Second malignant neoplasms in five-year survivors of childhood cancer: Childhood cancer survivor study

Joseph P. Neglia, Debra L. Friedman, Yutaka Yasui, Ann C. Mertens, Sue Hammond, Marilyn Stovall, Sarah S. Donaldson, Anna T. Meadows, Leslie L. Robison

Research output: Contribution to journalArticlepeer-review

644 Scopus citations


Background: Because survival rates among childhood cancer patients are increasing, assessing the risk of second and subsequent malignant neoplasms (SMNs) is ever more important. Using the Childhood Cancer Survivor Study cohort, we identified the risk of SMNs. Methods: A retrospective cohort of 13581 children diagnosed with common cancers before age 21 years and surviving at least 5 years was constructed with the use of data from patients treated at 25 U.S. and Canadian institutions. SMNs were ascertained through self-administered questionnaires and verified by pathology reports. Information on therapeutic exposures was abstracted from medical records. The risk of SMN was evaluated by standardized incidence ratios (SIRs) and excess absolute risk. Poisson multiple regression models were used to assess the impact of host and therapy factors on the risk of developing SMNs. All statistical tests were two-sided. Results: In 298 individuals, 314 SMNs were identified (SIR = 6.38; 95% confidence interval [CI] = 5.69 to 7.13). The largest observed excess SMNs were bone and breast cancers (SIR = 19.14 [95% CI = 12.72 to 27.67] and SIR = 16.18 [95% CI = 12.35 to 20.83], respectively). A statistically significant excess of SMNs followed all childhood cancers. In multivariate regression models adjusted for therapeutic radiation exposure, SMNs of any type were independently associated with female sex (P<.001), childhood cancer at a younger age (P for trend <.001), childhood Hodgkin's disease or soft-tissue sarcoma (P<.001 and P =.01, respectively), and exposure to alkylating agents (P for trend = .02). Twenty years after the childhood cancer diagnosis, the cumulative estimated SMN incidence was 3.2%. However, only 1.88 excess malignancies occurred per 1000 years of patient follow-up. Conclusions: Success in treating children with cancer should not be overshadowed by the incidence of SMNs. However, patients and health-care providers must be aware of risk factors for SMNs so that surveillance is focused and early prevention strategies are implemented.

Original languageEnglish (US)
Pages (from-to)618-629
Number of pages12
JournalJournal of the National Cancer Institute
Issue number8
StatePublished - Apr 18 2001


Dive into the research topics of 'Second malignant neoplasms in five-year survivors of childhood cancer: Childhood cancer survivor study'. Together they form a unique fingerprint.

Cite this