How to Sort Circulating Tumor Cells Part III: Microscopic Characterization

x-posted at Erica Pratt’s Blog


CTC: Circulating tumor cell. Read about what they are and why they’re important here.
Cell Fixation: Chemical preservation of a cell. Post-fixation, a cell is no longer alive.
Cell Staining: Using different markers to visualize cells, or components of cells.
Fluorophore: A fluorescent molecule that is excited at one wavelength of light (excitation), and emits light at another wavelength (emission).
Leukocyte: White blood cell (WBC).
Pathology Slide: A fixed section of unhealthy tissue that can be analyzed with various cell visualization markers.
Phenotype: Observable characteristics of a cell.
RBC: Red blood cell or erythrocyte
Systemic disease: Disease that has spread throughout the body.

Why use microscopic characterization to identify CTCs?

CTCs, which are shed from tumors into the vasculature, are considered to be key players in metastasis, and ultimately cancer patient death. Therefore, the goal of many CTC isolation systems is to separate these abnormal (cancer) cells from normal (blood) cells. Post-capture, analyses can be performed to analyze morphological differences between individual CTCs. Additionally, more researchers are investigating an underpinning assumption in CTC isolation: are the blood cells of a patient with systemic disease normal? One method used to answer that question is high-resolution, microscopic characterization of blood samples.

This technique is unique when compared to size– or immunocapture– based sorting I described previously. Microscopic identification of CTCs does not rely on physically separating them from native blood cells; instead, it uses imaging in combination with rapid scanning to look at almost all cells present in a blood sample. Many devices use microscopy to identify CTCs post-capture, but view other blood cells as contaminants to be identified so as not to confound results. Microscopic characterization aims to look at CTCs and other blood cells to further understand the pathology of the disease1-6. This method requires extensive image processing, and cell categorization algorithms.

What platforms are used for microscopic CTC characterization?

Most techniques focused on blood cell and CTC characterization have an initial stage to remove RBCs and small WBCs, either through lysis or filtration. The remaining cells are fixed (killed), and analyzed for a number of distinguishing characteristics. I will divide techniques by the substrate used, either non-porous surfaces (i.e. glass slides), or polymer porous surfaces (i.e. microfilters).

Non-porous Surfaces, such as modified glass slides1,2,3, are used to deposit and fix blood samples after minimal pre-processing (usually RBC lysis). Multiple slides can be produced from one 10mL blood sample3, enabling researchers and clinicians to perform multiple assays on one blood draw with slides left over for storage. The example to the left is an x-y plane image of stained pancreatic CTCs and a 3-D reconstruction using multiple x-y plane images at different depths in the sample (z).

Porous Surfaces use devices like microfilters4,5,6 to eliminate RBCs and small leukocytes prior to cell fixation and analysis. Cells are captured at regular intervals along the filter, making it simple to create a registry system to store unique information about each isolated cell. Multiple filters are used for one blood draw5, enabling the same analyses and storage as in the non-porous surface case.

Both of these techniques allow for morphological analysis of different types of captured cells, while also producing images very similar to standard pathology slides, making them attractive for clinical implementation.


What are the drawbacks of microscopic characterization?

While being able to identify CTCs to count them is clinically informative, the next logical step is to ask what is going on inside the CTCs. The example I showed above used three fluorescent probes to identify three cell characteristics: presence of a nucleus (blue), a leukocyte cell surface marker (green), and a CTC cell surface marker (red). I plotted the emission spectra of the three fluorophores used below, where the x-axis is relative intensity versus the y-axis which is emission wavelength.

As you can see, there are areas where the emission spectra touch, called spectral overlap. One artifact of spectral overlap that affects cell imaging is bleed-through, when the signal from one line “bleeds” to another. This could, for example, make a cell appear to be a leukocyte (red line), when in reality it is simply bleed-through from the nuclear label (green line). Fluorophores have a limited wavelength range, so as you try to look for more and more things in the cell, spectral overlap becomes a bigger and bigger problem. This is amplified in microscopic characterization because more cell types are captured, so more flurophores are needed simply for cell identification.

Thankfully, there are various engineering strategies, for example multi-spectral imaging, being employed to deconvolute very tight emission spectra. This allows researchers to use more colors, and therefore look for more things going on inside the cell.

Additionally, in microscopic characterization cell fixation and staining is required to identify CTCs from billions of other nucleated blood cells. This increases storage potential of blood samples, and allows you to answer many genetic questions, but also means you can’t ask questions about what a live CTC does.

When should you use microscopic characterization?

Microscopic characterization is an excellent tool for CTC analysis when:

  • You want something as close to an unbiased technique as possible.
    There is little to no pre-selection of CTCs, minimizing the risk of missing an unknown phenotype.
  • You want to look not only at CTCs, but at surrounding cells that may be affected by the metastasis process.
  • You want to minimize mechanical and fluid shear forces on blood samples.
  • You do not require live CTCs for biological or functional assays.


1. Marrinucci et al. Circulating tumor cells from well-differentiated lung adenocarcinoma retain cytomorphologic features of primary tumor type. Archives of Pathology & Laboratory Medicine 2009; 133(9):1468-1471

2. Marrinucci et al. Cytomorphology of circulating colorectal tumor cells: a small case series. Journal of Oncology 2010: 861341

3. Cho et al. Characterization of circulating tumor cell aggregates identified in patients with epithelial tumors. Physical Biology 2012; 9:016001

4. Farace et al. A direct comparison of CellSearch and ISET for circulating tumour-cell detection in patients with metastatic carcinomas. British Journal of Cancer 2011; 105(6):847-853

5. Hofman et al. Cytopathologic Detection of Circulating Tumor Cells Using the Isolation by Size of Epithelial Tumor Cell Method. American Journal of Clinical Pathology 2011; 135(1):146-156

6. Pinzani et al. Isolation by size of epithelial tumor cells in peripheral blood of patients with breast cancer: correlation with real-time reverse transcriptase polymerase chain reaction results and feasibility of molecular analysis by laser microdissection. Human Pathology 2006; 37(6):711-718