In our previous report, proliferative activity of the colon epithelium was compared between SSA/Ps and HPs using Ki-67 immunostaining . Our findings showed that the proportion of Ki-67 positive cells was higher in SSA/Ps as compared to HPs. A difference in the distribution pattern of the proliferative compartment revealed that Ki-67 immunostaining is useful to differentiate between SSA/Ps and HPs. However, the number of cells was visually counted, resulting in lack of objectivity and reproducibility of measurement. In this study, although we strived to circumvent the previous issues by counting the number of cells in microscopic images (x10) processed by Photoshop Elements 9 using a two-dimensional image analysis software , several problems still occurred. The targeted area of the colon mucosa was traced and defined, followed by a count of the number of cells based on color recognition of the cells. In order to maximally differentiate the target cells from the adjacent ones, the margins of the nuclei were identified by enhanced contrast for RGB separation. Although G (green) image was the easiest to assess, maximal and accurate differentiation of the targeted cells from the adjacent cells was still not possible using just the color recognition method. The measurement was, therefore, done with the “Separate Circular Figure” function using the 2D Image Analysis Software, Win-ROOF, by Mitani Corporation, Japan. The maximum and minimum diameters of the cells, the number of candidacy center points, edge threshold, and contrast were set to mark the cell border and in turn, visualize the cells as donut-shaped. Thus, the number of cells could be more accurately counted. However, due to the complicated adjustments of these settings, manual counting function of the image analysis software was utilized, and the number of cells was counted with a touch pen by introducing a liquid crystal touch panel. Moreover, the distance from the base of the crypts to the Ki-67 positive cells on the surface was also defined by a line drawn with a touch pen and automatically measured. These measurements could be automatically transferred to Microsoft Excel and were also useful for data processing. The above-mentioned measurements were performed as a pilot study, which was followed by commencement of the main study.
The protein antigen Ki-67 has been widely used as an indicator of proliferation due to its restricted expression between the S and the M phases of the cell cycle in differentiating cells. In this study, it was confirmed that SSA/Ps had a higher proliferative activity as compared to HPs as shown by the difference in incidence of Ki-67 positive cells. Furthermore, based on the assumption that the proliferative compartment might be an indicator of distribution range, the distance from the base of the crypt to the Ki-67 positive cells on the surface was measured, and it was found that SSA/Ps exhibited a significantly more asymmetric distribution compared with HPs . Moreover, it was revealed that SSA/Ps had a higher proliferative activity with more asymmetric distribution, as demonstrated by proportion of Ki-67 positive cells and distance in a single crypt, respectively, as compared to HPs even in lesions with diameter <10 mm. These results demonstrated that, according to the diagnostic criteria provided in Table 1, in SSA/Ps, even lesions with a diameter < 10 mm had a higher proliferative activity as compared to similar sized lesions in HPs. Thus, we verified here that the common features of SSA/Ps including crypt dilation, irregularly branching crypts, and horizontally arranged basal crypts (inverted T- and/ or L-shaped crypts) are indeed appropriate histological findings to show “none dysplastic, but have an extended proliferative zone [14, 15]” as indicated by Riddell. In this study, the intermediate group showed intermediate characteristics between HP and SSA/P, and this may be because this type of lesion is a transitional stage from HP to SSA/P. However, further investigation is required in order to support this finding.