Age Research for Childhood Cancer

Research background

Childhood cancer treatment has improved in the past several decades, with a current average five-year childhood cancer survival rate of 80%. Despite advances, there is still a small group of children with cancer who do not survive long enough after diagnosis to receive cancer treatment or they die early in treatment.

It's not known why this group of children experiences such poor outcomes, primarily because most pediatric oncology outcomes are reported from clinical trials. Pediatric patients who experience early death (defined here as death within one month of diagnosis) may be under-reported since they likely died before enrolling in a clinical trial (or were ineligible).

Research methods

Lead author Adam Green, MD, pediatric oncologist in the Center for Cancer and Blood Disorders at Children's Hospital Colorado, sought to define the scope and characteristics of this group of pediatric cancer patients using the National Cancer Institute's Surveillance, Epidemiology and End Results (SEER) database. The study described demographics, disease and socioeconomic status to identify at-risk groups of children in which to target interventions.

Data was retrieved from the SEER 13 registries on 36,337 childhood cancer patients under the age of 19 who were diagnosed between 1992 and 2001. Socioeconomic data was pulled from Census 2000 for each patient’s county of residence.

SEER 13 data includes the following cancer registries: Atlanta, Connecticut, Detroit, Hawaii, Iowa, New Mexico, San Francisco-Oakland, Seattle-Puget Sound, Utah, Los Angeles, San Jose-Monterey, rural Georgia and the Alaska Native Tumor Registry. This database represents 13% of the U.S. population and was designed to represent geographic and social diversity.

The early death and non-early death pediatric cancer groups were compared on the following variables: age, sex, disease type, socioeconomic status, race and ethnicity. Cancers were divided into three broad tumor categories: non-central nervous system solid tumors, hematologic malignancies and central nervous system tumors.

Research results

Of the 7,403 cancer patients who died, 555 died within one month of diagnosis (7.5% of total deaths). Additionally, the study revealed:

• Upon multivariable analysis, pediatric cancer patients under the age of 1 represented the strongest predictor of death among all disease groups.
• Pediatric patients diagnosed with acute myeloid leukemia, infant acute lymphoblastic leukemia, hepatoblastoma and malignant brain tumors had the highest incidence of early death.
  • Malignant melanoma, Hodgkin lymphoma, osteosarcoma, low-grade glioma and Wilms tumor represented the lowest risk of early death.
• In multiple disease groups, Black race and Hispanic ethnicity were risk factors for early death.
• Pediatric cancer patients who lived in counties with lower than average income were associated with high risk of early death in hematologic malignancies.
• Percentages of early death decreased over time, particularly in hematologic malignancies.

Research conclusion: Childhood cancer diagnosis at a very young age is the strongest predictor of early death

Risk factors of early death in childhood cancer patients include an age of under 1 year, certain diagnoses, race and ethnicity, and disadvantaged socioeconomic status. While early death rates are decreasing, the scope of early deaths reported in clinical trial literature is significantly less than what was found in the population-based SEER database. For example, early death from non-infant acute lymphoblastic leukemia (ALL) was observed to be twice as common than what clinical trial data reports. Wilms tumor early deaths were 12 times more common in the SEER data versus the clinical trial data.

Study authors recommend further work to determine reasons for these disparities, and the development of interventions for early identification and prevention of this problem.