026; adjusted HR=1 017 (1 002�C1 032)) (Table 4, Figure 3b) Thus

026; adjusted HR=1.017 (1.002�C1.032)) (Table 4, Figure 3b). Thus, the predictive power of the three-gene predictor is consistent across two validation sets, that is, one from our study patients and the other from published data. Table 4 Cox regression analyses of the three-gene DAPT secretase Gamma-secretase predictive index percentile, as a continuous variable, for published DNA microarray data from 40 metastatic gastric cancer patients treated with either FU-based chemotherapy or cisplatin/irinotecan combination … Interestingly, the three-gene predictor was found to be an independent predictor for poor survival, when the same Cox regression analysis was performed only on a subset of these patients (n=16) treated with cisplatin in combination with irinotecan, a topoisomerase I inhibitor (adjusted P=0.011; adjusted HR=1.

038 (1.008�C1.068)). Patients treated with irinotecan were not included in the original training set patients. Hence, the predictive power of three-gene predictor may not be specifically associated with only CF therapy, although further large-scale studies need to be performed to address the predictive value of the three-gene predictor for other therapeutic regimens. Three-gene predictive index and radiographic response Although the radiographic tumor response was not the main end point of this study, we also evaluated the association between the three-gene predictive index and radiographic response of patients with measurable disease. When published data4 were also included, 104 patients had either partial response or stable disease (clinical benefit) as the best response, whereas 46 patients had progressive disease.

The three-gene predictive index was significantly associated with radiographic response at a univariate P-value of 0.039, which is higher than the Cox regression P-value for the overall survival of all study patients (Table 5). This statistical association was at borderline significance in a multivariate regression analysis. Table 5 Logistic regression analysis on the three-gene predictive index for radiographic response of 150 patients with measurable disease, including patients represented by the published data set Three-gene predictor is not prognostic but predictive Although we showed that the three-gene predictor predicted time to progression and overall survival for CF-treated patients, we wished to further address whether it represents a prognostic signature, using the published data set from 88 gastric cancer patients who were treated by surgery alone and not with chemotherapy.

11 The three-gene predictive index percentile Batimastat was not a prognostic factor in this data set as a continuous variable (P=0.506). There was no difference in survival in the surgically treated patients between the high- and low-risk groups predicted by the three-gene predictor (P=0.972).

Leave a Reply

Your email address will not be published. Required fields are marked *

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>