Перегляд вихідного коду сторінки The Basis 3-Methyladenine Fees Will Remain High

Baseline covariates considered included age (as a continuous variable); race (Caucasian, Black, Asian, or other); ECOG performance score (0/1 or 2); PFI (as a continuous variable); cancer antigen 125 (CA-125) ([http://www.selleckchem.com/products/Sunitinib-Malate-(Sutent).html http://www.selleckchem.com/products/Sunitinib-Malate-(Sutent).html] are all statistically significant predictors of survival [14]. In particular, CA-125 remains the only well-defined, validated tumor marker to have a significant impact on the clinical management of ovarian cancer [15]?and?[16]; it has been shown that a decrease in CA-125 can be predictive of radiological response [http://www.selleckchem.com/products/3-methyladenine.html buy 3-Methyladenine] [17]. Therefore, to ensure no biases were introduced by these differences, we adjusted the survival distributions by PFI, CA-125, and ECOG performance score; 12 patients were excluded following adjustment because of missing CA-125 data. A similar process was undertaken by the ERG during initial assessment, but only CA-125 was adjusted for because imbalances in continuous PFI were not reported during the interim analysis [14]. Goodness-of-fit criteria based on Akaike information criterion (AIC) and Bayesian information criterion (BIC) of each survival distribution were used to select the preferred distribution for PFS and OS by treatment arm. The lower the AIC and BIC, the better fit the distribution is to the observed data. In addition, the survival distributions and Kaplan-Meier plots were analyzed to ensure that the survival distributions closely predicted the observed PFS and OS events. Table 2 summarizes the AIC and BIC scores for each survival distribution adjusting for PFI, CA-125, and ECOG performance score: ? The Weibull distribution was chosen for PFS for both trabectedin plus PLD [https://en.wikipedia.org/wiki/Ceftiofur Ceftiofur] and PLD alone Finally, we examined the functional form of the continuous variables considered for the chosen distributions, namely, PFI. By using linear, logarithm, and exponential forms, we found that the logarithm was a significantly better fit compared with the linear and exponential; AIC was reduced by approximately 3 or more for the four selected distributions (Table 3). Therefore, mean PFS and OS for trabectedin plus PLD and PLD alone were based on Weibull and log-logistic distributions, which adjusted for log (PFI), ECOG performance score, and CA-125. Based on the Weibull distributions, the mean PFS for trabectedin plus PLD and PLD alone was 11.26 and 8.25 months, respectively.