In our population-based study of AYA patients with cancer, we found higher rates of cancer trial participation (3% to 34%) than prior studies of this age group (2% to 15%)14,15
but comparatively lower participation than in pediatric or older adult populations15
or those treated in children's hospitals.10
We also demonstrated in the AYA population that uninsured, older patients and those treated by nonpediatric oncologists were less likely to enroll onto clinical trials. Furthermore, we found a shorter time from diagnosis to treatment in patients with low-stage disease or those treated in hospital settings compared with those diagnosed at later stages or treated in outpatient facilities. These results add to the growing body of literature identifying the need to create effective mechanisms for improving access to and participation in clinical trials among AYA patients with cancer.
Our findings are consistent with national estimates identifying steep declines in clinical trial enrollment as patients enter young adulthood.15
Since 1997, the NCI Cancer Therapy Evaluation Program has collected patient accrual data from all sites conducting NCI-sponsored clinical trials. It estimated that from 1997 to 2003, 10% to 15% of patients diagnosed between age 15 and 19 years participated in clinical trials. However, for patients age 20 to 30 years at diagnosis, clinical trial enrollment declined to 2% of patients, or just 5% of the rate in children and half the rate in adults age 40 to 65 years.14,15
Although our population-based study of clinical trial enrollment reflects similar patterns of decreasing enrollment by age, we found higher rates of enrollment than these prior estimates, potentially because of the exclusion of less common AYA cancers with lower than average accrual (eg, CNS tumors, sarcomas) and availability of single-year enrollment estimates.15
However, accrual rates for this population are still unacceptably low, falling well below clinical trial accrual in large, specialized institutions focused on treating younger populations (eg, children's hospitals), despite the higher likelihood of patients in SEER registries to live in urban areas with higher concentrations of these facilities.10
Additionally, our findings draw attention to the continued underenrollment of uninsured AYA patients. In a 1999 study of more than 2,300 patients with cancer in NCI-sponsored clinical trials in the southeastern United States, Klabunde et al16
found self-pay enrollees age 18 years and older were 60% less likely to participate than privately insured patients. More recently, in a 2010 study of 4,600 adult patients from a single institution, Klamerus et al17
found over 13% of patients with health insurance who were eligible for and willing to participate in a clinical trial were denied therapeutic trial enrollment because of lack of insurance coverage approval for participation. Because routine care costs associated with clinical trials can be substantial, this barrier can present significant challenges to clinical trial enrollment.17
Consistent with these results, our findings add to the limited data on the disproportionately low enrollment of uninsured AYAs onto clinical cancer trials.18–20
Considering that uninsurance rates peak during young adulthood,21
incorporating newly available reimbursement and coverage mechanisms for this population after a cancer diagnosis will be important to reduce barriers to clinical trial enrollment.22
Our study also points toward potential barriers to efficient treatment initiation. Our results identified advanced stage and outpatient-based care as important contributors to longer time from diagnosis to treatment. Although some cancers require more extensive workup before treatment (eg, computed tomography scans, bone marrow biopsies), leading to longer time from diagnosis to treatment, over 18% of patients treated in outpatient facilities had not begun treatment more than 60 days after diagnosis. Additionally, although stage I germ cell cancers are predominately treated with surgery at diagnosis, we continued to find longer time to treatment initiation in those with high- versus low-stage cancers when each site was examined separately. These findings are consistent with research identifying the multifactorial nature of delayed treatment, which may be driven by a number of patient, provider, and system factors.23,24
Delays may result from combinations of insurance coverage at diagnosis,25
patient preferences surrounding treatment,26
or the necessity to refer patients to specialists for care (eg, ALL). Future research should focus on identifying efficient transitions from diagnosis to treatment, including mechanisms for referral of AYA patients to tertiary care facilities that offer specialized cancer treatment or clinical trial participation (eg, comprehensive cancer centers, children's hospitals, AYA cancer programs).
Although many factors contribute to low clinical trial participation, one obstacle often identified is the lack of clinical trials for a patient's specific disease and stage at diagnosis. This may result from the relative concentration of clinical trials around common malignancies such as ALL and sarcoma, which have higher incidence among younger patients.7,27
As a result, protocols for these cancers often have less lag time between the end of one trial and initiation of another because of the increased number of patients overall and their ability to reach enrollment targets faster.7,28,29
Several mechanisms exist to provide information to physicians about available trials; however, future research should identify effective ways to promote physician utilization of these resources.
In addition, ambiguity surrounding standards of care may also result in treatment delays. Lack of evidence-based guidelines specific to the AYA population often results in a decision to treat these patients based on arbitrary age cutoffs,30,31
such as 18 years of age. Even though treatment practices directed at adults and children can differ substantially, those younger than age 18 or 21 years are often referred to pediatric oncology specialists, who as we demonstrated are associated with higher clinical trial enrollment, whereas older patients are referred to medical oncologists.28
Multiple studies of ALL in North America and Europe have shown AYA patients treated with pediatric-based protocols have better outcomes than AYA patients treated with protocols used for adults older than 18 years of age.32–38
Currently, the CALGB (Cancer and Leukemia Group B) 10403 study is evaluating whether pediatric-based protocols for patients with newly diagnosed ALL produce similar outcomes when administered by pediatric oncologists compared with hematologists/oncologists.
In response to the lack of available clinical trials for AYA patients with cancer, several approaches have been instituted. In 2000, the Children's Oncology Group started its own AYA committee to establish disease-specific study groups to increase therapeutic and biologic research protocols for this population. Additionally, numerous Children's Oncology Group protocols have extended upper age limits to 30 years for ALL and 50 years for rhabdomyo- and Ewing sarcomas. Furthermore, in June 2009, the NCI and Lance Armstrong Foundation sponsored a workshop to investigate whether the cancer biology in AYA patients diagnosed with breast cancer, colon cancer, and ALL are unique compared with that in other age groups. These initiatives provide promising new opportunities to address issues of effective treatment and clinical trial enrollment in the AYA population.
In summary, our study identifies low cancer trial participation in AYA patients with cancer, particularly among young adults. These findings support the need for improving access to clinical trials in the AYA patient population through continued education of patients and treating physicians, which will translate into better understanding of the biology of AYA cancers and ultimately into improved survival.