Long-term outcomes among survivors of childhood central nervous system tumors: A report from the childhood cancer survivor study
Reviewer: Christine Hill, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 1, 2008
Presenter: G. T. Armstrong Presenter's Affiliation: St. Jude Children's Research Hospital, Memphis, TN Type of Session: Scientific
Over 270,000 childhood cancer survivors reside in the United States, and this number is expected to grow as survival rates after childhood cancer continue to increase. Currently, one in 640 adults aged 20-39 years in the US is a childhood cancer survivor.
Of all adult survivors of childhood cancer, approximately 60% report having a chronic health condition, and 28% report having a severe or life-threatening medical condition.
Children treated for central nervous system (CNS) malignancies may be at particular risk for late effects from cancer treatment, including neurocognitive delays, endocrinopathies, and neurologic deficits.
Increased risk may be due to tumor location, as well as late-effects from therapy employed.
As multimodality treatment with surgery, radiotherapy, and chemotherapy results in improvements in long-term survival for children with brain tumors, increased potential exists for late morbidity and mortality.
The Childhood Cancer Survivor Study (CCSS) is a database of patients who have survived at least 5 years after being diagnosed with childhood cancer. Patients have been enrolled from 26 institutions across the US and Canada, and data is maintained and accumulated via review of medical records as well as healthcare provider and survivor survey-based reporting.
The study presented here utilized the CCSS database to retrospectively review the cases of children treated for CNS malignancies between 1970 and 1986.
The intent of this study was to investigate not only long-term survival, but the incidence of second malignant neoplasm and chronic medical conditions, as well as sociodemographic outcomes, in adult survivors of childhood CNS malignancies.
Materials and Methods
The authors utilized the CCSS database to identify 2,887 adult survivors of childhood brain tumors.
Of these, 491 had been lost to follow-up, and 520 refused participation in the study.
A remaining 1,876 survivors were included in this study.
Data on initial tumor diagnoses and treatments was abstracted from medical records.
All deaths were confirmed via review of the National Death Index.
Second malignant neoplasm (SMN) tissue was reviewed by a central pathology review center.
Survivors and their siblings completed surveys to provide information on late-effects and quality of life.
Outcomes including mortality, SMN, health status, education, employment, insurance, and marital status were assessed to determine risk associated with CNS tumor type and treatment modality.
Of the 1,876 adult survivors considered, 1,305 were survivors of astrocytoma/ glial tumors, 306 survivors of medulloblastoma/ PNET tumors, and 115 survivors of ependymoma.
Patients were diagnosed with primary malignancy between 1970 and 1986, and median follow-up was 22.1 years (range 16.1 – 34.6 years).
Of all patients considered, 26% had received surgical treatment alone, 42% had undergone surgery and radiotherapy, and 27% had undergone surgery, radiotherapy, and chemotherapy.
Sixty-eight percent of patients had received radiation to the whole or partial brain; 50% of these received dose of at least 50 Gray (Gy).
Overall survival in the cohort considered was 74.2% at 30 years; this is compared to a 98.5% survival rate in the age- and gender-matched United States population.
Overall survival differed for men and women (71.9% versus 76.9%, respectively).
A total of 546 deaths were observed > 5 years after treatment completion, resulting in a cumulative late mortality rate of 18.2% at 25 years.
The majority (61%) of deaths were due to tumor recurrence.
Second malignant neoplasms accounted for 9% of deaths, and cardiac and pulmonary problems together accounted for an additional 6%.
Late mortality rates varied according to initial tumor diagnosis, and were 23.3% for survivors of ependymoma, 19% for survivors of medulloblastoma/ PNET, and 17.3% for survivors of astrocytoma.
The overall mortality ratio was 18.0, describing an 18-fold increase in risk of death for adult survivors of childhood CNS malignancies when compared to age- and gender-matched controls.
Mortality ratios also varied according to diagnosis, and were 13.5 for survivors of ependymoma, 13.1 for survivors of medulloblastoma/ PNET, and 9.8 for survivors of ependymoma.
Annual risk of death was found to be 1.26% for ependymoma survivors, 1.1% for medulloblastoma/ PNET survivors, and 0.64% for astrocytoma survivors.
In the overall group of adult survivors, a 14% risk of death from recurrent tumor was observed, as well as a 2.8% risk of death from SMN at 30 years.
Second malignant neoplasms developed in a total of 9.1% of survivors (n = 223).
Of these, 53 were meningiomas, 100 were non-melanoma skin cancers, and 70 were other malignant tumors (14 new CNS malignancies, 12 soft tissue sarcomas, 12 thyroid cancers, and 3 leukemias).
An 18.5% increased risk of development of novel CNS malignancy was observed in the CNS tumor survivor population compared to the baseline adult population. Risk of meningioma development was increased by 2.9%, and risk of another SMN increased by 3.8%.
Risk of SMN appeared to be affected by cranial radiation, and a dose relationship was observed:
Risk of SMN was 12.2% in patients who received cranial radiation to a dose of 50 Gy or greater.
For patients who received doses from 1 Gy – 49 Gy, risk of SMN was 8.6%.
For patient who received no radiotherapy, risk of SMN was 4.5%.
Adverse health status was reported by 57% of adult survivors. The presence of chronic health conditions in survivors who did not have previous chronic health conditions was investigated:
New endocrinopathies were observed in 32% of survivors, representing a 20-fold increased risk over the general US population.
New neurologic deficits were observed in 72% of survivors, including seizures in 33%, difficulties with balance in 52%, and blindness in 16%.
Treatment with radiotherapy and/ or chemotherapy versus surgery alone was associated with increased risk of severe impairment in general health (odds ratio 2.8, p < 0.001).
When adult survivors were compared to their siblings, they were found to be less likely to have graduated from high school, be married, be employed, or have health insurance.
Survivors were 12 times less likely to be employed than their siblings, and 4 times less likely to be married or to have graduated from high school.
Radiation to at least 30 Gy was associated with statistically significantly lower rates of high school graduation, marriage, and employment.
The authors conclude that adult survivors of childhood CNS malignancies are at high risk for late mortality and morbidity.
Survivors have a 9% risk of development of second malignant neoplasm, as well as significant risks of development of other medical conditions.
Compared with their siblings, they are less likely to have attained a high-level education, to be married, or to be employed. Certainly, these factors affect quality of life.
The authors note that this population represents a high-priority group for intervention strategies.
This study represents a large, cohort analysis of adult survivors of childhood CNS tumors. It provides long-term follow-up, and an opportunity to observe late mortality and health effects in this population.
Based on the data presented here, adult survivors of childhood CNS tumors are at considerable risk for late mobility and mortality.
This population represents a group that should have access to diligent screening for SMNs, as well as other health issues such as cardiac and pulmonary disease.
Previous groups have shown that survivors of childhood cancer may not receive adequate risk-based health care (Nathan, ASCO, 2007); certainly, interventions from the oncology community are necessary to ensure that survivors are actively cared for as needed.
Survivors receiving care at a dedicated cancer center have previously been demonstrated to receive more active risk-based care than those followed in the community setting (Nathan, 2007); survivors of childhood CNS malignancies should likely be encouraged to seek follow-up care at a cancer center, ideally in a clinic setting dedicated to late-effects.
Although neurocognitive deficits in all likelihood contribute to decreased rates of high school graduation, marriage, and employment in this population, access to adequate social services could potentially increase these rates. Obtaining health insurance is particularly important for this survivor population which remains at risk for innumerable health issues and concerns.
The results presented here are somewhat limited by the era in which survivors were treated, as treatment modalities have changed considerably since the 1970s and 1980s.
Unfortunately, however, newer treatments are still associated with late health risks, many of which are the same as those presented here.
An expansion of the CCSS to include patients diagnosed with childhood cancers from 1987 – 1999 is planned; this will provide more current information, particularly regarding chemotherapies, many of which were not introduced until the mid-1980s. This expansion will also provide a larger and more modern population for analysis.
Having said this, the data presented here is an important contribution to our understanding of risks of late morbidity and mortality in survivors of childhood CNS tumors. Even as efforts are made to reduce these risks for patients treated today with advancements in surgical technique and radiotherapy technology, a large adult survivor population currently resides in the United States.
Diligence to the care of these and future survivors is essential to our ultimate global goal of improving outcomes for children treated for CNS malignancies.
Apr 1, 2011 - Double informant data indicate that a considerable percentage of adult survivors of childhood central nervous system tumors have long-term unmet health care needs, according to a study published online March 8 in Cancer.