Automated identification of circulating tumor cells by image cytometry

Correlative Multi-Modal Microscopy: A Novel Pipeline for Optimizing Fluorescence Microscopy Resolutions in Biological Applications

The modern fluorescence microscope is the convergence point of technologies with different performances in terms of statistical sampling, number of simultaneously analyzed signals, and spatial resolution. However, the best results are usually obtained by maximizing only one of these parameters and finding a compromise for the others, a limitation that can become particularly significant when applied to cell biology and that can reduce the spreading of novel optical microscopy tools among research laboratories. Super resolution microscopy and, in particular, molecular localization-based approaches provide a spatial resolution and a molecular localization precision able to explore the scale of macromolecular complexes in situ. However, its use is limited to restricted regions, and consequently few cells, and frequently no more than one or two parameters. Correlative microscopy, obtained by the fusion of different optical technologies, can consequently surpass this barrier by merging results from different spatial scales. We discuss here the use of an acquisition and analysis correlative microscopy pipeline to obtain high statistical sampling, high content, and maximum spatial resolution by combining widefield, confocal, and molecular localization microscopy.

ALICE: a hybrid AI paradigm with enhanced connectivity and cybersecurity for a serendipitous encounter with circulating hybrid cells

A fully automated and accurate assay of rare cell phenotypes in densely-packed fluorescently-labeled liquid biopsy images remains elusive. Methods: Employing a hybrid artificial intelligence (AI) paradigm that combines traditional rule-based morphological manipulations with modern statistical machine learning, we deployed a next generation software, ALICE (Automated Liquid Biopsy Cell Enumerator) to identify and enumerate minute amounts of tumor cell phenotypes bestrewed in massive populations of leukocytes. As a code designed for futurity, ALICE is armed with internet of things (IOT) connectivity to promote pedagogy and continuing education and also, an advanced cybersecurity system to safeguard against digital attacks from malicious data tampering. Results: By combining robust principal component analysis, random forest classifier and cubic support vector machine, ALICE was able to detect synthetic, anomalous and tampered input images with an average recall and precision of 0.840 and 0.752, respectively. In terms of phenotyping enumeration, ALICE was able to enumerate various circulating tumor cell (CTC) phenotypes with a reliability ranging from 0.725 (substantial agreement) to 0.961 (almost perfect) as compared to human analysts. Further, two subpopulations of circulating hybrid cells (CHCs) were serendipitously discovered and labeled as CHC-1 (DAPI+/CD45+/E-cadherin+/vimentin-) and CHC-2 (DAPI+ /CD45+/E-cadherin+/vimentin+) in the peripheral blood of pancreatic cancer patients. CHC-1 was found to correlate with nodal staging and was able to classify lymph node metastasis with a sensitivity of 0.615 (95% CI: 0.374-0.898) and specificity of 1.000 (95% CI: 1.000-1.000). Conclusion: This study presented a machine-learning-augmented rule-based hybrid AI algorithm with enhanced cybersecurity and connectivity for the automatic and flexibly-adapting enumeration of cellular liquid biopsies. ALICE has the potential to be used in a clinical setting for an accurate and reliable enumeration of CTC phenotypes.

Tumor-Cell-Macrophage Fusion Cells as Liquid Biomarkers and Tumor Enhancers in Cancer

Although molecular mechanisms driving tumor progression have been extensively studied, the biological nature of the various populations of circulating tumor cells (CTCs) within the blood is still not well understood. Tumor cell fusion with immune cells is a longstanding hypothesis that has caught more attention in recent times. Specifically, fusion of tumor cells with macrophages might lead to the development of metastasis by acquiring features such as genetic and epigenetic heterogeneity, chemotherapeutic resistance, and immune tolerance. In addition to the traditional FDA-approved definition of a CTC (CD45-, EpCAM+, cytokeratins 8+, 18+ or 19+, with a DAPI+ nucleus), an additional circulating cell population has been identified as being potential fusions cells, characterized by distinct, large, polymorphonuclear cancer-associated cells with a dual epithelial and macrophage/myeloid phenotype. Artificial fusion of tumor cells with macrophages leads to migratory, invasive, and metastatic phenotypes. Further studies might investigate whether these have a potential impact on the immune response towards the cancer. In this review, the background, evidence, and potential relevance of tumor cell fusions with macrophages is discussed, along with the potential role of intercellular connections in their formation. Such fusion cells could be a key component in cancer metastasis, and therefore, evolve as a diagnostic and therapeutic target in cancer precision medicine.

Detection of Androgen Receptor Variant 7 (ARV7) mRNA Levels in EpCAM-Enriched CTC Fractions for Monitoring Response to Androgen Targeting Therapies in Prostate Cancer

Expression of the androgen receptor splice variant 7 (ARV7) in circulating tumor cells (CTCs) has been associated with resistance towards novel androgen receptor (AR)-targeting therapies. While a multitude of ARV7 detection approaches have been developed, the simultaneous enumeration of CTCs and assessment of ARV7 status and the integration of validated technologies for CTC enrichment/detection into their workflow render interpretation of the results more difficult and/or require shipment to centralized labs. Here, we describe the establishment and technical validation of a novel ARV7 detection method integrating the CellSearch^® technology, the only FDA-cleared CTC-enrichment method for metastatic prostate cancer available so far. A highly sensitive and specific qPCR-based assay was developed, allowing detection of ARV7 and keratin 19 transcripts from as low as a single ARV7⁺/K19⁺ cell, even after 24 h of sample storage. Clinical feasibility was demonstrated on blood samples from 26 prostate cancer patients and assay sensitivity and specificity was corroborated. Our novel approach can now be included into prospective clinical trials aimed to assess the predictive values of CTC/ARV7 measurements in prostate cancer.

Clinical applications of the CellSearch platform in cancer patients

The CellSearch® system (CS) enables standardized enrichment and enumeration of circulating tumor cells (CTCs) that are repeatedly assessable via non-invasive "liquid biopsy". While the association of CTCs with poor clinical outcome for cancer patients has clearly been demonstrated in numerous clinical studies, utilizing CTCs for the identification of therapeutic targets, stratification of patients for targeted therapies and uncovering mechanisms of resistance is still under investigation. Here, we comprehensively review the current benefits and drawbacks of clinical CTC analyses for patients with metastatic and non-metastatic tumors. Furthermore, the review focuses on approaches beyond CTC enumeration that aim to uncover therapeutically relevant antigens, genomic aberrations, transcriptional profiles and epigenetic alterations of CTCs at a single cell level. This characterization of CTCs may shed light on the heterogeneity and genomic landscapes of malignant tumors, an understanding of which is highly important for the development of new therapeutic strategies.

Statistical performance of image cytometry for DNA, lipids, cytokeratin, & CD45 in a model system for circulation tumor cell detection

Detection of circulating tumor cells (CTCs) in a blood sample is limited by the sensitivity and specificity of the biomarker panel used to identify CTCs over other blood cells. In this work, we present Bayesian theory that shows how test sensitivity and specificity set the rarity of cell that a test can detect. We perform our calculation of sensitivity and specificity on our image cytometry biomarker panel by testing on pure disease positive (D⁺ ) populations (MCF7 cells) and pure disease negative populations (D^- ) (leukocytes). In this system, we performed multi-channel confocal fluorescence microscopy to image biomarkers of DNA, lipids, CD45, and Cytokeratin. Using custom software, we segmented our confocal images into regions of interest consisting of individual cells and computed the image metrics of total signal, second spatial moment, spatial frequency second moment, and the product of the spatial-spatial frequency moments. We present our analysis of these 16 features. The best performing of the 16 features produced an average separation of three standard deviations between D⁺ and D^- and an average detectable rarity of ∼1 in 200. We performed multivariable regression and feature selection to combine multiple features for increased performance and showed an average separation of seven standard deviations between the D⁺ and D^- populations making our average detectable rarity of ∼1 in 480. Histograms and receiver operating characteristics (ROC) curves for these features and regressions are presented. We conclude that simple regression analysis holds promise to further improve the separation of rare cells in cytometry applications. © 2017 International Society for Advancement of Cytometry.

Automated Classification of Circulating Tumor Cells and the Impact of Interobsever Variability on Classifier Training and Performance

Application of personalized medicine requires integration of different data to determine each patient's unique clinical constitution. The automated analysis of medical data is a growing field where different machine learning techniques are used to minimize the time-consuming task of manual analysis. The evaluation, and often training, of automated classifiers requires manually labelled data as ground truth. In many cases such labelling is not perfect, either because of the data being ambiguous even for a trained expert or because of mistakes. Here we investigated the interobserver variability of image data comprising fluorescently stained circulating tumor cells and its effect on the performance of two automated classifiers, a random forest and a support vector machine. We found that uncertainty in annotation between observers limited the performance of the automated classifiers, especially when it was included in the test set on which classifier performance was measured. The random forest classifier turned out to be resilient to uncertainty in the training data while the support vector machine's performance is highly dependent on the amount of uncertainty in the training data. We finally introduced the consensus data set as a possible solution for evaluation of automated classifiers that minimizes the penalty of interobserver variability.

In vivo cell characteristic extraction and identification by photoacoustic flow cytography

We present a photoacoustic flow cytography with fast cross-sectional (B-scan) imaging to precisely identify specific cells in vivo. The B-scan imaging speed of the system is up to 200 frame/s with a lateral resolution of 1.5 μm, which allows to dynamically image the flowing cells within the microvascular. The shape, size and photoacoustic intensity of the target cells are extracted from streaming images and integrated into a standard pattern to distinguish cell types. Circulating red blood cells and melanoma cells in blood vessels are simultaneously identified on melanoma-bearing mouse model. The results demonstrate that in vivo photoacoustic flow cytography can provide cells characteristics analysis and cell type's visual identification, which will be applied for noninvasively monitoring circulating tumor cells (CTCs) and analyzing hematologic diseases.

Apoptotic circulating tumor cells (CTCs) in the peripheral blood of metastatic colorectal cancer patients are associated with liver metastasis but not CTCs

Enumeration of circulating tumor cells (CTCs) by the CellSearch system provides prognostic information in metastatic colorectal cancer, regardless of metastatic site. We found that CTCs generally represent <1% of observed events with CellSearch analysis and adapted scoring criteria to classify other peripheral blood events. Examination of twenty two metastatic colorectal cancer patients' blood revealed that patients with high CEA or liver metastases, but not lung or distant lymph node metastases, possessed significant numbers of apoptotic CTCs prior to treatment initiation by Fischer's exact test. Six out of eleven patients with liver metastasis possessed apoptotic CTCs whereas one of nine patients with other metastases had measurable apoptotic CTCs. An elevated CTC number was not necessarily associated with apoptotic CTCs or CTC debris by Spearman's correlation, suggesting the metastatic site rather than CTCs per se as contributing to the origin of these events.

The heat-transfer method: a versatile low-cost, label-free, fast, and user-friendly readout platform for biosensor applications

In recent years, biosensors have become increasingly important in various scientific domains including medicine, biology, and pharmacology, resulting in an increased demand for fast and effective readout techniques. In this Spotlight on Applications, we report on the recently developed heat-transfer method (HTM) and illustrate the use of the technique by zooming in on four established bio(mimetic) sensor applications: (i) mutation analysis in DNA sequences, (ii) cancer cell identification through surface-imprinted polymers, (iii) detection of neurotransmitters with molecularly imprinted polymers, and (iv) phase-transition analysis in lipid vesicle layers. The methodology is based on changes in heat-transfer resistance at a functionalized solid-liquid interface. To this extent, the device applies a temperature gradient over this interface and monitors the temperature underneath and above the functionalized chip in time. The heat-transfer resistance can be obtained by dividing this temperature gradient by the power needed to achieve a programmed temperature. The low-cost, fast, label-free and user-friendly nature of the technology in combination with a high degree of specificity, selectivity, and sensitivity makes HTM a promising sensor technology.

Cytometry in the brain: studying differentiation to diagnostic applications in brain disease and regeneration therapy

During brain development, a population of uniform embryonic cells migrates and differentiates into a large number of neural phenotypes - origin of the enormous complexity of the adult nervous system. Processes of cell proliferation, differentiation and programmed death of no longer required cells, do not occur only during embryogenesis, but are also maintained during adulthood and are affected in neurodegenerative and neuropsychiatric disease states. As neurogenesis is an endogenous response to brain injury, visible as proliferation (of to this moment silent stem or progenitor cells), its further stimulation can present a treatment strategy in addition to stem cell transfer for cell regeneration therapy. Concise techniques for studying such events in vitro and in vivo permit understanding of underlying mechanisms. Detection of subtle physiological alterations in brain cell proliferation and neurogenesis can be explored, that occur during environmental stimulation, exercise and ageing. Here, we have collected achievements in the field of basic research on applications of cytometry, including automated imaging for quantification of morphological or fluorescence-based parameters in cell cultures, towards imaging of three-dimensional brain architecture together with DNA content and proliferation data. Multi-parameter and more recently in vivo flow cytometry procedures, have been developed for quantification of phenotypic diversity and cell processes that occur during brain development as well as in adulthood, with importance for therapeutic approaches.

Isolation and mutational analysis of circulating tumor cells from lung cancer patients with magnetic sifters and biochips

Detection and characterization of circulating tumor cells (CTCs) may reveal insights into the diagnosis and treatment of malignant disease. Technologies for isolating CTCs developed thus far suffer from one or more limitations, such as low throughput, inability to release captured cells, and reliance on expensive instrumentation for enrichment or subsequent characterization. We report a continuing development of a magnetic separation device, the magnetic sifter, which is a miniature microfluidic chip with a dense array of magnetic pores. It offers high efficiency capture of tumor cells, labeled with magnetic nanoparticles, from whole blood with high throughput and efficient release of captured cells. For subsequent characterization of CTCs, an assay, using a protein chip with giant magnetoresistive nanosensors, has been implemented for mutational analysis of CTCs enriched with the magnetic sifter. The use of these magnetic technologies, which are separate devices, may lead the way to routine preparation and characterization of "liquid biopsies" from cancer patients.

Detection and isolation of circulating tumor cells: principles and methods

Efforts to improve the clinical management of several cancers include finding better methods for the quantitative and qualitative analysis of circulating tumor cells (CTCs). However, detection and isolation of CTCs from the blood circulation is not a trivial task given their scarcity and the lack of reliable markers to identify these cells. With a variety of emerging technologies, a thorough review of the exploited principles and techniques as well as the trends observed in the development of these technologies can assist researchers to recognize the potential improvements and alternative approaches. To help better understand the related biological concepts, a simplified framework explaining cancer formation and its spread to other organs as well as how CTCs contribute to this process has been presented first. Then, based on their basic working-principles, the existing methods for detection and isolation of CTCs have been classified and reviewed as nucleic acid-based, physical properties-based and antibody-based methods. The review of literature suggests that antibody-based methods, particularly in conjunction with a microfluidic lab-on-a-chip setting, offer the highest overall performance for detection and isolation of CTCs. Further biological and engineering-related research is required to improve the existing methods. These include finding more specific markers for CTCs as well as enhancing the throughput, sensitivity, and analytic functionality of current devices.

Selective identification of macrophages and cancer cells based on thermal transport through surface-imprinted polymer layers

In this article, we describe a novel straightforward method for the specific identification of viable cells (macrophages and cancer cell lines MCF-7 and Jurkat) in a buffer solution. The detection of the various cell types is based on changes of the heat transfer resistance at the solid-liquid interface of a thermal sensor device induced by binding of the cells to a surface-imprinted polymer layer covering an aluminum chip. We observed that the binding of cells to the polymer layer results in a measurable increase of heat transfer resistance, meaning that the cells act as a thermally insulating layer. The detection limit was found to be on the order of 10(4) cells/mL, and mutual cross-selectivity effects between the cells and different types of imprints were carefully characterized. Finally, a rinsing method was applied, allowing for the specific detection of cancer cells with their respective imprints while the cross-selectivity toward peripheral blood mononuclear cells was negligible. The concept of the sensor platform is fast and low-cost while allowing also for repetitive measurements.

A label-free DC impedance-based microcytometer for circulating rare cancer cell counting

Quantification of circulating tumor cells (CTCs) in blood samples is believed to provide valuable evidence of cancer progression, cancer activity status, response to therapy in patients with metastatic cancer, and possible cancer diagnosis. Recently, a number of researchers reported that CTCs tend to lose their epithelial cell adhesion molecule (EpCAM) by an epithelial-mesenchymal transition (EMT). As such, label-free CTC detection methods are attracting worldwide attention. Here, we describe a label-free DC impedance-based microcytometer for CTCs by exploiting the difference in size between CTCs and blood cells. This system detects changes in DC impedance between two polyelectrolytic gel electrodes (PGEs) under low DC voltages. Using spiked ovarian cancer cell lines (OVCAR-3) in blood as a model system, we were able to count the cells using a microcytometer with 88% efficiency with a flow rate of 13 μl min(-1) without a dilution process. Furthermore, we examined blood samples from breast cancer patients using the cytometer, and detected CTCs in 24 out of 24 patient samples. Thus, the proposed DC impedance-based microcytometer presents a facile and fast way of CTC evaluation regardless of their biomarkers.

A computational platform for robotized fluorescence microscopy (I): high-content image-based cell-cycle analysis

Hardware automation and software development have allowed a dramatic increase of throughput in both acquisition and analysis of images by associating an optimized statistical significance with fluorescence microscopy. Despite the numerous common points between fluorescence microscopy and flow cytometry (FCM), the enormous amount of applications developed for the latter have found relatively low space among the modern high-resolution imaging techniques. With the aim to fulfill this gap, we developed a novel computational platform named A.M.I.CO. (Automated Microscopy for Image-Cytometry) for the quantitative analysis of images from widefield and confocal robotized microscopes. Thanks to the setting up of both staining protocols and analysis procedures, we were able to recapitulate many FCM assays. In particular, we focused on the measurement of DNA content and the reconstruction of cell-cycle profiles with optimal parameters. Standard automated microscopes were employed at the highest optical resolution (200 nm), and white-light sources made it possible to perform an efficient multiparameter analysis. DNA- and protein-content measurements were complemented with image-derived information on their intracellular spatial distribution. Notably, the developed tools create a direct link between image-analysis and acquisition. It is therefore possible to isolate target populations according to a definite quantitative profile, and to relocate physically them for diffraction-limited data acquisition. Thanks to its flexibility and analysis-driven acquisition, A.M.I.CO. can integrate flow, image-stream and laser-scanning cytometry analysis, providing high-resolution intracellular analysis with a previously unreached statistical relevance.

Comparative analysis of circulating endothelial progenitor cells in age-related macular degeneration patients using automated rare cell analysis (ARCA) and fluorescence activated cell sorting (FACS)

Background

Patients with age-related macular degeneration (ARMD) begin with non-neovascular (NNV) phenotypes usually associated with good vision. Approximately 20% of NNV-ARMD patients will convert to vision debilitating neovascular (NV) ARMD, but precise timing of this event is unknown. Developing a clinical test predicting impending conversion to NV-ARMD is necessary to prevent vision loss. Endothelial progenitor cells (EPCs), defined as CD34⁺VEGR2⁺ using traditional fluorescence activated cell sorting (FACS), are rare cell populations known to be elevated in patients with NV-ARMD compared to NNV-ARMD. FACS has high inter-observer variability and subjectivity when measuring rare cell populations precluding development into a diagnostic test. We hypothesized that automated rare cell analysis (ARCA), a validated and FDA-approved technology for reproducible rare cell identification, can enumerate EPCs in ARMD patients more reliably. This pilot study serves as the first step in developing methods for reproducibly predicting ARMD phenotype conversion.

Methods

We obtained peripheral venous blood samples in 23 subjects with NNV-ARMD or treatment naïve NV-ARMD. Strict criteria were used to exclude subjects with known angiogenic diseases to minimize confounding results. Blood samples were analyzed in masked fashion in two separate laboratories. EPCs were independently enumerated using ARCA and FACS within 24 hours of blood sample collection, and p<0.2 was considered indicative of a trend for this proof of concept study, while statistical significance was established at 0.05.

Results

We measured levels of CD34⁺VEGFR2⁺ EPCs suggestive of a trend with higher values in patients with NV compared to NNV-ARMD (p = 0.17) using ARCA. Interestingly, CD34⁺VEGR2⁺ EPC analysis using FACS did not produce similar results (p = 0.94).

Conclusions

CD34⁺VEGR2⁺ may have predictive value for EPC enumeration in future ARCA studies. EPC measurements in a small sample size were suggestive of a trend in ARMD using ARCA but not FACS. ARCA could be a helpful tool for developing a predictive test for ARMD phenotype conversion.