Abstract Introduction In metastatic breast cancer (MBC), discordant expression levels of the human epidermal growth factor receptor 2 (HER2) have been noted between primary tumors (PT) and matched metastatic lesions. Reassessment of HER2 status during treatment decisions in patients with advanced disease might help to optimize outcome. Circulating tumor cells (CTCs) offer the potential to provide a repeatedly accessible source of tumor cells for the real-time assessment of actual tumor characteristics. However, little is known on the concordance of HER2 expression on CTCs measured by immunofluorescence and the amplification status. Here we report on a preclinical study, using five spiked breast cancer cell lines, comparing semi-quantitative HER2 scoring on CellSearch (Riethdorf 2010) with objective DEPArray analysis, and subsequent FISH analysis on DEPArray-sorted tumor cells. At the moment these data are also being generated for 10 patient CTC samples. Expression and amplification status of CTCs will be compared with primary tumor tissue. Materials and methods MDA-MB-436, MCF-7, BT-20, KPL-4, and SKBR3 cells (increasing HER2 status) were spiked into donor blood and subjected to CellSearch enrichment. HER2/FITC intensity was scored manually on the CellSearch analyzer. All cell lines were injected into the DEPArray and exposure settings were optimized (FITC: exposure time 800 ms, gain 5%). These settings are further used for all preclinical and clinical samples. HER2 scoring was based on relative fluorescent units (rfu) of the HER2/FITC signal with background subtraction. Cells were sorted into pure batches of HER2 positive (DAPI+/CK+/HER2+/CD45-) and negative (DAPI+/CK+/HER2-/CD45-) tumor cells. Cytospins were formalin fixed and subjected to DAKO IQFISH. Results HER2 expression on CellSearch turned out to be very heterogeneous within the same cell line. DEPArray data was highly reproducible for all cell lines (p<0.001) and also showed a broad range of FITC rfu within the HER2 positive cell lines. Significant differences were observed between every cell line (p<0.001). The SKBR3 cell line sample also harbored a minor population of HER2- cells while this was the most positive cell line. However with FISH analysis, both HER2- and HER2+ SKBR3 cells were highly amplified (absolute HER2 count of 12-20 and HER2/CEN17 ratio of >4). MDA-MB-436 and MCF-7 cells showed no gene amplification on FISH, while in KPL-4 there was a HER2/CEN17 ratio of >2. Four patient samples with HER2 positive status on CellSearch have been run on the DEPArray. For patient 1, 1005 CTC were analyzed, 32.4% were HER2+. This was 53 (69.8% HER2+), 352 (5.7% HER2+), and 622 (6.7% HER2+) for patient 2-4 respectively. These numbers are comparable with CellSearch analysis. Discussion HER2 expression analysis by immunofluorescence is comparable between CellSearch and DEPArray, however DEPArray has the advantage that it is user-independent and highly reproducible. Furthermore, CTCs can be sorted into pure batches for downstream analysis. The FISH technique on DEPArray sorted cells is now optimized and will be used to determine the correlation between the immunofluorescent HER2 scoring and the actual amplification status of the CTCs. These data will be incorporated prior to upcoming SABCS. Citation Format: Brouwer A, Verhoest F, Vermeulen P, Rutten A, Prové A, Van Laere S, Peeters M, Dirix L. Evaluation of HER2 expression and amplification on CTCs using DEPArray analysis and sorting followed by FISH [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P3-05-10.
Read full abstract