Robert T. Krivacic

Multiple Biomarker Expression on Circulating Tumor Cells in Comparison to Tumor Tissues from Primary and Metastatic Sites in Patients with Locally Advanced/Inflammatory, and Stage IV Breast Cancer, Using a Novel Detection Technology

Patients with locally advanced/inflammatory breast cancer (LABC/IBC) face a high likelyhood of recurrence and prognosis for relapsed, or de novo stage IV metastatic breast cancer (MBC) remains poor. Estrogen (ER) and HER2 receptor expression on primary or MBC allow targeted therapies, but an estimated 10–18% of tumors do not exhibit these biomarkers and survival in these cases is even poorer. Variations in discordance rates for the expression of ER and HER2 receptors have been observed between primary and metastatic tumors and such discordances may lead to suboptimal treatment. Circulating tumor cells (CTCs) are considered the seeds of residual disease and distant metastases and their characterization could help guide treatment selection. To explore this possibility, we used multiple biomarker assessment of CTCs in comparison to primary and metastatic tumor sites. Thirty-six patients with LABC/IBC, or stage IV MBC were evaluated. Blood samples were procured prior to initiating or changing therapy. CTCs were identified based on presence of cytokeratin and nucleus staining, and the absence of CD45. A multimarker assay was developed to simultaneously quantify expression of HER2, ER, and ERCC1, a DNA excision repair protein. Novel fiber-optic array scanning technology (FAST) was used for sensitive location of CTCs. CTCs were detected in 82% of MBC and 62% LABC/IBC cases. Multiplex marker expression was successfully carried out in samples from18 patients with MBC and in 8 patients with LABC/IBC that contained CTCs. In MBC, we detected actionable discordance rates of 40 and 23%, respectively for ER and HER2 where a biomarker was negative in the primary or metastatic tumor and positive in the CTCs. In LABC/IBC, actionable discordances were 60 and 20% for ER and HER2, respectively. Pilot trials evaluating the effectiveness of treatment selections based on actionable discordances between biomarker expression patterns on CTCs and primary or metastatic tumor sites may allow for a prospective assessment of CTC-based individualized targeted therapies.

High-speed detection of occult tumor cells in peripheral blood

Although detection of tumor cells in peripheral blood using imitiunocytochemistry and optical scanning is a promising method for screening and monitoring cancer, it poses a major technical challenge due to the extremely low tumor cell concentration in blood. The preferred detection method - digital microscopy - is far too slow for analysis of the large numbers of cells required for statistical validity. We describe here a novel prescan instrument that rapidly identifies a small number of candidates for subsequent examination by digital microscopy to determine if they are genuine tumor cells. The prescan is 500 times faster than digital microscopy and yet has a similar sensitivity. The high prescan speed is accomplished by trading resolution for field of view. The resolution of the prescan is determined by the laser spot size of about 10 microns. While this resolution is much coarser than the submicron resolution of microscopes, it is still sufficient for detecting fluorescent cells because it matches the size of a typical cell. The wide field of view and high scan rate are enabled by a novel application of fiber optics.

Sensitive Characterization of Circulating Tumor Cells for Improving Therapy Selection

For metastatic disease, biomarker profiling of distant metastases is done only when feasible because biopsy of metastases is invasive and associated with potential morbidity without proven benefit. So although biomarker expression may differ in distant metastases, treatment with targeted therapies is almost always based on biomarker targets derived from a patient’s primary breast tumor, usually excised years before development of metastatic disease. This work addresses measurement of biomarker expression on circulating tumor cells (CTCs) as a source of current biomarker expression. CTCs are rapidly located on a planar substrate with a sensitive detection instrument using Fiber Array Scanning Technology. The instrument targets abundant cytokeratins rather than EpCAM. The assay includes quantitative measurement of expression levels of 3 breast cancer markers (HER2, ER and ERCC1) that predict efficacy of treatment. We have observed high discordance rates in cancer markers between CTC and tissue. Multiplex testing may allow for personalized therapy for patients.

Abstract PD04-09: Multiple Biomarker Expression in Circulating Tumor Cells from Metastatic Breast Cancer Patients

Background: Numeration of circulating tumor cells, CTCs, from metastatic breast cancer (MBC) patients (pts) is predictive of outcome. Biomarker characterization of CTCs may be a useful adjunctive guide for personalized targeted and systemic treatment (Rx) selection. Method: A multimarker assay was used to simultaneously quantify expression of HER2, ER and ERCC1. A fast laser scanning instrument was used for sensitive location of CTCs on large glass substrates. CTCs are identified using automated digital microscopy by morphology, the presence of cytokeratin and a nucleus, and the absence of CD45. At the COHCC pts with newly diagnosed/progressing MBC were accrued. Blood samples (10 ml) were procured prior to initiating systemic Rx and at subsequent 3 month intervals and sent to PARC for analysis. Cell lines with expression of each marker were used for normalization of the cell intensities. Sample scores were derived from the percentage of CTCs expressing the marker and the average expression level. Results: The multiple-marker assay was done on CTCs at repeat time points and results were compared to findings from the original primary BCs (P) and biopsied metastases (M) in 30 and 20 MBC pts respectively. While P and M tissue scores were concordant for HER2, the CTC score was discordant in 58% of the samples; HER2 expression changed during Rx in 19% of pts. While the status for ERCC1 was discordant between P and M tissue in 13% of the pts, CTCs scores were discordant with P and M tumors in 63% and 67% of the patients respectively, and CTC expression status changed during Rx in 15% and 7% of pts respectively. While the status for ER was discordant between P and M tissue in 15% of the pts, CTC scores were discordant from the P and M tumors in 42% and 71% of pts respectively, and CTC ER status changed in 7% and 17% respectively of pts during Rx. Conclusions: Significant discordances in expression level of ER, HER2 andERCC 1 was observed between CTCs, and both primary and metastatic BC tissue. Changes in CTC expression patterns were also observed during the course of Rx for all three markers. Correlation of CTC biomarker expression patterns and changes with response to Rx therapy is ongoing to validate medical significance. Multimarker testing may ultimately lead to improvements in personalized Rx for pts with MBC. Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr PD04-09.

Multiple Biomarker Expression in Circulating Tumor Cells (CTCs) from Metastatic Breast Cancer (MBC) Patients (Pts).

Background : Numeration of CTCs from MBC pts is predictive of outcome. Quantitative changes in CTC-s are currently tested for their potential to monitor therapy (Rx). Biomarker characterization of CTCs may be a useful adjunctive guide for Rx selection. Method: At the COHCC between 5/1/2008 and 4/31/09, consecutively treated pts with newly diagnosed/or progressing MBC were accrued. Blood samples (10-40 ml) were procured prior to or during systemic Rx, and were sent to PARC for analysis. A novel high-speed scanning instrument located CTCs from cytokeratin (CK) labeling enabling high resolution images to be selectively acquired using digital microscopy. From these images, CTCs were identified by CK, DAPI (nuclear marker) and CD45, and protein expression levels were determined for HER2, ER, ERCC1 and EGFR. Cell lines with expression of each marker were used for normalization of the cell intensities, and a scoring system was used to account for relative number and expression levels of markers on the CTCs. Results : Of 21pts tested 81% were found to have detectable CTCs. CTCs were further analyzed from 13 such pts, some of whom had multiple specimens. Expression of EGFR and ERCC1 were detected in 77% and 92% of specimens tested. Expression of HER2 was detected in 47% and ER in 91% in samples tested. Discordance rates for the expression of the above 4 markers on the primary tumors vs. CTC were measured either before, during systemic treatment, or at progression on therapy. We observed significant discordance rates for all markers tested:ER 36%; ERCC1:20%; EGFR:60%; and HER2: 50%, respectively. Conclusions : Multiplex tumor marker testing of CTCs from pts with MBC is feasible. Following additional validation of expression patterns and the high discordance rates observed between CTCs and primary or metastatic tumor sites, prospective trials incorporating CTC expression into personalized treatment strategies may be justified. Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 3007.