Abstract
At first, to identify a set of genes related to progression and metastasis of cervical carcinoma after radiotherapy (RT) and to establish a predictive method, genome-wide data were investigated using a cDNA microarray. A total of 28 patients with stage III-IV cervical cancer who underwent definitive RT were included in this study. The expression profiles of 14 tumors with local failure and multiple distant metastasis and 14 tumors with no evidence of disease (NED) obtained by punch biopsy before treatment. The most accurate prediction was achieved at 63 genes (sensitivity: 78.8%, specificity: 38.1%). One of these genes was already known to be associated with metastasis via chromosomal instability (TTK). A “Predictive Score” system was developed that could predict the probability for development of metastases.
Next, to identify the most important gene related to prognosis after RT, we have validated some genes by mRNA expression from blinded 60 test patients using a real-time PCR. We compared mRNA expression of 29 with NED and 31 with cancer-caused death (CD) tumors after RT obtained by punch biopsy before treatment using a real-time PCR. Taqman probe was designed according to the sequence of BAX, TEGT (BAX-inhibitor), XRCC5, PLAU, HIF1A, CD44, and TTK cloned genes, with normalization to housekeeping genes. The endpoint of this study was the correlation between mRNA expression and the overall survival rate of these patients according to Cox’s proportional-hazard model. HIF1A and TTK mRNA expression values in CD group were higher than in NED group, however, not significantly (p = 0.079, p = 0.092, respectively). On univariate and multivariate analysis, TTK was a statistically significant independent poor prognostic factor for overall survival rate after RT (HR, 1.06; 95% CI, 1.0-1.12, p=0.044, and HR, 1.35; 95% CI, 1.02-1.78, p=0.035). In conclusion, the relatively high incidence of TTK may explain, at least in part, the poor prognosis of advanced cervical carcinomas after RT.
