Long-term causes of death among pediatric patients with cancer

Samantha R. Horn, Kelsey C. Stoltzfus, Heath B. Mackley, Eric J. Lehrer, Shouhao Zhou, Smita C. Dandekar, Edward J. Fox, Elias B. Rizk, Daniel M. Trifiletti, Pooja M. Rao, Nicholas G. Zaorsky

Research output: Contribution to journalArticle

Abstract

Background: The objectives of this study were to characterize the risk of death (1) from the primary cancer vs competing cause of death; and (2) from various causes of death vs the general poplation. The relative risk of death after a pediatric cancer diagnosis versus the general population and the risk of death from a primary cancer diagnosis versus competing causes of death. Methods: This retrospective, population-based study used the Surveillance, Epidemiology, and End Results database (1980-2015) and included patients aged 0 to 19 years at the time of diagnosis. Observed deaths were calculated; the risk of death versus the general population was assessed with standardized mortality ratios (SMRs). Competing risk models for the cause of death were performed. Results: There were 58,356 patients who were diagnosed, and the mortality rate was 22.8%. To assess causes of death, 6996 patients who died during the study period were included (45,580 total person-years at risk): 5128 (73%) died of their primary cancer, and 1868 (27%) died of a competing cause. Among all patients, the rate of death from the index cancer was higher than the rate of death from another cause within the first 5 years after diagnosis. The risk of death from a nonprimary cancer began to supersede the rate of death from the primary cancer 10 years after diagnosis for patients with germ cell tumors, lymphomas, and sarcomas. SMRs for the primary cancer were highest within the first 5 years after diagnosis for all cancers (SMRs, 100-50,000; P <.0001). The risk of death from competing causes (heart disease, suicide, and sepsis) was elevated (SMR, >100; P <.001). The risk of dying of heart disease was high, especially for patients with astrocytomas (SMR, 47.84; 95% confidence interval [CI], 27.87-76.59) and neuroblastomas (SMR, 98.59; 95% CI, 47.28-181.32). The risk of dying of suicide was high in most patients, particularly for those with osteosarcomas (SMR, 111.40; 95% CI, 2.82-620.69), Hodgkin lymphomas (SMR, 62.35; 95% CI, 34.89-102.83), and gonadal germ cell tumors (SMR, 28.97; 95% CI, 12.51-57.09). Conclusions: The cause of death for patients with gonadal germ cell tumors, lymphomas, and sarcomas is more commonly a secondary cancer or noncancerous cause than the primary disease; their risk of death from competing causes (heart disease, suicide, and sepsis) rises throughout life.

Original languageEnglish (US)
Pages (from-to)3102-3113
Number of pages12
JournalCancer
Volume126
Issue number13
DOIs
StatePublished - Jul 1 2020

Keywords

  • competing risks
  • oncology
  • pediatrics

ASJC Scopus subject areas

  • Oncology
  • Cancer Research

Fingerprint Dive into the research topics of 'Long-term causes of death among pediatric patients with cancer'. Together they form a unique fingerprint.

  • Cite this

    Horn, S. R., Stoltzfus, K. C., Mackley, H. B., Lehrer, E. J., Zhou, S., Dandekar, S. C., Fox, E. J., Rizk, E. B., Trifiletti, D. M., Rao, P. M., & Zaorsky, N. G. (2020). Long-term causes of death among pediatric patients with cancer. Cancer, 126(13), 3102-3113. https://doi.org/10.1002/cncr.32885