Analysis of bone marrow micrometastasis in primary breast cancer: Automated cellular imaging analysis in relation to quantitative RT-PCR and tumor clinicopathologic features

The presence of bone marrow micrometastasis (MM) has been reported to be an adverse prognostic indicator in primary breast cancer, and may be of value in adjuvant therapy decision-making. To date, MM detection has been pursued by a variety of cell-based and molecular methods, and it remains unclear which methodologies are most useful, including recently developed techniques. In a prospective study, we have evaluated MM using novel methods applied to primary breast cancer patients. Methods: We have developed and optimized a highly sensitive cell-based MM assay (cMM) involving multiple antigen immunomagnetic capture (positive selection for MM), anti-cytokeratin immunocytochemical (ICC) staining, and automated cellular imaging (ACIS, ChromaVision Medical Systems, Inc.). The resulting cellular images are immediately available for review by laboratory professionals and pathologists. In addition to this assay, we have developed quantitative RT-PCR for MM-associated mRNAs mammaglobin, PSA, HER2, alternatively spliced HER2, and EGFRvIII. Results: The ICC-based assay has been documented to provide reproducible detection and enumeration of rare carcinoma cells (lte] 1 MM per 10e8 mononuclear cells). Initial results from 21 consecutive cases analyzed using the cMM assay detected marrow MM in 10/21 (48%) of specimens. MM in matched blood samples were frequently detected but generally at lower levels. Quantitative RT-PCR has been performed on 165 patient marrow samples, including all 21 analyzed by the ICC-based assay. RT-PCR assays detected MM in 10-20% of patients depending upon target gene. Results did not correlate closely with each other or with the ICC-based assay. Updated results will be presented, along with correlations with clinicopathologic parameters including tumor size, grade, nodal status, and primary tumor ER/PR/HER2 results.Conclusions: This prospective study compares recently developed and highly sensitive molecular and cell-based methods for MM analysis. Understanding performance differences of new methods should enhance both our understanding of the biology of MM and facilitate the assessment of MM assay(s)in clinical settings.

Duke Scholars

Author Eun-Sil Shelley Hwang Surgical Oncology