Publication-only abstracts (abstract number preceded by an "e"), published in conjunction with the 2018 ASCO Annual Meeting but not presented at the Meeting, can be found online only.
Increase in tumor burden at disease progression as a predictor of survival in advanced NSCLC patients.
Metastatic Non-Small Cell Lung Cancer
Lung Cancer—Non-Small Cell Metastatic
2018 ASCO Annual Meeting
J Clin Oncol 36, 2018 (suppl; abstr e21114)
Author(s): Yuri Kogan, Moran Elishmereni, Eldad Taub, Zvia Agur; Optimata Ltd., Bene Ataroth, Israel
Background: Extending survival in advanced non-small cell lung cancer (NSCLC) is of great importance. Finding survival predictors and assessing the potential for improving patient outcomes through their modification can assist these efforts. Here we tested tumor burden changes at disease progression as survival predictors in NSCLC. Methods: Data were taken from 2 clinical trials in advanced NSCLC patients under 1st line CarboTaxol (projectdatasphere.org). Magnitude and rate of changes in tumor size (sum of longest diameters; SLD) close to progression, and other patient parameters, were evaluated as predictors of post-progression survival (PPS) using univariate/multivariate Cox proportional hazards models. The latter were built by stepwise regression, p = 0.2 being the entrance criterion. Results: In both trials (n = 381 patients with progressive disease (PD) on target lesions), the SLD rise from nadir to progression (dSLD) was significantly correlated with PPS (Table). Patients with lower-than-median dSLD showed longer survival, with a 4.9 and 9.6 month gain in PPS over large-dSLD patients. The impact of dSLD persisted in a multivariate model that included also baseline performance status, gender, appearance of new lesions, and time to progression as significant factors; the additive contribution of dSLD to the multivariate model (Table) was verified by model comparison. Conclusions: Tumor burden increase near first progression was found to be a unique survival predictor for advanced NSCLC. Continuous control of SLD features in NSCLC patients may thus assist in predicting outcomes and suggesting appropriate therapy. We have used these features to develop a new predictive tool that alerts to imminent progression and suggests a treatment switch to limit tumor growth and extend survival.
|Clinical Study -Identifier||Target lesion |
(% of all PD patients)
|Median PPS (months)||Median dSLD (cm)||Median PPS (months)||dSLD|
Hazard ratio [95% CI], p value
|> median dSLD||< median dSLD||Univariate model||Multivariate model|
|CA031 -NCT00540514||159 (44%)||7.8||2||4.8||9.7||1.46 [1.22, 1.74], |
|1.26 [1.04, 1.53], |
|IPASS -NCT00322452||222 (49%)||12.4||1.5||9.4||19||1.21 [1.12, 1.32],|
|1.18 [1.08, 1.29], |