Search

Background: Autocrine motility factor (AMF) is secreted by cancer cells and acts in an autocrine or paracrine fashion to bind to its receptor, AMFR, at the cell surface of tumor cells. The AMF-AMFR pathway has been shown to promote proliferation, anti-apoptosis, motility and migration, invasion, and metastasis pathways in various cancers. However, our understanding of the AMF-AMFR pathway in breast cancer is limited. We propose that the AMF signaling pathway is an unexplored mechanism for breast cancer tumor aggression and its receptor, AMFR, may be a potential therapeutic target. Methods: Tumor tissue obtained from consented female (n = 31) and male (n = 1) patients were analyzed by the Affymetrix OncoscanTM genome-wide microarray platform and examined for somatic copy-number alterations (SCNAs) of AMFR. cBioPortal was used to investigate SCNA of AMFR and gene expression of AMF on primary breast cancer tumors from METABRIC (n = 1,784) and TCGA Pan-Cancer Atlas (n = 981) datasets. In vitro, AMF and AMFR gene expression in luminal A (MCF-7) and triple-negative breast cancer (MDA-MB-231) cell lines were assessed by qPCR. Cell migration assays were performed on MDA-MB-231 cells to investigate their migration towards AMF with AMFR present or knocked down by siRNA. Results: Microarray analysis of 32 tumors revealed that a single-copy loss of AMFR occurred 79% of the time in luminal A tumors (n = 19); 67% of the time in luminal B tumors (n = 6); 33% of the time in ER+, PR+, HER2+ tumors (n = 3); and 0% of the time in triple-negative breast cancer (TNBC) tumors (n = 4), suggesting that the loss of AMFR results in less aggressive tumors that have good overall prognosis. To extend our findings to a larger patient cohort, SCNA analysis of the METABRIC and TCGA Pan-Cancer Atlas datasets revealed that single-copy loss of AMFR occurred in 64.53% and 68.14% of luminal A tumors, 58.51% and 59.90% of luminal B tumors, 23.62% and 48.72% of HER2 overexpression tumors, and 29.80% and 39.77% of TNBC tumors, respectively. Therefore, AMFR appears most frequently deleted in the tumor genomes of good prognosis breast cancer molecular subtypes (luminal tumors). Gene expression analysis of AMF in the METABRIC (using z-scores) and TCGA Pan-Cancer Atlas (using batch-normalized numbers) datasets revealed median mRNA expressions of -0.3351 and 4862 in luminal A tumors, -0.05415 and 5841 in luminal B tumors, 0.5758 and 9390 in HER2 overexpression tumors, and 0.754 and 8798 in TNBC tumors, respectively, suggesting that the AMF-AMFR pathway is more active in aggressive breast cancers. Similarly, we observed that AMF and AMFR are transcriptionally overexpressed by 7-fold and 16-fold, respectively, in the TNBC cell line MDA-MB-231 compared to the luminal A breast cancer cell line MCF-7. When AMFR is knocked down in MDA-MB-231 cells, focused migration towards AMF is abolished. Conclusion: AMF-AMFR pathway activity correlates with aggressive cancer cell behavior and enhanced migration in breast cancer. Single-copy loss of AMFR in tumor genomes is associated with less aggressive tumors with better overall prognosis. AMFR may be an attractive therapeutic target. Citation Format: Liu C, Price JD, Maroni D, Stevens JM, VanDyke AZ, Althof PA, Mondal P, Sanmann JN, Thayer SP. Autocrine motility factor signaling pathway promotes aggressive behavior and migration in breast cancer [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P2-03-09.

Background: Most modern methods of detecting circulating tumor cells (CTCs) involve identifying cells with epithelial markers. This approach presents challenges, as not all epithelial cells found in circulation originate from the tumor and not all CTCs express epithelial markers. We propose using a size-exclusion filtration system to enrich for CTCs in peripheral blood followed by fluorescence in situ hybridization (FISH) of the filtered cells to identify cells of tumor origin in the early-stage breast cancer patients. We further hypothesize that the presence of CTCs may be indicators of therapy failure in early-stage breast cancer patients. Methods: Patients diagnosed with breast cancer (n = 9) were consented for CTC evaluation. Primary tumor DNA was analyzed by the Affymetrix OncoscanTM genome-wide microarray platform and investigated for somatic copy-number alterations (SCNAs). For each patient, two FISH probes were then identified for two regions of gain or a region of gain and a region of loss from the microarray results. Blood samples from patients were obtained before surgery, radiation therapy, endocrine therapy, and at 6-month or 1-year follow-up visits. Blood samples were filtered using ScreenCell® Cyto V2 devices, and FISH was performed. Cells were categorized as normal (diploid for all FISH probes), suspicious (single SCNA detected by FISH), or CTC (two SCNAs detected by FISH). Patients were identified as having CTCs present in their circulation when ≥2 CTCs were observed or when one CTC and >15 suspicious cells were observed. Results: The microarray data revealed that luminal A tumors ranged from 2-43 SCNAs; luminal B tumors ranged from 15-20 SCNAs; and ER+, PR+, HER2+ tumors ranged from 46-98 SCNAs. Although a correlation appears to exist between tumor genetic complexity and molecular subtype, the degree of complexity was highly varied within each subtype. We found that neither complexity of tumor profile, molecular subtype, nor stage could predict the presence of CTCs in patients. Molecular SubtypeSCNAsCTCsSuspicious Cells OnlyLuminal A2-433/52/5Luminal B15-200/21/2ER+, PR+, HER2+46-982/20/2 In pre-surgical blood samples, we detected CTCs in 63% of patients with stage 1 disease and in 60% of patients with luminal A tumors, 0% of patients with luminal B tumors, and 100% of patients with triple-positive tumors. StageCTCsSuspicious Cells OnlyIA/B5/82/8IIA0/11/1 Although limited in number, ongoing investigation revealed that one of our patients in early follow-up with a luminal A, stage IB tumor was identified to have persistent CTCs at 1-year after starting hormonal adjuvant therapy, suggesting residual tumor burden not detected by standard clinical modalities; this finding also suggests that this patient may be at highest risk for relapse and should be considered for additional therapies. Conclusion: Size-exclusion filtration followed by FISH analysis can accurately identify CTCs in early-stage breast cancer patients. Tumor complexity, molecular subtype, and stage did not predict the presence of CTCs in circulation. Our method for CTC detection may be able to serve as a diagnostic tool for treatment failure. Citation Format: Liu C, Althof PA, Maroni D, Stevens JM, Grabow CE, VanDyke AZ, Price JD, Sanmann JN, Thayer SP. Circulating tumor cells of breast cancer origin identified by fluorescence in situ hybridization and may be an early predictor of therapy failure in early breast cancer [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P3-01-21.