Abstract
BACKGROUND AND OBJECTIVES: The aim of this study was to analyze late mortality among 5-year survivors of childhood cancer, in Piedmont (Italy), in terms of risk factors and causes of death. DESIGN AND METHODS: From 1967 to 1999, the Childhood Cancer Registry of Piedmont recorded 3164 incident cases. Patients identified only by a death certificate (n = 59), lost to follow-up (n = 32), alive with a period of observation shorter than 5 years at the end of follow-up (n = 65) and records corresponding to a second malignant tumor during childhood (n = 9) were excluded from the analyses. RESULTS: Within 5 years after diagnosis, 1301 children died, and among the 1698 5- year survivors, 144 children subsequently died. Among 5-year survivors, cumulative mortality percentages increased from 5.1% (95% CI 4.0-6.2) at 10 years after diagnosis to 16.0% (12.2-19.8) at 35 years. Period of diagnosis (p = 0.006), age at diagnosis (p = 0.002), and tumor type (p = 0.003) were associated with late mortality. Most deaths were related to cancer recurrence (62.2%) and treatment-related sequelae (22.4%), including second malignant neoplasms, cardiac diseases and other late effects. Compared to the general population, children included in this study had a 9-fold increased risk of overall mortality, and experienced an absolute excess of 4.4 deaths per 1000 person-years. INTERPRETATION AND CONCLUSIONS: Among 5-year survivors, patients treated more recently (after 1979) had a statistically significant lower risk of late death than those treated earlier. However, long-term survivors still experienced higher mortality rates than those in the general population, and recurrence or progression of the primary tumor was the first cause of death.
Vol. 91 No. 8 (2006): August, 2006 : Articles
Published By
Ferrata Storti Foundation, Pavia, Italy
Statistics from Altmetric.com