Guilherme Cantuaria

E23285 Background: Centrosome amplification (CA) and aberrant mitotic figures are well recognized as beacons of malignancy. However, it has recently been brought to light that mitotic indices are much lower in patient tumors than in continuously cultured cells, which may explain the clinical failure of many mitosis-targeted drugs that exhibited preclinical efficacy. We investigated the concordance between clinical tumor samples and cultured cancer cells by evaluating mitotic activity and CA. Methods: Using indirect immunofluorescence confocal imaging, we compared the number of mitotic figures (phosphohistone H3 staining) and CA (γ-tubulin staining), in formalin-fixed paraffin-embedded clinical tumor samples of breast, pancreas and bladder (n = 20 for each cancer type), immortalized cancer cell lines (MDA-MB-231-breast, MIA PaCa-pancreas and T24-bladder), tumor xenograft (derived from MDA-MB-231) and patient-derived cell line (breast). Results: We found while < 5% cells were mitotically active in clinical tumors, 5-15% of cultured cells were mitotically active. By contrast, CA ranged from 35-50% in tumor samples and xenografts as opposed to only 5-15% in continuous cell cultures. Patient-derived tumor cells displayed the highest CA (~50%), consonant with human tumors, albeit at low passage numbers only. Mechanistically, we found that hypoxia can induce CA; thus intratumoral hypoxia may explain the discordance between the extent of CA in patient tumors and cultured cancer cells. These findings were validated in silico by probing the publicly-available Kao microarray (n = 327) dataset using Gene Set Enrichment Analysis. Results suggested that high hypoxic breast tumors are associated with (a) high expression of genes related to CA and chromosomal instability (CIN25 gene signature), and (b) worse distant metastasis-free survival. Conclusions: These results bring to light the previously unrecognized discordance in mitotic frequency and CA between various cancer model systems and highlight the importance of utilizing low-passage-number cell lines, preferably patient-derived, in the study of CA to more accurately represent cellular phenotypes in patient tumors.

E17083 Background: Centrosome amplification (CA) is a hallmark of cancer. In order to evade spindle multipolarity and mitotic catastrophe, cancer cells cluster their extra centrosomes at the two spindle poles; this tactic allows them to survive but engenders low-grade aneuploidy. KifC1 is a centrosome clustering molecule essential for viability of cancer cells with CA. Since ovarian cancers display robust CA, we examined KIFC1 expression in human ovarian tumors. Methods: Paraffin-embedded formalin-fixed tissue microarrays of epithelial ovarian cancer (EOC) (n = 120) and normal ovarian epithelial tissue (n = 13) were immunostained for KifC1 and nuclear KifC1 weighted indices (WIs) were determined. KIFC1 expression levels in serous ovarian adenocarcinoma (SOC) (n = 154) patients were evaluated in single channel microarray data from publically available Gene Expression Omnibus (GEO) (GSE9899) database. Kaplan-Meier survival analyses, uni-and multi-variable (MV) analyses were performed to evaluate if KIFC1 expression levels have a bearing on patients’ overall survival (OS). Results: Nuclear KifC1 WI was higher in EOCs when compared to normal tissues (p < 0.01). Also, nuclear KifC1 WI increased with tumor grade (p < 0.05). In silico gene expression and survival analyses showed that irrespective of histological subtype, patients with above-threshold KIFC1 expression had shorter OS (p < 0.067). In SOC (n = 154), this association stayed significant (HR = 2.6, p = 0.006) in MV analysis after adjustment for tumor grade and stage. KIFC1 expression levels in high-grade SOC (HGSOC) correlated highly with expression of genes driving CA. GSEA revealed that genes enriched in KIFC1-high group were mainly related to microtubule-based processes or were centrosome components. Notably, in vitromodels of HGSOC (OVSAHO, KURAMOCHI, OVCAR3 and SKOV3) displayed robust CA and KifC1 levels were directly associated to levels of mitotic centrosome clustering, enabling cells to bypass mitotic catastrophe. Conclusions: KIFC1 expression level is a potential prognostic biomarker in ovarian adenocarcinomas and may help identify patients who could benefit from inclusion of KifC1 inhibitors in their therapy.

1091 Background: Ki67 Index (KI) and Mitotic Index (MI) are proliferation markers with established prognostic value in breast cancer. These indices are evaluated individually and on disparate measurement scales; they therefore fail to capture information about cell cycling kinetics of proliferating cells. Methods: Pathology reports of breast cancer patients (n = 10,504 from Northside Hospital, Atlanta and n = 1560 from Nottingham Hospital, UK) were retrospectively analyzed for mitotic scores, KI and clinical outcomes. Mitotic counts in 267 H&E-stained breast carcinoma samples were evaluated by two pathologists to transform mitotic scores into %MI based on cellularity. %MI: %KI ratio was defined as the Ki67 Adjusted Mitotic Score (KAMS), which reflects cycling kinetics of proliferative cells. Ability of KAMS to stratify triple negative breast cancers (TNBCs) was tested in three cohorts who received only adjuvant chemotherapy (n = 478 from Northside Hospital, USA; n = 322 from Nottingham Hospital, UK, and n = 108 from Olabisi Onabanjo University, Nigeria). Stratification of KAMS, KI and MI were performed on the thresholds that produced the lowest AIC (best model fit). Slow cycling and fast cycling TNBC subgroups from Nottingham Hospital were analyzed for biomarker expression. Results: Kaplan-Meier survival analyses, AIC and c²values showed that KAMS based stratification of TNBCs into two subgroups was superior to that by either KI or MI, regardless of hospital, and KAMS retained its significance in multivariate analyses. Fast-cycling TNBCs have poorer prognosis than slow-cycling TNBCs, perhaps due to higher intratumoral heterogeneity in fast cycling tumors. Fast-cycling TNBCs showed high expression of proteins implicated in DNA damage response, sumoylation, EGFR signaling and metastasis. By contrast, slow-cycling TNBCs showed extensive chromatin modification. Conclusions: KAMS quantifies cell cycling kinetics, stratifies TNBCs and yields new risk-predictive information that is not revealed by either KI or MI. KAMS reveals the underlying heterogeneity in cycling kinetics among TNBCs and helps identify TNBCs who might benefit from treatments that target the cell cycle machinery.

1089 Background: Triple Negative Breast Cancer (TNBC) is a highly heterogeneous and aggressive breast carcinoma subtype typified by high recurrence and mortality rates. To date, no targeted therapies are available for TNBC. Recently, androgen receptor (AR), a nuclear steroid hormone receptor expressed in 10-43% of TNBCs, has emerged as a promising new biomarker with potential predictive value for certain TNBC subtypes. Compared to European-American (EA) women, African Americans (AAs) are twice as likely to develop TNBC and have poorer outcomes, than EA TNBC patients, after adjusting for treatment and healthcare access. This study aimed to determine the prognostic value of AR in a large TNBC cohort and evaluate racewise trends in AR expression status among TNBCs. Methods: AR expression was evaluated immunohistochemically in formalin-fixed paraffin-embedded samples from 822 TNBC patients (142, 95 and 264 patients from Emory, Northside and Grady Memorial Hospitals, respectively, in Atlanta, and 321 from Nottingham Hospital, UK) for whom complete clinicopathologic and overall survival (OS) data were available. Ethnicity data were available for 138 AA and 675 EA TNBC cases. Samples with < 1% nuclear-stained cells were considered quadruple-negative (QN). Associations between nuclear AR status, race, other clinicopathologic variables and OS were evaluated through survival analyses and multivariate Cox regression. Results: In this study, 45.6% of the TNBC cases were QN. QN receptor status was significantly associated with older age (age > 25) at diagnosis, AA ethnicity and lower (KI < 14%) proliferative index (p < 0.001 for all) in univariate analyses. In multivariate analyses, loss of AR expression was associated with poorer OS (p < 0.05, HR = 2.8) after adjusting for grade, stage and adjuvant therapy. Interestingly, AAs had a much higher proportion (80.8%) of QN cases than EAs (40.1%, p < 0.0001). Conclusions: These data strongly support a prognostic role of AR in TNBC and suggest that a higher prevalence of QN breast cancer among AAs may be a plausible reason for the ethnic disparity in outcomes among TNBC patients.

Amplified centrosomes are widely recognized as a hallmark of cancer. Although supernumerary centrosomes would be expected to compromise cell viability by yielding multipolar spindles that results in death-inducing aneuploidy, cancer cells suppress multipolarity by clustering their extra centrosomes. Thus, cancer cells, with the aid of clustering mechanisms, maintain pseudobipolar spindle phenotypes that are associated with low-grade aneuploidy, an edge to their survival. KIFC1, a nonessential minus end-directed motor of the kinesin-14 family, is a centrosome clustering molecule, essential for viability of extra centrosome-bearing cancer cells. Given that ovarian cancers robustly display amplified centrosomes, we examined the overexpression of KIFC1 in human ovarian tumors. We found that in clinical epithelial ovarian cancer (EOC) samples, an expression level of KIFC1 was significantly higher when compared to normal tissues. KIFC1 expression also increased with tumor grade. Our In silico analyses showed that higher KIFC1 expression was associated with poor overall survival (OS) in serous ovarian adenocarcinoma (SOC) patients suggesting that an aggressive disease course in ovarian adenocarcinoma patients can be attributed to high KIFC1 levels. Also, gene expression levels of KIFC1 in high-grade serous ovarian carcinoma (HGSOC) highly correlated with expression of genes driving centrosome amplification (CA), as examined in publically-available databases. The pathway analysis results indicated that the genes overexpressed in KIFC1 high group were associated with processes like regulation of the cell cycle and cell proliferation. In addition, when we performed gene set enrichment analysis (GSEA) for identifying the gene ontologies associated to KIFC1 high group, we found that the first 100 genes enriched in KIFC1 high group were from centrosome components, mitotic cell cycle, and microtubule-based processes. Results from in vitro experiments on well-established in vitro models of HGSOC (OVSAHO, KURAMOCHI), OVCAR3 and SKOV3) revealed that they display robust centrosome amplification and expression levels of KIFC1 was directly associated (inversely correlated) to the status of multipolar mitosis. This association of KIFC1 and centrosome amplification with HGSOC might be able to explain the increased aggressiveness in this disease. These findings compellingly underscore that KIFC1 can be a biomarker that predicts an aggressive disease course in ovarian adenocarcinomas.

Amplified centrosomes in cancers are recently garnering a lot of attention as an emerging hub of diagnostic, prognostic and therapeutic targets. Ovarian adenocarcinomas commonly harbor supernumerary centrosomes that drive chromosomal instability. A centrosome clustering molecule, KIFC1, is indispensable for the viability of extra centrosome-bearing cancer cells, and may underlie progression of ovarian cancers.