Analysis of SWOG Trials Indicates No Survival Difference After 1 Year in Cancer Patients Treated In vs Out of Clinical Trials

Key Points

  • In good-prognosis patients, there was no significant difference in overall survival between patients treated in vs out of clinical trials.
  • In poor-prognosis patients, there was a significant difference in overall survival during the first year after diagnosis (favoring trial participation), but not thereafter.

In a study reported in the Journal of the National Cancer Institute, Unger et al evaluated whether cancer patients from SWOG clinical trials were similar to nontrial patients in baseline characteristics and survival. They found that, overall, trial participation in standard treatment arms did not influence overall survival after 1 year, suggesting both that findings in clinical trials can be generalized to the wider population and that eligibility requirements might be relaxed to encourage greater clinical trial participation.

Study Details

The study involved comparison of demographic factors, disease stage, and overall survival in 5,190 cancer patients in the standard treatment arms in 21 phase III SWOG trials vs 69,187 nontrial control cancer patients selected from the Surveillance, Epidemiology, and End Results (SEER) program who were treated between 1987 and 2007. The trials included those in brain, breast, gastric, pancreatic, bladder, prostate, kidney, cervical, and lung cancers as well as leukemia and melanoma.

Corresponding SEER patients for each SWOG trial had to have a diagnosis of the same cancer on a date during the SWOG study’s enrollment period, and each corresponding SEER patient group had to reflect the SWOG study’s age limits. Since nearly all SWOG studies excluded patients with prior malignancies, only SEER patients with first primary malignancies were included.

Multivariate analysis of survival was performed by Cox regression. Survival functions from aggregate data across all studies were compared separately by prognosis of ≥ 50% (good) vs < 50% (poor) average 2-year survival. Of the 21 SWOG trials, 11 had good-prognosis populations and 10 had poor-prognosis populations.  

Trial Patients Younger

Trial patients were younger than nontrial patients (P < .001), with corresponding SEER groups having a mean of 18.1% more patients aged ≥ 65 years. SEER groups also were more likely to contain more women, with a mean 4.7% difference compared with SWOG trials (P < .001). There were no significant differences between trial and nontrial populations in disease stage or proportions of black patients.

No Overall Survival Differences After 1 Year

On multivariate analysis, there was no significant difference in overall survival between SWOG patients in the 11 good-prognosis studies and corresponding SEER patients, but SWOG patients had significantly better overall survival (P < .001) in 9 of 10 poor-prognosis studies. Among good-prognosis patients, the mean of the multivariate hazard ratios (HRs) for overall survival for SWOG vs SEER patients was 0.96 (P = .12). The hazard ratio remained nonsignificant on landmark analysis including only patients who survived ≥ 1 year (mean HR = 1.05, P = .22). Whereas the mean of the hazard ratios was significant among poor-prognosis patients overall (HR = 0.74, P < .001), it was no longer significant on landmark analysis including only patients with ≥ 1 year survival (HR = 1.05, P = .27), indicating that the beneficial impact of trial participation occurred only during the first year after diagnosis in this setting.

In non–sex-specific studies, it was estimated that disease and stage explained 92.2% of the variation in survival outcomes, with age explaining 5.2%, trial participation 1.5%, race 0.6%, and sex 0.5%. During the first year, the variation attributable to disease and stage was 88.4%, with trial participation accounting for 4.9%, whereas the respective proportions were 92.7% and 1.2% after 1 year.

The investigators concluded: “Trial participation was associated with better survival in the first year after diagnosis, likely because of eligibility criteria that excluded higher comorbidity patients from trials. Similar survival patterns between trial and nontrial patients after the first year suggest that trial standard arm outcomes are generalizable over the long term and may improve confidence that trial treatment effects will translate to the real-world setting. Reducing eligibility criteria would improve access to clinical trials.”

Joseph M. Unger, PhD, MS, of the SWOG Statistical Center, Fred Hutchinson Cancer Research Center, is the corresponding author for the Journal of the National Cancer Institute article.

The study was supported Public Health Service Cooperative Agreement grants from the National Cancer Institute, National Institutes of Health, and Department of Health and Human Services.

The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.