Tel.+49 30 838 62450 Fax.+49 30 838 62522
In breast cancer CTC have been subject of much attention in the past years. CTC and their detection have proven clinical usefulness in many types of cancers.
In fact through CTC enumeration lower CTC counts were correlated with better disease-free and overall survivals in patients with breast cancer, before and after treatment, hence providing valuable information in terms of prognosis and response to treatment (Cristofanilli, 2006; Hayes et al., 2006; Mostert et al., 2009).
Based on in these findings it was hypothesized that CTC are also present in dogs baring metastatic mammary tumors (MT), their detection using nucleic-acid based methods is possible and may correlate with the histological evidence of vascular invasion by tumor cells in the vicinity of the primary tumor.
The first phase of the project consisted in the identification of potential canine specific MT CTC mRNA markers. To this end a total of 106 genes identified by three different approaches were tested for their sensitivity and specificity: (1). mRNA markers with proved usefulness for CTC detection of breast cancer CTC; (2). genes known to be overexpressed in CMT; (3). genes identified by microarray technology. These genes were then tested for their expression in the peripheral blood (PB) samples of healthy dog donors (n=12 to 14), metastatic mammary gland carcinomas (n=10) and malignant CMT cell lines (n=2) by RTPCR. To be considered a potential CTC marker, the markers had to follow the selection criterion of not being expressed in the PB and to be expressed in metastatic carcinomas and malignant cells lines. A total of 12 potential candidate mRNA markers were identified. Furthermore the ability of these candidate mRNA markers to detect single tumor cells was tested in serial dilutions of tumor cells from cell lines CMM 26 and CMM115 in the PB were executed. Based on the frequency of occurrence of CTC in humans (1 CTC per 105 to 107 peripheral blood leukocytes (PBL)) a target sensitivity of 1 tumor cell per 106 to 107 PBL was established. From all the candidates only maspin was not able to achieve the desired sensitivity. In a second phase a panel of seven of the previously identified markers were tested in 120 PB samples from female dogs with malignant mammary gland carcinomas with histological evidence of vascular invasion of tumors cells near the primary tumor or presence of metastatic cells in the regional lymph node (n=40), with malignant mammary gland carcinomas without histological evidence of vascular invasion and metastasis free lymph nodes (n=40) and patients with benign mammary gland adenomas (n=40), in order to correlate their expression with the existence of histological evidence of vascular invasion and therefore with the presence of tumor cells in vascular compartments. From the panel of seven markers tested CRYAB was the most promising marker revealing the best relation sensitivity/specificity. Through the establishment of multimarker assays combining the detection abilities of individual markers (CLDN7, CRYAB, ATP8B1 and EGFR) we were able to significantly increase the sensitivity. From all the multimarker patterns MM1 was the best displayed with the best sensitivity/specificity relation. Although multimarker assays significantly enhance the sensitivity of the assay, a moderate loss of specificity is observed, however the multimarker approach seems to be more advantageous the single marker approach. The assays ability to differentiate blood from dogs with malignant tumors with and without vascular invasion with tumor cells, suggest that they may allow the identification of dogs with CTC in the PB, nevertheless further studies that include a follow-up period of 24 month with regular physical examinations and imaging scans are required to determine the real prognostic significance of these CTC markers in the dog.