Advertisement

A New Genetic Risk Score to Predict the Outcome of Locally Advanced or Metastatic Breast Cancer Patients Treated With First-Line Exemestane: Results From a Prospective Study

Open AccessPublished:November 23, 2018DOI:https://doi.org/10.1016/j.clbc.2018.11.009

      Abstract

      Introduction

      Approximately 50% of locally advanced or metastatic breast cancer (MBC) patients treated with first-line exemestane do not show objective response and currently there are no reliable biomarkers to predict the outcome of patients using this therapy. The constitutive genetic background might be responsible for differences in the outcome of exemestane-treated patients. We designed a prospective study to investigate the role of germ line polymorphisms as biomarkers of survival.

      Patients and Methods

      Three hundred two locally advanced or MBC patients treated with first-line exemestane were genotyped for 74 germ line polymorphisms in 39 candidate genes involved in drug activity, hormone balance, DNA replication and repair, and cell signaling pathways. Associations with progression-free survival (PFS) and overall survival (OS) were tested with multivariate Cox regression. Bootstrap resampling was used as an internal assessment of results reproducibility.

      Results

      Cytochrome P450 19A1-rs10046TC/CC, solute carrier organic anion transporter 1B1-rs4149056TT, adenosine triphosphate binding cassette subfamily G member 2-rs2046134GG, fibroblast growth factor receptor–4-rs351855TT, and X-ray repair cross complementing 3-rs861539TT were significantly associated with PFS and then combined into a risk score (0-1, 2, 3, or 4-6 risk points). Patients with the highest risk score (4-6 risk points) compared with ones with the lowest score (0-1 risk points) had a median PFS of 10 months versus 26.3 months (adjusted hazard ratio [AdjHR], 3.12 [95% confidence interval (CI), 2.18-4.48]; P < .001) and a median OS of 38.9 months versus 63.0 months (AdjHR, 2.41 [95% CI, 1.22-4.79], P = .012), respectively.

      Conclusion

      In this study we defined a score including 5 polymorphisms to stratify patients for PFS and OS. This score, if validated, might be translated to personalize locally advanced or MBC patient treatment and management.

      Graphical abstract

      Keywords

      Introduction

      Despite that personalized medicine has assumed a crucial role in recent years, pharmacogenetic investigations, aiming at identifying predictive/prognostic biomarkers, usually represent only secondary aims in clinical trials. In this scenario, setting up prospective studies, designed ad hoc to investigate the effect of genetic variants in candidate genes and pathways, could be an effective strategy to produce reliable and clinically useful results.
      Exemestane is a steroidal aromatase inhibitor (AI) for the treatment of postmenopausal patients affected by hormone receptor-positive (HR+) breast cancer (BC). It is registered for the adjuvant and the advanced settings for the treatment of locally advanced or metastatic BC (MBC).
      • Smith I.E.
      • Dowsett M.
      Aromatase inhibitors in breast cancer.
      Exemestane has shown efficacy in several clinical trials,
      • Coombes R.C.
      • Kilburn L.S.
      • Snowdon C.F.
      • et al.
      Survival and safety of exemestane versus tamoxifen after 2-3 years’ tamoxifen treatment (Intergroup Exemestane Study): a randomised controlled trial.
      • De Placido S.
      • Gallo C.
      • De Laurentiis M.
      • et al.
      Adjuvant anastrozole versus exemestane versus letrozole, upfront or after 2 years of tamoxifen, in endocrine-sensitive breast cancer (FATA-GIM3): a randomised, phase 3 trial.
      • Paridaens R.J.
      • Dirix L.Y.
      • Beex L.V.
      • et al.
      Phase III study comparing exemestane with tamoxifen as first-line hormonal treatment of metastatic breast cancer in postmenopausal women: the European Organisation for Research and Treatment of Cancer Breast Cancer Cooperative Group.
      but almost the 50% of patients, after an initial benefit, experience disease progression.
      • Paridaens R.J.
      • Dirix L.Y.
      • Beex L.V.
      • et al.
      Phase III study comparing exemestane with tamoxifen as first-line hormonal treatment of metastatic breast cancer in postmenopausal women: the European Organisation for Research and Treatment of Cancer Breast Cancer Cooperative Group.
      • Asten K.V.
      • Neven P.
      • Lintermans A.
      • Wildiers H.
      • Paridaens R.
      Aromatase inhibitors in the breast cancer clinic: focus on exemestane.
      • Mauri D.
      • Pavlidis N.
      • Polyzos N.P.
      • Ioannidis J.P.
      Survival with aromatase inhibitors and inactivators versus standard hormonal therapy in advanced breast cancer: meta-analysis.
      Unfortunately, there are still no biomarkers to predict the outcome of exemestane-treated patients.
      • Musgrove E.A.
      • Sutherland R.L.
      Biological determinants of endocrine resistance in breast cancer.
      • Osborne C.K.
      • Schiff R.
      Mechanisms of endocrine resistance in breast cancer.
      • Dowsett M.
      • Martin L.A.
      • Smith I.
      • Johnston S.
      Mechanisms of resistance to aromatase inhibitors.
      A great effort has been done in the past years to identify germ line variants as predictive and/or prognostic biomarkers in patients treated with AIs, including exemestane.
      • Oesterreich S.
      • Henry N.L.
      • Kidwell K.M.
      • et al.
      Associations between genetic variants and the effect of letrozole and exemestane on bone mass and bone turnover.
      • Peterson A.
      • Xia Z.
      • Chen G.
      • Lazarus P.
      Exemestane potency is unchanged by common nonsynonymous polymorphisms in CYP19A1: results of a novel anti-aromatase activity assay examining exemestane and its derivatives.
      • Ferraldeschi R.
      • Arnedos M.
      • Hadfield K.D.
      • et al.
      Polymorphisms of CYP19A1 and response to aromatase inhibitors in metastatic breast cancer patients.
      • Ingle J.N.
      • Schaid D.J.
      • Goss P.E.
      • et al.
      Genome-wide associations and functional genomic studies of musculoskeletal adverse events in women receiving aromatase inhibitors.
      Polymorphisms in genes affecting either the activity or the expression of the drug target (aromatase, cytochrome P450 19A1 [CYP19A1]), or influencing drug/hormone activity (CYP17A1,
      • Dunning A.M.
      • Dowsett M.
      • Healey C.S.
      • et al.
      Polymorphisms associated with circulating sex hormone levels in postmenopausal women.
      • Tworoger S.S.
      • Chubak J.
      • Aiello E.J.
      • et al.
      Association of CYP17, CYP19, CYP1B1, and COMT polymorphisms with serum and urinary sex hormone concentrations in postmenopausal women.
      estrogen receptor 1 [ESR1],
      • Oesterreich S.
      • Henry N.L.
      • Kidwell K.M.
      • et al.
      Associations between genetic variants and the effect of letrozole and exemestane on bone mass and bone turnover.
      • Henry N.L.
      • Skaar T.C.
      • Dantzer J.
      • et al.
      Genetic associations with toxicity-related discontinuation of aromatase inhibitor therapy for breast cancer.
      ESR2,
      • Oesterreich S.
      • Henry N.L.
      • Kidwell K.M.
      • et al.
      Associations between genetic variants and the effect of letrozole and exemestane on bone mass and bone turnover.
      PR/SET domain 2 [PRDM2]
      • Grundberg E.
      • Akesson K.
      • Kindmark A.
      • et al.
      The impact of estradiol on bone mineral density is modulated by the specific estrogen receptor-alpha cofactor retinoblastoma-interacting zinc finger protein-1 insertion/deletion polymorphism.
      ), metabolism (Catechol-O-methyltransferase [COMT],
      • Dunning A.M.
      • Dowsett M.
      • Healey C.S.
      • et al.
      Polymorphisms associated with circulating sex hormone levels in postmenopausal women.
      • Tworoger S.S.
      • Chubak J.
      • Aiello E.J.
      • et al.
      Association of CYP17, CYP19, CYP1B1, and COMT polymorphisms with serum and urinary sex hormone concentrations in postmenopausal women.
      CYP1B1,
      • Tworoger S.S.
      • Chubak J.
      • Aiello E.J.
      • et al.
      Association of CYP17, CYP19, CYP1B1, and COMT polymorphisms with serum and urinary sex hormone concentrations in postmenopausal women.
      uridine diphosphate [UDP] glucuronosyltransferase family 1 member A1 [UGT1A1],
      • Kuo S.H.
      • Yang S.Y.
      • You S.L.
      • et al.
      Polymorphisms of ESR1, UGT1A1, HCN1, MAP3K1 and CYP2B6 are associated with the prognosis of hormone receptor-positive early breast cancer.
      CYP3A4,
      • Abubakar M.B.
      • Wei K.
      • Gan S.H.
      The influence of genetic polymorphisms on the efficacy and side effects of anastrozole in postmenopausal breast cancer patients.
      • Hertz D.L.
      • Kidwell K.M.
      • Seewald N.J.
      • et al.
      Polymorphisms in drug-metabolizing enzymes and steady-state exemestane concentration in postmenopausal patients with breast cancer.
      CYP3A5
      • Abubakar M.B.
      • Wei K.
      • Gan S.H.
      The influence of genetic polymorphisms on the efficacy and side effects of anastrozole in postmenopausal breast cancer patients.
      ), or transport (adenosine triphosphate [ATP] binding cassette [ABC],
      • Kuo S.H.
      • Yang S.Y.
      • You S.L.
      • et al.
      Polymorphisms of ESR1, UGT1A1, HCN1, MAP3K1 and CYP2B6 are associated with the prognosis of hormone receptor-positive early breast cancer.
      and solute carrier organic anion [SLCO] transporters
      • van der Deure W.M.
      • Friesema E.C.
      • de Jong F.J.
      • et al.
      Organic anion transporter 1B1: an important factor in hepatic thyroid hormone and estrogen transport and metabolism.
      ) have been investigated as potential predictive or prognostic biomarkers of efficacy.
      • Turkistani A.
      • Marsh S.
      Pharmacogenomics of third-generation aromatase inhibitors.
      In this context, another mechanism of interest is the DNA repair pathway, that is deregulated in BC. The deficiency in DNA repair capacity is considered a hallmark of breast carcinogenesis.
      • Matta J.
      • Morales L.
      • Ortiz C.
      • et al.
      Estrogen receptor expression is associated with DNA repair capacity in breast cancer.
      A recent report on the outcome of patients treated with AIs has highlighted somatic mutations on genes involved in DNA replication, repair, cell cycle, and tumor protein p53 signaling pathways. These somatic mutations were associated with AI resistance.
      • Ellis M.J.
      • Ding L.
      • Shen D.
      • et al.
      Whole-genome analysis informs breast cancer response to aromatase inhibition.
      However, the role of the germ line variants in these pathways has not been investigated.
      To assess the clinical value of germ line polymorphisms as potential indicators of exemestane outcome, we performed a prospective multicenter study, specifically designed to investigate the role of germ line DNA variants to predict survival in patients treated with exemestane. This study was carried out on 302 HR+ locally advanced or MBC patients treated with exemestane as first-line hormonal therapy. Seventy-four polymorphisms in 39 candidate genes involved in drug activity, hormone balance, DNA replication and repair, and cell signaling pathways were investigated to identify biomarkers of progression-free survival (PFS) and overall survival (OS).

      Patients and Methods

       Patients

      This prospective study involved 23 Italian centers. All procedures were approved by the ethics committee of the sponsoring center, National Cancer Institute Centro di Riferimento Oncologico di Aviano (protocol number 1003/D; November 11, 2005), and investigations were performed in accordance with Declaration of Helsinki. HR+ locally advanced or MBC postmenopausal patients treated with exemestane were enrolled from 2007 to 2012. Patients signed a written informed consent for the purpose of this research.
      Eligibility criteria included: blood sample availability, measurable and nontarget lesions defined according to Response Evaluation Criteria In Solid Tumors version 1 (RECIST) criteria, Eastern Cooperative Oncology Group performance status 0 to 2, absolute neutrophil count ≥1500/μL, platelets ≥100,000/μL, hemoglobin ≥9.0 g/dL, and HER2-negative. Exclusion criteria were: previous exemestane treatment or other hormonal therapy in the advanced/metastatic setting, brain metastasis, serious infectious disease, serious functional alteration of visceral and metabolic disease, radiotherapy or major surgery within 4 weeks from start of exemestane treatment, previous or concomitant neoplasm (excluding in situ cervical cancer), and inability to attend periodical clinical and/or radiological evaluations.
      Clinical data were collected on case report forms. Details on primitive tumor, previous treatments, exemestane therapy, and follow-up information are reported in Table 1. Data were reviewed by an internal board and stored in a database. Hormone receptor status was assessed according to the 2010 American Society of Clinical Oncology guidelines.
      Table 1Patient Clinical and Demographic Characteristics
      Patient and Tumor CharacteristicValuePFS HR, POS HR, P
      Total, n302
      Age, Years
       Median (range)71 (35-93)
       <60, n (%)58 (19.2)RefRef
       61-70, n (%)88 (29.1)NSNS
       71-80, n (%)108 (35.8)0.69, .051NS
       >80, n (%)48 (15.9)0.44, .001NS
      Sex, n (%)
       Female301 (99.6)
       Male1 (0.4)
      Stage at Diagnosis, n (%)
       I-II130 (43.0)RefRef
       III-IV170 (56.3)1.38, .0152.04, .001
       Unknown2 (0.7)
      Dominant Metastatic Site, n (%)
       Visceral183 (60.6)
       Bone
      Ten of 302 patients had bone and soft tissue lesions.
      102 (33.8)
      Ten of 302 patients had bone and soft tissue lesions.
       Soft tissue17 (5.6)
      Liver Involvement
       No257 (85.1)RefRef
       Yes45 (14.9)2.49, .0011.83, .008
      Metastatic Sites at Recruitment, n (%)
       191 (30.1)
       2-3157 (52.0)
       4-554 (17.9)
      ER/PgR Status
       ER+/PgR+257 (85.1)
       ER+/PgR-44 (14.6)
       ER/PgR+1 (0.3)
      ER Expression, n (%)
       0-50%100 (33.1)
       51-75%49 (16.2)
       76-100%153 (50.7)
      PgR Expression, n (%)
       0-10%87 (28.8)RefRef
       11-100%215 (71.2)0.58, .0010.41, .001
      Surgery, n (%)
       No60 (19.9)RefRef
       Yes242 (80.1)NS0.44, .001
      Previous Chemotherapy, n (%)
       No138 (45.7)RefRef
       Yes164 (54.3)1.56, .001NS
       Neo- or adjuvant124 (41.1)
       First-line
      Seventeen patients also underwent neo- or adjuvant therapy.
      40 (13.2)
      Seventeen patients also underwent neo- or adjuvant therapy.
      Previous Hormone Therapy, n (%)
       No140 (46.4)RefRef
       Adjuvant tamoxifen98 (32.5)NSNS
       Adjuvant AI
      Thirty-one patients have also taken tamoxifen.
      64 (21.1)
      Thirty-one patients have also taken tamoxifen.
      1.49, .018NS
      PFS
       Progression, n (%)238 (78.8)
       Median PFS (range)15.4 (0.8-115.6)
      OS
       Deaths, n (%)141 (46.7)
       Median OS (range)26.8 (1.5-152.6)
      Significant results are shown in bold text.
      Abbreviations: AI = aromatase inhibitor; ER = estrogen receptor; HR = hazard ratio; NS = not significant; OS = overall survival; PFS = progression-free survival; PgR = progesterone receptor; Ref = reference category.
      a Ten of 302 patients had bone and soft tissue lesions.
      b Seventeen patients also underwent neo- or adjuvant therapy.
      c Thirty-one patients have also taken tamoxifen.

       Data Statement

      The data set generated and analyzed during the current study is not publicly available because the biological material and clinical data were collected from patients only for the purposes of this study. However, the data set is available from the corresponding author upon reasonable request.

       Efficacy Evaluation

      Patients were treated with 25 mg of daily oral exemestane. Efficacy evaluation occurred every 8 weeks for at least 24 weeks according to RECIST criteria. Treatment was continued until disease progression, unacceptable toxicity, death, or consent withdrawal.
      Progression-free survival was defined as the time between treatment initiation and tumor progression or death from any cause. OS was defined as the time between treatment initiation and death from any cause. Patients lost at follow-up were censored at the time of their last follow-up.

       Gene Variants and Genotyping

      Genes and polymorphisms of interest were selected upon literature analysis on the basis of association with cancer, BC, and AIs outcome. Moreover, to assess the involvement of selected genes into pathways of interest, each gene was searched in the Gene database of the National Center for Biotechnology Information . Genes and polymorphisms investigated are listed in Supplemental Table 1 in the online version. Linkage disequilibrium between variants was assessed using the Genome Variation Server (http://gvs.gs.washington.edu/GVS150/). A 3-mL blood sample was obtained from each patient before the first administration of exemestane. Genomic DNA was automatically extracted with BioRobot EZ1 kit (Qiagen SPA, Milan, Italy).
      Polymorphisms were genotyped with Automated Fragment Analysis on Genetic Analyzer ABI PRISM 3100 (Applied Biosystems, Foster City, CA), TaqMan Assays on a 7500 Real-Time PCR System (Applied Biosystems), PSQ96 MA Pyrosequencing (Biotage AB, Uppsala, Sweden), or a custom-designed Illumina GoldenGate Assay on a BeadXpress Reader (Illumina, San Diego, CA). Analyses were performed according to manufacturer’s instructions including negative and positive controls. Twelve polymorphisms analyzed using the GoldenGate Assay were also analyzed using TaqMan (Applied Biosystems) or Pyrosequencing (Biotage AB) as control.

       Statistical Analyses

      Progression-free survival and OS were assessed using multivariate Cox regression; clinical variables were included in the model if associated with a P value < .05 at univariate analysis, with one of the outcomes considered (Table 1). Proportional hazards assumption was assessed using Schoenfeld residuals. All tests of statistical significance were 2-sided and medians were reported with their relative minimum and maximum range or 95% confidence interval (CI). Robust standard errors were calculated to take into account the possible lack of independence between patients from the same hospital.
      • Rogers W.
      Regression standard errors in clustered samples.
      Additive, dominant, and recessive genetic models were evaluated, and the most statistically significant model for each polymorphism was selected. Ninety-five percent CIs and P values were estimated using a bootstrap resampling technique with 1000 replications. A P value cutoff of .05 after bootstrapping was considered to be statistically significant. Unadjusted differences in PFS and OS according to genotypes were assessed using Kaplan–Meier estimates and the statistical significance using the log rank test. Genotypes individually associated with a higher risk of progression or not showing a protective effect were used to generate a genetic risk score. All the analyses were performed using STATA version 13 software (StataCorp LLC, Cary, NC).

      Results

       Patients and Clinical Outcome

      Three hundred two patients with locally advanced or MBC treated with first-line exemestane were enrolled in this prospective pharmacogenetic study (Figure 1). All of the subjects were of Caucasian origin (self-reported). Patient median age was 71 (range, 35-93) years and the median follow-up was 35 (range, 2-153) months. The median treatment duration was 11.7 months (range, 0.7-84.8). Patients, tumor, and treatment characteristics are summarized in Table 1.
      Figure thumbnail gr1
      Figure 1Study Flow Chart. Three Hundred Two Patients Were Enrolled in the Period From 2007 to 2012. Among Them, at Least 298 Were Genotyped for All of the 74 Polymorphisms Investigated. All of the Polymorphisms Were Tested for Association With Progression-Free Survival (PFS) and Overall Survival (OS). Cox Multivariate Regression and Bootstrap Analyses Highlighted 5 Polymorphisms as Significantly Associated With PFS. None of the Tested Polymorphisms Was Significantly Associated With OS. The Risk Score Obtained From the Combination of These 5 Polymorphisms Was Subsequently Tested for Association With PFS and OS, Showing a Better Stratification Ability Than the Single Polymorphisms for PFS and a Trend for Association With OS

       Genetic Analyses

      Seventy-four polymorphisms in 39 genes involved in drug activity, hormone balance, DNA replication and repair, and cell signaling pathways were identified. The allele frequencies are reported in Supplemental Table 1 in the online version and were consistent with those previously reported (https://www.ncbi.nlm.nih.gov/snp). As a test for genotyping quality control, genotype data were obtained using 2 different techniques in 247 patients for CYP19A1-rs4646, CYP3A5-rs776746, COMT-rs4680, ESR1-rs2234693, ESR2-rs1256049, ESR2-rs4986938, CYP17A1-rs743572, CYP19A1-rs700519, CYP19A1-rs10046, CYP3A4-rs2740574, and PRDM2-rs2308040 variants. These analyses had a complete concordance rate (100%).

       Progression-Free Survival and OS Analyses

      The following clinical variables were associated with either PFS and/or OS and were used as covariates for multivariate analyses: age, stage at diagnosis, progesterone receptor expression, liver involvement, surgery, chemotherapy, and adjuvant hormonal therapy (Table 1).
      In multivariate analysis, 5 polymorphisms were significantly associated with PFS: CYP19A1-rs10046, solute carrier organic anion transporter 1B1 (SLCO1B1)-rs4149056, ATP binding cassette subfamily G member 2 (ABCG2)-rs2046134, fibroblast growth factor receptor–4 (FGFR4)-rs351855, and X-ray repair cross complementing 3 (XRCC3)-rs861539 (Table 2, Figure 2).
      Table 2Variants Significantly Associated With PFS
      GenersIDVariantModelAdjHR (95% CI)P95% CI Bootstrapped ValueP Bootstrapped Value
      ABCG2rs2046134G>ADominant0.62 (0.41-0.93).0200.39-0.97.038
      CYP19A1rs10046T>CAdditive1.15 (1.02-1.29).0211.01-1.31.038
      SLCO1B1rs4149056T>CDominant0.54 (0.41-0.71).0000.36-0.80.002
      FGFR4rs351855C>TRecessive1.85 (1.15-2.99).0111.03-3.34.039
      XRCC3rs861539C>TRecessive1.72 (1.15-2.57).0081.02-2.90.043
      The results were adjusted for the clinical variables significantly associated with PFS and OS. 95% CI and P values were estimated using the bootstrap resampling method by drawing 1000 samples from the original data set.
      Abbreviations: AdjHR = adjusted hazard ratio; OS = overall survival; PFS = progression-free survival; rsID = reference single-nucleotide polymorphism identification number.
      Figure thumbnail gr2
      Figure 2Kaplan–Meier Estimates for Progression-Free Survival (PFS) According to 3 Polymorphisms in Hormone Balance Pathways: (A) CYP19A1-rs10046, Additive Model (add), (B) SLCO1B1-rs4149056, Dominant Model (dom), and (C) ABCG2-rs2046134, dom; and 2 Polymorphisms in DNA Repair and Cell Signaling Pathways: (D) FGFR4-rs351855, Recessive Model (rec), and (E) XRCC3-rs861539, rec. Adjusted Hazard Ratio (AdjHR) and P Values Were Determined Using a Multivariate Cox Regression and Bootstrap Analyses: CYP19A1-rs10046: AdjHR, 1.15 (95% CI Bootstrapped Value [bootsr], 1.01-1.31), P bootsr = .038; SLCO1B1-rs4149056: AdjHR, 0.54 (95% CI bootsr, 0.36-0.80), P bootsr, .002; ABCG2-rs2046134: AdjHR, 0.62 [95% CI bootsr, 0.39-0.97], P bootsr = .038; FGFR4-rs351855: AdjHR, 1.85 (95% CI bootsr, 1.03-3.34), P bootsr = .039; XRCC3-rs861539: AdjHR, 1.72 (95% CI bootsr, 1.02-2.90), P bootsr = .043). “+” Indicates Censored
      In particular, the variant C allele of CYP19A1-rs10046, on the aromatase gene, was significantly associated with an increased risk of progression (additive model, adjusted hazard ratio [AdjHR], 1.15; bootstrapped 95% confidence interval [95%CIbootsr, 1.01-1.31; bootstrapped P value (pbootsr) = .038). A reduced risk of progression was observed for SLCO1B1-rs4149056, and for ABCG2-rs2046134, on genes involved in steroid transport (SLCO1B1-rs4149056 dominant model, AdjHR = 0.54 [95%CIbootsr, 0.36-0.80], pbootsr = .002; ABCG2-rs2046134 dominant model, AdjHR = 0.62 [95%CIbootsr, 0.39-0.97], pbootsr = .038). The homozygous variant TT genotypes of FGFR4-rs351855 and of XRCC3-rs861539 were associated with a higher risk of progression compared with wild type or heterozygous patients (FGFR4-rs351855 recessive model, AdjHR = 1.85 [95%CIbootsr, 1.03-3.34], pbootsr = .039; XRCC3-rs861539 recessive model, AdjHR = 1.72 [95%CIbootsr, 1.02-2.90], pbootsr = .043).
      To assess the role of the simultaneous presence of different risk alleles, 0, 1, or 2 points were assigned to the 6 genotypes previously described and a risk score was generated grouping the 5 polymorphisms associated with shorter PFS. A point was assigned to each genotype of the 5 polymorphisms according to its risk of progression, as shown in Table 3: additive model: 0 points if patients had 0 risk alleles, 1 point for 1 risk allele, 2 points for 2 risk alleles; dominant models: 0 points if patients had at least 1 protective allele, 1 point for 0 protective alleles; recessive models: 0 points if patients had 0 or 1 risk allele, and 1 point for 2 risk alleles. According to the genotype of each polymorphism, patients had a total score derived according to the sum of the assigned points, ranging from 0 to 6. Patients were then aggregated into 4 risk groups: 0 to 1 risk points (n = 26 patients), 2 risk points (n = 84 patients), 3 risk points (n = 111 patients), and 4 to 6 risk points (n = 75 patients).
      Table 3Points Attributed to the Risk Genotypes Associated With PFS and OS
      PolymorphismPatient’s GenotypeType of AlleleRisk Points AttributedGenetic Model
      CYP19A1-rs10046TTC = risk allele0Additive
      CYP19A1-rs10046TC1
      CYP19A1-rs10046CC2
      SLCO1B1-rs4149056TC or CCC = protective allele0Dominant
      SLCO1B1-rs4149056TT1
      ABCG2-rs2046134AA or AGA = protective allele0Dominant
      ABCG2-rs2046134GG1
      FGFR4-rs351855TC or CCT = risk allele0Recessive
      FGFR4-rs351855TT1
      XRCC3-rs861539TC or CCT = risk allele0Recessive
      XRCC3-rs861539TT1
      Abbreviations: OS = overall survival; PFS = progression-free survival.
      Figure 3A and Table 4 show PFS in the 4 groups of patients stratified according to the risk score. As shown, patients in the 4 to 6 risk group performed worse than those in the 3, 2, and 0 to 1 groups (log rank P = .0002). Accordingly, the risk of progression increased with the number of risk points: patients in the 4 to 6 risk group and those in the 3 risk group had a significantly greater risk of progression than patients in the 0 to 1 risk group (AdjHR for the 4-6 risk group = 3.12 [95% CI, 2.18-4.48], P < .001; AdjHR for the 3 risk group = 2.01 [95% CI, 1.20-3.36], P = .008). Even if the difference was not statistically significant, the group of patients carrying 2 risk points had a greater risk of progression than patients with 0 to 1 risk points (AdjHR for the 2 risk group = 1.39 [95% CI, 0.88-2.20], P = .149).
      Figure thumbnail gr3
      Figure 3Kaplan–Meier Curves for (A) Progression-Free Survival (PFS) and (B) Overall Survival (OS) According to the Genetic Risk Score. Patients Were Divided Into 4 Groups On the Basis of the Number of Risk Points. Kaplan–Meier Log Rank Survival Analysis (2-Sided) Was Used to Calculate P Values (PFS, P = .0002; OS, P = .07). “|” Indicates Censored
      Table 4Median PFS and OS According to the Risk Score: Univariate and Multivariate Analysis
      Risk PointsMedian Survival (95% CI), MonthsLog Rank PUnivariateMultivariate
      HR (95% CI)PAdjHR (95% CI)P
      PFS
       0-126.3 (15.9-39.0).0002RefRef
       222.3 (13.2-30.3)1.32 (0.98-1.79).071.40 (0.89-2.20).149
       313.4 (9.7-17.9)1.80 (1.21-2.66).0042.00 (1.20-3.36).008
       4-610.0 (8.1-14.4)2.54 (1.90-3.39)<.0013.12 (2.18-4.48)<.001
      OS
       0-163.0 (43.9-not reached).07RefRef
       258.9 (48.5-69.8)1.44 (0.79-2.63).2331.34 (0.77-2.34).300
       348.3 (36.5-65.1)1.72 (0.85-3.47).1311.77 (0.97-3.20).061
       4-638.9 (34.3-45.8)2.28 (1.61-4.50).0172.41 (1.22-4.79).012
      Significant results are shown in bold text.
      Abbreviations: AdjHR = adjusted hazard ratio; HR = hazard ratio; OS = overall survival; PFS = progression-free survival; Ref = reference category.
      Patients with the highest score (4-6 risk points) exhibited the lowest median PFS: 10.0 months (95% CI, 8.1-14.4 months), whereas patients with the lowest score (0-1 risk points) showed a median PFS of 26.3 months (95% CI, 15.9-39.0 months; P = .0002).
      None of the germ line variants individually investigated was significantly associated with OS. Nonetheless, the risk score allowed a global stratification of patients according to OS with the same trend (P = .07) observed for PFS (Figure 3B). Intriguingly, by comparing only the groups with the highest risk score (4-6 points) versus the lowest risk score (0-1 points), the difference was significant in the univariate and in the multivariate models (Table 4).

      Discussion

      In the modern armamentarium of the treatment of HR+ locally advanced or MBC, AIs still retain a crucial role. However, although a first stabilization of the tumor burden takes place in several locally advanced or MBC patients treated with AIs, disease progression almost always occurs, frequently after only a few months of treatment. Prognostic or predictive biomarkers for exemestane outcome are, hence, a clinical need.
      Recently, somatic DNA variations have been described to affect AI response during neoadjuvant treatment, but the role of germ line variants in PFS and OS after first-line exemestane treatment remains to be elucidated. A prospective multicenter study was designed to identify germ line variants associated with PFS and OS in locally advanced or MBC patients treated with first-line exemestane. Polymorphic variants in genes involved in exemestane pharmacokinetics and pharmacodynamics, hormone balance, DNA repair, and cell signaling pathways were considered. Multivariate and bootstrap analyses highlighted 5 polymorphisms as significantly associated with PFS: CYP19A1-rs10046, XRCC3-rs861539, ABCG2-rs2046134, SLCO1B1-rs4149056, and FGFR4-rs351855. The main findings reported in the literature regarding these polymorphisms are summarized in Supplemental Table 2 in the online version.
      Cytochrome P450 19A1 codes for the aromatase enzyme, the target of exemestane. Aromatase catalyzes the conversion of C19 androgens into C18 estrogens, a critical step in estrogen biosynthesis, inhibited by exemestane. Extensive research has been done to investigate the role of CYP19A1 polymorphisms in hormone therapy efficacy, but data produced so far are contradictory. A recent meta-analysis
      • Artigalás O.
      • Vanni T.
      • Hutz M.H.
      • Ashton-Prolla P.
      • Schwartz I.V.
      Influence of CYP19A1 polymorphisms on the treatment of breast cancer with aromatase inhibitors: a systematic review and meta-analysis.
      described an association between the CYP19A1-rs4646 polymorphism and time to progression in AI-treated BC patients. The authors concluded that the effect of CYP19A1 polymorphisms on clinical outcomes were most often detected in individual studies, underling the necessity of performing prospective validation studies. Our prospective study failed to confirm any association between rs4646 and PFS. The only CYP19A1 polymorphism associated with PFS was the rs10046, a 3′ untranslated region variation, which has a low level of linkage disequilibrium with rs4646 (r2 = 0.39 in the European 1000G phase 1 population). A recent work of Magnani et al
      • Magnani L.
      • Frige G.
      • Gadaleta R.M.
      • et al.
      Acquired CYP19A1 amplification is an early specific mechanism of aromatase inhibitor resistance in ERα metastatic breast cancer.
      highlighted how AI-resistant patients show acquired CYP19A1 amplification in their recurrent tumor. Amplification of this gene also occurred in vitro in AI-resistant models, showing a higher aromatase activity. In light of these results it seems that CYP19A1 amplification at the tumor level might be more important than germ line variations in determining the response to exemestane.
      It is well known that genotoxic estrogen metabolites might cause DNA damage.
      • Santen R.J.
      • Yue W.
      • Wang J.P.
      Estrogen metabolites and breast cancer.
      XRCC3 is involved in the DNA synthesis and repair pathways. In our study, patients carrying the XRCC3-rs861539TT genotype (241 Met/Met) had an increased risk of progression compared with the TC and CC genotypes. The 241 Met/Met variant was associated with a decreased DNA repair capacity
      • Matullo G.
      • Palli D.
      • Peluso M.
      • et al.
      XRCC1, XRCC3, XPD gene polymorphisms, smoking and (32)P-DNA adducts in a sample of healthy subjects.
      and it has been considered a biomarker of survival in MBC patients treated with DNA-damaging chemotherapy.
      • Bewick M.A.
      • Conlon M.S.
      • Lafrenie R.M.
      Polymorphisms in XRCC1, XRCC3, and CCND1 and survival after treatment for metastatic breast cancer.
      Similarly, patients carrying the XRCC3-rs861539TT genotype might be unable to repair genotoxic estrogen metabolite damage or other genotoxic insults, allowing cell proliferation and cancer progression.
      • Caldon C.E.
      Estrogen signaling and the DNA damage response in hormone dependent breast cancers.
      Steroid transporters have a critical role in tumor response to hormone therapy in BC.
      • Kiyotani K.
      • Mushiroda T.
      • Imamura C.K.
      • et al.
      Significant effect of polymorphisms in CYP2D6 and ABCC2 on clinical outcomes of adjuvant tamoxifen therapy for breast cancer patients.
      In our study we observed that polymorphisms in 2 genes (ABCG2 and SLCO1B1) encoding for steroid transporters, were associated with an increased PFS. Patients carrying at least 1 variant allele of either ABCG2-rs2046134 (A) or SLCO1B1-rs4149056 (C-174Ala) had a significantly reduced risk of progression compared with the wild type alleles. ABCG2 and SLCO1B1 genes encode for BC resistance protein (BCRP) and organic anion-transporting polypeptide 1B1, respectively. These are transporters also involved in the detoxification process of xenobiotic and antineoplastic drugs.
      • Mo W.
      • Zhang J.T.
      Human ABCG2: structure, function, and its role in multidrug resistance.
      • Niemi M.
      Role of OATP transporters in the disposition of drugs.
      Very recently the SLCO1B1-rs4149056 polymorphism has been associated with exemestane pharmacokinetics. In a study involving few healthy volunteers, women carrying at least 1 variant C allele showed a statistically significant higher area under the time/concentration curve for exemestane and its metabolite.
      • Gregory B.J.
      • Chen S.M.
      • Murphy M.A.
      • Atchley D.H.
      • Kamdem L.K.
      Impact of the OATP1B1 c.521T>C single nucleotide polymorphism on the pharmacokinetics of exemestane in healthy post-menopausal female volunteers.
      Moreover, SLCO1B1-rs4149056 is a predictor of statins and 3-hydroxy-3-methyl-glutaryl-coenzyme A reductase inhibitors disposition. Intriguingly, in our study, patients carrying at least 1 C allele had a prolonged PFS, probably as a result of an increased drug exposure because of the effect of this polymorphism. Regarding BCRP, at present, no data have been reported on its effect on exemestane. However, it must be considered that exemestane and its metabolites share similar chemical structures of the steroidal derivatives that are BCRP substrates.
      The FGFR is a transmembrane tyrosine kinase receptor involved in multiple biological processes, including cell proliferation, differentiation, and apoptosis. Previous reports indicate that endocrine resistance involves a cross-talk between growth factor pathways and estrogen signaling.
      • Dowsett M.
      • Martin L.A.
      • Smith I.
      • Johnston S.
      Mechanisms of resistance to aromatase inhibitors.
      • Tomiguchi M.
      • Yamamoto Y.
      • Yamamoto-Ibusuki M.
      • et al.
      Fibroblast growth factor receptor-1 protein expression is associated with prognosis in estrogen receptor-positive/human epidermal growth factor receptor-2-negative primary breast cancer.
      In BC, aberrant FGFR signaling, including the FGFR4-rs351855 polymorphism (Gly388Arg), has been involved in tumor progression and resistance to tamoxifen.
      • André F.
      • Cortés J.
      Rationale for targeting fibroblast growth factor receptor signaling in breast cancer.
      In vitro reports showed an increased motility of mammary cells with 388 Arg/Arg variant and also an increased extracellular matrix degradation.
      • André F.
      • Cortés J.
      Rationale for targeting fibroblast growth factor receptor signaling in breast cancer.
      According to this detrimental effect, we observed that patients carrying FGFR4-rs351855 homozygous variant genotype (TT-388 Arg/Arg) were at a significantly increased risk of progression compared with TC and CC genotypes.
      The 6 risk genotypes significantly associated with a shorter PFS (CYP19A1-rs10046TC/CC, SLCO1B1-rs4149056TT, ABCG2-rs2046134GG, FGFR4-rs351855TT, and the XRCC3-rs861539TT) were combined into a genetic risk score, allowing the definition of 4 risk groups of patients. As a result, we obtained a better stratification of patients according to their PFS, with a substantial improvement in the predictive ability of the score compared with the individual polymorphisms. Patients carrying 2, 3, and 4 to 6 risk points had a 40%, 100%, and more than 200% increased risk of progression, respectively, than patients with 0 to 1 risk points. We observed an overall not statistically significant trend for the association of OS with the risk score. The fact that we have not taken into account treatments after progression, which can affect OS, might explain this lack of significance. However, when considering only the groups with the highest and lowest risk points (4-6 vs. 0-1), the difference in OS became statistically significant, indicating that this score might be helpful in OS prediction for at least these extreme groups of patients, allowing identification of the ones at the highest and the lowest risk of death.
      Another limitation of this study is that an external validation cohort is missing, due to the difficulty of finding a prospective study with similar characteristics. However, the bootstrap analysis, consisting in the replication of the findings by drawing 1000 samples from the original data set, allowed an internal assessment of reproducibility. Moreover, this study could evaluate only the prognostic role of the germ line polymorphisms. The assessment of the predictive value of the biomarkers could not be established, because of the lack of a control arm with patients not treated with exemestane.
      To the best of our knowledge, this is the largest prospective study specifically designed to evaluate pharmacogenetic biomarkers of PFS and OS in patients treated with exemestane. In addition, the long follow-up (median: almost 3 years) is another point of strength of the study.

      Conclusion

      Our findings show that germ line polymorphisms in hormone balance, drug activity, DNA replication and repair, as well as in signaling pathways are associated with PFS and OS of exemestane-treated patients. The joint effect of polymorphisms from multiple pathways included in a multifactorial genetic score might better define groups of patients with different prognoses. Replication studies, in external cohorts of patients, are nonetheless required to finally ascertain the clinical utility of these markers.

       Clinical Practice Points

      • Several studies investigated the role of genetic variations in genes involved in exemestane pharmacokinetics, pharmacodynamics, and hormone balance pathways. Even if several associations have been found with exemestane outcome, no consensus has been reached on polymorphisms to be translated into clinical practice as predictive or prognostic biomarkers.
      • This study highlighted 5 polymorphisms on genes involved in AI pathways and also in DNA repair pathways. The combination of these 5 polymorphisms into a score allow better stratification of patients according to their PFS and OS.
      • The use of this pharmacogenetic score might help, through a simple blood test, to identify patients who might require a different or more aggressive therapeutic approach because of their higher risk of progression and/or death.

      Disclosure

      The authors declare the following conflicts of interests: National Cancer Institute Centro di Riferimento Oncologico di Aviano, IRCCS, Aviano, Italy, received a grant from Pfizer Italia (Italy) (CUP: J31J06000270007 ) during the time of this study.

      Acknowledgments

      The authors thank the patients, institutions and clinicians who participated in this study. The authors thank Dr Anna Scalvini, Dr Daniela Quitadamo, Dr Barbara Venturini, and Dr Elisa Perfetti, for the data management, and also Franca Sartor and Loredana Romanato for the laboratory support.

      Supplemental Data

      Supplemental Table 1List of the Analyzed Polymorphisms
      PathwayGenersIDPolymorphismPatients, nVariant Allele%
      Estrogen SynthesisCYP19A1rs10046T>C302C48
      Estrogen SynthesisCYP19A1rs60271534(TTTA) 7-13302L50
      Estrogen SynthesisCYP19A1rs4646C>A302A29
      Estrogen SynthesisCYP19A1rs700519C>T302T3
      Estrogen SynthesisCYP17A1rs743572A>G302G42
      Estrogen ActivityESR1rs9340799A>G302G38
      Estrogen ActivityESR1rs2234693T>C302C46
      Estrogen ActivityESR2rs1256049G>A302A2
      Estrogen ActivityESR2rs4986938G>A302A41
      Estrogen ActivityPRDM2rs2308040D>I302I37
      Estrogen MetabolismCOMTrs4680A>G302A50
      Estrogen MetabolismCYP1B1rs1056836C>G302G44
      Estrogen MetabolismUGT1A1rs8175347(TA) 5-8300L35
      Estrogen MetabolismCYP2C9rs1799853C>T301T11
      Exemestane MetabolismCYP3A4rs2740574A>G302G2
      Exemestane MetabolismCYP3A5rs776746G>A302A5
      Steroids Transport, MDRABCB1rs10276036T>C299C43
      Steroids Transport, MDRABCB1rs2235013A>G300G48
      Steroids Transport, MDRABCB1rs2235015G>T299T19
      Steroids Transport, MDRABCB1rs2235033C>T299T48
      Steroids Transport, MDRABCB1rs3213619T>C300C4
      Steroids Transport, MDRABCB1rs3842A>G298G14
      Steroids Transport, MDRABCC1rs2074087G>C301C17
      Steroids Transport, MDRABCC1rs212088C>T300T16
      Steroids Transport, MDRABCC1rs2230671G>A302A23
      Steroids Transport, MDRABCC1rs35587T>C299C33
      Steroids Transport, MDRABCC1rs35588A>G302G31
      Steroids Transport, MDRABCC1rs35605C>T302T19
      Steroids Transport, MDRABCC1rs3765129C>T299T14
      Steroids Transport, MDRABCC1rs4148356G>A302A0
      Steroids Transport, MDRABCC1rs60782127G>T302T1
      Biliary Transport, MDRABCC2rs17216177T>C302C7
      Biliary Transport, MDRABCC2rs2002042C>T302T23
      Biliary Transport, MDRABCC2rs2273697G>A301A19
      Biliary Transport, MDRABCC2rs3740066G>A302A37
      Biliary Transport, MDRABCC2rs4148396C>T302T38
      Biliary Transport, MDRABCC2rs717620G>A302A18
      Biliary Transport, MDRABCC2rs8187710G>A302A7
      Steroids Transport, BCRPABCG2rs2046134G>A299A5
      Steroids Transport, BCRPABCG2rs2231142C>A302A9
      Steroids Transport, BCRPABCG2rs2622604C>T299T22
      Steroids Transport, BCRPABCG2rs3219191D>I299I45
      Steroids TransportSLCO1B1rs4149056T>C300C16
      Cell Cycle, DNA RepairATMrs1801516G>A302A14
      Cell Cycle, DNA RepairCDKN1Ars1801270C>A302A7
      Cell Cycle, DNA RepairMDM4rs4245739A>C299C30
      Cell Cycle, SignalingFGFR4rs351855C>T301T27
      DNA RepairAPEX1rs1130409T>G302G46
      DNA RepairMSH6rs3136228T>G301G37
      DNA RepairOGG1rs1052133C>G302G21
      DNA RepairERCC5rs17655C>G302G22
      DNA RepairXRCC1rs1799782C>T302T6
      DNA RepairXRCC1rs25487G>A302A33
      DNA RepairXRCC1rs25489G>A300A6
      DNA RepairXRCC3rs1799794A>G302G22
      DNA RepairXRCC3rs1799796A>G301G27
      DNA RepairXRCC3rs861539C>T302T42
      DNA Repair, NERERCC1rs11615T>C302C40
      DNA Repair, NERERCC1rs3212986G>T302T27
      DNA Repair, NERERCC2rs13181T>G301G44
      DNA SynthesisATICrs2372536C>G302G36
      DNA Synthesis, Folate CycleFOLR1rs2071010G>A302A6
      DNA Synthesis, Folate CycleFOLR1rs9282688C>T302T2
      DNA Synthesis, Folate CycleFPGSrs10106A>G302G37
      DNA Synthesis, Folate CycleGGHrs11545078C>T301T11
      DNA Synthesis, Folate CycleMTHFD1rs2236225C>T302T43
      DNA Synthesis, Folate CycleMTHFRrs1801131A>C302C32
      DNA Synthesis, Folate CycleMTRrs1805087A>G301G19
      DNA Synthesis, Folate CycleMTRRrs1801394A>G301G49
      DNA Synthesis, Folate CycleSHMT1rs2273029C>T302T25
      DNA Synthesis, Folate CycleTYMSrs16430I>D302D38
      DNA Synthesis, Folate CycleTYMSrs2790A>G302G25
      DNA Synthesis, Folate CycleTYMSrs699517C>T301T38
      TP53 SignalingTP53rs1042522G>C302C29
      List of the analyzed polymorphisms with the pathway they are involved in, the number of patients genotyped, and the frequencies of the variant alleles. Underlined, polymorphisms in linkage disequilibrium, according to r2 threshold = 0.8. For the polymorphisms CYP19A1-rs60271534 and for UGT1A1-rs8175347 alleles were associated into 2 groups, long (L) and short (S) alleles: L ≥ 7 TTTA, S < 7 TTTA repeats, and L = 7-8 TA, S = 5-6 TA repeats, respectively.
      Abbreviations: BCRP = breast cancer resistance protein; D = deletion; I = insertion; MDR = multidrug resistance; NER = nucleotide excision repair; rsID = reference single-nucleotide polymorphism identification number; TP53 = tumor protein 53.
      Supplemental Table 2Role, Function, and Associations Found in the Literature for the Polymorphisms Significantly Associated With PFS
      GeneSNP (Amino Acid Change)LocationPredicted or Functional RoleCancer TypeTreatmentEnd PointPatient nAssociation With the Variant Allele or GeneReference
      CYP19A1rs100463′ UTR↓ Estrogen levelsHealthy subjectsNoneEstrogen levels1975↓ Estrogen levels
      BCAdjuvant LETEstrogen levels204None
      • Fasching P.A.
      • Loehberg C.R.
      • Strissel P.L.
      • et al.
      Single nucleotide polymorphisms of the aromatase gene (CYP19A1), HER2/neu status, and prognosis in breast cancer patients.
      BCNRDFSTotal 1257, premenopausal 439↑ DFS (premenopausal)
      • Leyland-Jones B.
      • Gray K.P.
      • Abramovitz M.
      • et al.
      CYP19A1 polymorphisms and clinical outcomes in postmenopausal women with hormone receptor-positive breast cancer in the BIG 1-98 trial.
      BCTAM and/or LETBone AEs4861↑ Bone AE risk
      • Pineda B.
      • García-Pérez M.Á.
      • Cano A.
      • Lluch A.
      • Eroles P.
      Associations between aromatase CYP19 rs10046 polymorphism and breast cancer risk: from a case-control to a meta-analysis of 20,098 subjects.
      BCNRBC Risk522 Cases/1221 controls↑ BC risk
      • Clendenen T.
      • Zeleniuch-Jacquotte A.
      • Wirgin I.
      • et al.
      Genetic variants in hormone-related genes and risk of breast cancer.
      BCNRBC Risk20,098 (meta-analysis)None
      • Clendenen T.
      • Zeleniuch-Jacquotte A.
      • Wirgin I.
      • et al.
      Genetic variants in hormone-related genes and risk of breast cancer.
      BCNRBC Risk1164 Cases/2111 controlsNone
      • Colomer R.
      • Monzo M.
      • Tusquets I.
      • et al.
      A single-nucleotide polymorphism in the aromatase gene is associated with the efficacy of the aromatase inhibitor letrozole in advanced breast carcinoma.
      BCLET; ANATTP67,272None
      • Liu L.
      • Bai Y.X.
      • Zhou J.H.
      • et al.
      A polymorphism at the 3′-UTR region of the aromatase gene is associated with the efficacy of the aromatase inhibitor, anastrozole, in metastatic breast carcinoma.
      • Colomer R.
      • Monzo M.
      • Tusquets I.
      • et al.
      A single-nucleotide polymorphism in the aromatase gene is associated with the efficacy of the aromatase inhibitor letrozole in advanced breast carcinoma.
      BCNeoadjuvant LETResponse95None
      • Garcia-Casado Z.
      • Guerrero-Zotano A.
      • Llombart-Cussac A.
      • et al.
      A polymorphism at the 3′-UTR region of the aromatase gene defines a subgroup of postmenopausal breast cancer patients with poor response to neoadjuvant letrozole.
      BCANA, LET, EXETTP, AEsNRNone
      • Artigalás O.
      • Vanni T.
      • Hutz M.H.
      • Ashton-Prolla P.
      • Schwartz I.V.
      Influence of CYP19A1 polymorphisms on the treatment of breast cancer with aromatase inhibitors: a systematic review and meta-analysis.
      BCAIsOS53↑ OS with WT allele
      • Miron L.
      • Negură L.
      • Peptanariu D.
      • Marinca M.
      Research on aromatase gene (CYP19A1) polymorphisms as a predictor of endocrine therapy effectiveness in breast cancer.
      SLCO1B1rs4149056 (Val174Ala)Exonic↓ Transport activityHealthy subjectsNoneActivityNR↓ Transport activity
      • Tirona R.G.
      • Leake B.F.
      • Merino G.
      • Kim R.B.
      Polymorphisms in OATP-C: identification of multiple allelic variants associated with altered transport activity among European- and African-Americans.
      BC cell lineNAActivityNA↓ Transport activity
      • Pu Z.
      • Zhang X.
      • Chen Q.
      • Yuan X.
      • Xie H.
      Establishment of an expression platform of OATP1B1 388GG and 521CC genetic polymorphism and the therapeutic effect of tamoxifen in MCF-7 cells.
      BCTAMOS296↓ OS
      • Zhang X.
      • Pu Z.
      • Ge J.
      • Shen J.
      • Yuan X.
      • Xie H.
      Association of CYP2D6*10, OATP1B1 A388G, and OATP1B1 T521C polymorphisms and overall survival of breast cancer patients after tamoxifen therapy.
      FGFR4rs351855 (Gly388Arg)Exonic↑ Tumor cell motilityBC cell lineNACell motilityNA↑ Tumor cell motility
      • Bange J.
      • Prechtl D.
      • Cheburkin Y.
      • et al.
      Cancer progression and tumor cell motility are associated with the FGFR4 Arg(388) allele.
      BCNone, CMF, and/or TAMDFS84↓ DFS
      • Bange J.
      • Prechtl D.
      • Cheburkin Y.
      • et al.
      Cancer progression and tumor cell motility are associated with the FGFR4 Arg(388) allele.
      PC cell lineNATumor invasionNA↑ ECM degradation
      • Sugiyama N.
      • Varjosalo M.
      • Meller P.
      • et al.
      Fibroblast growth factor receptor 4 regulates tumor invasion by coupling fibroblast growth factor signaling to extracellular matrix degradation.
      BC transgenic mouseNABC progressionNA↑ Tumor progression
      • Seitzer N.
      • Mayr T.
      • Streit S.
      • Ullrich A.
      A single nucleotide change in the mouse genome accelerates breast cancer progression.
      Multiple, including BCNAOS2537 (Pooled analysis)↓ OS
      • Miron L.
      • Negură L.
      • Peptanariu D.
      • Marinca M.
      Research on aromatase gene (CYP19A1) polymorphisms as a predictor of endocrine therapy effectiveness in breast cancer.
      BCAdjuvant CMF and/or TAMDFS/OS372↓ DFS/OS
      • Frullanti E.
      • Berking C.
      • Harbeck N.
      • et al.
      Meta and pooled analyses of FGFR4 Gly388Arg polymorphism as a cancer prognostic factor.
      BCTAMCB/PFS285↓ CB/DFS
      • Thussbas C.
      • Nahrig J.
      • Streit S.
      • et al.
      FGFR4 Arg388 allele is associated with resistance to adjuvant therapy in primary breast cancer.
      ABCG2rs2046134IntronicTranscription factor binding site (predicted)
      Predicted by ConSite.
      Tissue from healthy donors/GIST

      AI-resistant cells/xenograft tumors
      Imatinib

      LET, EXE
      Functional effect

      AI resistance
      44 + 60 + 28 Tissues/82 GIST patients↑ Protein expression

      Involvement of ABCG2 in AI resistance
      ConSite
      • Meijer D.
      • Sieuwerts A.M.
      • Look M.P.
      • van Agthoven T.
      • Foekens J.A.
      • Dorssers L.C.
      Fibroblast growth factor receptor 4 predicts failure on tamoxifen therapy in patients with recurrent breast cancer.
      • Poonkuzhali B.
      • Lamba J.
      • Strom S.
      • et al.
      Association of breast cancer resistance protein/ABCG2 phenotypes and novel promoter and intron 1 single nucleotide polymorphisms.
      • Gilani R.A.
      • Kazi A.A.
      • Shah P.
      • et al.
      The importance of HER2 signaling in the tumor-initiating cell population in aromatase inhibitor-resistant breast cancer.
      ,
      Predicted by ConSite.
      XRCC3rs861539 (Thr241Met)ExonicPossibly damaging (predicted)
      Predicted by PolyPhen-2.
      GISTImatinibOS81↓ OSPolyphen-2
      • Kazi A.A.
      • Gilani R.A.
      • Schech A.J.
      • et al.
      Nonhypoxic regulation and role of hypoxia-inducible factor 1 in aromatase inhibitor resistant breast cancer.
      ,
      Predicted by ConSite.
      BCAnthracyclinesOS150↑ OS
      • Ravegnini G.
      • Nannini M.
      • Simeon V.
      • et al.
      Polymorphisms in DNA repair genes in gastrointestinal stromal tumours: susceptibility and correlation with tumour characteristics and clinical outcome.
      BCNRBC risk70 Cases/70 controlsBC risk
      • Castro E.
      • Olmos D.
      • Garcia A.
      • Cruz J.J.
      • González-Sarmiento R.
      Role of XRCC3, XRCC1 and XPD single-nucleotide polymorphisms in survival outcomes following adjuvant chemotherapy in early stage breast cancer patients.
      BCNRBC risk19,575 cases/21,125 controls↑ BC risk
      • Smolarz B.
      • Makowska M.
      • Samulak D.
      • et al.
      Association between single nucleotide polymorphisms (SNPs) of XRCC2 and XRCC3 homologous recombination repair genes and triple-negative breast cancer in Polish women.
      The literature analysis was focused on the identification of the role of the SNPs with cancer risk, toxicities, patients’ prognosis, and cellular transformation. In case of no functional data available in literature, we referred to the results obtained with bioinformatic analysis exploiting ConSite (http://compbio.cs.queensu.ca/F-SNP) and PolyPhen (http://genetics.bwh.harvard.edu/ggi/pph2/e8dbeaa52a8642d83df5575a0830d51a00e71f38/5502954.html).
      Abbreviations: AE = adverse event; AI = aromatase inhibitor; ANA = anastrozole; BC = breast cancer; CB = clinical benefit; CMF = cyclophosphamide/methotrexate/fluorouracil; DFS = disease free survival; ECM = extracellular matrix; EXE = exemestane; GIST = gastrointestinal stromal tumor; LET = letrozole; OS = overall survival; PC = prostate cancer; PFS = progression-free survival; SNP = single nucleotide polymorphism; TAM = tamoxifen; TTP = time to progression; UTR = untranslated region; WT = wild type.
      a Predicted by ConSite.
      b Predicted by PolyPhen-2.

      References

        • Smith I.E.
        • Dowsett M.
        Aromatase inhibitors in breast cancer.
        N Engl J Med. 2003; 348: 2431-2442
        • Coombes R.C.
        • Kilburn L.S.
        • Snowdon C.F.
        • et al.
        Survival and safety of exemestane versus tamoxifen after 2-3 years’ tamoxifen treatment (Intergroup Exemestane Study): a randomised controlled trial.
        Lancet. 2007; 369: 559-570
        • De Placido S.
        • Gallo C.
        • De Laurentiis M.
        • et al.
        Adjuvant anastrozole versus exemestane versus letrozole, upfront or after 2 years of tamoxifen, in endocrine-sensitive breast cancer (FATA-GIM3): a randomised, phase 3 trial.
        Lancet Oncol. 2018; 19: 474-485
        • Paridaens R.J.
        • Dirix L.Y.
        • Beex L.V.
        • et al.
        Phase III study comparing exemestane with tamoxifen as first-line hormonal treatment of metastatic breast cancer in postmenopausal women: the European Organisation for Research and Treatment of Cancer Breast Cancer Cooperative Group.
        J Clin Oncol. 2008; 26: 4883-4890
        • Asten K.V.
        • Neven P.
        • Lintermans A.
        • Wildiers H.
        • Paridaens R.
        Aromatase inhibitors in the breast cancer clinic: focus on exemestane.
        Endocr Relat Cancer. 2014; 21: R31-49
        • Mauri D.
        • Pavlidis N.
        • Polyzos N.P.
        • Ioannidis J.P.
        Survival with aromatase inhibitors and inactivators versus standard hormonal therapy in advanced breast cancer: meta-analysis.
        J Natl Cancer Inst. 2006; 98: 1285-1291
        • Musgrove E.A.
        • Sutherland R.L.
        Biological determinants of endocrine resistance in breast cancer.
        Nat Rev Cancer. 2009; 9: 631-643
        • Osborne C.K.
        • Schiff R.
        Mechanisms of endocrine resistance in breast cancer.
        Annu Rev Med. 2011; 62: 233-247
        • Dowsett M.
        • Martin L.A.
        • Smith I.
        • Johnston S.
        Mechanisms of resistance to aromatase inhibitors.
        J Steroid Biochem Mol Biol. 2005; 95: 167-172
        • Oesterreich S.
        • Henry N.L.
        • Kidwell K.M.
        • et al.
        Associations between genetic variants and the effect of letrozole and exemestane on bone mass and bone turnover.
        Breast Cancer Res Treat. 2015; 154: 263-273
        • Peterson A.
        • Xia Z.
        • Chen G.
        • Lazarus P.
        Exemestane potency is unchanged by common nonsynonymous polymorphisms in CYP19A1: results of a novel anti-aromatase activity assay examining exemestane and its derivatives.
        Pharmacol Res Perspect. 2017; 5: e00313
        • Ferraldeschi R.
        • Arnedos M.
        • Hadfield K.D.
        • et al.
        Polymorphisms of CYP19A1 and response to aromatase inhibitors in metastatic breast cancer patients.
        Breast Cancer Res Treat. 2012; 133: 1191-1198
        • Ingle J.N.
        • Schaid D.J.
        • Goss P.E.
        • et al.
        Genome-wide associations and functional genomic studies of musculoskeletal adverse events in women receiving aromatase inhibitors.
        J Clin Oncol. 2010; 28: 4674-4682
        • Dunning A.M.
        • Dowsett M.
        • Healey C.S.
        • et al.
        Polymorphisms associated with circulating sex hormone levels in postmenopausal women.
        J Natl Cancer Inst. 2004; 96: 936-945
        • Tworoger S.S.
        • Chubak J.
        • Aiello E.J.
        • et al.
        Association of CYP17, CYP19, CYP1B1, and COMT polymorphisms with serum and urinary sex hormone concentrations in postmenopausal women.
        Cancer Epidemiol Biomark Prev. 2004; 13: 94-101
        • Henry N.L.
        • Skaar T.C.
        • Dantzer J.
        • et al.
        Genetic associations with toxicity-related discontinuation of aromatase inhibitor therapy for breast cancer.
        Breast Cancer Res Treat. 2013; 138: 807-816
        • Grundberg E.
        • Akesson K.
        • Kindmark A.
        • et al.
        The impact of estradiol on bone mineral density is modulated by the specific estrogen receptor-alpha cofactor retinoblastoma-interacting zinc finger protein-1 insertion/deletion polymorphism.
        J Clin Endocrinol Metab. 2007; 92: 2300-2306
        • Kuo S.H.
        • Yang S.Y.
        • You S.L.
        • et al.
        Polymorphisms of ESR1, UGT1A1, HCN1, MAP3K1 and CYP2B6 are associated with the prognosis of hormone receptor-positive early breast cancer.
        Oncotarget. 2017; 8: 20925-20938
        • Abubakar M.B.
        • Wei K.
        • Gan S.H.
        The influence of genetic polymorphisms on the efficacy and side effects of anastrozole in postmenopausal breast cancer patients.
        Pharmacogenet Genomics. 2014; 24: 575-581
        • Hertz D.L.
        • Kidwell K.M.
        • Seewald N.J.
        • et al.
        Polymorphisms in drug-metabolizing enzymes and steady-state exemestane concentration in postmenopausal patients with breast cancer.
        Pharmacogenomics J. 2017; 17: 521-527
        • van der Deure W.M.
        • Friesema E.C.
        • de Jong F.J.
        • et al.
        Organic anion transporter 1B1: an important factor in hepatic thyroid hormone and estrogen transport and metabolism.
        Endocrinology. 2008; 149: 4695-4701
        • Turkistani A.
        • Marsh S.
        Pharmacogenomics of third-generation aromatase inhibitors.
        Expert Opin Pharmacother. 2012; 13: 1299-1307
        • Matta J.
        • Morales L.
        • Ortiz C.
        • et al.
        Estrogen receptor expression is associated with DNA repair capacity in breast cancer.
        PLoS One. 2016; 11: e015242
        • Ellis M.J.
        • Ding L.
        • Shen D.
        • et al.
        Whole-genome analysis informs breast cancer response to aromatase inhibition.
        Nature. 2012; 486: 353-360
        • Rogers W.
        Regression standard errors in clustered samples.
        (Available at:)
        • Artigalás O.
        • Vanni T.
        • Hutz M.H.
        • Ashton-Prolla P.
        • Schwartz I.V.
        Influence of CYP19A1 polymorphisms on the treatment of breast cancer with aromatase inhibitors: a systematic review and meta-analysis.
        BMC Med. 2015; 13: 139
        • Magnani L.
        • Frige G.
        • Gadaleta R.M.
        • et al.
        Acquired CYP19A1 amplification is an early specific mechanism of aromatase inhibitor resistance in ERα metastatic breast cancer.
        Nat Genet. 2017; 49: 444-450
        • Santen R.J.
        • Yue W.
        • Wang J.P.
        Estrogen metabolites and breast cancer.
        Steroids. 2015; 99: 61-66
        • Matullo G.
        • Palli D.
        • Peluso M.
        • et al.
        XRCC1, XRCC3, XPD gene polymorphisms, smoking and (32)P-DNA adducts in a sample of healthy subjects.
        Carcinogenesis. 2001; 22: 1437-1445
        • Bewick M.A.
        • Conlon M.S.
        • Lafrenie R.M.
        Polymorphisms in XRCC1, XRCC3, and CCND1 and survival after treatment for metastatic breast cancer.
        J Clin Oncol. 2006; 24: 5645-5651
        • Caldon C.E.
        Estrogen signaling and the DNA damage response in hormone dependent breast cancers.
        Front Oncol. 2014; 4: 106
        • Kiyotani K.
        • Mushiroda T.
        • Imamura C.K.
        • et al.
        Significant effect of polymorphisms in CYP2D6 and ABCC2 on clinical outcomes of adjuvant tamoxifen therapy for breast cancer patients.
        J Clin Oncol. 2010; 28: 1287-1293
        • Mo W.
        • Zhang J.T.
        Human ABCG2: structure, function, and its role in multidrug resistance.
        Int J Biochem Mol Biol. 2011; 3: 1-27
        • Niemi M.
        Role of OATP transporters in the disposition of drugs.
        Pharmacogenomics. 2007; 8: 787-802
        • Gregory B.J.
        • Chen S.M.
        • Murphy M.A.
        • Atchley D.H.
        • Kamdem L.K.
        Impact of the OATP1B1 c.521T>C single nucleotide polymorphism on the pharmacokinetics of exemestane in healthy post-menopausal female volunteers.
        J Clin Pharm Ther. 2017; 42: 547-553
        • Tomiguchi M.
        • Yamamoto Y.
        • Yamamoto-Ibusuki M.
        • et al.
        Fibroblast growth factor receptor-1 protein expression is associated with prognosis in estrogen receptor-positive/human epidermal growth factor receptor-2-negative primary breast cancer.
        Cancer Sci. 2016; 107: 491-498
        • André F.
        • Cortés J.
        Rationale for targeting fibroblast growth factor receptor signaling in breast cancer.
        Breast Cancer Res Treat. 2015; 150: 1-8