This article is more than 5 years old. Information may no longer be current.
Shift in NSCLC trial design, interpretation linked to inefficiency
Click Here to Manage Email Alerts
Changes in primary endpoints and an increase in trials that report positive outcomes have coincided with fewer statistically significant endpoint improvements and a decreased net survival benefit among phase 3 randomized controlled trials for non–small cell lung cancer, according to study results.
Researchers evaluated data from 203 phase 3 randomized control trials of systemic therapy for NSCLC conducted between 1980 and 2010.
Overall, the size and number of advanced NSCLC trials has increased over time, from 32 trials with a median sample size of 152 patients in the 1980s to 118 trials with a median of 413 patients between 2001 and 2010.
OS served as the primary endpoint in most trials in the 1980s (97%) and 1990s (96%); however, significantly fewer trials used OS as the primary endpoint from 2001 to 2010 (81%; P˂.002). Thirteen percent of trials used PFS as the primary endpoint during that time.
The percentage of trials achieving significant improvements in the primary endpoint has remained consistent (29% for 1980 to 1990 vs. 31% for 2001 to 2010). However, significantly more trials in the last decade reported positive outcomes without demonstrating significantly improved endpoints (30% in 1980 to 1990 vs. 53% in 2001 to 2010; P˂.001).
Median survival gain among trials reporting statistically significant improved survival decreased from 3.9 months in the 1980s to 2.5 months from 2001 to 2010 (P=.11). This difference became statistically significant when researchers evaluated all trials that reported positive outcomes (3.9 months in the 1980s vs. 0.9 months from 2001 to 2010; P˂.001).
Overall, average median survival increased from 6.7 months in the 1980s to 9.5 months between 2001 and 2010.
“The intention of multiphase clinical trials is to focus resources on promising agents, with each phase eliminating either toxic or minimally active agents,” the researchers wrote. “Our research suggests that we are progressively less efficient in accomplishing this task in advanced NSCLC trials. If we can alter these trends in trial design and interpretation, there is great potential to minimize trial and drug development costs on marginally effective, or ineffective, agents.”
Disclosure: The researchers report no relevant financial disclosures.
Perspective
Back to TopRathi Pillai, MD
The treatment landscape for NSCLC has dramatically changed with the availability of agents effective beyond first-line therapy and newer targeted agents for subpopulations of patients with EGFR and ALK mutations. As our patients are living longer due to increased therapy options, clinical trial design has been impacted. PFS has become a more attractive endpoint for phase 3 studies due to the availability of subsequent lines of treatment and the frequent use of crossover design to allow patients access to newer promising drugs. Although the FDA has accepted this, PFS compared to OS has the potential to be more subjective and may not be clinically relevant.The study by Sacher and colleagues examines the changes in clinical trial design and interpretation of results in phase 3 studies of patients with advanced NSCLC over the last 30 years. The authors found that more phase 3 studies in NSCLC in the last decade use PFS as the primary endpoint. Although trials today are requiring larger patient sample sizes to achieve their primary endpoint, there has been a trend toward a smaller absolute survival gain compared with the 1980s. More concerning is the rise in studies that report a positive result, although there has been no increase in the number of studies achieving the primary endpoint over this same time period.
This study confirms that the bar for interpreting clinical benefit in NSCLC is indeed dropping. We must be cognizant of designing phase 3 trials to achieve clinically relevant improvements in outcomes for patients. We should examine novel trial designs to more quickly pick up a signal of drug efficacy and also focus on biomarker development in earlier phases of drug development. Otherwise, we will waste resources on the development of new drugs that do not offer a meaningful benefit to what is already available.
Click Here to Manage Email Alerts