M. Burgio

BACKGROUND: Nivolumab, a fully human IgG4 programmed death 1 (PD-1) immune-checkpoint-inhibitor antibody, disrupts PD-1-mediated signaling and may restore antitumor immunity. METHODS: In this randomized, open-label, international phase 3 study, we assigned patients with nonsquamous non-small-cell lung cancer (NSCLC) that had progressed during or after platinum-based doublet chemotherapy to receive nivolumab at a dose of 3 mg per kilogram of body weight every 2 weeks or docetaxel at a dose of 75 mg per square meter of body-surface area every 3 weeks. The primary end point was overall survival. RESULTS: Overall survival was longer with nivolumab than with docetaxel. The median overall survival was 12.2 months (95% confidence interval [CI], 9.7 to 15.0) among 292 patients in the nivolumab group and 9.4 months (95% CI, 8.1 to 10.7) among 290 patients in the docetaxel group (hazard ratio for death, 0.73; 96% CI, 0.59 to 0.89; P=0.002). At 1 year, the overall survival rate was 51% (95% CI, 45 to 56) with nivolumab versus 39% (95% CI, 33 to 45) with docetaxel. With additional follow-up, the overall survival rate at 18 months was 39% (95% CI, 34 to 45) with nivolumab versus 23% (95% CI, 19 to 28) with docetaxel. The response rate was 19% with nivolumab versus 12% with docetaxel (P=0.02). Although progression-free survival did not favor nivolumab over docetaxel (median, 2.3 months and 4.2 months, respectively), the rate of progression-free survival at 1 year was higher with nivolumab than with docetaxel (19% and 8%, respectively). Nivolumab was associated with even greater efficacy than docetaxel across all end points in subgroups defined according to prespecified levels of tumor-membrane expression (≥1%, ≥5%, and ≥10%) of the PD-1 ligand. Treatment-related adverse events of grade 3 or 4 were reported in 10% of the patients in the nivolumab group, as compared with 54% of those in the docetaxel group. CONCLUSIONS: Among patients with advanced nonsquamous NSCLC that had progressed during or after platinum-based chemotherapy, overall survival was longer with nivolumab than with docetaxel. (Funded by Bristol-Myers Squibb; CheckMate 057 ClinicalTrials.gov number, NCT01673867.).

In this paper, Neural Network (NN) models for the real-time simulation of gas turbines are studied and developed. The analyses carried out are aimed at the selection of the most appropriate NN structure for gas turbine simulation, in terms of both computational time of the NN training phase and accuracy and robustness with respect to measurement uncertainty. In particular, feed-forward NNs, with a single hidden layer and different numbers of neurons, trained by using a back-propagation learning algorithm are considered and tested. Finally, a general procedure for the validation of computational codes is adapted and applied to the validation of the developed NN models.

All measurements, although taken as accurately as possible, are subjected to uncertainty. So the analysis of errors and uncertainty is crucial in all applications since such errors need to be estimated and, when possible, reduced. In particular, when gas turbine mathematical models based on the processing of field measurements (such as the Gas Path Analysis models) are used, the evaluation of measurement reliability is a key point. In fact, it has been demonstrated that these kinds of techniques are sensitive to measurement errors: thus, tools for field data processing to evaluate the presence of the so-called outliers are advisable. In this paper, some statistical methodologies for the assessment of the reliability of the measurements taken on a gas turbine are presented. The methodologies, taken from literature and used for historical measurements, are discussed. Moreover, a new methodology, based on a modified t-Student distribution, is proposed.