Individual participant data meta-analysis for external validation, recalibration, and updating of a flexible parametric prognostic model.

Individual participant data (IPD) from multiple sources allows external validation of a prognostic model across multiple populations. Often this reveals poor calibration, potentially causing poor predictive performance in some populations. However, rather than discarding the model outright, it may be possible to modify the model to improve performance using recalibration techniques. We use IPD meta-analysis to identify the simplest method to achieve good model performance. We examine four options for recalibrating an existing time-to-event model across multiple populations: (i) shifting the baseline hazard by a constant, (ii) re-estimating the shape of the baseline hazard, (iii) adjusting the prognostic index as a whole, and (iv) adjusting individual predictor effects. For each strategy, IPD meta-analysis examines (heterogeneity in) model performance across populations. Additionally, the probability of achieving good performance in a new population can be calculated allowing ranking of recalibration methods. In an applied example, IPD meta-analysis reveals that the existing model had poor calibration in some populations, and large heterogeneity across populations. However, re-estimation of the intercept substantially improved the expected calibration in new populations, and reduced between-population heterogeneity. Comparing recalibration strategies showed that re-estimating both the magnitude and shape of the baseline hazard gave the highest predicted probability of good performance in a new population. In conclusion, IPD meta-analysis allows a prognostic model to be externally validated in multiple settings, and enables recalibration strategies to be compared and ranked to decide on the least aggressive recalibration strategy to achieve acceptable external model performance without discarding existing model information.

Ferritin heavy chain in triple negative breast cancer: a favorable prognostic marker that relates to a cluster of differentiation 8 positive (CD8+) effector T-cell response.

Ferritin heavy chain (FTH1) is a 21-kDa subunit of the ferritin complex, known for its role in iron metabolism, and which has recently been identified as a favorable prognostic protein for triple negative breast cancer (TNBC) patients. Currently, it is not well understood how FTH1 contributes to an anti-tumor response. Here, we explored whether expression and cellular compartmentalization of FTH1 correlates to an effective immune response in TNBC patients. Analysis of the tumor tissue transcriptome, complemented with in silico pathway analysis, revealed that FTH1 was an integral part of an immunomodulatory network of cytokine signaling, adaptive immunity, and cell death. These findings were confirmed using mass spectrometry (MS)-derived proteomic data, and immunohistochemical staining of tissue microarrays. We observed that FTH1 is localized in both the cytoplasm and/or nucleus of cancer cells. However, high cytoplasmic (c) FTH1 was associated with favorable prognosis (Log-rank p = 0.001), whereas nuclear (n) FTH1 staining was associated with adverse prognosis (Log-rank p = 0.019). cFTH1 staining significantly correlated with total FTH1 expression in TNBC tissue samples, as measured by MS analysis (Rs = 0.473, p = 0.0007), but nFTH1 staining did not (Rs = 0.197, p = 0.1801). Notably, IFN γ-producing CD8+ effector T cells, but not CD4+ T cells, were preferentially enriched in tumors with high expression of cFTH1 (p = 0.02). Collectively, our data provide evidence toward new immune regulatory properties of FTH1 in TNBC, which may facilitate development of novel therapeutic targets.

Growth and metastatic behavior of molecularly well-characterized human breast cancer cell lines in mice.

Breast cancer (BC) is a disease with intra- and inter-tumor heterogeneity, and models representing the complete variety of clinical BC phenotypes are not available. We explored the tumor growth potential and metastatic behavior of human BC cell lines and determined whether these cell lines can recapitulate subtype-related biological characteristics of tumors. Eighteen human BC cell lines were implanted under the mammary fat pad of nude mice. Subtype-specific differences in tumor growth, metastatic ability to distant sites, and tumor-related survival of mice were recorded. Eighty-nine percent of the cell lines gave rise to xenografts of which 56 % showed metastasis to distant sites. A clear difference was observed in growth of xenografts from cell lines of different molecular subtypes (P = 0.001; Kruskal-Wallis test). Mice bearing the basal-like and the normal-like xenografts showed poor tumor-related survival (HR: 10.50; P = 0.002 and HR: 9.89; P = 0.003, respectively) compared with those bearing the ERBB2-positive xenografts, which had the longest survival. Subtype-specific metastasis to distant sites between xenografts was not however observed. Comparable to clinical behavior in humans, we observed that the basal-like and the normal-like cell lines grew more aggressively in mice than the cell lines of other molecular subtypes. However, in contrast to clinical findings, we observed no relationships between intrinsic subtype and preferences for site of relapse. Importantly, we have established xenograft models from 16 phenotypically and molecularly diverse human BC cell lines, which can be exploited as useful tools to perform functional studies and screening of interfering drugs.

Multilevel mixed effects parametric survival models using adaptive Gauss-Hermite quadrature with application to recurrent events and individual participant data meta-analysis.

Multilevel mixed effects survival models are used in the analysis of clustered survival data, such as repeated events, multicenter clinical trials, and individual participant data (IPD) meta-analyses, to investigate heterogeneity in baseline risk and covariate effects. In this paper, we extend parametric frailty models including the exponential, Weibull and Gompertz proportional hazards (PH) models and the log logistic, log normal, and generalized gamma accelerated failure time models to allow any number of normally distributed random effects. Furthermore, we extend the flexible parametric survival model of Royston and Parmar, modeled on the log-cumulative hazard scale using restricted cubic splines, to include random effects while also allowing for non-PH (time-dependent effects). Maximum likelihood is used to estimate the models utilizing adaptive or nonadaptive Gauss-Hermite quadrature. The methods are evaluated through simulation studies representing clinically plausible scenarios of a multicenter trial and IPD meta-analysis, showing good performance of the estimation method. The flexible parametric mixed effects model is illustrated using a dataset of patients with kidney disease and repeated times to infection and an IPD meta-analysis of prognostic factor studies in patients with breast cancer. User-friendly Stata software is provided to implement the methods.

Hotspot mutations in PIK3CA associate with first-line treatment outcome for aromatase inhibitors but not for tamoxifen.

PIK3CA mutations occur frequently in breast cancer, predominantly in exons 9 and 20. The aim of this retrospective study is to evaluate the PIK3CA mutation status for its relationship with prognosis and first-line endocrine therapy outcome. PIK3CA exon 9 and 20 were evaluated for mutations in 1,352 primary breast cancer specimens by SnaPshot multiplex analyses. The mutation status was studied for their relationship with metastasis-free survival (MFS) in 342 untreated lymph node-negative (LNN) patients and to time to progression (TTP) in estrogen receptor (ER)-positive patients with metastatic disease treated with first-line tamoxifen (N = 447) or aromatase inhibitors (AIs; N = 84). We detected in 423 patients hotspot mutations for PIK3CA (31 %). Mutations in exon 20 were detected in 251 patients (59 %), with H1047L and H1047R mutations in 37 (15 %) and 214 (85 %) cases, respectively. Mutations in PIK3CA exon 9 were discovered in 173 patients (41 %), with E542K and E545K mutations in 57 (32 %) and 104 (60 %) cases as most prevalent ones. Evaluation of the untreated LNN patients for prognosis showed no relationship between MFS and PIK3CA mutations, neither for exon 9 [HR = 1.04 (95 % CI 0.57-1.89), P = 0.90] nor for exon 20 [HR = 0.98 (95 % CI 0.63-1.54); P = 0.94] when compared to wild-type. The PIK3CA mutation status was also not associated with treatment outcome after first-line tamoxifen. On the other hand, patients treated with first-line AIs showed a longer TTP when having a PIK3CA mutation in exon 9 [HR = 0.40 (95 % CI 0.17-0.95); P = 0.038] or exon 20 [HR = 0.50 (95 % CI 0.27-0.91); P = 0.024] compared to wild-types, both significant in uni- and multivariate analysis including traditional predictive factors. All results remained when only HER2-negative patients were evaluated for each cohort. PIK3CA mutations in ER-positive tumors were significantly associated with a favorable outcome after first-line AIs, which needs further confirmation in other datasets. Mutations were not associated with prognosis in untreated LNN patients nor predictive outcome after first-line tamoxifen therapy in advanced disease patients.

Ancestry-shift refinement mapping of the C6orf97-ESR1 breast cancer susceptibility locus.

We used an approach that we term ancestry-shift refinement mapping to investigate an association, originally discovered in a GWAS of a Chinese population, between rs2046210[T] and breast cancer susceptibility. The locus is on 6q25.1 in proximity to the C6orf97 and estrogen receptor alpha (ESR1) genes. We identified a panel of SNPs that are correlated with rs2046210 in Chinese, but not necessarily so in other ancestral populations, and genotyped them in breast cancer case:control samples of Asian, European, and African origin, a total of 10,176 cases and 13,286 controls. We found that rs2046210[T] does not confer substantial risk of breast cancer in Europeans and Africans (OR = 1.04, P = 0.099, and OR = 0.98, P = 0.77, respectively). Rather, in those ancestries, an association signal arises from a group of less common SNPs typified by rs9397435. The rs9397435[G] allele was found to confer risk of breast cancer in European (OR = 1.15, P = 1.2 x 10(-3)), African (OR = 1.35, P = 0.014), and Asian (OR = 1.23, P = 2.9 x 10(-4)) population samples. Combined over all ancestries, the OR was 1.19 (P = 3.9 x 10(-7)), was without significant heterogeneity between ancestries (P(het) = 0.36) and the SNP fully accounted for the association signal in each ancestry. Haplotypes bearing rs9397435[G] are well tagged by rs2046210[T] only in Asians. The rs9397435[G] allele showed associations with both estrogen receptor positive and estrogen receptor negative breast cancer. Using early-draft data from the 1,000 Genomes project, we found that the risk allele of a novel SNP (rs77275268), which is closely correlated with rs9397435, disrupts a partially methylated CpG sequence within a known CTCF binding site. These studies demonstrate that shifting the analysis among ancestral populations can provide valuable resolution in association mapping.

High TWIST1 mRNA expression is associated with poor prognosis in lymph node-negative and estrogen receptor-positive human breast cancer and is co-expressed with stromal as well as ECM related genes.

INTRODUCTION: The TWIST homolog 1 (TWIST1) is a transcription factor that induces epithelial to mesenchymal transition (EMT), a key process in metastasis. The purpose of this study was to investigate whether TWIST1 expression predicts disease progression in a large breast cancer cohort with long-term clinical follow-up, and to reveal the biology related to TWIST1 mediated disease progression. METHODS: TWIST1 mRNA expression level was analyzed by quantitative real-time reverse polymerase chain reaction (RT-PCR) in 1,427 primary breast cancers. In uni- and multivariate analysis using Cox regression, TWIST1 mRNA expression level was associated with metastasis-free survival (MFS), disease-free survival (DFS) and overall survival (OS). Separate analyses in lymph node-negative patients (LNN, n = 778) who did not receive adjuvant systemic therapy, before and after stratification into estrogen receptor (ER)-positive (n = 552) and ER-negative (n = 226) disease, were also performed. The association of TWIST1 mRNA with survival endpoints was assessed using Kaplan-Meier analysis. Using gene expression arrays, genes showing a significant Spearman rank correlation with TWIST1 were used to identify overrepresented Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG)-annotated biological pathways. RESULTS: Increased mRNA expression level of TWIST1 analyzed as a continuous variable in both uni- and multivariate analysis was associated with shorter MFS in all patients (hazard ratio (HR): 1.17, 95% confidence interval, (95% CI):1.09 to 1.26; and HR: 1.17, 95% CI: 1.08 to 1.26; respectively), in LNN patients (HR: 1.22, 95% CI: 1.09 to 1.36; and HR: 1.21, 95% CI: 1.07 to 1.36; respectively) and in the ER-positive subgroup of LNN patients (HR: 1.34, 95% CI: 1.17 to 1.53; and HR: 1.32, 95% CI: 1.14 to 1.53; respectively). Similarly, high TWIST1 expression was associated with shorter DFS and OS in all patients and in the LNN/ER-positive subgroup. In contrast, no association of TWIST1 mRNA expression with MFS, DFS or OS was observed in ER-negative patients. Genes highly correlated with TWIST1 were significantly enriched for cell adhesion and ECM-related signaling pathways. Furthermore, TWIST1 mRNA was highly expressed in tumor stroma and positively related to tumor stromal content (P <0.001). CONCLUSIONS: TWIST1 mRNA expression is an independent prognostic factor for poor prognosis in LNN/ER-positive breast cancer. The biological associations suggest an involvement of the tumor microenvironment in TWIST1's adverse role in breast cancer.

Correlation of breast cancer susceptibility loci with patient characteristics, metastasis-free survival, and mRNA expression of the nearest genes.

To understand the biology of low-risk breast cancer alleles, and to investigate whether these loci also contribute to disease progression that was once established, we examined the association of SNPs tagging the low-risk breast cancer loci in or near FGFR2, LSP1, MAP3K1, H19, TOX3, POU5F1P1, MYC, and 2q35, with clinical, pathological characteristics, prognosis, and mRNA expression of the nearest genes. Tumor DNA samples of 2,480 breast cancer patients were available. Out of this cohort, 1,290 patients with lymph-node negative disease who did not receive adjuvant systemic therapy, the SNP status was associated with metastasis-free survival (MFS). In 1,401 patients, the mRNA expression levels of FGFR2, LSP1, MAP3K1, H19, TOX3, POU5F1P1, and MYC were determined and correlated with SNP genotypes. The SNP rs2981582 in FGFR2 was significantly associated with positive ER and PgR status (P < 0.001 and P = 0.003, respectively). No other significant associations with patient or tumor characteristics were observed. Only rs2107425 near H19 was significantly associated with shorter MFS in uni- and multi-variate analysis (HR: 1.53, CI: 1.12-2.08, P = 0.006 and HR: 1.59, CI: 1.16-2.20, P = 0.004, respectively), with the more aggressive minor allele displaying a recessive trait. The minor allele of SNP rs3803662 located near the TOX3 gene was associated with lower mRNA expression of this gene. In conclusion, except for the association of rs13283662 with TOX3 gene expression indicating a tumor suppressor role of TOX3, our findings suggest that breast cancer low-risk loci generally do not affect expression of the nearest gene in breast tumor tissue. Also the prognosis of patients is largely not affected by low-risk breast cancer loci except for the SNP near H19. How, this SNP affects prognosis warrants further study as it does not operate through altering H19 mRNA expression.

MicroRNA-30c expression level is an independent predictor of clinical benefit of endocrine therapy in advanced estrogen receptor positive breast cancer.

MicroRNAs (miRNAs) are small RNA molecules that modulate gene expression and which have been implicated in cancer. We evaluated whether five candidate predictive miRNAs, derived from a pilot study in which 249 miRNAs were assayed, were associated with clinical benefit of tamoxifen therapy in advanced breast cancer. These five miRNAs were measured in an independent series of 246 estrogen receptor (ER)-positive primary breast tumors of patients who received tamoxifen for advanced disease by quantitative Real Time PCR. Univariate analysis showed that higher expression levels of hsa-miR-30a-3p, hsa-miR-30c, and hsa-miR-182 were significantly associated with benefit of tamoxifen treatment and with longer PFS (all P-values <0.01). In multivariate analysis, corrected for the traditional predictive factors, only hsa-miRNA-30c was an independent predictor (P-value <0.01). Finally, in an attempt to understand the biology connected to this miRNA, Global testing pathway analysis showed an association of hsa-miRNA-30c expression with HER and RAC1 signaling pathways. We identified hsa-miRNA-30c as an independent predictor for clinical benefit of tamoxifen therapy in patients with advanced breast cancer. Assessment of tumor levels and connected pathways could be helpful to improve treatment strategies.

Prevalence of the variant allele rs61764370 T>G in the 3'UTR of KRAS among Dutch BRCA1, BRCA2 and non-BRCA1/BRCA2 breast cancer families.

Recently, a variant allele in the 3'UTR of the KRAS gene (rs61764370 T>G) was shown to be associated with an increased risk for developing non-small cell lung cancer, as well as ovarian cancer, and was most enriched in ovarian cancer patients from hereditary breast and ovarian cancer families. This functional variant has been shown to disrupt a let-7 miRNA binding site leading to increased expression of KRAS in vitro. In the current study, we have genotyped this KRAS-variant in breast cancer index cases from 268 BRCA1 families, 89 BRCA2 families, 685 non-BRCA1/BRCA2 families, and 797 geographically matched controls. The allele frequency of the KRAS-variant was found to be increased among patients with breast cancer from BRCA1, but not BRCA2 or non-BRCA1/BRCA2 families as compared to controls. As BRCA1 carriers mostly develop ER-negative breast cancers, we also examined the variant allele frequency among indexes from non-BRCA1/BRCA2 families with ER-negative breast cancer. The prevalence of the KRAS-variant was, however, not significantly increased as compared to controls, suggesting that the variant allele not just simply associates with ER-negative breast cancer. Subsequent expansion of the number of BRCA1 carriers with breast cancer by including other family members in addition to the index cases resulted in loss of significance for the association between the variant allele and mutant BRCA1 breast cancer. In this same cohort, the KRAS-variant did not appear to modify breast cancer risk for BRCA1 carriers. Importantly, results from the current study suggest that KRAS-variant frequencies might be increased among BRCA1 carriers, but solid proof requires confirmation in a larger cohort of BRCA1 carriers.

Identification of a putative protein profile associated with tamoxifen therapy resistance in breast cancer.

Tamoxifen resistance is a major cause of death in patients with recurrent breast cancer. Current clinical factors can correctly predict therapy response in only half of the treated patients. Identification of proteins that are associated with tamoxifen resistance is a first step toward better response prediction and tailored treatment of patients. In the present study we intended to identify putative protein biomarkers indicative of tamoxifen therapy resistance in breast cancer using nano-LC coupled with FTICR MS. Comparative proteome analysis was performed on approximately 5,500 pooled tumor cells (corresponding to approximately 550 ng of protein lysate/analysis) obtained through laser capture microdissection (LCM) from two independently processed data sets (n = 24 and n = 27) containing both tamoxifen therapy-sensitive and therapy-resistant tumors. Peptides and proteins were identified by matching mass and elution time of newly acquired LC-MS features to information in previously generated accurate mass and time tag reference databases. A total of 17,263 unique peptides were identified that corresponded to 2,556 non-redundant proteins identified with > or = 2 peptides. 1,713 overlapping proteins between the two data sets were used for further analysis. Comparative proteome analysis revealed 100 putatively differentially abundant proteins between tamoxifen-sensitive and tamoxifen-resistant tumors. The presence and relative abundance for 47 differentially abundant proteins were verified by targeted nano-LC-MS/MS in a selection of unpooled, non-microdissected discovery set tumor tissue extracts. ENPP1, EIF3E, and GNB4 were significantly associated with progression-free survival upon tamoxifen treatment for recurrent disease. Differential abundance of our top discriminating protein, extracellular matrix metalloproteinase inducer, was validated by tissue microarray in an independent patient cohort (n = 156). Extracellular matrix metalloproteinase inducer levels were higher in therapy-resistant tumors and significantly associated with an earlier tumor progression following first line tamoxifen treatment (hazard ratio, 1.87; 95% confidence interval, 1.25-2.80; p = 0.002). In summary, comparative proteomics performed on laser capture microdissection-derived breast tumor cells using nano-LC-FTICR MS technology revealed a set of putative biomarkers associated with tamoxifen therapy resistance in recurrent breast cancer.

Four miRNAs associated with aggressiveness of lymph node-negative, estrogen receptor-positive human breast cancer.

In this study, we quantified 249 mature micro-RNA (miRNA) transcripts in estrogen receptor-positive (ER(+)) primary breast tumors of patients with lymph node-negative (LNN) disease to identify miRNAs associated with metastatic capability. In addition, the prognostic value of the candidate miRNAs was determined in ER(-)/LNN breast cancer. Unsupervised analysis in a prescreening set of 38 patients identified three subgroups predominantly driven by three miRNA signatures: an ER-driven luminal B-associated miRNA signature, a stromal miRNA signature, and an overexpressed miRNA cluster located on chromosome 19q23, but these intrinsic miRNA signatures were not associated with tumor aggressiveness. Supervised analysis in the initial subset and subsequent analysis in additional tumors significantly linked four miRNAs (miR-7, miR-128a, miR-210, and miR-516-3p) to ER(+)/LNN breast cancer aggressiveness (n = 147) and one miRNA (miR-210) to metastatic capability in ER(-)/LNN breast cancer (n = 114) and in the clinically important triple-negative subgroup (n = 69) (all P < 0.05). Bioinformatic analysis coupled miR-210 to hypoxia/VEGF signaling, miR-7 and miR-516-3p to cell cycle progression and chromosomal instability, and miR-128a to cytokine signaling. In conclusion, our work connects four miRNAs to breast cancer progression and to several distinct biological processes involved therein.

Clinical significance of the nuclear receptor co-regulator DC-SCRIPT in breast cancer: an independent retrospective validation study.

INTRODUCTION: In this study we aimed to validate the prognostic value of DC-SCRIPT mRNA expression in a large independent breast cancer cohort. In addition, since DC-SCRIPT is a transcriptional co-regulator of nuclear receptors, we explored its prognostic value in relation to estrogen-receptor-α (ESR1) and -ß (ESR2) and evaluated its predictive value for response to tamoxifen treatment. METHODS: DC-SCRIPT mRNA levels were measured by real-time PCR in 1,505 primary invasive breast cancers and associated with outcome (disease-free survival (DFS), metastasis-free survival (MFS) and overall survival (OS)) using univariate and multivariable Cox regression analysis. Logistic and Cox regressions were used to associate DC-SCRIPT levels with clinical benefit and progression-free survival (PFS) for 296 patients treated with first-line systemic tamoxifen for advanced disease. RESULTS: In univariate and multivariable analysis higher DC-SCRIPT levels were associated with a favorable outcome for both the entire cohort and patients with lymph node-negative (LNN) disease that did not receive adjuvant therapy (DFS, MFS and OS; all, P < 0.001). This association was most pronounced in small (pT1) tumors, in ESR1-positive tumors and in tumors with low ESR2 expression. For first-line endocrine therapy for advanced disease no predictive association was seen with clinical benefit or PFS. CONCLUSIONS: This study provides a higher level of evidence that DC-SCRIPT is indeed an independent, pure prognostic, factor for primary breast cancer and shows that DC-SCRIPT mRNA expression is most informative for either ESR1-positive and/or ESR2-low pT1 tumors.

Distinct inherited metastasis susceptibility exists for different breast cancer subtypes: a prognosis study.

INTRODUCTION: Previous studies in mouse models and pilot epidemiology studies have demonstrated that inherited polymorphisms are associated with inherited risk of tumor progression and poor outcome in human breast cancer. To extend these studies and gain better understanding of the function of inherited polymorphism in breast cancer progression, a validation prognosis study was performed in a large independent breast cancer patient population. METHODS: The study population consisted of 1863 Dutch patients with operable primary breast cancer from Rotterdam, The Netherlands. Genomic DNA was genotyped for the missense Pro436Leu RRP1B single nucleotide polymorphism (SNP) rs9306160 and the intronic SIPA1 SNP rs2448490 by SNP-specific PCR. RESULTS: A significant association of variants in RRP1B with metastasis-free survival was observed (P = 0.012), validating the role of RRP1B with inherited metastatic susceptibility. Stratification of patients revealed that association with patients' survival was found to be specifically restricted to estrogen receptor positive, lymph node-negative (ER+/LN-) patients (P = 0.011). The specific association with metastasis-free survival only in ER+/LN- patients was replicated for SIPA1, a second metastasis susceptibility gene known to physically interact with RRP1B (P = 0.006). Combining the genotypes of these two genes resulted in the significant ability to discriminate patients with poor metastasis-free survival (HR: 0.40, 95% CI: 0.24 to 0.68, P = 0.001). CONCLUSIONS: These results validate SIPA1 and RRP1B as metastasis susceptibility genes and suggest that genotyping assays may be a useful supplement to other clinical and molecular indicators of prognosis. The results also suggest that lymphatic and hematogeneous metastases are genetically distinct that may involve different mechanisms. If true, these results suggest that metastatic disease, like primary breast cancer, may be multiple diseases and that stratification of late stage patients may therefore be required to fully understand breast cancer progression and metastasis.

TSC22D1 and PSAP predict clinical outcome of tamoxifen treatment in patients with recurrent breast cancer.

Purpose Two genes, TSC22 domain family, member 1 (TSC22D1) and prosaposin (PSAP) were identified in an in vitro functional screen for genes having a causative role in tamoxifen resistance. These genes were also present in our previously established 81-gene signature for resistance to first-line tamoxifen therapy. The aim of this study was to investigate the predictive value of these genes for tamoxifen therapy failure in patients with recurrent breast cancer. Experimental Design The mRNA levels of TSC22D1 and PSAP were analyzed by quantitative real-time polymerase chain reaction (qRT-PCR) in 223 estrogen receptor-positive primary breast tumors of patients with recurrent disease treated with first-line tamoxifen therapy. The main objective of this study was the length of progression-free survival (PFS). Results High mRNA levels of TSC22D1 and PSAP were significantly associated with shorter PFS and both were independent of the traditional predictive factors (HR = 1.30, 95% CI = 1.04-1.64 P = 0.023; and HR = 1.40, 95% CI = 1.03-1.88, P = 0.029, respectively). In multivariate analysis, patients with high mRNA levels of both genes associated significantly with no clinical benefit (OR = 0.19, 95% CI = 0.06-0.62, P = 0.006) and had the shortest PFS (HR = 2.05, 95% CI = 1.29-3.25, P = 0.002). Conclusion These results confirm our previous in vitro and tumor-related findings and are indicative for the failure of tamoxifen treatment in breast-cancer patients. Both TSC22D1 and PSAP are associated with clinical outcome and may have a functional role in therapy resistance.

Downregulation of SIAH2, an ubiquitin E3 ligase, is associated with resistance to endocrine therapy in breast cancer.

PURPOSE: In our microarray analysis we observed that Seven-in-Absentia Homolog 2 (SIAH2) levels were low in estrogen receptor (ER) positive breast tumors of patients resistant to first-line tamoxifen therapy. The aim of this study was to evaluate SIAH2 for its (a) predictive/prognostic value, and (b) functional role in endocrine therapy resistance. PATIENTS AND METHODS: SIAH2 expression was measured with quantitative Real-Time-PCR (qRT-PCR) in 1205 primary breast tumor specimens and related to disease outcome. The functional role of SIAH2 was determined in human breast cancer cell lines ZR-75-1, ZR/HERc, and MCF7. Cell lines were treated with estrogen (E2), anti-estrogen ICI164.384 or epidermal growth factor (EGF). Moreover, MCF7 was treated with ICI164.384 after silencing SIAH2 expression. RESULTS: SIAH2 was not prognostic in 603 lymph node negative patients who had not received adjuvant systemic therapy. In multivariate analysis of ER-positive tumors of 235 patients with recurrent disease, SIAH2 as continuous variable, significantly predicted first-line tamoxifen treatment failure (OR = 1.48; P = 0.05) and progression-free survival (PFS) (HR = 0.79; P = 0.007). Furthermore, in primary breast cancer patients treated with adjuvant tamoxifen, SIAH2 predicted metastasis-free survival (MFS) (HR = 0.73; P = 0.005). In vitro experiments showed that SIAH2 silencing in MCF7 cells resulted in resistance to ICI164.384-treatment when compared with mock silenced cells (P = 0.008). Interestingly, in ZR cells transfected with EGFR (ZR/HERc), SIAH2 expression was induced by E2 but downregulated by EGF. CONCLUSION: In primary breast tumor specimens as well as in vitro low SIAH2 levels associated with resistance to endocrine therapy. Moreover, SIAH2 expression showed an opposite regulation by E2 and EGF.

Tumor tissue levels of Tissue Inhibitor of Metalloproteinases-1 (TIMP-1) and outcome following adjuvant chemotherapy in premenopausal lymph node-positive breast cancer patients: A retrospective study.

BACKGROUND: We have previously demonstrated that high tumor tissue levels of TIMP-1 are associated with no or limited clinical benefit from chemotherapy with CMF and anthracyclines in metastatic breast cancer patients. Here, we extend our investigations to the adjuvant setting studying outcome after adjuvant chemotherapy in premenopausal lymph node-positive patients. We hypothesize that TIMP-1 high tumors are less sensitive to chemotherapy and accordingly that high tumor tissue levels are associated with shorter survival. METHODS: From our original retrospectively collected tumor samples we selected a group of 525 pre-menopausal lymph node-positive patients (adjuvant treatment: CMF, 324 patients; anthracycline-based, 99 patients; no adjuvant chemotherapy, 102 patients). TIMP-1 levels were measured using ELISA in cytosolic extracts of frozen primary tumors. TIMP-1 was analyzed as a continuous variable and as a dichotomized one using the median TIMP-1 concentration as a cut point between high and low TIMP-1 groups. We analyzed the benefit of adjuvant CMF and anthracyclines in univariate and multivariable survival models; endpoints were disease-free (DFS) and overall survival (OS). RESULTS: In this selected cohort of high-risk patients, and in the subgroup of patients receiving no adjuvant therapy, TIMP-1 was not associated with prognosis. In the subgroup of patients treated with anthracyclines, when analyzed as a continuous variable we observed a tendency for increasing TIMP-1 levels to be associated with shorter DFS (multivariable analysis, HR 1.75, 95% CI 1.00-3.07, P = 0.05) and a significant association between increasing TIMP-1 and shorter OS in both univariate (HR 3.52, 95% CI 1.54-8.06, P = 0.003) and multivariable analyses (HR 4.19, 95% CI 1.67-10.51, P = 0.002). No statistically significant association between TIMP-1 and DFS was observed in the CMF-treated patients although high TIMP-1 was associated with shorter OS when analyzed as a dichotomized variable (HR 1.64, 95% CI 1.02-2.65, P = 0.04). CONCLUSION: In the subgroup of patients receiving adjuvant chemotherapy we found an association between shorter survival after treatment in TIMP-1 high patients compared with TIMP-1 low patients, especially in patients receiving anthracycline-based therapy. This suggests that high tumor tissue levels of TIMP-1 might be associated with reduced benefit from classical adjuvant chemotherapy. Our findings should be validated in larger prospective studies.

Association of an extracellular matrix gene cluster with breast cancer prognosis and endocrine therapy response.

PURPOSE: We previously discovered an extracellular matrix (ECM) gene cluster associated with resistance to first-line tamoxifen therapy of patients with metastatic breast cancer. In this study, we determined whether the six individual ECM genes [collagen 1A1 (COL1A1), fibronectin 1 (FN1), lysyl oxidase (LOX), secreted protein acidic cysteine-rich (SPARC), tissue inhibitor of metalloproteinase 3 (TIMP3), and tenascin C (TNC)] were associated with treatment response, prognosis, or both. EXPERIMENTAL DESIGN: In 1,286 primary breast tumors, mRNA expression (quantitative real-time PCR) was related to clinicopathologic factors and disease outcome in univariate and multivariate analysis including traditional factors. RESULTS: TIMP3, FN1, LOX, and SPARC expression levels (continuous variables) were significantly associated with distant metastasis-free survival (MFS) in 680 lymph node-negative untreated patients (P<0.03). Using a calculated linear prognostic score, these patients were evenly divided into five prognostic groups with a significant difference in 10-year MFS of approximately 40% between the two extreme prognostic groups. Furthermore, high TNC expression as continuous variable was associated with (a) shorter MFS in 139 estrogen receptor-positive and lymph node-positive patients who received adjuvant tamoxifen therapy (hazard ratio, 1.53; P=0.001), and (b) no clinical benefit (odds ratio, 0.81; P=0.035) and shorter progression-free survival (hazard ratio, 1.19; P=0.002) in 240 patients in whom recurrence was treated with tamoxifen as first-line monotherapy. These results were also significant in multivariate analyses. CONCLUSION: FN1, LOX, SPARC, and TIMP3 expression levels are associated with the prognosis of patients with breast cancers, whereas TNC is associated with resistance to tamoxifen therapy. Further validation and functional studies are necessary to determine the use of these ECM genes in decisions regarding treatment and whether they can serve as targets for therapy.

Fibroblast growth factor receptor 4 predicts failure on tamoxifen therapy in patients with recurrent breast cancer.

Tamoxifen treatment of estrogen-dependent breast cancer ultimately loses its effectiveness due to the development of resistance. From a functional screen for identifying genes responsible for tamoxifen resistance in human ZR-75-1 breast cancer cells, fibroblast growth factor (FGF) 17 was recovered. The aim of this exploratory study was to assess the predictive value of FGF17 and the receptors FGFR1-4 for the type of response to tamoxifen treatment (clinical benefit) and the duration of progression-free survival (PFS) in patients with recurrent breast cancer. mRNA levels of FGF17 and FGFR1-4 were quantified by real-time reverse transcriptase PCR in 285 estrogen receptor-positive breast carcinomas with clinical follow-up. All patients had recurrent disease and were treated with tamoxifen as first-line systemic therapy for local or distant relapse. FGF17 and FGFR1-3 mRNA levels had no significant predictive value for this group of patients. However, high FGFR4 mRNA levels analyzed as a continuous log-transformed variable predicted poor clinical benefit (odds ratio=1.22; P=0.009) and shorter PFS (hazard ratio=1.18; P<0.001). In addition, in multivariable analysis, the predictive value of FGFR4 was independent from the traditional predictive factors. Our analyses show that FGFR4 may play a role in the biological response of the tumor to tamoxifen treatment. In addition, as altered expression of FGF17 causes tamoxifen resistance in vitro, the FGF signaling pathway could be a valuable target in the treatment of breast cancer patients resistant to endocrine treatment.

IGF1R signaling drives antiestrogen resistance through PAK2/PIX activation in luminal breast cancer.

Antiestrogen resistance in estrogen receptor positive (ER+) breast cancer is associated with increased expression and activity of insulin-like growth factor 1 receptor (IGF1R). Here, a kinome siRNA screen has identified 10 regulators of IGF1R-mediated antiestrogen with clinical significance. These include the tamoxifen resistance suppressors BMPR1B, CDK10, CDK5, EIF2AK1, and MAP2K5, and the tamoxifen resistance inducers CHEK1, PAK2, RPS6KC1, TTK, and TXK. The p21-activated kinase 2, PAK2, is the strongest resistance inducer. Silencing of the tamoxifen resistance inducing genes, particularly PAK2, attenuates IGF1R-mediated resistance to tamoxifen and fulvestrant. High expression of PAK2 in ER+ metastatic breast cancer patients is correlated with unfavorable outcome after first-line tamoxifen monotherapy. Phospho-proteomics has defined PAK2 and the PAK-interacting exchange factors PIXα/ß as downstream targets of IGF1R signaling, which are independent from PI3K/ATK and MAPK/ERK pathways. PAK2 and PIXα/ß modulate IGF1R signaling-driven cell scattering. Targeting PIXα/ß entirely mimics the effect of PAK2 silencing on antiestrogen re-sensitization. These data indicate PAK2/PIX as an effector pathway in IGF1R-mediated antiestrogen resistance.