Definitions, measurement, and reporting of progression-free survival in randomized clinical trials and observational studies of patients with advanced non-small-cell lung cancer treated with immunotherapy: a scoping review

Abstract

Background

Evidence from observational studies is increasingly used in oncology to complement evidence from clinical trials. Commonly used endpoints to evaluate oncology medicines are overall survival (OS) and progression-free survival (PFS). However, comparing PFS across observational studies and with clinical trials can be challenging due to differences in its definition and measurement. This scoping review investigated how PFS was defined, measured, and reported in randomized clinical trials (RCTs) and observational studies of patients with advanced non-small-cell lung cancer (NSCLC) treated with immunotherapy.

Materials and methods

This scoping review included RCTs and observational studies that measured PFS in advanced NSCLC patients treated with immunotherapy. ASReview, an open-source artificial intelligence-assisted tool, was used to screen and prioritize relevant studies from records identified from PubMed and Embase between 2012 and 2023. Information on study characteristics, PFS definitions, and measurements was extracted.

Results

Forty RCTs and 144 observational studies were included. Most RCTs were conducted across multiple continents (70%), while most observational studies were conducted in Asia (62%). In contrast to RCTs, many observational studies lacked reporting on the end date of PFS measurement (69%), the type of radiological imaging (59%), and the imaging reviewer (78%). For observational studies that did report on PFS definitions and measurements, these often differed from those in RCTs, particularly regarding event definitions, the start and stop dates for PFS measurement, and tumor assessment schedules.

Conclusions

In contrast to RCTs, observational studies often lack reporting on PFS definitions and measurements, and if reported, they differ across observational studies and between them and RCTs. Since observational studies are important for complementing evidence, aligning PFS definition and measurement criteria with those used in RCTs, along with detailed reporting, is needed. However, some variability in PFS measurement characteristics is unavoidable, and therefore, PFS estimates from observational studies should be interpreted critically and carefully.