsomes (HLM) from UGT2B17(*2/*2) genotype subjects compared to wild-type UGT2B17(*1/*1) HLMs.

Aim The aim of this study was to investigate whether the UGT2B17 deletion is associated with increased serum levels of EXE and 17-dihydroEXE and to assess if this deletion predicts the anti-proliferative effect of EXE in BC tissue as measured by Ki67 changes 6 weeks apart.

Methods In a phase II pre-surgical trial, 50 postmenopausal women with histologically-confirmed ER positive BC; stage T1-2, N0-1, M0 were assigned to EXE 25 mg/d for 6 weeks before surgery. Morning fasting blood was collected at baseline (B) and after 6 weeks and stored at -80° C until assayed. Time of last drug intake was collected at blood draw. DNA was extracted from whole blood (Qiagen, Italy). We used Taqman copy number variation assay (Life Technologies, Monza, Italy) for the UGT2B17 genotyping. EXE and 17-dihydroEXE concentrations were determined by mass spectrometry (MS) using Waters® Xevo™ TQ MS in electrospray positive ionization mode (Waters, Manchester, UK), optimized by multiple reaction monitoring mode. We used Zorbax Eclipse Plus C18 columns and mobile phase MeOH/H2O with 0.1% formic acid. EXE pure substance was provided by Pfizer Inc.,17β-hydroxy EXE and EXE-19-d3 purchased from Toronto Research Chemicals (Toronto, Ontario, Canada).

Wilcoxon Signed Rank Test was used to assess differences in serum concentrations of EXE, its metabolites, and post-pretreatment tissue Ki67 changes according to UGT2B17 genotypes.

Results Median age and BMI were 62 years and 26 kg/m2, respectively. The UGT2B17 genotype (n=50) was 24 homozygote wt (*1/*1), 20 heterozygote (*1/*2) and 6 homozygote variant (*2/*2). Minor allele frequency = 0.32. Hardy-Weinberg Equilibrium (P=0.57) was respected.

Median and interquartile ranges (IQRs) of EXE and 17-dihydroEXE at 6 weeks, by time elapsed since last drug intakeTime since last drugEXE (nM)17-dihydroEXE (nM)≤24 hours (n=32)10.25 (5.55, 17.25)2.75 (1.85, 3.6)&gt;24 hours (n=15)2.10 (1.20, 2.20)*1.50 (1.20, 2.00)*** P=.0001 vs ≤24h; **P=.0003 vs ≤24h

;

Median and IQRs of EXE, 17-dihydroEXE and Ki67 change from B by UGT2B17 genotypesGenotypeEXE (nM)17-dihydroEXE (nM)Ki67 changeBMIUGT2B17 *1/*1 (n=22)5.3 (2.2, 11.4)1.85 (1.4, 3.06)-9 (-16, -5)28 (24, 32)UGT2B17 *1/*2, *2/*2 (n=25)9.3 (2.6, 17.6)*2.5 (2.0, 3.6)**-11 (-19, -3)♦26 (23, 30)♦♦P-values between genotype groups:*P=.28; **P=.04; ♦P=.82; ♦♦P=.41

Conclusions Our study demonstrates a significant association of the UGT2B17 deletion with increased serum concentrations of 17-dihydroEXE, irrespective of time since last drug intake. Neither 17-dihydroEXE, nor genotype explained the significant anti-proliferative effect of EXE on BC tissue. Larger studies should determine the clinical implications of this gene on EXE efficacy.

Citation Format: Harriet Johansson, Valentina Aristarco, Jennifer Gjerde, Sara Gandini, Aliana Guerrieri-Gonzaga, Matteo Lazzeroni, Serena Mora, Debora Macis, Davide Serrano, Antonio Toesca, Luca Bottiglieri, Gunnar Mellgren, Giuseppe Viale, Andrea DeCensi, Bernardo Bonanni. A genetic deletion in the uridine diphosphate glucuronosyltransferase 2B17 affects the pharmacokinetics of exemestane [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P3-06-43.

Abstract P4-11-16: Low serum adiponectin level is an independent risk factor of DCIS in postmenopausal women at increased risk of breast cancer

Background:The assessment of breast cancer (BC) risk is a key step for an effective preventive treatment. Besides the established risk assessment models, validation of independent predictive factors such as circulating biomarkers would improve patient selection and treatment efficacy.

Obesity and metabolic imbalance play an important role in BC risk in menopausal women. The role of adipocytes in energy homeostasis is currently under investigation for their emerging relationship with BC. Adipokines (such as leptin and adiponectin) are linked to insulin sensitivity and have been related to BC risk and prognosis. Adiponectin, a peptide hormone secreted by the adipose tissue, has been inversely related to BC risk both in observational studies and in a phase II chemoprevention trial in premenopausal women at increased risk.

Aim:We measured baseline serum adiponectin and leptin levels as well as HOMA index, in 534 postmenopausal women enrolled in 16 Italian centers and randomized in one of the two international phase III trials for BC prevention -the IBIS-II(Prevention) and IBIS-II(DCIS) trials- to assess whether these biomarkers were different in the healthy women at increased risk for BC cohort compared to the DCIS cohort.

Methods:Healthy postmenopausal women (aged 40-70) at increased risk for BC (on an age-dependent risk model; n=186) or DCIS patients who underwent radical surgery in the previous 6 months (n=348) were eligible according to the two separate protocol entry criteria. At baseline, fasting blood was collected, processed and stored at -80°C till biomarkers measurement.

Insulin and glucose levels were measured with the Architect system (Abbott Laboratories, Abbott Park, IL 60064 USA). Serum adiponectin and leptin levels were determined with Immunoassays by R&D (Minneapolis, USA).

Results:Participant characteristics and biomarker levels (median, IQ range) by disease status are reported below.

Table 1Healthy (n=186)DCIS (n=348)Age at entry59 (55, 63)60 (56, 65)BMI (kg/m2)25.2 (22.9, 28.4)25.0 (22.4, 28.1)Adiponectin (ng/mL)13063 (10279, 18157)11498 (7722, 16909)Leptin (pg/mL)16181 (9594, 26391)17284 (9675, 26173)L/A ratio1.33 (0.60, 2.09)1.46 (0.69, 3.19)Glucose (mg/dL)89 (81, 97)93 (86, 103)Insulin (μU/mL)6.6 (5.0, 9.6)7.5 (5.4, 10.6)HOMA index1.39 (1.04, 2.02)1.79 (1.18, 2.61)L/A, Leptin/Adiponectin ratio

Adiponectin was significantly negatively correlated with leptin, L/A ratio, HOMA index and BMI, while leptin was positively correlated with L-A ratio, BMI, HOMA.

Logistic regression has been used and Odds Ratios (ORs) have been calculated to assess the association of DCIS with biomarkers.DCIS patients were significantly more frequent in the lowest quartile of adiponectin compared to the highest quartile (60% vs 75%; OR=2.49; 95% CI, 1.39-4.44, p= 0.0003) adjusting for center, BMI, HOMA index and age.

Conclusions:Low serum adiponectin levels in postmenopausal women are more frequent in DCIS patients compared to healthy at risk subjects independently of BMI, HOMA index and age and results are similar to premenopausal women.

Future investigations in both trials will assess whether adiponectin is also associated with BC events.

Acknowledgments: J.Forbes and T.Howell, Co-Chairmen,IBIS-II Steering Committee.

Citation Format: Aliana Guerrieri-Gonzaga, Debora Macis, Sara Gandini, Valentina Aristarco, Harriet Johansson, Giorgia Bollani, Teresa Roth, Maria-Teresa Sandri, Jill Knox, Jack Cuzick, Bernardo Bonanni. Low serum adiponectin level is an independent risk factor of DCIS in postmenopausal women at increased risk of breast cancer [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P4-11-16.

Abstract P1-12-01: CYP19A1 and ESR1 polymorphisms and selected early-onset side effects during combined endocrine therapy in the IBCSG TEXT trial for premenopausal women with hormone receptor-positive (HR+) early breast cancer

Background: Single nucleotide polymorphisms (SNPs) of the aromatase enzyme (CYP19A1) and estrogen receptor alpha (ESR1) may be associated with breast cancer susceptibility and endocrine-mediated side effects. The IBCSG Tamoxifen (Tam) and Exemestane (Exe) Trial (TEXT) includes a translational research project to assess whether selected SNPs may influence treatment efficacy and toxicity. We report on early-onset hot flashes and sweating (HF/S) and musculoskeletal symptoms (MS; myalgia, arthralgia, stiffness) with respect to CYP19A1 and ESR1 SNPs under combined endocrine therapy.

Patients and Methods: 2,672 premenopausal women with HR+ early breast cancer were randomized to treatment with the GnRH-agonist triptorelin (Trip)+Tam or Trip+Exe for 5 years. Randomization was stratified according to intended use of chemotherapy (yes/no) and lymph node status (N- vs N+). Estrogen-depletion side effects (HF/S and MS) were recorded at baseline, 3-monthly during the first year and 6-monthly thereafter using the NCI CTCAE v3.0. DNA was centrally extracted from whole blood with Qiagen kits. SNPs of CYP19A1 (rs4646 and rs10046) and ESR1 (rs207764, rs2234693 and rs9340799) were analyzed by a pyrosequencing method (Diatech Pharmacogenetics S.r.l., Jesi, Italy). Control genotypes (wt/wt; wt/v; v/v) for all SNPs were processed in each run. Logistic regression was used to analyze two endpoints: presence or absence of grade (gr) 2-3 HF/S during first 6 months and gr 2-4 MS during first 12 months. Four genetic models were used to explore associations with side effects: genetic (wt/wt; wt/v; v/v), additive, dominant and recessive.

Results: DNA was isolated and genotyped for 1970 (74%) consenting women. Clinical characteristics and outcomes of this cohort were consistent with the overall trial. At baseline median age was 44, median BMI 24 kg/m2 and 86% had regular menses. During follow-up, 43% reported gr 2-3 HF/S and 27% reported gr 2-4 MS. The 5 SNPs did not deviate from Hardy-Weinberg equilibrium (p&gt;0.30) and minor allele frequency ranged from 29%-48%. The CYP19A1 SNP rs10046 (C&gt;T) was associated with gr 2-3 HF/S. Specifically, women with variant homozygote genotype (T/T) had a reduced risk of HF/S (39% T/T vs 45%). The univariate odds ratio (OR) was 0.77 (95%CI: 0.62-0.96; p=0.02) compared with other variant groups. The multivariate model showed consistent results after adjusting for age, BMI, menstrual status, chemo use, treatment allocation (Tam vs Exe) and baseline HF/S. The other 4 SNPs were not associated with the selected side effects.

Conclusions: Our analysis indicates association of HF/S with CYP19A1 rs10046 genetic variants. The lack of association between early-onset of selected estrogen-depletion side effects and the other 4 SNPs may be masked by the concurrent Trip. Ovarian reserve before treatment may also influence the impact of SNP genotypes on early-onset side effects. Further analysis with details of concurrent medications and treatment adherence will contribute to the interpretation of these results.

Funded by Susan G. Komen for the Cure.

Citation Format: Harriet Johansson, Kathryn P Gray, Olivia Pagani, Meredith M Regan, Giuseppe Viale, Valentina Aristarco, Debora Macis, Antonella Puccio, Susanne Roux, Rudolf Maibach, Marco Colleoni, Manuela Rabaglio-Poretti, Alan S Coates, Richard D Gelber, Aron Goldhirsch, Roswitha Kammler, Bernardo Bonanni, Barbara A Walley. CYP19A1 and ESR1 polymorphisms and selected early-onset side effects during combined endocrine therapy in the IBCSG TEXT trial for premenopausal women with hormone receptor-positive (HR+) early breast cancer [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P1-12-01.

Circulating adiponectin and breast cancer risk: a systematic review and meta-analysis

BACKGROUND

We conducted a meta-analysis in order to investigate whether circulating adiponectin, an insulin-sensitizing hormone produced by adipocytes, is associated with breast cancer risk.

METHODS

A systematic literature search was performed in PubMed, Medline, EMBASE, ISI Web of Knowledge and the Cochrane Library. The summary relative risk (SRR) was calculated by pooling the different study-specific estimates using the random effect models. Meta-regression, subgroup and sensitivity analyses were carried out to investigate between-study heterogeneity and to test publication bias.

RESULTS

Data from 15 observational studies, published between 2003 and April 2013 for a total of 4249 breast cancer cases, were analysed. The SRR for the 'highest' vs 'lowest' adiponectin levels indicated a 34% reduction in breast cancer risk [95% confidence interval (CI): 13%-50%]. Between-study heterogeneity was not substantial (I(2)=53%). Ten studies were included in the dose-response analysis: the SRR for an increase of 3 µg/ml of adiponectin corresponded to a 5% risk reduction (95% CI: 1%-9%). The comparison between 'highest' and 'lowest' levels of adiponectin showed an inverse association in postmenopausal women (SRR=0.80; 95% CI: 0.63-1.01) and an indication of an inverse relationship in premenopausal women (SRR=0.72, 95% CI: 0.30-1.72). No evidence of publication bias was found.

CONCLUSIONS

Low circulating adiponectin levels are associated with an increased breast cancer risk. However, properly designed studies are needed to confirm the role of adiponectin as breast cancer biomarker, and clinical trials should be performed to identify those interventions that may be effective in modulating adiponectin levels and reducing breast cancer risk.

Circulating hormones and breast cancer risk in premenopausal women: a randomized trial of low-dose tamoxifen and fenretinide

Tamoxifen and fenretinide have been extensively studied and exhibit breast cancer-preventing activity. We aimed to assess their effect on sex hormones, sex hormone binding globulin (SHBG) and retinol, and their association with mammographic density (MD) and breast cancer events. In a double-blind, placebo-controlled trial, premenopausal women at risk for breast cancer were randomized to tamoxifen 5 mg/day, fenretinide, both agents, or placebo for 2 years. We measured MD and circulating concentrations of follicle-stimulating hormone, luteinizing hormone (LH), estradiol, progesterone, testosterone, androstenedione, dehydro-epiandrosteronesulfate, prolactin, SHBG, and retinol at baseline and on yearly intervals. The associations with breast cancer events were evaluated through competing risk and Cox regression survival models. Low-dose tamoxifen markedly and enduringly increased SHBG, whereas the increases in testosterone, estradiol, and prolactin and reduction in LH weakened after 1 year. Fenretinide increased testosterone and androstenedione and decreased retinol. MD correlated directly with SHBG and inversely with retinol. After a median follow-up of 12 years, the 10-year cumulative incidence of breast cancer events was 37 % in women with SHBG ≤ 59.3 nmol/L, 22 % in women with SHBG between 59.3 and 101 nmol/L, and 19 % in women with SHBG > 101 nmol/L (P = 0.018). The difference among SHBG tertiles remained statistically significant at multivariable analysis: HR = 2.26 (95 % CI 1.04, 4.89) for the lowest versus the highest tertile. We conclude that low-dose tamoxifen or fenretinide exhibits favorable hormonal profiles as single agents, further supporting their administration for prevention of breast cancer in premenopause. Notably, SHBG levels were inversely associated with breast neoplastic events.

A randomized phase II presurgical trial of weekly low-dose tamoxifen versus raloxifene versus placebo in premenopausal women with estrogen receptor-positive breast cancer

Introduction

We previously demonstrated that 1 or 5 mg per day of tamoxifen (T) given for four weeks before surgery reduces Ki-67 in breast cancer (BC) patients to the same extent as the standard 20 mg/d. Given the long half-life of T, a weekly dose (10 mg per week (w)) may be worth testing. Also, raloxifene (R) has shown Ki-67 reduction in postmenopausal patients in a preoperative setting, but data in premenopausal women are limited. We conducted a randomized trial testing T 10 mg/w vs. R 60 mg/d vs. placebo in a presurgical model.

Methods

Out of 204 screened subjects, 57 were not eligible, 22 refused to participate and 125 were included in the study. The participants were all premenopausal women with estrogen receptor-positive BC. They were randomly assigned to either T 10mg/w or R 60 mg/d or placebo for six weeks before surgery. The primary endpoint was tissue change of Ki-67. Secondary endpoints were modulation of estrogen and progesterone receptors and several other circulating biomarkers.

Results

Ki-67 was not significantly modulated by either treatment. In contrast, both selective estrogen receptor modulators (SERMs) significantly modulated circulating IGF-I/IGFBP-3 ratio, cholesterol, fibrinogen and antithrombin III. Estradiol was increased with both SERMs. Within the tamoxifen arm, CYP2D6 polymorphism analysis showed a higher concentration of N-desTamoxifen, one of the tamoxifen metabolites, in subjects with reduced CYP2D6 activity. Moreover, a reduction of Ki-67 and a marked increase of sex hormone-binding globulin (SHBG) were observed in the active phenotype.

Conclusions

A weekly dose of tamoxifen and a standard dose of raloxifene did not inhibit tumor cell proliferation, measured as Ki-67 expression, in premenopausal BC patients. However, in the tamoxifen arm women with an extensive phenotype for CYP2D6 reached a significant Ki-67 modulation.

Abstract 167: A comparison of 25-OH-vitamin D levels and VDR polymorphisms in healthy and patients with breast intraepithelial neoplasia or cancer

Background: Vitamin D is mainly synthesized in the skin through UVB irradiation, while a minor intake comes from the diet. Vitamin D is converted in the liver to 25-hydroxyvitamin D-(25(OH)D) and circulating levels of this metabolite is an appropriate index of vitamin D status. A further hydroxylation occurs to produce the biologically active metabolite 1,25-dihydroxyvitamin D.

Epidemiological studies have suggested a role for vitamin D in cancer prevention showing an inverse relationship between incidence of cancers and mortality and 25(OH)D plasma levels. Vitamin D exerts its activity through a nuclear receptor (VDR) that regulates the expression of multiple genes, including genes for cell cycle regulation, differentiation, and apoptosis

Methods: We have collected plasma from pre and post menopausal healthy women (HW) and newly diagnosed patients with an intraepithelial neoplasia (IEN) or breast carcinoma (BC). We measured circulating levels of 25(OH)D3 using LC-MS/MS. The median values were then correlated with the disease status. In a subgroup, we performed a correlation with VDR polymorphisms (SNPs) and disease recurrences. BsmI, FokI, TaqI and ApaI SNPs were determined using the Applied Biosystems’ Taqman Allelic Discrimination Assay.

Results: 25(OH)D3 was measured in 249 HW, 175 subjects with IEN and 168 BC patients undergoing surgery. Age and BMI were taken into account for adjustments the median age being 51, 46 and 59 years and median BMI being 24, 23 and 25.6 kg/m2 in the HW, IEN and BC categories, respectively. Median 25(OH)D3 were respectively: 37.2, (inter-quartile range, IQR 25.0-49.7) 43 (IQR 30.0-60.0) and 42 (IQR 30.0-60.0) nmol/L respectively. The LS means with BC + IEN combined versus HW were 43 (IQR 40.8-45.2) vs 46.4 (IQR 43.9-49.0) nmol/L. The levels of 25(OH)D3 are statistical different between healthy versus affected subjects (p = 0.05). We also conducted an exploratory analysis in a subgroup of 228 subjects to evaluate the correlation of VDR polymorphisms and the disease free survival (DFS for breast recurrence or new event). With a follow-up of 160 months and a total of 70 breast events, the homozygote and heterozygote subjects for VDR ApaI polymorphism had a worst DFS compared to the wild type ones (log rank test p= 0.04).Conclusions: our data confirm the possible role of the vitamin D and its receptor in the carcinogenesis process. It is not clear when 25(OH)D level starts to decline and thus may influence the disease progression. In accordance with other research groups the ApaI polymorphism may have, together with other VDR SNPs, an influence on BC incidence and recurrence. Further correlations and analysis of the VDR genotype are ongoing. A prospective study to clarify is recommended the issue of the timing of 25(OH)D levels measurement is warranted.

Citation Format: Davide Serrano, Harriet Johansson, Debora Macis, Ernst Lien, Jennifer Gjerde, Antonella Puccio, Sara Gandini, Aliana Guerrieri-Gonzaga, Irene Feroce, Matteo Lazzeroni, Andrea Decensi, Bernardo Bonanni. A comparison of 25-OH-vitamin D levels and VDR polymorphisms in healthy and patients with breast intraepithelial neoplasia or cancer . [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 167. doi:10.1158/1538-7445.AM2013-167

Abstract 161: Effect of Simvastatin and Nimesulide on inflammation and cardiovascular risk biomarkers in a phase II breast cancer prevention trial

Background: Almost one-third of breast cancers (BC) will not be influenced by hormonal interventions because of the absence of ER expression. Thus, there is an urgent need to confirm in the clinical setting the potential efficacy of new compounds in contrasting hormone non-responsive BC, which are often marked by higher biological aggressiveness, early onset and worse prognosis. Over expression of COX-2 and HMG-CoA pathway seems to be involved in breast carcinogenesis and several studies have shown the ability of NSAIDs and statins to reduce the incidence of BC.

Material and Methods: We conducted a randomized phase II, double blind, placebo controlled trial with simvastatin (20 mg/day) and nimesulide (100 mg/day) for 1 year in 150 high risk women for ER negative BC in a 1:1:1 fashion. Participants were treated for one year and followed-up for an additional year. The primary endpoint was the change in prevalence of atypical cells obtained by ductal lavage or ultrasound guided fine needle aspirate and cellular proliferation (measured by Ki-67), while the secondary objective was to analyze the efficacy of these drugs in modifying the levels of circulating inflammatory and cardiovascular risk biomarkers. Circulating concentrations of high-sensitivity C-reactive protein (hsCRP), triglycerides, total cholesterol, LDL-cholesterol, HDL-cholesterol, antithrombin III and fibrinogen were measured on morning fasting blood samples drawn at baseline and after 12 and 24 months.

Results: Analysis of the primary endpoint is ongoing and here we present results on circulating biomarkers. After 12 months of treatment, simvastatin was associated to a statistically significant mean decrease of hsCRP compared to the placebo arm (0,90 mg/L vs 3,57 mg/L; p=0.008), while an attenuated effect of nimesulide was observed (1.61 mg/L vs 3.57 mg/L; p=0.077). As expected, simvastatin significantly decreased LDL and total cholesterol levels with respect to placebo (89.1 vs 130 mg/dL and 176.8 vs 218 mg/dL respectively; both p&lt;0.0001). Interestingly, after 1 year of treatment interruption, at 24 months from baseline, hsCRP serum levels were significantly lower with both simvastatin (p = 0.012) and nimesulide (p = 0.017) as compared to placebo. We did not observe a significant change in mean concentrations of triglycerides, antithrombin III and fibrinogen during treatment and follow-up.

Conclusions: Our findings demonstrate a favourable anti-inflammatory effect of both agents on circulating levels of hsCRP, although simvastatin achieved a greater decrease. These results will be correlated with the main endpoint to further assess the potential role of simvastatin and nimesulide in breast carcinogenesis and the potential use of hsCRP as a surrogate endpoint biomarker.

Citation Format: Valentina Aristarco, Davide Radice, Harriet Joahnsson, Debora Macis, Serena Mora, Massimiliano Cazzaniga, Laura Cortesi, Isabella Marchi, Maria Teresa Sandri, Bernardo Bonanni. Effect of Simvastatin and Nimesulide on inflammation and cardiovascular risk biomarkers in a phase II breast cancer prevention trial. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 161. doi:10.1158/1538-7445.AM2013-161

Dual effects of metformin on breast cancer proliferation in a randomized trial

519

Background: Metformin lowers breast cancer risk in observational studies in diabetics, but evidence for its clinical activity is scanty. We studied the metformin antiproliferative effect in a pre-surgical study. Since the antidiabetic effect of metformin is heterogeneous according to obesity and insulin resistance (IR), we also determined whether its antiproliferative effect was modified by risk biomarkers. Methods: After tumor biopsy, we randomly allocated 200 non-diabetic women with breast cancer to either metformin, 850 mg/bid (n=100) or placebo (n=100) for 4 wks. The primary endpoint was the post-pretreatment change in Ki-67 between arms. We explored effect modifications by STEPP and tested biomarker thresholds that showed an interaction with treatment on Ki-67. Results: Overall, median (IQR) Ki-67 was 19 (14-31) at baseline and 21 (14-32) after 4 wks and 18 (12-29) at baseline and 20 (13-31) after 4 wks in the metformin and placebo arm, respectively with mean increase of 4.0% (95%CI, -5.6 to 14.4) on metformin versus placebo. Multivariate analyses showed an increase of Ki-67 after 4 wks placebo in the following subgroups: HOMA> IR threshold, IGFBP3> highest quartile, IGFBP1< lowest quintile, IGFratio (IGF1/IGFBP3)< median, CRP> inflection point at STEPP analysis, HER2+ve tumors (versus –ve). Metformin blunted the increase of Ki-67 noted on placebo in these subgroups. Conclusions: Overall, metformin did not affect Ki-67 in most subjects with breast cancer. However, in exploratory analysis we identified subgroups of patients where metformin showed antiproliferative effect. Further studies to a personalized approach are warranted with selection of study populations. [Table: see text]

Abstract 2664: Biological relevance of CYP2D6 phenotype and low dose tamoxifen activity within a randomized phase II pre-surgical trial

Background: To further investigate tamoxifen (T) minimal active dose, we conducted a window of opportunity study in estrogen receptor (ER) positive breast cancer (BC) patients. The study was a three arms randomized phase II double blind clinical trial. The participants were randomized to tamoxifen 10 mg a week vs. raloxifene 60 mg a day vs. placebo for 6 weeks, which was the average waiting list in our Institute at the time of this study was conducted. The primary endpoint was tissue change of Ki67LI. This result was previously reported and overall we did not observe any KI67LI modulation by treatment. The clinical relevance of CYP2D6 polymorphisms is controversial, and recently negative data have been published. Here we present a significant biomarkers modulation by low dose tamoxifen based on CYP2D6 subjects phenotype. Methods: DNA was extracted from whole blood by the use of QIAamp DNA blood kit. We employed the INFINITI analyzer for CYP2D6 genotyping, according to the protocol of the manufacturer. A total of 14 SNPs were analyzed. To determine the concentrations of tamoxifen and its metabolites, we used a high-pressure liquid chromatography-tandem mass spectrometry system. Expression of Ki-67, ER and PgR were determined by immunohistochemistry. Serum concentration of SHBG and estradiol were measured by chemiluminescent immunometric assay; IGF1 and IGFBP3 were determined by enzymatic colorimetric methods. Results: CYP2D6 was analyzed in the 50 women who took tamoxifen. We found 40% of EM (extensive metabolizer), 48% of IM (intermidiet metabolizer), 6% of PM (poor metabolizer) and 6% UM (ultra-rapid metabolizer). The phenotypes were grouped in two classes: active (EM + UM) and with a reduced activity (IM + PM). Among T metabolites a significant accumulation of N-desmethyl-Tamoxifen was present in the reduced activity group (mean plasma level 19.9 vs. 14.4 ng/ml reduced vs. active group p=.05). Endoxifen and 4-OH-Tamoxifen were not significantly different in the two phenotype groups. Then we correlated the phenotype with biomarkers modulations. The active phenotype showed a significant different modulation in Ki67 with a median chage of −2.5 in the active group and +3 in the reduced one. No differences on ER or PgR expression were determined by the phenotype. Among the circulating biomarkers, SHBG in the active group showed a significant difference compared to the reduced one (median change 8.35 and 2 p=0.05) whereas the estradiol and IGF1/BP3 did not show any difference between groups. Conclusions: CYP2D6 phenotypes interact with the biological activity of low dose T. A reduced activity of the enzyme induces an accumulation of N-desmethyl-Tamoxifen, whereas no significant differences for Endoxifen. In spite of a similar Endoxifen concentration, a reduced activity phenotype may influence T activity as demonstrated by the different modulation of Ki67 and SHBG biomarkers.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 2664. doi:1538-7445.AM2012-2664

Prognostic effect of circulating adiponectin in a randomized 2 x 2 trial of low-dose tamoxifen and fenretinide in premenopausal women at risk for breast cancer

PURPOSE

Adipokines are linked to obesity and insulin sensitivity and have recently been related to breast cancer risk and prognosis. We investigated the associations of plasma leptin and adiponectin with mammographic density and disease status and assessed their prognostic effect on recurrence-free survival in premenopausal women at risk for breast cancer.

PATIENTS AND METHODS

We measured circulating lipids, insulin-like growth factor 1, glucose, insulin and insulin sensitivity (calculated by homeostasis model assessment [HOMA] index), leptin, adiponectin, and leptin-to-adiponectin ratio in 235 premenopausal women with pT1mic/pT1a breast cancer (n = 21), intraepithelial neoplasia (n = 160), or 5-year Gail risk of 1.3% or greater (n = 54) who participated in a 2 × 2 trial of low-dose tamoxifen, fenretinide, both agents, or placebo over a 2-year period.

RESULTS

At baseline, adiponectin levels were directly associated with mammographic density and HDL cholesterol and negatively associated with leptin, leptin-to-adiponectin ratio, body mass index (BMI), and HOMA index. Median adiponectin levels were lower in affected than in unaffected women (P = .006). After a median of 7.2 years and total of 57 breast neoplastic events, there was a 12% reduction in the risk of breast neoplastic events per unit increase of adiponectin (adjusted hazard ratio, 0.88; 95% CI, 0.81 to 0.96; P = .03). There was no interaction between treatment and adiponectin levels.

CONCLUSION

Low adiponectin levels are associated with a history of prior intraepithelial neoplasia or pT1mic/pT1a breast cancer and higher risk of second breast neoplastic events in premenopausal women. The associations are independent of BMI, mammographic density, and treatment. Our findings support the role of adiponectin as a potential target for premenopausal breast cancer prevention and treatment.

Lapatinib activity in premalignant lesions and HER-2-positive cancer of the breast in a randomized, placebo-controlled presurgical trial

Dual epidermal growth factor receptor (EGFR) and HER2 targeting with the tyrosine kinase inhibitor lapatinib is approved for treating advanced HER2-positive breast cancer and can prevent estrogen receptor (ER)-negative mammary tumors in HER2 transgenic mouse models. Ki-67 labeling index (LI) has prognostic and predictive value and can be used to screen drugs' therapeutic and preventive potential in a clinical model of short-term presurgical therapy of breast cancer. We conducted a randomized, placebo-controlled trial of lapatinib (1500 mg/d) administered orally for three weeks between biopsy and surgery in 60 women with HER-2-positive breast cancer to assess lapatinib biomarker (including the primary endpoint, Ki-67 LI) and clinical activity in invasive breast cancer, adjacent ductal intraepithelial neoplasia (DIN, which comprises ductal carcinoma in situ and atypical ductal hyperplasia), and distant ductal hyperplasia without atypia (DH). Ki-67 LI increased progressively in association with disease stage, increasing in the placebo arm, for example, by medians of 3% in DH to 20% in DIN to 30% in invasive cancer. Ki-67 LI in cancer tissue decreased by a mean (±SD) of 9.3% (±34.2) in the lapatinib arm and increased by 15.1% (±30.9) in the placebo arm (P = 0.008). Compared with placebo, lapatinib reduced Ki-67 significantly more in ER-negative tumors (by 34.8%; P = 0.01) but not significantly more in ER-positive tumors (by 12.3%; P = 0.2) and reduced Ki-67 more (nonsignificantly) in cytosol PTEN-overexpressing tumors (P = 0.057). The prevalence of DIN in post-treatment surgical specimens of both arms was similar (70%-76%), with a median Ki-67 of 15% (range, 5%-35%) on lapatinib versus 20% (5%-60%) on placebo (P = 0.067). The prevalence of DH also was similar in both arms (>90%), with a median Ki-67 of 1% (1%-7%) on lapatinib versus 3% (1%-5%) on placebo (P = 0.006). Other results of lapatinib versus placebo, respectively, were as follows: Median tumor diameter at surgery of 18 mm (11 mm-57 mm) versus 24 mm (10 mm-37 mm; P = 0.009); partial response of 13.6% versus 3.7%, stable disease of 59.1% versus 40.7%, and progression of 27.3% versus 55.6% (P-trend = 0.035). In conclusion, short-term lapatinib decreased cell proliferation in DIN, DH, and invasive HER-2-positive (especially ER-negative) breast cancer, thus providing the rationale for further clinical development of lapatinib for breast cancer prevention in high-risk patients, including those with HER-2-positive DIN.

Abstract 3675: Effects of cytochrome P450 2D6 polymorphisms on tamoxifen metabolism in a pooled analysis of breast cancer prevention trials designed to define the minimal active dose of the drug

Background Tamoxifen is the most widely used drug for the prevention and treatment of hormone-receptor positive breast cancer (BC). Nevertheless, its clinical effectiveness varies among individuals. Tamoxifen is extensively metabolized by cytochrome P450 (CYP) enzymes, and recent in vivo studies have shown that women with genetically impaired CYP2D6 have reduced production of endoxifen, the main active metabolite. This impairment has been linked to a higher risk of breast cancer recurrence. Despite these observations, the contribution of endoxifen to the overall drug effectiveness of tamoxifen remains uncertain. Methods and results This study presents a pooled analysis of four chemoprevention trials assessing three different low dose tamoxifen schedules: 1mg/day (n=52), 10 mg/week (n=152) and 5 mg/day (n=158). Each participant had signed an informed consent. Genomic DNA was extracted from whole blood by the QIAamp DNA Blood Kit (Qiagen, Valencia, CA, USA), and the CYP2D6 polymorphisms analyzed by the use of the INFINITI CYP450 2D6T Assay (AutoGenomics, Carlsbad, CA, USA). The following alleles were identified: Full enzyme activity: *1, *2; reduced activity: *9, *41 *29; no activity: *3, *4, *5, *6. A total of 336 out of the 362 study participants were genotyped. Circulating concentrations of tamoxifen and its metabolites were determined on cryo-conserved (−80°C) serum samples collected at different treatment intervals according to study design. The women were defined as extensive metabolizers (EM) if they had 2 fully active alleles; intermediate (IM) if they carried reduced alleles or one non-functional allele, while poor metabolizers (PM) carried two non-functional alleles. We found the following phenotype frequencies: 146/336 (43%) EM, 171/336 (51%) IM, 19/336 (6%) PM. When analyzing the effects of phenotypes on serum concentrations of tamoxifen and its metabolites we found a statistically significant effect of phenotype on tamoxifen and metabolites. EM had lower levels of tamoxifen (8,6 ng/mL; P&lt;0.02) and its main metabolite N-desmethyltamoxifen (Ndestam: 17,4 ng/mL; P&lt;0.0001) as compared to impaired metabolizers (IM+PM), who had median tamoxifen levels of 15 ng/mL and Ndestam of 30 ng/mL. No difference was observed on endoxifen levels, possibly due to the low dose schedules. Nevertheless, the increase in Ndestam concentrations in the group of impaired metabolizers is an important indirect effect of lower CYP2D6 activity, showing an accumulation of its main substrate. The relationship between CYP2D6 phenotype and tamoxifen pharmacokinetics will be extended, including women taking tamoxifen at 20mg/day for one year. Conclusions These findings support the relevance of pharmacogenetics in tailoring tamoxifen treatment in the setting of breast cancer prevention.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 3675. doi:10.1158/1538-7445.AM2011-3675

Abstract 3671: Associations between a polymorphic locus in the promoter region of the IGFBP-3 gene, circulating IGFs, mammographic breast density and survival in a 2×2 trial of low-dose tamoxifen and fenretinide in premenopausal women

Background The Insulin-like Growth Factors (IGFs) pathway has been reported to play a role in breast carcinogenesis. A single nucleotide polymorphism in the promoter region of the IGFBP-3 gene (rs2854744), located at position -202 from the transcription start site, results in reduced promoter activity. Some evidence of an association between this polymorphism and circulating IGFs levels, mammographic breast density and breast cancer risk have been reported.

Methods and results A total of 235 premenopausal women with intraepithelial neoplasia (IEN, n=181) or 5-year Gail risk &gt;1,3% (n=54), were randomly assigned in a double-blind four-arm trial to receive tamoxifen 5 mg/d, fenretinide, a retinoic acid derivative, 200 mg/d, both agents, or placebo for 2 years.

Mammography and fasting blood samples for serum IGF-I and IGFBP-3 measurements were collected at baseline and every 12 months for 3 years, within the same phase of the menstrual cycle. A total of 223 subjects were genotyped for IGFBP-3 A-202C with a Taqman real time PCR method for allelic discrimination.

The IGFBP-3 A-202C genotype frequencies for AA, AC and CC were 31.4%, 52.5% and 16.1% respectively. According to baseline risk stratification, we observed a higher frequency of IGFBP-3 -202 CC genotype (92%) as compared to the AA+AC genotype group (76%, p&lt;0.04) in the IEN group. Baseline serum IGFBP-3 (μg/ml) and IGF-I/IGFBP-3 molar ratio median values were 4.15 and 0.19 in the AA; 3.82 and 0.21 in the AC; 3.57 and 0.23 in the CC genotype group respectively. There was a significant trend for higher baseline IGFBP-3 levels and lower IGF-I/IGFBP-3 ratio with increasing number of the A allele at IGFBP-3 -202 locus (p=0.0005 and p=0.005 respectively) adjusting for risk stratification at baseline. Median baseline mammographic percent density values for IGFBP-3 AA, AC and CC genotype groups were 48.4%, 46.1% and 42.8% respectively. Interestingly, women with at least one A allele at IGFBP-3 -202 locus (AA or AC) had a 4-fold increase in the odds of having a high percentage of breast density (&gt; 42.8%) compared to women with IGFBP-3 -202 CC genotype (OR=4.21; CI 1.17-15.18, p&lt;0.05) adjusting for BMI and dense area. After a median follow-up of 7.5 years, 59 new breast events were observed. Subjects with IGFBP-3 -202 AA genotype had a borderline significantly greater disease free survival compared to AC+CC genotype group (p=0.06). There was no significant interaction between IGFBP-3 genotype and treatment.

Conclusions Our results provide further support for an association between IGFBP-3 A-202C polymorphism and circulating IGFBP-3 levels. Moreover, our data provide first evidence of an association between IGFBP-3 genotype and disease free survival in women with breast IEN. Further studies are warranted to fully elucidate these relationships.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 3671. doi:10.1158/1538-7445.AM2011-3671

Abstract 3670: A randomized, placebo-controlled, phase II, presurgical biomarker trial of celecoxib or exemestane in postmenopausal breast cancer patients

Background Treatment of breast cancer (BC) before surgery provides a window of opportunity to screen the activity and mechanism of action of selected agents. Exemestane (Exe) is an irreversible inactivator of aromatase, the enzyme converting androgens in estrogens. Celecoxib (Cel) is a selective inhibitor of cyclooxygenase-2 (Cox-2), the enzyme involved in the production of prostaglandins that play a key role in tumor proliferation. We conducted a double-blind, placebo-controlled, randomized trial to evaluate the activity and efficacy of Cel and Exe given for 6 weeks before surgery. The main endpoint biomarker was Ki-67 labeling index (LI), a well known marker of tumor proliferation. Secondary endpoints were the modulation of biomarkers associated to BC and cardiovascular disease.

Methods Postmenopausal women with histologically-confirmed ER positive BC; stage T1-2, N0-1, M0 were randomly assigned to either Exe 25 mg/d (n=50), or Cel 800 mg/d (n=50) or placebo (n=25) for 6 weeks before surgery. Fasting blood samples were withdrawn at baseline and the day before surgery. Baseline median age was 62, 65, 63 years, and median BMI was 26.56, 24.89, 26.67 in the Exe, Cel and placebo group, respectively. Changes in biomarkers were analyzed through an ANCOVA model adjusting for baseline values.

Results Compared to baseline, Exe showed a 10% statistically significant reduction in KI-67% LI (95%CI; -12.52, -7.75) and a 26% reduction in progesterone receptor (PgR) expression (95% CI; -32.67, -18.62) while in the Cel group, we observed a 5% decrease in PgR expression (95%CI: -12.08; 1.84). No change was observed in ER expression. In the placebo arm Ki-67% LI, ER and PgR expression did not change significantly. Compared to baseline, Exe significantly increased testosterone levels by 0.25 ng/mL (95%CI; 0.21, 0.28), while no modulation was observed in the other two arms: Least square means of changes in testosterone were -0.005 (95%CI:-0.05, 0.04) in the placebo arm and -0.001 (95%CI:-0.03, 0.03) in the Cel arm (P for differences between treatment arms&lt;0.0001), respectively. As regards high-sensitivity C-reactive protein levels, no change was observed in any arm. Exe induced a statistically significant -10.87 (95%CI:-17.32, -4.41) decrease in total cholesterol and -6.61(95%CI: -9.63,-3.59) in HDL cholesterol. Triglycerides levels were unchanged. Conversely, Cel induced a borderline statistically significant reduction of antithrombin III levels (P among active treatments =.08) while no difference was observed for fibrinogen levels.

Conclusions Our study confirms the anti-proliferative effects of Exe on BC, together with a favourable effect on lipid profile. Further investigations, including the modulation of markers of bone metabolism and the activity of Cox-2 will be conducted to better elucidate the mechanism of action of both drugs in BC prevention.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 3670. doi:10.1158/1538-7445.AM2011-3670

Abstract P5-11-14: Distribution of Tamoxifen and Five Metabolites into Normal and Malignant Human Breast Cancer Tissue during Conventional and Low Dose Treatment

Background: Biotransformation of tamoxifen leads to metabolites with different pharmacological and toxicological effects. Profiling and screening for active metabolites are important to assess their contribution to the overall therapeutic and adverse effects. The anticancer effect of tamoxifen is believed to be due to the hydroxylated metabolites 4hydroxytamoxifen (4OHtam) and 4hydroxyNdesmethyltamoxifen (4OHNDtam, endoxifen). Its demethylated metabolites have been related to side effects of the drug, whereas effects of tamoxifenNoxide (tamNoxide) have not been evaluated. In the present study we wanted to examine the distribution of tam and five of its metabolites in serum and normal and malignant breast tissues during normal dose (20 mg/day) and low dose (5 or 1 mg/day) regimens after 28 days of continuous dosing.

Methods: Serum and biopsies from 56 patients were available for analysis. The concentrations of tamoxifen and the metabolites 4OHtam, 4OHNDtam, Ndesmethyltamoxifen (NDtam), Ndesdimethyltamoxifen (NDDtam), and tamNoxide in serum and tissues were measured by means of a recently developed highly sensitive liquid chromatography tandem mass spectrometry (LCMSMS) method.

Results: Serum and biopsies from 56 patients were available for analysis. All compounds including the potent metabolites 4OHtam and 4OHNDtam were detected and measured in all samples. The tissue levels of tam and its hydroxylated and demethylated metabolites were highly related to the serum levels (p &lt; 0.000 — 0.0069). The tumor tissue to serum concentration ratios of tamNoxide decreased by increasing doses (p &lt; 0.0001) whereas this ratio was unchanged over the dose range 1-20 mg/day for the other compounds measured.

Conclusion: Tamoxifen and its metabolites were detected in all samples examined including those sampled during the low dose regimens. The tissue levels of tamoxifen and its hydroxylated and demethylated metabolites were strongly corresponding to the serum concentrations, in contrast to those of tamNoxide. The strong relations between the concentrations of the potent active 4OHtam and 4OHNDtam (endoxifen) in serum and target tissues suggest that implementation of therapeutic drug monitoring may represent a tool to improve tamoxifen therapy.

Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P5-11-14.

Abstract PD03-02: A Randomized Pre-Surgical Trial of Metformin in Breast Cancer. Preliminary Feasibility and Safety Results

Background: Metformin has been associated with decreased breast cancer (BC) incidence in diabetic patients in epidemiological studies. Moreover, this drug results in initiation of an AMPK-dependent energy stress response which can adversely affect survival of breast cancer cell lines and inhibition of PI3K/Akt/mTOR signaling leading to reduced proliferation of BC cell lines.

Methods: We conducted a randomized, pre-surgical, phase IIb, placebo-controlled, biomarker trial in women with stage I-III BC candidate to elective surgery. The primary endpoint was the change in cell proliferation in malignant, dysplastic and hyperplastic tissue as measured by Ki-67 labeling index (LI). With 150 subjects, the study was 80% powered to test for the interaction between metformin activity and ER status. Two interim analyses are being planned after 100 and 200 women enrolled. Results: As of June 18, 2010, 175 subjects have been randomized and 162 have completed treatment. Here we report data on the first 100 women enrolled. At the symposium, full data on the first 200 women enrolled will be available. As of December 31, 2009, a total of 163 women were screened, 26 were not eligible and 40 refused to participate, 6 dropped out during the study for AEs (n=2) or refusal to continue treatment (n=4), thus leaving 95 subjects assessable for the primary endpoint. The main subject and tumor characteristics blinded as to the allocated arm were mean age 52 (31-77), Pre/postmenopause, 54/41, mean BMI, 23.6 (18.0-40.2). At baseline, median Ki-67LI was 18% (range 4%-65%) at biopsy and 19% (4%-70%) after 4 weeks at surgery. All adverse events except for 1 SAE were grade 1 or 2, consisting of G2 nausea and G2 diarrhea in 4% and 6% of the cases, respectively. The prevalence of ductal intraepithelial neoplasia was 91% (median Ki-67 LI, 10%) in samples both adjacent and distant from the tumor, whereas the prevalence of ductal hyperplasia was 77% (median Ki-67LI, 2%) in samples distant from the tumor. Conclusions: Our preliminary results show the feasibility, high compliance and safety of a metformin trial in breast cancer patients. Assessment of tissue and circulating biomarkers is currently ongoing to characterize the whole spectrum of metformin activity in malignant and dysplastic tissue.

Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr PD03-02.

Abstract A61: Circulating adiponectin levels as a predictive factor of breast cancer in a double-blind 2x2 phase II trial of low-dose tamoxifen and fenretinide for breast cancer prevention

Background: The assessment of breast cancer risk is a key step for an effective preventive treatment. Besides the established risk assessment models, independent predictive factors like circulating biomarkers may also be important to better select high risk subjects and to determine efficacy of chemopreventive treatment.

Adiponectin is a peptide hormone secreted from the adipose tissue with anti-inflammatory, anti-atherogenic and anti-diabetic properties. Epidemiological studies have shown that plasma adiponectin level is inversely related to breast cancer risk.

In the present study, we measured plasma adiponectin levels in premenopausal women participating in a phase II trial of low-dose tamoxifen and fenretinide for breast cancer prevention.

Methods: Premenopausal women (n = 235) were randomly assigned in a double-blind 2x2 trial to receive tamoxifen 5 mg/d, fenretinide, a vitamin A derivative, 200 mg/d, both agents, or placebo for 2 years. A total of 181 premenopausal women with ductal intraepithelial neoplasia (DIN) and 54 unaffected women at higher risk according to the Gail model were enrolled.

Mammographic percent density was centrally measured using the computer-assisted method described by Boyd. Plasma adiponectin was measured by use of a commercial enzyme-linked immunosorbent assay kit.

Preliminary results: According to disease status (DIN vs Gail) at baseline, median plasma adiponectin levels were lower in women with a DIN (10 ug/ml; interquartile (IQ) range: 7.1-14) compared to unaffected women with a Gail risk (12 ug/ml; IQ range: 8.7-14.7) (p=0.05).

Importantly, after a median follow-up of 5.5 years, plasma adiponectin levels were lower in women who had a breast cancer event (7.7 ug/mL; IQ range: 6.46-12.78) compared to women without event (11.02 ug/mL; IQ range: 7.8-14.4) (p=0.015). This difference is also maintained after adjusting for BMI and treatment allocation (p=0.011).

Considering the distribution of adiponectin levels at baseline, we observed a 9% reduction in the hazard ratio of breast cancer events by each unit increase of adiponectin levels (95%CI: 0.83-0.96; p=0.02, cox model adjusted for BMI, treatment allocation, mammographic density and disease status at baseline).

Conclusion: We found that subjects with lower plasma adiponectin levels had a significantly higher risk of breast cancer event. The association appeared independent of known risk factors such as BMI and mammographic breast density. Further studies to better understand and confirm the role of adiponectin as independent predictive risk factor are warranted.

Citation Information: Cancer Prev Res 2010;3(12 Suppl):A61.