Observed versus modelled lifetime overall survival of targeted therapies and immunotherapies for advanced non-small cell lung cancer patients – A systematic review

Martijn Simons, Bram Ramaekers, Andrea Peeters, Joanne Mankor, Marthe Paats, Joachim Aerts, Wim van Harten, Valesca Retèl, Manuela Joore*

*Corresponding author for this work

Research output: Contribution to journalReview articleAcademicpeer-review

5 Citations (Scopus)
10 Downloads (Pure)

Abstract

Outcomes used for the effectiveness (median) and cost-effectiveness (mean) on overall survival (OS) are different and can vary from one another. Therefore, we compared median and mean OS gains of targeted therapies and immunotherapies for stage IIIB/IV Non-small cell lung cancer and explored underlying aspect. Eligible trials were searched in PubMed, survival curves were digitized, and parametric survival models fitted to model the mean OS. Twenty-seven trials were found for targeted therapies (n = 17) and immunotherapies (n = 10). Differences between median and mean OS gains in months ranged from −2.8 to 6.8 and −4.9 to 0.3 for two different subgroups of targeted therapies, and −2.4 to 11.4 for immunotherapies. The mean OS gain was substantially larger for most immunotherapy trials, due to relatively long survival. Median and mean OS gains did not differ for targeted therapies. Our findings imply a potential discrepancy between the estimates of effectiveness and cost-effectiveness of cancer treatments.

Original languageEnglish
Article number103035
JournalCritical Reviews in Oncology/Hematology
Volume153
DOIs
Publication statusPublished - 1 Sept 2020

Keywords

  • Immunotherapies
  • Non-small cell lung cancer
  • Overall survival
  • Survival analysis
  • Targeted therapies
  • n/a OA procedure

Fingerprint

Dive into the research topics of 'Observed versus modelled lifetime overall survival of targeted therapies and immunotherapies for advanced non-small cell lung cancer patients – A systematic review'. Together they form a unique fingerprint.

Cite this