2.1.

System of Optical Coherence Tomography

The OCT system used in this study mainly includes resource, fiber conduction, reference arm, and sample arm modules. Briefly, a broadband light source emitted from a superluminescent diode is the central wavelength at $830\mathrm{nm}$ with a bandwidth of $50\mathrm{nm}$ . The light is delivered via a single-mode fiber with a mode field diameter of $5.3\mu \mathrm{m}$ . This OCT system provides an axial resolution of $10\text{to}15\mu \mathrm{m}$ . The transverse resolution of the system is about $20\mu \mathrm{m}$ , determined by the focal spot size produced by the probe beam. The signal-to-noise ratio (SNR) of the system is measured at $100\mathrm{db}$ . A visible light source $(\lambda =645\mathrm{nm})$ is used to guide the probe beam. The OCT system operation is controlled automatically by computer. The detector current is demodulated using a lock-in amplifier and a low-pass filter in software prior to storage. Each in-depth scanning (A-scan) consists of 10,000 data points. Lateral scanning (B-scan) image is obtained by moving the mirror relative to the tissue sample, which takes about $1.0\mathrm{s}$ . The data acquisition software is written in LabView 7.2-D. OCT images are obtained in each experiment and stored in a PC for further processing.

2.2.

Material and Animal Tumor Model

20 BALB/c nude mice weighing between $15\mathrm{g}$ and $18\mathrm{g}$ , purchased from the Laboratory Animal Center at Sun Yat-Sen University, were used in this study. Animals were maintained under specific pathogen-free conditions, in which chow and tap water were available ad libitum. The animals were divided into two groups: the normal control group (ten samples) and the tumor model group (ten samples). The tumor model group received an orthotopic (gastric mucosa) graft operation.

The human gastric cancer cell line MCG-803 cells (CTCC) were cultured in DMEM culture medium supplemented with 10% fetal calf serum, $50\text{-}\mathrm{U}∕\mathrm{ml}$ penicillin, and $50\text{-}\mu \mathrm{g}∕\mathrm{ml}$ streptomycin. Suspensions of MCG-803 cells ( $5\text{to}10\times {10}^6$ viable cells/mouse) were implanted into the right dorsal portion of BALB/c nude mice. After $20\text{days}$ , the inoculated tumors reached diameters of about $3\text{to}6\mathrm{mm}$ , the mice were killed, and tumor tissues were surgically extracted for establishing the animal model of gastric carcinoma in situ. The resected tumor tissues were sheared to small tumor tissues about one cubic millimeter. The ten mice in the tumor model group were opened at their abdominal cavity by operation, the gastric walls were exposed, and then the small tumor tissues were implanted into the gastric wall. After $20\text{days}$ , when the diameter of the implanted tumors became about $2\text{to}4\mathrm{mm}$ , the animals were prepared for OCT imaging.

Then, the gastric walls in the normal control group and the tumor model group were cut into approximately $1\times 1\text{-}\mathrm{cm}$ samples and placed in $4\text{-}\mathrm{ml}$ 0.9% saline solution. Continuous OCT imaging was performed for approximately $1\mathrm{h}$ at $22°\mathrm{C}$ . The region was imaged for approximately $7\text{to}10\min$ before the glucose solution was applied to establish a baseline for the OCT signal slope graph. Then $4\text{-}\mathrm{ml}$ 40% glucose solution was added to the saline solution in a 1:1 ratio to maintain the end level of 20% glucose.

2.3.

Methods of Measurement and Calculation

The permeability coefficients of glucose in tissues ex vivo were measured by monitoring slope changes of OCT signals in the tissue. The permeability coefficient of glucose in the tissues was computed with the OCT signal slope (OCTSS).¹⁷ The glucose that diffused through the tissues changed the local optical property, which was detected by the OCT system. Glucose penetrated through the tissue to replace water in tissues. Due to the increase of in-depth concentration, the light scattering coefficient is altered. The changes of the optical properties were shown in biological tissues through time induced by the glucose diffusion.

The normal stomach mucosa tissues and stomach tissues are performed by OCT at a central length of $830\mathrm{nm}$ . 2-D OCT images were acquired by continuous imaging of the tissue. 20 consecutive scans were averaged to obtain OCT intensive profiles as a function of depth. A region with minimal alterations in the OCT signal was selected for time-dependent analysis of its optical properties as glucose diffused through these tissues. The permeability coefficient of glucose in the normal stomach and tumor tissues was calculated by using the following equation:

2.4.

Statistical Analysis

Data are presented as a $\text{mean}\pm \mathrm{SD}$ for a number of sample animals. Data were analyzed by Student’s t-tests for unpaired data. $\mathrm{P}<0.05$ was considered statistically significant. All statistical analyses were performed with the statistics software SPSS 10.0 for Windows.

3.1.

Optical Coherence Tomography Images of Normal Stomach Tissues and Stomach Tumor Tissues

Figures 1a, 1b, 1c show the OCT images of the normal stomach tissues ex vivo at 0, 20, and $40\min$ during the 20% glucose diffusion experiment. It is shown that the submucosal layers of normal stomach tissues are visible with layer structures at 0 and $20\min$ . The imaging depth is improved after the application of 20% glucose. Figures 1d, 1e, 1f display the OCT image of the stomach tumor tissues ex vivo. It is shown that there are no layer structures in the gastric cancer tissues in the tumor model group. It can be seen that there is an improvement in the imaging depth after the application of 20% glucose.

Fig. 1

OCT images of (a), (b), and (c) the normal stomach, and (d), (e), and (f) tumor tissues at 0, 20, and $40\min$ during the 20% glucose diffusion experiment.

3.2.

Optical Coherence Tomography Signal Slope Graphs of Normal Stomach Tissues and Stomach Tumor Tissues

Figure 2a shows a typical OCTSS graph in the normal stomach tissues during a glucose diffusion experiment. The sample was monitored for approximately $10\min$ before glucose solution was applied. The monitored region was about $135\mu \mathrm{m}$ thick and $195\mu \mathrm{m}$ away from the surface. From Fig. 2a, there was a decrease of the OCT signal slope caused by a reduction of scattering in the tissues due to the local increase of glucose concentration. 20% glucose solution reached the monitored region approximately $20\min$ after administration. This decrease of the slope could be due to the concentration gradient differences between both sides of the tissue; the fluid (mainly water) moved from areas of high concentration to those of lower concentration. And the monitored region was completely diffused through by the 20% glucose solution. At this time, a reverse process, the increase of OCT signals, in the OCTSS was seen. Water diffused back into the tissues due to the difference, which means water was diffusing back into the tissue due to the difference of the concentration gradient of the glucose solution on both sides. After $20\min$ , equilibrium is reached. There is no evident change in the graph. The permeability coefficient of 20% glucose from different experiments was found to be $(0.94\pm 0.04)\times {10}^{-5}\mathrm{cm}∕\mathrm{s}$ . Figure 2b displays an OCTSS graph during a 20% glucose diffusion experiment in the tumor tissues. The same procedure was used in the tumor tissue experiments. The tumor tissue region monitored was $310\mu \mathrm{m}$ thick and $420\mu \mathrm{m}$ away from the surface. The permeability coefficient of 20% glucose in the tumor tissues was found to be $(5.32\pm 0.17)\times {10}^{-5}\mathrm{cm}∕\mathrm{s}$ .

Fig. 2

OCTSS graphs as a function of time from (a) normal stomach tissues and (b) stomach tumor tissues during the glucose 20% diffusion experiment. The tick marks indicate the interval of glucose diffusion.

Comparing the permeability coefficient in the tumor model group with that in the normal control group, it is found that there is a significant difference in the permeability coefficient of glucose between tumor tissues and normal tissues $(\mathrm{p}<0.01)$ (see Fig. 3 ).

Fig. 3

Comparison of permeability coefficient of glucose diffusion.