Risk of somatic diseases in offspring of survivors with childhood or adolescent central nervous system tumor in Sweden

Abstract With the improvement of treatments, a growing number of survivors with childhood or adolescent central nervous system (CNS) tumor are parenting their own children. We aimed to explore the risk of somatic diseases among children of these survivors compared to population controls. Children of survivors with CNS tumor below age of 20 were identified between 1973 and 2014 by combining the several Swedish registers. Five children without parental CNS tumor were matched randomly to generate the population comparisons. Relative risk (RR) and absolute excess risk (AER) were calculated for overall somatic diseases, and hazard ratio (HR) was calculated for specific type of somatic diseases. A total of 2231 somatic disease diagnoses were identified in children of survivors with a cumulative incidence rate of 94.77 per 1000 person‐years, whereas the rate was 92.79 in matched comparisons thus resulting in an overall RR of 1.02 (95% CI = 0.98‐1.07) and AER of 1.98 (95% CI = −2.06, 6.13). Specifically, five of 1364 children of survivors had CNS tumor with an incidence rate of 0.21 per 1000 person‐year, whereas the rate was 0.04 in children of matched children, generating a HR of 4.91 (95% CI = 1.42‐16.96). Children of male survivors were at a statistically increased risk of malignancy, as well as infectious and parasitic diseases. In conclusion, no significantly higher risk of overall somatic diseases was found in children of survivors with CNS tumor before the age of 20, but children with a paternal diagnosis of CNS tumor had significantly increased risk of malignancies and infectious and parasitic diseases.

2010s being 75% for females and 61% for males in Sweden. 1 Many of these survivors might plan to have children when they reach parenting age. 2 Available evidence from population-based studies suggested that cancer survivors experienced an increased risk of infertility and adverse birth outcomes; this indicates cancer itself or treatments play a role in the reproductive system. [3][4][5][6][7][8] Adverse birth outcomes have been linked to long-term morbidity. 9 Besides, damage in reproductive organs may lead to genetic or epigenetic mutations in gametes, which may subsequently influence the physical health of their children. 10 Unfortunately, findings of the physical health in children of cancer survivors are limited and mixed. 11,12 We previously observed an elevated risk of being born preterm in children of survivors with CNS tumor before the age of 20 years by linking several Swedish nationwide registers. 13 In this populationbased study, we further explored the somatic disease burden among offspring of these survivors by comparing the cumulative incidence rate of somatic diseases between children of survivors and the comparison children. Furthermore, we classified diagnoses of somatic diseases into 12 main diagnostic groups to provide a detailed assessment for specific somatic diseases and investigated the risk of specific type of disease among children of these survivors. We selected children whose parents were ever diagnosed with CNS tumor under the age of 20 years and their parents had survived for at least 5 years after the diagnosis. Five children, whose parents did not have a diagnosis of CNS tumor, were randomly matched to each child of survivors conditional on the same birth year (continuous), gender of offspring, maternal and paternal age at birth (continuous). To ensure that the child was conceived after parental diagnosis, children were excluded if they were born within 1 year after parental diagnosis with CNS tumor. Children with neonatal death were excluded if they died within 3 months after birth.
In Sweden, a unique individual national identification number is assigned to each resident living in Sweden longer than 3 months, which was replaced by a serial number to provide anonymity, and used to link several registers in our study.

What's new?
Survivors of central nervous system (CNS) tumors are at higher risk of experiencing late adverse effects on reproductive function. However, it remains unknown if the tumor or related treatments subsequently affect the physical health of their offspring. This population-based study did not find an association of parental CNS tutors in childhood or adolescence with the risk of overall somatic diseases in the offspring. Nonetheless, an increased risk was observed in preterm born children, and children of male survivors had an increased risk of malignancies and infectious and parasitic diseases, calling for a tailored surveillance strategy.  diseases of the genitourinary system (ICD-10 codes: N00-N99). If an individual was recorded more than once for a specific type of somatic diseases, only the first record was retained (ie, only the first incident diagnosis for somatic diseases was retained).
The primary outcome was the number of overall somatic diseases, which was calculated by summing the 12 main types mentioned earlier. The secondary outcome was a specific type of somatic diseases.
In addition, these children were further linked to the Cause of Death Register to identify the date of death and the cause of death.

| RESULTS
A total of 1364 children were born after parental diagnosis of CNS tumor, and 6820 children born from parents without CNS tumor were selected as the reference group matched by gender, year of birth, and maternal and paternal age at birth ( Table 1).
The association between parental diagnosis with CNS tumor and risk of somatic diseases is presented in  The HRs and 95% CI for a specific type of somatic disease are shown in Figure 1 and Supplementary Table 1, and the stratified analysis by maternal or paternal diagnosis is shown in Figure 2 and Supplementary Table 2. Children of survivors were related to a 14% higher risk of being diagnosed with infectious and parasitic diseases than the corresponding children (95% CI = 1.01-1.30), which was amplified in children of male survivors (HR = 1.23, 95% CI = 1.03-1.47). Moreover, the highest HR was found for malignant neoplasm (HR = 1.58) and benign neoplasm (HR = 1.16) but they were not significant ( Figure 1).
However, as shown in Supplementary Table 3 A previous cohort study conducted among the Danish population found that offspring of cancer survivors in childhood and adolescence were not associated with a higher risk of overall hospitalization with a HR of 1.05 (95% CI, 0.98-1.12) when compared to the general population, but our study lacked detailed information for specific cancer type. 12 Other two register-based studies from Sweden demonstrated no difference concerning mortality between offspring of female or male cancer survivors compared to the general population. 16,17 In the present study, we also did not observe an increased risk of overall somatic diseases in offspring of survivors of childhood and adolescent CNS tumor. It is well-known that preterm birth is an important factor regarding long-term growth damage and morbidity. 18 Our previous study found a significantly increased risk of preterm birth among based studies also demonstrated an elevated incidence of experiencing preterm birth or other adverse pregnancy outcomes in female cancer survivors but not in male survivors. [3][4][5][6][7][8] After stratifying by preterm birth, we found that preterm birth strengthened the association of somatic disease risk with maternal diagnosis but played a small role in the association with paternal diagnosis. These findings might indicate that maternal diagnosis probably affects the physical health of their children via adverse birth outcomes, whereas paternal diagnosis might take an effect via genetic or epigenetic mutation. Emerging evidence suggested that epigenetic changes in male sperms could transfer to their children, which may contribute to disease susceptibility in offspring of mammals. [19][20][21][22][23][24][25] For example, chemotherapy in male adolescents was found to alter sperm DNA methylation, which could be transmitted to the next generation and promote the epigenetic transgenerational inheritance of disease. 24 Besides, age of parental diagnosis also modified the observed association with a higher risk in children of adolescent survivors than those of childhood survivors, but preterm-born children of childhood survivors were at the highest risk.
Furthermore, the risk of somatic diseases became weaker in children of survivors diagnosed later than 1990, which may be related to the improvement of treatments or the development of in vitro fertilization. in vitro fertilization was first adopted in 1982 and was very rare before 1990 in Sweden. 26  Cancer susceptibility is heritable, and several malignant neoplasms were demonstrated to have familial aggregation, including nervous system tumor. [31][32][33][34][35][36][37][38] The causes of CNS tumors are largely unknown, but family history is an established risk factor although a large number of familial aggregations cannot be explained by specific genetic mutations. 37 Our findings supported that parental CNS tumor was a risk factor for CNS tumor. Besides, the increased risk of overall malignant neoplasms was significant among offspring of male survivors rather than female survivors. The incidence rate of infectious and parasitic diseases was also significantly higher only in children of male survivors. As mentioned earlier, available evidence found that epigenetic changes in male sperms may lead to disease susceptibility in offspring of mammals, including cancer susceptibility. 21 The mechanism behind the difference between maternal and paternal diagnoses is worth deep investigation. This is, to the best of our knowledge, the first population-based study to explore the physical health among children of survivors with CNS tumor in childhood or adolescence. The strengths of our study include the high quality and nationwide coverage of registers, the verified disease diagnoses and the large randomly selected matched comparisons. The main limitation is the lack of treatment data, making it unavailable to examine the impact of specific treatments on somatic health in offspring. Data about diagnosis in outpatients were not completed until 2001. However, the matched comparisons were selected conditional on the birth year to ensure the comparability of the cumulative incidence rate of somatic diseases between the two groups.
In conclusion, offspring of survivors with CNS tumor in childhood or adolescence were not associated with a higher risk of overall somatic diseases. But an increased risk was observed in preterm-born children. Children whose father were ever diagnosed with CNS tumor in early life were related to an increased risk of malignancies, as well as infectious and parasitic diseases, which calls for a tailored surveillance strategy.