Long term risk of subsequent malignant neoplasms after treatment of childhood cancer in the DCOG LATER study cohort: Role of chemotherapy

Long term risk of subsequent malignant neoplasms after treatment of childhood cancer in the DCOG LATER study cohort: Role of chemotherapy
Childhood cancer survivors (CCS) are at increased risk for secondary malignancies. Although radiotherapy is one of the major known risk factors, less is known about the impact of chemotherapeutic agents. This is a cohort study that evaluated a large Dutch population of CCS and studied their risk of developing subsequent malignant neoplasms.
This study used a cohort of 6165 CCS with detailed information on individual treatments and information on subsequent malignant neoplasm follow-up. The cohort included patients treated before the age of 18 years in one of seven Dutch paediatric oncology centres between January 1st 1963 and December 31st 2001. Treatment-associated risks of breast cancer, sarcoma and all solid tumours were assessed.
The median follow up was 20.7 years since first diagnosis. The cumulative subsequent malignant neoplasm incidence at 25 years after first diagnosis was 3.9% and interestingly did not change noticeably among childhood cancer survivors treated in the 1990s compared with those treated earlier. Doxorubicin was associated with a dose dependent increased risk of female breast cancer (p = 0.001) whereas cyclophosphamide was associated with dose dependent increased risk of sarcoma (p = 0.01). The doxorubicin related breast cancer dose response was stronger in survivors with Li-Fraumeni syndrome-associated cancers suggesting a gene-antracycline interaction in the development of breast cancer.
A limitation of the study is that despite the large size of the cohort the number of events for most of the subsequent malignant neoplasm sites was fairly low as a result of the wide age distribution at the end of the follow-up (5.3 - 65.1 years). As with other large survivorship cohorts the authors acknowledge that correlations between patient and treatment factors can potentially hamper the ability to extract meaningful information in the multivariable analysis.
This study suggests that doxorubicin exposure in childhood cancer treatment is correlated with the risk of subsequent solid cancers and breast cancer whereas cyclophosphamide exposure is correlated with an increased risk of subsequent sarcomas. It is also shown in this cohort that genetic susceptibility may influence doxorubicin-associated breast cancer risk. However, it is unable to establish a causative role of these chemotherapeutic agents.