Start Your Search
O27 - Clinical Trials and Practice (ID 142)
- Event: WCLC 2013
- Type: Oral Abstract Session
- Track: Other Topics
- Presentations: 1
- Moderators:J.S. Lee, J. Bishop
- Coordinates: 10/29/2013, 16:15 - 17:45, Bayside Auditorium A, Level 1
O27.03 - Meta-analysis of progression-free survival and objective response rate as predictors of overall survival in locally advanced or metastatic non-small-cell lung cancer (ID 3170)
16:15 - 17:45 | Author(s): S. Norris
Improving overall survival (OS) in locally advanced or metastatic non-small-cell lung cancer (NSCLC) remains a goal for clinicians. OS is often the primary endpoint in Phase III clinical trials, however with the increased use of multiple lines of treatment and crossover/rescue therapy in NSCLC trials, overall survival differences are frequently confounded. Other clinically meaningful endpoints such as progression-free survival (PFS) are becoming increasingly important. A systematic review and meta-analysis was undertaken to investigate whether the treatment effect on PFS, or objective response rate (ORR), was predictive of the treatment effect on OS in locally advanced or metastatic NSCLC.
Embase, MEDLINE and the Cochrane Library databases were searched to identify relevant published randomised controlled trials (RCTs). Included RCTs compared pharmacological treatments in two or more groups in patients with locally advanced or metastatic (Stage IIIb or IV) NSCLC. The reported outcomes had to include median PFS and median OS. Data were analyzed with locally weighted least squares regression (LOWESS) to validate the linear relationship. Unweighted linear regression and weighted linear regression were used to estimate the relationship between OS treatment difference and PFS treatment difference. A similar analysis was also done to compare ORR difference and OS difference. The analysis also considered whether potential baseline prognostic factors could also change OS difference. Univariate weighted linear regression was used to assess the relationship of each prognostic factor with OS difference, followed by multivariate weighted linear regression.
A total of 124 RCTs (N=40,568 patients) were included in the analysis. Of the 124 trials, 98 (79%) were published since 2005. The age of patients ranged from middle aged cohorts to the elderly. There were 87 studies (70%) with first-line therapy and therapy type was predominantly chemotherapy (71 studies, 57%). The LOWESS of OS difference versus PFS difference was a relatively straight line, therefore a linear relationship appeared to be justified. Weighted linear regression showed a highly significant relationship between OS treatment difference and PFS treatment difference [slope 1.317 months (95%CI 1.000, 1.634) p<0.001 R = 36.6%]. Diagnostic plots revealed several statistical outliers. A weighted linear regression of OS difference versus PFS difference with six outlier studies removed also showed a highly significant linear relationship [slope 0.994 months (95%CI 0.749, 1.240) p<0.001 R = 36.7%]. The relationship between ORR difference and OS difference was also significant [slope 12.503 (95%CI 7.042, 17.963) p<0.001 R = 16.0%]. Univariate weighted linear regressions showed that the only baseline prognostic factor with a significant linear relationship with OS difference was the proportion of patients with non-squamous histology. Multivariate weighted linear regression showed none of the baseline prognostic factors was significant when PFS difference or ORR difference was included in the model.
The treatment effect on PFS was predictive of the treatment effect on OS. For a 1 month increase in PFS treatment difference, OS treatment difference increases by approximately 1.3 months. A 10% increase in ORR treatment difference was also associated with an increase in OS of 1.25 months. PFS is an important outcome for healthcare decision-making.
Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.