Clinical Microfluidic Chip Platform for the Isolation of Versatile Circulating Tumor Cells

Summary

The clinical microfluidic chip is an important biomedical analysis technique that simplifies clinical patient blood sample preprocessing and immunofluorescently stains circulating tumor cells (CTCs) in situ on the chip, allowing the rapid detection and identification of a single CTC.

Abstract

Circulating tumor cells (CTCs) are significant in cancer prognosis, diagnosis, and anti-cancer therapy. CTC enumeration is vital in determining patient disease since CTCs are rare and heterogeneous. CTCs are detached from the primary tumor, enter the blood circulation system, and potentially grow at distant sites, thus metastasizing the tumor. Since CTCs carry similar information to the primary tumor, CTC isolation and subsequent characterization can be critical in monitoring and diagnosing cancer. The enumeration, affinity modification, and clinical immunofluorescence staining of rare CTCs are powerful methods for CTC isolation because they provide the necessary elements with high sensitivity. Microfluidic chips offer a liquid biopsy method that is free of any pain for the patients. In this work, we present a list of protocols for clinical microfluidic chips, a versatile CTC isolating platform, that incorporate a set of functionalities and services required for CTC separation, analysis, and early diagnosis, thus facilitating biomolecular analysis and cancer treatment. The program includes rare tumor cell counting, clinical patient blood preprocessing, which includes red blood cell lysis, and the isolation and recognition of CTCs in situ on microfluidic chips. The program allows the precise enumeration of tumor cells or CTCs. Additionally, the program includes a tool that incorporates CTC isolation with versatile microfluidic chips and immunofluorescence identification in situ on the chips, followed by biomolecular analysis.

Introduction

Circulating tumor cells (CTCs) are significant in cancer prognosis, diagnosis, and anti-cancer therapy. CTC enumeration is vital since CTCs are rare and heterogeneous. The enumeration, affinity modification, and clinical immunofluorescence staining of rare CTCs are powerful techniques for CTC isolation because they offer the necessary elements with high sensitivity¹. Rare number of tumor cells mixed with normal blood closely mimics real patient blood since 2-3 mL of real patient blood only contains 1-10 CTCs. To solve a critical experimental problem, instead of using a large number of tumor cells introduced in PBS or mixed with normal blood, the use of rare number of tumor cells provides us with a low number of blood cells, which is closer to reality when performing an experiment.

Cancer is the leading cause of death in the world². CTCs are tumor cells shed from the original tumor that circulate in the blood and lymphatic circulation systems³. When CTCs move to a new survivable environment, they grow as a second tumor. This is called metastasis and is responsible for 90% of deaths in cancer patients⁴. CTCs are vital for prognosis, early diagnosis, and for understanding the mechanisms of cancer. However, CTCs are extremely rare and heterogeneous in patient blood^5,6.

Microfluidic chips offer a liquid biopsy that does not invade the tumor. They have the advantage of being portable, low cost, and having a cell-matched scale. The isolation of CTCs with microfluidic chips is classified mainly into two types: affinity-based, which relies on antigen-antibody binding^7,8,9 and is the original and most widely used method of CTC isolation; and physical-based chips, which utilize size and deformability differences between tumor cells and blood cells^{10,11,12,13,14,15}, are label-free, and are easy to operate. The advantage of microfluidic chips over alternative techniques is that the physical-based approach of big-ellipse microfilters firmly captures CTCs with high capture efficiency. The reason for this is that ellipse microposts are organized into slim tunnels of line-line gaps. The line-line gaps are different from the traditional point-point gaps formed by microposts such as rhombus microposts. Wave chip-based capturing of CTCs combines both physical property-based and affinity-based isolation. Wave chip-based capture involves 30 wave-shaped arrays with the antibody of anti-EpCAM coated on circular microposts. The CTCs are captured by the small gaps, and the big gaps are used to accelerate the flow rate. The missed CTCs have to pass the small gaps in the next array and are captured by the affinity-based isolation integrated inside the chip¹⁶.

The goal of the protocol is to demonstrate the counting of rare numbers of tumor cells and the clinical analysis of CTCs with microfluidic chips. The protocol describes the CTC isolation steps, how to obtain a low number of tumor cells, the clinical physical separation of small-ellipse filters, big-ellipse filters, and trapezoid filters, affinity modification, and enrichment¹⁷.

Protocol

Patient blood samples were supplied by Longhua Hospital Affiliated to Shanghai Medical University.The protocol follows the guidelines of Peking University Third Hospital's human research ethics committee. Informed consent was obtained from the patients for using the samples for research purposes.

1. Pre-experiment to check the capture efficiency with cultured tumor cells

1. Culture the tumor cells MCF-7, MDA-MB-231, and HeLa for determining the capture efficiency. Dilute the tumor cell suspension, count the number of tumor cells, and repeat until the desired number of cells in 1 mL of PBS is obtained.
   1. Culture the cells in a cell culture flask with a starting cell number of ~ 1 x 10⁵ cells in 1 mL of Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin. Incubate in a humidified atmosphere at 37 °C with a 5% CO₂ atmosphere.
   2. When the cell lines grow as adherent monolayers to 95% confluence, detach them from the culture dishes with 0.25% trypsin solution for 2 min as described in Chen et al.¹⁶.
   3. Stain the tumor cells with calcein AM. Put 3 µL of calcein AM in the culture dish, and keep the dish in the incubator for 30 min. Then,digest all the cells with trypsin.
   4. Count the cultured tumor cells in PBS with a cell counting chamber, and dilute until 100 tumor cells per 1 mL of PBS are obtained.
      NOTE: In order to precisely enumerate cell number, take 50 µL of the obtained cell suspension to see whether the number of tumor cells in it is five or not with a microscope. Decide the volume of cell suspension that needs to be taken to obtain 100 tumor cells according to the actual number of tumor cells in 50 µL of cell suspension.
2. Introduce the cell suspension containing 100 tumor cells per 1 mL of PBS into the microfluidic chip using a syringe with a syringe pump at varied flow rates of 0.5 mL/h, 1 mL/h, 2 mL/h, 3 mL/h, 4 mL/h, and 5 mL/h. Obtain the capture efficiency for the various flow rates, and determine the optimal flow rate.
   1. Count the number of tumor cells captured on the chip and flowing out from the outlet. Calculate the capture efficiency as below:
      ​Capture efficiency = Number of cells captured/(Number of cells captured + number of cells flowing out) × 100%
   2. Repeat to obtain the capture efficiency for different numbers of tumor cells from 10 to 100 (10, 20, 30, 40, 50, 60, 70, 80, 90, 100).
3. Test and validate the microfluidic chip for rare numbers of tumor cells. Inject these 10 sample suspensions into the microfluidic chips using a hollow needle made with a micropipette puller to aspirate 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10 tumor cells diluted in 1 mL of PBS. Detect and enumerate the number of tumor cells for each sample after their capture on the chip.
   NOTE: The hollow needle is around 3 cm in length with a 1 mm external diameter.
4. Perform clinical pre-experiment testing for tumor cells spiked into normal blood samples.
   1. Stain the tumor cells with calcein AM, and then enumerate 100 tumor cells in 5 µL of PBS. Spike these cells into 1 mL of whole normal blood samples. Introduce these cells into the chip, and enumerate the number of tumor cells captured on the chip with green immunofluorescence. Perform in vivo enumeration on the chip as described above, and calculate the capture efficiency after capture.
   2. Repeat step 1.4.1 for nine additional concentrations of tumor cell numbers from 10 to 90 (10, 20, 30, 40, 50, 60, 70, 80, 90).
   3. Enumerate 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10 tumor cells as described in step 1.3 without staining, spike into 1 mL of whole normal blood, introduce these samples into the microfluidic chip, and capture the tumor cells on the chip.
   4. Stain on the chip with Hoechst, CK-FITC, and CD45-PE. Put 3-5 µL of fluorescent dye into 20-30 µL of PBS, and then introduce this solution onto the microfluidic chip with a syringe. Enumerate the number of tumor cells captured on the chip with both blue and green immunofluorescence to determine the capture efficiency.
5. Perform modification of the microfluidic chip for the affinity-based capture of anti-EpCAM.
   NOTE: Since the wave chip combines both affinity-based and physical property-based isolation, modify the chip with agents.
   1. Modify the surface of the chip with 100 µL of 4% (v/v) 3-mercaptopropyl trimethoxysilane in ethanol at room temperature for 45 min. Introduce this solution into the chip carefully and slowly in case it destroys the internal structure of the chip, especially the bonding of the top surface and substrate glass.
      NOTE: The chip surface is modified using mercaptosilane. The interactions that occur on the chip are determined by chemical bonds. Chemical bonds are established to realize the bonding of the antibody modified with the antigen. Various reagents are modified inside the chip to establish chemical bonds that connect with each other. The last reagent to be modified is an anti-epithelial adhesion molecule (anti-EpCAM). The tumor cell surface antigen of EpCAM combines with anti-EpCAM inside the chip to realize the capture of the CTCs.
   2. Wash with ethanol 3x. Add 100 µL of the coupling agent N-y-maleimidobutyryloxy succinimide ester (GMBS, 1 µM) onto the chip, and allow it to interact for 30 min.
   3. Wash with PBS 3x. Use a syringe to introduce 1 mL of PBS onto the chip to wash.
   4. Treat the chip with 30-40 µL of 10 µg/mL neutravidin at room temperature for 30 min, leading to immobilization of the cells onto the GMBS, and then flush with PBS to remove excess avidin.
   5. Modify the chip with 3 µL of anti-biotinylated EpCAM antibody at a concentration of 10 µg/mL in 100 µL of PBS with 1% (w/v) BSA. Keep this overnight.

2. Clinical experiment on the chip to enumerate the circulating tumor cells (CTCs)

1. Pre-process clinical cancer patient blood samples using red blood cells lysis (RBCL) solution, or introduce 2-3 mL of the blood sample directly onto the microfluidic chip using a syringe.
   1. Perform pre-processing, which takes around 30 min. Collect whole blood samples in anticoagulation tubes. Add 6-9 mL of RBS lysis solution into 2-3 mL of blood. Centrifuge at 111 x g for 5-8 min at room temperature, and discard the upper layer of red clear liquid.
2. Capture, stain, recognize, and enumerate the cells on the chip as described below.
   1. Capture the cells on the chip as described in step 1.Stain the chip for the CTCs captured using Hoechst, CK-FITC, and CD45-PE. CK-FITC is a specific stain for tumor cells, and CD45-PE is for white blood cells.
   2. Add 3 µL of CK-FITC to 20 µL of PBS. Introduce this into the syringe, and pump the diluted CK-FITC onto the chip. Allow to stain for 30 min. Introduce 300 µL of PBS onto the chip to wash the chip.
   3. Add 3 µL of CD45-PE to 20 µL of PBS. Introduce this into the syringe, and pump the diluted CD45-PE onto the chip. Allow to stand for 30 min. Introduce 300 µL of PBS onto the chip to wash the chip.
   4. Identify the CTCs with an inverted fluorescence microscope at 20x or 40x magnification. CTCs emit both blue and green fluorescence, and white blood cells (WBCs) emit both blue and red fluorescence. Identify the CTCs with both blue and green fluorescence and the WBCs with both blue and red fluorescence.
   5. From the immunofluorescence images, enumerate the number of CTCs captured on the chip. Enumerate the CTCs as Hoechst+/CK-FITC+/CD45− and the WBCs as Hoechst+/CK-FITC−/CD45+.
3. Enrich the captured CTCs by flushing with PBS through the syringe in the reverse direction onto the microfluidic chip to collect the CTCs captured on the chip from the inlet. Use a syringe with 1 mL of PBS, introduce it from the outlet to enrich the CTCs captured on the chip within 2-3 min, and collect them from the inlet. Use 1 mL of PBS for each washing step, and repeat 3x.
4. Stain the tumor cells or CTCs captured on the microfluidic chip as described below.
   1. Stain using calcein AM. Add 5 µL of calcein AM to 20 µL of PBS, stain the cells for 30 min, then centrifuge at 111 x g for 2 min at room temperature to obtain tumor cells, and suspend in 1 mL of PBS.
   2. To identify the cellular nuclei, add 20 µL of DAPI solution (10 µL of DAPI reagent in 20 µL of PBS) to the chip at the optimal flow rate of 1 mL/h for big-ellipse microfilters and 1.5 mL/h for trapezoid ones. Pass through 20 µL of anti-cytokeratin stock solution (3 µL of anti-cytokeratin antibody in 20 µL of PBS) to react with the chip for 30 min.
   3. Stain the WBCs captured on the chip. After the CTCs have been captured on the chip, add 25 µL of anti-CD45 antibody solution (5 µL of anti-CD45 solution in 20 µL of PBS) to the chip, and allow to stain for 30 min. Wash with PBS, and identify the epithelial cells with both blue and green fluorescence.
5. Perform immunofluorescence identification with fluorescence microscopy as described below.
   1. Use a fluorescence microscope and excite the sample with the blue laser. Find the cells emitting blue fluorescence, which are nucleate cells. Use one of the following magnifications: 10x, 20x, or 40x. Find a clear field for the tumor cell. For the blue laser source, use an excitation plate wavelength of 420-485 nm and an emission plate wavelength of 515 nm.
   2. Without moving the samples, use another laser source. Rotate the fluorescence microscope and excite the sample with the green laser. Find the cells emitting green fluorescence, which is indicative of tumor cells. Take images of the same field with the same magnification with this green laser source. The cells that emit both blue and green fluorescence are recognized as tumor cells. For the green laser source, use an excitation plate wavelength of 460-550 nm and an emission plate wavelength of 590 nm
   3. Without moving the samples, use another laser source. Rotate the fluorescence microscope and excite the sample with the red laser. Find the cells emitting red fluorescence. Take images of the same field with the same magnification with this red laser source. The cells that emit both blue and red fluorescence are recognized as white blood cells.
   4. Save the image for obtained by using each of the laser sources above. Take several images of the same field with different colored lights.
   5. Use ultraviolet light with an excitation plate wavelength of 330-400 nm and an emission plate wavelength of 425 nm. Use purple light with an excitation plate wavelength of 395-415 nm and an emission plate wavelength of 455 nm.
6. Calculate the capture efficiency as in step 1.2.1.

Representative Results

The whole setup includes a syringe pump, a syringe, and a microfluidic chip. The cell suspension in the syringe is connected to the syringe pump, and the cell suspension is introduced into the microfluidic chip to capture the cells. The capture efficiency for all the microfluidic chips utilized was around 90% or above. For the wave chip, we designed microstructures with varied gaps. The small gaps are used to capture the CTCs, and the big gaps are used to accelerate the flow rate. The cell suspension flows quickly in the big gap areas. The missed CTCs tend to be captured by the small gaps in the subsequent array¹⁶. For ellipse chips, we designed line-line gaps instead of point-point gaps to form a slim tunnel to enhance the capture. Therefore, high capture efficiency was achieved¹. We designed an ellipse structure to avoid edges and corners to maintain viability. The trapezoid filters have two circular spiral channels with embedded trapezoid and circular micropost barriers. For trapezoid filters, the capture efficiencies for MCF-7, MDA-MB-231, and HeLa were 94%, 95%, and 93%, respectively¹⁸.

Figure 1 shows that all the tumor cells were captured by the wave chip. Since all the tumor cells were concentrated around the wave micropost array, this indicates high capture efficiency for this microfluidic chip, as demonstrated by the number of tumor cells that were captured. Therefore, this setup makes it much easier to capture rare tumor cells; indeed, the chip is fabricated to capture a large number of tumor cells as well as rare number of tumor cells. For example, if the chip is solid or reproducible enough to capture 10,000 tumor cells, it is easy for the chip to capture 10-100 cells. Video 1 shows how rare number of tumor cells were obtained for the pre-experiment. A hollow needle made using a micropipette puller was used to aspirate 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10 tumor cells from a culture dish with diluted tumor cells stained with calcein AM in PBS. The cells were absorbed by the silica gel tube connected with the hollow needle. A tumor cell with green immunofluorescence was suctioned into the hollow needle. Blowing into the hollow tube led to the tumor cell inside the hollow needle being discharged into a microcentrifuge tube¹. This is the procedure to obtain rare tumor cells. Figure 2 shows CTCs from a gastric cancer patient captured using small-ellipse microfilters. Figure 3 shows the CTCs from a colorectal cancer patient captured using trapezoid microfilters and emitting both blue and green fluorescence. Figure 4 shows tumor cells grown on the chip after capture, which are ready to be treated with anti-cancer medicine. These findings illustrate that big-ellipse microfilters do not have any negative effects on cell viability.

Figure 5 shows the clinical immunofluorescence analysis of colorectal CTCs captured on the microfluidic chip. These are seen in brightfield and stained with Hoechst, CK-FITC, and CD45-PE staining. The CTCs were recognized as DAPI+/CK+/CD45−, and the WBCs were identified as DAPI+/CK−/CD45+. Figure 6 shows colorectal tumor cells cultured on the big-ellipse chip after capture. Figure 7 shows colorectal tumor cells captured on the big-ellipse microfilters. Clinically, CTCs in patient blood were captured by wave chips, trapezoid microfilters, and big-ellipse microfilters, indicating that these three chips are successful in capturing CTCs. Potentially, they could be applied in CTC products, such as CTC isolation products, with high efficiency.

Figure 1: Wave chip capture. All the tumor cells of MCF-7 were captured around the array of the wave chip without any cells missing. Since there were not any tumor cells in any other area besides the array, this indicates the high capture efficiency of the chip. A large number of tumor cells were captured. Therefore, rare tumor cells can also be easily captured. Tumor cells of MCF-7 were captured by wave chip and (A) stained with Hoechst and (B) stained with calcein AM. Scale bar: 100 µm. Please click here to view a larger version of this figure.

Figure 2: Clinical sample for CTCs captured by small-ellipse microfilters. Clinical CTCs of a gastric cancer patient captured by small-ellipse microfilters. The CTCs were identified in brightfield and with both blue and green fluorescence. The blue circle in (A) indicates a CTC of a gastric cancer patient captured inside the chip. From the images taken, it can be seen that there were no other cells in any other areas, indicating that the capture purity was high for this chip. Tumor cells of MCF-7 were captured by small-ellipse microfilters and seen in (A) brightfield, (B) stained with Hoechst, and (C) stained with CK-FITC. Scale bar: 20 µm. Please click here to view a larger version of this figure.

Figure 3: Clinical sample of CTCs captured by trapezoid microfilters. Clinical CTCs from a colorectal cancer patient were captured by trapezoid microfilters. In these images, it can be seen that six CTCs were captured, indicating that the capture efficiency was high in the small field. Additionally, no other cells appeared, indicating that the capture purity was extremely high for this chip. Tumor cells of MCF-7 were captured by trapezoid microfilters and seen in (A) brightfield, (B) stained with Hoechst, and (C) stained with CK-FITC. Scale bar: 20 µm Please click here to view a larger version of this figure.

Figure 4: Culture of CTCs captured by big-ellipse microfilters. Tumor cells of MCF-7 were captured by big-ellipse microfilters in front of a big-ellipse micropost array. No tumor cells passed through the array, indicating high capture efficiency for this chip. After capture, the tumor cells grew for 24-48 h. This indicates that both the capture efficiency and viability were very high for the big-ellipse microfilters. The cultured tumor cells of MCF-7 were captured by big-ellipse microfilters and seen at (A) 0 h after capture, (B) 24 h after capture, and (C) 48 h after capture. Scale bar: 50 µm. Please click here to view a larger version of this figure.

Figure 5: Example of a clinical sample used for CTC capture by trapezoid microfilters. Clinical CTCs from a colorectal cancer patient were captured by trapezoid microfilters. In these images, it can be seen that two CTCs were captured, indicating high capture efficiency. There was no other cell disturbance except residues of RBCs in the chip. Thus, for the clinical pre-processing of CTC capture, it is better not to use red blood cell lysis. CTCs of colorectal cancer were captured by trapezoid microfilters and seen in (A) brightfield, (B) stained with Hoechst, (C) stained with CK-FITC, and (D) seen in merged images. Scale bar: 20 µm. Please click here to view a larger version of this figure.

Figure 6: Example of cultured CTCs captured by a big-ellipse chip. Tumor cell cultures of MCF-7 cells in front of the big-ellipse micropost array and behind the big-ellipse microfilters. The tumor cells stained with calcein AM emitting green fluorescence grew as desired, indicating that the cell viability for this chip was very high. In total, there were 15 arrays, with varied gaps for each array organized by big-ellipse microposts. MCF-7 tumor cells were cultured on the big-ellipse chip after capture (A) in one array and (B) in another array. Scale bar: 50 µm. Please click here to view a larger version of this figure.

Figure 7: Example of a clinical sample used for CTC capture by big-ellipse microfilters. Clinical CTCs from a colorectal cancer patient were captured by big-ellipse microfilters. From the images, it can be seen that there was RBC contamination. This indicates that the whole patient blood sample must be diluted and that the chip needs to be flushed after capture. CTCs of clinical colorectal cancer patient blood samples were captured on the big-ellipse chip and seen in (A) brightfield, (B) stained with Hoechst, and (C) stained with CK-FITC. Scale bar: 100 µm. Please click here to view a larger version of this figure.

Discussion

The prognosis and early diagnosis of cancer have a significant effect on cancer treatment¹. CTC isolation with microfluidic chips offers a liquid biopsy with no invasion. However, CTCs are extremely rare and heterogeneous in the blood¹, which makes it challenging to isolate CTCs. CTCs have similar properties to the original tumor sources from which they originate. Thus, CTCs play a vital role in cancer metastasis¹.

The proposed protocol allows a complete analysis of CTC isolation with the microfluidic chip. The protocol includes all the key procedures related to CTC isolation. For example, this protocol includes RBCL analyses, which are recorded meticulously and organized into different packages, rare tumor cell acquisition, optimal flow rate determination, capture efficiency, clinical characterization, and enumeration. The careful management of the operations facilitates the clinical patient sample processing. The protocol allows for the pre-processing and processing of clinical patient samples and offers vital services for clinicians.

For CTC isolation, the pre-processing of rare tumor cells is extremely difficult to perform. This difficulty was solved by using a hollow needle pulled through a needle puller to suck 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10 tumor cells as desired. Then, a close-to-real clinical patient blood sample was prepared with the desired number of tumor cells mixed into normal blood, mimicking a patient blood sample. With these artificial patient blood samples, a close-to-real experiment was carried out. This method has solved a difficult problem usually met in CTC isolation for pre-experiments. This approach is not easy to perform and has never been reported elsewhere; however, it is a useful approach before performing a real clinical experiment.

CTC isolation with microfluidic chips is classified into three types: affinity-based, physical-based, and immunomagnetic-based. For affinity-based isolation, the microfluidic chip needs to be modified. The specific modification procedures have been included for modification with anti-EpCAM. The CTCs were identified through immunofluorescence staining with Hoechst, CK-FITC, and CD45-PE. Since many tumor cells express EpCAM, anti-EpCAM is an important antibody to capture CTCs in patient blood. However, anti-EpCAM is very expensive; thus, many aptamers have been developed to solve this problem. We mainly utilize physical-based microfluidic chips such as small-ellipse microfilters and trapezoid microfilters to capture CTCs since they are simple, easy to operate, and effective.

The advantage of this technique over alternative techniques is that the physical-based chip of big-ellipse microfilters firmly captures CTCs with high capture efficiency. The reason for this is that there are slim tunnels of line-line gaps. Wave chips capture CTCs by combining both physical property-based and affinity-based isolation. The CTCs are captured by the small gaps, and the big gaps can be used to accelerate the flow rate. The missed CTCs have to pass through the small gaps in the following arrays and are also captured by the affinity-based isolation.

The protocol has resolved several major problems in CTC isolation, especially for clinical experiments. However, it is impossible to include every detailed step in CTC isolation, such as regarding the role of nanoparticles and nanostructures and aptamer modification. These aspects also play an essential role in CTC isolation. However, they are not critical in solving problems such as improving capture efficiency. Instead, these aspects enrich the CTC isolation with new contents.

This work concentrates on the clinical isolation of CTCs with the designed microfluidic chip. Most microfluidic chips used are mainly physical-based, such as big-ellipse filters and small-ellipse filters. Wave chips are made by combining both affinity-based and physical-based chip properties¹. The microposts of big-ellipse filters were designed to be shorter to enhance the capture purity in this work. The small-ellipse filters can also be improved by designing them with varied micropost sizes for different arrays. Wave chips can be better designed by adding more arrays to enhance the capture.

Big-ellipse filters are organized with long elliptical microposts to form line-line tunnels that achieve high capture efficiency. However, the limitation of this capture is that the capture purity is not high enough or high capture purity is not easily obtained. For wave chip, the small gaps are used to capture the CTCs, and the missed CTCs are captured by the following arrays. The big gaps are used to accelerate the flow rate and eliminate disturbance from RBCs, thus improving the capture purity; however, the capture efficiency is solid at above 90% and is reproducible.

Clinical validation is significant in CTC isolation with microfluidic chips. This work presents the clinical isolation of CTCs from colorectal patients with big-ellipse filters, small-ellipse filters, and wave chips. The aim of the designed microfluidic chips is to clinically enumerate CTCs. For the existing methods of some other systems or platforms, the capture efficiency is low, or the capture efficiency is not high enough for the method to be effectively used in clinical applications.

Based on the clinical performances of these three microfluidic chips, which have high capture efficiency, they can potentially be applied in CTC products, especially after modification. The strength of our protocol is that it demonstrates the enumeration of rare number of tumor cells to mimic real clinical samples. In addition, the clinical experimental procedure is feasible and practical. This method can separate CTCs from whole patient blood. Therefore, the method is appropriate for clinical applications.

Materials

Name	Company	Catalog Number	Comments
Calcein AM	BIOTIUM	80011
calibrated microcapillary pipettes	Sigma- Aldrich	P0799
CD45-PE	BD Biosciences	560975
CK-FITC	BD Biosciences	347653	cytokeratin monoclonal antibody
DMEM	HyClone	SH30081.05
fetal bovine serum (FBS)	GIBCO,USA	26140
Hoechst 33342	Molecular Probes, Solarbio Corp., China	C0031
penicillin-streptomycin	Ying Reliable biotechnology, China
Red blood cells lysis (RBCL)	Solarbio, Beijing	R1010