Temporal Trend of Overall Survival in Patients with Advanced Non-small Cell Lung Cancer Given First-Line Cytotoxic Treatments Andrea Messori • Dario Maratea • Valeria Fadda • Sabrina Trippoli Received: 23 June 2012 / Accepted: 22 September 2012 / Published online: 19 October 2012 Ó Springer Science+Business Media New York 2012 To the Editor, In elderly patients with advanced non-small cell lung cancer, first-line treatments are frequently based on either doublet or single third-generation cytotoxic agents. The results of the meta-analysis by Qi et al. [1] indicated that doublet therapy was superior to a single third-generation agent, but the sta- tistical significance of this difference was borderline (end point: overall survival [OS]; hazard ratio = 0.84; 95 % con- fidence interval: 0.71–1; P = 0.053; data from 11 trials). We reanalyzed the data of median OS from these 11 trials for a total of 22 patient cohorts corresponding to 2,616 patients. Using standard meta-regression techniques [2–5], we investigated whether these values of median OS showed any temporal trend from 2000 to 2011. Figure 1 illustrates the results of our analysis. According to the regression line, the values of median OS varied from 5.2 months in 2000 to 11.4 months in 2011. More impor- tantly, the annual improvement (0.56 months; 95 % confi- dence interval: 0.27–0.85 months) was clinically relevant and showed a high level of statistical significance (P \ 0.001). We are aware that one limitation of our analysis is that we did not account for the randomized design of the trials. Anyhow, the statistical impact of this temporal trend (P \ 0.001) seems to be greater than that resulting from the comparison of doublet therapy vs. single-agent (P = 0.053 according to Qi et al. [1]). Conflict of interest None. References 1. Qi WX, Tang LN, He AN, Shen Z, Lin F, Yao Y (2012) Doublet versus single cytotoxic agent as first-line treatment for elderly patients with advanced non-small-cell lung cancer: a systematic review and meta-analysis. Lung Jun 19. [Epub ahead of print] Fig. 1 Meta-regression analysis of temporal trends of median overall survival in patients with advanced non-small cell lung cancer given first-line cytotoxic agents (22 cohorts published in 11 randomized trials; data obtained from Table 1 of ref. [1]). The regression equation was: MEDIAN = 5.18 ? 0.56 9 (YEAR-2000); P \ 0.001. Study- specific weights were calculated as inverse variance (where variance was the reciprocal of the number of patients); the meta-analysis was based on the random-effect model as implemented by the OMA software (Open Meta-Analyst, version 4.16.12, Tufts University, http://tuftscaes.org/open_meta/). Symbols: each study is represented by a circle the diameter of which is proportional to its statistical weight A. Messori Á D. Maratea Á V. Fadda Á S. Trippoli HTA Unit, ESTAV Toscana Centro, Regional Health Service, 50100 Florence, Italy A. Messori (&) HTA Unit, Area Vasta Centro Toscana, Regional Health System, Via Guimaraes 9-11, 59100 Prato, Italy e-mail: andrea.messori.it@gmail.com; andrea.messori@estav-centro.toscana.it 123 Lung (2012) 190:591–592 DOI 10.1007/s00408-012-9426-4