How to Sort Circulating Tumor Cells Part I: Size

x-posted at Erica Pratt’s blog


Leukocyte: White blood cell.
Erythrocyte: Red blood cell.
Thrombocyte: Platelets, important in the formation of blood clots.
Reynolds number: is the ratio of inertial to viscous fluid forces. In pressure-driven flow, this is the ratio of the pressure force applied (to actuate flow) as compared to the resistance of the fluid to deformation or shearing. You can read more about the derivation of the Reynolds number here.

Why sort CTCs based on size?

Many researchers have anecdotally observed that CTCs can be larger than other cells in the blood (e.g. leukocytes, erythrocytes, thrombocytes). A commonly quoted range for CTCs is 12-25 microns1, which is bigger than 90-95% of the largest blood cell population, leukocytes. For this reason, size-based sorting is an attractive, label-free, isolation method if the ratio between CTC and blood cell size is large, and if CTC biochemical properties are not well understood.

What types of engineering techniques are used?

The three types of size-based sorting of cancer cells I’ll describe today (in order of frequency of use) are centrifugation, microfilters, and hydrodynamic sorting. Centrifugation is the current state of the art2,3,4, while microfilters1,5 and hydrodynamic sorting6,7 are engineering technologies developed more recently.

Centrifugation uses centrifugal force to separate the cells based on density, resulting in distinct zones of blood components as one moves from the top to the bottom of the centrifugation tube. CTCs, leukocytes, and thrombocytes end up in a zone called the buffy coat, where they can then be extracted and analyzed. There are different types of media (i.e. fluids) that are synthesized to optimize this process; two common clinical density centrifugation medias are Ficoll and Oncoquick.

Microfilters isolate cells based strictly on size, or a combination of size and cell deformability. Some devices use polymers with microfabricated pores or cavities to act as miniature sieves for cells of above a certain size1,5,8. Other devices use geometric obstacles that cancer cells have to squeeze through, systematically isolating cells based on both their deformability and size6. An example of a microfilter from Zheng et al. is shown to the left.

Hydrodynamic sorting uses low Reynolds number fluid flow in combination with microdevice geometries6, or parallel fluid flows of different flow rates7, to passively sort or separate cancer cells. An advantage of using hydrodynamic sorting is this technique exerts a low fluid stress on the cell instead than physical compression through a micro-pore or -cavity. An example of a hydrodynamic sorter using a technique called dean flow (movie!) is shown from Bhagat et a. Breast cancer cells are directed towards the outlet at the bottom while the rest of the blood exits the outlet at the top.

Both microfiltration and microfluidic flow fractionation are highly efficient at isolating cancer cells (i.e. high sensitivity on a ROC curve).

What are the drawbacks for sorting based on size?

As you might suspect, sorting based on size is a challenging engineering problem. Strictly size-based devices are generally very efficient at recovering cancer cells present in solution (high sensitivity), but also capture a lot of other cells (low specificity). The major contaminating features in these systems are the blood cells closest in size, leukocytes. Why is that, when I stated at the beginning for the post that CTCs are larger than 90-95% of the leukocyte population?

If we breakdown leukocytes by size distribution and population size, we see that roughly 90-95% of leukocytes are under 15 microns. However, 2-10% of the leukocyte population (i.e. monocytes), are roughly identical in size to CTCs, but exist at nearly 10,000 higher fold concentration! If you put this in terms of a 1 milliter blood sample, that means for every 10-100 CTCs, there are about 1,000,000 identically sized leukocytes floating around.

A secondary issue with size-based sorting techniques is avoiding device clogging, since blood is a very dense suspension. However, this can be circumvented with clever devise design (e.g. cross flow filters9) or dilution of blood before processing.

When should you sort CTCs based on size?

Sized-based sorting is an excellent capture/sorting method when:

  • The size differential between the blood cell population and CTCs is large.
    This can be informed by cell size in the tissue of cancer origin, average size of relevant
    cancer cell lines, or anecdotal observation of actual CTCs.
  • When you want to ensure capture of all target cells (high sensitivity), and are not sensitive to contaminating blood cells (low specificity).
  • If you want a method that is unbiased by changes in cell surface properties (e.g. protein expression).


1. Zheng et al. Membrane microfilter device for selective capture, electrolysis and genomic analysis of human circulating tumor cells. Journal of Chromatography A 2007; 1162: 154-161

2. Könisberg et al. Circulating tumor cells in metastatic colorectal cancer: Efficacy and feasibility of different enrichment methods. Cancer Letters 2010; 293(1): 117-124

3. Rosenberg et al. Detection of circulating tumor cells in blood using an optimized density gradient centrifugation. Recent Results in Cancer Research 2003; 162:149-155

4. Ghossein et al. Detection of circulating tumor cells in patients with localized and metastatic prostatic carcinoma: clinical implications. Journal of Clinical Oncology 1995; 15(3): 1195-1200

5. Lin et al. Portable Filter-Based Microdevice for Detection and Characterization of Circulating Tumor Cells. Clinical Cancer Research 2010

6. Bhagat et al. Dean Flow Field Fractionation (DFF) Isolation of Circulating Tumor Cells (CTCs) from Blood. 15th International Conference on Miniaturized Systems for Chemistry and Life Sciences (MicroTAS) October 2-6, 2011, Seattle, Washington, USA

7. Moon et al. Continuous separation of breast cancer cells from blood samples using multi-orifice flow fractionation (MOFF) and dielectrophoresis (DEP). Lab on a Chip 2011; 11:1118-1125

8. Hosokawa et al. Size-selective microcavity array for rapid and efficient detection of circulating tumor cells. Analytical Chemistry 2010; 82(15):6629-6635

9. Ji et al. Silicon-based microfilters for whole blood cell separation. Biomedical Microdevices 2008, 10:251–257