Quantum Dots-Based Multiple Imaging On Ki67 And Cytokeratin In Breast Cancer

BackgroundBreast cancer (BC) is the most common malignant tumor among females in the world. It has been widely accepted by the oncology community that BC is a highly heterogeneous disease, with different biological behaviors for the same stage of BC among different patients.As the result of continuous insights into the BC molecular biology, personalized therapy based on the different tumor molecular classifications has become the main direction in BC treatment. The prerequisite for personalized medicine is the correct understanding of prognostic and predictive parameters, which remains an important and active research field in BC management.With the development of molecular bio-medicine, many BC-related molecules have been explored, and the most commonly used markers in clinical practice include human epidermal growth factor receptor-2 (HER2), estrogen receptor (ER), progesterone receptor (PR),and Ki67. Among them, Ki67, an antigen discovered in the early 1980s has been reported to have close relationship with the prognosis of BC.Ki67 is a nuclear antigen originally identified by Gerdes et al in the early 1980s, by using a mouse monoclonal antibody against a nuclear antigen from a Hodgkin lymphoma-derived cell line. The antigen named after the researcher’s location, Ki for Kiel University, Germany, with the 67 label referring to the clone number on the 96-well plate. Ki67 expression varies with different cell cycle phases. Cells express Ki67 during G1, S, G2, and mitotic phases, but not during the resting phase GO. Ki67 levels are low in G1 and S phases, but rise to a peak level in mitotic phase. In later periods (anaphase and telophase) of mitotic phase,Ki67 levels decrease sharply. As a wildly used proliferation marker, Ki67 has attracted increasing attention in recent years.Many studies showed that Ki67 expression levels were negatively correlated with BC prognosis. However, it is difficult to manually measure the Ki67 expression accurately and objectively, significantly limiting its clinical applications. Automated counting of Ki67 by computer software could be helpful, however,there is no evidence to show it is better than manual counting. And the results obtained by automated counting with traditional computer software is the sum of Ki67 expression, not the percentage of Ki67 which may be influenced by the number of cancer cells. Thus, it is urgent to develop a more accurate and objective method to evaluate the Ki67 expression in BC.Quantum dots (QDs) are fluorescent nanoparticles with unique size and surface effects which have been wildly used in bio-imaging, targeting and detecting due to their excellent properties, such as high fluorescence intensity, strong resistance to photobleaching and chemical degradation, size-tunable emission wavelength and simultaneous multiple fluorescent colors under single source excitation. In this study, we aimed to establish a QDs-based quantitative and in situ multiple imaging on Ki67 and cytokeratin (CK, a kind of epithelial specific marker used to label cancer cells in this study), so as to better evaluate their impacts on BC prognosis. In order to evaluate Ki67 expression more accurately and objectively,the Ki67 percentage was replaced by Ki67/CK ratio.ObjectiveThe aim of this study is to establish a QDs-based quantitative and real-time, simultaneous, in situ multiple imaging on Ki67 and CK, so as to better evaluate their impacts on BC prognosis.MethodsA comprehensive BC database has been established at our cancer center. The TMAs included 240 BC specimens (480 cores,1.5 mm each core) with five years follow-up. Complete clinico-pathological information on the patients including major clinico-pathological information including age, menopausal status, T stage, N stage, histological grade, ER and HER2 gene status was recorded in detail. As this work was based on our systematically established database, ER and HER2 status were determined by the expert pathologists in clinical laboratories based on the most updated pathological guidelines, immunohistochemical (IHC) for ER and fluorescent in-situ hybridization for HER2.A QDs-based in situ multiple fluorescent imaging method was developed. Ki67 immunostaining was performed using anti-Ki67 rabbit monoclonal antibody,biotinylated anti-rabbit IgG and streptavidin conjugated QDs-605 were applied to develop the red fluorescence of Ki67 at cell nucleus. CK immunostaining was performed using anti-CK mouse monoclonal antibody,QDs-525 conjugated goat anti-mouse IgG was applied to develop the green fluorescence of CK at the cytoplasm. QDs fluorescence information on Ki67 and CK was simultaneously obtained based on 6 photographs for each TMAs core In order to validate the above imaging technique and compare the technical features of different imaging methods in BC, we conducted direct comparisons of three imaging methods,i.e. conventional hematoxylin and eosin (HE) staining, IHC staining of CK and Ki67 separately, and QDs-based multiple imaging of CK and Ki67 simultaneously.After imaging acquisition, the unmixing and quantification on Ki67 and CK were performed by the software package within CRi Nuance multispectral imaging system,both Ki67 and CK values, and Ki67/CK ratio were obtained for each patient.The X-tile software was applied to automatically judge the cut-off points of Ki67 sum and Ki67/CK ratio to assess biomarkers and find optimal cut-point of biomarkers based on the outcome and divided them into two categories, designated as grade I and grade II according their values, respectively.Statistical analysis was performed with SPSS 17.0 software. Correlation test was calculated by Pearson chi-square. Five-year disease free survival (5-DFS) was the primary end-point and determined by the Kaplan-Meier method and analyzed by the logrank test. Multivariate analysis was performed by the Cox proportional hazards model. Receiver operating characteristic (ROC) curve analysis was used to calculate the predictive value of the independent prognosis factors.ResultsWe first compared the technical features of different imaging methods in BC. HE staining can show the histological structures of tumor, with cancer cells compacted into tumor nests invading the stroma, but the invasion edge of tumor nests sometimes could not be clearly revealed by such technique, because there is no definite image differentiation between the tumor and the stroma. IHC staining of Ki67 can only label Ki67 clearly, but could not clearly show tumor nests, rendering it technically difficult to precisely quantify the percentage of Ki67 in the tumor nests. Likewise, IHC staining of CK can label tumor nests clearly, but could not label Ki67 well. In contrast, QDs-based multiple imaging on CK and Ki67 can simultaneously label Ki67 and tumor nests well, with clear imaging differentiation between tumor nests and the stroma and between CK and Ki67.QDs-based multiple imaging simultaneously stains nuclear Ki67 and cytoplasmic CK with clear signal contrast,making it easy for signal separation and quantification. The X-tile software based on the best P valueprinciple was adopted to automatically judge the cut-off points of Ki67 sum and Ki67/CK ratio as 3.58×107 and 0.186, using 5-DFS as a prognosticator. Subsequently, Ki67 sum and Ki67/CK ratio each was divided into two grades based on these cut-off points.Attentions were also paid to the patterns of Ki67 expression in BC tissues. The absolute expression of Ki67 was not related to the proliferation, but the proportion of Ki67 positivity among the tumor tissue correlated with the proliferation activity. Ki67 is mainly expressed in cancer cell nucleus with few exceptions in stroma but not in blood vessel. Ki67 expression is higher in invasive ductal carcinoma nests than in situ carcinoma nests. For carcinoma in situ, Ki67 expression is more prominent at the tumor nest periphery than at the center.For invasive carcinoma, haphazard Ki67 expression is observed in the disorganized tumor nests.High Ki67 expression is observed at the tumor nests invasion edges both for carcinoma in situ with minimal invasion and invasive ductal carcinoma. When a big tumor nest produces smaller seeding tumor nests, high Ki67 expression is observed at the conjunction between the two tumor nests.Ki67 grade was negatively related with 5-DFS in overall patients and lymph node-positive subgroup, but not in lymph node-negative subgroup. Ki67/CK grade was negatively related with 5-DFS in overall patients, lymph node-negative subgroup and lymph node-positive subgroup. Ki67 grade or Ki67/CK grade and traditional prognostic parameters that had been verified by Kaplan-Meier survival analysis were subjected to Cox proportional hazards model analysis. The hazard ratio (HR) of Ki67 /CK grade (HR 2.019,95%CI [1.254-3.251]) was higher than N stage (HR 1.819, 95%CI [1.501-2.204]) and HER2 gene (HR 1.748,95%CI [1.087-2.812]), but lower than the histological grade (HR 3.370,95%CI [1.125-5.364]). Moreover, the significance and HR of Ki67/CK grade (P= 0.004; HR 2.019,95%CI [1.254-3.251]) was higher than Ki67 grade (P= 0.047; HR 1.773,95%CI [1.009-3.117]). Furthermore, area under curve (AUC) of ROC analysis for Ki67/CK grade (AUC: 0.683,95%CI:0.613-0.752) was higher than Ki67 grade (AUC:0.665,95%CI: 0.596-0.734) and HER-2 gene (AUC:0.586,95%CI:0.510-0.661), but lower than N stage (AUC:0.760,95%CI:0.696-0.823) and histological grade (AUC:0.756,95%CI: 0.692-0.820) on predicting the risk for recurrence.ConclusionUsing QDs-based multiple molecular imaging technology to evaluate Ki67 and CK expression in BC tissues, this study demonstrated that Ki67/CK grade may be better than Ki67 grade in predicting BC prognosis.