EphB and ephrinB mRNA levels were determined

EphB4 and ephrinB2 mRNA levels were determined from three parts taken from each tumor, and each sample was analyzed in triplicate. The differences in the histoscores and mRNA levels of EphB4 and ephrinB2 were analyzed by Student's t test. The correlation coefficients were evaluated both by linear regression analysis and bivariate Pearson correlation. Survival rates were calculated using the Kaplan–Meier method and analyzed by the log-rank test. Differences were considered significant when p was less than 0.05.
Results Immunohistochemical staining for EphB4 and ephrinB2 of a representative case of squamous cell carcinoma of uterine cervix is shown in Fig. 1. All of the uterine cervical cancer specimens revealed strong staining for EphB4 and ephrinB2 in the cancer calcium sensing receptor and very faint staining in vascular endothelial cells. Both EphB4 and ephrinB2 histoscores in cancer cells and mRNA levels in uterine cervical cancers significantly increased according to clinical stage (I in 25 patients with positive pelvic lymph node metastasis (LN+) compared to 37 patients with negative lymph node metastasis (LN−). Similarly, ephrinB2 histoscores and mRNA levels were also significantly higher (p<0.001, p<0.01) in 25 patients with LN+ compared with 37 patients with LN− as shown in Fig. 3. Therefore, there was no significant difference in histoscores or mRNA levels of EphB4 and ephrinB2 according to histopathological types as shown in Fig. 4. We had selected the patients who had been diagnosed with stage Ib and II uterine cervical cancers to analyze the correlation between tumor size and histoscore/mRNA level of EphB4 and ephrinB2. The tumor sizes significantly correlated with EphB4 histoscores (histoscore=1.47×tumor size+86.86; r=0.53; p<0.001), ephrinB2 histoscores (histoscore=1.83×tumor size+100.36; r=0.61; p<0.001), EphB4 mRNA levels (mRNA=2285.68×tumor size+34337.73; r=0.639; p<0.01) and ephrinB2 mRNA levels (mRNA=48032.367×tumor size+610331.58; r=0.75; p<0.01) as shown in Fig. 5. We analyzed the prognosis of the 62 patients who underwent surgical resection. The two groups, determined independently by the EphB4 or ephrinB2 histoscores and mRNA levels, consisted of exactly the same patients. EphB4 histoscore of 175 and 1.6×105 DNA copy/μg total RNA in EphB4 mRNA level and ephrinB2 histoscore of 223 and 2.8×106 DNA copy/μg total RNA in ephrinB2 mRNA level were the median values, and were adopted to divide the 62 patients into two groups of 31 patients each. The 36-month survival rate of the 31 patients with high EphB4 (cases with EphB4 histoscore over 175; the same as those with EphB4 mRNA levels over 1.6×105 DNA copy/μg total RNA) was 31%. The survival rate of the 31 patients with low EphB4 (cases with EphB4 histoscore below 175; EphB4 mRNA levels below 1.6×105 DNA copy/μg total RNA) was 72%. The survival rate of the 31 patients with high ephrinB2 (cases with ephrinB2 histoscore over 223; the same as those with ephrinB2 mRNA levels over 2.8×106 DNA copy/μg total RNA) was 19%, while the rate of the other 31 patients with low ephrinB2 (cases with ephrinB2 histoscore below 223; ephrinB2 mRNA levels below over 2.8×106 DNA copy/μg total RNA) was 73%. There was a significant difference (p<0.05 in EphB4 and p<0.01 in ephrinB2) between the 36-month survival rates of the 31 patients with high or low histoscores and mRNA levels of EphB4 and ephrinB2, as shown in Fig. 6.
Discussion In the present study, coexpression of EphB4 and ephrinB2 revealed a significant positive correlation of clinical stage, tumor size, and lymph node metastasis with poor patient prognosis. Overexpression of EphB4 and ephrinB2 in tumor cells indicates that EphB/ephrinB signaling drives interactions among the tumor cells, which may result in destabilization and affect cell–matrix attachment leading to promotion of invasion and metastasis [20].