Low Enrollment of Adolescents and Young Adults onto Cancer Trials: Insights from the Community Clinical Oncology Program

Michael E. Roth, Ann M. O'Mara, Nita L. Seibel, David S. Dickens, Anne Marie Langevin, Bradley H Pollock, David R. Freyer

Research output: Contribution to journalArticle

27 Scopus citations


QUESTION ASKED: Given the importance of cancer clinical trials and the fact that most adolescents and young adults (AYAs) are treated in the community setting, did historical Community Clinical Oncology Program (CCOP) sites enroll a higher proportion of AYAs to Children's Oncology Group (COG) studies than non-CCOP sites? SUMMARY ANSWER: Over the 10-year period studied, CCOP sites enrolled a significantly lower proportion of AYAs onto COG studies than non-CCOP sites. Further, while proportional enrollment of AYAs declined over time at both CCOP and non-CCOP sites, the decline was greater at CCOP sites. METHODS: This was a retrospective cross-sectional analysis of clinical trial enrollment onto COGtrials utilizing the National Cancer Institute (NCI) Division of Cancer Prevention database. For the period 2004 through 2013, the proportional enrollment of AYAs (number of AYAenrollments divided by total enrollments) onto relevantCOGstudies was compared between CCOP and non-CCOP sites and potential changes over time were evaluated. All sites were COG member institutions. MAIN RESULT(S): During the study period, a total of 9,821 and 81,164 patients were enrolled onto COG trials at CCOP and non-CCOP sites, respectively. In aggregate, a significantly lower proportion of AYAs was enrolled onto applicable COG studies at CCOP than at non-CCOP sites (24.1% [1,606/6,672] v 28.2% [16,357/58,059], P,.001). During the intervals 2004 through 2008 and 2009 through 2013, the proportional enrollment of AYAs at CCOP sites and non-CCOP sites declined significantly (from 26.7% [844/3,156] to 21.7% [762/3,516], P,.001 and from 29.1% [8,189/28,115] to 27.3% [8,168/29,944], P,.001, respectively). Between those intervals, proportional enrollment of AYAs at CCOP sites declined for acute lymphoblastic leukemia studies (33.9% [108/319] v 23.6% [144/611], P =.001) but not for other cancers (Fig). BIAS, CONFOUNDING FACTOR(S), DRAWBACKS: A major strength of this study is the use of proportional rather than absolute AYA enrollment because this is less dependent on the changing availability of trials for AYA patients. Limitations of this study include being (1) restricted to enrollments onto COG and not adult cooperative group trials; and (2) unable to account for the observed differences between CCOP and non-CCOP sites and within site categories over time. REAL-LIFE IMPLICATIONS: This study suggests that despite many CCOPs being located in the community where AYAs tend to be treated, having a commitment to serving disparity populations, and having a clinical research infrastructure involving both medical and pediatric oncologists, these characteristics alone are not enough to result in higher AYA enrollment than in traditional treatment settings. However, the NCI Community Oncology Research Program (NCORP), which replaced the CCOP in 2014, is well-positioned to leverage these advantages through development and study of targeted interventions aimed at increasing AYA clinical trial accrual. Within NCORP sites, pediatric and medical oncologists should collaborate more actively in the recruitment and management of AYAs on clinical trials to improve care and outcomes.

Original languageEnglish (US)
Pages (from-to)e388-e395
JournalJournal of Oncology Practice
Issue number4
StatePublished - Apr 1 2016
Externally publishedYes


ASJC Scopus subject areas

  • Oncology
  • Oncology(nursing)
  • Health Policy

Cite this